The_Proceedings_of_the_Institution_of_El

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project to make the world’s books discoverable online. Ithassurvivedlongenoughforthecopyrighttoexpireandthebooktoenterthepublicdomain. Apublicdomainbookisonethatwasneversubject tocopyrightorwhoselegalcopyrighttermhasexpired. Whetherabookisinthepublicdomainmayvarycountrytocountry. Publicdomainbooks are our gateways to the past, representing a wealth of history, culture and knowledge that’s often difficult to discover. Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book’s long journey from the publisher to a library and finally to you. Usage guidelines Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the public and we are merely their custodians. Nevertheless,this work is expensive, so in order to keep providing this resource,we have taken steps to prevent abuse by commercialparties, including placing technical restrictions on automated querying. We also ask that you: +Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for personal, non-commercial purposes. +Refrainfromautomatedquerying DonotsendautomatedqueriesofanysorttoGoogle’ssystem: Ifyouareconductingresearchonmachine translation,opticalcharacterrecognitionorotherareaswhereaccesstoalargeamountoftextishelpful,pleasecontactus. Weencouragethe use of public domain materials for these purposes and may be able to help. +Maintainattribution TheGoogle“watermark”youseeoneachfileisessentialforinformingpeopleaboutthisprojectandhelpingthemfind additional materials through Google Book Search. Please do not remove it. +Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other countries. Whether a book is still in copyright varies from country to country, and we can’t offer guidance on whether any specific use of any specific book is allowed. Please do not assume that a book’s appearance in Google Book Search means it can be used in any manner anywhere in the world. Copyrightinfringement liability can be quite severe. About Google Book Search Google’s mission is to organize the world’s information and to make it universally accessible and useful. Google Book Search helps readers discovertheworld’sbookswhilehelpingauthorsandpublishersreachnewaudiences. Youcansearchthroughthefulltextofthisbookontheweb athttp://books.google.com/ 1 Proceedings gee, Pere “0. é; o°@::2: celeoOs#9 eg29‘c=, @ @: 32aNcee (eg@ i es ‘2 Bee @ASINS < yexog®cg penne we8g.,@%...es @a >&2 .< *We s fe= Gan —Pe: e@..@. 2 AG oa |SS. og ae NA al -ZAEDyt panS cs 4 wa fri&~_ |§ Years Nitten Haye JOURNAL orm INSTITUTION OF ELECTRICAL ENGINEERS, THE SOCIBTY OFTELBGRAPH-ENGINERES AND BLECTRICIANS. FOUNDED 1871. INCORPORATED 1883. ORIGINAL COMMUNICATIONS ON TELEGRAPHY AND ELECTRICAL SCIENCE. PUBLIBHED UNDEE THE SUPERVISION OFTHE EDITING COMMITTEE, ¥. H. WEBB, Srorerazy. ———— \VoL. XVIN.—1889, . London: E.axw F. N.SPON, 125, STRAND, W.C. Sew York: 12, CORTLANDT STREET. 1690. —————————————— Ke aes {HARVARD\UNIVERSITY LIBRARY fahec7 CONTENTS. fi TABLE OF OONTENTS, Vor. XVII. Precedings ofaSpecialGeneralMostingofMembers,heldonJanuary108, 1889 :— Confirmation ofSpecialResolution passedonDecember20th,1888..1 Yrosedings ofthe One Handred and Bighty-third Ordinary General Meoting, held January 10th, 1699:—TranaferofAssociates totheClassof Members...wane aTransferofStudentstotheClassofAmociates ovvowsAsoouncement bythe Chairman that allLegal Requirements inreferencetotheChangeofTitlehavebeenfalflled ....2letterfromProfeworSilvanusP.Thompson, requesting tobeallowedtoDeclineAcceptancs ofthe«Society'sPremium,” awardedtohim;andAnnouncement oftheCouncil'sAmendedAwardsin Praeatation ofPremiums awarded toMr. A.0.Cockburn and Mr. award Btallibrass, and Statement astoPremium awarded to Mee BO. Walker ee aes Tnsagural Addrom oftheNew President, SirWilliam ‘Thomson— Ether, Electricity, and Ponderable Matter” awe‘YaojofThankstothePresidentforhisAddreu— SirWilliamThomion,inacknowledgment wewoe 40VoteofThanks totheRetiring President, Mr.Edward Graves— Major-General Webber enews ewe AL »Graves, inacknowledgment ne eae ae AB BeetionofNewForeignMember,Members,Amociates,andBtadents43 Prcetings ofthe One Hundred and Bighty-fourth Ordinary General |. Meeting, held January 24th, 1869:—TranaferofAssociatetotheClausof Members.onenessTranaerofBtadentstotheOlasofAmociates ne nes A Donation byMr Preece ofanElectric Clock fortheLibrary... 44 “The Insulation Resistance ofElectric Light Installations,” by ProfesorA.Jamieson,FR8.E.,M.Inst.OE,Member..“6 ‘TelegramfromMr.GeipelastoInnalationResistanceoftheEdin- ‘arghConservative ClubInstallation. awweaeveOL | iw CONTENTS. mor LetterfromM.Picou,ForeignMember,inreferencetohisFormulaquotedbyProfessorJamieson... oneness asaOL Remarks onProfemsor Jamieson’s Paper by— MrRE. Crompton eos OL wAlexanderSiement aS ‘TheChairman (Profetsor Ayrton). ose BSMr Homan eae wySydneyEversbed eee 8. yyBernardDrake eae ee OD weWeAChamen ate TL TheChairman ooo woe OTAdjournmentoftheDiscussion saa OF Election ofNew Members, Avmociates,and Stadents .. 77 Anstascra— W.Kohlrausch—« Lightning Conductors” vse os 7B J.Chappuis and G,Maneuvrier—"The Mechanism ofElectrolysis “bymeansofAlternatingCurrente” seaeae7D G,Maneuvrier and J.Chappuis—' Spontaneous Explosions occurring “during the Electrolysis ofAcidulated Water bymeans ofwanAlternating Current” weeee ease aesBO G,Chaperon and E,Mercadier— Electro-chemical Radiophooy” ... 80 W.E,Ayrton and J,Perry— ficiencyofGlowLampswithDirect andAlternatingCurrenta”0scene oaBE ©,Ferraris—“Electrodynamic Rotation produced bymeans ofAlternating Currents”... aeseems BD.ProfemorJA.EwingandW,Low—Influence ofaPlaneof“Transverse Section on the Magnetic Permeability ofan H.A,RowlandandL.Bell—"Actionof@MagnetonChemical A.Voller— Earthing Lightning Conductors bymeans ofGas andWaterPipes”aeee ea (Classified List ofArticles relatingtoElectricityandMagnetismappearing {insome ofthe principal English and Fore'gn Technical Journals uringtheMonthofJanuary,1889.00 moa B. Proceedings oftheOne Hundrcd and Bighty-ffth Ordinary General ‘Meeting, held February 14th, 1889:— ‘Transfer ofStudentstotheOlessofAssociates... ewe 8D CONTENTS. ’ ron Donations totheLibrary ae noe ae ae 8D Continuation ofDiscussion onProfessor Jamleton’s Paper on“The “Insulation Resistance ofElectrio Light Installations "—etterfromProfestorJamieson wm mss aee80LetterfromMr,B.D.Smelliew, aa ewe OD oyKempece nee eae 88. yyOE, Spagmotetti ee ee 8BBGTFleetwood ocean ene aesWByeHolroyd Smith oso A eTfeLamtCarpenter no eam NG tyANDERECappeliaeocean meHB Dr.John Hopkinson none ea aw Election ofNew Foreign Member, Members, Astociates, andStudents 120 Gmmanications inreference toProfestor Jamicson's Paper from— Me, William McWhirter cn sete vee me TBM mwBW. Beckinguale.. cn oe eae TD Professor Jamieson's Reply totheDiscussion onhisPaper... 125 Ummunication from Sir William Thomson, President, inreference Proceedings oftheOne Hundred and Eighty-sixth Ordinary General ‘Meeting, held February 21st, 1889 -— Tranafer ofAssociate totheClass ofMembers os wae ae 189 “On certain Phenomena connected with Imperfect Earth in “Telegraph Circuits," byA.E,Kennelly, Amociate ..., 129, Remarks onMr,Kennelly's Paper by— Mr. W.P.Granville se ayo eee ae UAT wyHO.Donovan cea 8 weWB Pree0e eae tee wee 149,187 teDO. Bate ae ese ane ate nee aes 8B,187yyHowardSwan nese ame 186 wi CONTENTS. . race ‘RemarksonMr.Kennelly'sPaper(continued)— Mr,C.W.8,Crawley senena nae ae8B ElectionofNewForeignMember,Members,Associates, andStudents 160 ‘Proceedings oftheOne Hundred and Eighty-seventh Ordinary General‘Meeting,beldFebruary28th,1889:—‘TranaferofStudentstotheClaasofAssociates oseae vueGLDonationstotheLibraryeeseeseason BL “Some Electric Lighting Central Stations inEurope, and their “Lemons,” byProfessor George Forbes, F.2.88.(L, &F,),Member 161 PostponementofDiscussionontheabovePapers.su.sxawe196 Election ofNew Members and Associates. ase sae oso 97. Accessions totheLibraryfromJanuaryIsttoMarch20th nysu.198 Assrascte—Dr. J.A.Fleming—“'A Design for Standard ofElectrical J.Parker—“Thermo-Electric Phenomena”... ose oe ous 201 F,Quincke— Electrolysis ofCuprous Chloride”... as20BellatiandLossana— PassageofElectricCurrentsthroughBail E,Drechsel—" Electrolysis bymeansofAlternating Carrents” —....208DrArvonval—"A Universal Dead-Beat Galvanometer™ —wesas208PF,Larroque— Permanent Changesproduced inCopperWiresbythePassageofCurrent” aeons anewee20 Dr. A.v.Waltenhofen—"Experimenta with the Accumulators of Parbaky and Sebenek™ aie aie wee cee 208,©,Heim—"The UseofAccumulators inTelegrapby” wusveane206 (Classified List ofArticles relatingtoElectricityandMagnetismappearing {insome ofthe principal English and Foreign Technical Journale uring theMonthofFebruary, 1889wesoe 08 Proceedings ofthe One Hundred and Bighty-eighth Ordinary General Meeting, held March 14th, 1889 :— Donations totheLibrary aeeeene eee MD‘Discussion onProfessorG.Forbes'sPaperon“SomeElectricLighting. “CentralStationsinEurope,andtheirLessons”— Mr.Gisbert Kappieseonevee aeesa UD Mr, J.Swinburne, oe oe ete BID weRV,Andere eee aeBDfyAlexander Siemens eng cece nee ‘CONTENTS. a DisastiononProfesorForbes'sPaper(ontinuad)—~ . wyMark Robina nae eee ®RE, Crompton eo BL WHEPreece eaAdjournmentofDiscussion ens e248 ectionofNowAuociatesandStudent, seem MO F 0 andEighty-ninth Ordi PromotingoftheOeHuntsandRighyainthOrdoaryGece! Donationst0theLibrary ame aneae ALContinuation ofDiscussion ofProfessor Forbee's Paper on“Some‘"BlestricLightingCentralStations,”&e.—MeWOHPrec eee meMAL,BEL Alexander Siemensiycea 24T,286(Arthur Wright cena 80 Major-General ,B,Webber(communicated)... iu258 ProfemorW.EAgronveee cee eaeBLProfessorGeorgeForbesinreply)ne eave 268 Becton ofNew Foreign Member, Members, Amociates, andStadent 281 Precedings ofthe One Hundred and Ninetith Ordinary General Meeting, held March 26tb, 1689:—TransferofAssociatetotheClaasofMembers... asoes282TransferofBtudenttotheClaseofAssociates ew oe288 AmnoancementoftheDeathofMr.0...Patey,0.8,andVotoof Condolence with Mr, Patey—Me,AlexanderSiemens0seesaeseas 288 ProfemorW.Grylls Adams.nae ae288Presentation andAdoptionofBalance-Sheet fortheYear 1688...284 “Laboratory Notes onAlternate-Current Circuits" byProfessor W.Ayrton, F.RS,V-P,andJohnPerry,P.RS,Member 286Discussion— MWR. Granville eee aoe HD MWB Been eee eo vit CONTENTS, OntheDisturbances arisingfromtheUseof‘Barth*forBlectrie “Lighting Purpoue,” byW.H.Preece, F.R.8., Past-President 314 Discassion— ‘lectionofNewMambers,Aucciates,and Students... ay823BalanceSheetforC80Oar1888vueeee ae BOL Communication (in reference tothe Discussion onProt, Jamieson's Paper on“Insulation Resistance,” &c.) giving Details’ oftheHastingsInstallation, byMr,F.B.Nicholson... wemow994 ‘Arrascrs:— H,Nageoka—" CombinedEtfectaofTorsionandLongitudinal Stress‘ontheMagnetisationofNickel".neane828 B.Dorn—"DeterminationoftheTrueOhm”enswe828 ©.vonWyis—"ResistanceofMagnetic Iron”...829 0.L,Weber—"ConductivityofSolidMercury”n,noe880 43.Klewencio— PlatinumIridiumforStandardResistances” =.$30B.Landmann—"BichromateBatteries"seawaeoweBBL MEDellatiand8,Lomana—"OcclusionofHydrogenby Nickel”.982 3.H, van Hoff and I.7,Reicher— Dissociation TheoryofElectro- (lamsided Listof Article relating toBlectricity and Magnetiam appearing 4msome ofthe principal English and Foreign Technical JournalsduringtheMonthofMareh, 1889.aeayene OM Proceedings ofthe One Hundred and Ninety-fst Ordinary General Meeting, held April 1th, 1889:—‘ravsterofStudentstotheClassof Associates.eee 88DonationstotheLIDARee aeae IT“Underground Conduite and Electrical Conductors,” by John B.Verity,Member eet 88Discussion ontheabove Paper—MrWOEGRY ene eave 868,872,975weAPPTyotter eee 885 SBS ERD eee eee 8ProfeworAyrton ees 810,878Mr. GisbertKapp oeeee an8BOS.PBIMPH ee eee TH GeorgeButton... oa onTTMajor-GeneralWebber(communicated)... S77 Me,Verity,inreply(communicated). weee 8 CONTENTS. ix HuctionofNewMembers, Amociates, andStadents ... ..1.86 Proceedings oftheOne Hundred and Ninety-second Ordinary General Meeting,heldApril26th,1889-—TransferofAssociatetotheClassof Members...wsvee 885,‘TranaferofStudentstotheClassofAmociates... a885 AnnouncementofthedeathofMr.WarrenDelaRue,M.A.,D.C.L, F.RS.,Member,andVoteofCondolence withMra,DelaRue...985 Professor 8.P. Thompeon ee ane nee oe 886 “On Lightning, Lightning Conductors, and Lightning Protectors,”byDr.OliverLodge,FR,Member... i nee 886Postponement ofDiscussionontheabovePaper...avweve80 Announcement astoExtension ofPeriod during which Associatestransferred fromtheGlassofStudentsmayAttendtheStudents’ ElectionofNewForeign Members, Associates, andStudent... ...430 Aamucre— &.P,Herroun— DivergenceofElectro-motive ForcesfromThermo- Webemical Data". ae ene see ate ama ABT A.Potier— Electrolysis ofMercurous Nitrate asaMeasure of J.Mooser—" Experiments onMicrophonic Contacts” wu. ass aus432 W.H.Schultze—'The Electrolytic Behaviour ofMica ataHigh wTemperature” eae ae teense 88 K,Schreber—Electro-motive Force ofThin Films ofHydrated HHertz"RaysofElectricForce”oeees aeeABE J.Bergmann— Observations ontheChangeintheConductivityof MetalsproducedbyHeating” aene anyone48 8.Tereschin— Specifle Inductive Capacity ofsome Organic “Wompounds™ a ae eee eae wee ABT D.Goldbammer—“Influence ofMaguetisation onthe Conductivity ot Metals ae ne ate ate ae 8B HEBéckstrom— Conductivity ofSpecular Iron Ores. avs 438 B.Nebel— Disintegration ofCopperbytheElectricCurrent”... 439 HLDube“CounterEMP.oftheAre”osesueare oe489‘Anon—*Fritsche'sDynamo” en)F.Neesen—M' Lightning Conductors” tues HO Clamiied ListofArticles relating toElectricityandMagnetimappearing insomeoftheprincipalEnglishandForeignTechnical JournalsdaringtheMonthofApril,1889i.ene oneones MAL © CONTENTS. Proceedings oftheOne Hundred and Ninety.thind Ordinary General Meeting,beldMay9th,1889-— Discussion ofProfessor Oliver Lodge's Paper, “On Lightning, Light- “ning Conductors, and Lightning Protectors”— QS.Wimabort, ence meOL,ATLProfesorW.GryllsAdame cenayeGRBitagerald ee TB«OliverLodge(inreplytotheabove)seTB ElectionofNewForeignMember,Member,and Associates...490 Proceedings ofthe One Hundred and Ninety-fourth Ordinary General ‘Meeting, beld May 16th, 1669:—‘TranaferofStudenttotheClassofAmociates am aeBLContinuation ofDiscussiononProfessorOliverLodge'sPaper,“On“Lightning, LightningConductor, andLightningProtectors”Me.J,Wimahurst(continuation of Remarks).me498 [email protected], PRS eee 40 Lieut-Ool,B.¥.Armstrong, BE,eye 05 wyBydney Evershed ..owa ee>.) BirWilliamThomgonne twas BIBMajorCardew,BLE.(communicated). unos 620LieutCol.J.T.Bucknill(communicated) xywonBRT Me.LeonardJoweph(communicated)wenveen2D ProfesorOliverLodge(infurtherreply)nvewae58L ElectionofForeignMember,Auocintes,andStudents. nw,505 Proceedings oftheOne Hundred and Ninety-Atth Ordinary General ‘Most, beld May 28rd, 1860:— “OntheSecurityagainstDisturbances ofShipsCompassesbyElectric“Lighting Appliances,” byBirWilliamThomson,D.O.L.,LID.PRSS.(L 8B),President aewe B8TDiseuasonontheabovePaper— StallCommander CreakyBN,PRS moe amTL ProfessorGeorgeForbetee eae TGAndrew Samienon nasa TB CONTENTS, a rack DiscussiononSirWilliamThomson's Paper(#ontieued)— Major P.Cardew, RE 1. ete BIB SirWilliamThomson(inreply) ssosesueev OTD“OnAlternate-CurrentWorking,”byW.M.Mordey, Member.583 BeetionofNewMember,Associates, andStudents wwau630 Proceedings ofthe One Hundred and Ninety.sixth Ordinary General ‘Meeting,heldMay0th,1889:— ‘TransferofAssociatetotheClassofMembersws.scesueves681 ‘Transfer ofStadenttotheClassofAssociates neceeaweOBISapplementary Remarks byMr. W.M.Mordey inreference tohisPaper,“OnAlternate-Carrent Working”... mewusoweO81DiscussionontheabovePaper—Dr.John Hopkinson (communicated)... awaeoweOAS.Professor W.GryllsAdams(communicated). msoweGAB, Dr. J.-A.Fleming, ose sateen ote 8. Mr.GiabertKAPweeeseeveene oneOB ‘Mr.Zipernowski (communicated) esse ueoneOTEv»GC.Fricker(communicated) nae aeossOTBhsMr.Mordey,inreply(communicated) wvuae ve815Lxhibition and Description ofhisElectricity Meter,byGeorgeHookham, Discussion thereon— Mr.B.E, Crompton eae nease OL ‘lection ofNewMember,Associates, and Student... 696, Original Communications— ‘Reealts ofTesta ofhisLightning Conductors atDhubri, Assam,by WW.MeGregor, Members. weaneweanes OEReportofaFatalAccidentfromLightning inIndia,byP,V.Lake, MoteronaStaticBlectro-Motor, byChatlesZipernowaki ww. TOL Accemions totheLibrary,fromApril1toJune 80...surveve708 xi CONTENTS. ClamifiedListofArticlesrelatingtoElectricityandMagnetismappearing insome ofthe principal English and Foreign Technical JournalsuringtheMonthsofMayandJune,1889 aoe 08; ‘Proceedings oftheOneHundredandNinety-seventh OrdinaryGeneral‘Meeting, held November 14th, 1889:—‘TransferofAssociatetotheClassofMemberswv ee 708‘TransferofBtudenttotheClaseofAssociates e700“The Lighting ofthe Centennial International Exhibition, “Melbourne, 1888 and 1389," byK.I,Murray,Member... ...10 ‘Discussion ontheabove Paper, Remarks by— Mr, WeM.Mordey aie aeons aeHSProfessorAyrton soe cee aee7494TOAMr5.8,Baworthe oncece nate TD»o (Communicatet) 7 Professor8.P,Thompson... see TSBMrRL.Couens oo oo TT ByGiabertKapp ean TOQL,Addembrooke eae 182wyW.HLPreece... a See Oh Dr.JohnHopkinson sens 08 Proceedings oftheOne Hundred and Ninety.cighth Ordinary GeneralMeeting,heldNovember 28th,1889;—‘TransferofArsociatetotheClassof Members... asa 170“«Blectrical Engineering inAmerica,” by G.I,Addenbrooke, ‘Discussion ontheabovePaper.Remarksby—ProfessorForbeswiewvseoneewes nea 188 weWP. Granville ean BS. Professor8.P.Thompson noes BO. MrT. Swinbame aan SirWilliamThomaon 2... eae ae806‘Mr,Addenbrooke (inpartreply) . oomBOTAdjournmentofDiscussionseaeoneone ne8 Election ofNew Foreign Members, Members, Associates, and Students ae ate aes 80 Anarascrs:— ‘W,Koblrauseh and (,Heim—"Exporimenta with Accumulators for Station Working” aus ane asa ee 10 ‘W.Wedding—" Are Light Photometry”... . 810‘W.Negbaur—“' Experiments onthePermeability ofdifferentSorts.MoronandSize) eee nea BL CONTENTS. ait Aermusct (continued) -— moe Dr. V.Wietlisbach—“TheInductionCoilsofMicrophones”... x.612 Dr.LeonardWeber—Atmoapheric Electricity” mwawe818 ©.Grawinkel—Connection ofAecumulators forTelegraphWork”818 RBoutyandL.Poincaré—ConductivityofMoltenSalta”......818 1M.Deprez—Electrie Tranamiasion ofPoweratBourganen!” —..815 (ClamiedListofArticlesrelatingtoElectricity andMagnetism appearinginsome ofthe principal English and Foreign Technical Journa'saringtheMonthsofJuly,August,September, andOctober,1889...816 Proceedings ofthe Eighteenth Annoal General Mecting, held on ‘December 12th, 1889 :— Presentation byLady Bright ofBust ofthelate SirCharles Bright 823, AppointmentofBeratineerswu.aeoneeeeaeBU ReportoftheSecretaryastotheLibrary... n,n aso 880Soggestions byMajorFloodPage neaneoneness 88‘VoteofThankstothePresident,Council,andMembersoftheTostita- tion ofCivil Engineers ase see ee ame em OST. ‘Vote ofThanks totheLocal Honorary Secretaries and Treasurers ... 887VoteofThankstotheHonoraryTreasurer owas 3BVoteofThankstotheHonoraryAuditors... ose 838 ‘VoteofThankstotheHonoraryBolicitory ne oeae888.Continuation ofDiscussiononMr.Addenbrooke's Paperon“Electrical “Engineering inAmerica.” Remarks by— )Addenbrooke (inreply) ce BHDReraltoftheBallotforPresident, Cvuncil,andOffcersfortheYear Vote ofThanks tothe Scrutineers w.. wie ones ne_ BB Dr,JohnHopkinson's Acknowledgment oftheHonourofbeingelected Accemions totheLibrary from Jaly 1toDecember 31,1889... 855 LustofArticlesrelatingtoElectricity andMagnetism appearing insome‘oftheprincipal Technical Journals during theMonths ofNovember Gmunication from SirWilliam Thomson inreference toMr. Preece’s‘Bemarkson Submarine Cables,intheDiscussion onMr,Addenbrooke's —_ TheInstitution ofElectrical Engineers, The Society ofTelegraph-Engineers and Electricians. Founded 1871, Incorporated 1883, OOUNOCIL—1889, ‘President.SmWILLIAMTHOMSON, DOL,LLD,PRIS,(le8B),‘Past-Presidents. Laviaen CLARE MelottO2—1805 femFABDRMOE AAGEY GS,DOsFakir, WH PREECE. Pik tase G80,‘pormor @.&,FostEh, FRE—Isol ‘Mason-Guxmnat C.E.WEBBER, O.B.,ae‘RE,—1883, WiLOUGHBY STH“ Prorsann W-GHYELS ADAMS, Fs81004, ‘O'E SPAGNOLELEL M.a’ GatePoormanDBHUGHES. PS86,EDWARD GRAVES. isaVice-Presidenta. Da. JOHN HOFKINGOM AEA, PRS, WHAT GoOKES RRA bre C8 Puree WB ATHION, PE. TEELANDER BLEMENG. Members. GouinPariseCanvew,RE.|GreseurKarr,Assoc.M.Inst.C.E..LaxtCanrentee, B.A.,B.Sc.|Prof.JounPerry, M.E.,D.Sc.,RE. Gromrrox. ERS. SirJaws Dovatass, F.R.S. SirDavip Saromons, Bart., M.A. 4.A.Fiesana, M.A., D.Sc. Avavstos Srros. Prof. Geo. Forses, M.A., F.R.SS.| Prof. Sivanvs P.Tompson, (L.& BE). B.A, DSc, F.RAS. Capt. SirDovaxas Garrow, K.C.B., D.O.L., LL.D., F.RS. 1 ‘Associates. ‘Srpxer Evensuen. | Gor Caney Fricxzn, Captain A.E,Wrorrester, R.E. OFFICERS. Honorary Auditors. J.Wacstarr Broxpsts (Wagetal Blundell, Biggs, &Co., Chartered Accountants), 12,Delahay Street, Westminster, S.W. Faepenice ©.Danvers, India Office, 8.W. ‘Hon. Treasurer.—Epwarp Graves, Past-President, "Hon, Solicitors. Mesars, Wirson, Buistows, &Canruant, 1,Copthall Buildings, E.C.Trustees ; ‘SirFarpenicx A.Aner, C.B., D.C.L., F.R.S., Past-President. Latiwer Crank, Past-President. Epwarp Graves, Past-President and Hon. Treasurer. Bankers, Messrs. Cooxs, Broputra, &Co.,43,Charing Cross, S.W. ‘Secretary andEditoroftheJournal —F.H.Wrens. ‘Library and Offices ofthe Institution. 4,The Sanctuary, Westminster, 8.W. LOCAL HONORARY SECRETARIES AND TREASURERS. ‘DonJoséAraxicto,2theDirectSpanishTokiHopremntative oftheDirectSpanishTele?spars, rapt Company, Calle deFernando, Vi,¢ SPAIN.ome1,tated” oun Avuurs,CivilEngineer, 4,RuedeNaples, Paris }FRANCE,JMGouuerteoplaertheNetbarandTelegraph,“TheNETHERLANDS.gos B.0,Cracanets,pperintendent ofTelegraphs, Sydney ‘NEW SOUTH WALES, Fannie Dessssn,Engineer feBelgian Telegraph, Brameat BELOTUM. Avot UmSetadent,MontrealDistrictgracaans‘Rorth-Westetn Telegraph Company "of -Canada,Montreal ammene W.3,tuseock,‘uperintchdent ofGovernmenttaentoveWESTERNAUSTRALIA, pelWestrnAtsala*Benangmpinkent |BOBSTA“EngineertotheMexicanandCentraland!MEXICOANDPERU. ‘Soc american Telegraph, Lima, Pera 4BWGticralManagerofthe‘TelephoneComeyofAustin, 2,Wipplingerstrase, 22AUSERO-HUMGARY. (unuat Urwox, Ph.D, GUMGpetineneatofelgrap,Weington|SEWZEALAND, wailWebra, Copntagen }DENMARK. Redan }Siannger,Government TelegraphDepert- VICTORIA. sSafedaaages*weDom o.Ninn, ‘Director-General ofthe Norwegian Tele-- NORWAY,raphe,ChristianiaPoMsfennger, RiverPlateTelegraphCompany, “TabsDEL,URUGUAY anagerRiverPlateTelegraph Company, QUAY ‘MonteVideo feraphComPenyoyREPUBLIC, J.R,Pace, )Indo-European Government: ‘Telegraphs, PERSIA.shine{LeCommandeueF.€uursvonDirector-General ofthetalianseerpiahaux, FS rea DirvctorofEngineeringBureau,Ministryof}JAPAN.Bisbang Bevro D4,Markgrafen Strate, Berlin |ORRMANY, ‘GovernnentTelegraph,Department,carTHECAPEANDSOUTHOusnuzs Toon, CHG,‘Director-General,’ SouthAustralianreesourAUSTRALIA, raph, Adeaite 0.0, Wiss, ‘ecrtaty and Gen, Manager, Commercialt NORTH AMERICA. Cable Company, 1,Browiay, New York Free aedene ‘Superintendent inTasmaniaoftheEastern vw.whet OnesetreBown *YauianUbvernment Telegraphs,Caleutta }NDIA. ov mus Gnstitution ofElectrical Engineers. Founded 1871. Incorporated 1883, Vou. XVIII. 1889. No. 77. AtaSpecial General Meeting ofMembers, held attheInsti- tution ofCivil Engineers, 25, Great George Street, Westminster, on Thursday, January 10th, 1889— Mr, Epwarp Graves, late President, inthe Chair— ‘The notice convening themeeting wasread bytheSecretary. ‘The CuatRMan moved that thefollowing resolution, passed at theSpecial General Meeting ofMembers onthe20th December, 1888, benow confirmed, viz,:— “©That theRegulations oftheSociety ofTelegraph-Engineers “and Electricians, ascontained intheir Memorandum and “Articles ofAssociation, bealtered bysubstituting the name “©The Institution ofElectrical Engineers’ for‘The Society of “«Telegraph-Engineers and Electriciatis,’ and also. bysub- “stitating the word ‘Institution’ forthe word ‘Society, “wherever the same respectively occur intheregulations.” Themotion, having been seconded byProfessor D.E.Hugues, Past-President, wascarried unanimously. ‘VOL, XVIIL. 1 2 CHANGE OF NAMB OF THE SOCIETY. —{Jan. 10th, ‘The OneHundred andEighty-third Ordinary General Meeting of theInstitution washeldattheInstitution ofCivilEngineers, 25,Great George Street, Westminster, onThursday, January 10th, 1889—Mr. Epwanp Graves, Past-President, inthe Chair. ‘TheCuatman: Inorder toshorten theformal proceedings thisevening, wepropose todefer reading theminutes oftheAnnual General Meeting ofDecember 13th. They arenot,particularly novel orinteresting, butarenecessarily rather long, onaccount ofthenumerous votes ofthanks recorded therein. With your per- mission, therefore, wewill defer the reading ofthem until the next meeting. . ‘The names ofnew candidates for election into the Institution ‘were announced and-ordered tobesuspended. The following transfers were announced ashaving been approved bytheCouneil :— From the class ofAssociates tothat ofMembers— JamesThomson Bottomley, |WilliamM.Mordey. FRS. Lieut.-Col. M. T. Sale, George Fuller. CMG. RE, William Geipel. Albion T.Snell. ‘WalterThotnasGoolden, M.A.|HenryUpton. William P.Granville, T.Fred. Toft. Frank King. |John B.Verity. From the class ofStudents tothat ofAssociates— Edgar T.Gideon. Percival 8.Tasker. John Rance, jun. Emest George Tidd. ©.P.Sparks. Walter R,Underhill. Edward Wythe Smith, ‘TheCuainwan: Ihave nowtoannounce that thelastlegal steps have been taken inreference tothechange ofname ofthe Society, andthat infuture the“Institution ofElectrical Engi~ “neers” willtaketheplace ofthe“Society ofTelegraph-Engineers “and Electricians.” Before resigning thechair Ihave onemore agreeable duty to 18803] ‘PRESENTATION OFPREMIUMS. 3 perform, andthatistopresentthepremiums awardedbytheCouncil; butIwillfirstcallupontheSecretarytoreadaletter whichhasbeen received from Professor Silvanus Thompson. ‘The Secretany: Members will recollect that theprincipal premium ofthe Institution was awarded toProfessor Silvanus ‘Thompson forhispaperon“TheInfluence Machine, from1788 “to1888.” Inreply tomyletter announcing that award, he writes asfollows :— 20,Axoxoxs Ganoxss, W., Dacember 16th, 1886. Dus Ma. Wens,—I am inreceipt ofyoursofthe14thinst,announcing (attheCounciloftheBocictyofTelegraph-Engineers andElectricianshave sevardedtomethe“Bociety'sPremium"for1888,IneednotaaythatIgrate- fillyacknowledge thehonourwhichtheCouncilhavedoneineinselectingme tobetherecipientofthisaward.Ishouldvaluehighlysuchanadditionto amyHbeary, and deeply appreciatetheexpreaionoftheCouncil'sgoodwill ‘ThattheoriginalintentionoftheSocietyineatablshing thispremium ‘rutoofferitfortheencouragement oftheeffortsofthejuniormembersof theSocietyis,however,obvious;andI,havingalreadyonceservedonyour(une,viz.,in1696,andhavingjastbeenagainelected,cannotbutfeelthat,thooghtechnically T'maybeentitledtothehonoar,theintorestaoftheBociety ilbebestservediftheCouncilwillreverttotheoriginalintentionofthe anard.1thereforeasktheCotneiltoallowmemortrespectfully todeclinethepremium. Believe me, dear Mr. Webb, ‘Yours very truly, (Gigued) —SULVANUS P.THOMPSON. PHWane,Bag. J ‘The Cuhmaan: Idonotdoubt that you will highly appre- inte, asthe Council do,the consideration shown byProfessor Silvanus Thompson inhisdecision; and Ithink also that the course recommended bythe Premium Committee under the circumstances, and which has been adopted bythe Council, will meetwithyourapproval, viz.,thatProfessor Silvanus Thompson having declined toreceive thepremium awarded tohimforthe reasons given byhim, Mr.A.C. Cockburn, whowastoreceive the second prize, should receive thefirst ;that Mr.E.Stallibrass, who rastoreceive the third prize, should receive the second; and that Mr. E.O.Walker, whose communication, “Earth Currents “inIndia,” was highly commended, should receive the third Prize. ‘ PRESIDENTS ADDRESS, (Jen, 10th, ‘The meeting having signified approval, thefollowing prizes were presented byMr.Graves, viz.:— ToMr.A.C.Cockburn, Member, the“Society's Premium,” value £10, consisting ofachemical balance, Prescott’s “Dynamo-Electricity,” and Prescott’s “Electricity and “Magnetism.” ‘ToMr. Edward Stallibrass, Member, the“Paris Electrical “Exhibition Premium,” value £5,consisting ofSpon’s «Dictionary ofEngineering,” infour volumes. ‘The Cuatnqan: Mr.E.0.Walker, towhom the“Fahie Pre~ “mium” hasbeen awarded, isinIndia, andtheSecretary hasnot yethadtime toascertain inwhat form hewishes toreceive it. ‘The Carman: Myremaining duty istointroduce toyou as, your President SirWilliam Thomson, inwhose favour Ivacate this chair. ‘The Presioenr, SirWilliam Thomson, then took thechair, and delivered thefollowing Inaugural Address :— ETHER, ELECTRICITY, AND PONDERABLE MATTER. Gentlemen,—My firstduty istogive youmywarmest thanks forthegreat honour you have done meinelecting metobethe firstPresident oftheInstitution ofElectrical Engineers. Fourteen years ago,when thethree-year-old Society ofTelegraph Engineers honoured mebyappointing metobetheir President, theSociety numbered 570members; itnow numbers 1,500. Itisgratifying toustothinkhowthatyoungSocietyhasgrown,andhowsuccess- fal ithas been. Butwhile wethink with pleasure ofthegreat increase ofour numbers, that pleasure issaddened bythethought thatagreat many oftheoldmembers aregone, and especially thefirsttwo Presidente—Sir William Siemens and Mr. Frank Scudamore. ‘Thegenial presence ofSirWilliam Siemens isahappy recollec- tion tomany ofus—I think Imight saytonearly allpresent— andthethought that heisnomore with usiscertainly avery +great grief toallwhohave known him, andalosstoScience and 1] BTHER, ELECTRICITY, AND PONDERABLE MATTER, 5 toEngland. Welook back upon Mr.Scudamore’s presidency with {great satisfaction, considering allthat hedidfortheSociety, and hislossisdeeplyfelt.‘ThreeofthePresidents sinceIhadthe honour toserve arealso gone—Mr. Walker, SirJohn Bateman- Champain, and, within thepast year, SirCharles Bright. The firstAtlantic cable of1857 gave methehappiness and privilege ofacquaintanceship withSirWilliamSiemens. Inthecourseofthework SirCharles Bright wasmycolleague. Hewasengineer tothe company, andduring thethirty-three days when wewere outofsight ofland intheever-memorable “Agamemnon” expedi- tionof1858, SirCharles Bright waswith us,fullofvigour and eathusiasm, Tohisvigour and earnestness andenthusiasm, ina greatmeasure wasduetheexistence ofthatcable—the temporarily successful cable of1858—and allthegreat consequences which followed from it,even although that cable itself hadavery short life,havingonlythreemonthsofworkingtime,andastillshorter timeofreally useful, practical work. Still, wemust always feel deeply indebted toSirCharles Bright asapioneer inthat great, work, when other engineers would notlook atit,andthought it wasabsolutely impracticable; andwemustalwayslookuponour latecolleague, lostwithin thelast.year, ashaving done much indeedforthesubjectoftheSocietyofTelegraph Engineers. The Society ofTelegraph Engineers hasgrown notonly in membership, but inthe extent ofitsprovince, from thetime of itsfoundation until now. Itbecame the“SocietyofTelegraph- “Engineers andElectricians” afewyears ago,andnowthemore Properly representative title ofthe “Institution ofElectrical “Engineers” hasbeen adopted. The original name included only telegraphy, but that was not then the only application of electricity toengineering. There was clectro-metallurgy. Heetro.metallurgy and telegraphy were the two, and Ithink theonly two, branches ofpractical science towhich electricity hadthen been applied; butsince that time wehave avast augmentation ofthe field. We have telephony; wehave tlectrie transmission ofpower. The Society ofTelegraph Engineers will recollect SirWilliam Siemens's introduction to itofthatgreatsubjectoftheelectrictransmission ofpoweron 6 PRESIDENTS ADDRESS, (dan, 100, ‘ascale ofpractical usefulness. Ithink wemay safely feelthat totheSociety ofTelegraph Engineers inalarge measure is due that. practical development ofelectricity. We have now power electrically transmitted through factories todrive separate machines by separate motor dynamos; we have electric haulage and electric tramcars; wehave the application of electricity tonaval and military purposes;andlast,notleast, we have the application ofelectricity toelectric lighting. With allthese grand subjects ofapplied practical science for our province, IthinktheInstitution ofElectrical Engineers may feelthat ithasagreat and noble dominion. ‘But wemust notforget that the province ofthe electrical engineer ‘necessarily touches upon thatofthecivilengineer. When fourteen years ago Igave my Inaugural Address, Iendeavoured toimpress upon engineers and architects that architects made agreat mistake innotbeing engineers— innotqualifying themselves asengineers, and doing thework ofengineers—that architects donot dotheir duty totheir clients innotbeing engineers and understanding the engineer- ingoftheir own works, and making engineering science, par- ticularly thedynamics ofengineering, anessential part ofthe training ofanarchitect. Itisnotnecessary tomake any ‘snimadversion, Ithink, upon electrical engineers inthis respect. Electrical engineers know well that they must, before all, be engineers. They must beengineers, and they must learn electricity. Toyoung persons who haveatasteforelectricalmachines (and who that isyoung hasnotatasteforelectricalmachines, andsparks, andfashes, andaurora borealis artificially made, and thesmellofozone—sulphur andphosphorus weusedtocallit!— which isone ofthe pleasing reminiscences ofone’s youth in working with electrical machines?) aword ofadvice may be useful. Every young person who hasafancy forelectricity thinks hewould like tobeanelectrical engineer. They think electrical engineering isallether and electricity. Now Ihave continually toimpress upon anxious fathers and mothers that their boys must condescend tolearn something ofgross ponder- ite] ETHER, ELECTRIOITY, AND PONDERABLE MATTER, ¢ able matter, and that electrical engineering isnot confined toetherandelectricity, butmechanics alsoisanessential part ofthesubject. Itis,Ithink, animportant practical point this —that theelectrical engineer, ortheyouth oraspirant tothat. honourable profession, ought tolearn mathematics and dynamics afterhaving obtained theelements ofagood general education. Heought tolearn mathematics and dynamics well. Then a good deal ofchemistry and regular mechanical and civil engineering should allbelearnt; and electricity learnt besides. Itmaybesaidjuvenilelifeistooshort.Idonotthinkitis Ithink ifthe other subjects arewell learnt, electricity may be learnt inafew months. Iamperfectly sure that iftheyouth isqualified inother departments, the mere addition ofelec- tricity totheeducation ofacompetent engineer willnottake 80 very long atime asmight beimagined, and that themerely educational part ofthework will notbeprotracted unduly by ‘aiding electricity tothebranches learnt ingeneral engineering. Idonotmean tosaythat ifelectrical engineering isthebranch Mdopted, there isnotanendless andprodigious field ofelectricity proper inwhich the worker will learn every day ofhislife, though helives formore years than anyperson present. Iwished justtomakethesefewremarksinthebeginning,becausethey doteem tome ofsome practical importance, and worthy, there- fore,ofbeing putinthefront oftheAddress Ihave tooffer as firtPresident oftheInstitution ofElectrical Engineers. Now these remarks have suggested tomeasubject forsuch a ‘<ientific exposition asIcouldpossiblyattempttogiveinthisopen- ingAddress—* Ether, Electricity, andPonderable Matter.” The demand forsomething likeamechanical explanation ofelectrical Phenomena isnotnew, butitisgrowing inintensity every year. TheproceedingsofrecentmeetingsoftheBritishAssociation— andespecially ofthe last meeting ofthe British Association— iilustrate thegrowing desire toknow something below thesur- face; toknow something oftheinterna} relations connected with thewonderful manifestations offorce and energy which areput before usintheaction ofthemagnet, intheworking even of common electrical machine, andinelectro-magnetic phenomena 8 PRESIDENT’S ADDRESS. (Jen, 108, The Addresses ofPast-Presidents ofthe parent Societies of Telegraph Engineers, and Telegraph-Engineers and Electricians illustrate also thegrowing desire toknow something ofthe molecular theory orthe dynamical theory ofelectricity and magnetism. Mr.Preece, jnhisAddress of1880, pointed out how Maxwell had shown thevelocity oflight toberelated to electricity insuch away that wecanscarcely doubt but that the propagation ofelectro-magnetic disturbance through space, which wehave every reason tobelieve does exist—which, infact, from known laws wemay saycertainly does exist—is effected with a velocity equal tothat oflight, and that the propagation of electrical disturbance andoflightmayperhapsbeidentical. In support ofthese remarks, Mr.Preece alluded tothedisturbances atthesun’s surface and thesimultaneous magnetic disturbances which hadbeen observed inthetelegraphs andinother operations ofanelectro-magnetic character onthesurface oftheearth. In1883 Mr,Willoughby Smith described shortly some experi- ments which Iconsider tobevery beautiful andvery instructive, with which hewasthen engaged. Those experiments demon- strated and illustrated thescreening effect ofsheets ofdifferent kinds ofmetal upon electro-magnetic andelectrostatic inductions. Electric induction, simply, wemay say, because webegin tofail todistinguish between electrostatic induction and electro-mag- netic induction, InWilloughby Smith's subsequent work he gave anexceedingly beautiful setofexperimental investigations ofthe sereening effect oflead, copper, and iron,ofwhich,‘asI have said,aslightsketchwasgiveninhisPresidential Address. Alittle earlier thesubject was mathematically worked outwith ‘great power byseveral mathematicians, butperhaps most notably byHorace Lamb. Ifeel italmost. invidious tomention names when there aresomany thorough workers who touched upon the same subject very closely. Charles Niven,* almost simultaneously with Lamb,t went through very much thesame kind ofwork— +OntheInduction ofElectric Currents inInfinite Plates and Spherical “Bhellsy” Phi, Trans, Roy. Soc, 1881, p.807(read Jan, 29th, 1880)."OnElectricalMotionsinaSphericalConductor,” Phil.Trans,Roy.Sey1865, p,619(road April Sth, 1855). 180] ETHER, ELECTRICITY, AND PONDERABLE MATTER, 9 infact, obtained thesame solutions ofsome important and interesting problems regarding electric currents inspherical conductors. Ispecially mention Lamb's name because the subject ofscreening ismore particularly developed inhismathe- matical paper. Inthe memorable Presidential Address ofProfessor Hughes, another allied branch ofelectro-magnetic induction was very admirably illustrated byexperiments which are now more or less familiar tousall,but which have been ofanimmensely suggestive and stimulating character, both tomathematicians and toexperimental workers. The very criticisms bymathe- maticians upon some oftheexperiments and modes ofstatement byProfessor Hughes have, with Professor Hughes's own ex- periments, given avery large body ofelectrie knowledge and electro-magnetic knowledge which, without such stimulus and such mathematical and experimental scrutiny asithas ledto, might have been wanting formany ayear. One oftheearliest problems inwhich electric induction had tobeconsidered was that ofthe submarine telegraph, The subject ofinduction intelegraph wires presented itself ina peculiarly perplexing way tothe first workers inthat depart- ment. There was the general knowledge ofelectro-magnetic indaction between two wires, which had been worked outby Henry and Faraday inavery full manner. That was theonly kind ofinduction which was thought ofbysome ofthepioneers ofsubmarine telegraphy. Another kind ofinduction wasmore thought ofbyothers. ‘That wastheelectrostatic induction due totheLeyden-jar charge oftheinsulated wire. Faraday, inthis department, asintheother department, wastheorigin ofnearly allthatwenowknow.Heexplainedinaverybeautifuland clear way the electrostatic charge ofthesubmarine cable, and showed how theelectricity conducted through thecable from one ‘end, togive what ofpotential isnecessary inthemiddle orthe other end ofthecable—in themiddle ofthe cable foranymode ofworking, attheotherendofthecableformodesofworkinginwhichtheotherendisinsulated—gave risetotheLeyden-jar charge. Hepointed out(without going intoanyofthese details, 10 PRESIDENT'S ADDRESS, (Jan, 10tb, however) the doctrine ofthe conduction ofelectricity through thewire tosupply theLeyden-jar charge which thewire must have, inthecourse ofworking, inorder toberaised tothe difference ofpotential from theearth required tocause thesignal current topass through it. Cromwell Varley made very im- portant advances inthat direction. Atthe meeting ofthe British Association in1854, atLiverpool, hebrought forward some important developments ofFaraday’s doctrine. And then came onthegreat Atlantic Cable question, Ialways remember how that question came upon me. IseeinProfessor Stokes's Presence with usthis evening areminder ofthe circumstances. Iwas hurriedly leaving themeeting oftheBritish Association, when asonofSirWilliam Hamilton, ofDublin, was introduced tomewith anelectrical question. Iwas obliged torun away togettoasteamer bywhich Iwasbound toleave forGlasgow, andIintroduced him toProfessor Stokes, who took upthesubject with apower which isinevitable when ascientific question issub- mitted tohim. Hewrote tomeonthesubject soon after that . time, and some correspondence between uspassed, theresult of which wasthat. little mathematical theory wasworked out, which constituted, infact, thebasis ofthetheory oftheworking ofthe submarine cable, Inthat theory, electro-magnetic induction was not taken into account atall. The leaving itout ofaccount wasjustified bythe speed ofsignalling which the circum- stances ofacable exceeding 200 or300 miles inlength dictated. Foracable more than 200 or300 miles long the speed ofworking was essentially limited bythese electrostatic considerations—limited somuch that the electro-magnetic induction certainly could have nosensible effect. But the possible speed ofworking inacable of20miles or50miles, oreven 100miles, wassogreat that inshort lengths like that the electro-magnetic induction might well come into play. Iworked outthesubject partially myself. Ifound itnecessary todosotosatisfy myself that the doctrine upon which the Atlantic Cable project, then growing up,wasultimately founded, wasthoroughly trustworthy. Ifound itnecessary toinvestigate the question ofelectro-magnetic induction. This question was 1800) ETHER, ELECTRICITY, AND PONDERABLE MATTER, 1 further forced upon mebycommunications that Ihad with nyfriends, Lewis Gordon, and the two brothers, Charles Wm. Siemens and Werner Siemens, with reference toMediterranean cables. Ttwas imagined that electro-magnetic induction alone asoperative—that embarrassment inworking through the submarine cable wasdue toelectro-magnetic induction alone, Onitsbeing demonstrated that electro-magnetic induction could have no sensible effect onthe signalling through proposed Mediterranean cables, the proposal tohave two thin wires close together inorder toobviate electro-magnetic induction was given up, Experiments inGermany had shown considerable electro-magnetic induction onshort lengths of cable,andithadbeensupposed thattherewouldbeembarrass-mentfromthiscauseintheworkingofthe cables, which would bediminished byusing wires very close together. But this diminution ofthe electro-magnetic inductive influence would produce acorresponding increase inthe electrostatic inductive influence; and when itwas pointed out that the electro- magnetic inductive influence would beabsolutely imperceptible atthehighest speeds ofworking ofthe proposed cables, and that itwould bethe electrostatic induction which would limit thespeed, theidea ofmaking them ofthin twin wires—two pairs ofwires close together inmetallic circuit—was given up,andthe Present type ofsubmarine cables was adopted. But now itis veryinteresting toustofindthatoldquestionrevived. Ihad myself laid itaside insome comer ofmymind and insome slightcornersofmynote-books forfortyyears.Withinthelastfortydays Ihave really worked itouttothe uttermost, merely formyown satisfaction. But inthe meantime ithad been worked outinavery complete manner byMr.Oliver Heaviside; andMr.Heaviside haspointed outand accentuated thisresult ofhismathematical theory—that electro-magnetic induction isapositive benefit: ithelps tocarry thecurrent. Itisthe ‘ame kind ofbenefit that mass istoabody shoved along against 4viscous resistance. Suppose, forinstance, you hadarailway carriage travelling through aviscous fluid, Take aboat on heels,sothatthemoremassivebodywillnotsinkdeeperin rr PRESIDENTS ADDRESS. (Jan,1088, thefluid than thelessmassive body. Take aboat onwheels ina viscous fluid. Wewill shove offtwoboats with acertain velocity —the boats ofthesame shape; butletoneofthem beloaded to tentimes themass oftheother: itwilltake greater force togive ititsimpulse, but itwill gofurther. That isMr. Oliver Heaviside's doctrine about electro-magnetic induction. Itrequires more electric force toproduce acertain amount ofcurrent, but thecurrent goes further. Itisavery crude way inwhich Iam putting it. Iamnot doing justice, ofcourse, Iknow,tohis statement inoneshort sentence. ‘The whole question istreated inthe most complete mathematical way. The effect ofelectro- magnetic induction and electrostatic induction taken together (and they cannot beseparated) isfully worked out. One thing that wasknown ofoldismade apoint ofinMr.Heaviside’s treat- ment ofthecable problem—that is,thebeneficial effect ofleakage inrespect toclearness ofsignals. (ldtelegraphists remember that. They always used tosaythree orfour good leaks ina cable, ifthey would butkindly remain constant,andnotintroduce extra trouble byearth currents, would make thesignalling more distinct. That used tobewell known, and thereason used tobe fairly well known; mathematical theory had pointed itout. Now ‘Mr.Oliver Heaviside hastaken upthatsubject again andincluded itinhiswork. Hehasincluded italong with electro-magnetic induction, and this point hehas particularly accentuated. Itis 4practicalpointofimportance thatthequestionofclearness of signals isnotsimply oreven very importantly this—How much is the current attenuated atthe remote end ofthe cable? how much istheamplitude oftheelectric current inonemode ofworking, orofthevariation ofelectric potential inanother mode ofwork- ing, altered intransmission through athousand miles ortwo thousand miles ofcable? Acertain range isgiven atthesending end;andwhatistherangeatthereceiving end?Thatisan important question, but itisnotthe most important question with reference toclearness ofsignalling; infact, wemight almost sayitisnotanimportant question atall, Itisnotthesmallness ofthesignals atthereceiving end that isthereal difficulty ina submarine cable just now atall;itistherunning ofonesignal 19] ETHER, ELECTRICITY, AND PONDERABLE MATTER, 13, intoanother;itisthewantofcorrespondingly definitedistinctions ofsingle signals orofagroup ofsignals atthereceiving endand atthesending end. Now inthe mathematical theory there aretwothings tobe considered inrespect tothedistortion (asHeaviside called it) ofthesignalsinpassingthroughthecable.Onethingtobe considered isthe retardation ofphase; another isthe diminu- tion ofamplitude, Ifthe retardation ofphase were the same foralternating currents ofallperiods, then this retardation of thephase would beofnoconsequence whatever—it could not, diminish the distinctness atall. Again, ifthe diminution of theamplitude were precisely inthe same proportion foralter~ mating currents ofallperiods, then when wecome tomake non-periodie signals weshould find that the signals would be transmitted with perfect sharpness. Infact, ifwecompare thetransmission ofelectric signals through wire with the transmission ofsound through air,wehave inthe course of transmission ofsound through airgreat attenuation bydistance— inversely asthe square ofthedistance, infact—but thesame for allnotes;and,again,retardationofphasedependinguponthe velocity ofthe sound, the same forallnotes. The result is that speaking, and musical performances, and signals ofall kinds inair, lose none oftheir clearness bydistance. Itisjust, aquestion whether atthe very greatest distance atwhich a sound can beheard there isany want ofclearness due to different attenuations ofthe different notes orofthe different elements forming the compound sound, ortodifference of rtardations ofphase. Imust notoccupy you toolong with this subject, but itisone oflarge practical importance. Heaviside points out that electro-magnetic induction causes a lessgreat difference intheattenuation ofdifferent periods than there iswithout it;and that electro-magnetic induction (as reknew forty years ago) tends toreduce the retardation of phase tothe same foralldifferent notes—that is,totheretar- dation equal towhat would depend onavelocity notvery different from thevelocity oflight—if thesignals have but taffcient frequency. That velocity was then and isstill known “ PRESIDENTS ADDRESS, (an. 1008, asthevelocity which istheconductance inelectrostatic measure, ‘and the resistance inelectro-magnetic measure ofone and the same conductor. But itsrelationship tothe velocity oflight ‘wasbrought outina manner byMaxwell tomake itreally a part oftheory which itnever wasbefore. Maxwell pointed out itsapplication tothepossible orprobable explanation ofelectric effects bytheinfluence ofamedium, and showed that that medium—the medium whose motions constitute light—must beether. Maxwell's “electro-magnetic theory oflight” marks fastage ofenormous importance inelectro-magnetic doctrine, and Icannot doubt but that inelectro-magnetic practice we shall derive great benefit from apursuing ofthe theoretical ideas suggested bysuch considerations. Infact, Heaviside’s way oflooking atthe submarine cable problem isjust one instance ofhow thehighest mathematical power ofworking and ofjudging astophysical applications, helps onthedoctrine, and directs itinto apractical channel, The telephone—one ofthe added subjects oftheInstitution ofwhich wearemembers—illustrates very splendidly these developments ofthetheory ofthetransmission ofsignals through thesubmarine cable, The telephonic signals have, infact, sufti- cient frequency tomake electro-magnetic induction very sensibly influential. ‘The frequencies intelephony correspond tofrom 250 periods per second, uptotwo, orthree, orfour times that; being thefrequencies involved inspeaking inthehuman voice —tenor and soprano—and inthe quality ofthe voiceasaffected bythe overtones. Isay frequencies offrom 250 persecond to1,000 or1,500 periods per second, areconcerned inthe fundamental notes, and inthecharacterising over-tones, ofthe sounds. transmitted bythe telephone. Now there seems no doubt butthat theclearness ofthe telephone through great distances istoalarge degree due tothecircumstance which Heaviside haspointed out—that wehave much less ofdifference ofattenuation and difference ofphasal retardation, fordifferent notes, with theactual frequencies ofthenotes insounds trans- mitted through thetelephone wire, and thepractical dimensions ‘ofthe telephone wire, than weshould have without electro- magnetic induction. 11] BYHER, ELECTRICITY, AND PONDERABLE MATTER, 15 Teannot speak onthis subject without just touching upon a question which Idonotatallpropose toenter upon to-night, andthat is,the relative efficiencies ofiron and copper astele- phonic conductors, Information given tomebyMr.Bennett, theengineer ofthe National Telephone Company inScotland, leaves nodoubt whatever but that there isavery considerably grater loss ofclearness inspeaking through aniron wire—such, frinstance, asthe first metallic circuit iron wire ofthe Post Ofice between Glasgow andEdinburgh—than there isinspeak- iagthrough the present copper wire circuit ofthe company tetween thesame stations. Ishall saynothing more ofthisjust ow. Itisanexceedingly difficult and complicated subject. BatHeaviside’s mathematical work, and Lord Rayleigh’s experi- entalinvestigations onthesusceptibility ofirontoverysmall lilrences ofmagnetic force—differences ofmagnetic force super- imposed oneven apowerful magnetic force oralarge residual magnetism—put us,Ithink,inaveryhopefulcondition.‘The subject isingood hands, mathematical andpractical, andIthink tefore many months pass over the Institution ofElectrical Hagineers,wemayhaveanabsolutelyclegrunderstanding of ‘eephony through iron wire ascompared with telephony through copper wire. Leaving allquestions ofsubmarine telegraphy, and oftele- mphy ortelephony, inwhich—whether from thegreatness of ‘hedistancethrough whichthecommunications aremade,ormallness ofdistance between insulated conductor and sheath, ‘xbetweenthetwinwireswheninsulated metalliccircuitisused—tteeffectsofelectrostatic capacitygiverisetosensibledifferencecistrength ofcurrentindifferent partsalongthelengthofthe ‘mnductor—I wish tocall your attention tothedifferences of ameat-density across different parts ofthe cross section, which wxproduced when alternate currents aresent through awire. (msider acopper wire, and 8copper tube surrounding itfor mum, Orconsider what is,after all,oneofthevery simplest ‘es—two parallel copper wires. Ifthedistance between them alarge multiple ofthediameter ofeach, asisthecase in Radical telegraphy andtelephony, theproblem isthesameas 6 PRESIDENTS ADDRESS. (Jan,100, theproblem ofasinglecopperwireinthecentreofacylindrical tubeofinfinitely conductive metal,andofradiusequaltothe distance between the wires. The distribution ofcurrent within thesolid conductor depends only ontheperiod ofthealternations, and onthe diameter and the specific resistance ofthe metal; and isquite independent ofthesurroundings, provided only they besymmetrical allround, orprovided, ifthecase bethat oftwo parallel wires, thedistance between the two wires bealarge multiple ofthediameter ofeach, sothat the current ineach is notsensibly disturbed, bytheinfluence ofthe other, from being arranged inco-axial cylindric layers ofequal current-density. Forthisproblem themathematical theory gives usaremarkably interesting andvery useful practical result;andIreally,inpro- posing tospeak upon such avery abstruse and uninteresting subjectas“Electricity,Ether,andPonderableMatter,”wishtotry togive onelittle piece ofpractical information to-night. Itwill benoinformation tosome, but itmay beinformation toothers. ‘The solution, expressed inaformula, and atable ofnumerical results calculated from it,Ihope, will appear complete inthe report ofthis Address. Ithink Imight give you just now two orthree ofthenumbers that arerather interesting. Take 80 periodspersecondasthefrequency —thatisaboutwhatisadoptedinthealternate-current systemofdistribution foreleetrie light;atalleventsinonegreatsystemIknow—the Grosvenor Gallery installation—that isthefrequency oftheperiod;andI believe itispretty much thesame generally. Letustake, then, ‘asanexample, the80periods persecond. First, consider coppet wire of1centimetre diameter: the omic effective resistance is greater than forsteady current through thesame wire, butnotas much asxqper cent. greater. Take, now, copper wire ofIt centimetres diameter: theohmic effective resistanceis24percent. greater than theresistance forsteady current. Next, take copper wire of2centimetres diameter: the ohmic resistance is8per cent. more forthealternating current than forthesteady current. Inround copper rodof4centimetres diameter, theohmic resist- ance is68percent. more forthe80periods persecond alternate currents than forsteady currents. Inround copper barof10 8.) ETHER, ELECTRICITY, AND PONDERABLE MATTER. 17 centimetres diameter, the ohmic resistance is3-8 times what it would beforthe steady current. Inasolid copper cylinder of 100centimetres diameter theohmicresistance is35timesgreater thanforsteady currents. From 10centimetres diameter upwards theohmic effective conductance—that is,the reciprocal ofthe ohmic effective resistance—increases scarcely more than asthe diameter simply, and not asthe square ofthediameter. The conductance forsteady currents isasthesquare ofthediameter allthrough. The effective conductance foralternate currents followsalawwhichcanonlybeexpressed byaidofFourier-Bessel fanctions tillwegettovery great diameters. When wegetto togreatadiameter thattheshell,orouterportion,ofthewire intowhich thecurrent ispractically confined, ismoderate orsmall inproportion tothe diameter ofthe wire, then, fordiameters exceeding that, youcanallseeperfectly without calculation, that theconductance isinsimple proportion tothecircumference, and therefore insimple proportion tothe diameter. This very imperfect. explanation oftheresults may give some idea which, think, isofrather aninteresting and important kind, butthe figures willspeak forthemselves. With quadruple frequency, the samefiguresapplytowiresofhalfdiameter. ‘Therewegetthe telephone problem. Four times 80is320, which isamong the frequencies fortelephonic notes;andforthe320frequency,take thefigures Ihave given, butwith half thelinear magnitudes. ‘Thus, for instance, forcopper wire of1centimetre diameter, transmitting musical notes of320periods persecond, theohmic resistance is8percent. greater than theresistance forsteady carrents;foracopperwire2centimetres diameter, andfre- queney ofmusical note 320persecond, theobmic resistance is 68percent.greaterthantheresistance forsteadycurrents;and on, Another important development from this theory is,that there ismuch less importance intheconductivity ofthemetal fortelephonic work through extreme distances, than forordinary dectric work. The formula and figures showing thekind of relations between obmic effective resistances fordifferent speeds are,a8wehaveseen,alittlecomplicated; butIwillonlysay ‘You xvm. 2 8 PRESIDENTS ADDRESS, (Jen. 108, that ultimately itisthe square root oftheresistance that we have todeal with instead ofthe simple resistance. Quadruple resistance isonly twice asbad, speaking roughly, when the frequency issogreat astocause the effect ofthe ohmic resistance tobeverymuchgreaterthantheresistance forsteady currents, The moral ofthisis,notthat youmay choose wire of badconductivity ;onthecontrary, takethebestconductivity you canget,whether fortelephones orforelectric light conductors, butshape theconductors sothat theohmic resistance shall not ‘betoo much augmented bythe unequal distribution ofthe current. Inrespect toelectro-magnetic theory. Wehave avery fine analogy with viscous fluid motion, which hasbeen obvious, more orless, from thetime theknown laws ofelectro-magnetic induc- tion were put into formule inthe beautiful manner inwhich Maxwell putthem,—we have avery fineanalogy, Isay,with the diffusion oflaminar motion into aviscous fluid, and itsanalogue inthe diffusion ofheat byconduction through asolid, first pointed outbyProfessor Stokes. ‘The actions concerned inthe distribution ofalternating electric current through aconductor such ascopper, and the distribution ofthe motion ofwater in ‘aviscous fluid disturbed byperiodical tangential motions ofits surface, follow identically thesame law. Mr.Heaviside, referring tothis, haswell said that this analogy invery useful, because we canseethemotions inaviscous fluid, and understand them, and picture them toourminds, while itismuch more difficult to fancy wesee the distribution ofelectric current inawire. Takenowdefinitively, thisanalogy forthedistribution ofelectric current ina round copper wire through which alternate currents ofelectricity aresent.Takeaviscousfluidinatube,inplaceof theconductor: move the tube toand frowith regular alternating motion—a simple harmonic motion, Inorder that wemay fulfil atallapproximately what Iamspeaking of,the length ofthetube must bevery great incomparison with the diameter, andtheplace inwhich weconsider themotion ofthe fluid must beatadistance ofmany diameters from theends, which wemay suppose tobeclosed byfrictionless pistons, W0) ETHER, ELECTRICITY, AND PONDERABLE MATTER, 19 limiting thefluid atitstwoends. Inthefirst place, ifthefluid vwerenotviscous—if itwereperfectly liquid—you mightmovethetubetoandfro,butthefluidinsideofitwouldremain atrest. Water, however, would move ;oilwould move;themoreviscous thefluid is,themore liable itwould betoexperience motion inthat way. Now there isaperfect analogy between the altemating motion ofthefluid transmitted inwards from the surface, and the distribution ofthe electric current inawire through which theeffect ofthealternating current machine is being conveyed. Another very interesting analogy inwhich exactly thesame lwholds, isthechange oftemperature ofaconducting solid, ‘duetovariations ofexternal temperature. Imagine acolumn ofrock orstone ormetal, and, instead ofmoving ourtube to tndfrolongitudinally, letthe atmosphere surrounding our column beperiodically varied intemperature: the law ofthe inwards progress ofchanges oftemperature, the law ofthe maximums and minimums and zeros oftemperature, isidentical with the law ofthe corresponding features ofelectric currents andoffluid motion. Ineach case wehave @propagation inwards, with diminishing amplitude. Ineach case the rate of diminution ofamplitude corresponds totheretardation ofphase ‘sccording toexactly thesame law. Ineed notattempt atthis time tostate the law—mathematicians know itperfectly well. Nowtake another case. Here the thermal analogy absolutely failsus,butthefluid motion analogy stillholds. Take atube of finidand give itanalternating motion—a periodically varying ‘motion round itsaxiswhich gives atangential drag tothefluid intheinside. Now you canallseethat theinwards penetration ofthetangential drag,ifthealternations ofthemotionbeveryquick,willfollowthesamelawforthetoandfromotionofthecylinderandfortherotatory motionofthecylinder.Thequestion isthis,Doesthevariation penetrate sensibly toalargedistance ia,from the outside ornot? If,forexample, itpenetrates in alytheone-hundredth ofthe radius, then itisobvious that we ‘tall have sensibly thesame lawofpenetration inwards forthe disturbance, whether forthecase oftherotatory motion ofthe 10 PRESIDENTS ADDRESS, (Gan,108, cylinder round itsaxis, oroflongitudinally toandfromotion. Exactly thesame thing holds with reference toelectro-magnetic induction, The one case ofelectro-magnetic induction that I mentioned firstisthemost important, being thetelegraph and telephone case;butanotherveryinterestingcase,andnotatall without practical importance, isthe penetration ofinduced currents into acopper orother metallic core within asolenoid. ‘Takeacommon helixorsolenoid: sendanalternating current through itscoil—you know what itdoes. It.produces magnetic force,withlinesofforceparallel totheaxis,intheinterior ofthe solenoid. But alternating magnetic force, through thecopper, induces electric currents incircles perpendicular tothedirection oftheforce.Thuswehavecurrentsinduced,asyouallverywellknow,inasolidmetalcoreofasolenoid,Ametalliccoreother than iron, isasubject forinvestigation ofanexceedingly easy kind. TheFourier-Bessel functions come inhere justasthey do inthe other cases inwhich we are concerned with circular cylinders. Ifwehave, instead ofcopper, aniron core, wemust take into account itsinductive magnetisation. ‘This presents no ‘mathematical difficulty ifwesuppose themagnetic susceptibility constant; and thesame lawofamplitudes andphasal retardation holds asforcopper orother non-magnetic metal. Thedifficulties, both experimental and mathematical, totake into account, are the enormous differences oftheinductive quality ofiron with different degrees ofmagnetisation, and with reversals ofmag- netisation ;andthegreatcomplications oftheinductive effects onaccount ofthe“magnetic friction” intheiron, introduce corresponding complications inthe theory ofthe induced ourrents, and they arecomplications ofakind that arevery formidable. ‘Now Icanonly just goontosaytwoorthree words about an extension ofthat viscous fluid theory that allows ustotake into account allthat goes onboth inairandinmetal, andindifferent metals, whether incontact with one another orseparated byair. For illustration, consider our twosimple cases—parallel wires with alternating currents through them, andthecylinder rotated with aperiodic motion ofrotation alternately inopposite direo- lie} ETHER, ELECTRICITY, AND FONDERABLE MATTER. 21 tion, The analogy issimply this: Torepresent different metals, densities offluid insimple proportion totheelectric conductivities must betaken; theviscosity must bethesame inall. ‘The representative ofaninsulator inthis analogyisa massless fluid. By“massless” Imean devoid ofinertia— perhaps Iought tosayan“‘inertialess” fluid, because people attach other ideas to“mass” sometimes than “inertia,” but in thestrictest,dynamical language ‘“mass”istakenasthemeasure ofinertia. Aninertialese viscous fluid must take theplace ofair orothernon-conductor ;aviscousfluidofacertaindensity,but thesamedegreofviscosity,musttaketheplaceoflead.Afluid oftwelve times thedensity oflead would take theplaceofcopper, theconductivity ofcopper being, say, twelve times theconduc- tivity oflead. ‘Time does not allow metopursue the subject further in theway ofillustration atpresent, butImust return tothe secondcaselateron,becauseIamgoingtospeakofironand tation, Now, with reference totheelectrostatic effect, thehopeless— Tmust not say“hopeless:” that istoolarge aword; weare ever without hope inscience—I wasgoing touseanother word, “despair ”—well, Ifeelitdesperately dificult;Ifeeltheprob- ability ofmyseeing thesolution ofitishopeless. Tomerely introduce into the ‘analogy electrostatic effect isvery simple. Simply imagine aninterface between thetwo fluids, andgive it suchstiffness against change ofshape asisrequired tocause itto falfl the conditions which electrostatic knowledge and our knowledge ofthelawsofelectric andelectro-magnetic influence, dictate tous. Isay,putinattheinterface therequisite normal force,andyoucanextend theanalogy toinclude thecomplete Problem ofthesubmarine cable, inwhich electro-magnetic and ectrostatic induction are both taken into account. But itis onlybyputting in,and inanarbitrary manner, aforce atthe rurface tofalfil therequisite conditions, that wecancomplete theanalogy. The analogy Ihave just sketched cannot beconsidered as being inanyrespect aphysical analogy. Inittheanalogue to 2 RESIDENT'S ADDRESS. (Jan, 1004, electric current isnot velocity ofthe fluid; itisnot the molecular rotation ofthe fluid; itisaquality derived from thesimple motion ofthefluid, mathematically bytheoperation knownasthe“laplacian” operation—the laplacianof(1u,0,10)is theelectriccurrent.Anotherwayofputtingitistherotation ofthe rotation of(u,v,t0), orthespin-flow ofthespin-flow of (wet). ‘The word “curl” was introduced byClifford and Maxwell, and has, Igrieve tosay, been adopted byMr. Oliver Heaviside. Itisthecurlofthecurl.Iobject to“curl,” because “rotation” or“spin” isgood, and “curl” isbad; I object toitonthat ground, but not only onthat ground. I ‘object toitalso because itisconnected with akind ofmathe- ‘matical symbolism which seems tomenot desirable and not instructive, and, above all,notconvenient forany practical use inmathematics. This “laplacian” istoodifficult asubject to explain, butIwould byacase trytoillustrate it. Take inthe viscous fluid analogue what corresponds tothe steady current inawire. Think ofthetube with viscous fluid and pistons as before. Atoneendofthetube press «piston inwith auniform motion, continued long enough tocause the fluid throughout thetubetocometoastateofsteadymotion, Intheneighbour- hood ofthepiston themotion isdisturbed bytherigidity of thepiston; butgotoadistance oftenortwenty diameters from thepiston, and themotion ofthefluid takes aperfectly regular character. [Illustrating onblackboard asFig. 1). Suppose that tobetheinner surface ofthetube. This dotted line represents aportion ofthe liquid which atone time is plane, Alittle later, while the fluid incontact with the containing surface remains unmoved, inthedoctrine ofviscous fluid, asgiven byStokes—there isabsolutely noslipatthe containing surface—this portion ofthe fluid which was plane becomes the paraboloid ofrevolution, which you seeshown inaxial section inthediagram. The velocity ofthefluid is nothing atthe bounding surface, anditisamaximum atthe centre, Well, wehave two functions derivable from the con sideration ofthatdistribution ofvelocity. Thefirstistherate ofshearing ofthefluid; thesecond istherate ofchange per 1660] ETHER, ELEOTRIOMY, AND PONDERABLE MATTER, 23 unit-change ofdistance from theaxis oftherate ofshearing.” The rate ofshearing represented graphically isequal tothe tangent ofthe inclination (TPN) ofthis curve tothe transverse surface, thetangent oftheinclination ofthecurve being the angle which isrepresented bytheletter i. Nowtherateofchangefrompointtopointoftherateofshearingistheanalogue tothestrength ofthecurrent, andthatisuniform, Sointhis analogy ofaviscous fluid forced through atube, wehavenotthefluidvelocityequaltotheelectriccurrent,butsome-thingelse,quiteintelligible; andthereasonforit,inouranalogy, isclear enough. Butthere issomething interesting, perhaps, in this idea—thatwehaveasuper-subtle mathematical definitionof electriccurrent whichisnotfluidvelocity. Well,now,perhaps Ne 4 | i s Fro 1. someone will say, “Had not webetter getananalogy inwhich a «fluid velocity isequaltothevelocityoftheelectriclow?” Well, *Suppose the parabolainthedrawingtorepresenttheAidwhichlay along the dotted line aanit oftime earlier, ‘The distance ofPfrom the ote ine Snequal tothevelocity oftheMuid ()atthedistance (°)ofPtromtheaxin,Wehavew=(o*-r8),whereadenotestheradiosof thetabeandtefdveloity slongitaxin,Wohave=Sth= deg, whiehistheratofshearing;and=!=2whichinoaropremtative ofthe electre-carmnt density, ‘The whole strength oftheelectric current bavee. “ PRESIDENTS ADDRESS, (Jan. 10%, Idonotsaywhetherwehadbetterdosoorhadbetternot,butwedonot,otherwise thaninthewayIhavedefined,getthework- inganalogy;andthereisanadvantage inthisanalogy. Itgives us@motion ofwhich therotation isthemagnetic force. Mr. Heaviside ratheroverlooks that.HeobjectstoMaxwell's vector- potential. Idonotagree with him inhisobjection. Ifhehad confined the objection tothe “vector” and the “potential” (‘<veotor” wholly bad, and “potential” badinconnection with the present subject), Iwould heartily have agreed with him, because Ithink itisanunhappily chosen name; butMaxwell’s useofthe thing (which heunhappily calls “‘vector-potential”) ismost happy and most instructive, asitseems tome, Maxwell does not translate this into realities ofmotion, butheputs down inhis formule, asthefoundation from which onestepleads tomagnetic force and thenext step toelectric current, something which, translated into realities ofmotion, gives usamotion ofwhich the rotation isthemagnetic force; andhere itseems tomethat if weareever tohave areal theory itmust. befounded upon this view. The hand oftheclock warms metime isgoing onso rapidly that wemust leave this analogy absolutely unfinished. Perhaps itiswelltobeobliged toleave itnow, because themorewelookatitthelesswelikeit,ifwewishtoseeandtolikea true mechanical explanation ofelectro-magnetism. The work is done inthewrong place. Inthedense liquid, work isdone and heatgenerated inproportion tothesquareoftherateofshearing. Inthe electric analogue, the work isdone and heat generated uniformly throughout theconductor. Wehave work done and heat generated intheviscous massless fluid taking theplace of thedielectric inourfluid analogue. Wemust discredit that, absolutely;butthereasonforjudgingtheanalogyworthsomuch notice aseven ithashadto-night, isthat itisaperfect mathe- ‘matical working analogy,andanexceedingly usefulandinstructive kindofanalogy, andavery potent onetohelp usinguessingout,andinthinking out,andestimating resultsinpractical problems ofelectro-magnetic induction indynamos and in alternate-current machines, and intelephones and inelectric instruments ofgreat varieties ofshape andmutual relations, 16a] ETHER, ELECTRICITY, AND PONDERABLE MATTER, 25 Butnow there isanother line ofthought inconnection with thissubject, and that istheelastic solid idea, Willyouallow metoread avery short statement which was published inthe Cambridge and Dublin Mathematical Journal fortheyear 1847? ItindatedGlasgow University, November 28th,1846.Itwas written after Ihad been twenty-eight days atwork inmy Professorship, and itisasfolloiws:—“Mr. Faraday, inthe 11th “seriesofhis‘Experimental ResearchesonElectricity,’ hasset “forth atheory ofelectrostatical induction which suggests the “idea that there may beaproblem inthetheory ofelastic solids “corresponding toevery problem connected with thedistribution “ofelectricity onconductors orwiththeforcesofattraction and “repulsion exercised byelectrified bodies. The clue toasimilar “representation ofmagnetic and galvanic forces isafforded by “Mr, Faraday’s recent discovery oftheaffection with reference to “polarised light, oftransparent solids subjected tomagnetic or “electro-magnetic forces.Ihavethusbeenledtofindthree “distinct particular solutions oftheequations ofequilibrium of “anclasticsolid,ofwhiclioneexpresses astateofdistortion, such “that the absolute displacement ofaparticle inanypart ofthe “wlid represents theresultant attraction atthispoint produced “byanelectrified body.Anothergivesastateofthesolidin“which each elementhasacertainresultantangulardisplacement, “representing inmagnitude anddirection theforce atthis point “produced byamagnetic body; and the third represents ina “similar manner theforces produced byanyportion ofagalvanic “wire; thedirections oftheforce inthelatter cases being given “by the axes ofthe resultant rotations impressed upon the “elements ofthe solid.” Then come the mathematics, inthree Pages,andthencomesthelastsentence: “Ishouldexceedmy “present limits were Itoenter into aspecial examination ofthe “states ofasolid body representing various problems inelec- “tricity, magnetism, andgalvanism, which must, therefore, be “reserved forafature paper.” Astothis last sentence,Ican ‘aynow,whatIsaidforty-two yearsago—“mustbereservedfora “future paper!” ImayaddthatIhavebeenconsidering the mbject forforty-two years—night and dayforforty-two years. 26 PRESIDENT'S ADDRESS, (an, 10th, Tdonotmeanallofeverydayandallofeverynight;Idonotmeansomeofeachdayandsomeofeachnight;butthesubjecthas been onmymind allthese years. Ihave been trying, many days ‘and many nights,tofindanexplanation, buthavenotfoundit. Letthere beanelastic solid body ofexceedingly small density, and letthere beatubular portion ofitporous, butwith thesame aggregate rigidityasthatofthecontinuous elasticmatteraround it,Letthepores befilled with adense viscous fluid, andletthis fluid beforced, byaidof@pistonorotherwise,tomovethrough thetube. The pull ofthefluid upon theporous solid willproduce static rotational displacement exactly proportional tothecontinued rotatory motion which wehad inthe case ofthe viscous fluid. Some ofthe most interesting practical problems ofelectro- magnetic induction canbedynamically realised,asitwere,in model, byfollowing outthis idea;infact,ifwehadnothingbut electricity and ether, thething would bedone. Ifitwere not forthegrossponderable matterthatweareforcedtoconsider,Ishould beperfectly satisfied with theproblem ofelectro-magnetic induction, bytaking theelectricity asaviscous fluid, andether an clastic solid, porous insome places, and continuous ornon-porous elsewhere. y\ \ A ALTERNATING CURRENT CONTINUOUS CURRENT Fie. 8. Fra. 2. Now, ifyouwillpardon me,though itisvery late forintro- ducing another topic onwhich tospeak—I shall confine myself toone,andthat ismagnetism, Imust return totherotational case. Imagine this(Fig. 2orFig. 3)tobethesection ofan 1m) ETHER, ELEOTRIOIY, AND PONDERABLE MATTER, 27 ordinary helix orsolenoid with asolid copper core. Imagine a continuous electric current (Fig. 2)oranalternating electric current (Fig. 3)ofelectricity sent through it. Whatever the carrentofelectricitymaybe,Ibelievethisisareality:itdoes pulltheetherroundwithinthesolenoid. Idonotthinkthisis adream ofelectro-magnetic theory ;Ibelieve ittobeareality. Whatever ether is,wemovethroughit—theearthmovesthrough it.Astronomers and opticians donoteryoutandmake their livesmiserable because oftheaberration oflight. Fresnel and Professor Stokes have done allthat man, uptothe 9thof January, 1889, hasbeen able todotoexplain thedynamics of theaberration oflight.Itmaybenotbeyondman’srangeto complete the solution—how the earth can tear through this tastic solid ether and yetthewaves oflight bepropagated through itasthey are, The aberration oflight isstill an absolute mystery. Yet people who deal with optics and ‘stronomy are not expected tobemiserable forlife because theyhave that difficulty ever before them. Well, arewetobe shwlately unhappy because, while weseeamobile wire caused, invirtueofanelectriccurrentthroughit,tomovebyelectro- magnetic force, wecannot seeany possibility ofexplaining how ‘medium capable ofthe“magnetic stress” can allow itto move? After all, great asthe mystery there is,there isa uystery greater than that. The actoffree-will with reference tothelawsofmatteris»greatermysterythananythingthathaseverbeensuggested orimagined inthedynamics ofether,and lectromagnetism, andlight. Somehow orother, however itis, theether ispulled round, theether does getaturning motion in theinterior ofasolenoid; somehow orother itdoes give a tuning motion toether within our supposed copper core; and somehow orother there isamotion following these very laws wehave been speaking of,and illustrated bytheviscous fluid maalogy. Butnowfortheiron. And now, instead ofanalternating curent through thehelix, take aconstant current through it. What canitdo? Onething ortheother itdoes: either a constant current through this helix drags theether round and Ey PRESIDENTS ADDRESS, (Jan,10, round inside, oritdrags itround toacertain angle proportionate tothestrength oftheelectric current, and brings ittostatic equilibrium sotured. Itdoes either one orother ofthose things. Now, howonearth caniron differ, intheprinciple ofthe interfacial law, from copper? Our interfacial law depending on equalviscosities isquiteclear,butwhenyouintroduce ironyou introduce aninterfacial difference depending onrotation, without anything that could possibly beacause ofanyviscous action, or 8cause ofany elastic action. Elastic action (unless ofcom- pression orrarefaction, and these arenot ofour present subject) requires distortion. You have noelasticity ofanincompressible clastic solid without distortion, Now if,byapplying tangential force allround thespace within cylinder, youkeep turning thecircumference, youwillkeepturningthecontents. .Ultimately thewhole fluid within will goround with the same angular velocity asthe circumference ofthe viscous fluid within it. ‘Thus ourviscous fluid analogue works outperfectly well forthe magnetic force within asolenoid having anynon-magnetic material within it,and illustrates the fact that itisthe same for con- ducting andnon-conducting matter. But with iron thecase is something quite different. Our viscous fluid analogue iscalled ontogive usagreater permanent angular velocity, oragreater static rotational displacement, inthespace occupied byiron in themagnetic analogue, than inthis surrounding space! Thus the primary phenomenon ofthemagnetisation ofabarofiron within ahelix, absolutely leaves usbehind, euts theground from underus,bothastoourviscousfluidanalogyandourelastic solid analogy. Ifitistobeafluid going round andround, we must have anaction between the portions offluid onthe two sides ofthe interface, depending, not ondistortion, but on rotation, Orifwetake ourelastic solid analogue, wemust have static equilibrium oftheelastic cylinder, with the inner part turned through #greater angle than therotational part ofthe displacement ofthesurrounding matter. Anirrotational circular displacement subtracted from this, procures fulfilment ofthe ‘noelip condition attheinterface. ‘The distortion due tothis irrotational displacement gives rise toatorque, tending toturn wm} ETHEB, ELECTRICITY, AND PONDERABLE MATTER, 19 the matter within the interface! Hence we must have an arrangement ofmatter inwhich aconstant torque produces a constant angular displacement inabody, anddoes notproduce continued rotation. The only thing that can dothat isan inherent rotation existing inthe molecules ofmatter. This seems theonly thing that candoit,andthiscandoitcertainly. Batconsider this—that thegyrostat shows usthething done; and Iwill just conclude, ifyou will allow me, byasimple gyrostatic experiment—a very wellknown oldgyrostatic experi- ment—and Iwant toaccentuate theapplication ofit. am going toshow bythis illustration, with reference tothe ideaofamedium, amedium which has theproperties ofan incompressible fluid, and norigidity except what isgiven toit gyrestatically. Here is,sotospeak, amolecular skeleton that angive ussuch afluid—p setofrigid squares with their neighbouring corners joined byendless flexible inextensible threads, running frictionlessly through holes inthecorners, or rund pulleys mounted inthecorners (Fig. 4). Here isamodel aoe me enone one saree e:BH Ge aaasees thusconstructed—sixteen rigid squares and nine endless cord segments connecting thecorners inthispattern, forming akind ofweb, Now,ifwetakeanordinary clothweb,andpullitin » PunsIDENT# ADDRES, cos. different directions: inthedirection ofthewarpandthedirection ofthewoof, you cannot stretch it;butat45degrees from the ‘warp andwoof youcanstretch itvery freely. Weallunderstand that. You know howthesurgeons take advantage ofitintheir diagonally cutbandages. Now here isaweb which isequally easily stretchable inalldirections, and yet which isofconstant area—a constant area forinfinitesimal displacements, notacon- stant area forvery great displacements. The circumference of each rigid andofeach flexible square isgiven. Well, now, ifyou infinitesimally alter thesquare intoanot-square rectangle, orinto arhombus, thearearemains sensibly unchanged. Thefirstchange ofthearea isadiminution inwhatever direction youstretch it; butthat isproportional tothe square ofthe strain, sothat you may say, inlanguage ofinfinitesimals, thearea isunchanged. ‘The constancy oftheperiphery, then, ofeach ofthese figures givesrisetoandentailsthecondition ofanapproximate constancy ofthe area. Here, then, we have inthis skeleton atwo dimensional working model ofamedium which isunchangeable l€_iL, lkiT inarea, butisfreely extensible inanydirection, provided you allowittoshrinkproportionately intheperpendicular direction. ‘Well, now, letusputagyrostat into each ofthose squares (Fig.5),andyouhaveallthatiswantedtofulfilthestrange— almost inconceivable—condition foradynamical model ofelectro- Ws] ETHER, ELEOPRIOITY, AND PONDERABLE MATTER. 31 magneticinduction inironwhichIhaveputbeforeyou.Iwill justmake anexperiment illustrating that, ifitisnotoccupying toomuch time. [Sir William Thomson then spun thegyrostat.) tum azimuthally thesquare frame bywhich Ihold it—firet in medirection, andtheredendofthebearingoftheaxleofthe fy-wheel turns up; Iturn theother way, andupcomes theblue erie Thegyrostat ismounted inasquare frame, asyousee, which Ihold inmyhand. The rigid case bearing theaxle ofthe fy-wheel is,asyou see,free toturn round theaxisofthese trunnions, mounted horizontallyonbearingsinoppositesidesof thesquareframewhichIholdinmyhand.Theaxisofthese tramnionsisperpendiculartotheaxisofthefly-wheel.Ishall walkroundandroundtoright,tokeeptheredsideup;Iwalk mundtotheleft,anditkeepsthebluesideup.Itisratheracariousthing. ‘Therearethreelittleobjectsonatray,asitwere. Imagine this tobeabutler’s tray, with wine-glasses onit represented bythese india-rubber corks. Aslong asIturn ever littletomyleftallgoeswell;ifIgostraightforward,itis doubtful; butifIturn byaninfinitesimal angle tomyright,overitgoesandeverything fallsoffit. Look, now, atthegyrostat resting inthis position onitstran- tions, theaxes ofthetrunnions andofthefly-wheel being both apresent horizontal. The outer square frame seems immovable inarimuth. When Iapplyacoupletendingtomoveitinazimuth itdoesnotmove.Itdoesnotmoveinazimuthtillthegyrostat tumsrounditstrunnion axisandbringsitsfly-wheel axistobe papendicular totheplane inwhich Iamtrying totum the square frame. AndImust apply acouple whose time-integral is equaltodoublethemomentofmomentumofthefly-wheel,before Tean getthegyrostat from theposition with theblue endup,to theposition with theredendup. Before Isucceed inturning thesquare frame even adegree inazimuth Imust have applied a tomne whose time-integral isequal totwice themoment ofthe momentum ofthefly-wheel. This closed brass case, with arapidly rotating fy-wheel mounted onbearings inside it,iscalled agyrostat because in ‘irtue ofrotation itstands, however youplace it,with anyofits a RESIDENT'S ADDRESS, {[¥an, 108, edgesrestingonehard,smoothtable.Yousee,placeitasI will, itcannot fall. IfIplace itwith itscentre ofgravity above thesupporting point, itstands atrest. With itscentreofgravity ‘notvertically over thebearing point, itgoes round inazimuth, butitdoce notfall. Nowimaginemountedineachoneoftherigidsquaresofthis webagyrostatexactlyasthisoneisinthesquareframe4cop hold inmyhand. Ifthefly-wheel speed begreat enoug...» ~~ ofthose rigid squares ispractically immovable inazimuth. Ido notsay itisimmovable, but Isayyou may make itpractically immovable bymaking thevelocity ofthefly-wheel sufficiently great. Thus wehave askeleton model ofaspecial elastic solid with fastructure essentially involving agyrostatic contribution to rigidity. Nowdonotimaginethatastructureofthiskind,gross asitis,isnecessarily uninstructive. Look atthe structures of living things; think ofallwehave toexplain inelectricity and magnetism;andallow,atleast,thattheremustbesomekindof structure intheultimate molecules ofconductors, non-conductors, magnetic bodies, andnon-magnetic bodies, bywhich their wonder- fulproperties now known tous,butnotexplained, aretobe explained. Wecannot suppose alldead matter tobewithout form and void, and without any structure; itsmolecules must hhave some shape; they must have some relation tooneanother. SothatIdonotadmitthatitismerelyplayingattheory, butitishelping ourminds tothink ofpossibilities, ifbyamodel, howeverroughandunpractical, weshowthatastructure canbe produced which isanincompressible frictionless liquid when no gytostatic arrangement isinit,andwhich acquires apeculiar rotational elasticity orrigidity astheeffect ofintroducing the gyrostats intothese squares. Imagine acorresponding model in three dimensions, with rigid cubes instead oftherigid squares which youseeinthemodel before you. Instead oftheendless flexible cords which you see, you may imagine elastic threads stretched between neighbouring corners ofthe eubes. Ineach cube mount three gyrostats, with their trunnion axes perpendi- culartothethreepairsofitsfaces.‘Thegyrostatic domination 1860} EYER, ELECTRICITY, AND PONDERABLE MATTER, 33 thus provided, causes thecubes tobepractically immovable in rotation, butleaves them perfectly freetotake translatory motion. There youhave abody, then, thatyoucould notdistinguish from anordinary elastic solid inrespect toany irrotational distortion, orinrespect totranslational motion ofthewhole, butwhich if ‘youtrytoturn it,will notabsolutely resist, but will only admit ofturning bystretchings oftheconnecting elastic bands. Itwill rotvgimmovable inrespect oftheturning, butitwillbebalanced‘byaconstant couplewithaconstant degreeofrotatory displace- ‘ment. Thus upon this solid, theeffect ofaconstant couple is nottoproduce continued rotation, buttoproduce andbalancea constant displacement;andthatbalancemightlastforanytime, howeverlong,iftherotationalmomentofmomentumofthefly- wheels isbutgreat enough. Now, lastly, Ishould just explain briefly that the rotational contribution torigidity ofether inironmust beenormously less * than incopper orinair. The total effective rigidity oftheether due toelastic action and gyrostatic effect inallnon-magnetic bodies isthesame. Iniron there isless gyrostatic contribution, with equal elastic contribution, tothe total rigidity. These conditions fulfil exactly what wewant fortherelation ofether between airandironinside thehelix ofanelectro-magnet. But, alas! weareonly ledontoinscrutable difficulties. How much does ourelastic solid gotowards theexplanation, when inthevery fndamental fact ofthe mutual motions bywhich electro- magnetic forces aremade manifest tous,wehave aforce asof astrained solid between. the bodies (magnets orwires) whose motions revealed to(Erated andAmpére theexistence ofelectro- magnetic force? Why isitthat those strains donotsimply ‘balance themselves inthe solid? How can there beasolid capable ofgiving risetothat wonderful condition which wehave intheairbetween thepolesofanelectro-magnet—for instance, such that apiece ofcopper will falldown through itattherate of,perhaps,aquarterofcentimetrepersecond?Lookonthe subject asengineers, and think ofthe“strength ofmaterials” wanted forether inair, with the molecules ofthe airitself tearing through itinalldirections atspeeds averaging 500metres Vou. xvu. 3 u PRESIDENTS ADDRESS, [Ban. 10th, persecond,ormoreorlessaccording totemperature. Thinkof theforces, amounting to110 kilogrammes weight persquare centimetre, with which two bars ofiron magnetised to1,700 C.GS., with faces separated byathin space ofair,and with Ewing's 46,000 C.G.S. ofmagnetic force intheairaround the bars, areurged towards one another. How can itbethat these prodigious forces aredeveloped inether, anelastic solid, and yet ponderable bodies beperfectly free tomove through that solid? Now Isimply say, allthat hasbeen done tothink outthis subject merely gives usadynamical theory ononepart ofit, T have absolutely—not ignored, because Ihave spoken ofittwoor three times—but Ihave leftoutinthecold, theelectrostatic part, thething weknew first. Our first love waselectrostatics. That isabsolutely leftoutinthecold; wedonottouch it.Wedonot getnear toexplaining themutual force between twoelectrified bodies, inany ofthese illustrations orattempted explanations ; wedonoteven getnear themutual attraction between theiron ofanelectro-magnet, orthesteelofapermanentmagnet,andits armature orkeeper; wedonot get near toexplaining the possibility ofthe motions ofthe bodies that demonstrate the forces. Weonly trytoexplain foraquiescent system ofcon- ductors and insulators, the variable distributions ofelectric currents which from mathematical theory and experimental observation we know toexist. And here, Iamafraid, Imust end bysaying that thediffi- cultiesaresogreatinthewayofforminganything likeacompre hensive theory, that wecannot even imagine afinger-post pointing toawaythat canlead ustowanls theexplanation. That isnot putting ittoostrongly. Ionly saywecannot now imagine it. But this time next year,—this time tenyears,—this time one hundred years, probably,—it willbejust aseasy, aswethink itis tounderstand thatglassofwater,whichseemsnowsoplainand simple. Icannot doubt butthat these things, which nowseem toussomysterious, willbenomysteries atall;thatthescales willfallfrom oureyes; that weshall learn tolook onthings in adifferent way—when that which isnow adifficulty willbethe only common-sense andintelligible wayoflooking atthesubject. 186] ETHER, ELECTRICITY, AND PONDERABLE MATTER, 35 Task you topardon meforleading you uptos0impotenta conclusion asthat wereally know nothing below thesurface of this grand subject which constitutes theprovince oftheInsti- tation ofElectrical Engineers. APPENDIX. Alternate Currents throughaStraightConductorofRound,or odof,Non-Magnetic Material. Let odenote thespecific resistance insquare centimetres per second (orthe“specific resistance C.G.S.") ;@—»_theradius ofthe wires R(S) 5thevalue ofol+wa? [ortheresistance (in centimetres persecond) ofanylength (1)of thewire, with steady current through it]; - RN) ,,the effective ohmic resistance ofthe same length (1),with alternate current ofNperiods persecond through it. ©(N) »thecurrent-density atdistance rfrom theaxis, and attime t. CN) 5,thecurrent-density intheaxisattime t,Wehave ¢(N)=C(N)(bergeos.8—beigain.6), where adenotes(24/2%); » a » @rNts ber andbeidenote twofunctions defined asfollows:— =1-% ¢_ _ ge,berg=l— aptpage—bos igoff,beig=2—grag+he ‘Andifpdenotethevalueofg,withr=a,wehave R(N)_,.berpbei’p—beipber’p RS)=}? (berpF elDF? where the accents denote differential coefficients. ‘The following table ofnumerical results hasbeen calculated formebyMr.Magnus Maclean, official assistant totheProfessorofNaturalPhilowophy intheUniversity ofGlasgow:— se savnxonx10enaenexreADoans,(Jn.10t, Z| Fe . prrrne ia pret rrrrrereg s . ZERERELERETR Eeifeveeeehitiiilic, pebrrne ae ~[seteeenaceeseeese We) -—=«VOTE OP THANKS TOTHE PRESIDENT. 3 Forcopper wehave o=1,610 square centimetres persecond. Hence, with N=80wefind q=198r=2r, ‘Thosin respect totheohmic resistance ofthewhole wire, wemay {orcoppertakethecolumnheadedqasthediameter ofthewire, andinrespect tothedistribution ofthecurrent through thewire (expressed bytheberbeiformula above) wemay take gthe diameter ofthecylindric shell inwhich thecurrent-density isto becalculated. Professor G,Fonsus:Ihaveanextremely pleasingdutytoPrtwerperform inrising topropose—*Thataheartyvoteofthanksbe “given toourPresident fortheadmirable, instructive, andhighly “interesting Address which hehasgiven tous,and that hebe “asked topermit ofitspublication intheJournal oftheInstitu- “tion.” Ithink everyone willagree with mewhen Isaythat we lavethiseveninghadoneofthosetreatswhichitrarelycomesto1toenjoy—to hearanexposition fromamanofSirWilliam Thomson's experience and position, unfolding toustheresult ofthelabours ofhislife,putting before usnotonly hispersonal reollections inconnection with the advance ofthe science in which we are allsomuch interested, and the recollections of thers who have worked with him, butalso giving,astheresult ofthis, advice tothe younger members ofthe profession and thers astowhat subjects topursue, and soforth. With the Mdvancement and development ofour science Sir William Thomson has been sointimately connected, that allofusfeel thatwearebeing more closely brought into contact with itspast history when itisdwelt upon byhim insuch anAddress ashe Isjustgiven tous. ButIthink everyone willagree that what vehavetothank himformostly, isthemanner inwhich hehas inspired uswith some ofthe enthusiasm which hehimself possesses ininvestigating the deeper problems ofelectrical ience, and the interest which hehas made every one ofus feelintrying tounderstand those brilliant analogies between Physical phenomena which aretolerably comprehensible and lectrical phenomena which arerather obscure tous. Ibelieve 8 VOTE OF THANKS TO THE PRESIDENT. (Jan. 100, Friar Tam theonly member present whohadthefelicity ofattending the three weeks’ course oflectures which Sir William Thomson delivered atBaltimore ontheundulatory theoryoflight,andI cangive you some conception ofthefeeling ofthose who were fortunate enough toattend those lectures, when Isaythat for three weeks wehad anuninterrupted exposition of similar modeoftreating thesubjectaswehavehadthisevening; and itisathing 0rare that Ifeel everyone will have appreciated the treat which wehave had inlistening totheAddress this evening. Among theaudience thisevening there aresome very young members—some who have but little knowledge of mathematics—and there arealso present among usthegreatest mathematicians ofthecountry, and yetIventure tosaythat there isnotone, inthelong range ofabilities included bythe audience, who hasnotlearnt agreat deal, andhadagreat deal of newlight thrown upon thesubject, inthecourse ofthisevening; and Iamquite sure that those who have nothitherto appreciated thebenefits ofmathematical investigation into thetheory ofsuch phenomena will beable tosee from this Address the way in which amathematician can, bytrying togetdeep into thesources ofthings, beable eventually toactually apply such knowledge to practical phenomena. Iremember very well in1874, just after theappearance ofClerk-Maxwell’s book, Professor Tait—who took, perhaps,atoosanguineviewofthework—considered, infact,that itsettled thewhole question astowhat electricity was—Professor ‘Tait said: “Some twenty-five years agoSirWilliam Thomson “said tome, ‘Ifyou will tell mewhat electricity is,Iwill tell «you everything else;”and,” continued Professor Tait, “Clerk- “Maxwell hasgiven usthis information; wemust nowapply to “Thomson fortherest.” This evening wehave been listening to SirWilliam Thomson inhisattempts toexplain, sofarasitis possible atpresent, what electricity is,andwhat everything else istoo. Ifeelquite sure that thevote ofthanks which Iam proposing will bemost heartily agreed to,and that everyone inthis room feels,asIdo,thatwehavebeenlisteningtoan Address ofaquality which itisalmost impossible toexpect that wecan ever listen toagain. Wemust, however, also thank ws] «VOTE OF THANKS 10 THE PRESIDENT. 3 SirWilliamThomson forhavingcomethislongjourneyfromPrtwaerGlasgow especially togive usthis Address, and tohave taken thetrouble and submitted tothefatigue oftravellingallnight forouredification. Butitisnotsurprising thatheshouldhave taken this labour upon himself when during thehour hehas occupied indelivering toushisAddress hehasbeen trying to compress the subject which hehas been working out, almost continuously, night and day, forforty-two years. Ibegtomove theresolution which Ihave already read out. Professor W.E.Avgton: Itismygood fortune—for ITetafeelittobe@goodfortune—to beaskedtosecondthisvoteof thnks, which Iamsure you will allaccord with the greatest eathusinsm, toour new President. Within thelastfewmonths vehave had the inestimable privilege ofhearing our new President—ofseeingournewPresident, Imaysay—inthisroom intwo totally different capacities, Afewmonths agoitwasas 0electrical engineer that heshowed and described toushis uuostbeautiful measuring instruments;to-nightitisinatotally iifferent capacity—as atheoretical electrician—that hehas ctammed usforsolong atime; anditisduetothemarvellous per that hepossesses ofcombining theory with practice that lehasbeenenabledtoaccomplish theworkthathehasalready done, and which weallhope hewill live foralong time to continue carrying on. Wehave inthis country many eminent ‘ogineers;wehavealsomanycompetentmathematicians; butI ‘enture tothink that wehave only onewho cancombine, inthe marvellous way that our President hascombined, theory with practice; infact, Imay shortly say that wehave but one Thomson, Onthe previous occasion, when asanelectrical ‘agineer hewasaddressing us,hepointed outhow important it Wstostudy theory. To-night asanelectrican hecommenced, ‘iththediffidence that hasmarked allhiswork, bytelling us batifweonlyfirstbecamecompetent engineers, theadditionalknowledge ofelectricity thatwasnecessary could beacquired in 4fewmonths;andIamsurethisthoughtwillbe#consolation forallengineers here to-night tosleep on—that ifthey only comeintothisroom wellarmed with engineering knowledge, all 0 VOTE OFTHANKS TO THE PRESIDENT, (Jan. 10%, Rotter thispurely electrical science they willbeabletograsp inafew months. Isayitisaconsolation, and Ishould bevery sorry to remove that consolation from them, At the same time Icannot Dutfearthet this previous knowledge ofengineering, ifitwere sufficiently complete tosatisfy SirWilliam, would besomething s0vast and socomprehensive initscharacter astoappal the ordinary practical man. You allknow what adeep debt ofgratitude isowed by electrical engineers, and indeed Imay saybythewhole world, tothewonderful mathematical investigations that ourPresidenthasto-nightsolightlytouchedon—that investigation whichmade submarine telegraphy commercially possible. How SirWilliam's pianoforte wire hassounded thedeep sea—the ocean Imean, of course, notthe musical note—is quite familiar toyou all; and howhisship's compass hasguided vessels totheir journey’s end iswell known toallthe world. Well, now, Idaresay, onsome ‘occasions ships have started from aport armed with hiscompass— have started onvoyages ofexploration—not perhaps quite know- ingwhere they were going, orwhat they would find; and I think to-night hehas supplied uswith another compass—a mental compass—and started usallonavoyage ofexploration ; and because hecannot telluswhat isthecountry that weshall arrive at,itmay befifty orahundred years hence, itdoes not make the compass that hehas provided uswith—that mental compass—the lessvaluable, ordiminish ourunbounded feeling ofgratitude foritsgift. ‘The motion wascarried byacclamation. Pisuen, ‘ThePuzsivenr: Gentlemen,—I thank youallverywarmly for thekind manner inwhich you have received the motion which hhasbeen putbefore you. Ithank themover and seconder for thetookind appreciation which they have shown ofmyimperfect, effortstospeaktoyouinpropertermsthisevening. Ifeelthat. your electing metobethefirst President of“The Institution of “Electrical Engineers” isreally tomyself avery great personal kindness, because Ibelieve that itwasunderstood that, living as Idoats0great adistance, Ishould notbeable todomyduty 189] VOTE OP THANKS TO RETIRING PRESIDENT. a properly asPresident—to beatallthemeetings, asIwould like7seu, tobe,oratthegreaterproportion ofmeetings, asIcertainlyought tobe,butonly atacomparatively small number. Ilive 400 bundred miles away, and,asyouknow,haveincessantoccupationat thstdistance inGlasgow, sothat itispractically impossible forme tobehere atmany ofthemeetings. Imust therefore askyouto lindly excuse myinability todomyself thepleasure ofbeing ithyou continuously, oreven frequently, and Imust trust to thekind consideration oftheVice-Presidents totake the place tbatIought tooccupy inthePresidential chair. Ihope tobe present atallthemeetings oftheInstitution after theendofour Gksgow University session, and ifpossible Ishall beatsome \wlore that time, butIam afraid itcanbebutatvery few; and Imust askyou tokindly excuse me,andtounderstand that itis ‘imply owing tothenecessity ofthecase. Itwasnecessary, in fut,formetoletitbeknownbeforeyoudidmethehonourofdecting me that this would bemycase with reference tomy (oties asPresident ofthis Institution. Imust say that, as girls the management ofitsbusiness, Ithink none ofus needhave any concern atallwhen weknow that itisinthe bands ofourexcellent Secretary, Mr.Webb. Heisalways alive tothe interests oftheInstitution, always attentive incarrying outitsobjects; and with thevery great ability which hehas ‘torn wemay feelthat themanagement isinexcellent hands. ‘ThenwiththeCouncil here, inwhom youhave allsuch perfect confidence, Ithink youwillfeelthat theInstitution loses nothing vhatever bymyabsence; butIhave deemed itright that Ishould rakemyapology, asyouhave been sokind tomethisevening. Major-General C.E,Wenser: SirWilliam andgentlemen,— wsjr.ce=- Onbehalf ofthe members ofthe Institution, and inafew words, Ihave thegreat pleasure togive expression toour fgtitude toMr. Graves, the last President ofthe Society ofTelegraph-Engineers andElectricians, forthesplendidwayiawhich hehasfilled that chair during thepast year. All ‘ldmembers aresorry that hehasdone soto-night forthe listtime. Mr.Graves, weknow, hasdistinguished himself lugagoastheEngineer-in-Chief ofthePost Office, andthere 42 VOTE OF THANKS TO RETIRING PRESIDENT, (Jan. 10tb, YafeSe isnothing Icansaywhich could addtohisreputation ;but during thepast year heendeared himself tousbythecareful and friendly way inwhich hehas presided over usonevery occasion when wehave met together. Iwillnotatthis late hour occupy your time byany further remarks onthesubject, butask“that youwill accord, asmembers oftheInstitution of “Electrical Engineers, your cordial thanks toMr.Edward Graves “for theadmirable manner inwhich hehasdischarged theduties “of President oftheSociety ofTelegraph-Engineers and Elec- “tricians during thepast year.” Meso, MtC.E,Sracxourrnt: Irisewithverygreat pleasure to second the proposition ofavote ofthanks toourretiring Presi- dent, Mr.E.Graves. Ihave hadthepleasure ofworking with him formany years, and therefore Icanspeak with some personal experience ofhisexcellent business qualifications, ofthevery ready waywith which hedeals with matters andsurmounts diffi- culties, and ofthevery courteous wayinwhich healways performs hisonerous duties. Onbehalf ofthe Council Iamsure Imay saythatweareverymuchindebtedtohimforhisconstantattend— ‘ance, and forthe way inwhich hehas conducted theaffairs of theSociety during hisyear ofoffice;andIthinkImayalsosay the same onbehalf ofthe members ofthe Institution. ‘The motion wascarried unanimously. ae ‘Mr.E.Graves: When youdidmethehonour ofelecting me tooccupy the chair which Ihave sorecently vacated, Isaid it wasacase ofthewrong man inthewrong place, andIhave notat allaltered that opinion;infact,ifIhad,theAddresstowhichI have listened to-night would have corrected theerror. But I told you also that Iwould domybest. Ihave striven todo80. ‘Wehave hadmany rheetings during theyear—many meetings of theSociety, many meetings oftheCouncil, many meetings of Committees, especially onextraneous matters such asthose con~ nected with theLord Mayor's Committee oftheParis Exhibition ; Thave attended them all,and now, attheend, Ifeelmyself asort ofchief mourner fortheSociety ofTelegraph-Engineers, and I resign toSirWilliam Thomson themuch pleasanter task ofacting 1580,] ELECTIONS, “ asgodfather totheInstitution ofElectrical Engineers whichtakesMs.itsplace. The SECRETARY announced that thenext meeting would take place onJanuary 24th, when apaper would beread on“The “Insulation Resistance ofElectric Light Installations,” by Professor Andrew Jamieson, Member. Aballot fornew members took place, atwhich thefollowing were elected :— Foreign Member: José Savall ySalvat. Members : Jobn Marshall Gorham. Professor Arthur Schuster, William Edwin Heys. PhD, F.RS. Captain R.Hippisley, RE. Tins Augustus Timmis, Professor Oliver Joseph Lodge, M.Inst. C.E, &ME. F.RS. James Wimshurst. Anthony Reckenzaun. Associates : Sydney William Baynes. Henry Francis Sharman- Samuel Joseph Coxeter. Crawford, William Fulton. Thomas Thomas. Charles Frederic Heywood. Charles Turner. EAmund Philip Jackson, T.M.Winstanley Wallis. Alexander Login Lineff. Jobn C,Waltham, Lieatenant W.Luard, RE. Michael Scaler Warton. David McNeill. George Richard Webb. EAward Thomas Mercer. Charles Aspull Wells. Edwin Oldroyd. Ebenezer Clarence Wood. Reginald Frederick Yorke, Studente: John F.Coote. Josiah Sayers. Wilmot Emest Lane. Ernest Waltham. ‘Max Salmony, ‘Themeeting then adjourned, “ DONATIONS, [Jan.24th, The One Hundred andEighty-fourth Ordinary General Meeting oftheInstitution washeldattheInstitution ofCivilEngi- neers, 25,Great George Street, Westminster, onThursday evening, January 24th, 1889—Professor W. E.AYRTON, F.RS,, Vice-President, intheChair. Theminutes oftheAnnual General Meeting held onDecem- ber 13th, 1888, and ofthe Ordinary General Meeting held on January 10th, 1889, were read and approved. ‘The names ofnew candidates for election into the Institution ‘were announced andordered tobesuspended. The following transfers were announced ashaving been approved bytheCouncil :— From the class ofAssociates tothat ofMembers— Léon Drugman, From the class ofStudents tothat ofAssociates— F.W.Chapman. W.Clark Fisher. J.Mountjoy Elliott. E,Loraine Helis. Donations totheLibrary were announced ashaving been received since thelast meeting from the Astronomer Royal; theElectrical Society ofJapan; Rev. G.T.Carruthers; P.A. Scratchley, Esq.;Messrs.Whittaker &Co.(Publishers); theCom-mander F,Salvatori, Foreign Member;ProfessorK.E.Zetzsche, Foreign Member; Wm. Ellis, Member; W.T.Hancock, Member; and E.March Webb, Member. ‘The Secretary announced that Mr. W. H.Preece, Past~ President, had presented anelectric clock forthe use ofthe Library oftheInstitution, Ahearty vote ofthanks wasaccorded tothedonors fortheir presentations totheLibrary,andalsotoMr.W.H.Preecefor hisuseful and valuable gift. 1980) ‘THEINSULATION RESISTANCE, Ere. 6 The Cuarruan: In the unavoidable absence of Professor Jamieson inGlasgow, theSecretary willread thepaper announced forthisevening. ‘TheSECRETARY then read thefollowing paper :— THE INSULATION RESISTANCE OF ELECTRIC LIGHT INSTALLATIONS. ByProfessor A.Jawiesoy, F.R.S.E., M.Inst. C.E., Member. Insulation Resistance ofLand Lines and Submarine Cables. —In the early days oftelegraph land lines andofsubmarine ables, inewlation resistance was neither definitely specified for, torsccurately measured duringthevariousprocesses ofmanu-fictore and ofmaintenance. Afewfailures, however (insome cases dearly bought), soon impressed electricians with the tecessity forpaying more attention tothisimportant quality, and they were not long indemanding from contractors theminimum ofinsulation resistance which would enable their lines tobe' worked with success. Not only arecareful insulation tests now made during the manufacture ofinsulators and theerection of theland lines under theBritish Postal Telegraph Department, but“daily tests” aretaken ofallthemore important lines, The minimum insulation resistance which each insulator should have depends upon thelength oftheline. Foralineof100| nileslongeachinsulatorisexpectedtogivearesistanceof| 500megohms, whereas foralineofonly 10miles 50megohms is‘considered sufficient. Whenalinehasbeenerectedandputinto working order, itisusual tofixupon aminimum standard insula-tionresistance permile.Ifthemeasured valueatanytimefalls below this standard, then thelineisconsidered faulty, andsteps are taken toremove the fault. ‘The Post Office standard is 200,000 ohms per mile, Inthe cases ofsubterranean lines and ofsubmarine cables aneven still more rigid and careful system ofinspection andoftesting isnowcarried outduring the manufacture ofthecoresandcables.Theresultsareinvariablyreduced toaninsulation resistance per unit oflength ata “6 THEINSULATION RESISTANCE OF—(Jan.24h, standard temperature. The following extract from thespecifica- tion ofasuccessfully made and laid cable willserve toillustrate this:—“The insulation resistance perknotofthe gutta-percha core shall notbeless,whentestedat75°Fah.,than300megohms, “14 days after manufacture andafter twominutes’ electrification, “If theresistance ofany portion ofthegutta-percha (reduced to «75° Fah.) atanyperiod ofthemanufacture fallbelow theabore “specified limit, the cable toberejected.” Throughout the whole oftheoperations ofcoiling and uncoiling acable intothe shore andship's tanks, during thelaying, andsubsequently during itsexistence, theinsulation resistance isperiodically measured and tabulated. Results ofNeglecting Insulation Resistance inEledric Lighting—We have been induced thus toparticularise the extreme care and attention which isnow considered absolutely necessary inthe case ofland lines and ofsubmarine cables, because theindustry ofelectric lighting hasjust passed through asimilar experience tothat which these industries met. with in their earlier days. When theindustry ofelectric lighting was first beginning totake root inthis country, electric light engineers pooh-poohed theidea ofaccurately measuring thein- sulation resistance ofanypart oftheir work, and they invariably turnedadeafeartothetimelywarnings oftelegraph-engineers andelectricians. Experience has,however, taught them, asitdid their elder brothers, that good uniform insulation resistance throughout aninstallation isofimmense importance, andthatit cannot beobtained without careful attention todetails and » rigorous system oftesting during manufacture anderection, and also after completion, Ifaninstallation isnot well insulated throughout, leakage ofthecurrent takes place, andconsequently waste ofpower andmoney;electro-chemical actionsetsinwhere damp orliquids arepresent, and theconductor isthereby eaten through orashort-cireuit occurs, which foratime stops the whole ofthelighting. Discredit thus notonly falls upon the contractor, butothers arefrightened from employing theelectric light, tothedetriment oftheindustry asawhole. 1989) ELECTRIO LIGHTINSTALLATIONS. " Necessity for Discussing aStandard ofInsulation Revistance—The terms usually employed inregard tothe inmlation resistance ofanelectric light installation are, that it shallbe“good” or“perfect,” according tothefancy ofthe purty purchasing theplant orofhisadviser. ‘These terms have, however,nospecificmeaning, andaretherefore veryvariouslyinterpreted. Nodoubt theSociety hashadthis subject before themincommittee when drawing uptheir setofrules forthe “prevention offire risks;” but the Institution asawhole has never had anopportunity ofdiscussing it,and thereby of ‘impressing itsmembers and the public generally with its importance, and the necessity forobserving some minimum standard of insulation resistance. The writer had his atten- timdirected the other day toacommunication read before theInternational Society ofElectricians, Paris, inNovember lust,byM.Picou, wherein that gentleman ably analysed this question, and proposed arule based upon tests ofshort lengths feotton-covered insulated wire. Hehasnothad anoppor- tusity, however, oflearning theextent and thenature ofthe disassion upon M.Picou’s paper, and hecanonly, therefore, place hisrule before the members, along with afewother mules, and leave them toform their own opinions, guided by their practical experience. General. Propositions.—The following points will nodoubt bereadily admitted inregard toinsulation resistance ofelectric lighting installations:— 1,That, owing tothecomparatively lowresistance ofthe conductor circuit ofordinary electric light installa- tions onthe parallel incandescent system, the minimum standard of insulation resistance does not require tobenearly sohigh per unit of length asinthe case ofsubmarine cables, 2,That itshould bedirectly proportional totheelectro motive force ofthe generator, ordifference of potential between the conductor and earth or ‘between the forward and return conductors. “ THEINSULATION RESISTANCE OF(Jan.4, 8,That itshould beinversely proportional tothetotal length ofinsulated conductor, ortothetotal current, ortothe number oflamps incircuit. 4,That, whatever rules orstandards ofinsulation resistance may beadopted, they should beapplied, notonly toalltheconduéting wires, butalso tothedynamo, the lamps, and allthe fittings, or,inother words, tothe whole circuit. Rules.—1882 SetofRules bySociety ofTelegraph-Engineers and Electricians.—In May, 1882, inthe first. setof“Roles and «Regulations forthePrevention ofFire Risks arising from Electric «Lighting,” issued bytheSociety ofTelegraph-Engineers, 10 standard ofinsulation resistance orofminimum leakage is recommended. All that we find there on this head is: “The «difficulties that beset theelectrical engineer arechiefly interoal “and invisible, andthey canonly beeffectually guarded against “by testing orprobing with electric currents;” “All wires “used forindoor purposes should beefficiently insulated,” &c.; “The insulation ofdynamo coils and conductors should be «practically perfect;”“Thevalueoffrequently testingthe «apparatus andcircuits cannot betoostrongly urged,” &e. Five years ago even, the writer isnot aware that any published rule existed, orthat any specification contained any definite rule fortheinsulation resistance ofany part ofany electrical installation. Admiralty Rule—About 1883 or1884, however, theAdmiralty officials requested that theinsulation resistance ofthedynamos supplied tothem should withstand thetest ofstroking thefree endofaconducting wire(theotherendbeingconnected toeither pole) overanypartoftheframing without showing anysigns of sparking when themachine was driven atitsnormal speed. ‘The writer does not know whether they still insert this clause in their specifications,orwhethertheyhavemodifiedorimprovedit; butatbestitisonly arough-and-ready test, andgives noindica- tion whatever astotheprecise locality ofafault oritsvalue in obms. 1988) ELBOTRIC LIGHTINSTALLATIONS. 0 Workman's Tests.—The only testapplied bycontractors when wiring aninstallation was, and often still is,that ofsimply inserting anordinary lineman’s detector orgalvanometer, G, between abattery, B(ofoneortwocells), andtheleads, with theother end orends freed, asinFig. 1.Ifthegalvanometer Pr Ome a\!@ia Foot gives nodeflection, oravery small one, then thesection ispassed asallright. ‘This issuch anexceedingly simple and handy testforwork- menthatitcanscarcely beimproved upon,butitgivesnoprecise indication whatever ofthe resistance inohms, ofthe dielectric, or ofthefittings. Itsimplyinformsthesbserver whether thereis abed fault ornot. Hecan also find out byitwhether the continuity oftheconductor iscomplete, byearthing thefurther end; and hecanalso ascertain whether there isashort-circuit or Pre Return Lead I® ime a Forward Lends } Fo notbetween anytwoormore “forward” and “return” leads by connecting uphisapparatus asshown inFig.2, In1884 Messrs, Wm. Denny &Bros.’ electrician used asmall portable magneto-electric machine andsensitive bell—the same as telephone companies employ—instead oftheabove-mentioned galvanometer and battery. Ifthebell rang, afault existed, and itsposition had tobefound out; ifitdidnot, then allwas vol. Xvi. 4 0 ‘THEINSULATION RESISTANCE OF{Jan.44, supposed toberight.He,however, inaddition tothismerely ‘workman's test, applied amore exact test byWheatstone bridge, andthusobtained theinsulation resistance ofthewholeorofany desired portion ofthecircuit andthefittings inohms. Professor Jamieson's Rule.—In 1884, whilst testing the electric light wires andfittings ofsome ships, thewriter firstdrew uphisrule, which hasbeen published inthelastthree editions of Munro and Jamieson’s “Pocket-Book ofElectrical Rules and “Tables.” Hehadpreviously (since 1881) tested, byWheatstone bridge, &c,, theinsulation resistance ofanumber ofland andship installations, andwasrather surprised atthe variety and lowness ofthedielectric resistance, forhis ideas ofinsulation resistance had been somewhat magnified byseven years’ experience with submarine cables. Atfirst, hecould notgetcontractors tocome uptohisrule, buthewasmuch gratified about this time bythe enthusiasm and interest which Messrs, Wm. Denny &Bros’ electrician tookinthematter,andbyfindingsuchgoodresults inhistests oftheir ships’ lighting leads, which results sofar surpassed thedemandsofhisrulethathedecidedtoadhereto it, Since then, hehas hadmuch less difficulty, and when care andattention tocleanliness injointing, tothefixing ofwires and fittings, aswell asinthemanufacture ofdynamos, areobserved, ‘hehasfrequently foundthattheinsulation exceeded thedemands ofhisrule, which isasfollows:—“A careful insulation resistance “test ofeach circuit, and finally ofthe whole ofthe circuits “(including allswitches and terminals, but.notnecessarilythe “Jamps) joined uptogether, should betaken bythe method «described above,* andnotpassed unless theresistance isatleast «equalto“19perlampforeveryvoltemployed. Theinsulation «resistance ofthedynamo coils should beequally good.” *SeeMunroandJamieson's Pocket-BookofElectricalRalesandTables” B17,atfoot.‘Thisruleappliesonlytothedynamo,leads,andfittingsfor incandescent lamps arranged onthesingle parallel system. Should thelamps ‘bearranged inparallels, with two orthree ormore lampe inseries ineach parallel, they count asbutoneforeach parallel, 193 ELEOTRIO LIGHT INSTALLATIONS. 8 Patintoformulashape,theruleis— R=kXR where R,=thetotal insulation resistance ofthewhole orany part ofthelamp circuits orofthegenerator in ohms5 k=econstant (10 =jy2=100,000w)foundfrom actual tests ofseveral well-erected installations; E=EMF. ofdynamo orinstallation involts; N.=numberoflamps(16-c-p.) oneachcircuitoron the whole circuit. The insulation resistance is therefore here taken to be diredllyproportional tothenormalE.M.F.ofthedynamo,and inversely proportional tothenumber of16-c.p. lamps incircuit. Thedifficulty ofascertaining theprecise length ofthemains, sub- using, leads, and dynamo windings ledthe writer tosubstitute thenumber oflamps incircuit forwhat might atfirst sight appear tobethemore exact term inthedenominator, viz., the leagth ofconducting wires employed. Although notstated in theabove quotation ofhisrule, thewriter usually applies the same test forthe resistance orpercentage leakage ofcurrent hetween the whole system offorward and return wires;also between thedynamo armature with field-magnet coils incircuit andtheframe orbody ofthemachine, aswell asbetween the sevies and shunt-magnet coils ifthedynamo isofthecompound orseries-shunt type. Thefollowing example andtable will make theapplication of this rule clearer :— Sappose that wehaveaninstallationof100lampsof16ep. ach, andthat thenormal difference ofpotential between the teminals ofthe dynamo is100volts, then theinsulation resist tee should not: beless than E RakX, =100,000.» 100” =100,000 =+19. a TE INSULATION RESISTANCE OF a8. Table ofInsulation Resistances required byJamieson's Ruls. . For16-c.p. Lamp Circuits and Dynamos. |tm|otme|ohm|im|otme 1|some|gsc|same|soeropn|xxom0 10|“zon|“ee|sean|men|sono |ammo|steam|aan|“smn|“asm| ©|tasom|ogo|Sng|am|sro| t=|tonom|aca|ioe|aon|opm so|“Sm|“Sam|So|ine|ina| im|mam|som|am|‘teow|"am sm|azo|sn|46,000|namo|5,000| so|ine|see|Scam|ana|nen ‘to|tnum|tea|Sto|sam|om 500 10,000 18,000|16,000 20,000 22,000 ‘The leakage allowed bythisrule inthecase of100-volt lamps, ‘using ‘64ampere ofcurrent perlamp, is c=;=topwnors =Too.ampereofleakageforeverylampincircuit. Eachlamp,however, requires“64ampere; consequently theleakage =q's part ofthetotal current. Or,thepercentage leakage =«ty=“0016 %/,, Since bytheabove rulethedynamo requires tohave thesame insulation resistance asthelamp circuit, theinsulation resistance ofthewhole cireuit, including thedynamo, ishalfofthatgivenby theformulaandtables,andthereforetheleakageisdouble,or seapartofthetotalcurrent, or35°/,=0032°/,. Phoenie Fire Office Rules—In May, 1888, inthethirteenth edition ofthePhenix Fire Office Rules, Mr.Heaphy recommends that “inany electric light installation inwhich thecurrent iscon- «tinuous, andhasanelectro-motive force of200volts orunder, the “insulation resistance over thewhole circuit inside anybuildings “should never run below 10,000 ohms. ...With currents of “1,000 volts theinsulation resistance should never drop below 189 ELECTRIC LIGHT INSTALLATIONS, 8 50,000 ohms, nomatter how wet the weather may be. For “alternate currents the minimum insulation resistance should be “twice theabove number ofohms respectively.” InNovember, 1888, inthefourteenth edition, Mr.Heaphy is more precise, and hehasadded atable based upon similar propor- tionstothewriter’s rule.* HisRuleNo.27readsasfollows :—*In “any electric light installation inwhich thecurrent iscontinu- “ous, and has anelectro-motive force of200 volts orunder, the ‘insulation resistance over the whole installation should not be “belowthefollowing :— “Installations of25lights a 500,000 ohms. » 5 vee 260,000 » 10 se 125,000, » 500, a 25,000 ,, » 1,000 see 12,500 yy “When thelights areproportionate between theabove numbers, “then theinsulation resistance should becorrespondingly pro- “portionate, ‘The insulation resistance ofveparate circuits ofthe “installation shouldalsobetaken,andshould beinaccordance “rith the above table. For alternate currenta the minimum “insulation resistance should betwice the above number ofohms “respectively.” ObjectionstoPhanizRule—The followingobjectionsmay teofferedinregardtothislastrule:— 1,Itties down acontractor toone and the same insulation resistance whether the E.M.F. ofthe installation lies anywhere below200volts.Surelya50-voltinstallation requires buthalf theinsulation resistance ofa100-voltone, and butone-fourth that ofonewith 200 volts ? 2.Itissomewhat doubtful (owing toaprevious note toRule ‘No.20)whether thewords “over the whole installa “tion”include thedynamo aswellasallthewires, fittings, andlamps? ‘itersraleforanECP.of126volts;ag,take100lampe:then2-1=s0000»28=sa5o00oe; ‘ndm00inproportiontothenumberoflamps—A.J. “ THR INSULATION RESISTANCE OF (Jan. 24, 3.Nodistinction ismade between circuits with one, ortwo, ormore lamps ineach parallel. 4,Are lampsof2,000cp.ormoreinparallel,andlarge incandescent lamps of500ep.otmore, which require strong currents, come under thesame general words, “any electric light installation inwhich thecurrent “iscontinuous,” justasmuchassmallincandescent lamps of,say,5c.p, which require only afraction of an ampere. 4question naturally arises here fordiscussion—Should not the insulation resistance insuch extreme cases bear some definite proportion tothecurrent? Westminster Fire Ofice Rule.—In 1888 the Westminster Fire Office issued asetofrules forguiding their officers inregard toelectric light installations, inwhich they demand that the insulation resistance shall inno ease beless than 150,000,000 ohms permile!!! Surely they have been blindly making extracts from.‘submarine cable specification!Nocontractoreouldbeexpected, under ordinary circumstances, tocomplete hiswork (including dynamo andfittings) tosuch ahigh standard. HM,Picow's Rule-—On November 7th, 1888, M,Picou com- municated hisrule tothe International Society ofElectricians, Paris.* Using thesame symbols asinthewriter's rule, itis— RakewhereRy=thetotalinsulation resistanceofthecireutinohms; &=a constant (500) found byexperiments with short lengths ofwire covered with three layers of cotton wound inreverse layers and pressed between metal plates; E=maximum E.MF. ofthegenerator involts; ©=total current generated orpassing through circuits inamperes; &=the resistance inohms ofthe whole circuit when the installation isinfallworking order. The differences between M,Picon’s rule and the writer's lie in a eeeredRavi,Bow33,188 res) ELECTRIC LIGHTINSTALLATIONS, thedifferent values given totheconstantk,andinthesubstitu- tionbyM.Picou ofthetotal current, C,forthetotal number of lamps, N,,inthewriter’s rule, ‘There isnodoubt anadvantageinthussubstituting CforN,,sincetheconstant ithereby expresses directly thenumber oftimes the insulation resistance isoftheapparent working resistance, éAnditsreciprocal, theproportion which the leakage current benrs tothe total carent. For, let¢=the leakage current, and adopting the same symbols asabove fortheother terms, then enEuEuG res «1¢_leakagecurrent,“B=6=totalcurrent We, however, consider that M.Picou’s constant, 500, isbyfar twolow toensure good and lasting work. Byit,acireuit of 1,000 lamps of16cp.each, requiring 100volts and ‘64ampere ach, the total insulation resistance ofthe circuit would only need tobe78°1 ohms, instead of10,000 ohms bytheauthor's rile(or5,000 obms ifthedynamo beincluded). The Society ofTelegraph-Engineers and Electricians’ Last Rule—In April, 1888, this Society issued itsvecond pamphlet of “Rales and Regalations forthe Prevention ofFire Risks,” and Rule 16states that “the insulation ofasystem ofdistribution “should besuch that thegreatest leakage from anyconductor to “earth (and inthecase ofparallel working from oneconductor to “the other when allbranches areswitched on,butlamps, motors, “&e,, removed) does notexceed sere part ofthe total current “intended forthesupply ofsaidlamps, motors, &e,; thetestbeing “made atthe usual working electro-motive force.” This rule includes allkinds and systems ofelectrical installations, anditis very neatly and concisely expressed. Some persons have, however, adifficulty inunderstanding how the yyy, part ofthetotal carrent istobetested fororascertained, andothersdonotteeexactlyhowthetestistobeappliedtodifferent partsofthe circuits attheusual working E.M.F. before theinstallation ispnt intoworking order. we {THYINSULATION RESISTANCE OP(Jan,44h Iftherule hadalso been putinto thefollowing form, with a table giving theinsulation resistances required forafewofthe more common electro-motive forces and currents (ornumberof lamps), itwould nodoubt: have recommended itself more readily tothose unaccustomed todeal with theform ofexpression used in therule. Using thesame symbols asbefore, wehave— sy Cac 8,5000°=°=Ry osRr=5,000§=ee which isthe form ofexpression used before, and most easily applied, ‘This gives constant tentimes asgreatasM.Picow’s, and about one-tenth that ofthe writer’s rule. Which ofthese constants ismost inaccordance with good practice? Table ofInsulation Resistances required bytheSociety ofTeegraph-Engineers and Electricians’ LaatRude,* vit;—Theleakageofanypartofaninstallation nottoexceedsdvyoftotalcurrenteupplyingthatpart; or,Ry=6,000 E/C. (EY[se[ee|vee[nae[ae inCheas, |Same lamps ‘lamps. acre” ‘Lape. :Ohms.|onm.|Ot|Obme|Ohm. 1|250000|823,000|400.000|so,000|ss,o0010 25,000 82,500 40,000 60,000 (55,0002|12800 as2s0|soo|2500|arco4|620|eis|10000|12,000|13750w|cao|eso|soo|roo|11000wo|200|2.200|4000|500|e50010|1,666 2,166 2,666 8,888, 3,066wo|1290|165|2000|2000|200sa ‘a|toss|nas|nee|ane400 ex|m2|100|10|1st0 wo|eo|‘so|000|1001,000, 200|825400 00 a Ifweincladethedynamoaawellasthewholeoftelamp,mowara3 ‘in the inmaation resistance ofthe whole invallation behalf ofthe 07 gore? 18) ELECTRIC LIGHTINSTALLATIONS. 8 Different Teate—The ordinary Wheatstone bridge test isthe handiest and easiest method ofascertaining theactual resistance inohms ofthe conductors aswell asofthe insulation resistance. Stould the insulation, however, betoohigh forthebattery porer and thegalvanometer, then thedirect deflection substitu- tionmethod may beadopted. These tests aresofully described inevery book ontesting, and are,besides, sowell known, that there isnoneed toexplain them here. The battery power need onlybe@fewLeclanché cells (10or12atmost) ifthegalvan- ometer used isofthesensitive-mirror Thomson type. When the installation hasbeen completed, then thewhole oftheresults may beobtained bymeans ofordinary ampere and rolt meters. First, join upthecirouit asfollows, inFig. 3:— sDriOdor Dibshd J Fi 3, Work theinstallation atitsnormal power and observe thetotal current, C,bytheampere-meter, AM,and the difference of potential,E,bythevoltmeter, VM.Thisgivesus& Second, join upthevoltmeter asinFig. 4,with oneside to earth, having previously ascertained itsresistance, r,ohms. Alsonote, C,ontheampere-meter, and seethat itisthesame as Iefore. SsSe Os Ow _— zoo Fe 4 Ifanyleakage toearth exists inthedynamo orinanypart of tbecircuit, adeflection onthevoltmeter, VM,willbeproduced 8 ‘THE INSULATION RESISTANCE OF (Jan, 24th, corresponding totheleakage current, ¢,passing through it,and iarepresented by,¢volts difference ofpotential between its terminals. Then,£=¢,thetotalleakage current. But©=&,theactual constantorcoeicientoftheinsulation resistance ofthesystem. Then, ifthisvalue isgreater than k,ourstandard constant, the installation may beconsidered allright asfarasinsulation resistance isconcerned. Ofcourse thisdoes notinform usifany leakage istaking place between theforward and the return system ofwires, orbetween theshunt and theseries windings of thedynamo. Toascertain theformer ofthese, disconnect allthe lamps andinsert thevoltmeter first intheforward lead, andsecondly inthereturn lead, asinFig. 5,andnote ifwegetany leakage {= Free FI VMLamps&MotorsPree BQWireop Fro. & current, ¢when thedynamo isrun atitsnormal speed. In regard toleakage between theshunt andseries windings ofthe dynamo magnets, recourse must behadtotheWheatstone bridge test.* Nodoubt some ofthemembers will suggest other practical tests, aswell asgive their experiences ofhow the insulation resistance ofinstallations withwhich theyhappen tobeintimately acquainted compare with thevarious rules herein mentioned, as ‘wellashowthey stood thetestoftime, wear andtearofevery- daywork, What ischiefly wanted istoascertain theminimum standard ofinsulation resistance compatible with good honest work andordinary careful usage. *Forothertestawhentheinstallationisatwork,seeMunro&Jamieson’s “Bleotical Rules and Table,” 6th edition, pp.215, 216. 1889.) ELEOTRIO LIGHTINSTALLATIONS, cy P.S.—Since correcting the proof-sheets ofthis paper the writerthoughtthatitmightbeinteresting andinstructive ifhe obtained the insulation resistance ofaninstallation that had been working forsome years, because members and others might say that some oftheforegoing rules were allvery well fornewwork, batwerequiteinapplicable tooldwork.Heaccordingly wentto- day(January 16th, 1889) toMessrs. Graham’s large East India offices and warehouses inCathedral Street, Glasgow, with his testing apparatus, and took careful tests ofthe insulation resistance ofthe whole circuit, with the results shown inthe following table :— Originaldateoffittingupinstallation aw 1881-82Date on which mostofthewiringwasrenewed 1888, Dateoftest(daywet—Glangow weather)... Jan,16th,1889‘Total namber oflamps incireuit of16ep.each 81BMF.atbrushesofdynamo,inwoe ones 108 Currenttotalfromdynamoforalllampsinamperes(about)160Insulation resistanceofdynamoarmatureandmagnet collealone,imohneaaeaevee 40,000 Inwulation resistance ofthe whole circuit, inelading ‘the dynamo, allleads, switehes, cut-outs, and Inwulation resistancetopassJamieson's rule,includingdynamoaswella8thewholeampcircuit,inokme19,217 Tngulation resistance topass Bociety of‘Telegraph- Engineers and Electricians’ rule (iftheir constant includes dynamo anwell asthewhole lamp circuit) ‘Thisoldinstallation (theoldest large incandescent light onein Scotland), which hasbeen infulldaily working order forover fireyears (since thewiring wasrenewed) when tested, without vaming oranyspecial preparation, andonavery damp day, gave better insulation resistance than that demanded bythewriter's rule,andoversixtimesthatrequired bytheSocietyofTelegraphRagineers’ rule(taking itinitsbroadest sense). Ttisworth mentioning that thePresident, SirWilliam Thomson, hastaken aspecial interest inthisinstallation from itsverycommencement. In1881,whensofewelectricians knew 0 THE INSULATION RESISTANCE, Ere, (Jan. 24, practically howincandescent lighting should becarried out,and when the Messrs, Graham were erecting their large new premises, theyconsulted SirWilliamThomson astowhetheror notthey could rely upon theelectric light. Upon hisrecom mendation they fitted their premises throughout with theelectric light, without introducing gas-burners into any oftheir offices and warehouses. The actual carrying outofthe present wiring ofthe building hasbeen entirely done byoneofthewriters students, who hashadcharge oftheinstallation forthelastsix years.TheMessrs.Grahamhavebeensopleasedwiththecare ‘and attention which hedisplayed that they have sent him to their new premises which they areatpresent erecting inOporto, Portugal, tosuperintend the building and put upanelectric light installation there. ‘The dynamo which thewriter tested, and which gave atotal insulation resistance of40,000 o,against framing anddead earth, isaManchester shunt-wound machine byMessrs. Mather & Platt, put down about two years ago toreplace the direst lighting ofthe building bythree ofthe oldest shunt-wound dynamos inexistence. ‘These three machines arenowemployed incharging accumulators which serve tofeed thelamps used in the darker parts ofthe building during thedaytime. The writer tested the insulation resistance of one of these old machines, and found that itgave forarmature and shunt oils just over 10,000 chms, asagainst theframe ofthemachine. Inaddition totheabove tests, that indicated byFig. 4was applied, and noperceptible leakage current wasobservable on thevoltmeter, .even although thesensitive coil, which reads to less than avolt, was inserted. Thewriter might give theresults ofseveral new installations which have surpassed hisrule, and which therefore farexceeded the demands ofthe Society's rule. Hetherefore thinks that theconstantofthelattermightberaisedfrom5,000to10,000 with perfect fairness, and with beneficial results totheelectrie lighting industry asawhole. wea] “piscussto. a TheSecretary: Thefollowing telegram arrived attheofficeTee oftheInstitution thisafternoon, addressed toProfessor Jamieson. Thave since telegraphed tohimtoknow ifImight read it,and behasreplied inthe affirmative. Itisfrom Mr. Geipel, at Blinbargh, andisasfollows :—“Tested Conservative Club to-day, ‘insulating dynamo andcircuit from earth. 140,000, orthree to “four times your standard. 65volts, 147lamps.” Thefollowing is letter received from M.Picou, Foreign Member, inconsequence ofhishaving hadacopy ofProfessor Jamieson’s paper sent tohim:— Couraoste Coxrimertate Eoitox, 45,Ruz pvPane, January28rd,1889. TotheSecretaryoftheInatitation ofElectricalEngineers,‘ReferringtoProfessorA.Jamieson’spaper,Eahouldliketomake ‘snmarkonthevalueF=500inmyformula.quite agree with him that this gure isvery low, and that inpractical intstral work higher values ofinsulation are easily obtained than those ‘malting from this formula, Ididnot intend togive thefigure that the‘anlationoftheinstallation shouldhave,tobeconsidered asgood,butratherenisimam standard, under which itshould beconsidered axfaulty, At ‘isminimwum the fault must besearched for and corrected. Itisgenerally personal judgment which decides whether insulation is ‘ctcaly good ornot. But the determination ofthe minimum insulation‘Slerwhichthereisriskoffireordestruction ofmaterial,offersaninteresttanimpersonal andabsolutenature,whichisofverygreatimportance.‘That limit may boadvantageously fixed bythe discussion ofProfessor daninsn's paper, andIam appy tohave, formypart, contributed inraising ‘Sindiscuasion, which will beextremely proftable toelectrical work, R,V.PICOU, Foreign Member. TheCuainwax: Perhaps Mr.Crompton willfavour’ theme ‘eeting with hisremarks onthis paper? Asamember ofthe Committee oftheInstitution which drew uptheelectric lighting nilesreferred to,hemay probably beable todefend theprin- plesupon which they were based. Mr.Croton: Itrustthatinthisdiscussion someofthewr... soanger members oftheprofession who areengaged insuper- iutending thewiring ofhouses may take aprincipal part. Its ‘aluewillbeincreased inproportion tothenumber offactssuch gratlemen canbring forward thisevening. Iconfine myremarks, a ‘THEINSULATION RESISTANCE, Bre (Jun.24th, Upton, therefore, toclassifying thesub-heads under which itmightbe convenient todiscuss Professor Jamieson’s paper. ‘The insulation resistance ofcomplete installations may be subdivided asfollows:—Ist, thedynamo; 2nd, the distributing apparatus orswitch-boards ;Srd,theinsulated conductors them- selves; 4th, thevarious fittings, such asswitches, cut-outs, and lamp attachments. Dealing with thefirst matter, Imust point outthat Professor Jamieson hasdealt with thesubject asifthegreat majorityof installations contained dynamo machines, andmostofhismethods oftesting relate tosuch installations, Now Imust point out that, thanks tothe increased number ofcentral stations either actually atwork orprojected, thenumber ofhouses wired without adynamo aretwenty-fold ofthose with adynamo, The total insulation ofaself-contained installation containing adynamo need notbenearly sohigh, and consequently need notbetreated 80carefully, asthat ofone forming part ofacentral station system, asitisobvious that aleak which might beofslight moment inthecase ofdetached installations may, onaccount of the addition ofmany other similar leaks, become ofserious moment, andexceedingly embarrassing totheengineer incharge ofthecentralstation. YourPresident hasjustsaidthatIam oneofthe members ofthe Committee responsible forthe rule contained inthe Institution’s pamphlet relating toinsulation. First,astothewording ofthatrule,whichProfessor Jamieson considers ismost obscure. Icannot agree with himonthispoint. Itissurely asimpler matter toprescribe that the leakage of current toearth shall notexceed fixed percentage ofthetotal current required forthe whole ofthe lamps forwhich the installation iswired, than tofollow Mr. Jamieson’s rule, which takes intoaccount varying E.M.F.’s; but!amdisposed toagree withProfessor Jamieson thatthe5,000th partallowedinourrule, although itsufficiently protects detached installations, isnot sufficient toprotect aninstallation forming part of central station, andIamquite prepared toadopt hissuggestion toreduce theleakage tolath part. . Returning tothequestion oftheinsulation ofdynamos, itis 110) DISOUSSION. 6 nolonger adifficult matter toobtain high insulation onevery Me. attofdynamo machines. Now that wedynamo makers havehadourattentiondirectedtothefactthattherearefarbetter | vwmishes than shellac, there isnodifficulty inobtaining any desired degree ofinsulation onthedynamo itself from thefirst, ‘moment that itiswound. Inthedays when weused shellac it ‘asimpossible toobtainahighdegreeofinsulation untilthe whole ofthe spirit contained inthe shellac varnish had eraporated. Thereareotherobjections toitsusewhichIneedotparticularise here, asitisforeign tothe subject under diseussion, but which have forced usinto the use ofother vwmishes having @high insulating quality even while theyare | stillmoist. \ Omthesubject ofthedividing apparatus forswitch-boardsI ‘ lavelittletosay.Ashasbeenbeforepointedout,theuseof | ‘ute, unless very carefully selected, has been #very common scarce oftrouble; bat this difficulty can beavoided, ashasbeen already pointed onthere, bycarefully selecting theslate, orby coating itwith paradin orvarnish. Nest inorder comes the insulating wires themselves. ‘The manufacture ofthese hassogreatly improved during thelast fewyears that theinsulating resistance ofthecovering isso high,eveninthecaseofsecond-class wire,thatIthinktheyleave litte tobedesired;sothatthereisno-excuseforanycontractor usingawirehavingacoveringoflowinsulatingquality.When Ispeak ofsecond-quality wire, Imean wire ofwhich the insalation isofpure rubber,asmostmakersvulcanisetheirfirst ‘quality. Solong asourcoilofwire remains uncut, theinsulation maybemeasured bymegohms; itisonly after ithas been cut, intolengths, jointed up,andconnected ontothevarious fittings, ‘thatthepracticaldifficulty ofensuringhighinsulation commences.ThequestionofobtaininggoodjointsIwillleavetootherswho treeverydaydealingwiththissubjectinapractical manner; batIwishtodwellsomewhatstronglyonthefourthsub-head, viz,theinsulation ofthefittings thatweareforced toemploy. Tsay“forced toemploy,” because theuseoffittings made of glass, glazed porcelain, earthenware, slate, and similar materials o ‘THE INSULATION RESISTANCE, Bre. [Jan 4, MScpton, basbeensostrongly preseribed bytheInsurance Offices tatwe™=P*™ havenooption inthematter; butIwould pointoutthatthe useofthesefittings,havingahighlyglazedorcoldsurface,hence liable tocondense on tothem afilm from our moist London atmosphere, renders itanextremely difficult matter toobtain« high insulation toearth foralarge London installation. Solong ‘aswewere able tousewood which could berendered practically fire-proof, and could bevarnished insuch amanner that itdid notcondense ontoitthese films ofmoisture, itwas acompars- tively easy matter toobtain ahigh insulation, and thequestion arises whether wehave notpurchased theadvantages wegain by using these fire-proof fittings attoohigh acost. Nodoubt for some situations their use isabsolutely necessary, but inmost other cases their advantage isquestionable. Astothemethods oftesting proposed byProfessor Jamieson, thelineman’s detector, ifused with asource ofpractically constant EMF, andcarefully calibrated with that E.MLF. soastogivea certain deflection foreach fraction ofanampere passing through it,isquite themost convenient instrument forusebyordinary workmen orforemen, Such aninstrument isquite asportable asthevoltmeter heproposes, and isgenerally wound soastobe much more sensitive. When such an instrument isused for testing dynamos, as,forinstance, atourown works, weuseitwith thestandard E.M-F, of100volts ofourlighting circuit, andonly pass ourwork ifthedeflection isbelow acertain limit, which limitisdependent onthemagnitude oftheworktested;but when testing installations connected with acentral station, itis obvious that wecannot use the E.MLF. ofthe mains, asinthat case theexisting leakage onthemains would trouble us. For such work, therefore, itisbetter toprovide theconstant source ofEMF. intheshape ofaportable hand dynamo machine which canbedriven atacertain speed, and, consequently, give the EMF, within therequired percentage ofaccuracy. AtKen- sington weuse aportable battery ofaccumulators. Messrs. Siemens, ofBerlin, make theportable dynamo apparatus men- tioned above, andIbelieve itisused universally throughout the Berlin central stations. Ihave also seen itused inother Continental towns. 1088 DISCUSSION. 6 Inconclusion, Iagree withProfessor Jamieson thatitisquite Xe. easy, ifdue precautions aretaken, toobtain excessively high insulation oninstallations ofconsiderable magnitude. Inone case,inLondon,wherewelindtoworktoaveryseverespecifi- ‘ation—which hasbecome rather celebrated through adiscussion which took place onitinthetechnical journals—we obtained a resistance of439,000 ohms, which isabout four times inexcess oftherequirements laiddownbyProfessor Jamieson's rulefor thenumber oflamps, viz., 130. Inother cases wehave got results sixfold inexcess ofthat required byProfessor Jamieson, Itisunnecessary toquote these. Mr.ALEXANDER Stemexs: Idonotthink Ihavemuch tosay,Mr. butmyideas ofinsulation aredifferent: from those which have teen brought forward. Iwill not say anything ofour own installations, but wehad toexamine aninstallation ofabout M11lights because acentral station inLondon raised thenestionofinsulation, andonmeasuring theinsulation resistance wefound itsomewhat over 5megohms, and the insulation reistance ofthe dynamos—I forget the maker—was over 1 megohm. This isvery farinexcess ofany ofthese rules, and vassomething like what itshould be. Ithink Professor Jamieson iswrong when hesays that theresistance does not require tobenearly sohigh perunit oflength asinthecase of submarine cables. Why should webecontent with amuch lower inmotion resistance inthese installations than insubmarine tables? Idaresay wecannot getitquite asgood, butweshould ttytogetasnearaspossible. TheCuamaax:: There mustbeagreatmany members present the hohave hadexperienceinthissubjectofinsulationresistanceof Aectriclightinginstallations. Ofcoursethereisnodefiniterule tien byNature astowhether thestandard should be5,000 ohms ‘5,000,000, oranything else. The insulation resistance asked {ornsubmarine cablesissimplytheresultofyearsofexperience‘towhatengineerscangetbymoreorlessfightingwiththe ‘eatractors;andconsequently, inthesamesortofway,the jasulation that will be obtained for the house installations will avetobesettled byactualpractice. Therefore theresultsof You. xvut. 5 6 THEINSULATION BESISTANOE, Bre.(Jan.44h, Beeman, Xperience arewhat weparticularly want;infact,Ipresumethe object that Professor Jamieson hadincommunicating thispaper was toelicit information rather than togive it. Ihope, there- fore, that some ofthe members whio have had experience will continue the discussion. Mr-tuman. Mr, H.Human: Iwish just tocorrect anerror that the author hasfallen intoinreferring totheWestminster Fire Office rules. The same error was made byMr. Preece last year in dealing with this question, and onmycalling hisattention toit, headmitted that hehad misunderstood the rule. The ruleto which theauthor refers asshowing that theelectrical resistance should be150 megohms per mile, has noreference whateverto theinsulation resistance ofaninstallation asawhole. Itsimply refers totheelectrical resistance ofthedielectric orcovering of theconductor. Now theauthors ofthose rules were very careful togettheopinions ofexperts onthis question ofinsulation. ‘They first ofallthought ofdrawing uparule defining what should betheleakage allowed inaninstallation, basing itupon theE.M.F.; buttheywereadvised,andIthinkwisely20,to leave the matter alone, and letthe electricians themselves deal with it,they having agreater interest inthesubject than theFire Offices. But having specified what thenature ofthecoating should be,itbecame also necessary tospecify what itsminimom electrical resistance should be,and wefound that inpractice « cablehavinganinsulation resistance of150megohms permile ‘would, innine cases outoften, besufficient. Therefore this 150 megohms, asreferred tobytheauthor, hasnothing todowith thesubject heisdealing with. Ifeel bound,onthepartofthe authorsofthoserules,tomakethiscorrection. Astoincombustible bases, Mr.Crompton truly remarks that theFire Offices were originally satisfied with wood bases,andofcoursetheresultsweregoodaofarasmereleakagewasconcerned; butImay tellhimthisastheexperience ofthemajority ofthe Fire Ofices—that whenever afailure happened oranaccident occurred itwasinvariably through those fittings, i.e.,either with regardtothecut-outsortheswitches, Wefound,therefore,that itbecame absolutely necessary toprohibit what was then the 18a DISCUSSION. or cammon practice, ofputting them upon very inferior wooden ur.Haman blocks,andwethoughtthatthebestthingunderthecircum- stances was tospecify forincombustible bases. Well, Iam pleased tohear from Mr.Crompton that, inonecase atallevents, hehashadnodifficulty inobtaining, even under these conditions, reryhighinsulation. Wehavegainedbyexperience inthesematters,andwhenwefindanaccident, orthecauseofanaccident, vemakeanoteofit,andarecarefulthatinfutureitshallbe povidedagainst.‘These'arethereasonstheFireOfficesinsist ‘ponincombustible bases tofuses and switches. IfIremember rightly, Mr.Cockburn lastyear, inhisadmirable paper dealing ‘ithfuses, pointed outthat itwasobjectionable inmany cases to ‘wewood ;that intheordinary fuse, where there wasconsiderable latingofthefuseitselfbeforeabsolutebreakage, itwasinsome cases possible toignite the base. These are probably the aception, butstill they areliable tooccur, and itisonly right istweshouldprovideagainstthem,Itmustnotbeforgotten thstwhen anaccident does happen someone has to“pay the “piper,” and theOffices arejustified inprotecting themselves sguiatsuchcases.ButIamquitecertainthatnoonewould ‘necessarily impose conditions that would retard theintroduo- timoftheelectriclight.Thefactis,wehavefaithinthe deetric light, andweareconvinced that inthefuture weshall derive great benefit from it,Atpresent,ofcourse,ourexperience islimited,andwecanonlyjudgebyresults,Wearestil,asit, were,suffering from theuseoftwomodes oflighting—that isto ‘ay,bygusandelectricityincombination, Ourexperiencehas beenvery good 20faraselectric lighting alone isconcerned, but ithasnotyetsufficiently gainedtheconfidence ofthepublicto ‘aable them todispense with other forms oflighting. Iamnot, course, speaking ofitsdanger, butofitsstabilityorreliability. Thepublicstillkeepgasasastand-byincaseofabreakdown, ‘iththeresultthatthegasfittingsarefrequently eitherputtoim- Poper uses orneglected, andthejoints become impaired. Thus, ‘henithappens—as unfortunately itwillhappen attimes—that )thereisatemporary breakdown oftheelectric light, andthegas |fittingsareagainbroughtintorequisition, aleakagemayoccur, \ 6 ‘THE INSULATION RESISTANCE, Br. (Jan. 4th, Yr.Hamas. with aprobable explosion following. Sothat wearesuffering in- directly insome respects from the useofelectricity forlighting purposes. When, however, public confidence isrestored—which dependsentirelyuponelectrical engineers—and whenotherforms oflighting areentirely dispensed with, then weshall doubtless better appreciate theadvantage oftheelectric light than wedo ‘atpresent. ewes, Mr. Syowey Evensuep: Ifully agree with Mr. Crompton that theauthor's treatment ofinsulation testing israther archaic. Inthefirstplace, onecannot carry amirror galvanometer about foruseinhouseswhicharehalf-finished. NearlyallthehouseswhicharenowbeingwiredinLondon areonlyjustbuilt,andin many cates thewiring and internal fittings arebeing putin together. Iinspected some eight ortenhouses afewmonths agowhich were completely fitted with electric bells andspeaking tubes, gaspipes, water pipes, and electric light wires. Allthese thingswereburiedintheplaster. ‘Thisseemsverybadpolicy 80farasthe latter areconcerned, considering the effect damp plaster will probably have ontheinsulating material ofthe cables. ‘Mr.Alexander Siemens made some objection totheuseof8 voltmeter intesting. Ofcourse themethod described inthe paper looks very pretty inadiagram, where thevoltmeter isput ontoan apparently good earth;inpractice,however,itwillbe difficult tofind good earth. But Mr. Siemens was mistaken when hesaid this test would burn thevoltmeter coil, foritean- notpossibly getmore volts onitthan thedynamo willgive, and presumably itwillstandthat.However, ordinary voltmeters are| useless forthis method,astheirresistance is0lowthat,yougetnoreadingfromlinetoearth.IimagineProfessorJoin| was thinking ofaThomson graded voltmeter, which bas resistance ofperhaps 10,000 ohms. Very small leaks canbe measured with these voltmeters, butIneed hardly saythat 20 such instruments are made sufficiently portable fortesting work, YSeenue,_-MrsW.Lan Canrenren: With your permission, Sir,Ishould | liketoputaquestionandmakeasuggestion.Aswehave"| ws] DISCUSSION « many gentlemen here to-night, Iwould askwhether there isus. anywell-authenticated instance ofamelted tinwireinasafety “™"™“"" fuehavingactually causedafirebyinflaming woodoranythingofthekind?‘Thesuggestionthatoccurstomerathersupple- ments what you, Sir, and Mr.Crompton have said—that there maybemanygentlemen herewhohavehadwideexperience of tating insulation resistance, and who may perhaps feel some delicacy ingiving their experience because they may feel, to tome extent, that their information belongs totheir employer nither than tothemselves. Ifthere beanysuch here to-night, I ‘would venture tosuggest tothem that anyinformation that theytaybedisposed toputattheserviceoftheInstitution might quite well begiven without aname, orother means ofidentifica~ tionofanysort orkind, being mentioned inconnection with it. Itwould bevery valuable totheInstitution, and itisagreat pitythat weshould nothave it,more particularly asIfully agree vithProfessor Ayrton’s remarkthattheobjectofthispaperwas quiteasmuchtoelicitinformation astogiveit.Iknowthat thatwasthe opinion ofthe Council inlooking over the paper aadinspeaking ofitdownstairs just now, and Itherefore venture tohope that weshall have some valuable information given, |Mr.W.B,Eason:Ishouldliketocallattention tothetwous.asm,‘ublesonpages52and56,whichseemtobeofanentirelydifferent, claracter, InJamieson’s table, page 52,youwillobserve that the insulation resistance issupposed toincrease with thevoltage, and thatthepercentage ofcurrent leaking goes upwith theE.M.F.; hileintheSociety'stable,givenonpage56,theidealcon- titionofaconstantpercentage ofleakageissetforth.Inactual installation work forincandescence lamp circuits neither ofthese ‘onditions obtain, theinsulation resistance being,asamatterof fue,approximately thesame whether theE.M.F. is50,100,or200 volt,andthepercentage ofcurrent leaking increasing, therefore, asthe aquare ofthevolts. Mr.Benwanp Daaxe: Ishould like, Sir,tomake aremark inxr.ris rlerencetoMr.Crompton'sopinionofpotteryforbasesofcut ts,Heisundoubtedly right insaying that leakage ismore 7 ‘THE INSULATION RESISTANCE, Erc. (Jan. 2th, ar.Dake,generally duetoswitches andcut-outs thantothecable,butI maintain that the Insurance Companies arequite justified in insisting onpottery orincombustible materials. Ihave seen several bases ofwood with unmistakable signs ofburning from leakage through moisture absorbed bythewood. Now this could notresult with pottery, the only objection to whichiscondensation andsurface leakage. This,Ibelieve, will befound tovanish ifthecurrent isputonforanhour previous totesting forinsulation. Insupport ofthis Imay mention that onarecentoccasionIhadtotesttheleadsatLordArmstrong's private house, and found apparently adead earth onthe mains. ‘These consisted ofbare copper rods supported onpottery slabs placed in wooden trough which waslying ontheground. I believe thissystem wasdesigned byMr.Campbell Swinton. ‘Thepotterywasfoundtobewet,andtheleakageevidentlyon the surface. After, however, thecurrent had been used forabout tenminutes, westopped, and were surprised tofind that the leakage had practically vanished. Itwould beinteresting to know whether this wasduetopolarisation orthedrying of portion ofthepottery near the conductor. Ifthe same takes place with thepottery bases, there isnopractical objection to them, butthefinal tests forinsulation must bemade after the ‘current has been on. There isamuch more serious question than this towhich I woulddirecttheattention ofFireOffices,andthatistheplacing ofwires innew houses under boards when thepugging isnot dry. Aprocess of“sweating” then takes place, which will ina month rotofftheinsulation ofthebest wires made, and there is then aleakage through thedamp casing. My own opinion is that youshould give your wires alltheairpossible under boards, andplace them well apart onthesides ofthejoists, butnotin casing. Another most fertile source oftrouble towhich noallusion has yet been made isthejointing ofthecables. ‘The usual course adopted istosmear thebared copper with Chatterton's compound andwind itupwith waterproof tape ;buttheinsulation will befound tobeweaker here than elsewhere inthe cable, 108 DISCUSSION. a spart from thedanger ofthecompound cracking andmoisture a.Date. penetrating thejoint. This matter should be-faced bythe cable makers intheir own interests, and they should insist, fortheir omcredit, onaspecified method ofjointing, and, ifpossible, provide thematerials themselves foruse with their own cables, Mr.GuyC.Farceer: Withregard totheinsulation ofslatexr.basesandfittings,Ifindthattheslatemaybegreatlyimproved inthisrespect bysoaking itinmelted paraffin wax, andthat the filmsofmoisture mentioned byMr.Crompton donotthen appear tobetroublesome. Ithinkthatthisplanmightwithadvantage beextended tothetreatment ofunglazed earthenware fittings, which might then insulate better than ifglazed. Mr.Crompton wasalsoratherharduponshellacvarnishasaninsulating com- : pound. Itisquite true that shellac varnish will notdryproperly ifitisused oneach layer ofwire when winding upacoil, unless itisaerwards baked foravery long time; butinwell-ventilated amatares itisaperfectly good varnish touse, and Imay sythat Ihave never used anything else. Inregard tothe inmlation leakage ofcables, itseems tomethat, speaking Mtrictly, theinsulation resistance should beinversely proportional tosurfaceofthecableratherthantothenumberoflampsfed byit,as,Ithink, isproposed inthepaper. Although thequantity ofcurrent carried byaconductor should beproportional tothe cooling surface, bytherules under which electrical contractors ‘ually work, itisgoverned bythecross section, andtherefore it would seem tobebetter tospecify definitely insuch away asto curure against more than acertain current-density ofleakage ‘trough theinsulation. Mr.W.A.Cuamex: Perhaps,Sir,Icanofferafewremarksue.hich may beuseful. Icansayfrom experience that there isa great deal ofdifficulty ininsulating the backs offittings, more tatticalarly ofdouble-pole cut-outs, which wehave been almost, forced toputinbysome oftheFire Insurance Offices. These ‘bringthetwoopposite polesofthecircuit closetogether, thus fering facilities forleakage between them. Insome FireOffice niles itisrecommended that the cables orwires should be Hattered inthewalls; Ithink insome cases Portland cement is 2 ‘THEINSULATION RESISTANCE, Erc. (Jan,940, MSsaep, Mentioned. Someyearsago,attheForth Bridge, Ifound that Portland cement was absolutely useless. ‘The insulation was allright forsome time—I think about atwelvemonth ;butafter that time itwas completely gone. Portland cement certainly seems tohave some deteriorating effect upon insulation. Plaster ofParis hasbeen often used, and indryplaces, perhaps, thst isallright; but inordinary circumstances Ithink itvery undesirable toplaster leads inatall, Where conductors pass through awall, particularly from onerisk toanother, some of the fire inspectors insist upon their being plastered inor cemented. What Ihave always done insuch cases, however,i toput inporcelain orglass tubes, or,better still, aniron tube lined with glass, andtoplaster round that, but onnoaccount to putany plaster next totheconductor. Referring tothefusing ofatinwire, orother cut-out wire, causing thecharring ofawooden base, Ihave never known such8 case. Ihave seen wooden cut-outs charred, but Ihave always been able totrace this charring either toacopper wire having been inserted through somebody's carelessness, orelse toanarc having been formed. The latter, ofcourse, cannot beguanied against byany form ofcut-out, and forthat reason itmay be desirable tomake the bases incombustible. With regard tomoisture, Icanquite understand Mr.Drake's remarks, although Irather think that thecable used incasing underneath thefloor cannot have been sound ifanyamount of dampness lying between thetwocables caused leakage. Ithink that the insulation ofmost ofthe cables which are now used is such that they may belaid under water without any serious leakage;atanyrate,thoseusedontheGrosvenor Gallery installations might. Ithink, however, there isasource ofdanger even inusing such thick insulation, especially when itisvul- canised rubber, for itiswell known tothose who have the practical handling ofthese cables that where acable hastogo round asharp corner theinsulation isoften broken. After & cable has been made some little time vuleanised rubber seems togethard, andbreaks more easily. Ithink, however, that pure rubber cables, taking them allround, stand abetter chance than 1800) ‘DISCUSSION, 18 thevuleanised ones. Icannot saywhether after anumber ofMx.years,providedtheywereproperly andcarefully laidinthefirstple,thepurerubberwouldprovetobebetterthanthe valeanised, forIhave notlived long enough orhad thelength ofexperience tofindout. Respecting joints,IquiteagreewithMr.Drake,butIhave longagodiscarded“Chatterton” altogether. Ifoundthatthe nenwouldplaster“Chatterton” aboutthejointwithouthavingit properly melted, and leave holes init. Ihave forsome time pastused india-rubber strip, which islaid upround the cable, aousing india-rubber solution forthepurpose ofsticking it together. If joint iscarefully made inthis manner Ithink itisperfectly good; itmay notbebetter than “Chatterton,” provided the“Chatterton” joint isproperly made. Itismost, iificult tokeep alight, even aspirit lamp flame, burning in ‘many places, forthere isoften draught enough toblow itout. Whenworking inhousessuchasMr.Evershed referred tojust tow,where there arenowindows ordoors, you cannot doany- thing atallwith alamp. Ihave actually had aman bind up ‘joint without soldering it,simply because hesaid hecould not keephisironhotenough. Ofcoursewiththeindia-rubber and ‘elation noheat isnecessary. After theindia-rubber strip isput ougenerally useaserving ofwaterproof tape. Idonotmean thatthesejointsarealwaysperfect,becauseIbelievethatnine atoftenworkmen, ifthey arenotwatched, will make bad ‘pints, Mr.H.C.DonovansaidthatexperiencegainedinonebranchMr... ofelectrical construction must ever beuseful inanother, and most especially inthemeans tobeadoptedtomaintainagood ttateofinsulation inmains,leadingsndconnecting wires.He masofopinion that toensure aconstant insulation inelectric lighting installations thechoice andquality oftheinsulating materials were not ofsogreat importance asstrict attention to (tail,s0astopreventunduelossandagradualfallininsulation, due,nottothelossofcurrentthroughtheinsulatingmaterial, buttowhatisknowntomanyassurfaceleakage—that is,the current escaping from thebared ends‘of theconductors inleading " ‘THEINSULATION RESISTANCE, Ere, [Jan.4, ar. andconnecting wires,alsocontact-pieces, plugs,switches, andcom- P=" rmutators, which arefitted onslabs, pedestals, andbrackets ofthe well-known materials used forthis purpose. Leading wires, whether insulated with gutta-percha orthe different forms of india-rubber coatings, aresure tofalloffininsulation when the ‘endsbecomecoatedwithdustconsisting largelyofcarbonparticles (especially inlarge towns where coal isused). This coating ofdustbecoming wet,asitissuretobewhentheatmosphere isdamp, itcanreadily beunderstood that asource ofleakage is setup,first through the moist dust and dirt, and thence tothe damp tape andfeltontheouter surface ofthelesding wires to “earth.” The same remarks, toagreat extent, apply tothe insulating base ofcontact-pieces, plugs, switches, and commu tators, When new they have perfect insulation, but intime,if they benotkept clean, damp dust willdoitsevilwork, conduction willtake place over thesurface, andrender thebest ofinsulating material useless asregards thepurpose forwhich itwasselected. Forthesame reason toomuch confidence should notbeplaced inparaffin wax. Itisagreat dust collector, andconsequently itssurface may become amoisture absorber. ‘The moral ofthese remarks isthattheends ofallleading wires, alsothesurfaceof allinsulating slabs, &.,should bekept clean. Itshould bebome inmind byallelectricians that theclimate oftheBritish Isles, except from February toJune, isexceptionally damp. Ttisnatural toseethat allthese precautions areofmore importance totheconsumers than tothecontractors, asthelatter aresure toget ridoftheir responsibility before thetrouble commences. ‘When thelate Professor Varley first visited theUnited States, hewasatonce struck with theutter disregard electricians there had toallinsulatory precautions—for instance, brass keys and fittings screwed direct ontowooden tables and benches, single cotton-covered wire used forconnections everywhere. Hepro- claimed, much tothesurprise ofthe Americans, that there was noinsulation ;butthere was,forNaturehadprovidedtheAmerican electricians with something more useful than indin-rubber oF gutta-percha, more insulating than paraffin wax—an exceptionslly 189) DISCUSSION. w ‘ryatmosphere. ItisthesameinAustralia, wheretheclimateus,isverydryforthegreaterpartoftheyear. Donovan. With regard tothemaking ofjoints, itshould beborne in nndthatitisessential thatthejointer,beforeheiscalledupon tomake ajoint, should wash hishands. Very often hehasother worktodo,andwhencalledupontomakeajointhishandsare ‘egrimed with every form ofmislocated matter, often largely ‘cusisting ofoilandbrass filings. Hefully admitted that this ‘masamatterofdetailwhichatfirstsightmightseemanitemsfnoimportance.Mr.W.E.Guay:Ishouldliketosayafewwordsonthisar.ory. feper. Mr. Jamieson, inreferring tosubmarine. cables, says thatthe engineers insist that these cables should give an insulation resistance of300 megobms, fourteen days after manu- ficture, and after two minutes’ electrification. Iamafraid this ishardly complete, ashedoes notsaywhat number ofcells areto- weused,andthetimeofimmersion. Forelectriclightinstalla-tionstheconditions ofwork aredistinctly different. Tpresume insuch installations one does not care somuch nerely about theinitial resistance inohms ormegohms, asabout thepermanent resistance—that istosay,aninsulation resistance ‘atshall remain always about the same, This may bealow teaistance,astheamountofleakagewouldprobablybeinconsider- ahle,thetests being taken merely toshow theactual state ofthe Gireuit. Mr.Crompton says hefinds that when using slate or prcelain casing there isadeposition ofmoisture onthesurface, tadthat ifwood casing were used instead, asecond-class wire coald alsobeused—ice,, that hedidnotdepend ontheinsulation ofthewirefortheinsulation ofthewhole circuit, butrather on ‘beinsulating qualities ofthecase. Insuch aclimate aswearefavoured with here inLondon, this, \think,isratherrisky.Anothergentlemen—who, unfortunately forme,Ldonotknow—saidthathebelievedincableswhichcould beusedunder anyconditions ofclimate, and, therefore, preferred ‘uchcables aswould testinwater. ‘This seems only reasonable, if Yeare tomeet with any great difficulty, due tomoisture, inin- stallation work. Hethen referred tovuleanised rubber, andeaid 6 THEINSULATION RESISTANCE, Fre. (Jan.4th, Me.ny. thatinbending thisround corners hefound ittocrack. This seems rather peculiar, asrubber issupposed tobeelastic; but probably hemay refer tothecracking ofthetapes, which arenot rubberatall.Hesaid,Iunderstood,thathehadnotlivedlong enoughtoknow whether ornot vuleanised rubber would last longerthanpurerubber. Well,Imaynothavelivedlongerthan hehas, but Ithink hewill find, asothers have found, that valeanised rubber will last longer; and this is,Ibelieve, the general ideaamongst manufacturers, andtheprinciple onwhich they work,astheyactuallyusevalcanisedrubber,andnotpure rubber, which they know willnotlast. Mr. Drake referred tothemaking ofjoints, and spoke of ‘using Chatterton’s compound forthem. Itseems rather abastard cutofthing todo,toputgutta-percha ontoanindia-rubber insulated cable, ascertainly, ifanyincrease oftemperature takes place, thejoint will leak. There isclearly nodoubt ofthis, owing tothedifferent rates ofexpansion ofthese materials, and Chatterton’s will not stand anything like thesame degree of temperature asvulcanised rubber. Tomake ahomogeneous joint, pure rubber tape tightly wrapped onand smeared with india-rubber solution, ashasbeen suggested, would probably be byfarthebest;andiftightlywrappedon,andtheworkman's hhandbeclean,thejointbeingproperlymade,thereisnodoubt thatsuchajointcantestinwater.Afteracertaintime,this joint, from itsown defects—i.e,, therubber starting slightly to decompose—will simply become ahomogeneous covering round theconductor joint. Itistrae that this willnotlastsolong ss thecableitself,butIsupposeittobethebestthatcambedoneunder the circumstances. ‘You, Mr. Chairman, made aremark that Ididnotquite understand. You,Ithink,saidthatyouhadnodoubtthe engineers could wring from the contractors agood quality ofmaterial. Itseemstomethatnoengineer canwringfromacontractor thatwhichhecannotmake,Ifhecanmakeafirst- class quality, hetells theengineer that hecan,andtheengineer afterwards specifies forit;but really agreat deal inthis installation work must depend onwhether ornotthecontractor 18) ELECTIONS, n isagoodone.Itisnogoodgivinganempirical sortoffigure™.ory. forinsulation resistance perlamp unless you arecertain that the insulationwilllast;andthat,itappearstome,isthegreatpoint. Tam afraid that afigure taken upon afavourable occasion, as pointed outbyMr. Crompton, oronanice dryday, really can lardly beassumed bytheFire Insurance Companies tobesuch aminsolation asthey require. They must gofurther, and examine into the class ofmaterial that hasbeen used; and ifthis material isbad,orunsuitedforthepurpose,althoughforthe timebeing itmay give ahigh insulation and appear tobeall right, still, inmyopinion, they would bequite justified in rejecting it. TheCaamman: Beforeadjourning themeetingIwilljustre remark thatIhavesohighanappreciation oftheimproving ™"™* porerofcontractors thatthespecification Ishouldmakeinall caseswouldbe“‘alittlehigher(whatever thethingwas)than “theyhadbeenhithertoabletoget,”andIamsurethatthey ‘would then beable togetwhat Iwant. Tam toldthat there areseveral members who wish tospeak omthis subject, and the discussion will beadjourned until February 14th. Abullot took place, atwhich thefollowing were elected :— Members: Alfred Upton Alcock. |Jorgen Henry Ferdinand Toll. Associates : Frederick FrancisBennett. |CharlesCesarHawkins.WalterR,Cassels. |WilliamJohnThorrowgood. ‘Studenta: Donald Frank Adamson. Edmund A.Hall.FrederickRobertConnell.|ArthurWilliamRankin. Archie Davidson. Joseph D.Rolls. ‘Themeeting then adjourned. ” ABSTRACTS. W. KOHLRAUSCH—LIGHTNING CONDUCTORS (Bletrtechniucha Bata, Vol,9,9.228and297,1888.) ‘The author first deals with the vexed question ofusing gas and water rainsas“earth,”andthenpamesinreviewsomesimpleapparatusfortexting theresistance oflightning conductors, ‘From actual experiments hehasfound that theresistance ofan“earthformedbygasorwaterpipeisleesthanthatoftheusualearth-plates, Iaallbuildings protected byaaystemoflightningrods,theconductors mastia somepartoftheircoursecomeintoproximity tosomeportionofthemajl pipes; and since theresatance ofthe“earth "of theconductor igreater Un that ofthepipes theelectric current, even though itmay forsome distance traverse the condustor, has atendency toleap across tothepipes, Theres, therefore, every reason forennecting theconductor from thelightaing radto thegatandwater pipes. The way inwhich thejoint between theconductorandthemaininmadeisnotofgreatimportance; thetwoessentialpointaare ‘thatthetwoshallbeinthoroughlygoodmetalliccontact,andthatthejoint shall have fairly large surface, ‘The connection may bemade atany cov-venientspo,providedtha,ifthpipesareofion,thesectionofmetalbetweenthejointandthe“earth”shallnowherebelessthan100squaremillimetres (01851q.i0.);andifoflead,Svetimesaslarge,Meter,whetherguscorwate,shoulbeshort-circuited by'slengthofatrandedcopperwire,‘The nature ofthe joints inthe pipes isofrome importance; but actaal ‘experience hasproved that therein norisk ofdanger tothe pipe ifthe joints aresoldered, screwed, flanged,orspigotjointsfilledinwithlead.Inthetown‘ofHanoveragreatnumberof“earths”madebymeansofthegasandwaterpipes have existed forover ten yours, and nodamage hasresulted from their ‘use,Thereisnotmuchtochoosebetween gaspipesandwaterpipes,though 4gengineers have raised objections totheuteoftheformerontheground oftheresistance whichmustbeintroducedatthejointabythenseofre led. Someactualmeasurements areofinterest.Thefratserieswascarriedout {intheHigh School atHanover onthegaspipe inthe building, some ofwhich ‘had been put upseven years previous tothetimeoftexting,andothersthree years, Tho total lengthofpipetentedwas76metres,inwhichtherewere117 Joint,allmadewithredlead,Thetotalresistancewasfoundbyexperiment todeO12ohm;theresistance ofthepipetherr-Ives,ascalculatedfromtheir sectionandlength,was0-086ohm,thusleaving084ohmastheresistanceof ‘the117jointsoFsmeanof0-0008ohmperjoint.Itappears,therefore,thatscrewed joints iniron pipes made with red lead give asu@lciently good metallic contact. ABSTRAOTS. 79 ‘Thesecondseriesofexperiments wascarriedoutonlargegusmains ‘viththeurualspigotjoints,madegas-tightbymoltenleadbeingranin. Inondertoobtaintheresistanceofthepipesalone,advantagewastakenof 4mainexistingintheguaworksatHanover,whichwasnotburied,butreedonawoodenflooring,‘Thissorvedalsoasaroadforthetracksladenvithcoalandcoke,andconsequently thepipelinewasexposedtomuch ritation, which seems tohave loosened thejoints asvery varying resistances ‘reeobtained—e.g,0016,0°04,10-4,420,007,44,85,26,40,and61ohmsre-spectively. Bixlengthsofpipeof10em,outerdiameterworepattogether‘pecally,andtheresistence ofthefivenewjointswasfoundtobeonly0-009hn,‘Theresistance ofthejointsis,however,ofsecondaryimportancein ‘tecueofgasmainsburiedintheground,as,owingtotheirlargesurface, {Ge“earth” resistance ofeven the fiat length ortwo will bemuch less tanthat of joint, wothat thecurrent will beearthed before ithastottenemorethanoneortwojoints,‘Toesecond portion ofthepaper deals with several new portable testing amaratag, which have been devised and manufactured bydifferent firms, toiwhich comprise batteries, magneto machines, bridges, galvanometers,‘oitelephonesforusewithbothcontinuous andalternatecurrents, J.CHAYPUISandG.MAWEUVEIER—THE MECHANISM OF ELECTROLYSIS BY MEANS OF ALTERNATING CURRENTS. (ComptesRendus,Vo.107,p81,1888) Theabvenceofanyelectrolytic actionbyanalternating currentina‘slaonofsulphateofcopper,forexample,isgenerallyexplainedbysaying {atthecopper deposited oneach electrode byone current isimmediataly aired bytheopposite current, Inthecourse oftheir experiments the‘sttorshavebeenabletorenderthisaotionvisible,Ttwouldseemthatthephenomenon dependsupontheexistenceofa ‘aigofequilibrium betweentherapidityofdecomposition oftheelectrolyte ‘adtherapidity ofrecombination ofitselements, The most important fucor indetermining the rapidity ofelectrolysis isthe current-density.Yowthiscaneasilybeincreasedbydiminishing thesurfaceoftheelec-‘oles,thecurrentbeingkeptconstant, Avoltameter, whenfilledwith‘etalated water, produced anabundance ofhydrogen and oxygen with & caratof$6amperes;butonreplacingtheacidulatedwaterbyasolution Amlphate ofcopper, noelectrolysis was sotup,until theelectrodeurtace {adbeen reduced tosbout 6square millimetres, when gaswas evolved and ‘oyper atthe same time deposited. The material ofthe electrode also fafuencestheresult,Ofgreatereffectinthenumberofalternations ofthe ‘arent.Analternate-current “‘chinewasdrivenatincreasingspeeds20as, ‘aythenumberofalternation.,.while,bymeansofanadjustableresistance ‘atheexcitingcireait,thecurrentwaskeptconstant. ‘With 188alternations persecond, thecurrent was adjusted until recom- 0 ABSTRACTS, binationoftheelomentajustbalancedelectrolytic action;ondiminishingthe number ofalternations to100per second, electrolysistookplace, Agaia,the current,having183alterations persooond,wasbroughttosuch«strengthat toproduce anabundant development ofgas; when the alternations wer‘increasedto178persecond,allevolutionofgusceased,‘Theeffectofachangeinthecurrent-density isjustcontrarytothatof& change inthe number ofalternations per second; therefore even currents of smoall surfuce density may produce electrolysis ifthe numberofalternation sresaficiently reduced, This explains how itwas that DelaRive wasabe even with alarge electrodeaurface toproduce electrolyais bymeans ofthe current of«magneto machine, since itgave only about 60alternations per second. @.MAMHUVRIER and J.CHAPPUIS—SPONTANEOUS EXPLOSIONS OCCURRING DURING THE ELECTROLYSIS OF ACIDULATED WATER BY MEANS OF AN ALTERNATING CURRENT. (Comptes Rendus, Yo, 107, p.92,1888) Tehasalwaysboonobservedthatwhenwateriselectrolysed bymeansof snalternate current, themixed gusesinthevoltameterexplodesoonerorlate DeaBive, who noticed the phenomenon, attributed ittothecatalyti action ofthe platinam black which hesaw formed onhis platinum electrode,‘Bertinattributedtheexplosionstothepolariaation oftheelectrodes,Daringtheexperiments referedtointheprecedingabstracttheseexp>-fionawereaconstantsourceoftroubletotheauthors,‘Theythereforecarefully investigatedthephenomenén,withaviewtoitsprevention,Theirrel showconclusively thattheexplosionofthemixedgasesisbroughtaboutby ‘theplatinumelectrodesbecomingheatedtosuchwpointastodeterminetherecombination ofthedimociated elements, ‘Thisheatingisduetothreecauses,allofwhich depend ultimately onthegradual descent ofthe lovel ofthe clectrolytainthevoltameter, thusleavingmoreandmoreoftheelectrodein contact withthegases,Firstly,owingtothesmallersurfaceoftheelectrode, ‘thecurrent-densityisincreased;secondly,thesurfaceresistanceisincreased; ‘thirdly,thecoolingactionoftheelectrolyteisdecreased.Itbecomes,there- fore,easytpreventtheexplosionbytakingdueprecantionstoguardaguntt theheatingoftheelectrodes, @,CHAPERON and 2,MERCADIER—ELECTRO.CHEMIOAL RADIOPHONY. (BadindetaSovietInternationaledesBactricionsVol,p,396,1888) ‘Arodcastfromsulphideofsilver,ndarodofpuresilver,areplacedinaninactiveelectrolyte, og,acidalatedwater;if,then,lightbeallowedtofallon therodofsulphideofsilver,«currentwillbesetup.venthelightfroma candleatadistanceof5feetwillproduceanappreciable defectionon& ABSTRACTS, a deletegleanomeler, Onopenczcitthesulphideofsilveriaitlactedtydifueddaylight,Detterefectamaybeobtainedbysingaplateofsilvercewhichathinlayerofvalphurhasbeendepositedelectrolytically, SerenaCatlingofvationsportionsoftherpetramdontinterferewiththephe-semenon, that probablyavarylargportionofthospectrumicapableof prodacing the eet. W. AYRTOM and J, PEERY—DPPICIENCY OP GLOW LAMPS WITH DIREOT AND ALTERNATING OURRENTS, (Pienphea Megane, Vt 26,» 476, 1888)TheproblemtobesavedinwhetheraBoardofTradeuniismorevaluable forlightingbyglowlampswhenthecurrentsuppliedtotheconsumerisdirect ccwienitisalternating, Thecompleteelutionoftheprotiaminvolvesthetectorofthefeoflampsunderthetwocondition,sincethocostofravewala caghttobetakenintoconsideration;but,asdataonthinpintarewanting, theesperimente dealtonlywiththeotherfactorofeffeiency.Byaningeniousteranguentofesiswithntercalaiedrsstanenandofwitchestheglow leap ner obervation could besupplied either with the alternate current faruished byaFerrnti machioe, orwith the ditect current oftho Gramme oie,‘Thecurenta weremeasuredbyaapecialyconstructedelectrodynamometer,wound with unusually fe wire, taordertoobtainasfara8patblouniform terreatdensity.Thisledtowasteofpowerintheinstrument,whic,how- trewasofaoconsequenceintheactualavetigatin, Thepolentaldie. toc atthe lamp terminus was meanared byanon inductive vllmeter, and {teintenty ofthe light compored onaphitometer with thatof «sacdard cane otervations being made with both ruby-red andsignl-gremglane, Wis known that with altemating curente the meamred watts donotqualthetruewatts,ToobtainthecorrectvaluowehaveTarettroewatts=7VRVsVewsr dh todfnthisparticular cam4/Zi ithoqoare rootofthemann aqoare of theamperes measured ontheelectro-dynamometer, 4/7s isthesquare rot ofthe mean aquare ofthe vote meamed on the non-aduetive oll Teer, Tistho time between one alternation and the next—ie, half a complete period—r istheresistance ofthelampfilament inohms,andJisicoetielent ofsltnduction insecohma‘The obeervatioue—T5 innumfber—were made alternately with thedirect sdwith thalterantng current, "With green light, thealteraatiog curent fare ?857 watts per candle, and tho direct current 2811 watta pr candle, ihe ight theNguren were reepectively3217and8-88,"Themeanofall thetobwrvations gavefortheallarmatingcortent30107wattspercandle, todforthedirectcurrent80180watspercandle.‘Thisverysmalldierence thsatioreconsidermaybesetdowntoexperimental errors,andtheythere-VOL, XVIII. 6 we ABSTRAOTS. forecomstotheconclusionthattheoflclenoyofglowlampisthemefr ‘bothdietandalternatingcurrents, @. FERBARIS—ELEOTHO-DYNAMIO ROTATION PRODUCED 3Y MEANS OF ALTERNATING CURRENTS. (Bitar, Vo.12,pA, 1888) Two carrunta produce two magnetic felds inplanes #and yatrightsoglottoeachother,whichactonapoint;then,ifthecurrentsamtlternatiog andfollowthesineew,omAdin:276andy=Bain,22048), and ifthere isnophase-difference, the point will move in@straight line, or,inothercases,themotionwillbeelliptical orcircular.‘Twofiatcoll,theonewithafewturusofthickwire,tbeotherwitsanyturoffinewire,areplacedatanangleoftoeachother,withthewindingsInverticalplane;theoneiconnectedintheprimarycireut,tbeother inthe secondary ofaGaulard &Gibbe transformer, with anequal umber oftaras ineach circuit. Iathe secondary cireuit iavariable resistance, Between the fat cols amall closed hollow eylinder ofcopar dangs byafre. Ifthe current passes only through the one cialhe cylinderremainsatreatbatiwothciraitsareclo,thoeylinderrotateit foneortheotberdirection,accordingtothedispositionoftheconnection. ‘Thocotsmayalobearrangedintwoparallelciruit,onoofwhichcotaiat ‘inductonlem resistance,andtheotheranelectromagnet with«large coefcant ofwll-induction, The copper cylinder may bereplaced byoneof Jron, inwhich case the rotation isproduced, not hythe induced earrste nit, butbythechangeofmagnetiation; of,insteadofhangingitfrom» ‘Abe,theeylindermaybeprovidedwithpivots,andifmadeofsufclentsi, raybeusedabamotor,andtheworkgivenoutmaybemearared,1the two sine currents aroofequal strength, and bave aphase-difereate ofaquarier ofanoscillation, the magneto fed rotates with «uniform velocity V. Ifthoeplinder revolves with anangola velocity ,andMCinthe momentofthecoupleexeriadbythecurrentsinducedintheoylinderbythermagootic field, then themechanical work ofthemotor porunit oftina i ‘W-=Mv,andtheworkdoneinthecylinder inproducingheatisP=M(V—1); tuareoregp=,ifthemseratsofthemellotheeine fn,thenthe-workepntinbeatioP=ETSYwhereinconta enceat=kT=!,theworkWinmatinif¢=}Vjand«misinanity=0,1¢Wisamaximom,thenW=P. {aanotherformofapparatusfourcolsareused,arrangedintwocoax ABSTRACTS. %s pair,whichformtheverticalsidesofacube,inthecentreofwhichthe ‘ealcylinder,oraquantityofmerearycontainedinoglavee,canbe rade torotate, - Frofwor J.A.EWING and W, LOW-INFLUENOE OF APLANE OPTRANSVERSE SECTION ON THE MAGNETIO PEBMEABULITY OF AN IRON BAR, (PhtorphioalMagarine,Vol28,p.274,1888.) ‘Toebartobeexperimenteduponpassedthrough«massiveyokeof ‘roagbtiron,thecroesectionofwachwasmorethanahundredtimesthat cfthebar,Thelowerendabuttedagainstavetcrew,whilstalever,whichcouldbeloadedtoanydesireddegre,reatodontheupperend.Theentirebarvuwoundwiththemaguetising coll,thecurrentbeingfurnishedbymeansof seromulatoraAsmallinductioncoilwaswoundatthemiddleofthebar, todwasconnectedtoaballisticgalvanometer, whichwascalibratedinthesalwaybymeansofanearthinductioncoil, ‘Thebarsworefirsttestedinonepice,thencutintotwo,four,andeight ioearacessively,‘Themagneticpermeabilitywasconsiderablydiminished tythents,Dutcouldbemadetore-approachitsformervaluebypressingthe ‘arts together, especially when thecut surfaces had been brought totrue aves; indeed, with trae planes »considerable force will destroy theresiat-s2eeofthejoimtalmostcompletely. EvenifaAlmofgoldleatisinterposedYetreen the iron faces, the resistance ofthe joint isonly very little moreshanwithoatiftherurfacesbetrueplanesandthepressureveryconsiderable A ROWLAND andL.BELL_ACTION OFAMAGNET ON CHEMIOAL ACTION, (PaitovephicatMagasin,Vol.26,p.105,1088) ‘Twomallpioosofironwirewereembedded,sidebysde,inaninmalating ‘elinm,andwereconnectedtogalvancmeter;thewholewasplacedbetween ‘tapolesofapowerfalelectro-magnet insmallbeaker,whichcouldbeAled‘wihanyLiquid,TheendofonepieceofwirewasbaredofinsulationandSedtoapoint;theotherpiecewaslaidbareonaportionofiteide,Variouslidacapableofactingchemically onironweretried,nitricacidgivingthetotrerltz, Tho experiments tond toshow that themagnetio Meld hasa Fectve action onthe iroa, this being more especially the caso when the Aiucking Liquid doce notevolve hydrogen. ‘The most complete protection aware where the iron has points from which the lines offorce can radiata, Thuyifthereinparticleofironratheranderthanthesurrounding ones, ‘SiwilbeJe attacked, and will reslt inalittle elevation from which thelstforcediverge,stillfurthershieldingit,0a8toproduceultimatelya ‘mall ridge, Nickel and cotalt show the same phenomenco, bat inalees ‘ove than toa, u ABSTRACTS, A.VOLLER—EARTHING LIGHTNING CONDUCTORS BY MEANS OF GAS AND WATER PIPES, (Bektrotechniache eitschrft, Vol20,p.478, 1888.) Ieisgenerally assumed that thepath ofthedischarge follows only thelie ‘ofleast refistance, andnoattention hasbeen paid tothefact,onwhichthe author insists, that the direetion ofthe discharge ischiefly influenced bythe state ofelectric potentialofthebuildingsinclosestproximitytothecharged oud. The better theconnection ofthemetallic masses inbuildings iswiththeearth,thehigherwillbethepotentialoftheinducedelectricity, andthe‘greater likelibood isthere ofadischarge taking place between thecloud and ‘the points inquestion. Since thegeneral introductionofgasandwaterpipet into our houses, itisthese which offer theleast resistance between therootsfandearth,Hence,ifachargedcloudshouldpassoversuchahouse,thegut ‘and water pipes must beatahigher potential, and there ismuch greater probability ofthelightning entering thehouse through them than ataxy fother point, Inother words, itismore likely that thedisaharge will take pplaco through the pipes than through the lightning conductor; and ifthe lightning rod isnot connected tothe pipes, the discharge will find iawaysomehowtothelatter,causingdestruction initspath,‘Atthe request ofthe Hamburg Fire Insurance Companies, the author undertook toinspect cases oflightning-strokes, and toascertain the point struck, aswell asthepath followed, Agreat many interesting cases invett- gated areduly recorded,butsomegeneralresultsonlycanbereproduced.It ‘generally happened that, when the building struck was unprovided with» Lightning conductor, the lightning struck some part ofthe roof orwalls, found itsway tothe gas and water pipes, and then passed harmleslyto earth, Inthe few cases where lightning struck abuilding ftted with » lightaing conductor, thedischarge jumped over from thecondhictor tothepipes,Infifteencaseswhichwerespeciallyinvestigated intheyears184and1888, after the lightning had done more orless damage atthepoint where it struck, and intheimmediate neighbourhood,itwasfoundthatinninocats thedischarge made itsway tocarth through thewater pipes, intwo exes through the gas pipes, intwo cases through rain pipes, inone case ‘probably through the lightning conductor ofatelephone line onthenestIhouse,andinonecasethroughanironcrane,Inalleasoswherethepipetwere theconductors, thepath ofthedischarge could boclearly traced up‘0 ‘them, andthen cessed. Oneofthecases ofdischarge through thegaspipes ‘courted inanordinary dwelling.house providell with lightning conductor, from which thedischarge liad passed over adistance ofabout metres to tthepipes; subsequent tests showed that theconductar-earth had aresistance of188ols, Inmocase wasapydamage done tothepipes bythedischarge ‘occurring through them, 5 LIST OF ARTICLES axurix0 10 ELECTRICITY AND MAGNETISM, ‘Appearing insome oftheprincipal Technical Journals during themonth of {ANUARY, 1880. L-BATTERIES AND ACCUMULATORS. AVoum—E.M.F. ofFleming's Standard Daniell Cell—Beibtter, vol. 18, #47, 1869,Avon—Bharf'sGasBattery—Lam.BL,vol.81,p.79,1889.4Avos.—Wright's Dry Battery,—Lam. H,vol.81,p,187, 1889,AbeWarewnorex—The Accumalators ofFarbakyandSchenk—Lwm.Et,vol81,p.143, 1889, He—UveofAccumalators inTelegraphy EI,Zit,vol.10,p.41,1889, 1L—DYNAMOS ANDMOTORS. ¥.Lanogee—Ancient Machines forProducing Electricity—Lum. Bl,vol. 81, B27, 1889, ¥.CRrcextewori—Changes intheMagnetic Flax inAlternate-Carrent Machines—Lum, Bt,vol81,p.101,1889. ©.Riewsno—Details ofDynamo Constraction.—Lam. St,ol.81,.120, 1889, ©..Semapiiex—Construction ofDynamos toobtain Best Remults.—El. Zeit, Vol.10,p94, 1889. ML-ELECTRO- CHEMISTRY ANDELECTHO- METALLURGY.P.Qotseas—Bleetrolysia ofCopperChloride.—Annalen, vol.86,.270,1869.XDaxaser—Electrolysis ofPhenol byAlternate Curronts.—Beibatter, Val18,p27, 1889, EDasconst—Electrolyris bymeans ofAlternate Currents.—Briblatter, vol.18, 38,1869, 2H.Lapesorx—Bleaching byBlectricity.—Lum, Et,vol3,p.151,1889, ¥.veFoxvieuuz—Jacobi’s Invention ofElestro- Deposition.—Zawm, Et, ol.31,p.189,1888, V—ELNCTRIC LIGHT. A.Pauar—Unit ofBight andPhotometric Standards.—Lum, Bl,vol.81,p.109, 1189, ‘door—PortableElectricLightPlant—Lum.BI,vol.3,p.191,1889,4Aror—Iastallation attheTheatreatPrague—Lum.,El,vol.91,p.19,1889,4sox—Itallations inGermany bySchuckert &Co—Bl, Zait, vol. 10, 2,186, Casctneallation attheBonMarché—Bull, Soc.Int,desBleety Vol.6,p-11, 1, 86 ARTICLES RELATING TOELECTRICITY, Ere. V.-ELECTRIC POWER. Axox,—Hlectric Tramways atHamburg.—Lum, H,,vol.81,p.178, 1869. A.pvBowe-Ruruoro—Diffenlties attending theTranmission ofPower by ‘moansofAlternateCurrents,—EI, Zeit,vol.10,p.1,1889. ‘Vi-MAGNETISN ANDNLECTRO- MAGNETISM. E,H,Hiti—Action ofMagneticForceontheEquipotential LinesofanEleotrioCurrent—Beitlattr,vol.18,p.82,1889. L,Ktue—Experiments onMagnetic Cosreive Force—Beiblitte, vol.18,p.84, 1889, (©,Rmoxiz—Magnetio Induction ofIron.—Lam, El,vol. 81,p.170, 1889, VIL-MEASUREMENTS AND MEASURING INSTRUMENTS. Dr.J,A.Fumavo—New FormofStandard ofElectrical Resistance. —Phil.‘Mag., vol. 21,p.24,1889. ¥F,J.Suru—Continuous Heat and Electric.Current Measuring Instrument— Phil, Mag., vol. 27,p.28,1889. 3,W. W. Waoxonx.—Determination ofElectro-magnetic Capacity.—Phd, ‘Mag., Vol. 21,p.69,1889, B,Doax—DeterminationoftheObm.—Annalen, vol.86,p.22,1880 Brutart and Lustaxa—Passago ofElectric Currenta through Bad ntacte,— Beibitter, vol. 18,p.21,1889. A.D’Ansoxvat—Universal Doad-Boat Galvanometer—Tum, Bt,vol. 81,p.18, 1880;Bull,Soe,Int,desElec,vol.6,p.6,1889. ‘Brcums, Lanwrres, and Prcato—Photographie Reconis ofElectric Carrente.— Lum, BL,vol. B1,p.16,1689, ‘A.Paat2ow—Measurement ofthe ResistanceofaConductor,—Ium.El.,vol.$1, P.80, 1889, Kurr and Zoxax—Electro-Dynamometer forMeasuring Telephonic Carrents— Lum, BL, vol, 81,p.81,1889.©.Dromsnxe—Some NewFormsofGalvanometers—Lam, El.,vol.81,pp.70 and 127, 1889. Avox.—Edison's Electrolytic Meter.—Lum, Et,vol. 31,p.180, 1889. P,Nrrxow—Electro-Dynamometer forAlternate Currents, —I, Zsit., vol, 10, P28, 1889, ‘VIIL-RAILWAY APPLIANCES. BE,Zarsoms—Couplers forConductors onTrains,—Lum,El,vol,81,p.189,1889,1M.Cossxaxn—Metzger's Automatic Signalling Apparatua—Lum, El.,vol, 31, P.164, 1889, Axox.—Siemens &Halake's Alarm Signal for Level Crosings—E1, Zeit, ‘ol. 10,p.89,1889, TX-STATIC AND ATMOSPHERIC ELECTRICITY. ‘Eovox—Some TheoriesofAtmospheric Electricity. —Beitlater, vol.18,p.48,1669, ARTICLES RELATING TOELECTRICITY, Bre at A,PauusOonstruction ofLightning Conductors.—Lam. BL,vol.81,p.58, 1882, (©Warm—cyclones andWaterspouts,—Lum, Hl,vol.81,p76,1889. X—TELEGRAPEY AND TELEPHONY. &,Zxrscus—Domestic Telephony —Lam,HL,vol.81,p.28,1889.Axox.—Use ofTelephonic Conductors forSynchronising Clocks.—Lum.Ft,‘Vol. 81,p.82,1889, ‘Scuaxrres—New Iron Telegraph Post—Lum, Bl.,vol,81,p.88,1889. ¥.Lansoqus—Permanent Effects produced inCopper Conductors byaLong. continuedPassageofaCurrent—Lum, Bt,vol.81,p.161,1880. ‘B,Prmon—Automatic Gall and Conclusion Signal forTelephone Exchanges, Bl, Zeit, vol. 10,p.12,1889, T.D,Locrwoop—Apparatas for Diminishing Induction.—EI, Zeit,, vol. 10, P14, 1889. Asox.—The Swiss Telephone System,—FI, Zeit., vol. 10,p.46,1889. Dr.$,8,Wammuxe—Overhead and Underground Conductors inNew York.— Bull, Boe. Int, desElect, vol. 6,p.22,1889. xL—THEORY. 0.Hasvisroy—Maxwell's Electromagnetic Equations inaHomogeneous Inotropic Medium.—Phil. Mog., vol. 7,p.29,1889. Prot, J.V. Joxes—Calculation ofthe Coefficient of Matual Induction ofa Circle and aCoaxial Helix.—PMit, Mog., vol. 27,p.66,1889. 4.Panxan—Thermo-electric Phonomenon.—PM. Mag., vol. 27,p.12,1889.H,Hexrz—The ForcesofElectricOscillations.—Annalen, vol.86,p.1,1889.W.G, Hawre:—Tho Electro-dynamic Law—Annalen, vol. 86,p.78,1889. GP, Gamuspr—Effect ofAnnealing onthe Thermo-clectri Properties of ‘Bismath,—Beibliter, vol. 18,p.25,1889. 4. Lonmr—Action ofElectricity onSteam.—Beiblattr, vol. 18,p.87,1889. ©.RGunsacwe—Notation and Symbols—Lum, E1., vol. 81p.10,1889.J.andP.Cunte—Electric Expansion, —Lum.EI.,vol.81,p.60,1889. ‘XIL—VARIOUS APPLIANCES. W,pxForvimiz—Applications ofElectrolysis inOperative Bargery.—Laum,BL, vol. 81,p.45,1889. ©,Rreuaxp—Some Mechanical Applications ofElectricity,—Lum, Ht,vol. 31 P64, 1869. Arox,—Electric Mine-Explodera.—Lum, El.,vol.81,p.76,1889.XMorrratix—Fire-Alarms.—Lvm, Mt,vol.81,p78, 1889. ‘W.oxFoxvmis—Death byElectricity.—Lam. E1.,vol.31,p.90,1889.P.Geom,—Maximam Thermometer. —Lum.Bl.,vol.81,p:169,1889. NOTICE. 1.The Society’s Library isopen tomembers ofallScien- tific Bodies, and (onapplication tothe Secretary) to the Public generally. . 2.The Library isopen (except from the 4th August to the16th September) daily between thehours of11.0 a.m. and 8.0p.m,, except onThursdays, and onSaturdays, when itcloses at2.0p.m. An Index, compiled bythe late Librarian,tothefrst tenvolumes oftheJournalcanbehadonapplication to the Secretary, ortoMessrs.E.andF.N.Spon,125,Strand, W.0. Price Two Shillings and Sixpence. or me Institution ofElectrical Engineers, Founded1871.Incorporated 1883, Vou. XVIII. 1889. No. 78. TheOne Hundred andEighty-fifth Ordinary General Meeting of theInstitution washeld attheInstitution ofCivil Engineers, 25,Great George Street, Westminster, onThursday evening, February 14th, 1889—Dr. Joun Horxinsox, M.A., F.RS., Vice-President, inthe Chair. The minutes ofthe Ordinary General Meeting held on January 24th, 1889, were read andapproved. The names ofnew candidstes forelection into the Institution were announced andordered tobesuspended. The following transfers were announced ashaving been approved bytheCouncil :— From the class ofStudents tothe class ofAssociates— Charles AlfredBaker. i] Frederick Bathurst. Gustav H.C.Risch, jun. Donations tothe Library were announced ashaving been received since thelast meeting from Mr.Charles Todd, C.M.G. (Local Hon. Sec. forSouth Australia), and Mr. A.C.Swinton, Awociate, towhom the thanks ofthe meeting were duly accorded. The CuarnMan: We will now resume the discussion on Professor Jamieson’s paper on“The Insulation ofElectric Light ‘VOL, XVI. 7 90 LETTER FROM PROFESSOR JAMIESON, =[Feb.Mth, “Installations.” Aletter has been received from Professor Jamieson, which theSecretary will read. The Secretary: Ihave received aletter from Professor Jamieson, which hehas requested metoread tothemeeting before thediscussion isresumed, asithassome bearing onthe paper. Itisasfollows:— February 12th, 1889. Dear Sm,—Hince theCouncil have given afurther opportunity forait cussing mypaper on Tho Ineulation ResistanceofElectricLightingCircuits,” T-would like ifyou would kindly point out tothe meeting,onmybehalf,tat inmypaper IIntentionally conned my remarks toénrulaton roitonce,and totherulesaswellasmethodsoftestingforthesamo,asfarantheywere known tome. Tpurposely avoided saying anything about iarulatin, orthe diferent kinds ofénelating materials, oFthe diferent ways inwhich they rv rade anderected, Ae. Tohave done a0would have defeated theobject ofthe paper, viz, toflod out what isthe minimum ingulation resistance which we may with safety allow aninstallation tobave without considering itbelow ‘ar, and what arethesimplest and best methodsofascertaining theinrulaton ‘reaistanceofthevariouscircuitsbforeandaftertheinstallationhasbeenwt tovwork. {IT had treated oninsulation and inrulating materials, dc., Imight have ‘written atrenton theaubject and not have exhausted it,and thedit catsion could easily have been continued formany nights without covering thewhole ground orcoming toadecision, Inregard totesting, there seems toboapopular belief that itisnecer sary toemploy ashigh anELF. astheworking current when soroaust ‘ccertaining theinsulation resistance. This isnot the case,asoldsubmarine cable electricians know very well. {¢you have asensitive mirror galvan-ometerandafewordinarychlorideofsilverofEaclanchécals,youwillbeable totall quiteaswellasifyouhad100ormorecellsandaleasaensitivegalvas ‘ometer, what the real resistance is. Ofcourse ifyou wish tomubmit thecore forcable toanelectrical tres equal toorgreater than that oftheworking ‘current, you may do#0,but you would not keep your testing galvanomeler‘andresistancecoilsinireuitatthetimeofapplyingthisstress? What Imaintain isthis—that we can ascertain the actual Soeolation resistance inohms, oFtheportion ofafault, with great accuracy andeasebyusingonlyafowcelleandasensitivemirrorgulvanometer, Ihaveneverused anything elathan amirror galvanometer with which totake theinsula. tion resistance ofelectric light lreuita,and Thave employed itinthemost awkward positions, uch ascable-huts, engine-rooms inahipe and onshore,‘banks,cellars,stores.Acoupleofordinarychairsserves,theonetorapportthegalvanometer and theother thescale, Galvanometer, lamp, scale, and Key allgointo oneamall case, and the whole can becarried bythelitle finger, andswang round the hend atarm’s length without affecting theAne‘cocoonsilkruapension! InfiveminutesafteracrivalattheapotwheretheteatshavetobetakenThavegotaspot,andintwominutesafterIhavegot 19,3 COMMUNICATION FROMMR,SMELLIE. om ‘thercistancethewholsin‘packedandreadytobetakenaway,Ifyouwish asketeh,IwillsendonefortheJournal, Tam, dear Sir, ‘Yours faithfully, ANDREW JAMIFSON, 7-H, Wann, Ea, Secretary. ‘The Secretary :Ihave received ashort communication from Mr.Smellie, who istheelectrician forMessrs. Denny &Co., and ‘whowasreferred toinProfessor Jamieson's paper. Heregrets hecould notcome totown totake part inthediscussion, and srrtes asfollows :— “Re Pnoresson Jammeson's Paren—InsuLation Reststaxce. |“The instrument uredbyme,andreferred tobyProfessor |“Jamieson, ianotquitecorrectly described.| «Itconsistedofoneofthesmallhand-magnet generatorsused “bythetelephone companies, inconjunction withasmallvibrating ;“needle ‘tell-tale’ usedinWheatatone ABC exchanges before “the advent oftelephones, serving toshow theoperator when ;“two lines were speaking; and was much more sensitive than “any polarised bell Ihave yetseen, Since then ithas been in “constant use,andalthough themagnets areweaker than they “were atfirst, itisstill capable ofringing through 16,000 w. “The value ofaninstrument ofthis description canonly be “appreciated bythose who have been accustomed touse a “detector galvanometer and battery, which gets out ofcon- “dition orisupset, often when most needed. “Regarding thetotal insulation resistance ofE.L. systems, “Iwould suggest that theILR.should begiven perlamp, asa “means ofcomparison between installations ofdifferent sizes. “At thetime theProfessor refers toIwasobtaining from “30to350perlight,butsincethenhavegotashighas950“onboardship.Withthesamemenandqualityofmaterial,i]“the LR. ofinstallations onshore farexceeds this. “The value ofatest taken with alow battery E.M.F. and \“rtecting galvanometer vanes with thedewpoint, andraises 1]“couiderable difficulty inmakingafaircomparison ofanytwo 4)“installations, Pa ‘THEINSULATION RESISTANCE OF[ehUd, “However satisfactory suchatestwouldbewithasubmerged“cable, whose conditions areconstant, itisbynomeans suitable “forasystem ofelectric light leads thatvaries from daytoday “according toatmospheric conditions, andthereby renders all “tests made inthis manner useless forcomparison. “Being dissatisfied with this method, Ihave made it «practice tousea‘breakdown’ E.M-F. of700volts ominstalla- “tions for100-volt lamps: thisgives afactor ofsafety ofseven, “andthehighpotential quicklydecomposes themoisture which «may exist, without sensibly heating theconductors. “The total leakage may then bemeasured, andthetotal LR. “calculated therefrom, “Contractors might demur atthis being insisted on,on «account ofasufficiently high potential current being unobtain- “able; but this iseasily got over byusing asmall constant “currenttranaformer havingaratioof7to1,andexcitedbythe “dynamo fortheinstallation. “A test ofthis kind gives asatisfaction which noother “method can, and Ifeel sure would assist inregaining the “public confidence, which has been sadly shaken from time to «time through failures caused byindifferent workmanship,—not «due totheindifference ofcontractors, buttoworkmen without “a conscience.” ‘The CHamman:Mr.W.H.Preecewillperhapsresumethe discussion? usPees. Mr, W. H. Preece: Ishould like tomake one ortwo remarks onthis paper from theuser’s point ofview. Something wassaidintheletterwhichhasjustbeenreadfromProfessorJamieson from thecontractor’s point ofview, but itisavery differentthingindeedwhenthisquestion islookedatfromthe result oftheexperience ofafewmonths’ orafewyears’ working. One ofthe most serious evils that followed the introduction| oftheelectric light into England wastheindifferent work that‘accompanied itintheearlydays,andtheveryimperfectmaterial| used forthepurpose. Ihave been told myself, byoneofour oldmanufacturers, that hehas absolutely refused toexecute orders that have been putintohishands forvery indifferent wire | 1889.) ELECTRIC LIGHT INSTALLATIONS, 8 intended forelectric light installations;andIamquitecertainus.reve. thatthefirst rule thatistobefollowed intheinstallation ofany lectric light equipment, andininstituting anyregulation to determine the insulation resistance ofthat equipment, isto specify that none butthevery best: materials shall beused. Now theconditions that determine thequality ofmaterials, andthe tests that have tobemade upon those materials in electric light and transmission ofpower installations, isavery different thing indeed tothat which determines theinstallation oftelegraphs and telephones. Inanelectric light installation theconductors, during thetime thelight isgoing, areunder 1constantelectro-motive force;andwherethesmallestleakispresent under the existence ofmoisture and of#constant dectro-motive force, the breakdown ofthe wire issimply a question oftime. Ihave been astounded tofind thequickness withwhich electric light currents ferret outafault anddestroy thewire. There isnoplace where faults occur somuch, for instance, asinstables, where ammoniacal vapours areconstantly about, which assist intherapid destruction ofconductors. When there isthe smallest leak, and acurrent constantly passing through thatleak, electrolytic action issetup,nascent oxygen— r,rather, ozone—is produced atthe point ofleakage, and one,asitisproduced,hasadestructive influenceverysimilar tobuming onorganic substances. When you examine a conductor that has been broken down inthis way, iti,forall theworld, exactly liketheresult ofaflash oflightning. Ihave tienhearditsaidthatunderground wiresandwiresofdifferent, kinds have been broken down bylightning, beciuse they bore |thedistinct marks ofburning, ofsomething verylikelytohappen |Titheflash oflightning; butIhavebeenabletofind,inmanyinstances,thatithasheenreallydue,nottoflashesoflightning, |buttothedevelopment ofozoneattheleakage, andtotheeatingsmayanddestruction ofthehydro-carbon that,asarule,forms the dielectric. Therefore itisthat faulte of this class are developed sovery rapidly under theinfluence ofelectric light indallations, andsoitisthat itismost essential that weshould tvercise great care intheselection ofourmaterials, andintheir “ THE INGULATION RESISTANCE, Bre. (Fe.140, rmabeingtestedbeforetheyareerected.‘Then,again,wemut remember this—that inallelectric light installations there area number ofpositions, anumber oflocalities, that arethem- selves identical with faults, such asswitch-boards, the drnamo itself, cut-outs, and fixings; also the frames ofarelamps, and, worst ofall,gas fixtures. Itfollows from this that with an installation practically fulloffaults atallpoints there must bea distinot rule fordeciding what shall betheminimum insulation resistance tobegiven tothe installation under the worst cit cumstances; and Itake itthet theprincipal point ofProfessorJamieson’s paperthathehasbrought strongly beforeusishis own personal view ofwhat should bethe minimum resistance that under any circumstances should beallowed. Wetried tobattle with this question when theCommittee of theSocietyofTelegraph Engineers discussed thequestionoffire risks, andarule wasestablished, which was defended onthelast oocasion byMr, Crompton, and who said—with which Ifally agreo—that now with our present experience wemight rather enlarge that rule, andinstead ofdefining theleakage that should beadmitted: asbeing 1-5,000th ofthemaximum output ofthe installation, wemightverywellraiseitto1-10,000th part.Ifwe didthat, weshould bring theleakage allowable more near tothe lawandformula ofProfessor Jamieson. This subject, although itbas not been received with quite the feeling that one would have expected from somany engineers who areassociated not only with the maintenance but with the construction ofthese installations, has met with aconsiderable amount ofattention inAmerica, There they arefarahead of.us inthe practical developmient ofelectric lighting;theretheabsolutenecessityofhavingsomerule todetermine the minimum insulation resistance has forced itself forward, and Ithink itwas there that Mr. Leonard was the first toinitiate arale that has been the basis ofseveral others. His law was that the insulation resistance (R;) should boequivalent toKE,Kbeinga,constantmultiplied intoafrac- tion having theelectromotive force asitsnumerator andthe 16.) DISCUSSION. 5 carrent asitsdenominator. That formula hasbeen taken upbyMs.Pres. A.Picou inFrance, and itisthe basis ofthe formula that Professor Jamieson hasgiven us; but Professor Jamieson bas veryproperlyreplacedthedenominator Cforcurrentbynforthe number oflamps. ‘The current may beconstant, but the lamps maybeinseries, and therefore vary. Take, forinstance, aBrush aecireuit with 60lamps: thecurrent isconstant ;thenumber of ralts does notdepend upon thecurrent inanyway, butupon the number oflamps. Hence the denominator should ben,the numberoflamps,andnotthecurrentflowing. Thereisanother dlement that enters into this question that, Ithink, hasnotbeen touched upon byProfessor Jamieson, andavery important oneit is,Professor Jamieson's paper hasbeen written from thepointofviewofexamining theinsulation resistance ofanindoorinstal- lation;andifsimplyanindoorinstallation beconsidered, then probably that formula willanswer very well. Butitisnotindoor installations thataregoingtogiveustrouble; thetroubleisgoing toatise from exterior installations. We aregoing toseegreat advances inelectric lighting; weshall seeinthenext twoyears, ‘=y,nearly every town ofanyconsequence intheUnited Kingdom taking upthesubject ofelectric lighting. ‘Thewhole atmosphereisnowsimplyredolent withelectriclighting, anditonlywantsthepurse-strings ofourLondoncapitalists tobeundonetofind very town inthis country swarming with electric light engineers; aodwhen that isthe case—which will bethecase inavery few ‘months’ time—it will not bealone the insulation ofinstallations oftheinterior ofbuildings that weshall have tothink about, bat itwill bethose circuits that will becarried round about oartowns, andoursparsely inhabited villages and places, where insolation will beamatter ofvery serious consequence—insu- lutionnotalone forthesafety oftheelectric cironit, butalso | frthesafety ofthe public—and itishere where arule of ‘ome kind iswanted. Itisinoutdoor circuite where really tome rule isnecessary. ‘Those who have studied Provisional Orders, and those who have studied theActs ofParliament, |wellkmowthatitwillrestuponthelocalauthorities todecideVat shall beadefective main orwhat shall beadefective 8 TRE INSULATION RISISTANGE, Bre (PALMA Mr.Prec.conductor; andhowisitpossible fortheseauthorities, scat- tered allover the country, todecide what isadefective main orwhat isadefective lead, ifwe, the Institution ofElectrical Engineers, arenotable todefine some lawthat shall decide the difference between abadandagoodconductor? Ithinkthatthisformula isthebasisofaverygoodlaw.Ithinkthatitis perfectly certain that theinsulation resistance must vary directly with the electro-motive force. We have installations of50volts, of100volts, of2,500 volts, and inoneinstance itisproposed to goashigh even as10,000 volts, and itisperfectly clear that the insulation resistance ofan installation must increase directly with the electro-motive force. Then, again, itmust depend upon thenumber oflamps, forevery lamp isafault. Again,especially withaerialwires,itmustvarywiththelength ofthe circuit, and particularly sointhe case. ofthose circuits thatareusedforhighpressure—are circuits. Ihaveacaseinmy mind’s eyeatthepresent moment where weshall have tolight upaseries ofrows andstreets where thecircuit must betwenty mileslong.Itisveryevidentthatwithacircuit twentymileslong youmust have aconstant that willgive you thesame insulation forthewhole circuit asforone mile long;sothatIsayforsuch circuits we must introduce in this denominator the function oflength. The lamps should berepresented byN+1,the lengthofthecircuitbyL.IsayN+1,becauseifwegaveN nothing theformula wouldbeabsurd, butbymaking itN+1 wedeal with allcases. The formula would therefore be E w=Krwety Now, then, wehave only todetermine what thecoefficient Kshall be. Professor Jamieson says itshould be100,000. InAmerica, forasomewhat similar formula, they have made itaslow as1,000; but onthis point itisreally aquestion ofexperience, and Iwould ask every electrical engineer present who has anelectric light installation under his charge tomeasure theinsulation resistance, and totransmit the results tothe Secretary forpublication inthe Journal ofProceedings. Ifhedoes so,hewill give usthe oppor- 1, DISCUSSION. on tunityofbeingabletosaytolocalauthorities, “Nowthereisws.Prec.“aformulawhichyouhavetofollow:thecoefficient Kshallbe “one thing forhigh-pressure arecircuits, itshall beanother “thing foralternate-current cireuits, itshall beathird thing “forlow-pressure circuits;” andwecanmake itagree with every condition that may arise. Speaking from experience inthe PostOffice, and intelegraphs generally, wehave absolutely neglected thetesting ofinterior installations, forwenever meet withfaults there. Wedonotrequire togowandering about thecountry with mirror galvanometer that youcan“swing “round your head bytheaidofyour little finger;”wehave instraments—galvanometers—of realpractical character, that will enubleustoseewhether thewiresaregoodorwhether theyare tad. Inelectric light installations, oranyother installations, if ‘riresareproperly tested before they areerected, and ifthey are occasionally tested toseethat nofaults exist, Iaminclined to thinkthatoarduty aselectrical engineers isfulfilled ;butwhat ‘vereally dowant isthis—that weshall beable tosubmit tolocal authorities, and tosuch powers asthe Board ofTrade, some formula that shall enable them todecide what the insulation thall beforcircuits when used forarelighting under high| pressure, forglow lamps under low pressure, orforalternate ‘ystems under pressures varying from 1,000 to10,000 volts; and vecanonlyarriveatthepropervalueofthecoefficient Kifwe tueassisted byallthoee electrical engineers who have installa- "tions under their charge andwillsend toustheresults oftheir measurements. Ioughttosaythis—that onsearching through ‘mypapers tofind any illustration ofatest, Ihave only come ‘crossone,andthatisthePostOfficeatGlasgow, whichhasbeen lighted byelectricity forthelastfour orfiveyears. Recently vehad afireontheroof there—a pure accident, nothing todo Withtheelectric lighting, although itwasintheelectric light leads;butwetookthatopportunity totesttheinstallation, and thereport, ofanextremely able electrician—Mr. Boldchild, of Glasgow—says: “Itested thewhole oftheglow system, including “allswitches andcut-outs, using 50bichromate cells [that would. “be100volts}, andfound thetotal resistance toearth tobe ry ‘THEINSULATION RESISTANCE, Ero.(Feb.4th, urPree.“about58,000ohms;theresistance between theleadswas “100,000 ohms, allthe lamps being inand theswitches open. “The areleads showed nothing with afall coil ofthetangent «galvanometer. When Itested inOctober, 1886, oneleadgave “60,000 ohms resistance, and theother only 1,500;thisproved «tobeadefect, which wasimmediately removed. Myopinion is that weshould have nodifficulty inmaintaining aninstallation offrom 50,000 to60,000 ohms, even with the present crude “workmanship.” Ofcourse heisspeaking oftheworkmanship inGlasgow. Imention that asbeing just what wereally want: we wanttohavemeasurements oftheinsulationresistance,not only ofcircuits inbuildings, but especially circuits working under high pressure, and there arenow agood many ofthem about thecountry. aekeope. Mr, H.R, Kempe: Sir, Ithink that Professor Jamieson has made amost important mistake innot distinguishing between the insulation ofthe cables and the insulation ofthe fittings inanelectric light installation, The leakage inacomplete installation isdue tolossinthefittings, ortoloss intheeable, ortoboth. The former will bechiefly, ifnot wholly, due to surface leakage, thelatter toleakage through thedielectrie, Now theresult ofleakage inthecase offittings would probably beto some extent toimprove theinsulation, owing tothe decom position ofthemoisture which would consequently pass off; but this iscertainly not the case ifthe leakage isdue to defect inthe cables. Inthe case ofthe cable the circumstances aretotally different: acurrent would undoubtedly open afault ifitexisted, andintime would certainly break itdown, even if thefault beexceedingly minute atfirst. Itisnotsufficient for the insulation ofacable, ifthe dielectric isofany thickness, tobeonly afew megohms;itoughttobeseveralhundred,if notseveral thousand, megohms; forifthat isnotthecaso, the lowinsulation, although itmay beashigh asseveral megohms, isalmost sure tobedue toadefect, which defect would go from bad toworse, sooner orlater, under the influence of strong current. This, Ithink, isanestablished fact, andis well known toallwho have had todeal with cable cores, or 190) DISCUSSION. % with insulated wires. The standard ofinsulation ofawire—of aekenpe snordinary cable core—perhaps may beashigh as500megohms permile, Now, ifitrans down to,wewill say, 100megohms, thatinvulation, itistrue, would beperfectly sufficient to Prevent any leakage ofthe current, provided itremained constant; but afall offrom 500 to100 megobms would certainly bedue toaslight defect, and acurrent, even ofaslow 4potential as100volts,wouldbealmostsureinashorttime tocompletely break the fault down, Ofcourse adefect like thatwould belikely tocause rather serious results, because it isquite possible that thecurrent might arcacross and setthe insolation onfire. Itisawell-known fact intelegraph work tataninsulated underground wire, ifitsresistance does: not considerably exceed afewmegohms, issure togobed sooner later, although the insulation ofafewmegohms would be quitesufficient toenable perfect signals tobecarried through” i,There isanother point. Nooneengaged intelegraph work vould ever think oftesting enunderground wire with the imtruments incircuit. ‘The instruments would betaken off, aadthewire would betested byitself; because weshould know tatifwegot alowresult—a result equal to,say, 10or12 uegohms—it might beentirely due tothe leakage over the iastraments, and notduetoleakage onthecable. Itisamost important point that intesting cables orinsulated wires theyshould beperfectly free. Ireoollect, some time ago, lavingtotestthirtyorfortyunderground wires,havingeacha leagthofabout1}miles.‘Theinsulationcameout,withall thewiresonthetestbox,about12megobms permileallrand, That insulation, ofcourse, was quite good enough for wrking purposes;butIknewperfectlywellthattheresistances good underground wires could not possibly beaslowasthat, Ihadthewires taken offthe test box, and tested the ‘umulation resistances ofthe terminals alone, and the results ‘aneoutexactly the same asthey were before the wires were takenoff;butthewires tested offthetestboxgave between 4%and300megohms permile.ThatshowedthatthetestWiththewiresontheboxwasperfectly useless. 100 THEINSULATION RESISTANCE, Ere,(eb.14th, us.kenpe. Inthe letter from Professor Jamieson which the Secretary hasjust read, itisstated that itwasnotofimportance that the insulation resistance ofan insulated wire should be taken with high battery power. Asregards that point I'am totally atissue with Professor Jamieson. Forthelastfifteen years Ihave hada greatdealofexperience intestingseveralthousand milesofwire, and myexperience tells meexactly thecontrary; and this is especially thecase with indis-rubber insulation, You may test anindia-rubber cable, and theresult may apparently beperfectly satisfactory ifthetest istaken with, say, 20or30cells: the insulation will beextremely high, and theelectrification, which isconsidered tobeanindication ofthesoundness ofacable,may beperfectly steady. Suppose 300cells areused, youmay find thatafter twoorthree minutes thedeflection onthegalvanometer seale will become unsteady, anditwill finally gooffthe scale; infact, afault will develop, and theinsulation rundown tosome- thing exceedingly small. Itherefore consider that itisvery necessary inanelectric light installation that thetesting ofthe cables should betaken quite independently ofthefittings, and thatacertainstandardshouldbetakenforthefittingsthemselves, andanother standard—which would bevery, very much higher— foreach ofthecables tested independently ofthefittings and ofthetest boxwith which they may beconnected. MeWaner, Mr, Stovey F.WatKer: Ishould just like tomake afew remarks from thepoint ofview ofthedynamo manufacturer and the contractor ;andasapreliminary IshouldliketosaythatI agree with Mr.Preece that thework and theconditions ofan electriclightingcircuitarequitedifferent tothoseofatelegraphcireuit, andthat therules which would apply tothetesting of| telegraph circuits willonly apply toelectric lighting circuits in ‘aminordegree:wehavelargercurrents asarule,andhigher electro-motive force always atwork. Ithink that thethanks of| electric lighting contractors aredue toProfessor Jamieson for bringing thismatter forward, andforattempting tofixastandardofinsulation, Whatwewantmorethananything isastandard ofinsulation. What most contractors are troubled with when we tender forelectric lighting work isthat those ofuswho have all 100 DISCUSSION. 101 theconditions ofworkbeforeustenderforahighstandard, whilear.wate. romebody else, with alight heart, tenders foralowstandard, and getsthecontract. Ofcourse that leads, asMr.Preece hassaid, todisgrace, Electrical engineersaretroubledatthepresenttime, ustheyhavebeeninthewholematterofelectricalwork,more vithmengoingintothethingwithoutbeingproperlytrained, andwithout proper knowledge, than from anything else, One striking point about thepaper, though, Inotice, isthevariation ithe standard. Professor Jamieson gives hisown standard as 1410,000th ofthe current; that oftheSociety ofTelegraph Engineers is1-5,000th; Mr. Preece hasjust told usthat the Americans take 1-1,000th;andIunderstandthatM.Picoutakes 1-500th, ifIunderstand itrightly. IfImaymakeasuggestion, Ithink that weought tohave anauthoritative standard. I ‘certainly agree with theauthor, andwith theremarks ofothers,(iatwithincertainlimitsyoucanhardlyhavethestandard too high. Itisasimple matter, comparatively, toget high stan- dardofinsulation atthestart, but itisbynomeans asimple tattertokeepitup.Myworkhappenstobeprincipallywhat Mr.Preecehasmentioned asoutdoorwork.‘Thegreaterportion ofmy work hasbeen amongst collieries, where wehave consider- abledifficulty inkeeping upinsulation; andIhave nodoubt that there areplaces, even indoors, where there isconsiderable diffi- caltyinkeeping upinsulation. Ihave places atcollieries, inside ‘ildings, where cables areliable to#considerable amount of wet;inpit-shafts, again,acablemuststandastreamofwater, cen impregnated with thesalts ofdifferent metals. Now I lavefound noinsulating substance yetwhich willstand astream cfwater,ifitisallowedtogettotheinsulating substance itself,‘cept gutta-percha, andthat only ifthere isplenty ofit—if itis ‘good thickness. Ihope thatthere aresome cable manufacturers, here,because Iunderstand that cable manufacturers object to make electric lighting cables with gutta-percha covering, Ipre-‘umeonaccountoftheheatingeffectofthecurrent, ButIfeel surethat there are many places where only gutta-percha will and, India-rubber andCallender’s core will stand ifyoustop the ‘elgetting tothemat all; otherwise they will notstand. India oz ‘THE INSULATION RESISTANCE, Ere. (Feb. 4, aM.Walier.rubber—valeanised rubber—may startatathousand megobms, or ‘asmany megohms asever youlike, butitwillnotkeep itup;it isonlyaquestionoftime.Ashasbeenpointedoutinthepaper, and byMr. Preece, one ofthe main difficulties inmaintaining insulation ofelectric light circuits isthe action ofthecurrent itself,Thecurrentitselfpassingacrossthecable—even thevery small current that isgoing—deteriorates it;andnotonlythat,it alsodeteriorates thefittings. Ishould differ from Mr.Kempe in that respect. ‘The effect ofacurrent pasting across fittings, especially iftheyarewoodenfittings,isnotonlytodryup'themoisture inthefitting,butalsotocarbonise theinsulation ofthe fitting itself, and eventually sparks pass across and something very like alightning flash occurs. ‘With regard tothetesting, Icertainly agree with those speakers whohave saidthat forelectric light work youmust test with ahigh voltage—as high asever you cangetit. You cannot take amirror galvanometer and asetofdelicate testing instru- ments into the majority ofplaces where anelectric light installation existe. Professor Jamieson isapparently very lucky with hisapparatus, butthebest apparatus forrough-and-ready testing isawell-constructed detector galvanometer withahigh voltage behind it. Iam bound tosaythat Ijoin with those who view thehigh tensions which areabout tobeused inLondon with considerable apprehension. Icertainly admire theskill which has been shown inarranging the conductors forthe Deptford cirenit, which aretostand 10,000 volts; butIshall be agreeably surprised ifthat insulating substance isnotgradually deteriorated tillinsome part, where possibly there isasmall bend orsome slight difference inthickness, aspark passes across, and then thewhole thing willbreak down, Iwillendbysaying that, Thope astandard willbearrived atthat wecanallwork to. Reimer Professor W.E,AvRtoN: Itwillbeobserved thatallthese rules quoted byProfessor Jamieson astowhat the insulation resistance should be,really lead totheresult that theinsulation resistance should besome definite multiple ofthe conductor resistance ofthecable,viz,aanditreallycomestothe 88) DISCUSSION. 108 question, What multiple oftheconductor resistance ofthecableRotmer sould bespecified asbeing theminimum insulation resistance allowable? Mr. Proece's formula, however, differs from allof theso rules incharacter, and not merely inthe value ofthe coeficient, forhewould make, according tohisformula, the insolation resistance diminish more rapidly than the length, if thereweremanylamps.Infact,ifyouassumethatthenumberoflampeincircaitisaboutproportional. tothelengthofthe |able,hewouldmaketheresistancevaryastheinversesquareof |ttelength;ie,therevistancewouldfalloffveryrapidlyforlong conductors. Ican hardly think that that can betheright way ofspecifying theresistance, because itwould make theresistance auchtoolowavalueforslongconductor. Lasttimetherewas 2sort offeeling that this question ofspecifying what the inmlation resistance should beforanelectric light cireuit should teletopersonaljudgment;butthat,ofcourse,isnotpossible |wier theparticular circumstances ofthecase. Inthematter efsubmarine cablesitisamerequestion ofacompactbetweenthecontractor andthebuyerofthecable:thebuyeriswillingtopeys0much,andthecontractor iswillingtogivesomuch insulation, among other things, forthat sum ofmoney; andas lugasthosetwoagreehowmuchinsulation istobegivenon theone hand, and how much money istobepaid ontheother lund, Idefy anybody toenter between thetwoandsay, “You “shall notmake that compact.” Inthecase ofanelectric light reait there is,fortunately orunfortunately, athird party tobe cmsidered, and that isthe Board ofTrade orthe Insurance Office;andtherecomesthedifficulty.ImaysayIwishmy louse fitted up by acertain firm; Ihave only acertain saount ofmoney tospend, and Iwant togetasmuch insula- tinasIcanforit;but the Insurance Company steps inand ‘sy,“No, weshall notbesatisfied with that; you must pay “more andgetbetter insulation.” Ofcourse there isavery ‘lidreason fortheir doing this—on account ofsafety—and there- foreitbecomes necessary tosettle some sortofstandard; it ‘becomes all-important, asMr.Kempe haspointed out,toseein hatwaythatstandard canbesettled. Iquite agree with him 108 ‘THE INSULATION RESISTANCE, Ere. [Feb14t, Potmer thatitisamistake, ifyouaregoing tosettle astandard whichis tobind theEnglish world generally, tomix upthe insulation of thefittings with theinsulation ofthecable, especially when dealing with long lengths ofcable. ‘The same sort ofmistake that Mr.Kempe haspointed out Ihappened topoint outnearly twenty years ago. Atthat time aspecification inconnection with telegraph insulators was drawn upbyavery eminent electrician inEngland forthe Indian Government Telegraph Department, and I,asanot eminent electrician, onleave in England, was desired towork under that specification, The specification was tothe effect that notelegraph insulator, tested inthe usual way, should have aless resistance than somany thousands ofmegohms ;but,forconvenience, theinsulators should bejoined upinhundredsandshouldbetestedinparallel,andif theresistance ofthe100exceeded one-hundredth ofthespeci- cified resistance perinsulator, then Iwastopass them. Well,I pointed outthat that waswrong, and infact Isaid that until the specification was changed Icould notwork under it,forthe simple reason that itmight bethat the 100insulators would have ahigher resistance than one-hundredth part ofthespecified insulation, andyetthe100 might contain one ormore very bad insulators. Forexample, 99oftheinsulators might have farmore than thespecified resistance, but one insulator might have a good deal less, andthe100might still have higher resistance than theone-hundredth ofthespecified resistance perinsulator. Iamhappy tosay (and that enables metocorrect awrong impression which Isee one remark Imade last time about contractors seems tohave produced) that the contractors, who were Messrs. Siemens Bros., thoroughly fellinwith theobjection Iraised, and they were quite willing toallow metotest toa much higher standard than they had arranged tosupply insula- tors toforthe Indian Government;indeed,theyallowedmeto makeanimportant change inthemode oftesting (achange which mighthavebeenserioustothemselves), becausetheyweremost anxious tosupply good material, and because they realised the point Iraised that, had afewinsulators aless resistance than they ought tohave, their resistance would probably diminish in 18 DISCUSSION. 105 time,andweshould havedefective telegraph Linesinspiteofthefrasertesting oftheinsulators attheworke, This corroborates what " Mr.Kempe has said: lettheresistance ofthecable plus the resistanceoftheslateswitch-board beprettygood;then,ifthe leakage beintheswitch-board andthecable begood, allright; ‘ntiftheswitch-board bemuch better, onavery dryday,than youwould suspect, andthecable bebad, itought toberejected, although theresistance ofthetwoinparallel may bevery much higher than the specified resistance. Ifthedefect beinthe cable, that should berejected. ‘Theresistance ofthecable should tethousands ofmegohms, even onawetday;whereas theswitch- ward onawetdaymight have comparatively lowresistanee and setnot bebad. There isnoquestion astothe importance oftesting with highelectro-motive force, andhere again Ifind myself quite in scord with Mr,Kempe; theimportance ofusing high E.M.F. fortestinghasbeenrealisedforanumberofyears.Evenlong ‘agowenever thought oftesting telegraph insulators with aless dectro-motive force than 200volts;oneneverthoughtofputting onafewvolts, butused ashigh electro-motive force ascould be emveniently obtained. Infact, asthelate Professor Fleeming Jenkin pointed outinhis“Electricity,” in1873, leakage resist- suceisnotaconstant, butdiminishes inamount astheEMF, ‘employed inereases. There will nodoubt beexceptions tothis talewheretheleakageisasurfaceoneandwheretheE.M.F,isficiently large toproduce adrying-up action;butinanycase, ifthecircuit istobeworked with ahigh E.MF,, itisall- important thatthathigh E.M.F, should beused fortesting the insolation. ‘Theideaofusing amagneto machine wassuggested {1donotknow whether itwasforthefirst time) bythelate Mr.Schwendler in1871, and outinIndia wehad magneto ‘machines fortesting insulators, Theform ofgalvanometer that ‘Yeused wasaverysimpleandnaturalone:itwasourtongue. ‘Weused towhirl round thehandle ofthemagneto, and ifwe wald taste anyofthecurrent that gotthrough aninsulator, thatinsulator wastemporarily rejected, tobetested more care- fully. Ofcourse wewereusing alarge electro-motive fore> and vou. xvi, 8 106 THEINSULATION RESISTANCE, Ere.[FebUh. Rotemer avery sensitive detector; themachine wasmade byMean. ‘Siemens forthepurpose, anditgave analternate current. With reference tothe modes oftesting which Profesor Jamieson suggested, Iwould like torefer tooneofthem, viz, Fig. 4.Hesays inconnection with Fig.4:“Second,joinup “the voltmeter asinFig. 4,with one side toearth, baving «previously ascertained itsresistance, r,ohms. Also note, C,on “the ampere-meter [why not‘ammeter?’ itismuch shorter}, andseethatitisthesameasbefore.Ifanyleakagetoearth “exists inthedynamo orinany part ofthe circuit, adeflection “on thevoltmeter, VM, will beproduced corresponding tothe “leakage current, ¢,passing through it,and isrepresented “by, 6volts difference ofpotential between itsterminals. “Then, ‘=¢,thetotalleakagecurrent.” Well,now,letus examine that statement. Iwill take the case ofthe lower red lineinFig. 4,having adead earth onit—a frightfully bedstate —and Iwill suppose that dead earth isnear thedynamo,andwe willjoin upthevoltmeter asshown inFig. 4.What willthe voltmeter indicate? ‘The voltmeter tells you, ofcourse, the difference ofpotential between thepoint ofthecircuit towhich itisattached and the earth. There isadead earth on the cirenit somewhere near thedynamo; thepotential ofthelower cireuit near thedynamo istherefore 0,‘The voltmeter will tell youtheinteresting fact that there isnodifference ofpotential between thelead andtheearth: there willbenocurrent through itstherefore thetotal leakage current, according toMr.Jamieson’s formula, is0,orthelineisallright. Butthelineisnotall right; itisinavery bad state. Infact, theindication ofthe voltmeter depends entirely upon where theleakage happens to be,and notsimply onthe amount ofthat leakage. Ifthe leakage isinsome other part ofthecircuit, Igrant youthat the voltmeter may indicate that there isacurrent through it;butif theleakage happens tobenear what isrepresentedasthepositive terminal ofthedynamo inFig. 4,thegreater theleakage orthe worse theinsulation, thelesswill thevoltmeter indicate;infact, the voltmeter will indicate 0when there isadead earth on,and 1803 DISCURSION. 107 therefore, obviously, thetotalleakage current cannot bemeasured eter asgiven inFig. 4bythedeflection ofthevoltmeter. Ifeelinterested inthis particular test, because mycolleague andmyself, after testing @certain installation, and ascertain- ingthatitcouldpassthestandard required bytheauthorities, amused ourselves bytrying this particular test. Wedidnot atthe moment see how foolish itwas, but immediately after- vanis saw the absurdity ofit, Ineed hardly saythat our report was not, based onthat. particular test, as,obviously, itdoes nottellyou anything whatever. Instead ofthedeflection ofthe roltmeter, asProfessor Jamieson thinks, telling you theamount ofleakage, the deflection depends onwhere theleakage is,and wehave the maximum leakage giving aminimum deflection if ithappens tobenear where thevoltmeter isattached. Tdo not agree with Professor Jamieson that amirror galvan- ometer and ascale putontwochairs intheupstairs ofabuilding vould beavery convenient method oftesting. Idoagree with him, however, when hesays, atthe end ofhisletter just read bytheSecretary, that within fiveminutes ofgetting tothespothewouldfindthespotindeed,Iventuretothinkthatifhedid totfind thespot before hegotthere, hewould never getthere tall. Professor Si:vaxus P.Tomson: IquiteagreewithmostofPrtew, theprevious speakers inentirely condemning theuseofavery small electro-motive force intesting the insulation resistance of installations. Itclearly canbeofvery little value asindicating ‘the likelihood of that installation tobreak down. InMr. Smellie’s letter hesuggests amethod oftesting with amagneto machine andasensitive bell, and Ithink says that hismagneto rachine rings bell through 16,000 ohms resistance. Surely that isnotest whatever ofthe insulating resistance ofthe lighting system, because, ifthere isonly aresistance of16,000 ohms, youhave practically amost abominable condition ofthings. Noreliancemustbeplacedonthefactthatthebelldoesnot |Ting, because, ifitdoes notring with over 16,000 ohms, itwould botdetectaleakof20,000ohms,whichwouldteavery |important fault.‘Then,again,Ithinkwhenwearelookingat 108 ‘THEINSULATION RESISTANCE, Ere,(Feb.4h, ‘Pucgee, formule, ofwhichever shape they are—whether those from Professor Jamieson’s paper, orMr.Preece’s formula ontheboard —we ought tobevery careful astowhat case weareapplying the formula. Weought tomake abroad distinction between three general methods ofsetting upelectric light cireuits—those that work atconstant potential with continuous currents ;those which work(principally arelightcircuits) withaconstant current,and with apotential which varies with thenumber oflamps incireuit; and, thirdly, theslternate-current systems, chiefest among them ‘atpresent being those worked bymeans oftransformers. What Iwant topoint out isthat arule that may beperfectly appropriate foroneofthese, may beperfectly inappropriate for another. Ido not want tofind fault with Professor Jamieson's rule ifitistobe confined tocontinuous currents worked at constant potential;butitisquitewrong,andmustbewrong,for constant current worked with varying potentials. See what it Innds usin. ‘The insulation resistanceistobeproportional tothe electro-motiveforce—Ihavenoobjectiontothat—andtobein- versely proportional tothe number oflamps. Take acase, and seewhat itlands usin, Suppose anaresystem oflamps in series, requiring, say,50volts each. Ifyouhave three ares, you want 150 volts. Suppose, itistead oftaking three arelamps, youhavesixty,thenyouwant3,000volts,Dividing yourvoltsby thenumberofarelampsbringsyoubacktothesameinsulationresistance forsixtylampsasforthree. IfEistoagreeinproportion toN,theinsulation resistance istobethesame when you have 150volts and when you have 3,000 volts onthecircuit, which is anabsurdity. Again, take Mr.Preece’s rule, where heintroduces thelength ofthe line, Idonot quarrel with that rule fora constant potential system, butforavarying current itisworse *than Professor Jamieson’s, because notonly istheelectro-motive force tobedivided bythenumber oflamps, which would bring it down toProfessor Jamieson’s, butitistobedivided bythelength ofthecireuit. Well,now,ifyouhavetenarclamps,andtheyare comparatively close together, you have very fewinsulators onthe line between onelamp and thenext lamp;butswpposetheseten Imps aresome distance apart, you have many insulators onthe 196,) DISCUSSION. 100 cireuit, Doyou want alessinsulation resistance because youPrtamor have many insulators and more sources ofleakage? Ithink that inthat case you ought toinsist onahigher insulation resistance. Ithink you want Labove the line intheformula, and notbelow it;youwant more perfect insulation astheline islonger, not les. Further, applied toalternate-current work, one has to remember that itisthe maximum, and nottheaverage, electro- motive force that breaks thecircuit down; andifthealternating EMF, isrising above and falling below theaverage value, then dlearly adifferent ruleiswanted foralternating-current systems, Takethecase oftheproposed Deptford system, with its10,000 tolts. Ifwearegoing todivide the 10,000 volts bythenumber {lamps onthecireuit—there arenolamps onthemain cables, batsuppose there are100,000 lamps onthesystem—dividing yourelectro-motive force bythe 100,000 lamps, youaregoing tobave only one-tenth oftheinsulation resistance that youwould laveifyou had 1volt with onelamp onit.According tothat formola youwould make theinsulation resistance one-thousandth ofwhat itshould bewith 1volt onanordinary 100-volt sytem. Mr.ALEXANDER StEMENS: Might Isay,asMr.Preece isnot here(Mr. Preece had leftthemeeting], that wespoke about thatpoint yesterday? Insuch acase Nwould be0,andthat is whyheintrodaced N+1,because that primary circuit from Deptford would have nolamps onitatall. Professor 8.P.THOMPSON: Iquite understood whyMr.Preece introduced the1,butmyobjection istoarulewhich gives less insolation for100,000 lamps than foronelamp. Mr.ALEXANDER SteMENS: Yes; butyouhave notgot100,000 lunps onthat circuit; youhave only thetransformers onthe Professor S.P.Taomrson: This formula wants writing differ- tally fordifferent cases;ifitisrightforone,itiswrongfor wouter. Mr.ALexanver Sremens: Ionly interrupted because your istration wasagainst Mr.Preece’s intention. Professor S.P,Tuomrson: Then Ithink Iamright inasking Mr,Preece todefine howfartheformula istoapply. 10 ‘THE INSULATION RESISTANOE, Ere, (Feb. 1, Fetemor Another point that Ithink weought tobecareful about is astotheway inwhich this formula should beused according to themethodinwhichthecablesorconducting wiresarecarried, Clearly the same formula isinadmissible forunderground con- ductors aswould beadmissible foroverhead conductors, and for overhead conductors therule ought surely tobedifferent accord ingtotheway inwhich theoverhead conductors arestrung. Thave spoken inthisroom before now oftheabsurdity ofusing Brightshackles foroverhead wiresateverypole.Theyareall right where you have toshackle offattheendofalines batit isabsurd touseshackles, with twochances ofleakage and bad insulation, atevery pole, instead ofusing some other insulator which has less chance ofleakage. That absurdity becomes multiplied when you aredealing with high electro-motive force with anydistributing system. Lastly, arewequite clear that these formule forinsulation resistance arenot going tomislead usaltogether? Isnotthe test wewant toapply tothevarious systems atest ofquite a different order? The test isnot what number ofohms the insulation isto-day orto-morrow, but what will break that insulation down. Isthemeasurement byamirror galvanometer, andwith.a battery ofapower below theusual electro-motive force *ofthecircuit,reallyanytestastothelikelihood oftheinsulation ofthe system tobreak down? Iwill give you aparallel case. Suppose youarebuilding adynamo, You have thecoils, inmany cases, made with cotton-covered wire which isvery carefully soaked, say, with shellac varnish, and these coils areput into a steam-heated chamber and heated forsome time, todrive offthe spirit from theshellac. When these coils arebrought out, after aweek's heating, and aretested fortheresistance between the copper and theframework onwhich they arewound, you may getsomething considerably under 100,000 ohms. Areyougoing tocondemn that dynamo because ittests solow? Ivery much doubt whether halfthedynamos inthemarket willshow anything like half amegohm between theiron and thecopper asthey leave thefactory. Itseems tomethat thequestion isnotwhat the tested insulation is,but really what will break down that 193 DISCUSSION. m ingalation, And surely, ifthepractical question that presents Protemr itself inthat way indynamo machines, inparts ofthecircuits, inswitch-boards, andsoon, iswhat electro-motive force willbreak: theinsulation down, then these elaborate systems oftesting for insulation resistance arereally very much beside themark. Mr.J.N,Snoouonep: Theremarks ofProfessor Silvanus me Thompson only confirm Professor Jamieson’s paper and Mr. Proece’s remarks upon it,inthat itisimpossible tostate any oneformula forinsulation that would apply toallcases, even of onekind ofinstallation. ‘Take Professor Jamieson’s formule, which isstated inhispaper tobebased upon thenumber ofincan- descent lamps of16candle-power. Professor Thompson could scarcely have borne that inmind when hespoke ofand applied thstformula toarelights, because Professor Jamieson speaks alrays ofincandescent lights. Histables show that, andIthink, therefore, itishardly fair that itshould bemade toapply toanumber ofarclights inseries—that is,with avariable potential. Professor 8.P.Tuoursoy: Will youexcuse me? Hedoes not ‘9ythatitapplies toincandescent lamps—with constant potentials certainly—only ;they areused constantly inarocircuits. Mr. J.N.SHooenep: Professor Jamieson's formula has been framedmoreasanassistanceforarrivingatafairinsulationfor |theordinary ran ofinstallations ofincandescent lamps. My |cbject ismerely topoint out,that from thevarious uses ofthecarrentitisimpossible ‘togiveoneformulaapplicable toall,as |Mi.Preece suggested thatweshould trytoarrive ataformula thstwould beapplicable forallcases. Professor Silvanus Thompson hasjustdistinguished between incandescent lights in panllel circuits, arelights inseries, transformers, andalternate- arent systems. Inaddition, particularly intowns, where thereaaybealonglengthofuntapped main,orwherethemajority dfthework consists ofincandescent lamps. Again, there may bemotors introduced, orarelamps, andother circumstances ‘hich would cause therequisite amount ofinsulation tovary much indegree. Hence the difficulty offorming one single formula that. shall apply inalleases; thecircumstances ofeach ne ‘THE INSULATION RESIBTANCE, Ere. (Feb. 1th, Msormea, MUSE beconsidered, anddulytaken intoaccount inprescribing therequired amount ofinsulation. Ofthethree formule named inthepaper, Professor Jamieson’s tries torepresent,bymeansof thenumber oflamps, that ofthecorresponding connections and joints which would occur inincandescent light installations; while M.Picou laysstress upon thelength ofwire used; andthe Society ofTelegraph Engineers’ rule dwells upon thedensityof the current. Each formula isbest suited to certain eireum- stances, but hardly canitbemade equally well adapted forall installations. Mique, MtC.E,Sragsourrri: Ishould liketoaddmyexperience totheremarks that have been made byMr. Kempe, byMr. Sidney Walker, and byother speakers, astothe necessity of testing electric light wires with avery high potential. ‘The difference ofcurrents carried bytelegraph wires and electric light wires isofcourse very great. Acase came under mynotice intheearlydaysofelectriclighting, ‘whentheSociétéde VElectricité came over here, M.Baudet being their representa- tive. Acable wasput down ontheDistrict Railway. That cable came into myhands totest. Itwas tested intheordinary way with battery power, andwecould get noleakage whatever; butassoon asthecurrent wasputonthecable from thedynamo, wecould positively seethesparks going into thewall; showing that, although wehad agood battery power fortestirig, the current wasquite insufficient foritspurpose, and useless when compared with what was necessary fortherequirements. Mr. Kempe also pointed outthe advisability ofdisconnecting the instruments and test-box connections from the main circuit. ‘That, again, Ithink, ismost necessary todo, Irecollect, some years ago, when theMont Cenis Tunnel was being built, the cable for itwas manufactured here atthe Gutta-Percha Com- pany’s works, and Ihadthetesting ofit:Iwent tothetesting room andsawtheapparatus. OfcourseIknewnothingofhow the connectionsweremade,allwiresbeingoutofsight.‘The test ofthefirst coilshowed nomovement ofthespot whatever on thescale, IsaidtoMr.Willoughby Smith, “Are you sure that “all theconnections areright 2” Wewere then standing ona 18835 DISCUSSION. us sheetofgutta-percha. ‘Thetemperature oftheroomwas,Ithink,ws 65°Fab.; thesun hadbeen shining onadesk bythewindow, hich was warm. Hetouched the instrument with one hand, andthemahogany desk inthe window with theother, and thespot went offthe scale altogether. That, Ithink, will thowthenecessity ofhaving thetest-box connections offwhile testing, themore 60asthey would inallprobability notbeinso dryastateasthemahogany deskwas. Some very good information hasbeen given usbyProfessor Siranus Thompson;butIamafraidthatifyouadoptthesystem thathehasspokenof—i.e., offinding outwhatcurrents would leakthecable down—this would notalways apply, andwould depend upon theageandthedeterioration ofthecable, forwhat might notbreak itdown thefirst sixoreight months would in ayearortwo'stimegiveaverydifferentresult.Ithinkithas wenclearly shown bytheremarks ofmost speakers to-night thatintesting electric light cables they should betested with as highavoltage asisgoing tobeused, ifnot with ahigher essure, and free ofallconnections. Mr.C.T,Fuzetwoop: Icanspeak from past experience, con- sr firming whatMr.Spagnoletti hasjustsaid,alsoMr.Kempe, about "*"°*" thenecessity ofdisconnecting wiresfromtestboxeswhentesting.Someyearsago,Iremember drawinginsixtywiresoverBlackfriarsBridge, made upof3x20,from different manufacturers, and ahterthecables were inplace instructions were given forthem to betested. Itwasfound that they only gave 30megohms instead ofeeveralhundred megohms. Everyleadthatwasputdownto ‘thetesting room was condemned. Atlastitstruck methat Pusiblyitwasthetestbox.Iwentandcutthewireofftheteattox—thesamewire—and joinedittothesamelead,andwentdowntoseittested,wheneveryonewaspronounced perfect. Fromthat timetothepresent wiresarealways disconnected whenbeing tested atthecentral station. Inreference totesting with a |magnetomachine, previous to1870itwasthepractice ofmen |belonging totheNational Provincial Telegraph Company totesttheirwiresbymeansoftheABCinstrument, andasarulethey\‘8%goodresults;buttheircablesatthattimewereoverhead, and aa THE INSULATION RESISTANCE, Bre. (Feb, Mh, MSsgeooa, OfVery low insulation. Iremember one ofthe circuits being transferred totheunderground system between Winchester House and Clerkenwell firestations, and thecircuit worked very badly indeed, The men who had been inthe habit oftesting with these magneto machines again and again complained that the underground wire was faulty, while the men who tested the underground wires declared that they were perfect. Iwent to theendofthecircuit with themen using themagneto machines, andfound that they gotamove ontheir indicator assoon asthey began togrind, butwhen tested with theWheatstone bridge the wirewasperfect. Wethenwenttothemiddleofthecircuit,at thecomer oftheOldBailey, divided thecircuit, putthemachine imagain, andfound that thewire wasperfect;therewasnotsufi- cient capacity inthedivided circuit forthe charge and discharge ofthe wire tomove the needle. When we tested the whole length there wassufficient ofthecharge and discharge alone to move the indicator round. Itmay bethat the same difficulty may arise iftheinstrument isused fortesting electrie light wires. wezewert Mr.M.Hotnoyn Surrt:Iwishtosay,Mr.Chairman,that Irather fancy the question ofinsulation resistance forelectric light installations isnotgoing tobedetermined bythis Society, oranyother Society, fixing thevalue ofKinanyoftheformule presented. Inmyown experience the manufacturers ofelectric light leads have always been surprised atthe lowinsulation that Thave specified, butthey have been equally surprised when they have come across the condition that that insulation should be maintained and guaranteed foraconsiderable length oftime. Itisvery much more aquestion ofthe endurance than ofany very high andfanciful insulation attheontset. Wearedepend- inginthis matter almost entirely upon themanufacturers, and itisforthem tobring ussome material forwhich they can guarantee long duration ofamoderate degree ofinsulation. It ‘would then beaquestion ofhow much wecan afford tolose in anyinstallation through leakage, when weknow that theloss is going tobeconstant. Itherefore wish todraw especial attention tothefact that itisnotthefixing ofahigh initial value forK, but itisthe ascertaining ofsome material that can berelied 1899) DISCUSSION. us ‘uponandervarying conditions, sayofheatandcold,drynessandMr,3elrora moisture,thecondition ofwhichwillnotbedeteriorated bythepassage ofthat very small current that now, likethelittle rift within the lute, will make themusic incomplete. Ifsuch material canbesupplied, itmay then bepossible tofixavalue forKthatwouldreallybeaconstant. Icannot agree with those speakers who arecontent with delicate tests. Imuch prefer that allwires should, when possible, betested under the maximum current likely topass through them. Anillustration oftheimportance ofthispractice cceurred tome only yesterday, when Ideclined tosend a machine abroad until ithad been soproved, though the instru- ment tests were perfectly satisfactory. There were four circuits inthemachine; three stood remarkably well, butthefourth gave sayatonce. Mr.F.Wytes: Ithink, Sir,with reference tothe figures ¥rWyle. given byProfessor Jamieson regarding indoor lighting, there canbenodifficulty inmaintaining hisstandardofinsulation; batdirectly you come tooutdoor lighting, when you have to deal with smoke, fog, and atmospheric influences, there is dificalty inkeeping uphigh insulation. With regard totesting, Iwould rather seesome system of regular testing. Myown plan istohave tests taken weekly. Agalvanometer iskept connected up,theforeman takes the deflections, and Iwork outthe insulation resistance from his ports inmyown office, ‘Nothing seems tobeshown onthediagramswithregardto testing the machine forcontact. My own experience isthat ifthemachine isrigorously tested @breakdown may besaved. Itdoes notmatter, probably, ifonly onecontact occurs onthe machine,twobeinggenerally required tocauseashort-circuit; byregular testing, therefore, thefirst contact may bedetected, andabreakdown prevented. AstoMr.Walker’s remarks about gutta-percha being used forelectric light cables. Myexperience isthat with the heavy wireused there isatendency toproduce decentralisation; itmaybeaquestion oftime, butsome day orother thecore us THE INSULATION RESISTANCE, Ere. (Feb. 1th, Me.Wyte will drop through thepercha, and the copper will bealmost dareatthebottom, Melast, Mr.W.Lant Canrenter: Idesire foramoment toem- phatically confirm theremarks just made byMr.Holroyd Smith, because Ihappen toknow ofseveral contracts where that point ‘wasinsisted upon bythose who were buying thecable. Fora yenr atleast—and inmany cases foramuch longer period—the insulation hastobemaintained uptoagiven standard atthe expenseofthemakeroftheeable;anditisnotaquestionof thevalue ofK,aspointed out. aie.Waner. Mr. Stoney F,Water: Ishould like tosaythat Ihave gutta-percha cables that have been standing fiveyears without that fault being met with;andifIweretogivespecification forduration, Ishould saynotless than five years; one year would benogood atall. Sclios «SirHenry C.MaNce: With reference tothedurability of theinsulators, Imay mention that inIndia wehave hadagood deal ofexperience both with india-rubber andgutta-percha. If thelatter material isnottobesubjected toagreater voltage than 100or200volts, Iseenoobjection totheuseofgutta- percha forelectric light leads. Iassume, ofcourse, that the carrying capacity ofthe wire issufficient, and that thecable is buriedindampground, Anexcellent exampleofthedurability ofgutta-percha isafforded bythe Persian Gulf cable, which at thehead ofthePersian Gulf islaidfortwoorthree miles through thedate gardens, from twotothree feetbelow thesurface. Ithas been there formore than twenty-five years, and hasnever given theslightest trouble. Under such circumstances gutta-percha is practically imperishable, but exposed tothe dry airitisun- doubtedly inferiorinlustingqualities toindia-rabber. Thelatter material wefound most subject todeterioration when exposed to alternate wetanddry. Inanelectric lighting system thequestion ofjoints becomes avery important one. You will find intheProceedings ofthe Society, vol. iv.,1875, anaccount ofaseries ofexperimental joints inold gutta-percha and india-rubber cores which were ‘under observation fortwo orthree years. There was some se) DISCUSSION. i Aificaltyinmakingwhatcableelectricians wouldconsiderageeuy perfect joint inoldgutta-percha; butalthough theinsulation vould gradually fall,thejoint appeared tobemechanically perfect, andIdonotthink anyoftheseries would have been prejudicially affected bythecurrents itwould beconsidered prudent touseon4gutta-percha insulated electriclightmain.Theindis-rubber jpints made inHooper's core showed nosign ofdeterioration. Asregards theformula before us,Ithink inpractice weshall, ashitherto, have torely onthe common-sense and experience of theelectrician incharge, who would probahly adopt different standards and apply different tests under thevarying conditions. Wemayassume that periodical tests willbetaken inevery well- managed installation, nottoascertain thepotential theleads will bear(that hasbeen done inthefirst place), buttokeep arecordoftheinsulation ofthewholesystem.Atthecentralstationthe electrician would probably prefer touseasensitive galvanometer tadamoderate battery power, but under conditions when a delieate galvanometer cannot beused hewould naturally avail himself ofahigher voltage;itisimpossibletolaydownahard- aud-fast rule. Ttmay sometimes benecessary, forthepurpose of leclising faults,totest.portionsofthesystemawayfromthe central station where shelter isnot available, When compelled totest under these conditions, Ihave always used asmall Siemens lvanometer with asuspended astatie needle, inconjanction with portable bridge andbattery. Professor W.E,Avaton: Without delaying themeeting aotesor moment, might Isuggest toSirHenry Mance thatfroman” electric light engineer's point ofview thecable is,asarule, notinwil,dampordry,batisintheair,eitherinaroomorina trench under the streets; and under those circumstances (Sir Henry) would your experience lead you toprefer gutta-percha to ‘ntirubber? For example, you will remember that inthe telegraph offices inIndia thegutta-percha wires have had tobe replaced byindia-rubber-covered wires, forthesimple reason thatthegatta-percha wasfoundtoseparateintosmallpiecesroundthewireabout aninch long, exactly likebitsoftobacco pipe 18 ‘THE INSULATION RESISTANCE, Ere. (Feb. 4, Fotewce strung along thewire, andwhich merely served tokeepthe copper away from thewall. SirHenny Mance: That isquite correct. Professor W. E, Ayrton: For the insulation of wires for electriclightpurposes, wherethewiresaretobeinair,andnot inwater ordamp earth, would you not prefer india-rubber to ‘gutta-percha?Sir Hesay Mance: Yes. Syayet SirAusent Carret: Imight justaddthatintheinterior of telegraph buildings inIndia weuseindis-rubber-covered wires solely, asProfessor Ayrton hasjust said. Pa Mr.C,E,Sracyouerri: May Isayjustaword onthisgutta- percha question ?In1852 some wires were putunder ground for ‘about amile inthePaddington yard, inearthenware pipes, with the oldgutta-percha covering pure and simple, without com- pounds; itwasofavery light colour. Being atelegraph circuit, weworked with only alowvoltage. About sixyears agowehad ‘occasion torenew some underground wires, and onopening the ground through the yard wecame across the pipes containing these oldgutta-percha wires. ‘They were ofNo.16gauge copper, with gutta-percha covering toNo.3—a thicker coating than is nowgenerally used fortelegraphic purposes. Weexamined and tested thegutta-percha, andfound itinperfect condition, although ithad been down forabout thirty years, We joined the old wires towires wewere then laying, asthey raninthe right direction forour work, and they have been working well up tothepresent. time, and Ihave nodoubt they will last many more years. Mews Mr,F,Wees: Might Ibeallowed tosay,Sir,that Ididnot criticise gutta-percha asused forordinary telegraph work, because weknow itisthe proper thing inEngland? But Idocriticise itwhen itisused forelectric light wires with heavy cores. ws, ‘MrH.C. Dosova: Might IaskProfessor Ayrton andothers whether they could not give usaformula forinterior surface leakage, where exposed contacts would come into theformula ? "tee Professor W.E.Avrros: Idonotthink there would beany difficulty ingiving aformula, butIdonotthink oneformula can 10993 DISCUSSION. 19 corerevenonecase—that is,onekindofinstallation—for theroteworsimple reason that onestandard ought toberequired forthe” dynamo, another standard for the insulation ofthe gutta- pereha, orindia-rabber-covered, conductor, and athird standard forthelamp-holders, fuses, and soon,where surface leakage is mainly important. Ifasingle formula betaken tocover the wholething,then,whatever numberbetakenforthecoefficient K, youarevery likely, ifyouarebound byoneformula, tocondemn ‘8good installation and topass abad one, forthesimple reason thatthere may beaserious leakage inthecable foracable, and which ought tocondemn thecable, butwhich may notdiminish thevalue ofthe parallel resistance ofthe whole installation ficiently tomake the result worked out bythe formula low enough toensure condemnation. TheCuaraan: IquiteagreewithwhatfellfromMr.HolroydBeSisson.Saith—that thegreatthingistobesureofmaintaining amode-ntelygoodinsulation, ratherthantomerelystartwithavery highone, ‘There alsoseems tobeaconsensus ofopinion that it isdesirable totest theresistance oftheconductors with ashighpotential asthatwhichistobeused.Inmyjudgment, itisdesirable that weshould beable totest the insulation resistance ofthetwoconductors fromtheearth atalltimes, whether running orstanding—particularly when running—and ithaslong been known that itisquite easy todo that. Itisdone inthisway:— A *o-B E Itissimply @modification oftheordinary Wheatstone bridge. ‘Suppose +and —tobethetwoterminalsofadynamoandthecon- ductorsconnectedthereto,andthatErepresentstheearth,witha certainleakagepassingfrom+toE,andacertainleakagegoing frum—toE. Connect +to—through aresistance which can 120 ELECTIONS. (resen, Pea,bedividedintotwopartsatA,adjustable inrelationtoeachother;Horie connect AtoFthrough agalvanometer; adjust theresistances +Aand A—tillthegalvanometer isnotdeflected: then the ratio +AJA —willbeequal totheratio ofthetwoearth insula- tions, Connect either conductor, say +,toearth through o known suitable resistance, +BE,andagain adjust theresistances +A,A—. These two experiments give ustwo equations to determine thetwounknown quantities, theinsulations from earth. Gentlemen, itisnow my duty topropose that the best thanks ofthis meetingbeaccordedtoProfessorJamiesonforhis interesting paper. Itcertainly hasgiven risetoaninteresting, and [have nodoubt avaluable, discussion, ‘The motion wascarried unanimously. ‘The Cnainaax: Another paper hasbeen announced forthis evening, butitisofsuch alength andofsuch interest that it would bevery unwise forustobegin atthis late hour, and in fact Iunderstand that there isarule that nofresh matter is tobetaken after half-past nine; wemust therefore postpone it, tothenext meeting oftheInstitution, which willtake place on the 21st instant. Aballot took place, atwhich thefollowing were elected :— Foreign Member: Virgilio Machado. Membere: Charles Laurence Baker. Norcliffe George Thompson. George Cameron Sillar. Edward Willmore. Associates: John Richard Bainton. John Macfee, jun.EmileGuitton. |GeorgeHenryRew. Robert Cattley Jackson. Brabazon Rutherford. Edwin Isidore Lloyd. H Studenta: Joseph Pratt Sleigh. |George Herbert Thornton. Henry Walker. ‘The meeting then adjourned. am COMMUNICATIONS. INSULATION RESISTANCE OF ELECTRIC LIGHT INSTALLATIONS, ByWmuam McWaurter, Member. Installation atDomira House, Partick—50 lamps, 16candle~ power, and60volts. Tested 9thJanuary, 1886, when insulation reistance =14megohms. Installation atDomira House, Partick—90 lamps, 16candle- Power,and60volts.Tested16thJanuary, 1889,wheninsulation istance =35,000 ohms. Noe—This installation inthree years hasbeen more than doubled, and thedrop ininsulation islargely due toleads and fittings inconservatory, which arealways dripping wet,asitis tathed down with ahose-pipe. Installation atHeriot-Watt College, Edinburgh—400 lamps, Weandle-power, 100volts. Tested April, 1888, when insulation resistance =30,000. Installation atHeriot-Watt College, Edinburgh—400 lamps, Weandle-power, 100volts. Tested 15th September, 1888 (after Inildings had been standing without heating forover three ‘onths), and found insulation resistance =6,000 ohms, Tested again 22nd September (heating having been going on ‘ace15th), andfound insulation resistance =11,000 ohms. Thefallininsulation wasentirely duetomoisture, which was (eposited heavily ontheslate baseswitches. Installation atFree Prese office, Aberdeen—200 lamps, 16 audle-power, 100volts, ‘Tested 30th August, 1888, and found inmlation resistance =70,000 ohms, Imayaddthattheaboveexamples areaveragepractice, and ‘Yaymuch what wegetininstallations onboard ship. VOL xvm. 9 an COMMUNICATIONS. Indynamos wefind itmuch more difficult towork upto Jamieson’s rule;infact,itrequiresgreatcareinthequalityof materials and inthe application ofsame tomaintain this standard. Fanapar Euzcrareat, Worx, Carnie Sranz, Govax, GLascow, 181k February, 1889 ByE,W.Becinasaus, Associate. Oftherules onthe insulation resistance ofelectric light installations presented tothe meeting ofthe Institution of Electrical Engineers fordiscussion byProfessor A.Jamieson, that whichappearstobethemostgenerally suitable isR=10'6; this constant, 10', being not too high forcontractors towork upto,andnottoolowforelectricians topass. Butsuch arulehasanarbitrary character, anddoes notmeet therequirements ofsome oftheelectric lighting cables andwires inusewhich have avery high resistance toleakage. Reference ismade tothe lead-cased cables inwhich the eon- ductor isinsulated with »fibrous material steeped inahydro- carbon. Though high insulation perseis notrequired toproduce agood sound cable orwire, yetitisdesirable that thenormal electric resistance ofthe insulation should bemaintained within some30percent.ofthefactorytestsafterthewireiserected; otherwise thecables contain defects, probably caused bydamp or byrough usage, which aredetrimental totheir lifeand contain theelements ofmischief.Itisevident thatwerequire different values ofKfordifferent typesofcable.IfweconsiderthefactorytestsofaNo.5B.W.G.Tead-cased fibre-insulated wireandofNo.18B.W.G.wireofsimilar make,whichare8,0002and15,0002permilerespectively, and‘compare theseresistances withthe300or6009permileofa vuleanised india-rubber-covered cable, itdoes notseem right that thesame constant (10*) should beapplied tothetests ofthese twowidelydifferent classesofwire.Dampness oftheinsulating coMmUNICATIONS. 188 naterial, especially ofthe lead-cased cables, will generally let down itsmegohms inasurprising manner, andwithin reach of themost humble instrument tomeasure; butthisisnotalways me'sfortune. Cases have occurred within thecognisance ofthe iterinwhichtheadmission ofwatercausedonlyapartiallossof resistance,Oneoftheseleakageswasmeasured—or, rather,that the cable was—which showed adecrease ofinsulation from $3000 0to1,700 9. The cause ofthefault wasapparent, and it wascutout, but under the rales itneed not have been. Theabundant use ofcompound intheinsulation ofjoints is tobecondemned. ‘Thewriterhasemployed asleeveofthesame, of similar, insulating material tothat which theconductor iscovered with: this isplaced over theendofoneofthewires before itissoldered, andthen drawn into position over thejoint. Thesleeves must belong enough tolapover and embrace the insulating material oftheconductor ateither side. India-rubber cement may beused sparingly, and the ends bound with fine twine. Aslightmodification ofthisinsulation jointisrequired forTjoints, Taraary 16th, 1889, ByCuanues Briont, Member. Iwas rather surprised tohear itsuggested during thedis casion that gutta-percha was, inageneral way, preferable to wuleanised india-rubber forelectric light purposes; though it ‘Appearstomethatitmayperhaps besuperior intheparticular case ofmines orcollieries, under certain circumstances—where theground issubject towater, orwhere itis,atanyrate, moist damp. Itiswell toremember thefollowing facts, however :—Gutta- percha is,sotospeak, essentially a“wet bob”: inwater itmay besaid tobeimperishable, and not only retains itselectrical qualities, but absolutely improves ininsulation byage toa considerable extent; whereas theresistance ofindia-rubber, even henvuleanised, goesoff,asarule, byagewhen inwater. Ina 1 COMMUNICATIONS, rysoil, however, vulcanised india-rubber generally proves. more durable asaninsulating medium than gutta-percha; andwhen exposed totheoxidising influence ofair(asinthecase ofan aerial line), gutta-percha rapidly deteriorates bycracking, espe cially inasmoky, orotherwise foul, atmosphere, and itsresistance issoon fatally lowered inconsequence. Vuleanised india-rubber isfound tobemuch more durable under the same circumstances. Invery lowtemperatures, moreover, gutta-percha becomes hari and cracks still more readily. ‘The highest temperature before which any species ofgutta- percha gum beginstosoftenis120°Fah.,whereasnomixtureof valeanised india-rubber softens until about theboiling point of water isreached. No.4oftheSociety ofTelegraph Engineer’ “Rules forthe Prevention ofFire Risks” suggests, very pro- perly Iventure tothink, that “whatever insulating material is “employed, itshould notsoften until atemperature of170° Fah. “has been reached.” India-rubber when vuleanised, being somuch less affected physically orelectrically bytemperature than gutta-percha, itis better suited asaninsulator toaconductor carrying high currents. Iflarge currents areused, oriftheline beexposed tothesun, theconductor ismuchmoreliabletofallexcentric inagutta-percha covering than inonecomposed ofvulcanised india-rubber, owing tothegreater sensitiveness ofgutta-percha totemperature; with agiven limit ofthickness forinsulating purposes, thisis liable tobethe cause ofserious faults, Inorder toavoid theabove illeffects ofoverheating, the conductor would, inagiven circuit, require tobeheavier ina gutta-percha core than with avulcanised india-rubber covering, and, consequently (which isamore important item from an expense point ofview), agreater weight ofgutta-percha willbe necessary than inthecase ofthe india-rubber core, inorder to givethesame thickness, soastooffer thesame insulation resist- ‘ance beyond theextra amount which might berequired owing tothedifference oftheir specific resistances, February906,1889, COMMUNICATIONS. 135 PROFESSOR JAMIESON'S REPLY TO DISCUSSION ON HIS PAPER ON “THE INSULATION RESISTANCE “OF ELECTRIC LIGHT INSTALLATIONS.” Tam very glad that mycommunication hasgiven risetosuch mextensive, complete, and instructive discussion, Inreading mertheprinted reports, Ifind that every point raised inthe per, and almost every fresh idea expressed bymembers, has wenmet and answered byoneormore ofthespeakers and ‘omespondents. Theduty ofreplying isthereby rendered much wer than itwould otherwise have been. Mr.Preece hit the nail onthe head when hesaid, “There “must beadistinct rulefordeciding what shall betheminimum, “insulation resistance tobegiven toinstallations under theworst “circumstances ;”and, “How isitpossible forlocal authorities, “eattered allover thecountry, todecide what isadefective “main, orwhat. isadefective lead, unless we,the Institution of “Electrical Engineers, areable todefine some lawthat shall“decidethedifference betweenabadandagoodconductor?” Igather from theremarks ofseveral ofthespeakers that it would benecessary tohave different rules foreach ofthe fllowing cases :— 1.Installations worked bycontinuous currents atconstant EMF. 2,Installations with constant currents, such asarecommon toarelight circuits orwith low-resistance incandescent lamps inseries. 3.Alternating-current systems, These would require tobesupplemented byconstants for— (@)Subterranean main supply cables. (0)Overhead main supply wires. (.)Thecomplete system ofleading wires within buildings. (4)The switch-boards, and thefittings, separately and combined. (¢)Thegenerators ofelectrical energy. (f.)Thetransformersofelectricalenergy. 1would therefore suggest thattheInstitution ofElectrical 138 COMMUNICATIONS. Engineers should appoint aCommittee todraw upandsend outforms, including each oftheabove cases, toelectricians andfirms, requesting them tofillthese upand return themto theSecretary. The desired information would then beobtained ‘upon auniform and comparable system, which would enable the Committee todecide upon themost suitable rules and constants. ‘Mr.Kempe draws attention totheomission inthepaper of notdistinguishing between theinsulation resistance ofthecables forlends and that ofthefittings. When Ifirst began totet electric light installations (i.e., when Iwas fresh from submarine cable work), Ialways tested theleads bythemselves, andlaid down arule of“one megohm perlamp forevery volt inthecase “of theleads bythemselves.” Now, however, when called into report upon completed installations, Ifindthat contractors object tothedisconnecting ofalltheir leads from thefittings andlamps, forthey say,“To disconnect allthewires every time aspecial “test hastobeapplied would domore harm than good;”and usually add, “Apply your tests tothe whole installation asit «stands, seeing that istheworking condition ofaffairs.” When Ihave theopportunity ofdrawing upspecifications and of inspecting the whole ofthe work, Inow insist upon insulation resistance tests being applied toallthe coils ofinsulated wire before they leave thecontractor's works;thecoilsbeingsub- merged inwater fornotlessthan sixhours immediately previous to,and during, thetests, The coils are, ofcourse, afterwards carefully dried, Inthe case ofcoils that donot considerably exceed the minimum insulation resistance allowed, Ihave them subjected whilst inthewater toan E.MLF. 50percent. greater than the proposed working E.M.F. forsome considerable time, andthen tested again. Should anyappreciable reduction intheir insulation resistance have taken place, duetothe stress orfault- ferreting currents, the coils arerejected. The leads areagain tested when fitted up,before theswitches, &.,areattached, and finally thewhole combination, immediately previous toandafter thepreliminary ortrial runofthecomplete plant. Istill think that theapplication ofhigh E.M.F’s isonly necessary forthe purpose ofputting asevere stress upon theinsulating material of coMMUNTOATIONS. 1 thecablesandleads. Ifaweakness orsmallfaultexists,itwill beopened upbytheproper application ofthis stress, and then thecorrect. measurement ofitsvalue (and even theposition in tomecases) canbemore accurately obtained bytheemployment ofalower E.M.F. combined with more delicate apparatus than ‘ould beconveniently orsafely used with thehigher E.M.F's Inthecaseofimportantdistrictmainsandsub-mains,itwill befound absolutely necessary tocarefully test them periodically adrecord the results oftheir insulation resistance quite independent ofthat ofthe distributing apparatus and the leads, &e., inthe various places orbuildings supplied with curent. The very simple and easily applied test described by Dr.Hopkinson (and referred tobyafoot-note inthepaper, see }-58) will, however, befound very useful and convenient, since itisspecially adapted fortheworking conditions ofinstalla- tions, ‘The person incharge ofaninstallation does notneces- swilyrequire toknow anything about theprinciple ofthetest, or tobeeven able towork outtheequations, ifheissupplied with table giving him theinsulation resistance corresponding tothe ‘employment ofthevarious resistances inthearms ofthebridge. Allhehastodoistovarytheseresistances untilhefindsnodeflec- tiononhisgalvanometer, and then refer tothe table and record theresult, Infact,itwouldbepossible toarrange foranastomatie recorder oftheworking insulation resistance ofan ‘astllation upon this principle. ‘Themoral effect ofadefinite setofrules and system of testingforinsulationresistanceuponcontractors, andstillmore ®upontheworkmenemployedinfittingupinstallations, was tttouched upon byanyofthe speakers. There cannot bethe lightest doubt but that acontractor and hisforemen have a ‘hiphand over their men ifthey cansay,“Well, now, Jack, you “must take special care thistime with theleads, joints, and “fitings, forProfessor Spot iscoming down with that magic “mirror ofhis,which willshow uptheweak points inyour work, ‘tadifthedeflection should betoogreat youwillget.time for ‘reflection atyourownexpense!” 128 COMMUNICATIONS, COMMUNICATION FROM SIR WILLIAM THOMSON, PRESIDENT. Inspeaking ofthepractical superiority ofcopper over ironfor high-speed telephony ortelegraphy inmyrecentPresidential Address, with reference towhich Iquoted Mr.Bennett's practical experience, Ishould also have quoted thefollowing from Mr. Preece’s paper ‘On therelative Merits ofIron and Copper Wire “for Telegraph Lines,” read before theBritish Association in September, 1885, which Ihad overlooked when Iwas pre paring myaddress, but which Ihave since seen with much interest. Itiscontained onpages 908 and 909oftheBritish Association Report ofthe1885 Aberdeen Meeting :— “<The socond series ofexperiments were couduoted between London andNewcastle,andweredesigned,asstatedabore,tooattheworkingeflelencyofthecopper wire ascompared with iron wires, ‘They were conducted by ‘Messrs, Chapman snd Eden.Sit?ORE#verydeadeaperirtyoveriron,thepetssng8|Copper Tron Simplexworking 418. ,345wordsporminate, Itisanticipated that thesuperiority ofcopper over iron indicated bythese‘experiments willhaveabeneficialandeconomical influenceonourtalegraptsyrtam, and that itaextended use will enable usnotonly towork better, bat todispense with intermediate repeaters inmany casos where, onlong line, they arenow necessary. ‘The most interesting point, however,inconnectionwiththeseexperiments isthat they apparently prove that thesuperiority ofcopper isnotsimply doe toitaamaller electrostatic capacity and resistance, but that itismore susceptible torapid changes ofelectric currents than iron;forwhanthe resistance andcapecityofthecopperandironwireswereequalisedbytheSasertionofresistancecoilsandcondenser,theapeedontheformerwat26t ‘thereby diminished. Posibly themaguetio susceptibility oftheiron isthe ‘causofthis,Themagnetization oftheironactsanakindofdragnthe ‘currents,Itiwellknownthattelephonesalwaysworkbeteroncopperthaa oniron wires, doubtles forthesume reason,‘Thesoexperiments alsoshowthehighspeedofworkingthatisnow attainedbythePostOffcoauthorities withtheWheatstone automaticapparatus, Thefollowingtablegivesaninteresting rarunéofthedirest‘Ragesoftheprogressmade,anditsrateofgrowth":— 1877..80wordsperminute.|1881... 190wordsperminute, iss.)100", ” ssa! 200 yy ry iso: 2180), : sss: |250 |, > 1880:170}, °|Bet)2880 : Ww. T. March 6th, 1889. ON CERTAIN PHENOMENA, Ere. 129 ‘TheOneHundred andEighty-sixth Ordinary General Meeting of theInstitution washeld attheInstitution ofCivil Engineers, 25,Great George Street, Westminster, onThursday evening, February 21st, 1889—Professor W.E,AYRT0x, F.R.S., Vice- President, intheChair. ‘Theminutes oftheprevious meeting were read andapproved. The names ofnew candidates for admission into the Institu- tionwere announced andordered tobesuspended. The following transfer was announced ashaving been approved bytheCouncil :— FromtheclassofAssociates tothatofMembers— Frederick Jonathan Down. ‘The paper fortheevening, which, astheauthor wasinthe United States, wasread bytheSecretary, was— ON CERTAIN PHENOMENA CONNECTED WITH IMPERFECT EARTH IN TELEGRAPH CIRCUITS. ByA.E.KEnveity, Associate, Since theearliest days ofcommercial telegraphy—the days thatfirstreaped theIsbours ofSemmering andSteinheil, Gauss tndWeber, Cooke andWheatstone—the substance ofourplanet hasbeen impressed intosuch universal conducting duty that twsearcelyacircuitcloses,beit,acrossanocean,orbeitacross meadow, butembraces itsshare ofthe earth initscompletion; tndthetask thus imposed isgenerally borne soreadily and treated soperfectly that, until the existence ofthe eaves- dropping telephone, itwasreally doubted whether intruth the ‘ath could ordid conduct the current entrusted toitscare. Considering, then, thegreat obligations telegraphy owes to theearth—that theexistence ofsome hundred thousand miles of 190ONCERTAIN PHENOMENA CONNECTED WITH[Febtit metallic conductor, atpresent indaily telegraphic operation, impliesinacertainsensethecorresponding twinexistence ofan ‘equal length ofterrestrial conductor ready made byNature ;that ithaseven been possible totelegraph over aspace ofwhich the metallic conductor formed butavery proper fraction—the cases that aremet with where the earth fails tofulfil itspart excite astonishment more by:their rarity than bytheforce ofthose laws whose mandates they obey. Almost allsuch cases ofimperfect earth are,weknow, trace- able tosome fault either inthe earth connection itself orinits immediate neighbourhood ;andaremedy isfound byadequately increasing thesurface area oftheburied electrode, orthecon- ductivity oftheground initsvicinity;sothattheyarenotreally failures ofthe earth's conduction, but failures insecuring the necessary degree ofconnection with theearth’s mass. There are, however, other cases, brought about byconditions more deeply seated andmore rarely metwith, which will occupy our present attention. ‘Apaper byMr. James Graves, “OnVibrations due toEarth- «Plates,” wasread before thisInstitution inJanuary, 1875, andit, will beseen that thephenomena therein described areprobably related tothose that will bepresently detailed. Although the matter hasnotoccupied theInstitution's direct attention since that date, nodoubt some ofitsmembers will have independently observed similar phenomena, and can bring forward such additional evidence ascannot failtoelucidate asubject atonce replete with technical interest and important practical bearings. These phenomena arepresented bytheAtlantic cables ofthe Direct United States Cable Company, with whose permission the facts are laid before the Institution. ‘Thetwocables ofthiscompany terminating inNova Scotia run toIreland and toRye Beach respectively. Their landing place wasoriginally Torbay, N.S., butthey were removed thence toHalifax some eighteen months since. ‘Thediagram map supplies theoutlines andmain features of Halifax Harbour. ‘The Irish cable wasfirst brought intothe cable-house, situated 40feetabove high-water mark, upon the Wwe.) IMPERFECT RARTH INTELEGRAPH CIRCUITS. 1 beach opposite George Island, onthe23rd October, 1887, andit vsputincommunication with the town office bymeans ofa linecableand@returnearthcable,sidebyside,inanironpipe.Eachcablewas0-7knotlong,hadaresistance of8:3w,acapacityof025g,andahighinsulation. Thelinecableunited withthe Irishshore endthrough lightning guards, while theearth cable made good soldered connection with theshore endsheathing. Some vibration ofthe mirror spot, due toelectrical dis- turbance, wasnoticed onfirst joining upthecircuit attheoffice through instrument and condensers, but this wasatthetime tributed tofaults which then existed inthe cable. Beyond thiscontinual vibration, nonoteworthy feature occurred until six dayslater (29th October), when theRye Beach cable waslanded atthesamecable-house, andjoinedupwiththeofficebyapre- cisely similar pair ofcables running through iron piping inthe ‘ame trench with the other above mentioned. This cable is worked onthedouble-current Morse system without condensers, anditwasimmediately noticedthatsendingtoRyeBeachwith 12voltsofbatterypowervisiblydisturbed themirrorspotonthe Irishcircuit, although each cable made earth separately through itsownsheathingviatheinsulatedreturncablestothebeach. Receiving from RyeBeach didnotdisturb themirror, nordidthe ‘peration ofthe Irish cable initsturn affect the less delicate ‘apparatus ofthe Morse circuit. Theonly remedy that could be,orhassince been, practically amailable, without adding totheretardation ofthecircuits, was thereduction oftheMorse battery power ononehand, and thesensitivenera oftheIrish mirror ontheother, asfarasthe peration ofthecircuits would conveniently permit. After the removal ofthe faults inthe Irish cable the ‘ibrations ofthe mirror still continued, and itwas noticed that notonly these, but periods ofdisturbance coincident Withlocal meteorological changes, affected thesignals more than had been onrecord with the same circuit when worked Previously from Torbay. Onthis was superposed the inter- ferencecausedbytransmission ontheRyeBeachcircuit, and thecombined disturbance attimes made the mirror signals difficult toread. WON CERTAIN PHENOMENA OONNEOTED WITH (Fa. The Rye Beach shore end was atfirst laidvery near to the Irish cable, and actually over itatone ortwo points, ‘The cables were, however, subsequently separated, although they still necessarily lieatdifferent points inclose proximity. ‘This separation ofthecables wasfound tohave sensibly lessened their interference. The operation ofthecables from thecable-house instead of from theoffice wasfound toproduce nochange inthenature or degree ofdisturbance, while experiment failed toproduce any visible interference between the trenched cable circuits from office toBeach, even onthemost delicate instruments, Athird cable was then laid from the cable-house across the harbour and outthrough theeastem passage inthedirection shown onthe diagram. Ithad atotal length of5°8knots, sheathed uptowithin half amile oftheseaend. This un- sheathed half-mile ofcore terminated inacopper earth-plate one yard square, nearly onaline between Thrumeap Rock and Hartland Point, ‘This well-insulated line offered every prospect ofsecuring agood andindependent “earth”byitslargeterminal plate. When, however, itwasjoined upatthecable-house with theIrish cirouit, intheplace oftheIrish shore endsheathing, theinterference from the Rye Beach transmission remained as marked asever. Todetermine the nature ofthis obstinate interference, a series ofexperiments wasmade atthecable-house, asfollows. Allexisting connections were forthetime dismantled, and only the two cables, the two earth-wires soldered totheir respective sheathings, and thecore ofthethird cable, orsix-mile earth, were brought uptothetesting table. The resistances between these three earth-wire terminals was measured asfollows bybridge :— BetweenlishsheathingandRyeBeachsheathingO16.)gy » ” » »six-mile earth ... 65°7@ 3Rye Beh, yy O57@fMM From which theresistance ofeach sheathing earth connection ‘would, ofcourse, be0-08 w,and ofthesix-mile earth 65:6 w. ‘The only apparatus then retained forinvestigation was a 1690] IMPERFECT EARTH INTELEGRAPH CIRCUITS, —188 Daniellbatteryoffourteen gravitycells—used with,andsometimes alsowithout, apole-changer—for exciting disturbance, and a sensitive speaking mirror—used with, andalsowithout, condensers 0f404capacity—to renderanyinterference visible. Ezperiment 1,SeeFig. 1. Sending ‘reversals from seven cells into theRye Beach cable produced vibrations ofthe spot onthemirror instrument in creuit with theIrish cable, itssheathing, and40$, Experiment 2.SeeFig. 2. Similarly, sending reversals into the Irish cable, using its sheathing, produced vibrations onthe mirror incircuit with theRyeBeach cable condenser andsheathing. Experiment 3.SeeFig. 3. Sending reversals into the Rye Beach cable produced vibrations onthe mirror incircuit with the Irish cable condenser and six-mile earth. Experiment 4.SeeFig. 4. Sending reversals into the Rye Beach cable, using the sirmile earth, produced vibrations onthemirror incireuit with theIrish cable condenser anditsownsheathing. Experiment 5,SeeFig. 5. Sending reversals into theIrish cable, using thesix-mile cath, produced vibrations onthe mirror incircuit with the RyeBeach cable condenser and itsown sheathing. Experiment 8.SeeFig. 6. Sending reversals into the Irish cable, using itsown sheathing, produced vibrations onthe mirror incircuit with theRye Beach cable condenser and six-mile earth. Experiment 7.SeeFig. 7. ‘Sending reversals onthecircuit formed bythesix-mile earth, battery, andRye Beach sheathing, thedisturbance onthemirror incireuitwiththeIrishcablecondenser andsheathing wasmore‘decided than inanyofthepreceding experiments, Experiment 8.SeeFig. 8. Similarly, sending reversals onthe circuit formed bythe 18 ON CERTAIN PHENOMENA CONNECTED WITH (Feb. Sie, six-mile earth, battery, andIrish cable sheathing, thedisturbance onthe mirrorincircuitwiththeRyeBeachcablecondenserand sheathing wasequally great, Experiment 9. ‘The two last experiments were repeated with the condenser removed and the mirror indirect circuit between cable and sheathing. The continual vibration ofthespot made observa- tion difficult; butduring periods ofcomparative quiescence it was soon determined that thedisturbance produced bydeprese- ingoneofthekeys was notofamomentary nature only, but consisted ofasmall deffection, permanent during the whole period ofkeyapplication, and ofdirection depending upon the keyselected. Itwasalso ascertained that this deflection wasnot duetoanylocal electro-magnetic action by Experiment 10. SeeFig. 9. Onsending reversals between thetwo cable sheathings in cireuit with thebattery, nodisturbance could bedetected onthe mirror incircuit with one ofthe cables and the six-mile earth. This small steady deflection produced byholding down the keyineither ofthetwo cases tried inExperiment 9was more plainly visible when themirror wasincircuit with theRye Beach ‘able, which, besides itsshorter length, had nocondenser in cirouit atthedistant station. Toobtain some quantitative infor- mation regarding thestrength ofcurrent represented, thefourteen cells were connected inseries and more careful observations made. Experiment 11, SeeFig. 10. Closing thecirouit formed bythefourteen cells, thesix-mile earth, and Irish sheathing, measurements were made ofthe steady deflection voproduced onthemirror joined upwith the Rye Beach cable andsheathing. ‘The deflection obtained was subsequently found toapproxi- mately represent thetwo-hundred-thousandth partofanampere —i.e,, 5micro-amperes—and itsdirection wasobserved tobethat indicating acurrent entering theRye Beach cable from that pole ofthe battery which was connected tothe Irish sheathing, 1a DOPERYECT RARTH IN TELEGRAPH CIRCUITS. 185 reversing when the battery was reversed. Thus, inthefigure thecopper pole ofthebattery being shown connected tothe sheathing, the deflection represented acurrent from thecopes poleentering theRye Beach cable through themirror. Experiment 12. SeeFig. 11. ‘TheIrish sheathing wasreplaced inthebattery circuit by thewater-pipe earth connection atthe office through the trenched cable. Ondepressing thekey themirror showed the ‘ame permanent degree ofdisturbance asbefore, ineach cable cireuit, Experiment 13. Thisexperiment wasarepetition ofNo.3,with thecondenser removed. Itwas subsequently observed inthis case that the disturbance was not permanent during thedepression ofthe lattery key,butconsisted ofasmall vibration ofthespot atthe opening andtheclosing ofthebattery circuit. Experiment 14.SeeFig.12. This was carried outatthe office, situated inthe principal Street ofthetown, and nearly opposite totheWestern Union Telegraph Office, fifty yards distant. AlltheWestern Union ‘ries areoverhead, andtheearth connections made forthem are thegasand water mains, which, with thetelephone lines, form theonlymetallic connection between thetwobuildings. With thecourteous assistance ofthe Western Union officials itwas smmged that atagiven word bytelephone several oftheir creuits should suddenly beoperated, andagain, allstopped. When this wasdone theeffect wasdistinctly visible upon a ‘imorincireuitwitheachcable,thedisturbance beinggreater theRye Beach line, the earth connections being thecable sheathings through theinsulated cables inthetrench. Only fourknown causes could under anyconditions produce inter- ference between thetwocable circuits, these being, ofcourse, tlectrostatic induction, electro-magnetic induction, leakage, or ‘adearth connection incommon;and,intheabsenceofevidence, ayoneormore might produce it,although noreason could be demandedfortheoperation ofany,sincecableslieinotherparts. 1% ON CERTAIN PHENOMENA CONNECTED WITH [Feh.tin, oftheworld forgreater distances atequal proximity without the least visible interference. Experiment No. 10,together with thepreliminary trials with thetrenched cables, showed that thecauses were nottobetraced locallytotheseortotheapparatus; andthehighinsulationof thecables andcireuits, together with thefactthat the tests were carried outatthecable-house, precluded allpossibility ofleakage being concerned, sothat theoperating causes were thus limited tothe two forms ofinduction and bad earth. ‘The first sixexperiments didnotdirectly point infavour of anyparticular cause, butNos. 7and8rendered thepossibilityof electro-magnetic disturbance untenable, since theexciting circuit, except inthecable sheathing itself, took acourse,asseenon themap, almost atright angles tothetwomain cables. ‘They further showed that the interference was such asadeficient earth connection would produce. Experiment No. 9confirmed this view, since noform ofinduction could setupthesteady defles- tions observed ;andleakagebeingoutofthequestion, badearth was the only cause that could produce them. Itwas further necessary tosuppose that notonly onebut both cable sheathings failed tosecure good earth, and that the resistance intheir respective earth connections wascommon toboth. ‘Experiment 11corroborated thisposition;andNo.12didmore, foritextended thelimits ofthis badearth region over thewhole water-pipe areaofHalifax,asitwouldhavebeenimpossible to ‘obtain interference under those conditions unless the cable sheathings and thetown water-pipes formed acommon conduct- ingsystem throughout whose limits imperfect earth connectionexisted. Finally,theevidence ofExperiment 14followedthesame direction, theexciting circuit being inthiscase theWestern ‘Union land lines. ‘Theimputation ofbad earth tothegas- andwater-pipe area ofatown, and that town aseaport,issufficientlyremarkable; butthepresumption ofageneralearthimperfectioninasystem ofextended water and gasmains, together with thesheathings ofthreecablesrunningdirectlyintoanarmofthesea,mayat, firstsight appear anoutrage upon credulity. Dueconsideration Iso) IMPERFECT EARTH INTELEGRAPH CIRCUITS, 137 oftheexperimental evidence renders theconclusion, however, inevitable;andthisistantamount totheassertionthatthewhole ofthissystem—that istosay,allthelandcoveredbythetown, aswell asthat forming theharbour basin—is composed ofsub- stance more orlessinsulated orinsulating. Oneofthemost interesting features inthegeology ofNova Scotia isthat theAtlantic seaboard presents anuninterrupted line ofLower Cambrian formation. The relation ofthis fact toourimmediate subject ispatent when weremember that these rocks—early inorder ofgeological structure, and said tobealmost destitute oforganic remains—are here almost catirely composed ofargillites, quartzites, and micaceous schists slates and grits—substances experimentally known tobe ton-conduetors; and according tothe published papers of Professor Honeyman, who has studied the geology ofNova Scotia for many years, noother formations mingle with the Lover Cambrian inthestructure ofthe Halifax basin except thestill lower archwan granites, allofwhich arepresumably won-conduetors. Infact, the renowned natural advantages of Yalifax Harbour aredue totheunrelieved presence ofthese abrupt slaty cliffs that immediately underlie thesoilformiles inland, The question then arises, Granting that on the above geological considerations therocky basin oftheharbour and foundation ofthe city form aninsulating crust, orone, at least, interposing adecided amount ofresistance between the superficial conducting system and theocean orearth's mass,— istheresistance thatcanbejointlyofferedbythethreecable ‘heathings and thewhole waterway between thecable-house andopen sea, ofsufficient magnitude toproduce the degree ofinterference observed? ‘Theanswer, which seems tobeaffirmative, strictly demands ‘computation ofthe resistance offered bythe entire mass of seawater inthe harbour under those conditions. This would ‘beagiganticproblem tosolvewithanypretencetoaccordanceithnature; butforthepurposes ofthecase such limitations naybeadmitted asshall greatly simplify thecalculation. You. xvi, 10 18§‘ONCERTAIN PHENOMENA CONNECTED WITH[Feb21s ‘The map shows that three channels open upon theinner harbour, one running inland through the narrows, and two passing seawards—the eastern passage, andthemain orsouthern entrance. The inland channel terminates inabay some miles beyond, and may beconsidered tobepractically insulated with therest ofthebasin. ‘The waterway inthe shallow andnarrow eastern passage will evidently offer acomparatively high resit- ance compared with that ofthemain entrance, which, therefore, demands principal attention. ‘The general mathematical theory oftheearth's resistanceis given inSchwendler’s “Testing Instructions,” following the reasoning ofSmaasen’s theory, published in1847. Itisshown that ifaperfectly conducting small spherical electrode ofradius ‘7were situated concentrically within alarger hollow perfectly ‘eonducting sphere ofradius R,andtheintervening space filed with ahomogeneous substance ofspecific resistance p,the resistance ofanyelementary spherical shell ofradius 2,measured fromthecommoncentre,wouldbe&&‘andhencethetotal resistance (w)between thetwospherical electrodes woiildbethe integral ofthis quantity between thelimits 2=Rand @=+; lou whence o=£(4-®) Iftheradius Roftheouter electrode beindefinitely increased, theresulting equation becomes o=7> which represents the resistance ofanunlimited medium of resistance ptoacurrent passing from asphere ofradius rtoits ‘ultimate confines. Similarly, asecond small spherical electrode ofradius7,wouldofferunderthesamecircumstances aresistance «=;"= Eee and ifthese twosmall spherical electrodes were placed sofar apartinthemediumthattheirdirectinfluence oneachother's lines ofcurrent-flow would bevery small, theresistance between them would be 1 o+m= P(E+5) vio) IMPERFECT EARTH TN TELEORAPHE cIROUITE, 189 aresult equivalent tothestatement that almost allthelines of carrent-flow issuing from thesurface ofeach electrode would ‘mite with least opposition atindefinitely great distances, and afterindefinite diffusion throughout themedium.Ifthesystemwerethensymmetrically dividedbyanunlimited insulating plane, theresistance (2)between thehemispheres on tesideoftheplane would bedouble theabove, or ‘11o=e([ +5) where ¢and ¢,arethecircumferences ofthetwohemispheres. ‘This would then betheresistance between twohemispherical electrodes imbedded inthebounding plane surface ofanindefi- aitely extending homogeneous medium; and iftheelectrodes vere not hemispheres, butvertical plates imbedding active surfaces (eand ¢,),the above formula would approximately apply, iteand ¢,bethe circumferences ofequivalent hemispheres exposing those areas respectively, and thetransformed equation weal be 1 1 9=FE(atv) —anequation independent ofthedistance between theplates. SimilarresultsareobtainedandpublishedbyMM.Mascart andJoubertbyseparatereasoning. ‘Theformulaappliestothe ‘arth's resistance inatelegraphic circuit, ontheassumption that,arplanethasapractical uniformconductivity, andformsprac- tially anindefinitely extended medium. Intheabsence ofexperimental evidence itwould bedifficult tosay how farthevariable conductivity oftheearth’s mass would affect theapplication ofthis formula. The conductivity ofthe waterial composing theearth's crust probably varies considerably indifferent localities, andisgenerally notthesame asthat of ‘theocean,forexample. Itmight evenbeassumed thatthe Tesitanee oftheprimary rocks issogreat astolimit thecon- docting area almost entirely tothesuperincumbent stratified formations. Inthatcase theearth's telegraphic resistance would bemore early comparable totheresistance between two small distant Mo OM GERTAD PHENOMENA CONNECTED WITH (Fat, electrodes imbedded inthesurface ofanindefinitely extended sheet ofuniform thickness andconductivity. Onthissupposition, letdbethedepth oftheearth’s conducting crust, assumed tobe uniform: then theresistance, ©(vide Appendix), between two earth-plates ofactive surfaces (#and«,)would be,approximately, e 1 lV ‘4D9258 G+zn)+fae (ZP) ‘This equation isthe same asthat ofSchwendler last stated, butwith theaddition ofasecond term involving D,thedistance between theplates; sothat onthis theory theearth’s resistance would increase with thelength ofthecircuit. Ifsubmarine cables were sufficiently well insulated, itwould bepossible toascertain which ofthese two theories isnearer tothetruth, bymeasuring theconductor resistances ofcables which arelaidinduplicate or triplicate between two stations; the difference between the resistance ofany pair looped, and the sum oftheir separate resistances employing theearth circuit, giving, ofcourse, double the earth’s resistance between the terminals. Itis,however, impossible todecide from present experimental evidence inthis way, since notonly does thedifficulty ofaccurately measuring these quantities probably increase with thesquare ofthelength ofcable, owing toleakage andother causes, butalso because this second term ofthelast equation whose existence isinquestion munat inalloases bevery small, Generally speaking, therefore, theexperimental evidence may besaid toestablish theresults given bySchwendler’s formula, namely, that when theearth-plates areseparated byany distance great compared with their dimensions, the resistance varies with their active area, and isindependent oftheir distance. Schwendler gives anestimated superior limit ofthespecifie resistance ofthe earth inIndia as3,300, orabout 100 times that ofsaturated zinc-sulphate solution at10°C. Returning totheresistance offered bythewaterway inthe main entrance ofHalifax Harbour, wemay suppose acurrent to beflowing seawards through itandthecable sheathings from the cable-house, The lines offlowwill permeate itswhole area, and ‘theequipotential surfaces may besupposed toform acrossitin wel] IMPERFECT EARTH INTELBORAPH CIRCUITS, 141 successive curves. ‘These surfaces will bepractically vertical, and2postionhastobeassumedfortheultimateonerepresentingthezero-potential, orthesurface that may beconsidered tomarkthelimitofperfectearthconnection, Theoretically, ofcourse,thislimitwouldbefaroutintheocean;butitmaybearbitrarily assumed that avertical plane shown onthemap byadotted line connecting theThrumeap with theMars Rock buoy represents thisboundary, and that theresistance toearth oftheocean mass ‘ntside this plane isaltogether negligible. ‘The twoother dotted lines converging from this base uptothe cable-house—one kiting Sandwich Point, and theother ranning over Maugher Baach—are intended torepresent what may beconsidered the simple boundaries ofanequivalent waterway equal incon- activity tothe actual one. The average depth may betaken 414 fathoms, Wehave, onthis supposition, amass ofsea-water rmaembling awedge with ablunt point, thebreadth ofthebase being 1-2Knots; thebreadth attheapex, say,5metres;the length, measured perpendicularly tothebase, 4knots;andthe depth, 14fathoms. Strictly speaking, theequipotential surfacesfflowwillbecurvesconvextotheocean;buttheerror introduced will not begreat ifweconsider them asplanes pnlleltothebase,orzerosurface.Sincethefallofpotential willtake place with equal rapidity through the water andthoughthesheathings, theresistanceofthelattermaybetakenintoaccount later andindependently. ‘The resistance ofany = Genentary lamina ofthickness dz,distant ©from theeable- bene,wllbe5Ease.gytherepistheapecficresistanceofthe a ON CERTAIN PHENOMENA CONNECTED WITH [Feb. 2is, water,Athedepth,whileaandbareconstants;sothatthetoll resistance between the two ends is B R=Fflogegy whereBandAarethebreadthsatthebaseandapex. Inthis case i” a=1=03; hh=14x6x305=2,560; Baese 42=1,505 p=3l. ‘The specific resistance ofasaline solution ofdensity 1-027 ha: been observed experimentally tobeabout 31wat5°C. With these data, R=31x272 x319 O03 x2,560 =035 0, about one-third ofanohm, equivalent toaconduc- tivity of. ake nee nae ee 249mos. Compared with this, theconductivity oftheeastern passage channel may beestimated atabout one-fourth, or...aeweooos0-7mho. ‘The conductivity ofthetwocable sheathings, each consisting oftenNo. 6wires, and length 4-2 knots,wouldbe20x0-020... a=0-4mho. And, finally, the conductivity of7knots ofeable sheathing inthe eastern passage, 10x0-012 =0-1mbo. Total conductivity ... ... 4+mhos, equivalent toajoint resistance of0:24 w.Sothat iftheabove reckoning issufficiently fair, theresistance between thecable- houseandtheoceamisaboutaquarterofanohm,The resistance actually offered canbeestimated from thedata ‘ofExperiment 11,asfollows :— The E.M.F. inthe exciting circuit was about 15volts; the resistance, 66winthecable and earths, and 24inthebattery— Wo] IMPERFEOT EARTH IN TELEORAPH CIROUITS. 143 total, 90,IntheRyeBeach circuit theresistances atthetime ‘were, approximately— Observing mirror... wu. vee 2,000 Cable conductor... eave 5,800 Relay atRye Beach wee ee 1,000 Total eee 8,800 Since the interference current observed inthis circuit was, as already mentioned, 5x10-* amperes, theE.M.F. impressed upon thecable due tointerference was 5x10x88x10°=0-044volts; andconsequently, ifxbetherequired resistance ofthegeneral earth connection, @2902:0-044:15, o 2=026 0, oranactual resistance ofaquarter ofanohm. Theagreement between the actual and measured values isreally much closer thantheaccuracy ofeither determination would warrant. ‘The only experimental results still left unaccounted forare those ofNos. 3,4,5,and 6,inallofwhich thesix-mile earth was ‘edeitherintheexciting orobserving circuit;andsinceExperi ‘ment 13proved that this earth-plate wasoutside thelimits ofthe ‘emi-insulated basin, asmight beexpected from itsposition on ,,themap, itwould beimpossible toaccount fortheinterference in thoseeases onthescore ofacommon imperfect earth. Reflection shows, however, that theinterference observed would beanindirect aecessary consequence oftheimperfect earth, since thesix-mile earth, when connected toeither main circuit, forms acable-loop of, ‘7,10 knots within thelimits ofthesemi-insulated basin, anda alight variation inthepotential ofthebasin iscalculated to Prduce@considerable electrostatic effectonthecable.Ithas, ifact, been found experimentally that when 1,000 feet of |itualatedcablecore(ofsimilarcapacity tothatofthecablesat |Halifax)arecoiledinaninsulatedtankandleftincommunication |withtheearth through amirror instrument, theelectrostatic disturbance ofthemirror isabout equal tothat observed in |theseexperiments (3-6) when thepotential ofthetank isaltered 4s ON CERTAIN PHENOMENA CONNECTED WITH (Peb.2i8, by0-2volts. The total charge passing through themirrorin Experiments3and 6would bethe resultant effect ofthe whole system ofharbour potential variation onthefour knots ofcable ‘under observation, except. insofarasthecapacity ofthecable seawards would modify theinfluence. InExperiments 4and5 theinterference would, ontheother hand, beduetoelectrostatic charge entering theobserving circuit owing tothevariationin theharbour potential, electrostatically produced bytheloopof exciting cable traversing it. Anexternal average variation of 0-01 voltinthepotential ofthebasin would suffice toaccount for thedegree ofinterference observed;andthisamountofvariation, itisevident, the conditions amply supply. Experiment 13, contrasted with Nos. 9and 11, showed that the interference observed incases 3,4,5,and 6wasofanelectrostatic nature. The sensitiveness ofacable toexternal electrostatic influence ‘can beshown experimentally byjoining upamirror galvan- ometer incircuit with 100 miles of cable and earth, Ifthe distant end besealed, and laid for two orthree feet inan insulated joint trough filled with water, thepotential ofwhich iscapable ofbeing varied suddenly through 100 volts (from +50to—50),the effect isimmediately visible onthegalvan- ‘meter; andinfact, simple signals have been transmitted inthis ‘wayfromshiptoshorethroughthecoatingsofajoint. ‘The hypothesis ofaninsulated orsemi-insulated basin, therefore, satisfactorily explains allthe phenomena observed. Ttaccounts, first, forthe meteorological disturbances which areobserved onthecables, since any alteration ofthechargeor Position ofelectrified clouds inthe vicinity ofthe harbour may affect thedistribution ofcharge inthecable toanextent that would notbepossible iftheharbour were inmore perfect elec- trical communication with the ocean orearth's mass; and allthe phenomena experimentally observed canbetraced eithertothe direct orindirect effects ofthis imperfect earth connection— directly byconduction, indirectly byinduction. Assuming, then, that theLower Cambrian rocks effectively insulate theHalifax basin, itisnecessary tosuppose that the same effect would beproduced atany landing place onthe Ween] IMPERFECT EARTH IN TELEGRAPH CIRCUITS, 45 Atlantic shore ofNova Scotia, where these rocks sogenerally prevail. This isalso corroborated bythefact that previously at Torbay some difficulty ofthesame kind appeared when attemptsvereoriginally madetoutiliseanordinaryearthconnection for anoverhead Morse circuit. ‘The interference with the cables thusset:uprendered aspecial earth necessary fortheland linein thebay. ‘The fact that nointerference wasvisible between the cables themselves isaccounted forbytheopen landing place and ateence ofany such basin asHalifax Harbour presents. InMr, James Graves's paper onearth vibrations, already alluded to,mention ismade oftheslaty character oftherock in theneighbourhood ofthestation, and sofavours thesupposition that the interference there observed was due tothe same causes, Itissaid that intheLake Superior region ofthe United States great, difficultyhasoccasionally beenfoundinobtaining eficient earth, even when nolocal cause forthedeficiency has teenapparent. Iamindebted toProfessor Moses G.Farmer for thefactsofacasewhichcameunderhisownnotice, wherean arth connection, apparently good and well situated, wasfound ‘useless,andmeasurement showedthatitsresistance toearthwas12,000w.Ialsoowetohimtheinformation thatinterferenceduetoimperfect earth wasfound in1869 atSt.Pierre Miquelon, theMorsesignalsontheSt.PierrePlacentia circuitaffecting the |Atlantic cable toBrest, andthat special measures hadtobetakentoeffectabetterearthconnection.Generalising fromthesefacts,itwouldseemthattheremaybedifficulty inobtaining good earth atcable stations opened on ‘shore ofprimary ormetamorphic rock, and that permanent interference beyond remedy byearth connection ispossible between cables landed inabasin ofsuch formation;sothatifthis “atement receive confirmation atthe hands ofthe members of thisInstitution, itwillbeevident thatadueregard willbegiven ‘ogeclogical conditions byelectrical engineers among themany incidental considerations that determine their selection ofa table's landing place. us ON CERTAIN PHENOMENA, Ere. mone, APPENDIX. Ontheresistance between twoelectrodes ofactive surface (¢and.) imbedded inthe surface ofanindefinitely extended homo- geneous medium ofspecific resistance pcomprised between two unlimited parallel planes separated byaperpendicular distance d,thedistance between theelectrodes being repre- sented byD. Let rand 7,bethe radii ofthe equivalent hemispheres exposing surfaces ofsands,respectively. Then, ifaninfinitely thin perfectly conducting hemispherical shell ofradius dcon- centrically surrounds each ofthese electrodes, these shells will touch the opposite bounding plane ofthe medium, and the resistance between them and their included electrodes will be 11 Lod a Fe(ea) 009g G-g)mpentively. IfDbelarge compared with d,these twoshells may beapproxi- matelyconsidered astwoparallelcylindersofradius$traversing the medium atright angles tothe bounding planes, and their surfacepotentials maybewritten*— a = op >); Vi=constant —°F|loge(?}:D ©D V,=constant—,faloge(:} where V,andV,arethepotentials and¢thestrength ofcurrent.. _ie 2D_log.dMio Va=zea(lee2?-Sy) +,Resi =, AD). -'-Resistance =<P.loge(47)5 80that the total resistance, =f (14+1-2)4 2 4D),oat a)ttale (Ps andsince dislarge, =P (gL) 4#tog,(42 aA:+72)*faloge(“> approximately. vce Mavcart andJoubert Electricity andMagaetiamy” wo.1,p.306 1988) DISCUSSION. ur The Cuarrman:NodoubttherearesomesubmarinecableProfesor engineers present, andIthinkthatperhaps thebestthingtodo” vould beforthem tofavour uswith anaccount ofany similar experience they may have had tothat described inthe paper, of sigoals ononecable being interfered with bycurrents onanother able, due possibly toimperfect earth. Mr,W.P.Graxviuiz: There areoneortwoparagraphs inthisme. admirable paper which, Ithink, should becarefully considered. Mr.Kennelly states that “the Rye Beach shore endwasatfirst “laid very near tothe Irish cable, and actually over itatoneor “two points.” Inthat case you would certainly expect induction tetweenthetwoparallelwires.Hethengoesontosaythat “the cables were subsequently separated, although they still “necessarily Lieatdifferent pointe inclose prozimity;” and that “this separation ofthecables wasfound tohave sensiblf lessened “their interference.” Now, asthepartial removal caused acon- siderable cessation ofthe disturbance, isitnot. probable thata futher removal would practically eliminate theinterference? Certain experiments arealeo described ashaving been made withthecondensers removed, and itwasthen found that afeeble ‘outinuous current wassetupinone cable circuit whenever the her was charged bythe strong battery employed; this feeble current being indicated byasmall permanent deflection ona wnsitive galvanometer. This also, according tomyexperience, ‘isnotanexceptional effect; infact, Idonotbelieve, iftwo “earths” areinproximity, that youcancharge theonewithout, affecting theother, however good the“earths” may be,especially ifnocondenser isinthereceiving circuit. Ittherefore appears thatthe phenomena here noticed may befully accounted for Without introducing thetheory ofaninsulatedbasin. Itoccurstomethatthepracticaldifficultyexperienced might teovercome byanadditionalcoiltothespeakinggalvanometer oftheIrish cable, preferably made eothat itsposition relative to theordinary coils could bereadily altered, and thereby the |rength ofitseffect upon the mirror magnet varied atwill. Ifthis extra coilwere coupled upinthesending circuit ofthe |‘RyeBeach cable,orasa“shunt” thereto, itisclearthatthe Ms ONCERTAIN PHENOMENA, Ere. (Fesun, Hay, Mirror would beaffected inonedirection bythedisturbance referred tobyMr. Kennelly, and intheother bythecurrent circulating intheextra coil; andthese twoeffects could,Ishould imagine, bycareful adjustment, bemade tocancel each other, and abalance bethus obtained. Koon, _Mr.H.C.Donovan :Idesiretocompliment Mr.Kennelly onhisable investigations into aphenomenon which hasbeen asource oftrouble tomore than one station, tomyknowledge. Thelate Professor Varley foresaw that geologic conditions should betaken into account intheworking oflong submarine cables. Consequently hetook the precaution tohave aspecial earth- wire laid from the cable-house tosome distance out tosea. Conditions somewhat similar to Halifax existed atValentis, The island isaslate rock. Iamunder the impression that thisprecautionary “earth” wasnotused, astheironsheathing of the cables themselves afforded all that was required when working from thecable-house. When, however, thecables bad subsequently tobeworked from theoffice atKnightown,some five miles distance from the cable-house, the iron wires ofthe subterranean cables, which connected the shore ends with the office, were used asthe“earth ;”then—owing, Isuppose, tothe non-conductivity oftheisland—there were thesame conditions, inalesser degree, which, judging from Mr. Kennelly’s paper, exist inHalifax Harbour, due, Iimagine, tothe impediment tothe complete diffusion ofthe return currents, which were conducted both bythe iron wires and the earth itself, A complete “earth” should bemass with the lowest possible resistance. The iron wires ofacable under conditions which exist inHalifax Harbour donotfulfil theconditions ofaperfect “earth,” hence thedisturbances noticed, and soably obviated byMr.Kennelly. we.timan, Mr.C.E,Prtmax, C.LE.: Theonly case ofaninsulated earth Iknow of,occurred during oneofourfrontier expeditions. We ‘were about toestablish atemporary office attheclose oftheday, and selected avery damp, swampy place fortheearth-plate, but toourhorrorwegotnosignalswhatever; ourlinewouldnotwork intheleast. Onexamining thenature oftheground inthe 9) DISCUSSION. wo vicinity ofthecamp, wefound that afewhundred yards away upa,Pitman. theriver thebank wasformed ofashelving rock, which, passinganderthecamp,eroppedoutagainsomedistanceoffintheoppo-sitedirection;butwesoongotoverthedifficultybyhitching» ‘ireontothetrees andrunning itback about halfamile.‘That isthe only case Ihave metofathoroughly insulated basin. Mr.W.H.Paevoe: Intheearly days oftelegraphy “badyr,rece “earths” were constantly cropping up,and oneinfallible mode of determining whether thedisturbances were duetobadearth orto ckher causes was very easy when thedouble-needle telegraph instrument was used. The double-needle instrument requiredtwowires,aspersketch,withvirtuallyagalvanometer oneach, == <—_<— ws =< rar = raking earth. Itwasrareinthose days tohave aninstrument always free from disturbance, contact, weather contact, orwhat- everitmight becalled: theslightest shower ofrain inanypart ofthecircuitwouldinvariably causedisturbance fromoneneedle totheother; theearly letters ofthealphabet—a, b,c,d,and so ca—mhich were formed with oneneedle, were repeated onthe other, Earth currents were also aconstant source oftrouble, as ‘ellascontacts from various causes, and theway inwhich we vee able toatonce saywhat was thenature ofthefault was this:Suppose acurrent issent upon onewire ofthedouble needle, inonedirection, itwould divide itself, and the result. would lwthat thetwo needles would bedeflected inthe same direction. Withsucharesult oneknows foradead certainty thatthefault istetocontactbetweenthetwowires.Supposethedisturbance‘anedthetwoneedlestodeflectinoppositedirections, withan‘yal dead certainty onewould know that thefault wasabed “earth ;”sothatbythedirectionofthetwodeflectionsthenature efthefault. cambeeasily distinguished. The same practice is flowedatthepresentday:ifanyonedoubtsthecharacter ofisearth,” hesimply hastoputtwogalvanometers incireuit in 190 ONCERTAIN PHENOMENA, Ere. [Fennit, Mr.Prece. thesame way asthey were used with double-needle instruments, andhewillquickly seewhether the“earth” isgood orbad. InMr.Kennelly’s very exhaustive and admirably worked out paper there arealotoffacts andplenty offood forthought, bat there isonepoint which Ihave not detected inthe paper: he does notappear tohave made aparticular test toascertain for certain that thefault was abad “earth;” hespeaks merely throughout thepaper ofvibrations, Histests nearly throughout thewhole paper aremade with reversals; and ifhehadused reversals only, then hewould have leftusinvery great doubts; butintheexperiments about Nos. 7,8,9,hehasused aconstant current, andthough hehasgiven facts that leadustothink the causetobesomething ofthekindthathepointsto,hehasnot given ussufficient data tomake one feelquite sure thatheis right inhispremises. ‘Thisquestionofbad“earth”wasverythoroughlyinvestigated and written about byProfessor Fleeming Jenkin, and Iama little surprised tofind that Mr. Kennelly was notacquainted with Professor Fleeming Jenkin’s work inthis matter. He ‘mentions the fact that he received information from Professor ‘MosesFarmerthattheislandofSt.Pierrehadabad“earth;” but Professor Fleeming Jenkin read acapital paper, entitled “An “Insulated Island,” apaper which isreprinted inthe“Life” written byStephenson (the second volume ofwhich contains» synopsis ofhisworks, collected and edited byProfessor Ewing), and inwhich theinvestigations ofMr. Gott, the superintendent ofthe French Atlantic Company atSt. Pierre, are given. Professor Fleming Jenkin, inhisremarks, leaves nodoubt whatever that the “earth” was bad, and that St. Pierre isan insulated island. : ‘Mr.Kennelly hasalsoreferred toMr.James Graves's paper— ‘Mr.Graves isstill thesuperintendent oftheAnglo-American Telegraph Company atValentia—in which observations were described ofavery curious phenomenon which wasaltogether aside from thesubject ofMr. Kennelly’s paper. Mr.J.Graves's observations were notprimarily onabad“earth,” but oncurious vibrations that were observed attheearth-plates, and heshowed 1988) DISCUSSION. as thattheywereduetosomedefectintheearth-plate; andat,Pree. erentually—in the discussion, Ithink—it waspointed outthat probobly theresults that hehad observed were due toelectro- Polarisation ofthe earth-plates: that the earth-plates were too small; that thecurrent decomposed thewater, gases were formed onthesurface, thegases asthey moved away caused avariation intheresistance orpotential oftheplate, and soproduced vibra- tion, There isnodoubt that the explanation wasquite true, because thevibrations entirely ceased when theearth-plate was calarged byusing theoutside sheath ofthecable. Iremember, too,agreat many years agonow, asimilar difficulty being met vith atTorquay;itwasdescribedbyMr.Culleyinthisroom, ‘When thetelegraph wasfirst carried toTorquay, there wasvery eat difficulty indeed ingettingan“earth,”and“earth”wasonly atlastobtained bycarrying abarewiretosea.Inmyownexperience Icanoulycalltomindagreat ‘ifcalty being met with inthechalk districts ofHampshire, when therailway from Basingstoke toSalisbury wasconstructed. Allthrough that district there wasvery great difficulty indeed inmaking earth—so great that between twoofthestations a retum wire had tobeused. Whether thedifficulty continues do not know. Tshould like tocallattention to point notalluded toby ‘Mr.Kennelly—that is,theuseofthetelephone indetecting the tatare and character ofthese vibrations. ‘The telephone isan atremely delicate instrument ofresearch;withconstantpractice itsindications canbemastered with very much greater confidence thantheindications ofagalvanometer:youcantellwithalmost absolute certainty whether theindications aredue toelectro- magnetic induction ortoelectrostatic induction. Itiswonder- |(towthetelephone seemstoepeaktoyouinscompleteImguage, which isonly acquired byconstant practice. You ‘ayremember that Ibrought before the British Association thesubject, and went into thewhole question ofthedisturb- ‘amees that are occasioned between parallel wires—between Fires that run incontiguity toeach other—overland, under goad,andatsea,Imadeagreatnumberofexperiments in 132 ONCERTAIN PHENOMENA, Ere. Fans, uePrewe. this direction, Some very accurate andvery reliable experiments were made formebyMr.Gavey intheneighbourhood ofCardif, onthe sands ofPorthcawl, onthe South Wales coast. There, asyou know, the tide rises and falls enormously:theriseand fallinspring tides issomething like40feet. Iwasvery anxious tofind out whether the water acted asascreen—whether it prevented theinfluence ofonewire from affecting another wire initsmeighbourhood—and the result ofthose experimentswas toshow that the water offered no obstruction whatever tothe disturbance between one wire and another. Iftwo wires were carriedwithin,say,20feetofeachother,inairorwater,inthe earth orunder the earth, the effect was precisely the same. Experiments were also carried outbymeintheGeneral Post, Office. Wires were arranged around thecorridors ofthedifferent floors;thecireuitwassomethinglike400feetlong.‘Thereare four floors, and thecurrents sent onthetopfloor were clearly indicated inthe cireuits onevery other floor; infact, between the top floor, where the experiment was made, and the lower floor—a distance of60or70feet—we found that, inanelectrical sense, “stone walls donotaprison make.” ‘There arenoknown present means ofpreventing thepro- duction ofthese electro-magnetic aud electrostatic disturbances through space. They areevidently, aswenow know from the investigations ofHertz, conveyed through theether;andwhether thespacebeoccupiedbystonewalls,whetheritbetheair,whether itbewater, orwhether itbethecrust oftheearth, theinterposi- tion ofthese elements does not affect the question atall. At Portheurnow, ontheCornish coast, where theEastern Telegraph Company have their landing place forcables toLishon,—wherethe cable tothe Scilly Islands, the cable toBrest, and several other cables land orstart from,—they areallwithin ashort distance of each other, There Mr. Ash, the superintendent ofthe Eastern Company, found that currents sent ononewire were clearly and distinctly traceableonallthecontiguouswires.Hesenttele- phone currents ontheLisbon wire, and indications ofthem were distinetly heard ontheBrest wire; sothat there was absolute evidence inthecharacter ofthesignals that theelectro-magnetic disturbance passed through thebody ofthewater. 18883 DISCUSSION. 188 1donotfeelthat Mr.Kennelly’s paper isperfectly exhaustive, sr.Pree orthat hehas sufficiently investigated ‘the electro-magnetic distarbance between one wire and another;forIthinkthatifhe ‘adused telephones hemight have added considerably tothe value ofhispaper, and theresult might have been negatived, thoogh Iwill not saythat itwould have been so. Hehas, however,doneallthatonemancandotoprovehispoint;but itis avery curious thing that when you have apoint to pore,whenyoufindanexperiment toconfirmyourviewofthat point, youoften, bysheer neglect, omit some very trifling thing |thatwilentirelyupsetyournotions: thathappensoverandover agin. Iought tomention that there isnremarkable case of thepropulsion,asitwere,throughwaterofanelectricdisturbance —Ivvill notcallitelectro-magnetic,Ibelieveitiselectrostatic disturbance—that is,the existence ofthunderstorms inmid- Mlantic, Itis@common and frequent thing attheend of Atlantic cables toseesudden, sharp, decisive currents knock the ainor about ormove the style ofthe recorder, and these things ueknown ns“kicks.” Itiswell known that they always occur henlightning ispresent. Itvery frequently happens that there ‘maybealightning storm athousand miles away from land, and setthedisturbances produced bythelightning discharges pass through the2,000 fathoms ofwater andthere induce inthecable 4thebottom oftheseathedisturbances that. produce thekicks atthelanding place ateach end. They arevery interesting fuety,and Ithink that, bearing thefact inmind that electro- magnetic and electrostatic disturbances can proceed through ater aseasily asthey canthrough air,theeffects that Mr. Kemelly hasobserved would beclearer ifhebad looked at tiemfrom this point ofview more than from theresistance iatofview.Mr.A.J.S.Apams: Themass offormule andfigures given Mr.Adama. iaMr.Kennelly’s paper appear, tomymind, tobebased upon Mactically nothing, andtheconsideration ofthem tohave been twute oftime. Idefy anyone tosay,from theexperiments Rentined inthe paper, what the disturbances really were, ‘tvhatitisintendedthatwearetoconsiderthem.‘Thereare You, xv, u 158ONCERTAINPHENOMENA,Ere.(Fen, uz,Adana.severalpossible causesofinterruption —induction, leakage, derived currents from “badearth,” and vibrations. Ifwehave disturbance duetoleakage, thentherewillbeamoreorles permanent effect solong as.the keyisdepressed;whilst,ifthe disturbance bedue toinduction, the effects will bemomentary. Surely itcould have been possible tothrow alittle more light, upon thepoint. Ithappens that the vibration disturbances referred to,and which formed thesubject ofaninteresting paper byMr.James Graves in1875, arenostrangers tome,although, unfortunately, myown opinion astoprobable causation differs from that of Mr. Preece, inasmuch asIam ofopinion that these peculiar vibrations arenotdue toelectrolytic nor toanyelectro-chemic action attheearth-plate.* Sofarastime may permit, Ishould like torelate some ofthe experiments made bymyself inconnection with thesubject, and which tend toshow that the vibrationsarenotnecessarilydueto so-called badearth attheplate, nortomere electrolytic action. Ateach extremity ofmygarden—north andsouth—an “earth” was putin,asalso right and left ofmyhouse todistances of 250and 500yards respectively. Each “earth” was ledinto my house bymeans ofNo. 16copper wire, without joint, thoroughly insulated, and brought directly ontotheebonite base ofthe reflecting galvanometer. Ihad thus ameans forsecuring six different pairs of“earths.” Each “earth” wasformed bybunching together andsoldering oneendoftwenty 2-foot lengths ofclean No.16copper wire, the line wire being included. The holes were about 5feetdeep, andineach case theearth- wires were spread outfrom thecommon soldered centre. Asthe earth wasfilled initwasthoroughly punned. Now inthecase ofeach pairof“earths”theseunmistakable disturbances were always visible, thejerky movement ofthe “spot” being altogether unlike the steady variation ofau ordinary earth current. ©Thedisturbance ofcurrent,notthecurrentitelfisherereferredto, 19993 DISCUSSION. 1s ‘Two other “earths” were also put in,the one 5feetmr.Atm immediately below the other. Here also, inaminor degree, vibrations presented themselves, although the flooding ofthe loeality with water hadlittle influence upon theresult, Incontinuation oftheseexperiments, abox18inchessquare wasparted off into two compartments, space being provided atthebottomofthepartition forproperconnection betweenthe two. Two copper plates were inserted, oneineach compartment, andthewhole filled inwith sandy earth, “Bymeans ofanindia- rubber tube attached tothe house tap, water was caused to dribble down the front ofone plate and drain offunder the ther. Ofcourse the usual electrolytic current was theresult; andsurely, ifvibrations aretheresult ofsuch action atthe earth-plates, they ought tohave appeared inthis instance, but therewasnosignofthem.Itwasevidentfromthisandfrom similar experiments that vibrations arenot due toelectro- chemic action attheearth-plate.* Itwas anoteworthy feature ofMr. Graves’s results that directly hemade earth connection bymeans ofthemain cable- sheathing thevibrations ceased: the“earth” had been cutont, amdpractically a“return” substituted, And so,Itake it, directly afoot even oftheearth’s surface isintroduced into any ireuit, s0surely also willthese vibrations present themselves. Inthe paper read thisevening itwould seem that thisvibra- tioneffect hasbeen somixed upwith some, orall,oftheother disturbances mentioned,astorenderaseriousconsideration ofthe figures and formule given awaste oftime, and itistobe regretted that the particular kind ofdisturbance witnessed in cachexperiment atHalifax wasnotstated clearly bftheauthor ofthisinteresting andimportant paper. Mr.D.C. Bare: Itmay beinteresting tostate that ononemr.mie cecasion, when Iwas connected with theshort-lived Central News *Varatims are notobserved iathe caso ofordinary electrolysis—the Teultant carrent ofwhich issteady initaeffects—unlow mechanically dis.{urbed,abyashakeorjar.‘Thereisthepomibility ofamechanical move-‘eatintheearth'scrustascauseforvibration;andalthoughseemingly otcalculated—ifexisting—to producethenecessarydisturbance ofearth Pita relationship, thepoint isworthy ofbeing followed upinthatdirection, 6 ON CERTAIN PHENOMENA, Ere. (resis, MzBat ‘Telegraph Company, weworked anoverhead and underground wire,parallelwiththePostOfficewires,fromFleetStreettothe Manchester Guardian office. ‘The latter office, ifIremember rightly, had anABC instrument, and atnight anews wireto Manchester. Our own wire was worked with about 200 volts through 3,000 ohms, and therefore had apretty considerable current on. Itused towork perfectly well attheopening, butas theevening wore onweused toupset thenews wire relays both atNewcastle Street and atManchester. Does notthis, therefore, pointalittletoMr.J.Gravesbeingcorrectinsayingthatsuch faults may bedue tothe polarisation ofthe earth-plate, oon- sidering that wewere using such alarge current? Another explanation, possibly, isthat wewere connected tothe house side ofthewater service (itistrue there were several moderately good “earths” on,because alarge number ofwater taps were supplied), and when thecistern was full ofwater weworked through almost without trouble; butifthecistern became empty, then theincreased resistance oftheearth, asmight beexpected, increased thedifficulties, although stillgood eriough forthesmall news current alone. The circumstance will probably beremem- bered byMr. Preece, asIthink wehad considerable trouble to find outwhere thedisturbance came from, and itwas finally traced totheCentral News asbeing theculprits. xe.sma. ‘Mr.Howanp Swan: Itmight beinteresting tomention, in connection with the remarks that have just been made, acase that was reported afewmonths ago from Bridgewater, Nova Scotia, inwhich atelephone line was run inthegold-mining district. Itwasfound that aconstant current wasflowing inthe wire—not very strong, butsufficient sometimestooverpowerthe exchange battery—and itwas difficult tofind out wherein the cause consisted. The electrician who investigated the case seemed to believe that there was what he considered avast natural battery, consisting ofthegold mine asonepole, andthe large deposits ofironorgold, orother metals, astheother pole. Ofcourse,ifthisweres0,itwouldbeabatteryonclosedcircuit, and most ofthecurrent would gothrough theearth, and part only ofthecurrent through thewire. From some reason orother 0 DISCUSSION, ur acurrent did gothrough thetelephone wire, andthere is,itismeseas said,tothis day asmall constant current flowing through the wirewithout anybatteryexcept naturalone.Thatpointis,I think, ofinterest inthis discussion. Mr.D.C.Bare:MightIaskM.Despointes, whoIseeisws.nu here, whether Iam correct inmymemory that when theRev. Mr.Highton’s gold-leaf receiver was placed onafaulty dead- earthed Dover-Calais cable, they gotsignals which were being sent over one ofthe North Sea cables? M.Desrorxtes: Icannotsay.Idonotremember thematter;Mapointes.and,iftried, theexperiment would beatDover perhaps, while my daties have been inLondon only. Mr.©.T,FLEErwoop: [remember making the “earth” forthe Me. London Central Station telegraph circuits, and Ihave always ,feltthat itwas anexcellent one, because ithas never been com- plainedofsincethetimethatitwasmadein1874. That“earth” jismade byaconnection with thepneumatic tube system, which rans forsome thirty-five miles under the London streets. I hoped that we should have heard more from Mr. Preece this evening, forIhave heard ofinduction between overhead and underground wires, and inthepaper referred tobyMr.Preece mention ismade ofinduction being heard inwires half amile apart. Surely, ifthis iss0,itwould bewell forMr. Kennelly to wea telephone andfurther investigate thecase that hehasin and. ‘Mr.W.H.Presce: Ishouldlikejusttomention someat.Preese caperiments that Imade inorder todetect thepresence, ifany, ofdisturbance betweentwocablesseparated fromeachotherbyadistance oftwenty-five miles. Onecable wastheoldcable con- uecting Dublin and Holyhead; theother wasanewcable Inid calytwoorthree years ago between Nevin, inNorth Wales, and Neweastle, near Wicklow. ‘Those twocables lieparallel toeach theratadistanceoftwenty-five milesapart;theyareeachabout tisty miles long. Ihad very powerful currents sent onthe Dublin-Holy head cable, andlistened forthesignals ontheother able, Intelephonic experiments Inever trust tomyown judg- ment, andIdonotthink anybody should ever relyonhisown 188 ONCERTAIN PHENOMENA, Ere, (Fenn, Mr.Prece. judgment with telephones—it should always beconfirmed by somebody else. Itisperfectly absurd how your imagination leads you adrift. with telephones, and therefore, unless obserrs- tions areconfirmed bythose who arewith me, Idonotmake use ofthem. Inthis particular instance Iambound tosaythatI knew what Iwasgoing tohear, andIam equally bound tosay that Iheard what Iexpected tohear, but nobody else did, although there were three accomplished engineers andelectricians assisting intheexperiment;buttheydidnotquiteknowwhat they were tolisten for. Myimagination may have putmewrong inthis case; butIhave notthe slightest doubt that the effect could beheard, andthat signals sent ononecable could beheard ‘onanother cable separated from itbywater atadistance of twenty-five miles, Itdepends onthemagnitude oftheprimary currents. Bee. ‘Mr.C.W.S.Crawney: Ibadsomeexperience ofearth-current disturbances when investigating with my friend Mr. Hawes the currents due topassing trains. Inconjunction with these wehad frequent occasion toobserve the currents between two zine earth-plates some 200 yards apart, aswell asthose between these plates and the rails, These currents were ina constant state ofvariation, Notonly, asshown bythegalvan- ometer, didthey vary inwaves ofconsiderable amplitude—say ene totwo minutes between the crests—but there was also a constant bubbling when atelephone was substituted, showing continuous quicker vibrations tobepresent. These wecould hardly putdown toconduction orinduetion from neighbouring telegraph wires, asthebubbling wasequally noticeable at2and 3a.m., and allSunday, when fewornomessages would beabout. The telephone would always bubble ifsetfine enough, even between “earths” afewfeet apart. mr. Mr.W.P.Granvitie: Withregard toMr.Preece’s remarks ‘nelle upon the sensitiveness and suitability ofthe telephone for researches ofthiskind, Ihave hadtheopportunity oflistening toatelephone joined incircuit with twoearth-plates submerged intheThames about 100 fathoms apart. Onputting the tele- phone tomyear,itbecame apparent, thattheriver wasperfectly 1999) DISCUSSION. 180 teemingwithMorsealphabets,andatalltimesofthedaythea.rapid clicks ofMorse instruments could beheard. Iamnot arapid reader bysound, butIcould now and then manage to interpret. aword, oreven part ofasentence. TheCuatRuaN: Thefirstduty,inclosingthisinteresting trierdiscussion, istopropose ahearty vote ofthanks toMr.Kennelly forthepaperthathehasbeensogoodastosendusfrom America, Assome ofthespeakers have pointed out, there seems tobesome little doubt—possibly some would saymuch doubt— sto thecause ofthe effects which Mr. Kennelly hasobserved. Theimportance oftaking anearth-wire outtoseahasbeen urgedbysomeofthespeakers, butitmustberemembered that when weusethesheathing ofacable asourearth wedoreally take anearth-wire outtosea; sothat Ihardly seehow such an carth-wire could improve matters, especially asthe cable which wastaken outtoseadid solittle toimprove matters that Mr. Kennelly came totheconclusion that thewhole basin isin- sulated. Some ofhisexperiments, however, donot seem tobe ‘consistent with, and arenotexplained by,that hypothesis. Take, forinstance, Experiment No. 9. Hesays: “The last two “experiments were repeated with thecondenser removed and “the mirror indirect circuit between cable andsheathing. The “continual vibration ofthe spot made observation difficult ;but “daring periods ofcomparative quiescence itwas soon deter- “mined that thedisturbance produced bydepressing oneofthe “keys was notofamomentary nature only, butconsisted ofa “small deflection, permanent during the whole period ofkey “application.” Icannot seehow that experiment can beex- plained byanywant ofgoodness oftheearth; howitispossible inFigs. 7and8togetasteady current ontheother cable which isnot connected with the battery. Hesays you do,butIdonot seehow theabsence ofgood earth would explain that, Take the ‘ase,again, ofExperiment No.7,where themirror galvanometer isconnected with the copper conductor which goes toIreland. Attheotherendofthatcircuit,hementions somewhere elsein thepaper, there isacondenser; sothat infact thecopper ‘conductor going toIreland iswell insulated, orassumed tobe 160 ELECTIONS, (Feb21, Rotewor welll insulated. How, then, asteady current canbeproduced through that conductor, assuming that thecable iswell insulated, byadefect ofthe earth atNova Scotia, Idonotatallsee, Some oftheexperiments doseem more orlessconsistent with bad earth, but these others arenot;so,ifIamsupposed inany way, asChairman, togive averdict onthediscussion, Icanonly saythat myjudgment will have tobereserved until further evidence isbrought forward. Really myownmind isundecided. Twould notgoasfarassome speakers appear tohave gonein concluding that Mr. Kennelly iscertainly wrong; atthesame time Iwould notgoasfarasMr.Kennelly hasgone inassuming that heisperfectly right. Iwill therefore leave thematter, as inthecase ofaScotch verdict, inastate of“not proven.” Aballot took place, atwhich thefollowing were elected :— Foreign Members: Colonel Huber. |A.Paoletti. Members? Emile Garcke. |Willoughby Statham Smith. Associates: Samuel Wells Cuttriss. Frederick Arthur Pock- Lawrence H.8.Ellson, lington. Charles E.Hodgkin. Robert Hodgshon Postle- Arthur Hough, thwaite.Robert Ewer’ Pemberton George Edward Pritchett.Pigott. WilliamStevenson. Alfred Mills Taylor. Students: George William Bousfield. |Edwin S.Jacob. John Leonard Thomson. Fig. Experiment N?1, TeRyeBeach~ie+L SS trv Cx Fig 2. Experiment:N?2, tyemast . fe Fig 3. Experiment, N°3. TeRyeBeack ToTred ijo? ; _ 1860.3 SOMEELECTRIC LIGHTING, Bre.- 1st TheOne Hundred andEighty-seventh Ordinary General Meeting of the Institution was held atthe Institution ofCivil Engineers, onThursday evening, February 28th, 1889— Dr.J.Horxixsox, F.RS., Vice-President, intheChair. The minutes ofthe Ordinary General Meeting held on February 21st were read and confirmed. The following transfers were announced ashaving been approved bytheCouncil :— From the class ofStudents tothe class ofAssociates— William Peto. i Frank Taylor, Donations tothe Library were announced ashaving been received since thelast meeting from 8.8.Wheeler, Esq., the Proprietors ofThe Electrician, and E.Hospitalier, Foreign Member, towhom thethanks ofthemeeting were duly accorded. ‘The following paper wasthen read:— SOME ELECTRIC LIGHTING CENTRAL STATIONS IN EUROPE, AND THEIR LESSONS. ByProfeisor Gronce Fonses, F.R.SS. (L. &E.), Member. Part I. Before commencing thesubject ofthispaper, Ithink itmay bewell ifIgive yousome idea ofthereasons whyIhave brought thesubject toyour notice. Youareaware thatgreat strides are about tobemade atthepresent time inelectric lighting inthis country, specially bymeans oftheGaulard &Gibbs system with theconverters putinparallel, Tocomplete thepractical details ofsuch asystem involves alarge amount ofexperiment, and vhether low-pressure orhigh-pressure work isinquestion, Ihave 162 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb, 28th, frequently maintained that agreat deal ofexperience hasstillto beboughtandtimetobewasted. Ithasappeared tomethat wecanonly hope toensure thebest success ofelectric lighting schemes nowinhand ifwebury ourpride alittle, andtrytoobtain theexperience offoreign countries without payingforitorwasting time over it,asthey have done. Some irresponsible persons have considered that Iammuch toblame forthus advocating thedis- cussion, and consequent approval ordisapproval, ofsystems which have been developed inforeign countries. Idonotsaythat we arenoteven now capable ofdoing much better work than these countries, aswehave done with steam, and Iam certain that we shall dosointhecourse oftime; butIdosaythat wearewanting intheexperience, andIseeallover thecountry work being done where therules ofexperience have often been setatdefiance; and myobject inintroducing this subject ischiefly forthesecond part ofthetitle ofthepaper, namely, adiscussion ofthe lessons which can bederived from astudy ofthework which bas been done abroad. The object ofthis paper, then, istoraise adiscus sion onthe details ofcentral station lighting and electrical distribution, and toenter aprotest against defective systems, which, without some such discussion, Ifeel sure will beasplen- tiful inthe near future with usasthey have been uptothe present time, Ofcourse Iamaware that many ofmyhearers have appre- ciatedthesepoints,andhave,likeme,triedtolearnfromthose who have hadmost experience ;andIknowthattheywillpardon mefordrawing attention towork that they have already ex- amined, since they know aswell asIdothe need wehave for such discussion. Ibelieve that myobject canbebest attained byproceeding, first, todescribe theprincipal central stations inEurope which I have lately visited, pointing out specially the features inwhich they differ from thepractice inthis country sofarashitherto developed; andthentodiscussthedetails,withaviewofderiving the most profitable lessons from them astowhat toadopt and what toavoid. Iwill deal only with what Ibelieve tobethe three most important central stations inEurope, namely, Berlin, 188) INEUROPE, ANDTHEIRLESSONS, 168 Rome, and Milan. Ihave already partially dealt with America ‘eliewhere.* Berlin issupplied onthelow-tension direct-current system, as sriginally developed byEdison, except thestreet: arelights, which areonahigh-pressure direct-current, circuit. Rome issupplied altogether, both forarcs and incandescent, ontheGaulard &Gibbs system, with converters arranged in parallel. - ‘Milan isworked byThomson-Houston apparatus forares, and byboth low-pressure currents and high-pressure alternating currents forincandescent lamps. AtalltheEuropean stations which Ihave visited Ireceived thesame cordial weleome which Ihave always hadinAmerica, andthesame desire that Ishould thoroughly inspect every part ‘oftheworks. Thestationshaveliterallybeenplacedatmydisposaltowatchtheactionofthemachinery andtoseewhat attention isrequired foritsmanagement. Ihave been allowed ‘tointerchange dynamos atwill, and generally tolearn allthat ‘aspossible, Imust especially record mythanks toSignor Pouchain, Professor Mengarini, Signor Trenza, Professor Colombo, Mr.Lieb, Herr Rathensu, andHerr Datterer forthewillingness with which they imparted tome somuch information inthe details oftheir working, and toMr. Zipernowski and Messrs. Ganz &Co, formuch valuable information. BERLIN. The Berlin mains areatpresent, Ibelieve, supplying a lager number oflights than any other connected system of mains inEurope. They supply 36,000 lamps of16candle-power, ortheequivalent current, and 144arelamps of15amperes. The conductors arealllaidunderground. Amap isshown (Plate 1, Fig.1)giving theposition ofmains already laiddown andthose inprocess ofconstruction. There arethree central stations at work, and two more ofimproved type areinprocess ofconstruc tion,‘Theexistingstationsare— ‘©British Amociation, 1888, 164 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb. 28th, 1,Markgrafen Strasse. 2,Mauer Strasse. 8.Friedrich Strasse. ‘Those about tobeadded are— 4,Spandauer Strasse. 5.Schiffbauer Damm. The first isthe most important; the third isquite un- important. Here the boilers arebySteinmiller, and anthracite coal isused toprevent nuisance from smoke. The foundations areofconerete iniron boxes built upwith brickwork and cement. ‘There aresixengines of160horse-power each, each driving old- fashioned Edison six-legged dynamos. There are also four engines of400 horse-power each, each driving anew type of dynamo direct at80revolutions perminute. ‘These arecom- pound tandem Corliss condensing engines byVan denKirchove, ofGhent, who isone ofthe few foreign makers ofengines who cancompete with this country. The attached dynamos areby Siemens &Halske. The Electric Company ofBerlin arealso making amachine somewhat similar. Adiagram (Fig. 1)is shown ofthisdynamo andoftheengine combined, theheight of which is23feet. ‘These continuous-current dynamos have never ‘been built inthis country, and areworthy ofsome notice. ‘They may bedescribed asmultipolar Gramme machines, the poles being radial, and the ring armature external tothe poles. A photograph ofthefour engines and dynamos isshown (Plate 2, Fig. 1). ‘There aretenpoles oneach machine. The armature is3 metres indiameter;thecommutator is14metresindiameter. ‘Therearetenbrush-holders. Eachpairofpoles,andthecorre-sponding part ofthe armature, with thecorresponding pair of brush-holders, constitute asection which isvirtually onemachine. Allthese sections arejoined inparallel. The field magnetsare notseen inthedrawing, being inside thearmature. ‘Two handles areprovided, oneofwhich adjusts allthebrushes atthesame time; theother puts them onalltogether. Each brush-holder carries four brushes. Another drawing isshown ofasimilar dynamo being made bytheBerliner Elektricitits-Werke, inwhich vm sUORE, aD THER Eason ws themdialpolesareexternal tothearmature. Ibelieve thatthistypeofmultipolar machine isagooddesignwheretheobject ees \ra\ zPT|qei= : he, Bo GANfle 5 eK D>} r=@ise ittogetalargeoutputwithaslowspeed.Iholdworkingcravingsinmyhandofasimilardesignprepared bymyselfin 105 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb. 28, 1881, but itwas mever executed. The number ofbars ofthe commutator andthenumber ofparts generally make itexpensive to construct. Atthis station the boilers arefedautomatically bymeans of ahydraulic accumulator. ‘The water forcondensing isdrawn from fifteen artesian wells which have been sunk forthis purpose. Steam isused atapressureof115Ibs.anditisclaimedthat 15°5 Ibs. ofsteam only arerequired per electrical horse-power. This isprobably themost economical production ofelectricity for central station work intheworld, and the engines and dynamos, though expensive, must. certainly have avery low depreciation, Themaximum work which this station candoistogenerate a current for26,000 lamps of16candle-power. Itisfound that themaximum number oflamps inuse atanyonetime isabout 66percent. ofthetotal number oflamps. The thing which strikes one most inthis station isthe enormous mass ofcopper. The conductors inside this station consistofeightstripsofcopper,6incheswideandrdofaninchthick.‘Theseareledintotheroomforcableterminals. ‘These strips, before entering this room, areinterrupted byagap, ands junction ismade bypinching seven similar strips, about 18 inches long, between thestrips and the interrupted ends, the seven strips being placed between. ‘Thus, incase ofany serious accident, thewhole ofthe mains ean, byoneturn ofascrew, be disconnected from the dynamos, the interlapping pieces all Groping out. The only exception isinthe case oftheatres, where, toprevent panic, aseparate connection ismade with the dynamos. Inthecable terminal room thepositive andnegative mains run horizontally, and vertical rods come down tothe different cables. ‘There areeighty-four ofthese cables, forming forty-two pairs offeeders supplying thenetwork ofmains with current onthetwo-wire system. These cables aregenerally of large size, being frequently 3inches indiameter complete. They consist ofstranded cables covered with jute prepared with bituminous compound, enclosed inlead, then covered with tape andpreservative compound, and finally armour-plated with two crossed spirals ofiron ribbon. The system offeeders andmains 189) INEUROPE, ANDTHEIRLESSONS. er isoneofthe features towhich Iwish todirect the special attention ofEnglish engineers, since the principle has been ‘opted oftrying toproduce assmall avariation ofpressure in themains aspossible. This isaccomplished byhaving alarge number offeeders. The cost ofthe underground cables has hitherto amounted toabout £90,000, half ofthis being spent in ‘mains and half infeeders. The greatest variation ofpressure allowedinthemainsis1}percent.‘Thelossofpressure inthe feeders during periods ofmaximum supply is15volts. The next point which Iwish todraw special attention toinconnection withtheBerlincentralstationworkisthe typeofcable which Ihave now described. This cable succeeded verywell forthree years, butlately they have been giving way verygenerally. The lead gets eaten into and water percolates tothe copper, which isthen destroyed. Atfirst itwassupposed thatthedeterioration came from thecable being pierced bythe borers used bythegaspeople, butitwasfound that thesame fanlt isproduced even iniron pipes which have remained injured. Itissupposed that thedeterioration ofthelead mmaybedue insome way toitsforming galvanic element with theironsheathing. Whatever thecausemaybe,thefactseems tobeestablished that such cable will notstand undergroundelectriclightworkformorethanaboutthreeyears.Thesecables generally run under the footways without any casing, and are connected bycircular junction boxes tothe houses, powerful lamps and nosoldering being used forthecontacts. Allthe dynamos feed direct ontothemains inparallel. One feeder alone, serving themains close tothestation, hasaresistance in circuitinthestationtoequaliseitslossofpressuretotheothers. Afewdetached facts strike measbeing worth noting. Fach 16candle-power lamp uses 50to55watts. ‘Thechargesforelectricsupplyare6s.perlampperannum, and96d.perBoardofTradeunit. TheAron meter isuniversally employed. ThepriceofgasinBerlinis4s.10d.per1,000cubicfeet. ‘Thedistances from central stations tolamps extend to1,000 1200 yards, 168 «SOME ELECTRIC LIGHTING CENTRAL STATIONS [Feb. 2b, 100 kilometres ofcable have been laid on the two-wire system, EachoftheJargemachinescansupply3,000to3,500lamps of16candle-power. ‘Thelargest cable is1}square inches ofcopper insection. Ispent along time inthisstation, andhadmachines taken offandputon—anoperation whichoccupied twoorthreeminutes, Ttook careful note ofthe amount ofskilled attention which was required, andsawthatitwasconsiderable. Aman ofhigh training, accustomed tothe use ofscientific instruments, is always inattendance and looking after theamperes and volts from thedifferent machines and pilot wires, and recording the measurements. Itstruck methat the working ofthis station wasthoroughly German—certainly notAmerican—but itwasmostmethodical andveryeffective, Oneofthemostinteresting parts ofthiscentral station istheswitch-board, with thespecial means forintroducing afresh dynamo inparallel, orcutting itout. This alsoisapoint towhich sufficient attention hasnotgenerally been given inEngland. Inallforeign stations that Ihave ever seen, when acurrent ofover 100amperes isswitched inorout,it isfoundnecessary tomakeuseofanauxiliary loadintheformof resistances, abank oflamps being generally employed asbeing thecheapest resistance available. When anew dynamo isgoing tobeintroduced inparallel, itisfirst worked onanartificial load, and theexciting current regulated until thevolts and amperes agree with those oftheother dynamos with which itistobeput inparallel. Assoonasthisequilibrium isattainedthemainsare switched into connection with the new dynamo and itsload. After that theload istaken off. This prevents thepossibility ofthedynamo thus introduced being caused torunasamotor bythe current from the other dynamos, and prevents too violent asparking atitsbrushes. ‘The switch-board atthis central station forthe four large dynamos, with thesccom- panying artificial load orresistance, and with iteammeters and voltmeters, isvery perfect. Facing theoperator, the switch- board isinfour sections forthefour dynamos. Each dynamo hasthree massive double-pole switches—one forthedynamo, one 13) INEUROPE,ANDTHEIRLESSONS, 169 forthe artificial resistance, andoneforthemains, Plugs can be inerted forvarying theresistance onthe artificial load. The ammeterisplacedabovetheseswitchesforeachmachine,anda voltmeter canbeattached toany one ofthem,asrequired.A Photograph ofthis switch-board isreproduced (Plate 2,Fig. 2). The second station isinthe Mauer Strasse. This station els tofeed the mains forincandescent lamps. Italso supplies aelights arranged intwelve circuits allarranged inparallel, achcirenit having twelve arclamps of14to15amperes. Here (bree Siemens machines ofthe oldtype aredriven byacounter- ‘aft. Three Siemens &Halske new-type multipolar machines axealsoatwork,andfourEdisonmachines andsixmultipolar machines oflowtension, 11,000 lamps of16candle-power can besupplied besides theares. Sinty-eight feeders, orthirty-four pairs, start from this Station, ‘TheFriedrich Strasse station issmall, and was thecommence- ment ofthe present system. Here four six-legged old-type Edison machines, each of75horse-power,aredrivenbyArmington- Sims engines, and feed the general system ofmains. Fifty- twomen are employed onthe three stations ineight-hour shifts, The Berliner Elektricitits-Werke paid lastyear adivi- dend of5percent.; theAllgemeine Elektricitate Gesellschaft, which does thecentral station work, paid lastyear 7}percent, Inspecialdistricts 800or900lamp-hours perannumareused; inresidences only 400 to500 lamp-hours per annum. Load diagrams showing thecurrent used atdifferent hours ofthe dayineach month, and theproportionate cost from different ‘auses, fixed andvariable, have already been published, but are herereproduced (Plate 3). Max. The central station inMilan isnotofthehighest importance ‘©faraswecanlearn anything from it,except ononeimportant Point. The principal streets throughout thewhole town are lighted byThomson-Houston arelights tothenumber of350. Thislighting gives ussome idea ofthemanner inwhich this metropolis will soon belighted; and anyone who sees itand You. xvm. 12 170 SOME ELEGYRIO LIGHTING CENTRAL STATIONS (Feb. 28th, realises that this clear, white, brilliant illumination may bemade tobrighten ourwider thoroughfares andtopenetrate ourdarkest slums, willrealise that when London isasfaradvanced inadapt ingthese possibilities ofcivilisation asMilan is,weshall lead healthier, happier, and purer lives, and ourstreets will beless notorious throughout theworld asthehotbeds ofimmorality andmidnight assassination. Iwill briefly refer tothepoints which struck measbeing most worthy ofnotice. ‘The station has 14,000 incandescent lamps of16candle power. The maximum load is60percent. ofthenumber of ights installed. Atthis station ithasbeen found quite practicable touse Siemens ratchet arelamps ofonly 4}amperes, theupper carbon deing 12mm. diameter and 200 mm. long, thelower one7mm. diameter and 200mm, long, lasting 7}to8hours. For longer duration larger carbons areused upto15hours. This isthe first station where Ihave seen anarclamp ofsolowacurrent in use. Acertain resistance isputinseries with thelamp, and two Tamps areputinseries onthemains feeding theEdison lamps. ‘The LaScala Theatre has2,800 glow lamps of16candles and 30ares onthe same mains, Edison meters areused and have given great satisfaction. ‘The price isfixed onasliding scale varying from 7d.to1s.per Board ofTrade unit. Gas now costs about 6s.9d.the 1,000 feet, Before theelectric light wasintroduced itwas10s.6d.the 1,000 feet. ‘TwoofthetheatresaresuppliedbytwoZipernowskimachines, each of40amperes and2,000 volts. Forthese circuits Siemens’ concentric cables, ofwhich aspecimen isexhibited here, are ‘used, thesection being 28square millimetres. Four Thomson- Houston machines aresupplying 35arclights each, and four more aresupplying 30. Half ofthese aresingle and half double carbon lamps, thesingle onesbeing putoutatmidnight. Ishould mention that thecoalused costs 27s.:6d. per:ton, but labour ischeap. ‘| ‘TenBabcock &Wiloox boilers supply thesteam. Theengines 1) INEUROPE,ANDTHEIRLESSONS. m ‘weallnon-condensing. Thelow-pressure mainsareon@two- iresystem, ‘Twenty-eight cables, orfourteen pair offeeders, supplycurrenttothemains. Nopilotwiresareused,butaman readstheampere-meter and sets thevolts byatable. There are tenEdisonmachines ofwhatisknownastheJumbotypefeeding mains onthetwo-wire system. They were originally supplied achwith aseparate regulator foradjusting theexciting current, batconnecting gear hasnowbeen introduced which works them allsimultaneously. Afanisprovided fordriving airthrough Ilesinthepolesofthefieldmagnets. Thewholeofthelow- tension system isantiquated, and wehave little tolearn from it eaept what toavoid. Ifound thearrangements forswitching dmmamos inand.out worked very well. Anartificial load. of lumpsisusedatonepartoftheroom,amaingoingthencepast whdynamo, Thustwomen-are: required toputinanew dmamo. Itisfirst put ontothelamps, themachine being ‘azitedtothesameextentastheotherones.Theloadoflamps ischanged until thevolts arethesame ason,theworking circuit. A.igualisthengiven,andthedynamowithitsloadisswitched ™tothe working mains. The load oflampsisthenremoved. ‘Thisissimpler thanatBerlin,owingtothefactthattheexciting arent isknown beforehand, but less simple owing tothe dmamoswitchesbeingdistantfromthelampsandvoltmeters.‘Tenmachines work thus inparallel without trouble, which isa ‘erygood performance. ‘AtMilantheaverage costforwiringhousesisonly£1per lamp. Wagesareone-fifthofthetotalworkingexpenses,coalis cochalf, lamp renewals (undertaken bythe company) is7per at,Thecapital is£120,000, ofwhich£24,000 hasbeenspent iamains, The company haspaid adividend forseveral years, huissteadilyincreasing,andwas4percent.lastyear.Thereis 4large reserve fund. Plane are exhibited, one showing the positions ofthe arc lamps, andalso thealternate-current circuits (Plate 4,Fig. 1),aadtheothershowingthelow-pressure mainsandfeeders,with thejunction boxes and distributing boxes (Plate 4,Fig. 2). Anotherdrawingshowsthestationinsection,withtheengines WtSOMEELECTRIC LIGHTING CENTRAL STATIONS (Feb.28th, anddynamos ontheground floor, and theboilers onthefirst floor (Plate 4,Fig. 3). Another shows theplan oftheengine ‘and dynamo room (Plate 4,Fig. 4). Rome. ‘The station atRome, started bythegascompany inthat city, isthe finest example ofanalternating-current central station which Ihave yetseen, But ithasbeen very costly toestablish. Many ofthedetails contain hints ofthe greatest usetothose who areatwork onthis subject. The whole plant hasbeen contracted forbyMessrs. Ganz, ofBuda-Pesth. ‘Ageneral plan isshown (Plate 1,Fig. 2)oftheelectrical plant inthis central station—all ontheground floor—the different parts being indicated bythefollowing numbers andletters :— ‘Nos. 1to8.—Eight Babcock &Wilcox boilers, of164 H.P. each, already inplace. Total, 1,812 H.P. Nos. 9to14.—Space where sixother boilers, similar toabove, will beerected. Aand B—Two self-exciting dynamos, 2,000 volts x45 amperes, and 150-H.P. engines. Land I—Two dynamos, 2,000 volts x160 amperes, and 600-H.P. engines. IIL. and IV.—Space fortwoother dynamos and engines similar toT.and IT. ‘n,n, n.—Three exciters, andWestinghouse engines, C.—Place where arefour feed-pumps forboilers. ‘M.—One 50-H.P. gasengine, with dynamo, 30,000 watts, andexciter. ‘Therearetwoself-exciting alternatorsof150horse-power each,, and two independently excited ones of600horse-power ead ‘These give 42amperes and 160amperes respectively, and2, volts. Atpresent 9,000 glow lamps of16-candle power supplied, and over 200arelamps, Allthemachines arecapable ofworking inparallel, even thelarge andsmall ones togeth ‘Thenumberofalternations is5,000perminute,or2,500compl periods. ‘The greatest distance towhich current isatpresent supplied is44kilometres, orabout$miles,Thefeedersallgo i 9) INEUROPE, AND THEIR LESSONS, 113 thePiazzaVenezia, andthenbranchoffintothreemainsin diferent, directions (Plate 1,Fig3). The loss inthesecondary éreuits due tolossinthefeedersatpresentisonly0-6voltat themaximum,andthatduetothemainsisthesame,thelength offeeders being 1,600 metres. Siemens’ concentric conductors areused throughout. The feelers areeach 220 square millimetres section, or0-35 inch, Drawings ofthejunction boxes bySiemens &Halske areshown (Fig.2),indicating themeans ofattachment byclamps ofthe SS ia) ‘Fro, 2-—Janction BoxbySiemens&Halske, twoendsoftheinnerandouterconcentric cores.‘Thecablesare |ad inswooden box, which isthen filled with cement. 17kilo- ‘tres ofcable have been laiduptoJanuary, 1889. Thefeeders tebeenworking since September, 1887. Theconverters are Wof10horse-power, withabout4amperes intheprimary and ‘Sinthesecondary. 110volts areused inthesecondary, but tesecondary coilhasthree terminals, onehalf-way along the lagthofwire.Thisistoenableathree-wire systemtobeused ‘thesecondaries when arclamps areused, consuming, with their ‘eistances, 55volts each. 174 SOME ELEOTRIC LIGHTING CENTRAL STATIONS (Feb. 25, Theoldtype ofZipernowski converter, with iron wire wound round thecopper coils, hasbeen abolished. ‘Theironpart isbuilt upofcircular flatring-shaped dises ofiron. These arefirmly ‘clamped together byiron clamps, which form thesupport ofthe instrument. Thecoppercoilsarewounduponthesegments ofthe rings between theclamps. ‘Theprimary coilisunderneath the secondary. The converter hasaniron circular diso attopand bottom fixed totheclamps, and this enables the’instrument tobe easily rolled about without injury, which facilitates handling. Onthetopisaporcelain disc, onwhich areplaced theterminals andfusiblecut-outs. Theseareshortstripsofthinmetalheld inpasteboard, andcanbeslippedinandoutwiththegreatest ease. Drawings ofthese converters areshown (Figs. 3and4). ‘Thesmallmachines areintended towork1,000lampsof60 watts, There aretwenty poles, and250 revolutions per minute, ‘Aderivation istaken from acommutator toexcite the machine. Sulze engines areused forthese dynamos, acting direct onthe dynamos. ‘The steam pressure is120Ibs. Aphotograph ofthese machines andengines isshown (Plate 2,Fig. 3). ‘Theautomatic resistance forregulating theexciting current, invented byM.Blathy, isvery effective. The exciting current passes through asolenoid actuating aniron rod which is supported atoneendofabalanced lever. ‘Thebarisweighted with afloat inwater, and water canbeputinto thefloat. On thetopoftheiron barisacupofmercury which, infalling, breaks contact with the ends ofwires dipping down into the mereury cup. These ends areatdifferent depths, sothat the wires arecutoutofcircuit insuccession. The’ wires goto resistances. Thisautomatic adjustment actswel’over’cén- siderable range. ‘The exciters forthelarge dynamos aredriven byWesting- house engines. ‘They are governed byhand forthe:large changes, andthen theautomatic regulation comes in,Such a method ofregulation isnecessary with machines having solarge ‘anamount ofself-induction. This isthe only satisfactory regulation ofthekindwhichIhaveseenatwork.Itisoneof thesmall details onwhich thesuccess ofastation depends, bat 1889.) INEUROPE, ANDTHEIR LESSONS. 16 [oso \\l1// aNfatsPegFWP oe Itienoesl 16:altonvrr, (Bale1hasaok) —LS SS | 1 i } 1 biel AAMT WG 3) | id Pu,.—Zipermowaki 7-5-UnitOonvertar. (Beale1}}inchestothefoot) 176 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb 28tb, which areoften forgotten bytheinexperienced until thewant is felt. ‘The twolarge machines have forty poles each ontherevolving field magnets. Both field magnets and armature aredivided into twoseries inparallel. When themachine isilluminated by fanarelight, towhich itsupplies thecurrent, acurious optical effect isproduced. ‘The arebeing periodically made andbroken, therevolving magnet-poles areseen fixed inposition, and the amount oflagwith different loads can beseen directly. ‘The efficiency ofthese machines issaid tobe90percent., including theexciting current, These machines aresoconstructed that thecover which contains thearmature coils can bedrawn away, bymeans ofahandle, shown inthephotographs exhibited, over the electro-magnet ring, without the necessity ofusing any hoisting tackle orother machinery. Moreover, the construction ofthefixedarmature issuchthatanysinglecoilcanberemoved inafew minutes,orexchanged. Theironofasectionofthe armature is in this case removed with the coil. The iron plates ofwhich these armature sections aremade areT-shaped,s0 that the centre ofthe coil isfilled with iron, In the smaller machines there isnotnearlysomuchironinthecoils.Inthe 600-horse-power machines the energy absorbed inexciting at fullloadissaidtobe34percent.ofthefullload.Photographsareshownofthelargeenginesanddynamos(Plate2,Figs.4and5). ‘The T-shaped iron plates ofthe armature (K') (Fig. 5)are pressed together bystrong bronze press plates (S',S*)andascrew (C"). Thewire bobbin isthen slidonandheld byserews. ‘These segments arefastened totwotraverses (T)bythescrews (R)y ‘which traverses areinturn screwed with aninsulating layer (U) totheside-plates oftheframework ofthemachine. . ‘The field-magnet wheel isconstructed asfollows:—U-shaped soft iron sheets (KK) aresoarranged together astoformastar, towhich isadded asimilar star, with aninsulating layer between, placed insuch away that theinterstices between the sheets composing thelower stararecovered bythesheets of theupper star. Inthis way many iron lamine areputoneupon theotheruntiltherearesufficient toformthewholeelectro- ae 118 SOME BLEOTRIC LIGHTING CENTRAL STATIONS (Feb. 26, magnet wheel. ‘The pile soformed isthen pressed together to form acompact whole bymeans oftwo stiff dises (S), twobosses (N), andtheserews (C). ‘Thebobbins (M) areslipped ontothe core (K), andheld down bymeans ofthebobbin-holder (H) and screws (C’). The subdivision oftheiron inthefield magnets is necessitated bytheteeth ofiron inthearmature projecting inwards through thecoils. ‘The large engines arebyGanz &Co. They arecarefully balanced, andareprovided with starting gear. (O © | ok|See BhOS wap ‘al| gpaOmesie8 aR TH OPA eS M3 ABH incdimaiecinemaoni 0+ornAne-DES,Tee a ER FI daabaLanada Was: iWicegiiede! gee Be Cah):Cea alaone amCTTAes i Scale i LLLrrrmente Fie, 6—Rome—Mercory Gwiteh-Board, Adrawing isexhibited oftheswitch-board (Fig. 6),which is very ingenious, butvery expensive. Theboardisabout8feethigh and 17feet long. Itisdivided into four sections, one foreach dynamo, andeach section isdivided into six,oneforeach pair of feeders. Apairof4-foot horizontal bars runs along each dynamo section, connected with the two poles ofthe dynamo. Two vertical barsdescend fromtheseateachofthesixpartsallotted tothesixcircuits. Sixpair ofhorizontal bars areconnected to thesixpairoffeeders (only three areasyetinuse), these pairs 198 INEUROPE,ANDTHEIRLESSONS, 179 being oneover theother. Ateach feeder section apair of vertical copper bars descend from the horizontal feeder bars ‘comesponding tothat section. Thus wehave ateach feeder sedion two vertical dynamo bars and two vertical feeder tury,thelowerendsallbeingatthesamelevel.Mercurycupscanberaisedtoconnect,eachdynamobarwithafeederbar,orloneredtodisconnect them.Thechiefobjectoftheswitch-boardistoenableustoswitchanynumberofdynamos inoroutof connection with any feeders simultaneously and instantaneously. Iwill nowdescribe this action, referring tothediagram. Inthe tormal condition thecrank (A)andhandle (B)would beasshown inthesideelevation, onecranktoeachmercurycup(C).IfitberequiredtoputontothedynamothefeederNo.6,andtakeoffNo,1fromthedynamorepresented bythispartoftheswitch- board, the handles (B), with cranks (A),correspondingtothese feedersarepushedbodilytotheleft;thewebofeachcrankis thusmadetofitinaverticalslotinthedisc(E),whichdiscisconnected tothebevelwheel(()byahollowspindletowhich(G).iskeyed,thisbevelwheelgearingwiththebevelwheel(H)keyed omtotheshaft (I),running thewhole length ofthe switch- tard. ‘The crank ofNo. 6would bediametrically opposite that atNo.1,since onewasdisconnected and theother connected toa cireait.Thelever(L)isthenthrownoveragainsttheotherstop, causing theerank No. 6toberaised from itslowest toits highest position, and that ofNo, 1tobecorrespondingly lowered. ‘Thevoltmeters inthis station bywhich thepressure atthe lamps iskept constant areregulated bymeans oftwocircuits, onebeing ashunt onthe dynamo terminals, and the other carryingthemaincurrent. Thesoarepassedthroughconverters,audthesecondary currentsactuateasolenoidinoppositedirec- tions. ‘This compound-wound voltmeter indicates thepressure at, thelamps,andismaintained constant. Professor Mengarini bas devised aningenious recording voltmeter formarking inadaily carve thepressure atthedynamo terminals ateach minuteof time, These show variations of12 volts atdifferent hours, Reproductions oftheregistered curves arenowexhibited (Figs. Tand 8). 189 SOME ELEOTRIO LIGHTING CENTRAL STATIONS [Fe 280, wo\\\ MARRRRRRURRRUN ananueed \ | \V]| VIVA F| | 1}]] |||HTT ro, 7Gurves taken onProfesor Mengurin's Antomate Reonding "Voltmeter 3 - Fg Fre, ‘Therearefiftyconverters nowatwork,eachof10horse-power. ‘The efficiency oftheseconverters is95pereent.atfulllosd. 180.) INEUROPE, ANDTHEIRLESSONS, 181 ThatofMessrs.Ganz’sS-horse-power convertersis92percent., andthe2}-horse-power is88percent. This isfarhigher than with many ofthose now made inEngland, The House of Parliament islightedbyfourbanksoffiveconverters each. Part II. LESSONS TO BE LEARNED FROM FOREIGN CENTRAL STATION LIGHTING. Ipropose nowtodeal with some oftheproblems ofcentral station lighting which weinLondon have now toface. Ijudge bythepractices Iseecommon inthis country that I shall, astomyopinions onmany points, beatvariance with some ofmybrother engineers, because intheworkwhichhashitherto beendone Iseethat theexperience offoreign countries hasvery often been neglected;andtheruleseemstobetoneglectthe opportunity ofutilising theexperience gained atgreat expense oftime and money inforeign countries, while our progress was being retarded bytheElectric Lighting Act. But myconclu- sions have been arrived atafter amost careful consideration of thesuccessful work which hasbeen done inEurope andAmerica, andIamready tocondemn foreign practice when necessary. All Imayis,that weought tobenefit bytheexperience abroad, and know thatthose whohave thebest interests ofelectric lighting inEngland atheart agree with me. ‘This neglect ofthe principles laid down inother countries I findtoexistwhetherthecentralstationbeusedforcontinuous oralternating currents, Iforesee great danger instore forusif ‘edo notdiscuss thedifficulties fully atthe outset, and Itrust thatthediscussion thatthispartofmypaperwillgiverisetoWillassisttomaketheprogress nowbeingmadeinEngland a thorough success. ‘Tanse-Wine System, Myfirst remarks will beabout thedirect system ofsupply at lowpressure—a system which maybemuch more frequently em- 82 “SOME ELECTRIC LIGHTING CENTRAT STATIONS [Feb 260, ployed with advantage than many people seem toimagine. It isnot suited foralarge district, ‘but ithas some advantages. First—That thesteadiness ofthe light canbeincreased, the fear ofaccidents diminished, and theamount ofmachinery re- duced, bythe employment ofstorage batteries. Second—If batteries arenotused,thelossofenergy intransformation is avoided. Let usdeal first with the case where allthe dynamos are working inparallel atthe same pressure. This isthe Edison plan, andisvery generally adopted. Ihave shown inmyCantor Lectures* that when weextend beyond acertain distance from thecentral station onthesystem atpresent used, thequestion of economy, asestablished bySirWilliam Thomson's law forthe size ofconductors, must bepartially abandoned, andadifferent size used, inorder toprevent the variation ofelectrical pressure over the area being too considerable under the varying loads. With the“two-wire” system thelawofeconomy holds only up toaveryshortdistance; withthe“three-wire” thedistance is doubled. ‘This isdue tothe fact that the pressure isdoubled, and this istheprincipal advantage possessed bythethree-wire system. InBerlin and Milan the two-wire system hashitherto been adopted, andtheenormous amount ofcapital expended in copper isaproof ofthecostliness ofthesystem, Ihold that in nocase ofcentral station work atlowpressure ought thetwo- wire system tobeadopted. Inthis Iamsupported bythe universal practice inAmerica, where thetwo-wire system forlow- pressure distribution isnow totally abandoned. Iamalso sup- ported inthisbythepractice inBerlin, where allfuture work is tobedone onthe three-wire system. Inspite ofthese con- siderations, Iamnotaware ofasingle case where itisinactual ‘useinEngland since theexperiments atLeamington have been abandoned. The two-wire system isused atagood many places inthis country. The result ofexperience probably isthat the three-wire system canbeprofitably worked toamaximum limit. of half amile. ~SGantorLectures,“OatheDatribution ofElectricity,” SccllyofarteFotenal, 1886. * ' 182} INBUROPE, ANDTHEIRLESSONS, 188 FREpens. Inow come toanother principle ofdistribution which is strangely, and Ibelieve very generally, neglected inthis country. Irefer totheuseof“feeders.” Let medistinguish between the use offeeders and mains. Mains are conductors fromallparts ofwhich branches may beledinto thehouses. Feeders areconductors leading from thecentral station tovarious pointsinthesystemofmains,andwhicharenottappedonthe reed. InBerlin andMilan thepractice, first started inAmerica, isfollowed oflaying down anetwork ofmains andsupplying currentatnumerouspointsinthembymeansoffeeders.This serves toequalise thepressure over theentire districts under sarying loads. InEngland thispractice isnotgenerally adopted. Theresult isthat ifyou wander over adistrict supplied by eeetrie light inEngland, you find that when the supply of curent issmall the lamps are equally bright over the whole town, butwhen there isalarge demand forlight theglow lamps vearthestationarefartoobrightandthoseatadistance arefar todall, InBerlin, where half the capital spent inconductors lasbeen used forfeeders, thelamps are, allover thedistrict, equally bright atalltimes. Not theslightest difference canbe perceived atdifferent hours oftheday ornight. Ihold the cpinion that astation with low-tension current without feeders is ladly engineered. ‘Usmic Dynamos atDIFFERENT PRESSURES. Itistheuniversal practice abroad tomake thesizeofthe feeders such that thelossofpressure inthese feeders isthesame inal. This involves theuseofvery large feeders when their length isgroat, andgreat, waste ofenergy when thelength of federissmall,butitenablesallthefeederstobesupplied by dmamos atthesame pressure, which may therefore work in wanllel. This arrangement issimple towork, butitiswasteful. find that wecansave enormously incopper ataslight extra ‘tyable inworking thecentral station. Iwould prefer toseethe feeders, divided into twoorthree groups, feeders ineach group taving the same length, andhaving thesame lossofpressure. 184BOMEELECTRIO LIGHTING CENTRAL STATIONS [Feb.18th, Allthesefeederswouldbemadeoftheeconomical section,accord- ingtoSirWilliam Thomson's law. When thestation isatfall load thedynamos working long feeders have ahigher pressure than those working short feeders. When theload islight allthe dynamos would work atthesame pressure, because thelossinthe feeders isinsignificant. Atsuch times allthedynamos might be {joined inparallel, orthefeeders might allbeupon onedynamo. Posrrion orPior Wines. Innearly allcases oflow-pressure distribution thepressure is regulated bya“pilot wire” coming from thepoint where the feeder meets the main, These pilot wires indicate thepressure atthose points upon avoltmeter atthestation, which iskept at constant reading byanattendant, Sometimes allthedifferent pilot wires aregrouped together atthestation andindicate the average pressure onthevoltmeter. This isdone atBerlin. At Milan, however, thepressure isvaried according toatabular rule when thecurrent varies. The useofpilot wires is,however, very desirable, butIhold that hitherto thepilot wires have genemlly een placed inthewrong position. Suppose that a4percent. variation exists over thedistrict ofmains supplied byany one feeder, andthat thepoint where thefeeder strikes themains is kept atconstant pressure, then themost distant lamp supplied bythat feeder hasavariation of4percent. If,however, the pilot wirecame from apoint half-way between themost distant lamp and thefeeder, and this point were kept atconstant pressure, themaximum variation ofpressure would bereduced to one-half, andwould beonly 2percent. Itappears, then, thatby properly placing thepilot wires thevariations inpressure canbe reduced toone-half; or,keeping thevariations ofpressure the same, thequantity ofcopper inthemains may bereduced toone- half, Inthesesuggestions astodynamosatdifferentpressuresand 1stothepositionofpilotwiresIamsupportingacoursewhich thasnotbeen generally adopted anywhere, buttheimprovement ‘ought onlytobementioned toreceive general recognition. sy INEUROPE, AND THEIR LESSONS, 185 Lames orDirrerent VOLTAGE. Ithasoftenbeenproposed tosupplydifferentpartsofadis- trictwith lamps ofdifferent voltage. Intheordinary plan of usingfeeders and pilot wires themaximum pressure atthemost distant lamp supplied byanyfeeder isthesame asthepressure inthe mains atthe feeder. ‘The minimum pressure may be,let assay, 4volts lese, Suppose 100-volt lamps areused, those near thefeeder will always have 100volts, butthemost distant ones illhave pressures varying from 96to100volts. 96volts would nakethelariaps very dull;andifatthedistantpointslampsof98 voltswereusedthepressurewouldnevervarymorethan2-per cent.from the normal, and abetter light would beobtained. Thiswould beavery good arrangement ifwecould layouta ‘istrict from thebeginning withallthepositions ofthelamps established, but itisfound that asthestation progresses ahouse ‘hich would have required 100voltlamps yesterday mayrequire a linitofonly 99voltlamps to-day, owing totheintroduction of ‘ditional lamps between itand thefeeder. Experiments which tavebeen made inAmerica.on this plan have ledtodisastrous results and hopeless confusion incompleting thescheme. This planhasbeen completely abandoned, andinthepresent state of theindustry Ithink that thisabandonment hasbeen wise. Idonotthink agreater variation ofpressure than 2percept. upand down should ever beallowed ifyouwish tosupply asatis- fuctory light. InBerlin itisonly }ofavolt. Tconsider that success orfailure ofmany electric light companies now doing central station work isvery much ‘dependent ontheir following orneglecting the proper rules fordistribution, and Iwill illustrate theimportance ofthese rulesby«fewexamples. 1.Aplain network bythe two-wire system, fedatthe eatral station with theeconomical size ofcopper conductors acording toSirWilliam Thomson's law, sayof1,000 amperes persquareinch,andgoingtoadistance of960°yards,wouldhave4difference ofpressure towhich thedistant lamps would be tubjected of48*volts. angestoremoveaneverponteontbyMiLantOarpesteritmayaloo Yeoticedthattheforeignpracticeistomaketheneutralwireequalinsection 1D positive and negative wires, ‘VOL, XVII. 13 188 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb. 8h, 2.Onthethree-wire system this would be12volts with one-fourth extra copper. 3.With thesame amount ofcopper asinthelast, properly arranged with feeders, thefeeding points being 320° yards apart, themaximum variation ofpressure inanylamp over thedistrict is2volts, ifthepilot wires come from thefeeding centres. 4.Everything remaining the same asthis last, except that thepilot wires areplaced intheproper position instead ofatthe feeding points, themaximum variation isonly 1volt. Putting itinother words, thequantity ofcopper required to limit thevariation ofpressure over thedistrict toacertain figure isinthe last case one-twelfth part ofwhat itisinthe second case, andone-thirty-eighth® part ofwhat itisinthefirst case, ‘Hion-Pressure Disrripution. Before Ileave thequestion ofdistribution, letmenowapply thesame rules tothe“Gaulard &Gibbs” system ofusing alternating currents with high pressure inthemains and low pressure inthehouses, thereduction inpressure being effected bymeans ofaninduction apparatus called a“converter.” At thepresent moment these converters areusually putinparallel upon themains, There isageneral impression inthiscountry that itisnotnecessary tohave feeders supplying themains, even though thedistance extends toseveral miles. Here again the experience ofother countries hasbeen setatdefiance, although inalarge town likeLondon theimportance ofthisruleisgreater than inmost American cities, orinRome. Ifsuch views continue tobeadopted, theresult willbeasdisastrous inthefuture asithas been inthepast. Inthemost extensive system inthiscountry, when there isalarge demand forcurrent, thedistant lamps look like red-hot hair-pins, even though the near lamps beatfull brightness. This isduetodefective distribution. ‘Asalready stated, there isageneral belief inthiscountry that towhatever extent the Gaulard &Gibbs system isused no trouble comes from variations ofelectricpressureoverthedistrict, Bee noteonprecedingpage. 1) INEUROPE, AND THEIR LESSONS, 187 senwhen theconvertersareplacedinparallel.Letuslookat thisquestionalittlecloser.Returningtothetwo-wiresystem forlowpressure,ifweuse1,000amperestothesquareinchas theeconomical section forourconductors, thefallinpressure is Svoltsfor100yardsthelossofpressureduetothehousewiring amounts very generally to2volts: hence the total variation hetween theperiods ofmaximum and minimum output is7volts per100yards, orabout double what ispermissible (4volts). letusseehow this distance isextended byusing @pressure 0f2,000 volts inthemains and 100volts inthesecondary. With thesamedensityofcurrentthelossofpressureinthehousewill benow5volts for2,000 yards, towhich must beadded 2volts for thehouse wiring and about 1}volts forloss intheconverter, making 8}volts lossofpressure in2,000 yards. Ifweallow 4volts variation inthe lampe, which isquite a marimum, this allows only }percent. variation inthemains, and thenthegreatest distance towhich conductors can extend is 20yards.‘Theonlymeansofextending thedistrictisbythe wef feeders. InAmerica thepractice prevails ofallowing only4per cent. lossofpressureinthemains.Ihavetakenaconcrete caseof1,000 amperes persquare inch, which may notalways apply toactual conditions, butitshows thenecessity ofadopting theuniversal foreign practice ofemploying feeders even with high-pressure mains. Pressure InicaTons oNHiH-Texsion Cincurrs, Itwould behighly inconvenient touse pilot wires with the Gaulard &Gibbs system, such asare used on low-tension cireaits;forthisreasoncompound-wound voltmetersareused, which indicate thepressure atthefarendofthefeeder. Ihave described those inuse atRome and Milan; Ihave also described euewhere those used byWestinghouse inAmerica, The prin- ciple ofthese indicators istohave onewinding connected like an onlinary voltmeter with the terminals ofthe dynamo, and auother winding inreverse direction tocarry themain current, ‘Thediminution intheindication owing tothis second winding is always equal tothefallofpressure inthefeeders, Different 188 SOME ELECTRIO LIGHTING CENTRAL STATIONS [Feb, 26%, inventors have applied different classes ofapparatus foraccom- plishing this result, butitisneedless tosaythat theworking out ofsuch instruments takes along time, more especially insuch an instrument asthat employed bythe Westinghouse Company, which ismade interchangeable forallcircuits bythe inser- tion ofsixdifferent plugs, representing sixdifferent losses inthe feeders, converters, and house wiring, varying from 3percent. to10percent., besides four plugs forthedifferent maximum currents inusewith any feeder. Idraw attention tothis point because Iconsider that the successful working ofsuch acentral station depends largely upon the use ofsuch anindicator of pressure; and Ifind manufacturers inthis country who have had little ornoexperience incentral station work prepared toequip centralstationswithallthenecessary measuring instruments, and Iwish tofeel confidence that these manufacturers have realised the importance ofsuch instruments; and Iwish earnestly to impress upon them, solely intheir own interests, thenecessity of working outsuch instruments ifthey wish toredeem electric lighting inthemetropolis from thecondition into which some of theso-called pioneersofthis industry have reduced it. INSULATION. Wehave now toconsider avery important, and perhaps the most difficult, part oftheproblem ofcentral station lighting— How arewetopreserve our insulation? Itmust benoticed that wedonotrequire ahigh insulation forthesake ofgetting a small loss ofcurrent, butonly because with many insulators the high insulation ismore permanent, andpermanency isthequality ofinsulation which ismost tobedesired. Iamnot going to discuss atpresent thewhole question ofinsulation, butrather to draw conclusions from theexperience ofthepast. Atthepresent moment vulcanised india-rubber and okonite are the substances which have thehighest reputation fordurability andhigh insula~ tion under allconditions. ‘These, like other cables, aresometimes covered with lead, but intheir case itisonly asamechanical protection, especially during theprocess oflaying, because the destruction ofthelead would notdestroy theinsulation. Oflate 130) INEUROPE, ANDTHEIRLESSONS. 189 years aclass ofcables hasbeen much praised, about which Iwish tospeak. The copper conductor, stranded orotherwise, is cwered with any fibrous material, preferably jute, impregnated vith bituminous oils. This covered core isenclosed inthe lead pipe. Inthis class ofcable theinsulation depends upon thelead being water-tight. Weallknow ofmany places where thistype ofcable hasdone good work foralimited time, andmany ofus vereinclined tobelievethatitwasquitesatisfactory;butIthink Tamrightinsayingthatnoneofusknowofitshavingworked successfully inany place forthree orfour years. Ihave said ‘thatacableofthiskindhashitherto beenexclusively usedin Berlin, thelead being covered with layers oftape orbraid soaked inapreservative compound, andthewholearmour-plated withtwo closed spirals ofiron ribbon. Iregret tohave totellyouthatthis insalation hasbeen afailure. Itgoes onwell enough forabout three years, then thelead gives way, thecovering onthecopper ispermeated, and chemical orvoltaic action sets upinthecopper, which itself becomes disintegrated. Sofarasexperience goes, thistypeofcableseemstometobeunsuitable forpermanent workwith theelectric light. Itrust that inthediscussion we stall have some facts brought out. Ihad hoped, from what Berthoud &Borel, andalsoWaring, haddone, that thiswould ‘totbeso,buttheBerlin experience leads metothink otherwise; andonlooking through thetestimonials ofmakers Idonotfind thatthese cables, when placed underground, have ever worked electric light cirouits satisfactorily beyond thethree years fixedby theBerlin people asbeing destructive. Ihave been told, butI cannot believe it,that there isone company inLondon actually proposing toputsuch lead-covered cables without covering into ironpipes. Ofcourse everyone knows they would soon beeaten intobygalvanic action. At Berlin and elsewhere Ibelieve there isthe intention of trying some modification ofthesystem ofusing bare conductors, fistpractically used byMr.Crompton atSouth Kensington. Takingeconomy intoaccount, itisprobable thatforlow-tensioncreuitsthisisanadmirable plan,Itisprobablethatitmayalsobeadapted tohigh-tension circuits byusing oilinsulators to 18 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb. 8th, support the bare conductors. When copper strip isused itis necessary toraise thecover ofthetrough inorder toincrease the number ofconductors: this isvery expensive. For my own part, Iwould prefer adrawing-in system, ifitwere otherwise equally good. Noinformation hasever been given asto(1)themaximum or(2) theminimum insulation resistance oftheconductors at Kensington Court, but perhaps this information will begiven during thediscussion. Atthepresent moment itseems tomethat theonly types ofunderground cable proved suitable forpermanent work are either bare copper supported oninsulators, orelse vulcanised india-rubber, orperhaps okonite, Especial care must betaken to avoid aninsulator which isinjured bythegases which permeate the soil ofatown, orwhich has the property, like piteh, of becoming viscous, andsolettingthecopperbecomedecentralised. Merens. ‘The next lesson which theexperience offoreign stations in Europe andAmerica hastaught usistheimportance ofcharging theconsumers ofelectricity bymeter, like gas. Ifind thatat present most companies inthis country arecharging thelarger proportion oftheir consumers bycontract, making anannual charge foreach lamp. ‘This appears, from evidence collected everywhere, tobeafatal mistake. Itistheuniversal experience that those stations pay best where meters areused. Inmany stations itisfound tobethe best plan tomake afixed annual charge,whichmaybeputdownastherentofmeter,andthen ‘anadditional charge proportionate tothe quantity ofelectric current supplied. This has the advantage ofpreventing customers putting inplace alarge number oflamps which during thegreater portion oftheyear areidle, butwhichon festivaloccasions mayallrequirecurrent. Suchidlelampsare very expensive tothe suppliers, because alarge reserve of machinery must belying idle during thewhole yeartobeready for such occasions. Within thelastmonth Ihave described before theSociety of 1983 INEUROPE,ANDTHEIELESSONS. 1 4Arts* thedifferent meters which urevaluable, without pretending toestimate their relative value. For continuous currents we havetheEdisonandAronmeters,foralternating currents we have theSchallenberger meter, and forboth classes myown “Windmill” meter. CONVERTERS, The induction apparatus used with alternating currents is alled a“converter,” “secondary generator,” or“traisformer.” Thereduction ofpressure from theprimary tothesecondary cirenit isgenerally 20to1,but sometimes more. Alarge‘umberoftypeshavebeeninvented. Ihaveelsewhere deseribed theWestinghouse converter, which Ihave tested and found tobe veryefficient, even though thelargest oneheisinthehabit of supplying isonly forforty lights. Only thelarger size ofZipemowski converter issoefficient. ‘Three sizes aremade, varying inefficiency from 95down to88 percent.,thoughthelatterisaconverter suitable fortwenty-five lamps. Many ofthetypes which aremade inEngland Ifeel surearefarfromefficient. Thereisageneral opinion thatthe lossintheconverter isnecessarily small. This isbynomeans theease. Agreat deal ofcare must betaken inthedesign to seta satisfactory result. Ithink there isageneral idea that any oftheconverters nowonthemarket inEngland willgivean ficiency fortheaverage number oflamps inuseofabout 90 1095 percent. Ishould bemuch surprised ifthere aremore thantwotypes inuseinthiscountry which have anefficiency for theaverage number oflampsover70percent.Yet,sofarasI 4maware, nomaker inthis country ever tests theefficiency of theconverters whichhesupplies. Theefficiency ofaconverter fulla when itisunder-loaded. Ifthe converter inahouse is Adapted for100lamps, and only three orfour lamps areinuse, then, with many converters, thecurrent which isbeing used is double ortreble what isrequired forthelamps. This defect in tomeconverters isclearlyshownintheloaddiagram(Fig.9) oftheGrosvenor Gallery installation, where during thehours of ©SocietyofArteJournal,January28,1899, 192 SOME ELECTRIC LIGHTING CENTRAL STATIONS (Feb, 28th, minimum supply thecurrent sentoutfrom thestation isseen tobeabnormally high—far abovethatoftheloaddiagramofothercentral stations—and atthe hours from 4to6a.m. inOctober current sufficient for4,000 lamps isindicated onthe station meter. This curve shows alossof20percent. The type of converter which Westinghouse hasperfected isoneoflowmagnetic resistance, Theworst ones nowbeing made inEngland areof gift Udoa ||U bee al SmROMOa2a ‘ok| i i nay ‘otute? ?Oe 28* 867 KR ONRI ZSeS CUMUNTOUTPUT,GROSYDORGALLERY Fug, high magnetic resistance. The waste inaconverter, indepen- dent ofmagnetic friction, varies aathesquare ofthemagnetic resistance.* The waste due tomagnetic friction varies asthe length ofthemagnetic circuit, iftheinduction intheiron is thesame, These high magnetic resistance converters require a great deal ofcopper. Themagnetic friction issogreat that they would probably work better with anopen magnetic circuit, as shown byLord Rayleigh (Phil. Mag., 1886). . Stow- anp Hicu-Sreep Eneives. Ineed hardly tellyou that themost startling difference I found between the best European andAmerican practice isthe *S00myformuleforconverters,JournaloftheBoilyofTlaraph-Enginaers andBlacrisans, vol.vi,part71,p.158 193 INEUROPE, AND THEIR LESSONS. 193 question ofspeed ofrevolution ofdynamos. InAmerica 1,000 revolutions aminute isthenormal speed foranalternator of 280horse-power ;inRomethe600-horse-power dynamosrevolve tt125revolutions perminute; atBerlin the400-horse-power dynamos at80revolutions perminute. InEurope there isa tendency tocondemn high speed assuch, whereas high speed is moreobjectionable only inthetype ofmachine where therevolv- ingpartismoremassive,lessperfectly balanced, lesscapableof anding centrifugal strains, andhasalonger distance between thebearings. Inevery other point high speed isanadvantage, ‘specially ineconomy ofplant. Now compare theZipernowski audtheWestinghouse dynamos, choosing oftheformertheonewithout Pacinotti teeth onthe armature, asmade byElwell- Parker, ‘These twomachines areelectrically identical, butinone thearmature isfixed, intheother thefield magnets. Thearma- tarecan,andthefieldmagnetscannot,beperfectly balanced. Theweights arenotverydifferent. Thearmature ismore capable of sanding centrifugal strains than thecoils ofthefield magnets. Forall these reasons theWestinghouse mayberunathigh speed andgain alltheadvantages ofeconomy. Ofcourse theneces ‘tyofintroducing belting hastobeconsidered inthis case. Batthen turn totheParsons alternator, which drives direct, and ‘wing thelightest, best-balanced machine, with great capacity of ‘maisting centrifugal strains, andwith theonly scientific bearings iathe market. Here we have @machine where allthe advan- ‘ugesofhighspeedmaybesafelyacquired andgreateconomy introduced. People areapttoforget that slow speed means lagemachines with noincrease ofoutput. SPEED OF ALTERNATIONS. Now Icome toavery important point. What speed of erations shall weuse? InRome they have 83alternations, 41complete periods, persecond. InAmerica theWesting- touse Company have 267alternations, or133periods, persecond. ‘Whataretheadvantages anddisadvantages, andwhyhavethesetakersadopteddifferentplans?Iwillanswerthisindetailtothebest. ofmyability. First, Westinghouse went torapid 19%SOMEELEOTRIO LIGHTING CENTRAL STATIONS (Feb.th. alternations because he wanted his converters tobe ofreasonable size, Second, hewanted toeconomise plant byusing high speeds, and itwould bevery difficult tohave slow alternations with high speed. Third, Ganz &Co.(Zipernowski) wanted to drive their machines atslow speeds, under which circumstances they cannot easily get rapid alternations. This ishow these things were arrived at, Now letusseewhat theresults are, 1.With rapid alternations thesizeandcostoftheconverters isdiminished. Most people seem tothink this isasmall matter. ‘Amoment’s consideration shows that itisnot eo. Itisfound that immost houses thesizeofconverter does notexactly ftthe numberoflamps,[email protected] 7,500-light plant converters for10,000 lights arerequired. The price ofthese, forthesizes used inhouses, is(astheaverage of many makers) £5,000. The cost ofengines and dynamos com- bined (Parsons) forthesame would beless than £6,000. You will now seethat thecost ofconverters isasimportant as anything else. Mr.Kapp hasstated, Iventure tothink erroneously, that the sizedoesnotvarywiththespeedofalternations. Thisisopposed alike toexperience andtheory. Westinghouse knows this. Tam informed that Mr.Ferranti knows it,and Mr. Zipernowski has told me that he finds it¢oinarecent letter. Itisalso shown in myformule, towhich reference hasalready been made. 2.There is,however, animportant gain byslowing the alternations, Itisthen easier toworkinparallel.Thisisdue tothefact that themachines have more timetogetintostep. IfIconsidered parallel working essential tosuccess, Ishould say, “Sacrifice thecostofconverters, anduseslow speed.” Otherwise Ishould say, certainly and without hesitation, “Use rapid «alternations toincrease theefficiency and economy.” Worktsa iNPaRaLur.. ‘Theexperience ofRome isencouraging toanyone whothinks parallel working essential. Formypart Ido not. Subdivision ofmainsisfarmorepractical inmostcases,andenablesyou 8) INEUROPE, AND THEIR LBSSONB. 195 cailytocutoutspecialdistrictsiffireorothertroubleoccurs. 1kalogets over thetrouble ofincreased resistance inlarge conductors with alternating current, dwelt onbyour President inisopeningAddress. IfIhavemorethanonestationinatown,andIwantonetoassisttheother,Iwoulddosobytrunk linesbetween thestations, towhich adynamo canbeattached at ‘aeend and feeders atthe other. Of course both stations must supply the same type ofconverters; hence insuch system it vwald notdotousedynamos ofvarious speeds ofalternations at tiedifferentstations, [wouldcertainly preferparallelworking being simpler forthemen atthe central station, ifnocom- cations areintroduced tomake themachines work inparallel, audifthesystem were equally economical, butIdonot consider itatallessential. Iventure tothink, however, that theplan toruniversally proposed formaking machines work inparallel vilnotlong betolerated. This istointroduce into themachine slarge amount ofinjurious self-induetion, thus diminishing the fant efficiency and rendering the equalising ofpressure with ‘aying loads very dificalt, Ihave confidence intheingenuity fourinventors, whowillsoondevisealessobjectionable plan. Avtowaric orHaxp Reautation. The question now arises whether the regulation ofpotential ‘ould beaccomplished byhand orautomatically. Inalllow- wreasure systems, such asBerlin, itisdone byhand. Aworkman vongets toknowatwhathoursspecialattentionmustbepaidto luis,andonthelow-tension systemwithshunt-wound orcom- jound-wound machines theregulation required isslight. “Where tesavinginpersonalattention isnotverygreatitisalwaysun-‘dvisabletotrasttoautomatic dodges. Thecaseofalternating- carent dynamos with high self-induction isvery different, forthe tctrio pressure then varies enormously with thecurrent. Inthis emnection such amachine resembles aseries-wound direct-current ‘mchine with high resistance. The changes aresofrequent that ‘omeautomatic adjustment isnecessary. Ferranti, Lowrie, and ‘dhers have used thesagging ofawire toeffect this, andthey 196 SOME ELECTRIC LIGHTING CENTRAL STATIONS “Feb. 28th, regulate thepressure atthedynamo, notatthecentres ofdistri- bution, which iswrong. ‘Theapparatus looks toodelicate foran engine-room, ButGanz &Co,useavery practical type. ‘They also regulate forthecentres ofdistribution, which isright. 1 also prefer the method used byGanz &Co.ofregulating the large changes byhand, andonly thesmall changes automatically. ‘This isnecessary, owing tothevery large adjustments necessary with alternators ofhigh self-induction, Inconclusion, Ifeelconfident that allthose who areworking tomaintain England's supremacy inthis department ofindustry will agree with me that the practical details necessary for successful central station working must bethoroughly well thought out beforehand, and that much saving oftime and money can beeffected bygathering hints from those who have spent years oftime and tens ofthousands ofpounds on these details, Ifrankly admit thatIhavelearntmorethatisofpractical valueinthislinefromtheinspectionofactualworkdonethan byreading books. But wemust remember that thelowprice ofgasinEngland, thecheapness ofcoal, and therelative cost oflabour, allintro- duce special factors which will partially differentiate theEnglish practice either from that ofthecontinent ofEurope orof America. Iwish now that thelastwords ofthis paper may bewords of thankstothegentlemen inothercountries—whether inEuropeorinAmerica—who have thecontrol ofcentral stations, forthe courtesy which they have extended tomeinmyendeavours to assist, tothe feeble extent possible tome, inmaintaining the position ofour own country intheprogress ofthis new and important industry. ‘ThePresiexr: Atthislatehour oftheevening Iamafraid wecannotproceedtodiscussthepaperwhichwehavejustheard,andmust defer itforthenext meeting, which will beheld on March 14th next. - Fig2. : x mseMains. || { i . Se el — we Di octose. J 1 ex\ a POWER NOVEMBER Pi FSTSSSSS DECEMBER, 1 1 be t 1 rosa i SSeS | 1889.) INEUROPE, ANDTHEIRLESSONS, 197 ‘Mr. W. H.Preece: May Iaskifinthemeantime thepaper will beprinted and circulated? ‘The Secretary: Yes. Aballot for new members took place, atwhich thefollowing were elected :— Members : : Henry Graham Harris, |H.Whitby Smith, Associates : William Arthur Britten, |Fiennes Barrett Lennard, M.A. ‘The meeting then adjourned. 198 THE LIBRARY. ACCESSIONS TO THE LIBRARY FROM JANUARY 1TO ‘MARCH 20,1889, (Frerksmarkedthus(#)havebuenpurchased, Ofthosematpurchasedorreteit ‘zchanye,wherethedonor?namasarenotgive,theworkshavebeenproved bytheauthors) Fr 18PARTICULARLY DEsTRABLE THAT XODCDERS sUOULD FRESENT coPtES OF Ht ‘WORKS TO THE LIBRARY A8 $00% A FORSIBLE AFPEX PUBLICATION. Bamber (B. Fj, Baitor, The Scientide Works ofC.W.Siemens, Kt, PRS, DOL, LED. Svola BV,Vol.1.HeatandMetallurgy. 496pp.47plates,»Ul,Electricity and Miscellaneous. 498pp. 97platen 1TIL, Addremes, Lectures,Bc.488pp.11plates.Londen,18 (PresentedbytheExecutorofthelateSirWilliamSiem} Barly (J.A} (Vide Reynier.) BoardofElectrical ControlfortheOityofNewYork.Fix‘andSecondReports,8v0.40pp.and78pp. XowYer8- [Presented by8,8.Whose] ‘©British Association. Reports, 1681-1887. (1889 missing.) 55vols Tendon, 682188 *Clausins [R.] TheMechanical Toory ofHeat, with itsApplications tote ‘Steam Engine, and tothe Physical Properties ofBodies, ated¥y ‘T,Archer Hirst; with anIntroduction byProfewor Tyndall, 8 316pp. Tandon, 1961 Electrical SocietyofJapan.Journalof,(PrintedinJapanese) Jals15,1888.Bro. ‘obi,88[Presented bytheSoci) “Blectrician, The.” he Blectrician Blectrical Trades Directory axlHandbook for1869,La,8vo,cxxxviil,+$18+cexii,pp.Tendon, 188 Harris [Oharles Hope]. [Vide South Australian ostitate ofSurveyors} iret(7.Archer](VideClausing,R] ACCESSIONS TO THE LIBRARY. 199 ‘Hospitalier (B:]YsComptours d’EnergicElectrique, 8vo.87pp. Paris, 1889 Indian Telographs. Administration Report ofthe Indian Telegraph ‘Department for1867-68.Fo.85pp. Sino,1888[Presented bytheDirector-General ofTelegraphs inIndia] Iron and Steel Institute, Journal of.No.2, 1888. 8v0, 406 +Ixvii. pp, Landon, 1888 [Bxchange.} Kelland(Bev.P.)[VideYoung,Thomas} Kew Observatory, Report ofthe Kew Committae fortheYear ending Get. 81,1888; with Appendices containing Raralts ofMaguetial, Meteorological,andSolarObservationsmadeattheObservatory.ro. 28pp. (Proc,Roy.Boe,1888.) London,1888Kichofl(VideRimington.) Mater, [Vide Preece andMaier.) Now York. [Vide Board ofElectrical Gontrol fortheityofNewYork.] Preece [William Henry] and Maier (Julius). The Telephone. sro. 496pp. (Specialist Series.) Tondon, 1689 {Presented byMouera, Whittaker&Co,(Publisher»)] *Remnier [mile]. The Voltaic Accumulator: AnElementary ‘Treatise. ‘Translated from thePrench byJ.A.Borly. 8yo. 202pp. Tendon,1889 Rinington (E. ©.) Kirchoffs Laws and their Application, (ReprintedfromTheBlectricalReview}v0.69pp. London,1888(Presented byMoers.H.Alabaster,Gatchouse,&Co.(Publishery}.) enens (Sir C.W.}, PBA, D.OL, LUD. The Scientifle Works of, Edited byKF. Bamber. Bvols.” vo.‘Vol.I.HeatandMetallargy. 496pp.47plates,»WLBloctrcity andMiscellaneous. 498pp.37plates.1MILAddreses,Lectures,dc.458pp.11platen,London,1889 {Presented bytheExecutor ofthelateSirWilliamn Siemens.) South Australian Institute ofSurveyors. Report ofAnnual Meeting eld Nov. 2,1888, ro, 22pp. [Contains Address byChas, Todd, andPaper entitled “Developments ofTerrestrial Maguetiam affecting “theCompass,” byCharles Hope Harris} ‘Adelaide,1688[Presented byCharles Toda, Member.] ‘winburne (J. Practical Bleotrical Measurement. vo. 156pp.London,1888 [Presented byMears, H,Alabaster, Gatehouse, &Co.(Publishers)) Swinton [Alan ACampbell). The Elementary Principles ofElectricLighting.2ndEdition,12mo.39pp. Landen,1989 200 ACCESSIONS TOTHELIBRARY. *Tait (P.G.j LecturesonsomeRecentAdvancesinPhyticalScience,20 Edition, vo, $62pp. London, 1806 ‘Thomson (Sir William}, LL-D, F.R.8, Popular Lectures and Addresee ‘Vol. 1.ContinuationofMatter.NatureSeries.Cr.8vo,460pp.Tandon, 1889 ‘Todd (Charles), [Vide South Australian Institate ofSurveyors] Trouenfeld (B. von Fischer]. Index ofPublications onMethods of ‘Communication intheField, and onTorpedo Warfare. Sv. 71pp. Landon, 189 (Presented byMesera,H. Alabaster, Gatehouse, &Co.(Publisher9)}‘Tyndall(Prof)[VideClausius,B.] *Young (Thomas), ACourse ofLectures onNatural Philosophy andthe ‘Mechanical Arts, ANew Edition, with References and Notesbythe Rev.P. Kelland. 2vola, [Vol.1, Text; Vol. IL,Plates] By Londen, 1807 201 | ABSTRACTS. Dr. J.A,PLEMING—A DESIGN FOR ASTANDARD OF ELECTRICAL RESISTANCE, (Philosphical Mapasine, Vol.27,p.14,1880) The two chief points tobeborne inmind indesigning astandard of Tesitance arethematerial andtheform oftheinstrument, Mercury, though itcanbeobtained and kept in» state ofparity,isyet,formanyobvious reuont,ratheraninconvenient substanceforastandardofresistance;itis lettertouseawireofplatinum-silver alloy,thepermanenceofwhichmaybe ‘hitly anticipated from previous experience,‘TheordinaryB.A.formofresistanceistoowellknowntoneeddescrip. tiobutsomeofitsdisadvantages maybealludedto,Owingtoitasizeand shape,itrequiressometimeforthewiretotakeupthetemperatareofthe liguldinwhichitisimmersed; andtheheatproducedbythepassageofthecarrentwillnotbeeaslygotridof.‘Theoldformalsocannotbeentirely immersed,asby#0doingthewaterwillshort-circuit thetwostoutrodswhich formtheterminalconnections; hencethebottomofthecoilmaybeatone ‘anperatare whilethetopisatadifferentone.Iftheresistances areusedat tbetemperature ofmeltingice,dewwillbedepositedonthesurfaceofthe‘arafinwax,andwillthustendtoshort-cireuit theresistance,‘The best form ofcoil will therefore beone which permite ofrapld iaipation ofheat, and ofbeing entirely immersed without risk ofshort.cireiting. ‘Theimprovedformdesignedbytheauthorconsistsof«fiatring,‘ithaninteriorrectangular channelinwhichthesilk-covered wireiswounddouble.‘Theconnectionsarebymeansoftwostoutcopperrodssolderedto tbetwoendsofthecoil.Theserodsriseupinsidetwobragstubes,from Which they are insulated byebonite rings, ‘The bottom rings are plain ‘ruhers; the two top rings are hollowed out fannel shape: into these twofemelsparaffinoilcanbepoured,thuseffectually preventing anyereepingof‘moisture.Inconstructing suchacoil,thewireisatfirstcutofLittletoolong;afterthecopperconnecting rodshavebeensoldaredon,theresistanceis ‘measured;toreduceittothestandard,thesilkcoveringisremovedearefally atthebendwherethewireisdoubledonitaelf;thebaredloopisthentwistedtogetheruntilexactadjustment isalmostobtained;thetwistedcoilisthentouchedwithsolder,andthebaredportionisre-invulated withsilk, 3. PARKER-THERMO-ELECTRIO PHENOMENA. (Philosophical Magasine, Vol. 2,p.72,1889.) ‘Theauthor imagines anarrangement ofcondensers andwires ofvarious ‘metals, which permitsofacompletereversiblecycleofoperationsbeingper- formed,asinOarnot'sheat-engine,includingthepasageofelectricityacross VoL. XVI. “ 202 ABSTRACTS thejanctions ofthedifferent metals without producing anythermal effec. From the considerationofOarnot’sprinciple,andofequationsestablishedin «previouspaper,itfollowsthat,iftwoportionsofthesamemetal,atdifeaat sboolute temperatures f,f,respectively, beincontact, andifV,V,bethe ‘potentials which they assume, then VaV,<9K(H-43), whereKisaconstantwhichhasthesamevalueforallmetals,Inother ‘words,thedifferenceofpotentialdependsonlyonthetemperatures, andiste sameforallmetals.If,however,unitchargebemadetocrossthejanction, ‘the heat that mustbeimpartedtothejunctiontokeepitstemperstareoo stantwillbedifferentfordifferentmetals.Ingeneral,ifathermo-electiccircuitbeformedofanynumberofdifferentmetals,andif8betheabraptFigoofpotentialatanyjunctionaswetravelinthedirectionofthecurrent then evidently E=23, Ifthe junctionsbeallatthesametemperatare,we have28=0,andthereforeE=0. ¥.QUINCKE—BLECTROLYSIS OFCUPROUS CHLORIDE. (Annalen, Vo. 86,».270, 1880.) ‘The authorhasrepeatedapreviousexperimentofBuffs.ontheelectrolyis ofcupronschloride.‘Thesaltwaspreparedpurebyboilingeupricchloride‘with copper filings inconcentrated hydro-chloric acid ;andwasthen, avoiding1sfaraspossibleallexposuretotheair,meltedintoaporcelaincrucible,Thit latterwasprovidedwithacoverofporcelainormica,throughwhichpum the two electrodes ofcopper wire made into spirals, The current from 20 Vichromate cells was sent through«tangentgalvanometer,«coppervolt ‘moter,andasilvervoltameter,andcouldthusbeaccuratelydetermined. It ‘varied indifferent experiments from about04toO4ampere.‘Theamountof copperelectrolysedwasfoundtobefromoneandahalftotwoand«ball ‘timesthatcalculated. ‘Thedivergencies dependpartlyupontheactionofthechloride itaelf onthecopper electrodes, which produces adecomposition ofthesaltowingtoaspontaneous electriccurrentwhichvariesverymuchbothit‘value anddirection, andisapparently caused byunequal heating ofthesalt. ‘The author concludes that cuprous chloride cannotbeusedfordetermis- {ngtheeloctro-chemical equivalent ofcopper; andthequestion of«double ‘equivalent forcopper, accordingly uathe electrolyte may be«cupric o‘8cuproussalt,cannotbesettledbyitselectrolysis. ‘BELLATI and LUBSANA-—PASSAGE OFELECTRIO CURRENTS THROUGH BAD CONTACTS, (Beibatter, Vet 18,p.21,1880.) ‘Asmall prismofironpyriteswasplacedwithitsendsintwoveal© ‘mercury, curtent was passed through, and thenoninterruption ofth current thepyrites was connected toadelicate gulvanometar, ‘The «all ABSTRACTS, 208 carreatsobervedcouldnotbetakenasevidenceofpolarisation,butweredue totharmo-electric causeaBymeansof»quadrantelectrometer thefallof‘oteatilttwopointaoneachsideofonecontactwasdetermined, aswellat sttwopointsontheprismitself.Theresistanceoftheselattervariedbetween, 187and740Biemensunits,andwasconstant; whiletheresistance atthe |contactincreased, andwasgreaterwhenthecurrentpassedfromthepyrites |tothemercurythanwhanitpassedintheoppositedirection,Ifthecurrent inkapton,theresistance graduallyincreases, whileareversaloftheourrentisstuadedbyadecrease, Byexertingpressureonthemercurytheresistance decreases,utrstrapidy,andthenmoreslowly.Ifacopperpointispressed ‘uainstthepyritesbymeansofaweightedlever,theresistancewillgreatlyininih—e,, from6099obmswith25grammesto97-2ohmswith636Fanner, Replacing thecopper point byavewing needle, theresistancewas ‘©great that itcould not bemessared, The direction ofthecurrent was slaysofimportance; ariteoftemperature alsohastheeffectofdiminishing thecontactresistance, 1%DRECHSEL—ELECTROLYSIS BY MEANS OF ALTERNATING OUBRENTS, (Beiblater,Vol18,p.28,1889.) ‘Theauthorclaimspriorityforhispublicationofobvervations onthis ‘iatoverMessrsManeuvrier andChappuis(ideJournal,vol.xviii,pp,79,80). Heremarks,besides,thattheexplosionofthemixedgasesisnotalonecaused ‘theheatingoftheelectrodes,astheyamsert,butalsothatitmaybenoticed ten the electrodes consist ofWood's fusible metal, which subsequently stowednotruceoffasion,andwithplatinumwireelectrodes#0arrangedas tobecapableofbeinggraduallydrawnoutofthevoltameter soastopresent,alwaysthesamesurfacetothogases.Eventhen,ifthegasisgivenoffViolently, sothat the bubbles ofnascent gas form elayer between the‘eeizodeandtheelectrolyte,asparkwillpasswhichdeterminestheexplosion. Thesumephenomenon maybenoticedwithelectrodesentirelycoveredwith ‘beLigaid, andwhen continuous currentsareused, DARSONVAL—A UNIVERSAL DEAD.BEAT GALVANOMETER, (BaetindataSoilsInternational desBlacrcions, Vo.6,p.6,1889.)‘Thsinstrument, whichwasspeciallydesignedforphysiological researches, isgaenlly similar toWiedemann’s form oftangent galvanometer; byan‘asylteration itcanbomadesuitableforballisticmeasurements,Onabaseofwoodisfixedagraduatedbar,atonedge,Thobaserests ‘cntheofeet,twoofwhicharolevellingscrews,thelinejoiningthembeing ‘erpendicalartothebat,6thatthegalvanometer canbeatoncesetupin ‘Postionbyplacingthesetwoscrewsinthelineofthemagneticmeridian. (Onthebartidetwoslottedpies,eachofwhichcarries«collofwire;these 208 ABSTRACTS, anboreadilyexchanged forothers,accordingtothenatureofthemeamure. ‘mentstobemade,Thewoodencoresonwhichthecoilsarewoundarehollowedoutononesideintosemi-epherical cavities.‘Fromthebaserisetwocolumns,rightandleftofthebar,carrying« ‘oroes-pisco onwhich reste the closed cylindrical box containing themagnet,mirror,anddampingarrangements; thiscamiscapableofbeingrotatedas ‘wholeintheholeinthecrose-piece,iawhichitrasta,‘Therodcarryingthe ‘mirrorandmagootissuspendedbyafbrepassingthroughatubefixedonthe topofthebox;themagnetisatthelowerendoftherod,andisshapedlike»horseshoe, ‘Therodpassesdownthroughatubeformingaprolongation of‘thebox,andhavingascrewthreadcutonitslowerend,sothatdampersof ‘coppersndbrass,andofvariousforms,maybescrewedon,enclosingthe magnet,‘Thesuspensiontubocarriesadirectingmagnet;ifthisberemoved, theinstrument becomes simple tangent galvanometer. The galvanometer‘eanborenderedastaticbyplacing«secondhorse.shoemagnetjustundertherirror,ontherodcarryingthelatterandthelowermagnet,Orthewhole‘magnetic systemmaybereplacedbyanordinarymirrorwithsmallmagueta,‘asinThomson'susualform.‘Thereadingsmaybetakeneitherwithaspotoflight oFwith atelescope, ¥. LARROQUE—PERMANENT CHANGES PRODUCED INCOPPER ‘WIRES BY THE PASSAGE OF ACURRENT. (LeLaumitre Bectrigua, Vol.81,P3161, 1889.) Tthasbeennoticedsince«longtimethatthecontinued pamageof‘current ofelectricity inacopper wire tends torender itbrittle, Dufournoticedadecreaseofmorethan80percent,inthetensilestrengthof«copper‘wire which wan traversed byacurrent ineuficient toheat itsensibly, ‘Wertheim thought that hehad found that the decrease inthe tenatle strength was only temporary, anddimppesred when thecurrent was in. torrupted. Edlund and Strvinte speak of«permanent elongation, bat the ‘experiments ofBlondlot have demonstrated that thisisan error. folong asthesmall currents chiefly used intelegraphy wore the only ‘ones tobeconsidered, their effect ontheir conductors was not ofpractical {importance ;batwiththespreadofthesphereofelectricityforlightingand tranmissionofpower,andthemoregeneralusoofheavycurrentaforthese ‘purposes, the permanent effecta produced bythem incopper wires claim attention, ‘Tho breaking ofanoldelectrio light cable ledtheauthortoinvestigate thequestion. A.pieceofwireusedduringtwentyyearstoconveythecarrentto&lighthousecouldbeeaslybrokenupwith«hammer;thebroken surfaces showed clearly that the metal hadagranularstractare,euchasis ‘wenincopperdepositedelectrolytically, althoughwhenfirstpatapthewire had been hard-drawn, Owing tothesection oftheconductor, itprobablyYecamesomewhat heatedbythepassageofthenormalcurrent,thoughtheriseoftemperature couldscarcelyhaveexceeded80°or4°, ABSTRACTS, 205 Itseomsthatthelong-continued passageofacurrentofelectricitytends tobring about «molecular rearrangement, resulting inagranular structare. Morvover, asfarasexperionce goes, itseems tobeimmaterial whether thecurrentisdirectoralternating, astheactionofbothiasimilar. ‘The secondaryofaninductioncoilhadbecomeaobrittlefromlonguse thatthewirecoaldnotbeunwound, butbrokerepeatedly. Theratioofthe‘sistance ofthe old wiretothatofanewwireofsimilarlengthanddiametar ‘wasfoundtobeas21to16.‘From hisown experiments, which were carried out with spirals and ‘igzags ofcopper strip, theauthor coneludes— (1)That theelongations referred toare excessively small, and donot follow any definite law. The lengthenings and shortenings probably remult from want ofhomogeneity, and arenot permanent rerults produced bythe pamageofthecurrent.(2)That thechanges inthetensile strength depend onthestrengthof thecurrent traversing the conductors, bat the relation isnot one ofsimple ‘Proportion ;thepasageofthecarrenthasmoreeffectonbard-drawn copper ‘thanonannealedcopper;alternating currentsseemtoactmorequicklythan iret currents, (8)Acurrent oftwo amperes, expecially when alternating, materiallychangedtheelasticityofacopperwirethreemm.indiameter,afteralapseof fouryears,Itwouldcertainlyseem,therefore,thatconductorscarryingheavy ‘arrentssuchasareusedinelectriclightingarefatedtoundergoakindof lowdisintegration which will affect notonly their tensile strength, butalso ‘heir resistance, Dr. A.vy,WALTENHOFEN—EXPERIMENTS WITH THE ACCUMU- ‘LATORS OF FARBAKY AND SCHENEK, (CoLumidre Bletrique, Vol.31,p.142,1889.) ‘The cells contained each seven positive plates and sixnegative, Rach Plate weighed 21°8 kilos,, and had «surface of877 square cm, ‘The normal‘ateofdischarge was20amperes,butforthepurpotesoftheseteststhecells vwerepushedfarbeyondthis,Inthefitexperimentthedischargewasat therateof100amperes,andlastedfor100minutes, ‘Thetotaldischarge was16587 ampere-hours, oF905-08 watt-hours, The potential difference fell from 147to1-78 volta percell. The cells were then recharged with 100ampere-hour,andagaindischarged withameancurrentof100amperesfornearlyun ‘hour oFatthe rateof4-7amperesperkilo.ofplater,‘Theefficiencyinampere- hoary was 891 percent,, orinwatt-hours 77-4 percent, ‘The decrease of Potential difference was5-22percent.;thedecreaseincurrentduringthelast tenminutesascomparedwiththefirsttenminuteswas2°82percent, ‘These satisfactory rerults induced the author totry the effect of= ischargeattherateof200amperesfor60minutes,or96amperesperkilo.of Plates, ‘The decrease inpotential difference was26percent., thepotential‘idlerencebeingatthebeginning2-26voltspercell,and1-98voltsattheend. 08 ABSTRACTS, ‘Thetotaldischargewas168ampere-hours, 181ofwhichwereobtainedbeforsthepotential diference fell10percent.‘Afinalexperiment wasmadewithadischargerateof270ampares,12:63amperesperkiloofplates.‘Thisgave120amperecbours, with«fallig offofpotential difference of805 percent, ©,MEIM—THE USE OF ACOUMULATORS IN TELEGRAPEY, (Blettrotehnivche ite, Vol, 10,p.4,1889.) [ULprimarybatteriesareused,asiealmottexclusivelythecase,andthere aremanycircuitsthelrnumberbecomesenormous, andthekeopingthemiaorderbecomesveryonerous, Dynamomachineshavebeentried—insomecasesraccesafuly, inothersnot—butthereareseveralobjectionstotheiruse; ‘perhapsthechiefaro,thattheengineanddynamomustbekeptconstantlyrunningdayandnight,ascurrentmaybewantedatanymoment,andthat theplantmustbeduplicatedsoastoprovidearaserveincaeofaccidents,‘Oneofthechiefcausesoraccumalators notbeingmoreusedhasbeenwantofconfidence intheirdurability; butthisobjectionwillfalltothe‘roundasimprovementainthemanufactureleadtotheturningoutofmore andmoreperfectcells,whichcanbecounteduponforsomelengthoftime,‘Thouaeofaccumolators fortelegraphworkpresentsmanyadvantages, ‘TheirEMP, of2volta iscomparatively high, while they have avery mnallinternalresistance; whenatrest,noenergyisspent;theyeaneasilybekept inorder, and require but little attention, andthat atcomparatively long intarvals oftime, ‘Asapracticalexample,thecaseofanofficewithonehundredcircaitemaybetaken,‘Themaximumcurrentonanyonecircuitwillbeatmot0-02ampere,andthehighestE.M.F.200volta,Onebundredcellswillthere-forebewanted,of&capacityof60ampere-hours, andcapableofstanding& normaldischargesttherateof6ampores;theinternalresistanceofthe100 callawillbe1obm,1fallthecircuitsareworkingsimultaneously, thetotalcurrent required will beabout 2amperes; and consequently the EMF, willdropaboat2volts.Inpractice two ruch batteries would beprovided,tobeconnectedaltr- natelytothecircuits,Fromeveryfifthcallwireswouldbetakentothecommutator board,thusgivingapotentialdifferenceof10voltfromterminal toterminal,‘Thetwobatterieswouldbeexchanged bymeansof«simpleswitchstafixedhourdaily,oreverytwodays.Theactaaltimeduringwhich the battory would bedischarging current during aworking day oftwanty-four hourswouldscarcelyexceedsixhours,orthetotaldischargefor thattimewouldbe1?ampere-houra; andasthetotalcapacityis60ampare- hoary,thedropinEMP,wouldbeverysmall,notexceeding9percent, stmost,‘Thechargingcanbeeffectedinfromfourtosixhours.Allthecallswillnot Ihavebeenequallydischarged, owingtothediferentdemandsofthevarious tireaita; therefore the charging would have tobecontinued until allthe callagave offgas. For working circuits with polarised instruments and guch asare operated byalternating currents, abattery of200accumulator cells would‘twwanted;theconnection betweenthe100thand101stcellwouldbeearthed,sndthepositiveandnegativecarrentssentfromeitherend.Localcircuits‘aldalsobesuppliedwithcurrentfromthelessusedsetsofcells,bywhich ‘mansthedischarge would bemore evenly distributed over thewhole battary. ‘The author estimates thefirst cost ofgasengine, dynamo, calls, c., at 12865; and theannual maintenance, allowing forinterest and depreciation, 384200. AprimarycellwithanE.MCF,of1voltandaninternalresistance ofSohmsgivesonetwentieth ofawatt;anaccumulator cellwhendischarged attherateof1amperegives2watis;hence200accumulator cellscould ‘place200x2x20=8,000primaryeells,orallowingforthefactthatthe latterarenotworkeduptotheirfallpowerallthetime,some6,000to6,000, Primary cells 08 LIST OF ARTICLES amare 70 ELECTRICITY AND MAGNETISM ‘Appearing insome oftheprincipal Technical Journals during themonth of FEBRUARY, 1889. -BATTERIES AND ACCUMULATORS. B,Laxowsne—Bichromate Batterieswithout«Diaphragm —Bailtter, wl18,p04, 1889, .‘Avox—Improvement inAmericanAccumulators.—Lumn, HL,vol.81,p.$36 18:9, .HintEffectoftheStrengthofAcidontheCapacityofAccumalators—Bl,Zeit, vol. 10,p8, 1889. Gnawnexa1—Use ofAccumulators inTelography.—Et Zait, vol,10,p.108, 1880. TL—DYNAMOS AND moTORS. B.Guounv—BalciencyofSmallMotorsoftheZarichCo.—Luu,H.,vol.81, p.268, 1880. 2B,Mrriax—Thury's Automatic Regulator.—Lau, Hl, vol 1,p.274, 1869, ‘Axon.—Desrozies’s New Dynamo—Laom. Bl, vol 81,p.278, 1889.2,Matian—Rechaiewaki's NewDynamo—Luu, Bl,vol1,p.801,1880.G.Ricaan—Details ofDynamoConstruction —Lun.Bt,Yol.81,p.34,1880.‘Axox—The Kummer Motor.—Lav. Bt, vol. 31,p.329, 1889, W,Lamcarss—New SystemsofRegulation—Bl,Zeit,vol.10,p.79,1889. Tit—ELNOTROCHEMISTEY AND ELECTRO-METALLUZGY. Bratt and Lussaxs—Occlusion ofHydrogen byNickel.—Belatter, vol. 1, 95,1880, ‘W,Oneiid—Relation between theComposition ofTons and their Velocity of MotioninanElectrolyte—Buibitter,vol.18,p.96,1889. 4.H.vaxPHore and L,'T.Rxtouen—Diswciation Theory ofElectrolytes.— Baibliter, vo. 19)ps98,1889. ‘A.Muuer—introdaotiontotheStudyofElectro-Chemistry.—Lwt. El,vol,81, PP.256,822, 360, 1899. Axox—Production ofHydrogen byElectrolysis forInfation ofBalloons,—Zam,Bl,vol.31,p.277,1889. M.K,Lavunsr—The HermiteBleachingProcess21,Zeit,vol,10,pb1889. IV.-ELEOTRIC LIGHT. ,Dinvpowné—The Biison Central Station atthePalais Royal atPatis.— am, Blvol. Blyp30), 1880, ARTICLES RELATING TO ELECTRICITY, Fre. 200 4.PurSome MoternPhotometric Arrangementa—Lam. Bt,vol3,p.220,1609,‘AvonElectricLightingoftheViennaCourtThestre—Lam, Ht,vol$1,P31, 1899. €iani—Blectric Light andNavigntion—Lum. Bt,vo.31,p878,1860. V—ELEOTRIC POWER.Axo.—TheBentleyKaightSystemofTramways.—Lam. Bl,vol.8,p895,1689,W.Surrmunx—Eficency ofElectricTransmissionofPowerRifai,vl.1 66, 1889, Asex—New Submarine Bosta—Fl, Zit, vol. 10,p75, 1889. ‘VIL-MAGNETISM AND ELECTRO-MAGNETISM. HMiaioes—Combined fect ofTorsion and Longitudinal Stree ontheMaguetiaation ofNickel—PAil.Mop.vol.27,p17,1889. 6.Hoosnaxe—Parmanent Magnetic Orcuits,—Phil Mg, vol 27,p.188, 1880. ?.Joonen—Hotary Magooti Dispersion.—Journ, dsPhy. vol8,p.59,1869, 6.8y,Wras—Eifoct ofMagnetination ontheElectricalResistanceofTron.— Annem, vo. 86,p.447, 1869.2Poauri—Thermo Maguetism.—Bebitter, vo.18,p.100,1889,.Dowex—New TheoryofMaguetination byInduction—Deilatter, vl.18P-101, 1889. Vi.-MEASUREMENTS ANDMEASURING INSTRUMENTS. RBiorbuor and P,Coate—Astatic Blectrometer—Journ. dePhys, vol. 8, 80,1689,&.Doax—Determination oftheOhm(continued)—Anaaln,vol.26,p398,1889.C1. Wanms—Conductivity ofSolid Mercaty.—danaten, vo. 86,p-87, 1689.1SexmainsUseoftheBlectroscope.—Beibliter, vol.19,p67,189. 4.Kunerscio—Rasearches ontheFitnemofPatinam-Iridium Wireandother‘Alloys forConstructing Standards ofRetstance.—Baitltter, vol18,p.89, 1188. ©. Guniamm—ResitancesotsomeAlloysLum.H.,vol.31,p.2141880 PH,Lapnsora—Gome Recent ResearchesonElestrometert,—Lum. B,vol.81, 1p.228, 1889. 8,Lowace—E.M.P,ofAmalgams.—Lum, El.,vol.31,p.293,1889, P-L Lavenoms—Use ofthe Ballistic Galvanometer forMeararument ofthe CoeticientofBelf-Induction—Lum. HL,Yo.31,p.80,1889. ©.RGunusee—Reoent Determinations ofthe‘Ohm.—Lam. EL,vol.Bt, B61, 189, ‘VIKL-—RAILWAY APPLIANCES. B.Gernuss—New Contact-faker forRaila—Bl, Zeit,vol.10,p.71,1850 ‘VoL, Xvi. 16 0 ARTICLES RELATING TO ELECTRIOITY, Bre. XX—STATIC ANDATMOSPEERIO BLECTRICITY. A.Nacoant—Protective Action ofLightning ondactor,—lum, Bi., vol. 31, .339, 1880, A.Riom—Volatilisation of«Fine Wire byStatic Dischargex—Lum. EU., vol 31, 340, 1688. X—TELEGRAFEY AND TELEPHONY. Baum, and Barumesi—Experimenta onTelophonic Ourrente,—Lum. Bt. vol. 31,p.240, 1889,‘Axox—Oables betweenJava,BalijandCelobot—Ht. Zeit,ol.10,p,71,1889.BR,Proe—Improvements inMaltiple Telephone Switch-Boards.—El. Zeit., vol 10,p96, 1880, xI_TEBORY. ‘A,BionEMP.ofSelenium,—Annalen,vol.86,p.484,1889. -G,Avum—Change produced inElectric Action byaConducting Plane,— ‘Bebtter, vol 1,p86, 1889. K.0,Rioursy—Induction inBolid Oonductors.—Beiittr, vol. 18,p.104,1869. 45.Poxs—Unipolar Induction.—Lum.Bl,vol.81,p.294,1889, ‘A.Maccant—Action oftheElectricSparkonElectrified Bodien—TZawm,2,vol.31,p.389,1889, P,H,Laptoorr—Modern TheoriesofElectreity—Lam. HL,ol.81,p.971,1889. X.—VARIOUS APPLIANCES. AxoxElectricTemperature-Regulator.—Lam., EL,vol.81,p.287,1869‘Avox.—Bloctrie Welding.—Bt, Zeit,vol.10,p.74,1889, ov ax Gnstitution ofElectrical Engineers, Founded 1871.Incorporated 1883. Vou.XVIII. 1889. No.79. |‘TheOneHundredandEighty-eighth OrdinaryGeneralMeeting of the Institution was held at the Institution of Civil Engineers, 25,Great George Street, Westminster, onThurs- \ dayevening, March 14th, 1889—Professor W.E.AYRTON, \ F.R.S., Vice-President, intheChair. The minutes ofthe Ordinary General Meeting held on February 28th were read andapproved. The names ofnew candidates for election into the Institution wereannounced andordered tobesuspended. Donations tothe Library were announced ashaving been neeived since thelast meeting from theExecutors ofthe late SirWilliam Siemens; the Director-General ofTelegraphs in India; and SirWilliam Thomson, President ;towhomahearty, voteofthanks wasduly accorded. ‘TheCaamman: The discussion will nowcommence upon theProtesor Paper read tousatthelastmeeting byProfessor G.Forbes on “Some Electric Lighting Central Stations inEurope, and their “Lessons.” Iwillnotdelay youbyanyintroductory remarks, as theimportance ofthesubject isobvious tothemeeting. ‘Mr.Gisszrt Karr: Iwill notdetain you formore than the™r.Karp. tenminutes allowed toeach speaker. Before entering upon the VOL, XVII. 16 212 NOME ELECTRIC LIGHTING CENTRAL STATIONS [March 1, MeKopp. paper itself, Iwould like toaskProfessor Forbesaquestion.In describing theBerlin installation hesays, “They supply 36,000 “lamps of16candle-power, and 144arelamps of15amperes” Later on,when hecomes tothequestion ofcost, hespeaks of feeders inconnection with theglow lamps, and says that the “cost. ofunderground cables hashitherto amounted tosboat «£90,000, halfofthisbeingspentinmainsandhalfinfeeders.” Ishould like toask him whether this sum isinclusive ofthe ‘underground are-light wires ornot. After reading, and again reading this paper, Icould not escape the impression that Professor Forbes hasnotavery high opinion ofcentral station engineering ascarried oninthis country;andhechargesusespecially withhavingneglected to use feeders. Professor Forbes isagreat authority onelectric lighting questions, andtherefore itisimportant toclear ourselves ofwhat appears tobeanunmerited censure. The question of feeders isone which has been considered from thevery earliest days when central station lighting wasthought of, Edison didit from thevery first, and ourfirst steps inEngland were taken mainly bytrying tocopy Edison asclosely aspossible,and everybody sawtheimportance offeeders and adopted them. I may mention that when Iwasemployed byMr.Crompton, some five orsixyears ago, weworked out together astation, the machinery forwhich was tobeerected atVictoria, andthe arrangement offeeders entered into ourplans from thevery first, ‘Wealsodesigned thevariousboxesforfacilityofconnection andtesting. There were, first, feeding boxes ;next, what were called distributing boxes, and athird kind which were called house or service boxes. We not only designed the station with the feeders, but also made provision tokeep thepressure atthe endofthefeeder constant, and bytheassistance ofMr. Willans wegotoutadifferential regulator. ‘TheWillans regulator, instead ofbeing simply shunt-wound, 4sitiswhen you want toregulate the pressure close tothe dynamo, wasprovided with adifferential coil. Bythat arrange- ment there wasinthemain coilofthesolenoid oftheregulator ‘acurrent proportional tothemain current going out,andinthe 1180) INEUROPE, ANDTHEIRLESSONS, 218 shunt coilacurrent intheopposite direction which waspropor- MtKarn tional tothepressure attheterminals ofthedynamos. This arrangement, aseverybody knows, willkeep thepressure constant attheend ofthefeeder, and ithasbeen adopted invarious tations, amongst which Imay mention Leamington ;#0that, eveninthisdetail, English engineers have been atleast asearly inthefield astheir foreign colleagues. Iam sorry that Professor Forbes didnotgive usmore details ofthewayinwhich feeders areworked. Aperson notunderstanding much about it,ora person who wishes tolearn from this paper, would get the impression that allyou have todoistoput plenty offeeders inand allwill bewell. That is,however, not so; itisnot enough tohave feeders, butyou must also have some means by ‘hich youcanregulate thecurrent andpressure inthose feeders. The earliest’ method ofcontrol has been byrheostats inthe feeders, and here isacurious point which hasbeen frequently overlooked. Itisnotsufficient tohavearheostatinthefeederof onepair, but you must have rheostats inboth theoutgoing and relumingfeeder.Ifyouregulateonlyonefeeder,youcankeepthe absolute potential attheendofthat feeder constant, butnotthe difference ofpotential between thetwofeeding points. Inthis caseyoucould only effect anequalisation tohalfofthedifference Which isdue totheresistance ofthefeeders. Togetperfect regulation with rheostats, you must have them inboth feeders, andthey must bemechanically coupled insuch awaythat the switches ofboth rheostats shall actsimultaneously. Ibelieve thisisthesystemwhichhasbeenactuallyadoptedatBradford,There is,however, abetter wayofregulating thepressure atthe feeding points. ‘The insertion of rheostat, inthefeeder isa verycrude device, because ifthecurrent islarge theresistance oftherheostat must beexceedingly small, asitisnotexpedient touse upmore power than absolutely necessary inthe rheostate. Ifthe current issmall, theresistance oftherheostat mustbevery large indeed tomake asensible difference, and forthis reason the Resistance coilsmustcontainagreatdealmorematerial than ‘conespondstotheirnormalwork.‘Thebetterplanistoinsert same counter electro-motive force cells, ashas been tried atSt. 24 SOME ELEOTRIO LIGHTING, Bre. [March 14t, arXapn.Austell, Withcellstheregulation islessgradual, butyouain regulate within one percent. ifyouwork at100volts atthe feeding centre. You can either rise bytwo volts,orlowerby two, sothat youhave amargin oferror ofonepercent.;butthe systemhasthegreatadvantagethatitdoesnotuseupany energyandiseffectiveoveraverylargerangeofcurrent. With \ x —a iF Ps B B analternating current system arheostat hasbeen used—or, atall events, patented—by Mr. Westinghouse,anditissufficientin thatcase toregulate onefeeder only. Butregulations canbe ‘obtained inabetter manner bythe apparatus shown inthe diagram, 1682] DISCUSSION, 218 Imagine that these two bars,B,aretheomnibusbarsinaMs.Kapp station from where allthefeeders start. Onefeeder, F,only is shown, andthisisshown feeding into anetwork ofmains, N,laid along thestreets ofapart ofthetown which may beseveral miles away from thestation, Let ussaythat wework at1,000 volts, Ifwekeep 1,000 volts atalltimes between theomnibus bars,andifthefeeders aresimply joined tothem, wewould have ‘dropinpressure atthefeedingpoint,C,proportional tothe caent supplied. Tomake upforthis drop, wecan raise the pressureattheomnibusbarsbyacorresponding amountwhich could beindicated byadifferential voltmeter, and thus forthis onefeeder wecould getconstancy ofpressure inthelimited area served bythenetwork ofmains connected toit.Butwemay luresome 10or20feeders going outfrom thestation, andeach fiving risetoadifferent drop inpressure. This willespecially be thecase ifthestation supply districts ofdifferent character; xe,letussay, containing residential houses and chambers, other warehouses orbusiness premises, another theatres, and s0 ou, Ifwefeed allthese districts from the same omnibus bars, it illbenext toimpossible toregulate thepressure soastosuit themall. Intheoffice district nolight isrequired after 7o'clock, ‘Whereasintheresidential districtamaximum oflightisrequiredatthattime. Again, intheresidential districts thesupply fallsoff veymuchbetween 8and10p.m.,whereasthetheatresremainfallonuptonearlymidnight. Tosuitthesedifferent casesitis ‘ecessary tobeable toadjust thepressure atthestation endofthe diferent feeders independently,which would either require theuse ofseparate machine foreach feeder, orifthey beworked from omnibus bars, thepressure between which iskept constant, it ‘Would requiretheuseofsomeapparatuswhichwillraisethepressure 4thestationendofeachfeederindependently ofthepressureontheomnibus barsandotherfeeders. Onaccount ofitsgreater simplicity, thelatter plan ispreferable, anditmay becarried out byinserting thesecondary coilofasmalltransformer intooneleadofevery pair offeeders. The size ofthis transformer depends onthemaximum current forwhich thefeeder isintended, andonthemost economical loss inthefeeder. Suppose ata 216 SOME ELECTRIC LIGHTING, Ere. [March 14th, Xc.Kapp. standard pressure of1,000 volts itwould beeconomical tomake ourfeeders ofsuch asizethat atfullcurrent they would lose50 volts. Theregulating transformer must inthiscasebesodesigned that itcanadd 50volts tothepressure inthefeeder when the fullcurrent flows. ‘Thearrangement isshown inthesketch;the primary, P,ofthetransformer ispermanently connected tothe omnibus bare,B;thesecondary,S,containingonlyafewturnsof stout. wire, issubdivided into sections ofwhich oneormore may beinserted bymeans ofaswitch,J.Thisarrangement ispro- vided toenable the attendant atthestation tovary theamount bywhich the pressure inany feeder israised. The secondary ‘may convenientlybedividedintofivesections,each representing 10 volts, sothat thepressure may beadjusted insteps of10volts. This isone per cent. ofthe standard pressure, and thegreatest possible deviation atthefeeding point cantherefore only be4per cent, Itdepends upon theposition oftheswitch byhow much the pressure willberaised bythepassage ofthecurrent through the transformer;andintheonlycaseinwhichIhaveasyet:applied this instrument, theswitch issetbyhand, butIpropose towork itinfuture byanautomatic device which iscontrolled bysome thermo-electric ormagneto-electric relay, R, Imust point out that the switch lever isdivided into two portions, eothat in passing from onetotheother contact theswitch never short- cirenits anysection ofthe secondary coil. ‘The only other point that Ishould like todraw attention to isaquestion upon which, unfortunately, Ifindmyself atvariance with Professor Forbes, and that isthe question offrequency. Hesays that high frequency isbetter because you getasmaller dynamo and asmaller and cheaper transformer. This may be right intheory, butifwelook toactual practice wefind thatit isnotthecase, Professor Forbes, attheBath meeting ofthe British Association, gave apaper ontheWestinghouse system, and hegave astheweight ofa40-light transformer 160Ibs.,or 4Ibs, per lamp. Now, my own S0-light transformer weighs 200 Ibs. :that isalso4Ibs.perlamp;butmyfrequencyis70,and that employed byWestinghouse is133. You seethat, although 180) DISCUSSION, ar theratiooffrequencyisalmost1;2,theweightofthetransformer Mr.Karn. isthesame forboth. The low frequency, Professor Forbes says, isadvantageous forparallel working, because giving time forthe machines topull each other into step. There Icannot agree with him; theinterval istoo short (the 100th or200th part ofa second) forany single pull tohave much effect. The action of palling into step isduetotherepeated tugs orpulls which the lagging armature receives from thecurrent, andmay roughly be expressed asthe product ofstrength ofone individual pull and number ofpulls perunit time. How this product ismade upis immaterial, andtherefore Idonotthink that lowfrequency gives wsanyadvantage forparallelworking. Dr.J.A.Fiza: Iwill follow Mr.Kapp's example, andbebe verybrief intheremarks Imake upon Professor Forbes’s interest~ ingpaper. ‘Thefirst,point Inote isthethree-wire system. Certainly itis ‘matterofsomesurprisethatthoughthissystemwasdevised independently in1883 byDr.Hopkinson inEngland and by ‘Edison inAmerica, and that whilst inAmerica there areabout 120stations worked onthethree-wire system, wehave notasingle onehere. Ihave here asample ofthethree-conductor tubing thatisnow manufactured inthe United States, and used inthose tations, Ithas not byany means been asimple matter to arrive atsuccess inthese conductors, Every one recollects, in theearly days ofthe Paris and Crystal Palace Exhibitions, tamples ofEdison’s tubing: semicircular pieces ofcopper rod weredrawn into iron tubing andfilled with bitumen insulation. Atfirstthepractice wastosimply separate thecopper strips with puteboard distance pieces, and thewhole putinto cast-iron gas Piping, into which theinsulation waspoured. The difficultiesthatthenoccurredintheuseofthoseconductors werethese:the copper rods very often were strained, andthebituminous insula tionwassqueezed outfrombetween them,andthepasteboard distance pieces didnotkeep theconductors apart satisfactorily. More than that, theiron gas-pipe tubing was notsufficiently strong towithstand hard usage, and especially notstrong enough towithstand theactive attacks ofthepickaxes ofroad repairers, 218 SOME ELECTRIC LIGHTING, Bre. [March 14h, Praing, Gnditgotdamaged, andtrouble ensued. Butnowtheprocess is somewhat asfollows:—The copper rods arerespectively spun over throughout thewhole oftheir length with drystring toseparate them;thecopperrodsarethenplacedcentrally intheirproper position intheinterior ofstout steel tubing; avacuumismadein thetube, and theinsulating material isdrawn in,and itpene trates into every portion. Adifficulty which appears intheuse ofanextended system ofthis kind istheresult ofthe stresses that arebrought tobear ondifferent parts oftheconductor net- ‘work buried intheroadway, owing tothedifferent pressures that come upon theroadways atdifferent times, causing severance of theconnection with thejunction boxes. Inorder torender the network ofconductors more flexible, andinsome way orother to make itgive and take under theroadway, asystem ofball-and- socket joints wasdevised bywhich theends ofthetubing were connected tothejunction boxes byakind offlexible junction, perfectly watertight, but, nevertheless, allowing acertain amount ofelasticity inthenetwork, which could thereby accommodate itself toslight changes ofthelevel intheroadway. With regard tothesubject offeeders. AsMr. Kapp has said, thefeeder system isvery oldindeed. Edison took outhis first patent forfeeders intheearly part. of1880, anditwasat ‘once applied inthefirst district station laid down inNew York.- Iknow that Edison attached very great importance tothat, even intheearly days. With regard totheuseofthese feeders, and themanner inwhich they were used tocontrol thepotential in thedifferent parts ofthenetwork, in1883, intheNew York first district station, the method employed was asfollows:—A pair ofsmall wires was run back from the ends ofthe feeders, where the feeders dipped into the network, tothe central station, toenable them toatalltimes test thepotential; and ifthey found that itwas toohigh, afeeder leading tothat Point was disconnected, and they were thus enabled tokeep the potential constant. Iam not aware what atthe present timeisthemethodadopted inthiscentralstation,butperhaps ‘weshall obtain information onthepoint inthecourse ofthis discussion, 1898] DISCUSSION, no With regard tothevery important question oftheperPais, manence oftheinsulation, Ithink there can benodoubt that whilst our experience has been extensive onthe insulation of continuous currents, there isagreat deal ofexperience which villhavetobebought expensively upon theinsulation employed indealing with high potentials on'thealternating system. When. vweconsiderthatelectricexcitedenergyisconveyed, notalonga conductor atall,but along adielectric, Ithink itbecomes avery important question todiscover whether theinsulators areofa ass that will stand reversed electrical stresses many times a second foryears; and ifthedielectrics now used will notstand that,thenIthinkthesoonerourattention isturnedtoothers thebetter. With regard tothepoint raised byProfessor Forbes onthe ‘atte oftheprimary current running through theconverters. I think hehas mentioned the case ofthe Grosvenor Gallery installation, where during the hours ofminimum supply the ‘amentsentoutfromthestationissaidtobeabnormally high; batitwould benecessary tofind out whether that current nesurement iscorrect. Ofcourse, ifitisthefact that itis60, thenwhatitindicates, Ithink,isthattheconverters arestarveddfiron(Professor Forszs: “Hear,hear”},andaneconomy has beenperpetrated inthewrong direction. Tthinkthosearethechiefpointsuponwhichitoccurstome toremark. Iwould saythat Ihave read thepaper with great interest,andIamsurethereisalargeamountofinformation thered together init,which willhelp allwho areendeavouring ‘ouolve this problem ofthedistribution ofelectricity. Mr,J.Swoveunve: ThefirstthingthatstrikesoneinProfessor x,Forbes's veryableandinteresting paperisthatalargeproportion ""“"™* cfhisconclusions will notbeapplicable toEnglish electric lighting. Abroad, distribution byalternating currents iscarried catwith atransformerineveryhouse.Inthiscountryitismuch orelikely that alarge transformerwillbeusedforeachblogkof buildings. Each will then have itsown leads from the station,60 ‘tatthepressureonthelampsofeachblockcanbemaintained cmstant, We start, unfortunately, late intheday, but, fortu- Iutely, with more modern ideas. 20 SOME ELECTRIC LIGHTING, Ere. [March 14th, Myame Professor Forbes over-estimates thedifficulties incoupling direct-current machines inparallel. Allthat isneeded, when throwing adynamo into circuit, istoregulate either itsspeed or itsexcitation tillitgives theright, ornearly theright, pressure, and then toswitch itin, ‘The engine isregulated forload, and the excitation forspeed. Asashunt, orseparately excited, dynamo runs inthe same direction when amotor, there isno accident, even ifthe pressure isalittle toohigh ortoolow,on switching on.Thecomplications described inthepaperarereally unnecessary. The difference inGerman design oflarge machines may perhaps bedue tolabour being cheaper there inproportion to material. The fewer poles thebetter inadynamo. Alarge machine cannot bemade towork with less than acertain number, butitisbesttokeepthisnumberassmallaspossible. Itis satisfactory tofindProfessor Forbes against slow-speed dynamos. Thave always advocated amoderately high speed, asitmakes « machine more efficient, less liable tobreak down, easier to construct, andcheaper. Compounding governors and voltmeters forloss inleads is quite well known inthiscountry, andIthink originated here. ‘We cannot assume broad general rules forthedesign or efficiency oftransformers. They have tobeworked outspecially foreach alternation frequency. Atransformer, curiously enough, isnotlike adynamo, inthat you canapproximately calculate the absolutely best form. The equations arevery long and tedious to work out, butarenotdifficult. The transformers atpresent in themarket are, Ibelieve, designed atrandom: askeleton is taken and aninduction assumed, and thenecessary wire wound on. There isthus room forconsiderable improvement inthe design. The main question indesign isgenerally overlooked— that is,nottheefficiency ofthe transformer as converter, but thevariation ofsecondary pressurewithvaryingload,andconstant primary pressure. The 88percent. efficiency transformers men- tioned inthepaper would cause anenormous variation inthe efficiency ofthelamps onthesecondaries, anditislight that people payfor. This fault cannot becorrected byany systems of 1882) DISCUSSION, an feeders orofpilot voltmeter leads, ifseveral transformers areinM*.... parallel‘on each circuit. Thetransformers insuch cases ought to bespecially designed forasnearly constant secondary pressure as practicable. Inotice aslip where Professor Forbes says that some trans- formers would bemore efficient iftheiron circuit were open. As Ifellinto thesame error myself, Imay point outthat thewaste energy isftGH,andisindependentofthesourceofthe magnetic force; thedemagnetising action oftheends does not lewen theenergy wasted. Professor Forbes seems tometoexaggerate thedifficulties in running large alternators inparallel. Iexplained asimple “induction coupler ”tothisSociety afewmonths ago. ‘Though Imuch prefer small frequency ofalternation, asit takes themachine easier tomake, better, and cheaper, Idonot weewhy,otherthingsbeingequal,alowfrequencymakesiteasier tocouple machines inparallel. Itseems tobeaquestion ofthe ationofself-induction tooutput,&c.‘Thequestionofsettling upon auniform alternation frequency inthis country isoneof aormous importance, butitistooearly todiscuss it,aswith our Present knowledge wemight easily tieourselves down toawrong frequeney. Mr, F,V.ANDERSEN: Allow metomake afew remarks ontheus general question oflow-potential distribution. Professor Forbes “"*"™ saysthat hedoes notthink that itpays inEngland tousethis system formore than adistrict with aradius of400 yards. Ifthisisthecase, then thesystem offeeders needs nottobe verycomplicated, oritwill notbenecessary togotofeeders with very different, lengths, ortohave rheostats insome ofthem. 200yardswillbeaboutthemaximumdistanceitwillbenecessary tocarry theappliances fordistribution into thehouses. Suppose ‘Yehaveadistrictwhichhasaradiusof400yards;themains— hich arethetwo strong black lines inMr.Kapp’s diagram— maybeInidintheformofaringwhichsurroundsthestation, Thestationwillveryoftencomewithinthedistrict,andsome-timesnearthecentreofthedistrict. Inthatcase,ifasetof aaa SOMEELECTRIC LIGHTING, Ere,(MarchMth, Yeon, Taine arerunatadistance from thestation ofabout 200 yards, with branches like those shown onthe diagram, into the streets, none ofthese branches will bemuch more than 200 yards long, and inthat case acurrent-density of600 amperes tothe square inch will, onthe three-wire system, only give 3volts fall of potential; thefeeders may have the same length, and allthe dynamos berun inparallel. This system ofusing allthe machines inparallel is,Iunderstand, theone adopted inBerlin, andonecanvery readily understand the great advantages that, they have found tolieinthis simple form ofdistribution. Of course itonly holds foralimited area, Tam surprised tosee that Professor Forbes against this system puts hisown idea ofhaving dynamos invarious groups, and, sometimes, when the load issmall,torunafewdynamos and have allthe feeders onthe one ortwo dynamos which are running, andchange them into more groups astheload increases; because itisacomplication, andattimes, such asonfoggy days, itwillbevery troublesome tochange from onesystem toanother. Professor Forbes mentions 1,000 amperes asthe efficient density ofcurrent, butIthink that itisnotvery difficult now towork outwhat thedensity should beaccording toSirWilliam Thomson's law. For Berlin load diagrams have been worked out, and pub- lished inexcellent papers, forallhours ofthedayalltheyear round, anditisquite possible from these tosolve thequestion of what should bethedensity inasystem like this, Ithink that, with thepresent price ofcopper and coal, and therate atwhich electricity canbeproduced, something like 600 to700amperes will befound tobeaneconomical density. ‘Astothequestion ofpilot wires, Professor Forbes indicates that they aregenerally attached inthewrong place, and that a great improvement canbeobtained byputting them intheright place, which hesays ismidway between thefeeding centres and thefurthestlamp.Idonotthinkthatthiscanberight.From theonefeeding centre onthemains totheother weshall have a very small fallofpotential indeed, and therefore shifting the pilot wires along themains hashardly anyeffect atall. But to gotothecentre between thefurthest lamp and the feeding 180) DISCUSSION. 333 centre isadifferent thing, because itwould mean toconnect them toa branch, andthis would notlead toaproper regulation ofthedistrict. Theywillhavetogotothepointscarryingthe marimam load, i.,topoints near thefeeding centres. There aresome questions astothe average voltage. We findinthepaper that “4volts may beallowed upanddown,” and Tcannot understand this inanyother way than that average voltages aresupposed tobeused; for,ifyoucanallow 4volts fall,youcertainly cannot allow 4volts inonedirection, Now Dr.Fleming said hethought there wasonly oneproper solution tothequestion, andthat wastousethesame voltage all through. Idiffer entirely from that view. Asarule, when we make aninstallation with acompound dynamo inahouse, we compound themachine sothatitshall give avoltmore when the fallloadisonthan itgives when there isaminimum load. That gives anideal regulation insuchacase;butincaseofdistribution fromacentral station, if«constant potential iskeptonthemains (whichisthebestwecando),thenregulation must,beobtainedin some other way; and thebest result will beobtained byusing average voltages. Itmay bequite possible toallow,atleastatfull load, 4volts fall into the branches. But that installation which isclose onthe main hasonly avery small variation, say°5volt, hich isduetothefallinsidethehouse,whileinthebranches there willbeafallupto4volts; therefore, ifitberight togive 100-rolt lamps tothehouses near themain, then weshould give lampe for98volts tothefurthest houses onthebranch, ifthefall from the maximum density amounts to4volts. The fall of Potential depends upon distance and density ofcurrent, and tothing else, andallbranches when laid down aredesigned to onlygouptoacertain density ofcurrent with thefullload;and sinceneither thisnorthedistance canchange, wecantellexactly thefallofpotentials with which wehave todeal. ‘There can thereforebenoobjection tovarythevoltsbyoneortwo.Itis dificult toseehowProfessor Forbes canreport that itleads to dimstrous confusion, since itisonly aquestion ofavoltortwo, anditcertainly isastep towards perfection touse average Potentials, Ofcoursethevoltagecannotvary,asProfessor Forbes oy SOME ELECTRIO LIGHTING, Br. [March 14h, Xs,gq,8878,inthewaythathousesrequiring 100voltsonedayshould require 99thenext, This canonly happen ifthehouses are moved upordown along thebranches.Xinem «Mr.ALEXANDER SteMENS: Myremarkswillrefertooneonly ofthevarions subjects that Professor Forbes brought before us insuch avery interesting and admirable manner atthelast meeting—that is,totheinsulationoftheBerlincables.‘These cables consist ofacopper conductor surrounded byjute, which is impregnated with acompound andthen enclosed inalead tube. ‘The lead tube isfurther protected byacovering ofjute and compound, and then byaniron armature,toscreenitagainst mechanical injury; andthat, again, iswrapped round with jute and compound toprotect itagainst rusting. Professor Forbes tolduslastmeetingthatcablesconstructed inthismannerwillcertainly break down after about three years’ use, and the explanation given himinBerlin was that agalvanic action was setupbetween theiron armour ofthecables andthelead. Now this subject isvery important, because those cables were designed tomeet agreat want forunderground conductors, You are,ofcourse, allaware that gutta-percha and india-rubber aretheprincipal insulators forcable work; butgutta-percha for electric light wires isnotatallsuitable, because, whatever you may do,theelectric light wires areoccasionally over-heated, and that would make the copper conductor sink through the gutta- percha and touch the lead;s0thatgutta-percha, whichwould otherwise bethe best, isout ofthe question, India-rubber is rather expensive, and there isatleast alittle doubt whether it willremainpermanent. Therefore agoodmanyexperiments were made, and the insulation just described was settled upon as being themost suitable forthese sort ofwires. Knowing that Professor Forbes hadbeen notlong ago inBerlin, and that he obtained hisintroduction totheAllgemeine Elektricitits Gesell- schaft through Dr.Werner v.Siemens, Ithought, when Iheard thisunfavourable account, that ithad been verified byreferring toMessrs. Siemens &Halske, but Ifind that such was notthe case. The reasons why Professor Forbes was supplied with inaccurate dataIdonotknow,butitiscertainthatthisgeneral 1880} DISCUSSION, 235 breakdown hasnottaken place. Thehistory ofthese cables isyr. that in1882 about 7kilometres ofthem were laid for the are light circuits inthe Leipziger Strasse. These cables were not armoured with iron at first. In 1885 the installation of the central station began, and atthe same time about half ofthese unarmoured cables were exchanged foriron-armoured cables; and in1886, 1887, and 1888 more cables were laid. 142 kilometres ofthesecablesareatthepresenttimeunderground inBerlin, anduptoAugust, 1888, nointerruption whatever occurred in thisextended cable system. About thistime thethree separate stations shown onthe diagram were united onaplan which hadbeen designed bytheAllgemeine Elektricitits Gesellschaft, against the advice ofMessrs. Siemens &Halske, because a testing ofthe cable system was made highly inconvenient— ‘onecould almost sayimpossible. Then, between August, 1888, andthebeginning ofNovember ofthat year, four different places inthe system were found faulty. Altogether thirty- fivecables are passing these four places, and ofthese only eighthadtobepartlyexchanged inordertorepairthe damage done. One oftheeight was pierced byapickaxe; another was making earth byone ofthe screws inthe coupling being badly fitted inand touching theiron ofthejointer; a third became also faulty through careless fitting; and inthe other cases the cables were sobadly burned that itwas not Possibletofindoutafterwards whatwasreallythefirstcauseof thefault, Messrs. Siemens &Halske arestrongly ofopinion thatineachcasethemischief wasstartedbysomemechanicalinjury tothecable, and notagalvanic action assuggested to Professor Forbes. Another circumstance isthat itwas not the cables which were laidin1885 which went wrong, butthose which ‘ere laid in1887 and 1888, ‘This shows that the statements were made toProfessor Forbes rather recklessly, and Iregret his having puttousinsuch avery definite manner that thelifeof thesecableswasonlythreeyears,whennoneofthethree-year-oldcables hadbecome faulty. ‘This corrosion ofthe armour covering which has been owerred, and which hasbeen shown toProfessor Forbes, and 216 SOMEELECTRIC LIGHTING, Ere,[March140, Seem 2180toMr.Crompton, Ithink, basonlytaken place incables which were electrically atfault; the other cables which were lying immediately alongside ofthe damaged cables were not touched stall,although they were exactly under the same conditions. Then another matter which Professor Forbes seems tohave overlooked isthat the mere juxtaposition ofiron and lead, with jute between, would notgive risetoanycurrent atall. Itcanonlygiverisetoacurrent ifthereismetallic connection elsewhere between theiron and thelead; and even ifthis, by someaccident, hadbeeneffected,thentheironwouldbeeateninto andcorroded, butnotthelead. There was, Ibelieve, agood deal offriction between theAllgemeine Elektricitits Gesellschaft and Messrs. Siemens &Halske about these faults being dis- covered; andasamatter ofprecaution agood many other places— Ithinkaboutfortyplaces—of thecablesystemwereopenedto ascertain thecondition ofthecables, anelectrical test being impossible onaccount ofthepeculiar arrangement ofthewhole system. Inallthose cases thecables were found tobeinperfect condition, only inseveral places itwas found that the cables had been heated tosuch anextent that theexternal compound of asphalte hadbeen melted together with thesurrounding earth, without injury tothe insulation:thatisaboutthehardesttrial which theinsulation ofacable could besubjected to;anditis aleo afact that these cables have, since November, 1888, given notrouble whatever after thefaults were cutoutbyreplacing about 250 yards outof144,000 yards, Onreferring toMr. Rathenau—the same gentleman who gave Professor Forbes that ‘unfavourable opinion—he confirmed that thecable system isin perfect order atthepresent date, that there isnothing the matter with itwhatever, andthat thewhole trouble was simply inthose four places ofwhich Ihave spoken. Ihave thought itright togosomuch into detail about this point because areliable insulation forunderground electric light wires isvery much desired, and itwould beapity ifthis insulation were condemned onsuch very slender grounds. Inaddition, Iwill only remind you that Professor Forbes bas told ushimself that inMilan and inRome these cables have 1.) DISCUSSION. on givenperfectsatisfaction, andthatinRomeconcentric cablesofur.thisconstruction have been inusewith 5,000 volts ofalternating currents fortwo years. Sothat Ireally think there must have been some other reasons than real failure which have prompted theadverse opinion which Professor Forbes has brought over toos. TheCuamman: AsMr.Shoolbred isinterested intheProtesor Bradford installation, perhaps hecan telluswhy thethree-wire system, which issocommon inthe United States, was not ‘sed there ? Mr.J.N.Sooousrep: Amongst themany andimportant uy ints ofthis interesting andinstructive paper theonethat parti~ calarly strikes myattention isthat which hasalready been dwelt upon atsome length byMr. Alexander Siemens; viz., thewhole- ‘alecondemnation oflead-covered cables forunderground work hich iscontained init,Professor Forbes uses strong language Withregard tolead-covered cables, and condemns notonly those used atBerlin, but also those ofother makers, and which are lugely used inAmerica and inEurope. Imyself would be dif hewould, inhisreply, give ussome evidence, andfurther information onthepoint. Probably during thediscussion some- thing may beforthcoming from the makers. Indeed, while making sosweeping acharge, and one which ismost impor- tantinthepresent stage ofelectric lighting, Icertainly think thatProfessor Forbes should have adduced some reasons, andalso smefacts, when stating that heregretted “to have totellus “that this insulation has been afailure.” With regard tothequestion which theChairman asked just towabout the Bradford installation. Ihave carefully avoided anyallusion toitbecause Ithink itishardly fair, either inthat rinanyother installation which isinprocess ofconstruction, to referspecially toituntil after ithasbeen completed. Respecting theinquiry,astowhythethree-wiresystemwasnot ‘opted atBradford. ‘Theimpression generally seems tobethat theBradford installation isintended tobeonly alow-pressure one, Itisultimately intended tobeahigh-pressure installation, tobeextended tolongdistances. ‘Theportion oftheinstallation ‘VOL, XVII. Ww 28 SOMEELECTRIC LIGHTING, Ere.(March1h, Muga BOWUnder construction issmall inproportion towhatmaybe ultimately carried out, Besides, inthepresent stage ofelectrical information itseems pretty generally conceded that, forhigh pressure, probably atwo-wire system would bemore convenient than athree-wire one. The merits ofthethree-wire system were very carefully con- sidered, notmerely bymyself, but also bytheElectricity Supply Committee oftheCorporation. But, after personal consultation with Dr.Hopkinson, they decided not toadopt thethree-wire system—at present atleast—as any economy inthe cables, arising from itsadoption atthat early stage, would have been more than counterbalanced bythe cost oftheincreased number ofsmaller dynamos which their series arrangement, inpairs, would have entailed. Furthermore, thethree-wire arrangement inthestreets might lead tocomplications inpremises where theproprietors preferred thetwo-wire system. Again, thefacility afforded bythethree- wire system fordoubling the E.M.F. (which hasalready been referred toduring this discussion, and asanadvantage) would, most probably, have ledtodifficulties with the Board ofTrade, owing tothelimiting E.MLF. forhouses having thereby been exceeded. ‘There isnothing, however, intheBradford installation,asat present being laid down, toprevent, ifthought advisable, the adoption ofthethree-wire system atafuture time. Yen, -MvManxRoutNsow: Inshowingustheexcellent workwhichhasbeendoneabroad,Professor Forbeswascarefultoexpress& beliefthatEnglishmen wouldinduetimedoworknolesssuccessful. Ithinkheisright;andinonesmallbranchofthe subject, onwhich alone Ifeelable tosayanything, Icangive some reasons forendorsing hishopes. Professor Forbes laid stress, very justly, upon the question ofeconomy inthemotor. Whatever may besaid about the first cost ofapparatus, and about thecost ofworking, inthelong runthemain point to keep inview inelectric lighting isthecoal bill; themore elec- tricity canbegotoutofatonofcoal, thelarger, ifother things areequal, thedividend willbe. Special reference hasbeen made 180] DISCUSSION. 29 tothevery fine 400-H.P. Corliss condensing engine inthe central station at;Berlin, and ithasbeen mentioned that itgave ‘anelectrical horse-power for15°5 Ibs.ofwater evaporated. When thepaper was read Iwasunder theimpression that Professor Forbes spoke ofindicated horse-power; andas15Ibs.perLH.P. isconsidered goodworkforalargecompound Corlissengineinthiscountry, themistake wasnotunnatural. But itappears that 15Ibs.was given toProfessor Forbes astheconsumption per dectrical horse-power—a marvellous result, because itappears to Pointeithertoaconsumption delow13lbs.perILH.P.intheengine,orelsetoaquitephenomenal efficiency inthedynamo. Onecannot butwish fordetailed figures ofsuch tests asthese, andforfuller particulars generally. Certainly wecannot match such figures in England yet;butfigures canbegiven, astowhich nodoubt exists, which areatleast ofhappy augury forourprogress inthefuture. Thave seen anEnglish condensing engine giving anindicated horse-power upon 15'1 Ibs.ofwater. Like theBerlin engine, itwas direct driving;anditwascoupledtoaCromptondynamo.The differences were these: itwasnotalarge engine, indicating 400 horse-power, butavery small engine, indicating only 40horse- power;itwasnotrunningatonly80revolutions, butat4005itwas tot,liketheBerlin engine, designed asacondensing engine, but expressly fornon-condensing work. Had ithad thesame advan- tages ofsize and ofspecial adaptation, itseems not anunreason- ablebelief that itsconsumption ofwater might have rivalied that oftheBerlin engine. Considering howmuch isgained inpoint offirstcostbythemuchhigherspeedoftheEnglishengine,it willprobably beconceded that, even asitstands, itsrecord isa good onetoshow forEnglish work; and asEnglish engineers engaged inother branches ofelectrical work, such asdynamo- making, have nodoubt, relatively quite asgood results toshow, itisclear that Professor Forbes's hopes onour behalf arewell founded. Mr.W.LaxCanrzwrer: Therearethreepoints,Sir,that,mctan ‘withyour permission, Ishould liketorefer toinconnection with thisinteresting andvaluable paper. ‘The points are—Feeders, thethree-wire system, andinsulation. | 310 SOMEELECTRIC LIGHTING, Br.[March14th, uslat -‘T-certainly felt, inhearing Professor Forbes's paper, and after readingit,thathewastosomeextentfloggingadeadhorse,or killing theslain, Itsohappens that within ayear ortwo ithas been mylottohear details oftheplans ofabout twenty central lighting stations inEngland, and ineach ofthem feeders have been provided for. Iwas inBradford some months ago, and sawthem being laid there inthemunicipal installation, andI know people who have laid noless than 200 feeding mains to separate installations intown and country indifferent parts of England. ‘With regard tothethree-wire system, Ishould likejusttopoint outthat thedoubling ofthe pressure ofwhich Professor Forbes speaks isnot exactly the chief advantage, but itisonly one of several advantages;andIamverysorrythatDr.Hopkinsonis nothere,sothatwemighthavehadtheadvantage ofalittle further exposition ofit.But byfeeding onthis system,asis probably well known, one outside wire comes from the positive pole ofonedynamo, and another outside wire comes from the negative pole ofthe second, and when these two areproperly balanced the current inthe third orneutral wire isnil; and hence, there being nocurrent, there isnoresistance toallow for inthereturn wire, and hence there isagreat saving; andI *venture toaskProfessor Forbes whether, inhispaper, where he isspeaking ofthedrop of480yards with 1,000 amperes, thedrop should not be6volts instead of12. Professor Forbes: No} the12comes inbythethird wire. Mr.W.Lanr Canrenten: Iamopen todiscuss itprivately, but Iadhere tomyview ofthe case.* Iwould just point out that,sofarasIamaware,inpracticethebestsizefortheneutral wire isone-third, andnotone-fourth, that oftheactive wire. ‘Then with regard towhat Professor Forbes has said about bituminous compounds. Ithink weought todistinguish be- tween the coal tarproducts ofvarious kinds which are often called bitumen, orbituminous, and that bitumen which Dr. Fleming hasalludedtoashavingbeenusedbyEdison,andwhich *ProfemorForbesultimately admittedthecorrectnea ofmyview—W.L.C. 1889.) DISCUSSION. 231 : is,onemaysay,produced inNature's laboratory, ieisanatural ast, product. think that itistothecoal tarbituminous products that failures aregenerally due, sofarashasbeen known. I venture toassert that with real bitumen no such effects will be got,andIcouldmentioncablesthathavebeenthusinsulated which have been laidforfarmore than three years without any signofinjurytotheinsulation. Therearealargenumberof stations which Icould mention, both inthis country and in theUnited States, which have been solaid, only thelistisso Jong that Ishould not like toname them, because weareso thortoftime this evening. With regard tothequestion ofvulcanised rubber cables,—if ‘uchcables arecovered with lead, thelead will beattacked by thesulphur asreadily asuntinned copper would be;sothat if tulcanised rubber istobeused, then alarge quantity oftinmust, beadded tothe lead, and that will make the cables hard to manipulate inthelaying. Mr.RE,Cnourtox: Whileagreeing withProfessor Forbes 4%tothe vast amount tobelearned bythe study offoreign central stations, Imust point outthat hehas, Ithink, inhisown Peron shown usthat there aredisadvantages attaching tothis odeofstudy,viz.,thatitisextremely difficultforanytravelling satsider togetattherealfacts when conflicting interests areat work. Inthis particular case ofthebehaviour ofthelead- covered cables atBerlin, nodoubt theofficials attheAllgemeine Hiektricitits Gesellschaft, put thecase before Professor Forbes verystrongly from their own point ofview. ‘They didsotome also;but,after what Mr.Siemens hasrecently said,itisprobable thatthere isanother side tothestory. Ifirst wish toask Professor Forbes ifhehasnotmade anerror indescribing the output ofthe Berlin central stations. Hegives itasifthe majority oftheoutput was inincandescent lamps;whereasIwas told, and from my own personal observation itseemed tobe correct, that the output from the low-tension mains was very lagely inarclamps :possibly two-thirds ofthe whole output was thus taken. Icannot speak toohighly ofthe ‘Wonderful engines made byVan den Kirchove, thedrawings of 288 SOMEELECTRO LIGHTING, Ere. (March14th, Mane, hich arebefore you. Theyaremostperfect specimens ofwork manship and design, Mr. Van den Kirchove has solved the extremely difficult problem ofmaking engines ofthis class run with perfect quietness and freedom from shock atthe time the connecting rodpasses over thelower centre. They work per- fectly noiselessly, thesole exception being theslight click of theCorliss valves, Asan English engineer Imust confess that ourmanufacturers ofsteam engines cannot teach Mr.Van den Kirehove anything inthis respect. ‘The previous speaker, Mr.Robinson, has, however, pointed outthat these engines are very bigand very costly forthe work they have todo; and itis quitepossible thattheuseofsuchlargeandcostlyengines will notbeasprofitable tothe company asifthey had used smaller, quicker running engines ofthesame power, but ofthe type we useinEngland, Speaking generally, thewhole ofthemachinery atBerlin iscarried outonamost magnificent scale; butIhardly think that, ifwewish togetthe best economic results, weshall copy thismachinery inEngland. The economy ofthese engines, viz., 154Ibs.ofwater perE.H.P., appears tobeextremely high. Such economy isjust possible, but Ishould like toknow inwhat manner the tests were carried out. Similar tests carried out at Vienna showed that recently wehave been obtaining the result of22Ibs.ofwater per E.H.P., but inthis case the tests were carried out byindependent authorities. Before Ileave the question ofthe Berlin and other central stations,Imustask Professor Forbes whether hethinks itisreally good modem practice tousebanks oflamps orother artificial resistance when low-tension dynamos aretobebrought into parallel working. I have eight large sets ofengines and dynamos running parallel atVienna, and from the commencement wehave not had the least trouble inthrowing them into parallel working without the useofany such devices. Mr.Melhuish, theresident engineer ofthat station, bas inone ofthe technical papers shown how easily and conveniently anadditional machine isthrown into circuit bymeans ofpilot lamps, which show when the E.M.F. of thefresh machine comes tobeequal tothat oftheline. At this moment the circuit isclosed, and the machine takes its share oftheload without anyflicker inthelights. 1689} DISCUSSION. 238 Tnow come tothethree-wire system. Ifully agree with what Professor Forbes hassaid about it;butIwish topoint outthat a great many engineers, including myown firm, have often been obliged tocontinue tousethetwo-wire system, although they know perfectly well that itwould beeconomically advantageous tochange tothethree-wire one. Several stations that Iknow of veredesigned forthethree-wire system originally, butsolong as thedemand wassmall andirregular itwasfound very much easier towork the station ononeparallel only, theintention beingto make thechange asthestation grew larger. Insome cases, however,thedemand hascomeonsosuddenly thatithasnot° been found convenient tomake the alteration atonce. The suthor has referred toSir William Thomson's formula forthe calculation ofthecopper inelectric mains asbeing constantly in ‘we, Imust point outthat, with every desire tomake useofthis formula, ithasbeen extremely difficult todosouptothepresent time,asitwasnexttoimpossible toobtainthetimefactorneces- sarytowork itoutcommercially. Forthefirst time Ihave got ‘time factor calculated from the average number ofhours the feeder maing areworked atcertain currents, butuntil Iobtained thistime factor itwasuseless toattempt toapply thelaw. On thispoint Icould have learned very little from thestudy of foreign stations. The habits ofthedwellers inContinental towns are sodifferent from our own that the time factors for Berlin, Vienna, orParis would befound todiffer widely from those useful tousinLondon. Inthose towns people mostly dine oat, Alarge portion ofthelamps areused inrestaurants or places ofpublic entertainment, sothat the hours ofmaximum lighting arespread over aperiod much longer than isthecase in England. AtBerlin, forinstance, inDecembertheloadonlyvaries slightly from 6to10p.m.; whereas atKensington itwould reach amaximum at7,and onehour later would bereduced nearly one- lalf. The maximum really corresponds with the time when everybody isdressing fordinner andtheshops still remain open. Asregards thepoint atwhich pilot wires should becoupled onto network,ImustagreewithMr.AndersenthatIcannotseehow ProfessorForbesisgoingtocoupleonthesepilotwiressoasto 34 SOMEELECTRIC LIGHTING, Br.[March14, Ysapin, obtain theendthatheseeks. ‘Thedesirable object thatheseeks toobtain hasbeen before usall,butitapparently cannot be applied except inthecase ofdistributing mains which arevery completely coupled upinto anetwork. Ifweimagine that the maintenance ofthe required difference ofpotential allover network wassomething likethat ofmaintaining afilmoffluid of ‘agiven thickness allover afiat porous absorbing surface, thesupply being kept upbymeans offeeder tubes spaced evenly over the surface, we should then find the film decreased in thickness towards that part ofthe surface which ismidway between the feeders, and where theabsorption ismost consider- able. Itisevident that where Professor Forbes would wish to join onhispilot wires would beatthepoints midway between the feeders andthethinnest part ofthefilm ;butthis appears tobe impossible, except invery exceptional cireumstances, partly owing tothedifficulty ofthoroughly connecting upandcompleting the network, andpartly owing totheuncertainty ofknowing atwhich points thedemand will beheaviest. Inowcometotheuseoftheairasadielectric, Professor Forbes begs that Iwould state authoritatively what isthemaxi- mum and minimum resistance of the air-insulated mains at Kensington, because, hesaid, nothing hadbeen stated definitely about it.This istrue. ‘The air-insulated mains atKensington are somixed with the continuously insulated cables that we have great difficulty intesting any considerable lengths ofthe formerbythemselves. Wehavefourorfivemilesofthesemainsat Kensington, butowing tothefactthat wenever have thecurrent offnight and day, ithas been found impossible tocarry out delicate quantitative tests. We doknow that the insulation is very high—far higher inthecase oftheair-insulated mains than inthecase ofthecontinuously insulated ones. Professor Adams, atmyrequest, made anattempt lastweek totestashort length ofmain which isonly partly finished. ‘This section wasrecently laid, and inallprobability itsinsulation wasatitslowest, asthe weatherwasdampanditwasnotinuse,sothatthesurfaceoftheinsulators couldnotbedriedbyelectrolysis, which,ofcourse,is thenormalcondition ofaworkingmain.Theinsulation resist- 180) DISCUSSION. 235 anceofalength (including onecrossing made upofcontinuously Mt. insulated cables drawn intubes) wasattherateof380,000 ohms permile, Although thisresult isfairly good, yetIdonotthink thatitforonemoment represents theinsulation resistance ofthe baremains,butratherthatofthefourjointsontothecontinuous cables attheextremities ofthecrossing. Asmost ofyoupresent ellknow, wealmost invariably findgreat tendency toleakage at thejoints ofcontinuously insulated cables. Itisacomparatively easymatter togethigh insulation where thecontinuous insula- tionisunbroken ;butevery joint, whether itisaTjoint forthe house services orajoint onthemain line, isalways asource of leakage andtrouble. Wefindthat thecontinuous insulation of 4lineofconductors isusually sohigh, andthat ofthejoints so lor,that when wetalk oftheinsulation ofthewhole line weare rallyspeaking ofthatofthejointsonly. Iwish toaddaword ofpersonal explanation. Some gentle- menhave recently seemed tothink that although lastyear I took up aposition hostile tothe alternating transformer system ofelectrical distribution, recently Ihave modified my views. Iwish topoint out that nothing ofthe kind has taken place. Inever was hostile tothealternating trans- former system. Ipointed out that probably itwas notso suitable orsocheap asthe battery transformer system for thelighting ofdensely populated towns such asLondon, but thatthealternating transformer system waswellsuited forless densely populated centres. Iamofthesame opinion still, and amconsequently consistent inadvocating each system forthe Purpose which suits itbest. All, however, that Professor Forbes hasputbefore ushasonly confirmed myown views that the tkernating transformer system ofdistribution isnotnearly so simpleorsoeasytomanageaslastyearhewishedtoleadusto believe, Mr. Frank Wrnne: Professor Forbes has referred totheas.wrane Westinghouse pressure indicator invery high terms, forusein central stations. InWestinghouse’s price list@description is givenofthis,butsoslightandinvolvedthattounderstand itIhad torefertoEngineering of22ndFebruary, wherealongdeseription 286 SOMEELECTRIO LIGHTING, Er.(Mareh14th, we.wane. isgiven ofitunder thename ofa“compensator.” But neither Westinghouse norProfessor Forbes make anyacknowledgment of thereal author ofthis important instrument being Dr. John Hopkinson, who describes itatlength inhiswell-known three- wire patent of1882. Mr,Westinghouse nodoubt deserves credit forthewayhehasworked itout; buttheonly difference that Tcanfindisthat Westinghouse uses thesecondary wires froma converter towind Dr.Hopkingon’s instrument, instead ofusing theprimary wires. ur.Prewe. Mr.W.H.PREECE: There isnopractice more conducive to thewelfare ofthis Institution than forourmembers tocarefully inspect what isbeing done abroad, andtocome here andgive usthe result oftheir inspection. Ihave been aculprit inthis direction myself, andtherefore Iamrather timid atdaring tocriticise the work that hasbeen done byProfessor George Forbes. But Iam strongly inclined tothink that Professor Forbes hasputtheboot onthewrong leg. Instead ofgoing abroad,—instead ofspending hisChristmas holidays inexamining the central stations at Berlin, Milan, and Rome, and coming here and throwing down the gauntlet toEnglish electrical engineers, and telling them that they didnotknow their ownbusiness,—he would have done wisely ifhehadvisited some ofourcentral stations, and gone to Berlin, toMilan, and toRome, and told them what todo. ‘Wearenotaltogether deficient incentral station working in thiscountry. Itisperfectly true that wearealong way behind America, and insome things wearealways prepared tofollow the lead ofAmerica, ‘There isacertain energy andago-aheadedness about that nation that imparts itself touswhen wegothere, and weallfeel better men when wecome back again. We are alwaysgladtofollowtheirlead;butwedonotliketobetold that Italy isleading usinthisparticular direction, Isaythat wearedoing something incentral stations here. HasProfessor Forbes inspected thecentral station inBond Street and under theGrosvenor Gallery since ithasbeen under the charge ofMr.Ferranti? There isakind offashion todecry the work that has been done atthe Grosvenor Gallery: wehear innuendos about thelamps being “like redhairpins,” andthings 1680.) DISCUSSION. 27 ofthat kind; but, gentlemen, the Grosvenor Gallery and itsmr.rene spirited proprietors have maintained electric lighting alive in England, and itisdue totheir energy that atthe present timewefindcapitalists readytoopentheirpocketsandsupport electric lighting inthis country. Itisscarcely afortnight ago thataquarter ofamillion ofmoney was wanted tocomplete the Deptford central station, and inless than one hour that money wasfound intheCity ofLondon;andifamillionpoundswere wanted to-morrow morning, amillion would beforthcoming. Well, there are other central stations. What dowehear about Briston? What dowehear about Holborn? Who isthere that valsalongHolbornanddoesnotseelamps,arcandincandescent, springing upallround and about? What have weheard about Pritchett’s station inRathbone Place? Who isthere that drives along Oxford Street and does not see those beautiful lamps lighting upshop after shop? Kensington Court isquite capable ofspeaking foritself;Whitehall Courtisnottobescoffedat, northeCadogan installation atChelsea. Again, there aretwo langestations each ofwhich tells usalot: oneisatSilvertown, there theIndia-Rubber andGutta-Percha Company have some- thing like 150arclamps and3,000 or4,000 glow lamps going, andworked inaway that isacredit toanyEnglish firm; the ‘ther isaninstallation that people aresometimes apttoturn theirnoses upat—that isPaddington, where over 100arelamps tnd4,000 glow lamps arelighted up,now under thecharge of carrespected Past-President, Mr.Spagnoletti, who will probably speakforhimself onthenextoccasion;butisthereanybodyin thisroomwhohasheardofanybreakdown there?No;itisonlyhenbreakdowns occuratalittleplacelikeBarnetthatwesee‘arpapersnarrating themwithseriousness. Why,therearenot %many lamps alight atBarnet asthere areinmyown house, Whilethevagaries oftheBarnet lamps have occupied columns of artechnical press.SomuchforLondon. Arewedoing anything inthe country? Has Professor George Forbes inspected thecentral station atLiverpool ? There,thereare4,000or5,000lampsgoing. Thereisaplace totfarfrom hisown native heath—Glasgow—where Messrs, Muir 238 SOMEELECTRIC LIGHTING, Ere, (March14th, uePree. &Mavor have acentral station, which, bythe bye, supplies our own Post Office. ‘Then, coming further south, wehaves central station atTaunton, and another which has met with a certain amount ofobloquy—that isatLeamington. ‘The reason that wehear somuch ofthe Leamington installation isthat ‘contractors supplied 16-candle-power lamps forthestreets, which wasanactoffolly. But apart altogether from street lighting, ifanybody goes toLeamington and examines thelight inthe houses, orsees theillumination oftheTown Hall, hewill saythat there isnobuilding inthis world that ismore brilliantly ormore beautifully illuminated than theTown Hall ofLeamington, Let usgofurther south again. Has nothing been done atHastings ? Hasnothing been done atEastbourne? Why, forthelastseven years, Mr.Sayers, Mr.Lowrie, and Mr. Hall have been working away with alltheenergy, nerve, and fervour ofEnglishmen, and have established aninstallation there that hasbeen transplanted toWestBrompton; andIdonotthinkProfessor Forbescould bring aninstance from America which has been carried outmore perfectly. Wewilltake another place—Brighton—and Ithought weshould have heard to-night from Mr. Wright something of what hehas done atBrighton. There aremany who may know ‘Mr.Wright, and itisacharming thing togotoBrighton and seetheway inwhich that gentleman has applied, unaided, the most exquisite automatic apparatus totellhim everything that isgoing on, Iwill not deprive you ofthe pleasure ofhearing him next time. Hehaspromised tocome and show uswhat an English engineer candoatanEnglish central station, There isone point upon which Ishould like you tohave clear anddefinite ideas, andthat isthequestion ofthedurability oflead. ‘That isapoint upon which Mr. Graves and Ican speak with some authority. Ihave here aspecimen oflead-covered wire. Itwasnotlaid down intheyear 1887,orintheyear1886, butitwaslaiddown intheyear 1844; forty-five years ago this ead-covered wire was laid down inthe streets ofLondon, It waspicked uplastyear, ortheyear before, Ithink intheneigh- bourhood ofVauxhall (Mr. Fleetwood will correct me ifIam wrong), and itisasperfect asthevery dayitwas putdown. 1088) DISCUSSION. 239 Thecopper wire iscoated with amixture ofpitch, resin, andmr.Prec. beeswax, and thereason thewires failed and were replaced by gutterpercha-covered wires was not because the lead failed, but because thecopper conductors fellthrough thepitch andcame ‘incontact with thelead, Ihave another specimen here that is wellworth your examination also, and Iamindebted forthis specimen toSirAlbert Cappel, who ishere to-night. Itisa syecimen oflead-covered wire which was dug upatKiddapoor, nearCalcutta: itwasputdowneitherin1851or1854;itwastakenupin1885,anditalsoisquiteperfect. Then,intheyear 1854,Mr.Latimer Clark, who ishere to-night toconfirm what, Inay, started thepneumatic tube system forthetransmission of telegrams. ‘Tubes oflead were placed iniron pipes unprotected inanywayorshape,andthoseleadpipeshaveremained intheir Wotecting casing ofiron from that daytothis—1854 to1889— thirty-five years; and Icansay,inthepresence ofMr.Graves, Mi.Bell, Mr. Fleetwood, Mr. Latimer Clark, and many other oficers who have had plenty ofexperience, that wehave never yetseen asample oflead that has inany way deteriorated from itscontact with iron.* But Ihave had specimens oflead which have decayed. I temember acase where alead-covered wire was laid through Windsor Park—it must have been about theyear 1868 or1869; itwasknown as“Marshall's cable,” and was alead tube with a copperwirerunthrough it,thecopperwirebeingwrapped with cotton andthetube filled with paraffin wax. That very speedily decayed. Itdecayed inpatches, anditwasfound that wherever thelead-covered wire layunprotected indecaying vegetable ‘matter, then thelead wasdestroyed. Well, now, gentlemen, thewayinwhich lead isconverted into thewhite lead used forpaint isbyburying grids oflead in mounds,asitwere,orlayers,oftannin,moistened withvinegar or‘eticacid,Thesameactioncommences wherever thereisleadinthepresence oftannicacid;andweallknowthatthebarksof *Load-covered cableswereplacedinthesewersofParisin1862,andsome ‘olthemarestillinuse,‘Thosethatfailed,failedininsulation, andnotinthedW. HP, 240SOMEELECTRIOLIGHTING,Bre.[March10, uePreetreesarefulloftannicacid,anditisfrombarkthatwegetall ourtannin. Wherever lead passes through such decaying matter asthebark oftrees ordecaying leaves, then wegetdecay. Itisavery easy thing toprotect thelead from theaction oftannicorotheracids,andI'amprepared tostakemyprofessional reputation that ifalead-covered wire, properly protected,islaid inironpipes,itwilllast.mylife,and,Ihope,thelifeofthe Institution ofElectrical Engineers. There isjust one other point connected with lead; itiss decided sliponProfessor Forbes's part, which hasbeen alluded to byMr.Siemens. Professor Forbestellsusthattheremustbeagalvanicactionbetweenleadandiron,andtherefore itisverybal forthe lead. Itisnot bad forthe lead; itisbad fortheiroo, because iron iselectro-positive tolead; and when thetwo things farebrought together that which goes istheiron, andnotthe ead.Iwillnotoccupyyourtimeanylongerthisevening,but, ifImay beallowed, Imay take uponeortwopoints onthenest ‘occasion when we meet. potewor ‘TheCriatnaan: Itisalmost impossible forProfessor Forbes toreply this evening, and, Iunderstand, moreover that there are several gentlemen who would like tospeak. Itherefore adjourn “thediscussion until Thursday, March 21st. Aballot fornew members took place, atwhich thefollowing were elected :— Associates : SidneyJohnCluer. ||WilliamPatrickHenderson, Emest Bonnell Hudson. ‘Students : Arthur E,Childs. Wilfrid Bartholomew Lloyd.DuncanW.Johnston, |HubertH.Nalder. ‘The meeting then adjourned. 80 DIBOUBSION. ty ‘TheOneHundred andEighty-ninth Ordinary General Meeting ofthe Institution was held atthe Institution ofCivil Engineers, 25,Great George Street, Westminster, onThurs- day, March 21st, 1889—Professor W.E.Avntox, F.RS., Vice-President, inthe Chair. The minutes ofthe Ordinary General Meeting held on March 14th were read and approved. The names ofnew candidates foradmission into the Institu- tionwere announced and ordered tobesuspended. Donations tothe Library since the last meeting were anounced (three books) ashaving been received from the publishers, Messrs. Alabaster, Gatehouse, &Co. towhom a hearty vote ofthanks wasduly accorded, The Cuamwan: We will now resume the discussion onrettuor Professor Forbes’s paper, which advancedacertainlengthlasttime. WeleftMr. Preece, wemay say, speaking touswhen thedis- ‘assion was adjourned ;and Iwill askhim, therefore, tocomplete hisremarks now. Mr.W.H.PREECE: There isavery important point onwhich 1,rece Professor George Forbes and Iareatvariance, and Ithink, as itis90important, itisjust aswell that you should have all thereasonsproandconbroughtclearlybeforeyou.Theques- tionis,What isthe most advantageous frequency togive toan alternating-current system? and bytheterm frequency Imean thetotal number ofcomplete periods that take place per second. Aconsiderable amount ofconfusion arises because tome persons speak ofthe number ofalternations per minute, ‘others ofthealternations persecond. When wespeak ofthe alternations perminute, wemean thenumber ofsingle reversals perminute; when wespeak ofthe term frequency, wemean thecomplete period ofthepositive andnegative current—the numberofcomplete periodsthatpasspersecond. a SOMEELECTRIC LIGHTING, Ere, [Marcha1, mnprece. Itisperfectly clearthatitismostdesirable thatthis question shall bethoroughly and clearly thrashed out, forwe have this strange variation inpractice, that inAmerica Mr. Westinghouse adopts afrequency of133 persecond—that is, 16,000 alternations perminute;inEngland,atDeptford,Mr. Ferranti isgoing toadopt afrequency of67; while Professor Forbes has shown usthat atRome Mr. Zipernowski adopts a frequency of42. Well, when wehave such astrange difference ofpractice, itisclear that something requires tobethreshed outtomake matters straight. The first fact that Ishould like tocallyour attention tois this, that those who have hadthelongest experience inworking alternating-current systems have reduced their frequency;for, starting attheGrosvenor Gallery with afrequency of150, Mr. Ferranti has come down to67; Mr. Zipernowski, Ithink, commenced with 200—he has now come down to42; but Mr. Westinghouse inAmerica commenced with 133, and like a good old conservative, hesticks toit. ‘There are reasons, mechanical and electrical, pro and coninthis question. The mechanical reason isavery important one, and itisthat which controls andguides thestrength ofmaterials. Every engineer knows that there isacertain peripheral velocity beyond which ‘youcannot step, forifyoudo,youendanger thestrength ofyour structure. InEngland weadopt,asafairpractical maximum,a velocity ofabout 6,000 feetperminute, Well, wehear ofstrange departures inrates ofrevolution. InAmerica, Professor Forbes tells usthat they adopt arate ofrevolution of1,000 per minute; inEngland theWillans and Robinson—the favourite engine ofthe day—bas been brought down tofrom 350 to 380; while inBerlin, Professor Forbes told us,they have come down toaslowas80. Here, ofcourse, mechanical questions control theday, anditisquite clear that with thesame diameter thelower thenumber ofrevolutions thesafer isyour machinery; but, however, Idonotrelyupon themechanical question, and will notconsider itfurther. Iwill take thedynamo. Now in thedynamo wehave onegreat disturbing element, anecessary evil, and that necessary evil isself-induetion. Professor Forbes 180) DISCUSSION. us haspointed outhow necessary itisthat weshould beable toarPeet work these dynamos inparallel, and towork them inparallel theexistenceofself-induction isnecessary. Weknowthatthe presence ofthis self-induction inalternating-current machines chokes orthrottles theoutput ofthedynamo, and the higher thefrequency with thesame E.MLF., themore does thischoking zis, and the less isthe output ofthe dynamo; hence ina dynamo thefewer thefrequency thegreater theefficiency of Jourapparatus.Let usnext take the conductor. When SirWilliam Thomson delivered his Presidential Address here, heshowed how itwas thatwhen you use copper conductors forthe distribution by your altemnating-current machines, the presence of“time” restricted the entry, asitwere, ofthecurrent into the body ofthewire;itdidnotsoakintothewholesubstance,andthe result was that itwas only akind ofskin-deep conductor, and that itreally was the outside only that acted efficiently; tadheshowed, moreover, that that was dependent onthe frequency. Ihave been incommunication with SirWilliam Thomson onthis matter, and Ithink his answer tomy inquiry issoclear that Icannot dobetter than read it. He sys: “The smaller the number ofperiods per second, con “sistent with thoroughly good action ofthe transformer, the “better. Itook 80persecond formyexample, because that “iswhat Ferranti uses intheGrosvenor Gallery installation, “and Iknow hegets good action with histransformers there. “Ifgood action canbegotwith lessthan 80,Ishould certainly “prefer lessthan 80,because notonly inthegenerator, but “inalltheconductors, both ofprimary and secondary circuits, “itismore difficult togetgood economy ofthecopper the “greater the number ofalternations per second;andyouwill “we from the appendix tomyAddress, that even with 80it “isdifficult enough toget good economy ofcopper inthe “conductors foranything more than 100amperes.” Soyou %:SirWilliam Thomson isvery decided onthe question of lowfrequency. Letustake thetransformer—and here isthe calypoint that Professor Forbes makes inhispaper. Heshows Vou, xv. 18 Bry SOMEELECTRIC LIGHTING, Bre, (Marchte, Netwee that practically thehigher thefrequency, thelesstheefficiency ofthe machine, and therefore that the higher thefrequency, the smaller you can make your converter;thatisProfessor Forbes’s argument. There isnodoubt that that isperfectly true—the lower thefrequency themore the loss ofefficiency inyour converter. But what istheamount ofthis loss, and how canitberemedied? The answer isgiven byProfessor George Forbes himself. Inanote that hegave toour Society last,year, hepointed outthat thefrequency may bediminished without loss ofefficiency ifthe resistance ofthe magnetic circuit bediminished, and ifthe mess ofiron beincreased ; sothat ifwereduce our frequency from 133 to100, we simply have toadd alittle more iron inorder tocom- pensate forthe loss ofefficiency, and the quantity ofiron that youhave touseisvery easily calculated. Infact, Imade some very careful experiments onthis very point with aLowrie- Hall converter, and itcame outthat there wasalossofefficiency of1-2per cent. forevery thousand reduction inalternations per minute—i.e, ifyou came down from 16,000 to15,000, there wasadiminution of1-2percent.,andifyoucamedown from 16,000 to12,000, itwasvirtuallyadiminutionof5percent. sothat youcancompensate forthat byincreasing theweight of your iron somewhat inthesame ratio. But while thebad effect oflowering the frequency has this one disadvantage, there is another great disadvantage inanother direction, and that iswhen ‘youincrease your frequency you bring into action theoperation ofthat mysterious property ofmagnetised iron that iscalled hysteresis, and the result ofthis isthat asyou increase your frequency you increase the temperature ofthecoils ofyour transformer ;and itwas thisvery question oftemperature that was the practical reason that ledFerranti and otherstobring down thefrequency from thehigh figure that heoriginally used tothemuch lower figure that wehave now; sothat, gentlemen, wecometothisconclusion, thatyouhavealltheobjections thatI have enumerated onthe one side tohigh frequency, and you have ontheother side only onereason, and that isthat you have toincrease the size ofyour converter tosecure the same 1988) DISCUSSION, as ficiency. ‘There isasecond reason: itisthat thehigh frequency ws.Pres. isused byMr.Westinghouse inAmerica, Idonotthink thelast reason isagood one, andIknow myAmerican friends sowellthat Tam sure, when they weigh carefully thereasons that Ihave given, and other reasons that remain behind, that MrWesting- house himself isofthat temperament andofthatenergy that the moment hebecomes convinced, ashewill, that 133iswrong, he villcome down to80or100, and wipe outallthose converters thatarenow inexistence. ‘That isonething where theAmeri causbeat us:themoment they areconvinced that athing is wrong, they donothesitate whatever atchucking itaway, and replacing bythebetter material. Another point Iwant toputbefore you is,that Iwant to correct Professor Forbes ontwo historical facts. ‘The one isthat, inreferring toalternating-current dynamos, hehints, notvaguely, butrather clearly, that the Elwell-Parker dynamo, towhich reference has been made, andwhich isused agood deal now in London,isamerecopyofthosethataredescribed inhispaper, andthat arerepresented bydiagrams onthewall. Idonot think that Professor Forbes would have hinted this ifhehad read, asThaveread,thepatentofElwell-Parker of1882;norwould hehave said itifhehad seen, asIhave seen, dynamos made in 1883, photographs ofwhich Ihave here. Theotherhistoricalpointis,thatProfessorForbeshasreferred tothefact that MrWestinghouse inAmerica introduced acom- pound indicating device, which hesays Inidtheprinciple ofthese indicators—that is,having one winding connected, like anordi- saryvoltmeter, with theterminals ofthedynamo, and another ‘winding inthereverse direction tocarry themain current, He quite forgets topoint outthat this compound device which has beenused byMr.Westinghouse inAmerica istheinvention and thepatent ofDr.John Hopkinson, There isanother point in connection with Dr.John Hopkinson that hasnotbeen taken sufficient notice of. Itis,that not only was heasimultaneous inventorofthethree-wire systemwithEdison,butitistoDr.Jobn Hopkingon alone that weareable towork alternating-current ‘machines inparallel. Itisquite truethat Mr.Wilde many years 6 SOME ELECTRIC LIGHTING, Ere, (March let wr.Preece.beforehadreferred tothematter, butitwasinalecture inthis room, delivered totheInstitution ofCivil Engineers, that Dr. John Hopkinson forthefirst time brought toournotice thefact that itwas possible towork alternating-current machines in parallel, andProfessor Adams wastheonewhoproved practically, down attheNorth Foreland, that such was the fact. urswat Mr, Stuart Russeit: There areone ortwo points onthe“ew question ofunderground cables upon which Ishould like to sayafewwords. Several systems have been mentioned inthepaper andinthe discussion, and Iwould speak first ofthelead-covered cables. It has always appeared tomethat there must beavery great danger inthe useofthese cables, from the fact that the occurrence ofsmall pinholes oreracks inthe lead may escape detection atthe time oftesting, because, unless the cable is immersed foravery long time indeed inwater, themoisture has not time totake effect on the fibrous material inside the lead. We bave heard from America that one ofthe great complaints about these lead-covered cables hasbeen that there isacontinual and increasing leakage taking place, starting a very short time after thecable hasbeen laid, This Ihave seen stated inmany reportsofdiscussionswhichhavetakenplaceatthe meetings oftheAmerican Societies, when thequestion ofunder- ground cables hasbeen discussed ;anditappears tomethat the readiest explanation ofthisisthefactthattheleadisnotperfect atthetime ofmanufacture, and that these imperfections escape detection atthetime oftesting. There isalso,Ithink,some danger owing tothecomparative want offlexibility ofalead- covered cable—i.e., that especially when dealing with heavy cables, cables with heavy conductors, thecoiling and uncoiling ‘and thehandling necessary inlaying the cable areapt, perhaps, tostart small faults inthe lead. These small faults inthe same way may take some time toaffect theinsulation, but Ibelieve myself that itismerely aquestion oftime, and that sooner or later, ifthere istheslightest pinhole inthelead, moisture will creep inand the insulation will fall, Itwould beinteresting to know ifinBerlin anysuch effect hasbeen noticed ashasbeen 1889) DISCUSSION a7 foandinAmerica—that is,agradualfallingoftheinsulation ofws,sua these lead-covered cables. Onthequestion oftheperishing of leadwehave really very little, or,atany rate, Ihave very little, experience atall;itseemstometodependalmostentirelyon thesoilinwhich itislaid. Undoubtedly, inAmerica, inNew York expecially, where they seem tomeet with very great difficulties, andwhere newly laid cables have frequently come togrief on that account, the acids orgases which are inthe soil have attacked the lead. But this question isonewhich must be made aspecial study ineach particular case; and the nature of thesoilmust certainly betaken into account inarranging the method oflaying thecable. ‘Anaccount wasgiven usbyMr. Alexander Siemens ofthe actual number offailures inthe Berlin cables towhich Professor Forbeshadreferred, whereeightcableshadfailedoutofsomething like35:that seems tometobeafairly large percentage. Six ofthose failures were unaccounted for,and Iwill suggest that posibly,Imayalmostsayprobably, theywerecausedbythis absorption ofmoisture coming inthrough theminutest fault in thelead, Mr,Aupxanver Siewens: May Ijust correct @wrongur impression? You must. notputitthat eight cables went wrong outof35. ‘There were four places atwhich there were 35cables passing, anditwasatthose four places where thefaults occurred, Theother figure that Igave ought tobetaken—that out of 144,000 metres 250metres had tobereplaced. Mr.StuartRusset: Ibegyourpardon,Iunderstood youtoms,start sayeight cables failed outof35;ofcourse that undoubtedly alters thepercentage offailuresverylargely. Iwasgoingtorefertothefailures which were attributed tomechanical injury, and that, I think, shows that thelead covering and iron sheathing isnot sufcient torender such acable really safe underground, and seems tometobeavery strong argument infavour oflaying all thecables inconduits oriniron pipes. Inhispaper Professor Forbes stated thathewasnotaware of ‘anycable which had lasted more than three orfour years. Professor Fonses: Lead-covered. 8 SOME ELECTRIC LIGHTING, Ere. (March alst, McQuart Mr,Stvant RUsseLt: Lead-covered ;andatpresent,inthisdis- cussion, nogentleman hasgiven anyinstance ofthis. Mr.Lant Carpenter said heknew ofmany cables which had been very successful;butIthinkitisamatterofregretthat,forfearof taking upthetime ofthemeeting, heshould have abstained from naming theplaces. These cases would,Ithink,havebeen ofthegreatest value; because, were weinformed astothecon- ditions under which these lead-covered cables had been success- ful,bycomparing them with theconditions under which they have been known tofail,weshould have gotsome very valuable information, andpossibly have gotavery good line towork onas towith what voltages and under what conditions itwould have been safe tousethem, and when not, Mr. Lant Carpenter also referred tothequestion ofputting vuleanised rubber cables under lead, andthere Ithink hisfears were unfounded. Heimagined that thefree sulphur intherubber would affect thelead. First ofall,theinsulation ofthecable does not depend upon the lead covering, andmany vulcanised cables have been laidwithout lead covering, and have been inconstant useforperiods upto three and four years. Secondly, theamount offree sulphur in rubber isvery small indeed ifithasbeen properly vulcanised, and, owing totheinterposition ofoneormore compounded tapes between therubber andthelead,theeffectofthesulphur onthe ead would bevery small, asthecompound would really absorb thesulphur and prevent itgetting atthelead. Ifamey, from experience with rubber-insulated cables covered with lead, that there isreally nothing tobefeared onthis point, and, atany rate, that thedestruction ofthelead ontheinner surface isvery much lessthan that which takes place ontheouter surface, even intheordinary soil. Speaking ofindia-rubber cables reminds methat Mr,A.Siemens saidthat oneofthedisadvantages of rubberasaninsulator wasitsvariability. ‘Thisappears tometo bereally @question ofmanufacture, and todepend onthe knowledge ofthequalities ofrubberthatthemanufacturer hasathisdisposal. Ofcoursetherearemanydifferent gradesofraw rubber, andmany different ways oftreating them, and itisonly natural toexpect that differences inthe manufactured article 8) DISCUBBION. 49 willoceur, justastheyoccur inother manufactured articles—wt.suas dynamos, for instance, where the efficiencies ofvarious makers arenotallthesame. But, however, Iaminformed bythose who have made aspecial study oftheparticular question, that by usingoneofthefourorfivedifferent kindsofgoodrubber,and bytreating itexactly inthe same way, there isnopractical ificulty whatever inreproducing rubbers ofthesame quality, andtherefore this disadvantage which isattributed torubber is really notanimportant oneatall. Iwill now pass toanother class ofmain altogether—that advocated byMr.Crompton—and here Ithink theonepoint to lookatis,that although airinsulation isavery excellent thing in itself, there isalways the danger ofitsperhaps atsome time becoming water insulation, Overhead telegraph lines, Ibelieve, arenotalways perfect, and Ithink that inmany respects they are inquite asgood acondition asabare wire inatrench. Wehave heard great outeries about the breakdown ofoverhead lines through asnowstorm; butsuppose weimagine acase, which is really notlessprobable than that ofthe snowstorm, viz., that of thetrench getting flooded from exceptionally heavy rains orthe bursting ofadrain; Ithink that theoutery raised onsuch an occasion asthat would beeven more serious still, ‘Wehave tothank Professor Forbes forthesuggestion inhis Peper which hasbrought forth from Mr.Crompton thestatement, ofatest ofthese mains, butour information wants supplement ingonthis point, Iremember, ataprevious meeting inthis ‘om, that Mr.Crompton, with afrankness which really cannot be toomuch praised, stated that “of course hechose histime for “testing.” Now thetime chosen would depend upon what you really wish toget, and itwould beinteresting toknow whether theresult given isamaximum oraminimum, orwhether weare totake itasthe happy mean. The actual result ofthetest— 380,000ohms,IthinkMr.Crompton said,permile—does notseemveryhigh,andMr.Crompton explained thatbysayingthatthegreater part oftheleakage hethought wasatthejoints where theordinary covered cables were joined tohisbare mains. If thatbeso,itseemstobeagreatdisadvantage tothesystem,for 200 SOMEELECTRIC LIGHTING, Ere,[Marchie, ur,2uuat_in advocating itMr.Crompton saidthathewaslooking forward to thetime when every second orthird house would haveabranch toitsand ifthe few branches which Iunderstand were onthe main tested, gave aresistance ofonly 380,000 ohms, when we come togetabranch every 20or30yards wemust expect the insulation resistance tobevery much decreased indeed. ‘Another system hasbeen mentioned (the Edison tube), andI have noticed with regard toitthat although ithasmetwitha very large amount ofsuccess, yetitisnot,according tomymind, perfect system. There hasbeen considerable trouble inoneor two Continental stations, somuch sothat the tubes have been pulled up,andIunderstand they arepossibly tobeabandoned andreplaced bysome other system. also find that inAmerica some ofthelocal Edison Companies have given uptheuseoftheEdison tubes, andareusing cables inplace ofthem; and inFrance, inthe recent installation atthePalais Royal,Iaminformedthatindia-rubber cablesare being used bythe local Edison Company, asalso intheOpera House. Idonotknow whether thisisaquestion ofcostorofthe efficiencyofthecable,andpossiblysomeothergentlemannow Present who ismore acquainted with the Edison work may be able tothrow some light onthesubject. Iwould conclude bythanking Professor Forbes forthepaper thathehasputbeforeus:andalthough somespeakersatthelast meeting seemed torather take exception tobeing taught byour Contintental brethren,Iwouldsuggestthat,eventhoughalltheir work had been afailure, which itundoubtedly has not, their experience wouldbeofmostundoubted usetous;asthenextbest thing toknowing exactly how todoit,istoknow what toavoid. te.wei, Mr. A.Waicuir: Although Ihave come upfrom Brighton un- prepared tospeak, Icannot, help expressing mydisappointment at nothearing from Professor Forbes certain data andfigures which would have been extremely valuable tocentral station engineers. Themost valuable facttobelearned from theexperience offoreign stations istheaverage costofproduction oftheelectric light, or theaverage number ofpounds ofcoal consumed perunit pro- duced forthe different systems employed ;also,Ilookedfora 1669.) DISCUSSION. 251 statement astotheratio oftheaverage current tothemaximum %s.weight required. These arefacts that areextremely important, and about which we have not heard aword from Professor Forbes. I think, also, wemight have hadsome information about thecon- centric mains, which isasystem interesting agreat many engineers now. Ishould like toknow whether, inthetowns where these concentric mains have been used, there hasbeen any interference inthetelephone ortelegraph systems. The question ofthe subdivision ofplant isalso asubject which Ishould have thought would have formed part ofProfessor Forbes’s paper. How todivide themaximum load curve; into what units the generating power was divided—I hope that Professor Forbes, inhisanswers, willgive usthebenefit ofthese ‘experiences. IfThadcome prepared tospeak to-night, Ishould have been veryhappy togive certain figures relating totheabove problems 4sdetermined inBrighton, The CaamMaN: The very information you mention incon- uection with Brighton will bemost acceptable. Mr.Waicat: IamsorryIamnotprepared togivethe information from memory. The CaatmMan: Surely, with your long and valuable experi- ‘ace, youcangive ussome information, Mr. Wrucut: Imay say that the average number of pounds ofcoal used throughout theyear per unit ontheold Brush multiple-series system was 24Ibs., while with thenew siternating system, with Elwell-Parker dynamos andconverters, driven from thesame engines, thisfigure comes uptonearly25iba.;showingthattheBrushmultiple-series systemseemstobe Alightly more efficient than themodern alternating system with Elwell-Parker dynamos. Ibelieve thehigher coal cost.ofthe herating system isduetotheexcessive dayandnight magnetis- ingcurrents required bytheLowrie-Hall transformers. Theratio cfthemaximum currenttotheaveragedailycurrentatBrighton, sypears inDecembertobefive,andtheproperdivisionofplant stBrighton would bebyfour equal units. Ishould like tomention afact about thecost ofrunning 282 SOMEELECTRIC LIGHTING, Ere. (March214, Me.wright transformers empty, orofmagnetising transformers intheday- time: this has, Iconsider, avery important bearing. Itis generally supposed, from theoretical reasons, that thework in magnetising the transformer ispractically nothing, thelag being nearly 90degrees; butwehave taken along seriesof power measurements tofindoutwhether the4or5amperes sentoutinthedaytimeatBrighton meantworkornot,andwefind that 4amperes with the secondaries open absorb asmuch power as4amperes and2,000 volts running ondead resistance. Imention this because itseems acommon impression that 4 or5amperes ofmagnetising current mean nowork onthe engine, whereas itmeansagreatdealofwork. MewsMr.F,Wyugs: Ihave never hadany difficulty incoupling machines inparallel. Professor G.Forses: Uptowhat sizes ? Mr.F.Wvues: Ihave hadnoexperience with large machines. Reger ‘TheCuamaan: WillMr.Eddison giveussome information about lead-covered wire? Misa, Mr.R.W.Eopisox: Ididnotattend thismeeting withtheexpectation ofsayinganything onthesubjectoflead-covered cables, butIcame rather inthehope ofhearing from Professor Forbes thereasons why Lead-covered Cables were supposed tohave failed sorapidly ashestated they had done insome instances. It would beinteresting toknow whether the cables alluded to had been drawn into lead pipes, orinwhat way thelead bad been applied, and also the nature ofthe dielectric. Ithink there isnoreason why the lead covering ofatelegraph ot electric light cable should fail any more readily than an ordinary lead water-pipe, ifthe lead beapplied inaproper manner, the cables laid with ordinary care, and alljoints properly made. Myexperience inthe manufacture oflead- covered cables does not extend for“three years,” but Ishall bevery glad inthe course ofanother twelve months, togive particulars oflead-covered cables that will then have been working beyond Professor Forbes's limit. Uptothepresent time Ihave notfound anyofourlead-covered cables tofail either inlowering ofthe insulation (asmentioned byMr. 18693 DISCUSSION. 258 Russell)orthrough deterioration oftheleadcovering, unlessMi, itbasreceived external injury. TheCuamwax: Somenoteshavebeensent,sincethelastpormormeeting, byGeneral Webber, who isunfortunately unable tobe present this evening. The Secretary read the following communication from General C.E.Webber :— “Most ofthespeakers inthisdiscussion haveeither commenced geen! “orfinished their remarks with cordial expressions ofadmiration “forProfessor Forbes's paper, but inthebody oftheir remarks “have been very scant ofanything butadverse comment. “The Professor hasbeen twitted with oblivion, and neglect of “what engineers inthiscountry have already accomplished;and “Mr. Preece hasgiven usalistofcompanies who aresuppliers “ofelectricity, and asks the author, have they not accom- “plished more than heseems inclined togive them credit “for, “Itisquite true that designs ofsystems which show themost “complete acquaintance with alltheconditions which thePro- “fessor seems toimply that English engineers areignorant of “have been carefully elaborated bymany ofus. “Inmyown case Iwasledtolook upsome papers connected “ith Chelsea, and Ifind that acomplete specification fora “rystem inthenorth-east part ofthat parish, requiring distribu- “tion bydirect low-tension current onthethree-wire system, was |“printed forprivate circulation in1884, andthatmyfigures, &c., “were checked and approved atthat time byProfessor Ayrton. |“Thecapital costwas£48,000. Thechief difference between |“thatestimateandothersmadelater,liesinanoverestimate ofthe “aumber ofhouseholders whoarelikely totake theelectric light |“within such anarea, We then estimated forevery house, “whereas experience hassince shown thatoneinthree isnearer “themark. “Many bere canpoint tosimilar labours, and cangive the “vell-anderstood reasons whysolittle realwork followed. “But Ifeel that wemust bona fide congratulate Professor “Forbes onthewayhehasoccupied theground onthesubject a4 SOMEELECTRIC LIGHTING, Ere, (March21s, Geagat “ofdistribution, ofwhich, asweallknow, hehasmade especial “study formany years, “Ifitisthecasethatthispartofelectric lighting hasreceived “little genuine development inthiscountry, hecannot have done “usagreater service than bytelling uswhat hasbeen done in “that direction elsewhere. “We have been called on,inthecourse ofthediscussion,to “admire what several enterprising companies have accomplished “inLondonandothertowns;butIfeelsurethatProfessor Forbes, “inhisreply, willbear meoutwhen Isaythat most ofthem,by “their manner ofdistribution, have done farmore tohinder electric lighting than toadvance it. “Idonotrefer totheunsightliness ofoverhead cables % “much astotheabsolute want ofsystem ofany kind inarrang- “ing fortheir routes asameans oftown distribution. “Tventure toput itthus: Would any onewho hashad “experience ofthis kind ofconstruction foraerial electrical “conductors intowns, estimate thevalue ofthelines weseehere “and there inLondon, atmore than their value asold material “after being taken down, supposing herepresented acompany “which isacting under thepowers andregulations of»Pro- “visional Order that had been asked totake them over? “Ithink not. Ihave never seen one, except perhaps foruse “as afeeder, that would beofany assistance whatever tounde “takers who had seriously toestablish asystem ofdistributing “mains inadistrict. Unless they already existed, noonewould “have anything todowith them even asfeeders. “Considering theabsolute insecurity oftenure forthepoles “and supports, and ofway-leave fortheconductors passing over «private property, Idonotthink that anycongratulation canbe “offered tothose who have been beguiled tocountenance their “money being spent inthat way. “Ishould beglad toknow what arrangements have guided “those responsible fordesigning these systems other than hazard “and thecaprice ofthepeople who areapplied toforway-leave. “Have they anyfeeding mains? Arethere anypoints where the “electrical pressure isequally maintained? Have they anypilot 80 DISCUSSION. 256 ‘wires? What precautions havebeentaken tosecure thesafety Gemnt“ofthepublic? Havetheylightning protectors ?Canpartofa “system beisolated incaseofafire? Arethemains welloutof “reach inpassing other structures? “No wonder Professor Forbes indirectly criticises thework in “this country, bydescribing what isdone inother parts of “Earope, when, with few exceptions, ourdistribution systems “bear little evidence ofgood engineering orscientific knowledge. “At one time ofmylife noone was astronger advocate of “that best ofallinsulators, namely, theair,inconnection with “telegraph and telephone lines, andnoonemore believed that it “required better engineering skill toputupagood overhead line “than tolayone down underground. | “But thepoints tobereached bythewires forthose purposes |‘weisolated andscattered. Ontheotherhand,thehouses tobe |“supplied with current forelectric lighting, under thecompulsory +“provisions ofaProvisional Order orlicense, arealmost next door “tooneanother and instraight lines. Hence, while overhead “lines aresuitable fortheoneservice, they areeminently ill- 1“adapted totheother. ‘We aretoldthatweought tobegrateful tothosewhohave “carried theirconductors overhead fromtheirstations inalldirec- “tions toisolated wealthy customers, . “Ichallenge anyonetoprove that theindustry hasbeen in i‘anywaybenefited orpromoted bytheunbusiness-like proce- “dure Ihave described, orthat theexample ofthose wealthy “castomers willpromote themore general useofthelight. On “thecontrary, their indifference toprice has,ifanything, tended “todiscourage their lesswealthy neighbours, andtogive the “impression that itisacostly luxury instead ofawelcome “necessity. “Professor Forbes has told uswhat the concessionaires under “Continental mtnicipalities have been obliged todo. Anyone “hohasseen(asIhave) many oftheregulations under which “theright tolight some oftheContinental cities isconceded, “villrecognise thewisdom ofthosecharged withthepreserva !“tion ofthepublic safety andinterests, andcanonlyregret that 256 SOMEELECTRIC LIGHTING, Ere. (March, Gaga“theextraordinary stateofourmunicipal lawdidnotlongago"«give ourlocal authorities similar powers. “Mycontentions arevery much supported bythefact that “some ofthose whom Itake theliberty tostyle ‘offender; “instead ofbenefactors, arenow seeking toplace themselves “under Board ofTrade control, when, ifIamnotmistaken, they wil have toretrace many aweary steps +100Moreh,1889." weoote Mr,D.©,Bate: May IaskProfessor Forbes what healludes toinhispaper when hesays that atBerlin “the system of “feeders and mains has been adopted toproduce assmall a “variation ofpressure inthemainsaspossible.Thisisaccom- «plished byhaving alarge number ofconverters.” Professor G.Forses: Ithink that isfeeders: itisamisprint, Mr. D.C.Bare: AsIthought: itwas adirect-working system ? Professor G.Forses: Yes, Mr. D.C.Bare: Iwould just add aword ortwo tothe discussion onthequestion ofunderground mains. Itappears that beyond doubt some mains failed atBerlin, whosesoever they were, and that, ashas been usually the custom, thoso mains were expensively insulated; probably when they were firstput down, they tested toahigh number ofmegohms ofinsulation resistance, and yet insomething less than three years there ‘were 200 and odd metres which had tobe cut ont. wen, ‘MrALEXANDER SIEMENS: Please donotforget thatitwas not the cables that were Inid in1885 which failed, but those which were laidlastyear andtheyear before. Professor Forbes wasuninformed onthat point: itwasnottheoldbutthenew cables that failed, and the cause was undoubtedly traced to mechanical starting ofthe faults. uni _Mr.D.C, Bare: Iquite understood Mr. Siemens tohave made itperfectly clear tousthat itwas not the oldcables, but itwas some cables that failed. ‘Mr,Auexanpen Steers: Yes ;mechanically. Mr.D.C.Bare:Ifitwasthenewonesthathadbeen 1989) ‘DISCUSSION. 387 worked forashort time only, Ithink that that only strengthens sr.Bats, myargument that cables failed within avery short time, by ‘mechanical injury insome eases; butinsome cases—six, Ithink 8previousspeakerhassaid—they failedfromsomeunexplained reason, which might beelectrical ormight not; atany rate, theydidfail. ‘The point Iwish tobring outisthis—that the ultimate court ofappeal which will decide whether electric lighting istobeasuccess orwhether itisnot, istobethe pocket ofthe shareholder, and the dividend which the central station will pay. Now, Sir,ifyouaretoputdownmainsin which your copper isbut asmall percentage ofthe total cost, andthose mains aretofail, asthese and other mains appear tohave done, pretty quickly, you have only thecopper asan asset tosell foroldcopper; the insulation isworth nothing aferithasbeenputontothecable,however muchitcostto pititthere, and however much itcost asinsulating material. Itherefore think that itisquite possible tooverdo thecost cfunderground mains.Ifancythatmanywell-known makerswold make cables with aninsulation of100megohms amile inwhich thecopper would costabout 50percent. ofthetotal atlay. These, wewill suppose, lastthree years; they arenot fring tofail allalong the line,—they arenot tocome full of iaholes like acolander,—but they will failinspecific points, %,to quote Berlin again, where only 200 odd metres outofa verylarge number failed, andtherefore itwill beperfectly easy toreplace them atacomparatively small outlay. If,however, yeuhave toreplace heavily insulated mains, even supposing theylastdouble thetime, they willcosttreble theamount or ore, and the interest oncapital that will beexpended will besogreat that Iventure tothink itwillprobably overbalance thecostofrepairs onalessinsulated cable. One hundred Regohms permile seems tobeapretty good insulation ifit embekept, and Iventure tosay that probably ten would besufficient forsafety andgood working. Suppose youput doraacable with 500megohms amile, isitgoing tolast fivetimes aslong asoneof100megohms would last? andif itdoesnot,where isthoadvantage ofputtingsomuchmoney 258 SOMEELECTRIC LIGHTING, Ere.[archlt, wr.tste,underground tohavetopayinterestoncapital,whenavery small amount ofrepairs onalessinsulated cable would probably Keep itjust aslong asthebetter insulated one? IthinkI have heard that one central station inEngland uses amain which costs something like £180 amile, whereas acable of 100megohms would only cost somewhere about £50 amile, Tam speaking of:%;wire. Passing from cables, Iwould remark, asregards theswitch- ing-in oflarge dynamos, that Ihave apretty good memory that atthe Edison station onthe Holborn Viaduct, with Dr Hopkinson's differentially wound voltmeter, wewere enabled to put the two “Jumbo” machines, alternately ortogether, into themains without having togeteither ofthem uptofall _ load before doing so. One high-resistance wire wasputacross themains oftheworking dynamo, thesecond onewasthen started with itspoles inconnection with the other winding; assoon asthepotential ofthesecond dynamo became asgreat asthe first, itswitched itself inwithin avolt ortwo ofthe right potential, and either speeded uporslowed down asmight be. westony Mr.W.M,Morney: Astoworking large continuous-current ‘machines inparallel, Imay point outthat itisrather easierto work such machines parallel, ifthey have aconsiderable drop inthe characteristic. Most modern machines change from & generator toamotor within avery short range ofspeed, but machines such asthe “Jumbo” Edison machines had, Ithink, « considerable drop, and agood many machines atpresent are made, perhaps purposely, with aconsiderable drop; and such machines, whether shunt orcompound, aremore easily putin parallel than machines that have anearly straight. characteristic. Inhispaper, page 172, Professor Forbes alludes totwo150H.P. alternators that areself-exciting, and totwo larger ones of 600HLP. each which areindependently excited. ‘That israther instructive, Itappears tobethe practice ofthose who have had most experience inalternators touseindependent exciters forlarge machines; the oldpractice which originated with Wilde, orbefore him, was tocommute thealternating current anduseitforthefields. Here wehave Zipernowski using large 1889 DISCUSSION. 250 machines of150 H.P. self-excited. Ihave not had any ex-Mr.Mortey. perience with self-exciting alternators, andshould like toknow if Professor Forbes hasanyinformation astotherelative efficiency ofexcitation, andoftherelativeefficiency ofthewholemachinewithself-excitation, andwith independent excitation. With self- excitation there must beagood deal ofinductional loss, and the difficulty ofgetting agood commutator collection isperhaps considerable. ‘TheEdison meter wasalluded tobyProfessor Forbes asbeing largely used atMilan, and healso spoke very highly ofitin hisrecentpaperattheSocietyofArts.IrefertotheEdison electrolytic meter, and Iwish toask for some information stout it,which Idare sayagood many here would also be ladtohave, astohow such meters areput into circuit. For costant-current work, where you simply want toputsuch a meter, forinstance, across agroup ofincandescent lamps oracross umarelamp, and togetaconstant effect aslong asitison, theproblem iseasy enough; butEdison inAmerica, andthe Eiison people inMilan, usetheelectrolytic meter exclusively for anllel, where theconditions arenotsosimple. The difficulty thatIhavefound,inafewexperiments Ihavemade,hasbeen thatunless aperceptible resistance isputinto thecircuit youdo ‘wtgetanydeposit atallwith very light loads. Ifyouputsuch ‘resistance into thecircuit astogive asufficient deposit with lightloads, then thefallofpotential duetothat resistance when theloadbecomes heavy isquite serious. Ibelieve thatthatmay havebeen overcome toacertain extent byusing zincandapure tne-sulphate solution, but Ifound that difficulty with ordinary ‘copper sulphate; andIshould liketoknow what theresistance is acts which thecellisplaced, what the electrolyte is,what the Belals are,what isthepercentage losswith thegreatest current, thatistaken through those meters, andwhat isthegreatest Proportional range. Ifally endorse allthat Professor Forbes hassaidabout the beautiful effect produced inMilan bythearelighting inthe ‘streets: itcertainly ismagnificent. Thephotographs onthewall sowthesquareinfrontoftheCathedral. ThatisanordinaryVou. xv, 19 260 SOMEELECTRIO LIGHTING, Bre.[March2, Ae.Morey: ease oflighting @square such asmay beseen anywhere, with the lamps onposts; but the lighting ofthe streets isdifferent, There areabout seven main streets branching out from the Cathedral tothegates ofthecity, andtheeffect ofthelighting inthese streets isvery fine. But weought toremember that the working conditions inMilan are very different tothe conditions existing here. InMilan they have the ondinary Continental arrangement ofhigh houses, generally white orlight coloured, andthelamps arehung inthemiddle oftheroadby steel cables passing across from onehouse totheother. Nor, with ourgrandmotherly system ofgovernment, and with allthe authorities and interests that wehave toconsult and consider, there would beconsiderable opposition ifanybody spoke of putting acable across thestreet tohang anarclamp on;yetitis farthebest system;itisnotatallunsightly.‘Thesteelwireis« verysmallthing;itcanbetightenedupbyascreweyeletletinto thewall, and thewhole thing isasconvenient asitcanbemade. ‘The wires arerun along the front ofthehouses under the windows, and arefrom there taken out tothe lamps. The lighting ofMilan isquite anexample ofwhat street electric lighting should be,butIamquite sure that itwould notbeper mitted inanEnglish town—not foranysound reason, butsimply because we are behind inthese matters. Ishould alsolike togetsome information astoanEnglish installation about which much has been said—that atWest Brompton. Ifind that they areputting lead-covered cables down iniron pipes, and that these lead-covered cables areput down singly. This isathing that probably agood many others here have noticed: Iknow that Mr. Fricker has noticed it,and probably others. Instead ofinsulating thetwowires andputting them into onecovering, Iamtold that each conductor isputin ‘separate lead covering. Inthediscussion onMr,Kapp’s paper atthe Institution ofCivil Engineers the other day, one ofthe engineers oftheWest Brompton station showed some very pretty and ingenious arrangements formaking awatertight and well- insulated cover joint attheTconnections tothe lead-covered cables. That arrangement showed that thewires were tobeused 1889.) ‘DISCUSSION, 261 singly. Now, isitnotacanon ofunderground work, where alter-ar.Money. nating currents areused, that the cables should beeither con- centriorthattheyshouldboth,ifenclosedatall,beenclosed | inthesame metallic envelope, otherwise themaximum waste is ||incurred? IwishtoaskifProfessor Forbes hasnoticed this, | andifhecan support thepractice that istobecarried outat | West Brompton. Ifhecannot, Ithink itisbest forallcon- cemed that attention should be drawn tothe mistake that is pethape being made atWest Brompton. Mr.W.H.Preece: Ishould just like tosayone word inwr.Mew. replytoMr.Mordey,topointoutthattheplanofsuspending lamps inthecentre ofthestreets was my proposition totheCity authorities, and some fiveorsixyears agoIexperimented on those lines atWimbledon. Themain thoroughfare atWimbledon vaslighted with lamps suspended over thecentre ofthestreets, and there isno doubt itisthe most efficient and the most effective way oflighting atown; butthepractical, serious, and almost insurmountable objection is,that thewires andsupports must becarried atasufficient height toallow the fire escapes to pasthrough without anydanger, andwere itnotfortheexistenceoffreescapes wecertainly should have ourstreets lighted inthe vaythat they arelighted inMilan. TheCHamMan: Iamafraiditwouldbeverydifficulttogoprisworroundasharpcornerwithafireescapeinahorizontal position: it“”*”‘ouldprobablybeawkwardforpersonsontheothersideofthecomer; however, itmight betaken atabetter angle than at prsent, and[like Mr.Mordey’s idea ofafrontwheel,butIwould totadvocate hissuggestion that theposition oftheescape in notion should beperfectly horizontal, otherwise perhaps, like Stevenson, who proposed tohave trains going asfastas10miles amhour,hemaythwarthisownplansbythedaringcharacter ‘thisproposition.‘Asnoone else appears towish tomake anyremarks, Iwill, Without detaining you very long, make afew suggestions in élocing thediscussion, Inhispaper Professor Forbes commenced byapologising forgoingabroad,seeingwhatisbeingdonethere,‘comingbackandtellingus.Whyheshouldthinkitnecessary | 262 SOMEELECTRIC LIGHTING, Ere.(March2, ‘ewer tomake anapology Ireally cannot understand. Ifheiswilling "togoabroad, tolearnallhecan,tocomebackhereand immediately give usfull information, Ithink heistobe extremely thanked byus,andsofarfrom apology being necessary from him, ourgratitude isduetohimforhislabours and forthe information hebrings us. Hehasbeen taken totask because he has not visited some ofour London central stations; but wedonot blame aStanley oraLivingstone because they are not perfectly familiar with the sub-tropical garden in Battersea Park; sothat, even ifProfessor Forbes has not seen some ofthe installations that exist inLondon, hedoes not merit blame atourhands onthat account, especially aswe possess thevery simple remedy ofgoing and seeing these for ourselves, Hehas,nodoubt, asMr.Wright haspointed out,not given usperhaps asmuch information asweshould have liked to have had, but probably hehas given usallthe information he could obtain, astothecostofworking thesevarious installations; butonething hehasgiven uswhich isofinterest, andthat isthe cost atwhich theelectric power issupplied toconsumers. I hhave made acomparison toseewhether itwas cheaper ordearer than gas—ie., what was about the proportion incost ofelectric light togasinplaces where theelectric current wasgenerated on afarbiggerscalethanithashithertobeendonein,say,GreatBritain—and Ifindthattheproportion seemstobesomewhat the same. InBerlin theprice ofthe electric power isequivalent to about 8s.3d.per1,000 cubic feetofgas—that is,nearly twice the actual price ofgasinBerlin, which is4s.10d. InMilan the price ofelectric power isequivalent to8s.8d.per1,000 cubic feet ofgas—gas therewasoriginally alittlemorethanthecostofthe electric light, butithasbeen reduced to6s.9d. InEngland, with ourlargest proposed distribution—the Deptford installation—the price willbeequivalent togasat6s.3d.per1,000 cubic feet. In these andother cases that Ihave been able toexamine, itseems that theprice ofelectric power isvery nearly 100percent, more than theprice ofgasattheparticular place. Whether that is because theconsumers ofelectric power appear willing topay about twice asmuch asforgas, and therefore thesuppliers of 3) DISCUSSION. 268 electricity havenotunnaturally fixedthatastheprice,orwhether rramorthatisthelowestpricethatsuppliers ofelectricpowercanfurnish theelectric light for, inorder toenable them topayafair dividend, Idonotknow, Butitisreally aninteresting point to ascertain what isthelowest price that itispossible tosupply lectricity atonalarge scale inagiven place, compared with the Price ofsupplying gasatthat place, both being produced bythe ‘consumption ofcoal, Isitpossible todistribute electric power onavery large scale cheaper than gasissold atinthehouses ? Reference wasmade more than once byProfessor Forbes asto thenon-adoption ofSirWilliam Thomson's law,which hequoted asbeing somany amperes persquare inch. Asamatter offact, SirWilliam ‘Thomson's lawshould notbestated assomany ‘amperes per square inch, but assomany inches perampere—a distinction thatmightappearatfirstsighttrivial,butwhichia all-important, seeing that theproblem SirWilliam Thomson solvedwas:—Given acertain current, tofindoutwhatwasthe proper size ofconductor tousewiththatcurrent. Thatistheonly vrayinwhich hislawcanreally beapplied. Mycolleague and myself have already pointed out that SirWilliam Thomson did notrightly interpret hisown lawinhisInaugural Address at York, The lawthat heworked outmay bethus stated: IfA bethecurrent inamperes that wehave decided tousewith a particular conductor ofresistance rohms permile, then thewatts ‘expended permile are e Art? A'rbeing thewatts expended inheating theconductor, and £tatheformofinterestonmoneysunkinthecopperconductor. (isequal to txexVAore%e, whereiistheratepercent.ofinterest onmoney,cthecostin Pounds sterling ofatonofcopper, ptheresistance ofamile of ‘copperonesquareinchinsection, atheyearlyvalueinpounds ‘sterling ofoneelectric horse-power forthenumber ofhours the on SOMEELROTRIO LIGHTING, Bre.[March21, velaer poweritused.If,forexample,wetakeiasequalto12,toallow fordepreciation, cas£125, pas0-04378 ohm, which isthe resistance at20°C.,iftheconductor has98percent. oftheconductivity ofpurecopper,andaasequalto£15,then¢equals2973. Now, ifAbefixed, then the minimum value ofthe above expression isobtained bymaking t rath, that is,bymaking thetwowastes equal tooneanother. This, fortheparticular value of¢that wehave calculated, leads tos+3. square inch ofcopper per ampere, oratrife over one-fourth square inch per 100 amperes. Now this solution can only beused onthe assumption,aswehavealreadypointedout,that1isthevariableandAis constant, and itcannot beused atall,and hasnopractical meaning whatever, ifrisfixed and Aisvariable. Ifthere be agiven line laid underground oroverhead, ofacertain resistance permile, andyou wish toapply SirWilliam Thomson's lawtofindoutwhatisthemosteconomical currenttoemploy,hislaw hasnoapplication whatever, forthesimple reason that ifyou taketheexpression are * andtrytofindoutwhatisthevalueofA,thatmakestheexpres- sion aminimum, itisAequal tonought, that is,donotuseanycurrentatall;thatanswergivesyoutheminimumwaste,butnottheconditions ofmaximum economy inthecommercial senseof theword. Therefore, theconclusion which SirWilliam Thomson came to,viz., that thetwo wastes (the waste due toheat and thewaste due tointerest oncapital sunk inconductors) should bemade equal, isonly applicable intheparticular case that the current isfixed and theresistance isvariable. Itisnotapplica- bleincases where, aspeople sofrequently areinclined toapply it,the resistance isfixed onaccount ofthe line beingalreadylaid,andthecurrentisvariable;infact,aswohave shown,underthesecircumstances adifferentlawmustbeapplied, 1682.) DISCUSSION. 265 foryoudonotthengetmaximum economy bymaking thetworotorwastesequal;youcangetgreatereconomy,infact,bymaking thetwowastes unequal, What theexact solution isdepends upon what isgiven. Suppose wehave todeal with alinealready made, then thelength ofthelineanditsresistance permile are known, LetV,theP.D. atthedynamo endoftheline, bealso fixed, then the value ofA,the current which gives maximum economy, is at VERE =ntrn where1isthetotallengthoftheconductor inmiles. Ifneither Anorrbefixed, butVbegiven, andif,inaddition, thepower, pwatts, that hastobefurnished atthedistant endof thelinebealsoknown, then, aswehave already shown, nt 4-50 +p) ives maximum economy. Inboth these cases, then, greater economy isobtained by akingtheinterestoncapitalsunkinconductors largerthanthe ‘annual coal billthan bymaking these twosums equal. If,onthe other hand, the P.D. atthe distant end befixed, andnotV,theP.D. atthedynamo end oftheline, and ifAand rhe either, orboth, variable, then Arst gives maximum economy. Ofcourse wehave tothank Sir William Thomson forhaving initiated thekind ofsolution, for having suggested themethod ofsolving such problems, andfor having drawn attention tothefactthat there were twothings to betaken into account, viz,, thewaste ofpower inheat, andthe ‘waste ininterest onmoney. Ionly, therefore, inalluding tothe nile, wish toremind you that you cannot apply hissolution ezcept intheonecase, that is,when thecurrent isthegiven thing, andtheresistance ofthelineisthevariable. Tthoroughly agree with Professor Forbes's remark aboutinsulation: “Wedonotrequireahighinsulation forthesake. “ofgetting asmall loss ofcurrent, butonly because with many “insulators thehigh insulation ismore permanent, andperma- 288 SOMEELECTRIC LIGHTING, Ere.[March2st, Foumer “neney isthequality ofinsulation which ismost tobedesired.” That remark isespecially pertinent inreference tothepaper we recently hadread tousastotheinsulation resistance proper for electric light cireuits, and where itwas proposed totake afixed resistance permile,irrespective ofwhether theleakagewasdue toleakage inthe cable, orleakage inthe fittings. Several speakers then pointed outwhat awrong method that must be, because aninsulation resistance which might befairly good, if theleakage waspartly infittings and partly inthecable, would beanextremely badresistance ifalltheleakage wasinthecable. Only afewdays ago Ihadanexample brought before me, since thelastmeeting, onthis very point. Aline some miles long, having anumber ofoutdoor arclamps oncircuit, hadaresistancewhichvariedfromabout,forthewholeline,1,000,000 toabout, 3,000, sothat itwould beseveral, perhaps four orfive, millions permile onavery dryday, andonly afewthousands permile on averywetday.Now,ofcourse,iftheresistance wentdownto3,000 or4,000 ohms inthecable, thecable would becondemned and cutout; but asthetests showed that the fluctuations from daytodaywere due totheleakage inthe fittings ofthe street arelamps, the installation was notcondemned inspite ofits ‘oceasional low insulation. Reference was made byMr.Preece tothe question ofspeed ofalternation, andhepointed outthat athigh speed theoutput ofadynamo inconsequence ofself-induction isseriously dimin- ished, and therefore alowspeed may bebetter. Ipresume his remarks apply toadynamo inwhich there isironinthearmature. Tcannot imagine that ifyou take adynamo liketheFerranti or Mr.Mordey's, inwhich there isnoiron inthearmature, that the ‘output would beincreased bylowering thespeed: theincrease ofelectro-motive force obtained byincreasing thespeed would far ‘more than make upforanylossdue toself-induction, andthere- foreyouwould certainly have thegreatest output atthegreatest possible speed you could run themachine at, Indeed, Ican hardly imagine hisconclusion can betrue foranyalternate- current dynamo, even ifthere beiron inthearmature. ‘Also, Mr. Preece referred tothe loss ofefficiency indifferent 1860.) DISCUSSION. 267 transformers, andstated howtheefficiency varied whenmoreoFfriswr {essironwasputin. Hewasdealing with very small differences ofpercentage—1} or1°]percent.Ishouldlikeifhecouldtell 1showtheefficiency wasmeasured. Itisvery easy tomake a nistake ofseveral percent. inmeasuring theefficiency ofa transformer, inconsequence ofthefact that you cannot apply a voltmeter, ammeter, orsuch meters, and saythat theproduct oftheir readings gives thewatts supplied. Imay mention, in passing, inconnection with the use ofwattmeters foralternate- carrentcircuits,thatsomeofourstudents, onmakingthecalculation,have found that the coefficient ofself-induction inanon- inductive resistance boxoftheZipernowski wattmeter specially intended foralternate-current circuits, isquite comparable with thecoefficient oftheself-induction ofthesuspended coil, and that people who have been making calculations have been wong inmerely assuming that thecoefficient ofself-induction ofthefine-wire circuitofawattmeter wasthatofthesus- pended coil only. Indeed, fordelicate measurements ofself- induction, itbecomes necessary nottouseordinary double-wound resistance coils atall,foreven those made bythe best makers lave avery considerable amount ofself-induction. Inour Wheatstone bridge attheCentral Institution, wehave totake into account the self-induction of the various double-wound «ails, andyoucannot byany means assume that this isnought. Thebest way tomake the self-induction ofresistance coils really nought would bebytaking aplatinoid wire, covering it with silk, then winding platinoid outside that again, and covering that again with silk, getting, infact, concentric conductors ofplatinoid; youthenwindthemupinacoilor leave them straight, asyoulike, andyouhave acurrent going onewayinawire,andbackbyaninsulated concentric outertube,Producing acircuit whichispracticallynon-inductive. Itwasproperly pointed outbyMr. Wright that you cannot wume that nowatts are given to transformer when the ‘steondary circuitisopen.Indeed,intheexperiments madeby ourstadents that were described inthis room ayear ormore ago, attentionwasdrawntothat,andthenumberofwattsgivento 268 SOMEELECTRIO LIGHTING, Br. [MarchAst, Retewor theprimary circuit ofatransformer wasshown,—the truewatts, notthewatts measured insomo imaginary way, butthetrue watts given forvarious currents inthesecondary, when afixed square root ofmean square ofvolts was maintained atthe terminals ofthe primary,—and itappeared that about one- twentieth ofthemaximum watts wasgiven tothetransformer when thesecondary circuit wasopen: that is,one-twentieth of themaximum watts that wassupplied onfullload was entirely wasted inheating thetransformer when allthelamps were turned off. Itmust indeed beafamiliar experience tomost ofyouthat transformers during thedaygetvery hot,andthat therefore you cannot assume that there ismopower wasted. Idonotmean to saythat thetransformer system isabad system, butIdomean tosaythat wemust not neglect thewaste that occurs intrans- formers with thesecondary circuit open. Iamalittlesurprised tohearfromMr.Wrightthattheyfind that itactually takes more coaltosupply power with transformers than with theBrush multiple-series eystem. The Brush multiple series system, worked atabout 1,500 volts, hasanefficiency of50 percent.,Iunderstand—that istosay,youcangetabout390watts developed inlamps foronehorse-power developed bytheengine; that, Ibelieve, istheresult ofactual experience, sothat thetotal efficiency isabout50percent.Onewouldhavehopedthatyou would getonthewhole more than that with transformers, but according toMr.Wright's experience itrequires more coal to workwithtransformers thantousethehigh-potential multiple- series system. Seeingthelengthyremarksthatwearelikelytohave,and which Ihope weshall have, from Professor Forbes, Idon’t think Tare detain youamoment longer. Iend, therefore, bythanking him formyself, and thanking him,Ihope,inthenameofthe Society, forgoing abroad andgetting information, andgenerously placing atourdisposal thisinformation whichwearenotableto ‘getforourselves. Frigmer ProfessorGzonczFonsss,inreply,said:Inthecourseofthediscussion ofthelasttwoevenings, Imust.say,inthefirstplace,thatIfeelthatIhavebeenagooddealmisrepresented, and@ 1099 ‘DISOUBSION. 269 geoddealmisunderstood. Ihavenothoisteduponapedestal Primereverything foreign, andcastdown into ahole everything British; onthecontrary,Ihavesimplytoldyouwhatisbeingdoneabroad insomelimitednumberofcases,andIhaveseenthereagreat dealthatistobedisapproved of. With regard totheforeign element, Iwould remind youthat latevening, among thevery first speakers, there were three |gentlemen offoreign birth who spoke, and Ithink that they, in |theirspeeches, taught ussomething that several ofthose who |followedthemmightlaytoheart.Thosethreegentlemen—Mr.|Siemens,Mr.Andersen,andMr.Kapp—showed that,indiscussing |mengineering question, itwaspossible forengineers todivest themselves ofallpersonal feeling andstick totheengineering pobleminhand.Ishalltry,asIwouldwishthatsomeofthosewho followedthemhaddone,totakethatlessontoheart,andresist thegreat temptation that has been thrown out byseveral sprakers,toleadmetoreplyinaveinofsarcasm. Iwillgo ‘through the points that have been raised inthe discussion, ‘eparating thewheatfromthechaffasmuchasIcan, Instead ofreplying toeach gentleman inturn, andholding himupbefore youtoridicule ortoadmire, Ihave taken up allthesubjects which have been discussed, intheorder inwhich theyweretreated inthepaper,andIhavetried,asfarasIam Abletodo,todividethedifferent speakers’ subjectsinclassified‘order,andIshalltakethedifferent subjects inorder,insteadofthedifferent speakers.Fint, astotheobject which Ihad inbringing forward this Peper, onwhich there hasbeen adeal ofmisunderstanding. WhatIhavesaidis,thatduringthepastfewyearstheelectriclighting industry inEngland hasbeen delayed—it maybebythe Heetric Lighting Act,itmaybebyother causes—I saythat it tasbeen delayed inEngland. Isaythat during that time there harebeen hundreds ofthousands ofpounds spent abroad by Paple trying tolearn theexperience which isnecessary tobe leamed before thething canbemade acommercial success. All Inowsayisthatweshouldtrytotakeadvantage ofthetimeand ‘money which hasbeen spent bythose people upon those experi- 210 SOMEELECTRIO LIGHTING, Ere, (Marchals, Peieger ments. ‘Theonlyperson whohasventured tooppose thislineof argument which Ihave taken isMr.Preece, who hasmaintained that Iamquite wrong; that there hasbeen nodelay inthe progress ofelectric lighting inEngland; that they have been ‘going ahead faster inEngland than anywhere else; and that there isabsolutely nothing that wecanlearn from foreigners, and that wecanteach them everything. Iamperfectly con- vinced that the English engineers canteach those abroad avery great deal. Iknow that the Americans, while they have been pushing on with their practice, have been deriving their theoretical notions from England, while practical progress was retarded here: they have confessed that themselves, and are perfectly willing toadmit theobligations they owe toEnglish- men. Nogreater advance hasbeen made inpractical work, I think Ican saywithout fear ofadoubt, than theadvance that hasbeen made bytheBrothers Hopkinson when they published their paper onthedesign ofdynamo machines—a paper which hasbeen thebasis ofdesign inevery part oftheworld. That andthediscovery oftheGaulard and Gibbs system ofdistribution arethetwo greatest steps that have been made oflateyears— onemaking dynamo-design possible, theother making distribu- tion possible onalarge scale. This subject hasalso been dealt with byGeneral Webber, inaspirit opposite tothat ofMr. Preece, but Ireally need notwaste more time onthesubject; Ifeelperfectly confident that thegeneral sense ofthose who have studied itisinfavour oftheview that Ihave expressed—that ‘wehave been retarded inouradoption ofelectric lighting. Questions have been asked mewith regard toBerlin, and as far asIcan Iwill answer them. IwasaskedbyMr.Kappaboutthenumberofare lampswhichthereareontheincandescent circuits. Iamreallynot quite sureoftheexact number, butitisnothing likesolarge proportion, Ibelieve,asheandMr.Crompton havethoughtit was. ‘Then Mr. Kapp asks mewhether the cost ofthe mains and feeders includes thearelamps. Itdoes. Ttwas assumed byMr. Andersen, because Isaid there was 180) DISCUSSION. mm 1jpercent.variationofthemains,thatthatmeant1percent.prtsmerupanddown,Thatisnotthecase.Thereisonly1}percent. mariation allowed inthemains inBerlin anywhere—that is,}per cent, upanddown. Theengines were praised very much byMr. Crompton on sccount oftheir absence from shock, noise, and soforth, and I quite agree with him that theengines arevery good ;but Ido totgosofarwithhimastosaythatweinEnglandcannotproduce engines asgood asthem. Ifeelconfident thatinthatdepartment, ofengineering wearestillatthetopofthetrae. Thavebeenaskedaboutthehorse-power—the steamconsumed pethorse-power inBerlin. The exact statement isthat theindi~ ‘atedhorse-power is134Ibs.ofsteam,theelectrical horse-power is15}Ibe.ofsteam, Complaint was made byMr.Wright that Ihave notgiven some facts about the cost, ofproduction, astothe Ibs. ofcoal wedper unit, astothe ratio ofthe maximum current tothe average current, andastotheuseofconcentric mains, andsoon, Ireally thought that Ihadsaidasmuch about these points asI ‘asentitled tointhelimits ofthepaper. Ihave given agood mny ofthese points, ashewillfind ifhereads thepaper. With regard to.Milan, Mr.Preece says there isnothing thatTaycanteachus.IsaythatMilan—and Mr.Mordeyhasalso ated 20most positively—can teach usthat acity lighted with electric arclights isathing very much tobedesired, and that theycanlight uptheir streets right andleft,allthrough thetown, inamost splendid wayandatasatisfactory cost. Iwas asked byMr.Mordey about theEdison meters which are ‘uedinMilan, astotheir varying deposit with light loads, because ifyoudogetanydeposit with light loads then theresistance becomes much higher, or,inother words, therange isnotlarge. ‘Therange ofthemeter isnotlarge, asIstated inmypaper at. theSociety ofArts, towhich MrMordey hasalluded. Ihave made very careful tests ofitmyself, andgenerally itvaries from fivefold totenfold range. ‘Theexcellent street lighting inMilan wasalso spoken ofby Mr,Mordey, andIthink thateveryone whohasvisited that town anSOMEELECTRIOLIGHTING,Ere.[March2, Pofenor= mustlookanxiously forwardtothetimewhenweshallhavealarge portion ofthecityofLondon solighted. Mr.W.M.Monpey: Could you give ustherange oflossof the E.MLF. inthe meters ? Professor G.Fornes: Onlyaverysmallfractionofthecurrent passes through the electrolytic cell, and the resistance ofthe whole thing isx¥ev ofanohm. ‘Mr.W.M.Monpey; That istheresistance that themeterisa shunt of? Professor G.Fornes: Yes, Mr.W.H.SNELL: Might Ibeallowed tosaythat Ihada letterfromEdison's laboratory onlyyesterday morning, andit mentioned, among other things, that their latest’ meters havea lossof“2percent.atfullload. Professor G.Fornes: That isaninteresting fact. ‘That Rome was dealing with alternating currents, onewould have thought would have been apoint which would befreely dis- cussed, butMr.Mordey alone hasnoticed it.Hehasasked about the selfexciters, how they work, and whether their efficiency comes out well. Inever sawaself-excited alternator with solittle sparking at thecommutators; itcertainly isextremely good. Iremember, afew years ago, having one ofthe Lontin type under test for some time, and the sparking was something terrible inaself- exciting arrangement like that; butinthese Zipernowski ones there isvery little sparking atall. But, asamatter offact, experi- ence hasshown, bytheresults ofMessrs. Ganz and others, that itisbetter touseanindependent exciter, and that theefficiency ofworking isreally better under those circumstances. 2DAstothelessonstobelearned,Iwilltaketheseinorder,and soIshall beable togetrapidly through thedifferent points that have been raised. Imust saythat every word almost that hasdropped from any speaker, inwhatever humour hehasbeen speaking, orupon what- ever part ofthesubject hehasbeen speaking, hasconfirmed me inmybelief that wehave lessons tolearn bystudying thework which hasbeen done abroad—lessons astowhat todo,andlessous 1stowhat toavoid. - 1384] DISCUSSION. a Fint,astothedistribution, whichmostspeakersweredealingPriveor vithlastevening. Thefirst point isastothethree-wire system. Dr.FlemingandMr.Crompton havebothsupported thestatement, which Ithink ought tobeamaxim, that thethree-wire system onght always tobeused inalow-pressure system ofdistribution. Theonly person who threw doubts onthe question was Mr. Shoolbred, who inavery cautions way suggested that “hitherto “the smocess ofthethree-wire system inEngland hasnotbeen “ficient towarrant itsgeneral introduction.” That isatype ofsrgument which isvery apttoprevail inEngland, that because thing hasnotbeen largely used inEngland yet,wehadthere- |forebetter wait; and Ithink those very cautious words will |indicateagreatmanyofthereasonswhymoreprogresshasnot, |teenmade inthiscountry. | Thesame thing appliestoelectrictraction:wefindthatthe same argument holds very much inregard toit. The railway agineers will tell youthat thesuccess ofelectric traction inthis country hasnotbeen sogreat astowarrant itsadoption. Yet, in America they have their 250 horse-power locomotive running on theelevated railroad inNew York, drawing itscomplete quota olearriages, anddoing itsregular work with thesteam locomotive everyday;andtheyhavetheirRichmond electrictramears, and.deetric traction isgoing ahead splendidly inAmerica, : Thenext, point isabout feeders. Mr. Kapp complained that +[didnotnowdescribe toyoutheconditions ofworking feeders. ‘Well,considering that Idevoted somuch time tothem inthe ‘cane ofCantor lectures totheSociety ofArts, Ihardly thought thatthiswasthetimetodealwiththesetheoretical pointsabout‘heprinciples ofdistribution. Hepointed outthat heand Me.Crompton hadconsidered thequestion thoroughly intheir designs fortheVictoria Station along time ago, andhewillfind intheCantor lectures that Ifully went intothese points, and stored howfarthequestion hadbeen dealt with uptothat time byMr.Crompton, andwhereinhis‘errorslay. iFeeders were stated byDr.Fleming tobemost important and ‘ential toanysystem ofdistribution, and Mr.Lant Carpenter ‘assidthatitislikefloggingadeadhorseorkillingtheslain m SOMEELECTRIO LIGHTING, Ere, (March1x, Folemer tospeak offeeders. Iwish,gentlemen, thateveryperson inthis country, andevery engineer inthiscountry, would hold thesame opinion. Icannot tellyoutheamount oftrouble Ihave attimes inhaving toargne thispoint with most practical engineers who ‘areopposed tousing feeders with their mains. There areagreat number ofpeople who wish tolaydown their central stations without feeders. The only person who spoke against feeders at thismeetingwasMr.Andersen, whotooksometroubletoshowthat itwaspossible, inalow-tension system, onthethree-wire system going upto400 yards, togetafairdistribution ofelectricity without using feeders, and without any very great expense in copper. Granted thatisso:undoubtedly wealladmitthatto Mr. Andersen; but Mr.Andersen hasnever attempted toshow that hewould notgetafarbetter distribution with thesame quantity ofcopper laiddown slong thesame conduits ifheused feeders tosupply themains; andthere isnottheslightest doubt, just asallthespeakers have said, that feeders areabsolutely essential forthat, andasMr.Andersen hasnotattempted todeny thefactthat feeders would improve thesystem, Ineedhardly waste more time upon thequestion. Mr.F.V,ANDERSEN: Iwould liketosaythat allmyremarkswerecertainly baseduponthesupposition oftheuse.offeeders. Ithink ifProfessor Forbes willread myremarks hewillfindthat 80. Professor G.Forses: Ibegyour pardon most sincerely. Mr.F.V.ANveRsEN: Imentioned themains inthefeeding centres, ofcourse, and Iunderstood that the mains around the station would besupplied byfeeders, andImentioned thefeeding centres, Professor G.Forses: {understood that thewhole system of mains thatwere laiddown were tobetapped bythehouses along which they passed, and that noseparate feeders would beled ‘away untapped uptothecentres ofdistribution. ‘Mr. F.V.ANpeRsen: Imust apologise ifIhave not made myself clear enough. Myremarks were based ontheprinciple of having radial mains, and feeding them radially from feeding centres. 1a88 DISCUBSION. 18 ProfessorG.Forses:Thenyouacceptedtheprinciple ofrorfeeders? ‘Mr.F.V.ANDERSEN: Certainly. Professor G.Forses: That isallright. Then everybody here admits that feeders aretobeused forlow-tension systems: that ispointIamgladtofindweareallagreedupon. The next question isastousing dynamos atdifferent poten- tial,andputtingthepilotwiresinadifferent positiontothat ‘hich hasbeen generally adopted. Mr.Andersen hassaid that using dynamos atdifferent potentials isalittle complicated, and thatisaquestion thatmustofcoursebeconsidered inanycase ‘here itisproposed todoso, Imay mention, however, that the complication, which consists inchanging feeders from onedynamo toanother, isanoperation which isbeing daily performed in lage numbers ofcentral stations everywhere, and Idonot think that there isany trouble indoing that atall. But undoubtedly itwould besimpler, asMr. Andersen very justly remarks, ifwecoulddowithout it;itisonlyaquestionofwhether thebenefit tobegained isworth it.Thesuggestion that Mr. Kapp made, that counter electro-motive force might beintro- daced intheform ofabattery celltochange thepressure given inthedifferent feeders, perhaps limits thecomplication that it would benecessary tointroduce. Objection was made byMr. Andersen toputting the pilot ‘wires into thebranches. Now, pilot wires, wherever you placo them,ifyouaredistributing withallyourdynamos atthesame Potential, your pilot wires must begrouped together, and you musttake anaverage ofthereadings ofthepilot wires; that is thepractice universally adopted where pilot wires areused, andif ‘thatisthecase, then theobjection which Mr.Andersen has nised hardly holds. Hisobjection isthat your pilot wires, ifthey sobeyondthefeedingcentres,willgointothebranches. Well,letthem gointothebranches, andletthem gointoevery one ofthebranches, butputthem inatthehalf-way place which Thave pointed outastheproper place, then take theaverage of them,andyouwillgetverycloseapproximation totheaverage Pressure which youought tosupply those points. VOL, XVIII. 20 26 SOME ELECTRIO LIGHTING, Ero. (Mareh 218, Fotmer ‘Ihaveputasideasheetofpaperforeachofmydifferent heads, and Ihave notgotany discussion note about alternate- current distribution, which Ithought wasabout themost impor- tantthingtotalkabout,therefore Ihavenothingtoanswerunder that head, Ithink itwas General Webber who was the only one who pointed outthat feeders arenotused atthe present time inthe largest central station inLondon, andIhave pointed outinmy paper that that islargely thecause oftheunequal lighting which wohave over London, Mr. Preece hascalled mystatements in question, butIamsorrytosaythatthatmustbeduetohisnot going about sufficiently inLondon toseethese lamps, Ifhedid, ThavenotanydoubtaboutitthathewouldfindthatwhatIsaid inmypaper iscorrect, and that the variation ofpotential over thedistrict lighted from theGrosvenor Gallery central station is enormous atthetimes ofmaximum supply; andhewould rightly ‘come totheconviction that wehave just arrived atuniversally, bygeneral consent, that inanysystem ofdistribution feeders are essential tokeep thepotential uniform. ‘The next question isastohigh-speed engines. Ihave had very fewremarks upon that question too. Mr. Swinburne has backed meupbysaying that inproperly constructed machines high speed isgood, andthat weshould always trytogetashigh aspeed asisconsistent with the safety and efficiency ofthe machinery. Idecline altogether toaccept Mr.Preece’s statement which hemade to-day, viz., that inthis country wehave got definite peripheral velocity ofsomany feet per minute, and that itisallthesame whether you have acomplicated system ofwire coils orasymmetrical mass ofiron which isrevolving. Idecline absolutely toaccept that asamaxim inengineering. The amount ofspeed which itissafe ordesirable touse depends onthe construction ofthe machine, and amachine which is well constructed—which isperfectly symmetrical, and ofsolid parts that arenotlikely tofyasunder—may beran atafar higher speed than amachine with complicated parts which cannot beeasily balanced, and with parts ofthe machinery which willbeliable toinjury from theenormous centrifugal forces which arepresent. 1888) DISCUSSION. am ThenextquestionisthatofcouplinglargemachinestogetherPrter inparallel with continuous current, Now thegeneral experience which hasbeen related here hasbeen that there isnodifficulty incoupling continuous-current machines inparallel. Ishallbe veryglad, indeed, ifthat isthecase: Iamcertainly notpre paredtosayitisnotthecase,butIshallbeverygladtoseeit prored that itis, Ithink that most ofthose who made this assertion, however, have been dealing with comparatively small current, Mr. Crompton says that there isabsolutely noneed of resistance orbanks oflamps toequalise theoutput before switch- inginadynamoinparallel;hesaysthatchildrencandoit,that intheVienna central station itisdone perfectly easily. Istill adhere, however, tothebelief that where this plan hasnotbeen opted inthepast, there isadisagreeable momentary flicker. AttheVienna central station IdonotthinkMr.Crompton hasgetanyverylargecurrent. AsfarasIremember, theoutputs ofhisdynamosaresomewhere between100and200amperes. Mr. R.E,Cnowrrox: 250, Professor G.Foxses: Well, itiswhen you arecoming to 1,000amperes orso,that anyone Ihave heard remark upon it, Satesthatitisessential tohavebanksofresistances. Itis stated inthepaper that for100amperes itiscertainly utterly unecessary, I'amofopinionthat,ifwebadanautomatic meansdiswitchingin,suchashasbeendescribed byoneofthespeakers, itcould bedone quite well, even with the largest current. I balieve that Mr.Andersen himself, though hedidnotmention itinhisremarks, hasdesigned anautomatic switch which is Petfectly successful, and will probably work with thelargest ‘arent. ‘Thefrequency ofalternations hasbeen spoken about byMr. KippandbyMr.Preece. With regard toMr.Kapp's statement, Iamsorryheisnothere.Mr.Kapphasmaintained hereandtlewhere that thesizeofaconverter does notdepend upon thefrequency ofthenumber ofalternations. IwanttoknowWherehegetsthisbelieffrom.Hebringsasaproofof this,thatintaking thefigures Igaveinspeaking about theWesting- houseconverters, Ihadspoken abouttheweight perhorse- a8 SOMEELECTRIO LIGHTING, Bre,[MarchMt, feteerpower,andhefindsinhismuchlargermachines hehasthesame weightperhorse-power. Now,Iwillsimplystatethisfact— thatwhenMr.Kapppointed thatouttomesometimeago,I . explained clearly tohimthat intheweight which Ihadgivenfor the Westinghouse converter, theweight ofacast-iron casing, weighing almostasmuchasthewholeconverteritself,wasincluded. That puts adifferent complexion onthecase. ‘The caving thus becomes very appreciable, and thefact hasbeen admitted by everybody else except Mr.Kapp. Ithasbeen admitted byMr. Preece, ithasbeen admitted byDr.Fleming; itisadmitted by those who have had large experience inmanufacture; Mr. Zipernowski mentioned ittometheother day;IbelievethatMr. Ferranti admits itthoroughly, andIknow thatMr.Westinghouse does. Itwas said byMr.Preece that ifMr.Westinghouse knew what wonderful things arebeing done inEngland, hewould change thisandreduce thespeedofhisalternations. IwilltellMr.Preecealittlefactaboutthis.WhenIwasinAmericas ‘year ago,thequestion ofrunning motors with alternating currents wasoneofvery great importance. Ihad some reasons forthink- ingthat probably areduction inthenumber ofalternations would assist therunning ofmotors, andIasked atthetime ifthey were perfectly sure that they had gotthebest speed ofalter- nations forsatisfactory working. Since that time they have been almost constantly atwork inmaking proper tests oftheir con- verters with icecalorimeters, and they have finally decided, after ‘ayear’s practical working andtesting ofthat sort, that they will notreduce thespeed that they have got—that thehigh speed which they areusing isthebest forconverter work. Generally, ‘astothefrequency ofalternation, about which Mr.Preece has spoken to-night, theconclusion comes tovery much asIsaid in thepaper. Ifyou want high efficiency usehigh speed ofalter- nation, butifyou want towork inparallel itisbest tosacrifice that and reduce thefrequency. Wewere told byMr.Swinburne that hehasdevised ameans forparallel working. That isvery interesting, and Iamsure he willgoonwith it,andwork itout,andgetitintothorough working 1880) DISCUSSION. a order.ButIdonotthinkithasbeenprovedinactualpracticeroimeryet.Tamsurenoonewillbemorepleasedthanmyselftoseeif itispossible toproduce parallel working without theheavy self- induction that isatpresent necessary. Now the important question ofinsulation comes on,and I havedelayed myremarksonthattillnow.Agooddealofvaluable information, Ithink, hasbeen gained upon this, and Iamglad thediscussion has been raised, ifonly toclear the ground a little;Iamonlysorrythatithasbeenclearedsolittle.Webavenotarrivedatanything verymuchmoredefinite,Ithink, than what westarted with, Let metake Mr.Siemens first, and, indoingso,letmeagainthankMr.Siemens forthecourteous vayinwhich hespoke onthesubject;andImustsaythatwhen Iwas raising this question Ihad noidea that Iwas treading |mthetoes ofMessrs, Siemens &Halske; orthatIthought +thtMessrs. Siemens &Halske would notreally have given me theinformation Iobtained, justasreadilyasthosefromwhomIid obtain the information. But atthe same time Ihave "simply stated what Ifound tobethecase inBerlin—at least{,thatIbelieved tobethecaseinBerlin. Mr.Siemens, however,|tastoldusthatageneralbreakdown hasnottakenplace.He |8mthat eight cables were replaced: one was injured bya 1Felaxe; twowere, Ithink hesaid, badly putintothejoint box; jadthefive others arenot definitely explained, except bythe balithat they were duetomechanical injury;andthatinother Places where tests had been made itwas found that the only injury wasthathigh temperature hadmelted theinsulation. ‘Mr,ALEXANDER SieMENS: No; theoutside covering. Ithad otinterrupted theinsulation ;that isjustthething—the outside‘emeringandthecompound weremelted,buttheinsulation was totdamaged. Professor G.Fonses: The insulation wasnotdamaged! 1 ‘annot decide thequestion astowhat isthetrue state ofthecase abouttheseBerlincables.Butletmetakeexception towhatMr.‘Crompton said. Mr.Crompton said,inspeaking ofthis,thatitis hisbeliefthatwhenIaminsuch#positionIameasilygulled.thasbeenwiththeutmost amusement thatIhavefrequently 280 SOMEELECTRIC LIGHTING, Er [Marchtut, i noticed thatheheldthisbeliefwhenhehasbeendescribing to mehislatest electrical hobby, such astheHowell battery. When Jfirstcametojointheengineering profession inLondon,Iwas amused bythe number ofpeople who held that opinion; Mr. Crompton, Ithink, istheonly oneleft,and hestill continues to afford methat great amusement. Iamnotapttotake the firststorywhichistoldbyinterested parties;Iamnotaptto take statements without sifting: Iinvariably trytosiftthe evidence asmuchasIcan,and,ifIcannotgetatitbyactual test, Iconsider very carefully theprobabilities. Ithink wemust agree with Mr.Eddison that lead cables areontheir trial. Let ‘ushope they will come out,successfully. ‘The question astolead-covered cable isvery important, Mr. Preece made agreat point when hesaidthat ithasbeen working well intelegraphy,whenhebroughttheexpertsofthePostOffice here tosupport him inthestatement that lead-covered cables have worked well inmany cases. That isvery good, butitmust beremembered that these lead-covered cables were generally well insulated without thelead;itmustberememberedthatthemodern electric light cables without thewaterproof lead coating would bebadly insulated;itmustberemembered thatinsomecases lead-covered cables have failed when they have passed through made-up ground ordecaying vegetable matter; itmust be remembered that intelegraphic work you arenotworking with 2,000 volts—all these things must beconsidered, andtheopinions thatMr.Preece’sexpertssupported donotbearuponthosepoints. ButMr,Preece hasstaked hisprofessional reputation upon this. Igrant the enormous value ofthat statement, and Icertainly shall beledfrom that statement tomyself carefully reconsider thequestion oflead-covered cables, which Ihadbeen inclined to look onwith disfavour. ‘Astocompound-wound voltmeters: ofcourseDr.Hopkinson wasthefirst inventor, that isperfectly well understood, andother people have only been helping towork the instruments outin8 practical way forspecial purposes. Inconclusion, letmesaytothegentlemenwhohavespoken —let mesaytoDr.Fleming, Mr. Kapp, Mr.Swinburne, Mr. 110] DISCUSSION. 21 Crompton, Mr. Siemens, Mr. Andersen, and also letmesayProfesor toMr,Parker and thehost ofother gentlemen who have not spoken, butwhoought tobementioned inthisconnection—that thereisnopersoninthisorinanyothercountrywhocancompete with me inthe intense admiration Ihave forthe enormous efforts - thatthey aremaking toraise thestandard ofelectrical manufac- turetothe proper standard which itought toattain inthis country, andthat Ilook upon theadvances which arebeing made inthiscountry withthefullestconfidence. Iamasconsciousas anybody canbe,ofwhat isbeing done inthiscountry, andIvalue theefforts ofthese gentlemen asmuch asanyperson intheworld cando. Aheartyvoteofthankswasunanimously accorded toProfessor Farbes forhisveryinteresting andvaluable paper. Aballot fornewmembers took place, atwhich thefollowing ‘were elected :— Foreign Member: Hidesuke Igarashi, M.E. Members : R.S.Erskine, ‘W.M.Shaw. A.L,H.Palmer. Arthur R.Simkins. Aesociates : M.S. Chambers. | Sydney Dobson. Henry T.Thornbury. Student: Charles Oliver Lloyd, ‘Themeeting then adjourned. 382 DEATHOFMB.0.H.B,PATEY,C.D,[MarchHb, ‘The OneHundred andNinetieth Ordinary General Meeting ofthe Institution was held attheInstitution ofCivil Engineers, 25,Great George Street, Westminster, onThursday, March 28th, 1889—Mr. ALEXANDER Stemes, Vice-President, inthe Chair. The minutes ofthe Ordinary General Meeting held on March 21st were read and approved. The names ofcandidates for admission into the Institution were announced andordered tobesuspended. ‘The following transfers were announced ashaving been approved bytheCouncil :-— From the class ofAssociates tothat ofMembers— Samuel S.Dickenson. From the class ofStudents tothat ofAssocintes— Charles Woodward Neele. ‘Adonation totheLibrary was announced ashaving been received since theInst meeting from Mr. Charles Streatfeild James, towhom thethanks ofthe meeting were unanimously voted. ‘TheCuamaan: Ihave nowthevery sadduty toperformof announcingtoyouthedeathofanoldmemberofthisInstitu- tion,theintelligence ofwhich reached usinthe Council Room only afewmoments ago: Mr. C.H.B.Patey, C.B., theThird Secretary ofthePost Office, andwho hadthemanagement ofthe Telegraph Department, died athis residence atBickley at five o'clock this afternoon; and Iwill call upon Mr.CE. Spagnoletti tomove aresolution which the Council desire to submit toyou, Mr.C.E.Sracyourrm: Mr.Chairman and gentlemen,—I risewith feelings ofvery great regret andsadness topropose that our Secretary beauthorised towrite aletter ofcondolence to 10) “VOTE OFCONDOLENCE WITH MRE. PATEY. 283 Mrs,Pateyontheverysadeventofwhichyouhavejustheard.IamsurethatIshallhavetheheartiest supportandsympathyofeverymember oftheInstitution inmaking thisproposition.Mr.Patey,asyouareallaware,wasagentleman inavery prominent. position inthePost Office, being theThird Secretary, andhebad raised himself tothis position, and distinguished himself therein, byhisexcellent business habits, and bythe performance ofhisonerousdutiestothesatisfaction ofeverybody during thetime heheld that position. Hisduties were princi- pollyconfined totheTelegraph Department. Onaccount ofhis excellent business qualifications and hisgreat tact, hewas appointed, in1879, President ofthe International Telegraph Conference then held inthis country, which Idaresay manyof youmay recollect. Ilearn also that hewas most active in ‘urgently pressing forward andbringing about theadoption ofthe sixpenny rate fortelegrams, which hasproved such aboon to thegeneral public. Mr. Patey wasalsoengaged, andmade verygreat reforms, intheIntelligence Department, bywhich thePress ofthiscountry hasbenefited very much, andbywhich allintelligence ofpublic interest isnow sowell circulated over allparts ofthe globe. For hisdistinguished talents he wasmade aCompanion ofthe Bath in1886; and latterly hehasbeen engaged, asIdaresay many ofyou know, as theprincipal negotiator forthepurchase oftheSubmarine Tele- gph Company's property. Mr.Patey wastaken illonSaturday lust,March 23rd, with congestion ofthelungs; hisfriends were hopeful on Monday that hishealth would improve, oving tofavourable symptoms; but, however, arelapse ubappily occurred, andhepassed away atfiveo'clock to-day. Itisvery sadtothink ofaman ofhisage—only alittle over foty—being cutoffintheprime ofhiscareer, when hewas doing such excellent work and making himself s0useful'to his fellow-creatures, and Iamquite sure that you will allagree ith methat itisour bounden duty torecord thedeep regret which weallfeel onthe occasion. Itherefore beg to more—“That theSecretary beinstructed tocommunicate toMrs. “Patey theexpression ofthedeep regret feltbytheCouncil and ey LABORATORY NOTESON (DMarch2004, “members ofthisInstitution atthedecease ofMr.Patey, andof «their sincere sympathy with herinherbereavement.” Professor W.Gryuts Apams, F.R.S.: Ibeg tosecond the proposition which hasbeen made byMr.Spagnoletti. ‘Themotion wasunanimously carried. ‘The Cuamaan: The first business ofthe meeting isthe presentation ofthe Balance-Sheet forthe year 1888, acopy ofwhich, having been sent toevery member inEngland, itwill notbenecessary toreadatthismeeting; Itherefore callupon anymember whomay have remarks tomake orquestion toask inreference tothe accounts todosonow. ‘After apause, ‘TheCHAmMAN: Asnooneappearsdesirousofmakingany comment, Ibegtomove—“That theBalance-Sheet andStatement “ofAccounts fortheyear ending 3lst December, 1888,asnow «presented, bereceived and adopted.”* ‘Mr, Lavoxerr seconded the motion, which was carried unanimously. ‘The following paper wasthen read :— LABORATORY NOTES ON ALTERNATE-CURRENT CIRCUITS. ByProfessors W.E.Avnrox, F.RS., V.P., and Joux Penny, F.R.S., Member. Atthelastmeeting Professor Perry andIreceived anurgent request from your indefatigable Secretary toread apaper atthis meeting, asunforeseen circumstances had prevented theauthors oftwootherpapersfromsendingtheminintime.Wepointed ‘outthat wehadnopaper ready;wewere,however,urgedtodo something, anditoccurred tousthatpossibly some short account: ofafewoftheexperiments that some ofthestudents ofthe Central Institution areatpresent engaged upon might beof interest tothe members. You areofcourse allaware that, when youaredealing with *ForBalance-Sheet ovepage8234, 1188 ALTERNATE. CURRENT CIROUITS, 385 varying currents, self-induction isofimportance;indeed,ifthe alternations are only sufficiently rapid, self-induetion completely replaces resistance, theresistance becomes entirely unimportant, andselfinduction isall-important. Inthe caso, forinstance, of awell-made transformer, the resistance ofthe primary circuit maybepractically neglected altogether, and allthat you have to consider inorder todetermine what current willflowthrough the Primary cireuit foragiven mean potential difference atthe terminals, isthe effective coefficient ofself-induction. In spite ofthe importance that self-induction possesses, and in ‘spiteofthefactthat thelargest distributions ofelectric power in thisandinothercountries arecarriedoutbymeansofalternating curents, there isawonderfulamountofignorancepossessedby tall astowhat isthe coefficient ofself-induction ofany puticular circuit. We have, infact, noinstinctive feeling what- everastothevalueoftheself-induction ofacoiloracircuitwhen veseeit. Some time ago wewere asked byawell-known Piysicist whether wecould tell him what was the self-induction olaThomson reflecting galvanometer oftheordinaryform,havingabout 7,000 ohms resistance. Hesaid, “Iamnotparticular to “50percent.,oreven100percent.,butcanyougivemeany“sortofideaastowhatitis?Isit,forexample,tobemeasured “incentimetres orinmiles? andisitalarge number ofmiles or “asnall number ofcentimetres ?”Well, that impressed upon us lowlittlewasanyinstinctive feelingastothemagnitude ofthecoefficient ofself-induction ofanyparticular coilcalled into ‘aistence bytheappearance ofthecoil. Asanexample, here is»coil wound sothat you can seethenumber oflayers, aswell asthenumber ofconvolu- tions ineach layer. The wire isthick, 90that thetotal tumber ofconvolutions.can beeasily counted, and isseen tobe 0, Now, although there isnoiron inthiscore, andalthough tkewindings arepractically symmetrical, wethink that, probably 1wsinglepersoninthisroomcansaywhatiseventheapproximate ‘ulueofiteself-induction, Were wetotellyouthat theresist ‘nceofthiscoilwere amegohm oramicrohm, youwould smile, ‘becauseyourexperience tellsyouthattheresistance mustbeof 286 LABORATORY NOTESON (March260, the order ofafew chms, Asamatter.of fact itis1°34 ohm. Possibly you could not sayfrom merely looking atthe coil whether itsresistance was half anohm or5ohms, butifthere benodiscontinuity norshort-circuit youareperfectly certainthat‘amegohmisfarabovethetruevalue,andamicrobm farbelow.Butifweweretotellyouthatitsself-induction wasamillion miles, notasingle brow would beraised inastonishment ;orifwe were tostate that itwasabout 20yards, you would receive that statement with perfect equanimity ;orevenifweweretosaythat itisabout 100miles, youwould notsmile,—well, you ought not tosmile inthat case, foritisabout 100miles, being inreality 92 miles. Now, how have wegotthis clear notion with reference to resistance? Obviously byalarge numberofmeasurements having been made, byourhaving hadtofindoutwhat wastheresistance ofallsorts oflengths ofwires and ofcoils ofvarious sizes and shapes, andtherefore weknow, from asort ofinstinct which has grown upastheresult ofalarge number ofmeasurements, what isroughlytheresistance ofanyparticular wirewhenweseeit.It therefore seemed tousthat itmightbeofsomeinteresttoyou tohave theresults that have been obtained byvarious students astowhat are the actual coefficients ofself-induction ofsome simple circuits. The experiments have been made byvarious groups ofstudents under thecharge ofMr.Sumpner;infact,forsometimepast, scarcely acoil hasbeen allowed toremain peaceably atrest inour laboratory, but ithas been operated on,and itscoefficient of self-induction measured. Henceourstudentsarejustbeginningtoacquire that instinctive feeling ofwhich wehave spoken, regarding themagnitude oftheself-induction ofdifferent circuits, anditisourdesiretogoalittlewaythiseveningtowardsimpart~ingthat feeling toyou. First,ofcourse,wemustbefamiliarwiththepractical unitof self-induction. Weshould like tobeable tosaythat ofcourse ‘you areallfamiliar with it,butinview ofrecent experience we fearthatthatwouldbetoogreatanassumption; foronlyquite recently, atameeting oftheInstitution ofCivil Engineers, a 1190) ALTERNATE. CURRENT CIRCUITS. 2st very large mistake was made byawell-known electrical engineer regarding themagnitude oftheunit ofself-induction—a mistake which,wepointedoutatthetime,wasverylikethatofconfusingthedistance between here and Charing Cross, which isabout a mile,withthedistance fromheretothesun,which isabout 100,000,000 miles. The practical unit ofself-induction is 99,77 x10°centimetres;itisnotexactlytheearth'squadrant, inconsequence ofthelegal ohm notbeing exactly the intended ortrueohm, andtherefore, ifyouwant theequation ao rO+LGene tobetrue where risinohms, 0inamperes, and Einvolts, must beexpressed inaunit which is99,777 x10*centimetres, about 6,200 miles. Forwant ofabetter name, wesuggested some fewyears ago that the name “secohm” should beemployed forthis unit, it being ofcourse asecond xanohm, sothat asecohm isreally 99,777 x10*centimetres, orabout 6,200 miles, ortwo-hundredths ofasecohm israther morethan300miles. When making that suggestion, weshowed you anarrange- ment, which wecalled asecohmmeter, forenabling thecoefficients ofself-induction tobemeasured withthesamefacilitythatyou have been accustomed tomeasure anordinary resistance with a Wheatstone’s bridge. Ofcourse, self-induction only shows its effect when currents arevaried. You cannot, ofcourse, measure itbysteady currents; but byperforming certain operations which were explained toyou some two years ago, youareable to make themeasurements with alternating currents orvarying currents nearly aseasily asyou are accustomed tomeasure resistances bytheuseofsteady currents. | Inthesecohmmeter thatwedevisedtheoperationthatwasper- formed was this: the battery circuit was alternately made and broken,andduringoneoftheseoperations—say, duringthemake ofthebattery circuit—the galvanometer wasrendered inoperative bybeing short-cireuited, sothat during every break ofthat battery circuit thebattery wasoperative, while during every make thegulvanometer ceased tobeoperative. Inthat waytheeffect 208 LABORATORY NOTESON (tare230, ofself-induction was apparently toincrease the resistance of particular circuit byadefinite amount, depending onthespeed, andfromtheapparent increaseofresistance andthespeedthe coefficient ofself-induction was determined. Since then wehave made various alterations inthe instrument. One improvement ‘wasavery obvious one—to reverse thebattery instead ofmaking andbreaking thebatterycircuit;and,lastly,toreversethe galvanometer instead ofshort-circuiting it;sothat, atpresent, with thelatest form ofthesecohmmeter wealternately reverse the battery circuit andthegalvanometer circuit, and, asfour reversals ofeachcircuit occurforeachrotation ofthecommutator spindle, thesensibility ofthelatest form iseight times asgrest ‘asthefirstformthatwehadthehonourofbringing beforethe Society twoyears ago. ‘The commutators canbedriven atoneorother oftwo speeds relatively tothat ofthedriving handle, With one arrangement there arerather more than eight reversals ofboth the galvan- ‘ometer and ofthebattery foronerevolution ofthehandle,and with theother twenty-four reversals ofeach foronerevolution of thehandle. The apparatus is60constructed that thespeedof thefly-wheel remains thesame relatively tothat ofthehandle, whatever speed ratio beemployed, andhence thesame uniformity ofspeedcanbeobtainedwitheitherspeedratio.‘Thesecohm-meter can beconveniently driven byhand soastoobtain constant speed ofreversal varying from 300to6,000 reversals per minute ofboth thegalvanometer andthebattery. Toshift from one speed ratio totheother, press down the endofthelocking lever attheright ofthesecohmmeter, and slightlypushinorpulloutthehandle,turningitslightlyto assist thetoothed wheels engaging properly; when engaged,letgo theend ofthelocking lever. With theoriginal form oftheinstrument weproposed touse8* speed indicator attached totherotating spindle, butwepointed outthat insome cases itwould bedesirable todispense with the speed indicator, especially inview ofthedifficulty thatexists in gettingagoodcheapspeed-indicator; andsomostofthemeasure-mentsthatwehavetobringbeforeyouto-nighthavenotbeen 1889.) ALTERNATE-OURRENT CIRCUITS. 289 made inaccordance with thefirstmethod. Weproceeded, notby comparing thecoefficient ofself-induction interms ofaresistance andatime, butbycomparing onecoefficient ofself-induction with another. You know,ofcourse,ifyouhaveaWheatstone’s bridge (Fig. 1)with twowires possessing certain resistances 7,,ryand NCO) Leh Se H } Fig1 coefficients ofself-induction Z,and L,,and two other wires Posessing resistances 7,and1,butnoself-induction, thatifyou firstbalance forsteady currents, that is,satisfy theequation te youwillalsohave balance forvarying currents ifinaddition you make hon - i ‘The second equation canbefulfilled without disturbing theresistances, thatis,withoutalteringtheequalityoftheresistance ratios, ifoneofthearms contains anapparatus ofadjustable self- induction; and further, if[the value ofself-induction ofthis apparatus can beread offinsecohms foreach position ofits 220 LABORATORY NOTESON (atareh180, adjustment, weeanatonceusethesecohmmeter forthemeasut-ment ofaselfinduction without theuseofaspeed indicator. InFig. 1,BCisthecommutator forperiodically reversing thebattery connections, andGCthat forperiodically reversing the galvanometer connections, thethick lines inthefigure repre-sentingpermanent connections inthesecohmmeter itself,and thedotted lines connections temporarily made outside it.When making such scomparison between anunknown and aknow coefiicient ofself-induction, thespeed atwhich the secohmmeter handle, H,isdriven need notbeknown, butthegreater thespeed the more sensitive the test the rate ofreversal must not, how ever, betoogreat forthe currents toreach their steady value: between two consecutive reversals, The variable standard of self-induetion that we have been employing isamodification ofthatusedbyProfessor Hughesand byLord Rayleigh, andconsists oftwocoils whose planes may be made tohave anyangle with oneanother, The speciality ofour apparatus isthat wehave determined theexact value insecohm: ofthe selfinduction ofthe arrangement forvarious positions of the coils relatively toone another. The values are recorded on thedialfixed tothelarger coil, andover which moves thepointer attached tothe smaller coil. The instrument thus constitutesa direct-reading variable standard ofself-induction. The wire onboth coils isofplatinoid, sothat thevariation of resistance bytemperature ispractically negligible. ‘The smaller and movable bobbin, which wecallA,iswound with 2994 turns; thelarger stationary bobbin with two coils, one of49tums, which wewill callB,andtheother of147turns, called C.Band Cmay beused separately ortogether, 60astohelp one another,or80a8toopposeoneanother, Thereare,therefore, foureom-binations, Aand B,Aand B—C, Aand C,Aand B+, and with each ofthese arrangements thesmaller coilmay beturned round through 180°. ‘The range oftheinstrument is— A&B{erinction naseFOO18St0.0-0175secobm, A,B-C ” 00157 ,,00229 ,, 4,0 » 00171,, 00283, 4,,B+C » 00215 ,,00365 ,, 1980) ALTERNATE. CURRENT CIRCUITS, 2 thetotal adjustable range isthus from 0°01356 to0-0365 secohm, orfromabout.84milesto226miles.Itwillbeobservedthatthe ‘various component ranges overlap oneanother fully,sothatmany ofthevalues canbeproduced with twoarrangements, which is courenient forthepurpose ofchecking theaccuracy ofameasure- ment. The smallest coefficient ofself-induction that can be cblained with theparticular standard isbytheuseofcoilB alone, which corresponds with 0-00096 secohm, The apparatus therefore constitutes adirect-reading secohm standard. Although the coefficient ofself-induction ofthe standard canonly bemade tohave successively progressive values from0°0135to0.0365secohm,coefficients ofself-induction whichdonotliebetween this limit can bemeasured: forexample, suppose theunknown self-induction Z,ofacoil ofresistance 7, chmsbeabout0°5secohm,thenweaddaresistance rtor,,sothat noninRAT ==20,soy, inwhich case balance forvarying currents will beobtained when 1,=201,5 andsince 20Z,varies between 0-27 and0°73 secohm, these limits indude 0-5secohm, and therefore balance can beobtained on ‘lating thesecohmmeter. ‘Thefollowing aresome oftheresults that thestudents, under theguidance ofMr.Sumpner, have obtained, using thesecohm- eter andtheadjustable secohm standard just described :— Coilwound with copper wire 34mills. indiameter, covered to about 48mills, consisting of48layers with 13convolutions ineach layer, wound onawooden core 2inches thick and4 |incheslong.L=00147secohm,1=1-34ohm,2=0011second; | that is,the “time constant” ofthis coil is0-011 second, |which isthetimerequired foranycurrent torisetoi}0-06321 ofitsmaximum valueafteritisstarted inthecoil'byanapplication ofafixedP.D,between theendsofthecoil. ‘VOL, XVIII, 21 2 LABORATORY NOTES ON (are 28, SinglecoilsofanordinaryMorsereceiver,ironcoreabout031 inch thick and3inches long, Outside diameter ofcoil0% inch, L=00936 secohm, = 32ohms. Another single coil ofanordinary Morse receiver: iron cow about 0°31 inch thick and 3inches long, outside diameter of coil 1°25 inches, 1=0-444seoohm,r=50chms,%=00089second ‘The resistance ofthis second coilisonly §that ofthe former, butitsself-induction isover four times asgreat. The two complete coils ofaMorse receiver inserie, with imasole-plateandarmature—the armature,however,beingmther asmall one, 1=0265,r=14ohms,%=0018second. Coil ofalow-reading magnifying spring voltmeter wound with copper wire onabrass bobbin, noiron inside. Length of coil 2°88 inches, external diameter 3inches, diameter of brass bobbin onwhich wire iswound 0-38 inch. L=1-462secohm, r=3335ohm,2=0-0044send. SinglecoilofanordinaryThomson's reflectinggalvanometer. TL=2:56secohms, r=2,700ohms,Fa=0-0007. Single coilofavery high resistance Thomson's reflecting gal- ‘vanometer, thecoilbeing ofabout theordinary size. L=70secohms, r=100,000ohms,z=0001. Comparing these twolastresults,weseethattheratioof the selfinduction isabout the same asthe ratio ofthe resistances, which iscorrect,aseachratioisproportional tothesquareofthenumberofturnsofwirewhentheevilsare ofthesame size and shape, Hence the time constantsare about the same. An ordinary Ruhmkorff induction coil intended togive a 2-inch spark. Secondary coil, I= 51-2 secohms, r=5,700 obms, + coeficient ofmutual induction, 0-46 secobm. 1w0] ALTERNATE-CURRENT CIRCUITS. 208 Ferranti dynamo intended togive 200volts and40amperes. Armature, L=0-013 secohm foronephase, =00011 ,, foranother phase, when nocurrent iapassing through thefield magnets. Field Magnets inseriesL=0°61secohmforasmallexcitation, 1=3ohms, Mather &Platt shunt dynamo, intended togive 100 volts and35amperes. Armature from brush tobrush, EL=0-005 secohm, r=0-215 ohm. Field. Magnets inseries, E=136 secohms forasmall excitation, r= 44ohms. ‘When there isiron inacoil the value ofthe self-induetion depends,ofcourse,onthestrengthofthecurrentemployedwhen making theobservation, andonthewayinwhich this current is varied fordetermining theself-induction. ‘This will befound veryfully worked out, with numerous experimental illustrations, inapaper byMr.Sumpner, “OntheVariation oftheCoefficients “ofInduction” (Proc. Phys. Soc., vol.ix,part iii.,forJuly, 1888), ‘The following areillustrations ofthe sort ofresults obtained :— Experiments ontheSelf-Induction of«Kapp and Snell 2-H.P, Transformer. ‘Through oneofthecircuits ofthetransformer ofapproximate resistance 0-07 ohm acurrent waspassed, andwhich wasmade to haresuccessively thevalues shown onthetable. Onchanging its value great care wastaken that thechange was steadily inone |diction without any fluctuations. “The self-induction ofthe |cher circuit ofthetransformer ofapproximate resistance 0-14 chmwasmeasured intwo distinct ways. With thefirst method thesecohmmeter was employed torapidly alternate asmall |arent of0-01 ampere through thiscircuit, while with thesecond method theself-induction ofthiscircuit wasmeasured bystarting |asmallcurrentof0-037amperethrough thiscirouit. Afterstopping thiscurrent theironwasmade togothrough thecomplete mag- etic cycle before making thenext observation oreven repeating theformer, Forinstance, supposing theprevious measurement 2 LABORATORY NOTESON (Maren283, wasmadebystartingacurrentof0-037amperethroughtheone circuit when thesteady current through theother circuit was —2amperes, then onstopping thecurrent of0-037 ampere thecurrent through theother circuit wasfirstincreased upto +10amperes, then diminished to—10amperes, andlastly increased upto—1ampere, forwhich value thenext observation vwas made, Currents steadily Increasing from Minus toPlus. ysletndactonferSeainoo] ‘SteadyCurrentinone|2thotherCirle,|measuredSywastingcarat ERE Cee -* 0010 008-6 o oo -4 o oss -2 o 0100 -1 0136 ° 0-090 ois1 om | +2 0-200 +4 o 0-084 +6 o ons +8 0-010 oor Ourrents steadily Diminishing from Plus toMinus. Satndocion ofSteadyCurrentinonehice|mcuaredbywari ‘Sreutsinameres”|RSShame” |homenstn Scrcie| +8 ~ 010 0-008 +6 - oon a4 oo 0-028 +t a ome |° 0-000 ome|-1 os 0-264 jon . om || o=6 ~ coe j |ce : coe |;oc oo10 0-020 1m) ALTERNATE. CURRENT CIROUITS, co Ashasbeen already pointed out, thepresence ofiron ina cireuit causes the self-induction ofthat circuit tohave very different values; and, further, itcauses the value ofthe self- induction todepend ontherate onwhich anyparticular change inthemagnetic induction iseffected. Forexample, ifthecurrent through one cirouit ofthe transformer bekept quite constant, and the secohmmeter be used to alternate adefinite small ‘carent through theother cirenit, thevalue ofthecyclical self- induction diminishes somewhat asthe speed ofalternation is increased, When wehadthehonour offirstbringing thesecohmmeter to your notice in1887, weexhibited certain curves (Fig. 3,page 307,Jour. Soc. Tel.Engrs, part 67,vol.xvi., 1887) showing the apparentincreaseofresistance ofacircuitcontaining noironwhenthesecohmmeter wastuned atdifferent speeds, andcertain other curves(Fig.4,page310)fortheapparent increase ofresistance ‘hen an iron core was inserted inthe coil. The first set of carves arestraight lines, while thesecond show adistinct curva~ tare, This curvature we attributed atthe time tothe fact that withhigh speeds ofthesecohmmeter thetime constant ofthe cireuit when theiron core was inserted was too large forthe current toreach itssteady value when thebattery circuit was closed, orforthecurrent todieaway when itwasopened. But. Mr, Sumpner’s subsequently attacking this subject both ‘experimentally and theoretically, hefound that theincrease ofthe time constant produced bythe insertion oftheiron was not saficient toexplain thelarge amount ofbending ofthecurves. Hence weconcluded atthebeginning of1888 that thebending ¥asmainly due toamagnetic lag. Atabout thesame time we sbtained evidence that when thespeed ofthesecohmmeter was verygreat, and thelead very small, theinsertion oftheiron core ‘in@coilactually diminished theself-induction ofthecircuit—a reault that wefeared atthe time must bedue tosome defect in thecoil,butwhich therecent investigations ofDr.Lodge now thowtobequite possible. This subject weareatpresent investi- gating further. Sometimeagowedrewattentiontothefactthattheaccuracy ofthemeasurement bymeans ofawattmeter ofthepower given 296 LABORATORY NOTESON [Mare28, byanalternate current toacirouit possessing self-induction depended ontherelative values ofthetime constant ofthefine- wire circuit ofthe wattmeter and the time constant of the cireuit thepower given towhich itwasdesired tomeasure. Our result wascriticised byMr.Blathy intheElectricianfor April 6th, 1888, who made certain calculations regarding theuse ofthewattmeter constructed byMessrs. Ganz &Co,tending to showthatthecorrection pointedoutbyourselves wasanunimpor- tant one, Wetake the opportunity ofdrawing Mr. Blathy’s attention to the fact that in his calculations he has under- estimated thevalue oftheself-induction ofthesuspended coil, and—what isequally important—neglected altogether theself- induction oftheso-called non-inductive portion ofthe fine-wire circuit. Taking theactual specimen ofthewattmeter wepossess,wefindthat thesuspended coilofresistance 2ohms hasaself- induction 0-00042 secohm. This isabout twice what Mr.Blathy allows forit. Further, two ofour students—Messrs. Lamb and E,W.Smith—on making acalculation astotheself-induetion of thevery carefully doubly wound resistance coilofsome 980ohms which isplaced inthe fine-wire circuit, find that itis0-00025 secohm, which ismore than half thevalue forthe inductive col. And if,inthehypothetical case taken byMr.Blathy,adoubly ‘wound resistance coilof100,000 ohms beemployed, made ofthe same wire asisused inthe coil of980 ohms—and this would be necessary, since, the suspended coil remaining the same, the current-density inthehigh-resistance cireuit must remain con- stant—they findthat theselfinduction ofthedoubly wound resistance will be0-025 secohm;s0thattheself-induction ofthe high-resistance circuit will beabout one hundred times asgreat asissupposed byMr.Blathy. Forthepurpose ofenabling practical men tocalculate the self-induction ofdoubly wound coils made ofconducting wireof diameter d,covered toadiameter D,and placed parallel toone another, with their insulating coverings incontact, wegive the following formula :— 422Xlogan?+173 Z,insecohms, =———y>y"—— Xtotallengthofgoing andreturn wire, inyards. 1888) ALTERNATE-CURRENT CIRCUITS. 207 ASaunt TRANsronMER. Thefollowing problem presented itself toour student Mr. Smith, whose name wehave just mentioned;andashissolution depends ontheIngproduced byself-induction, wehave thought thatthismay beafitting opportunity forbringing thematter forward. The problem isasfollows:—If acertain alternate Potential difference bemaintained between two mains, MM (Fig. 2),between which there aretwoincandescent lamps, Pand iisitpossible tomake thesumofthemean potential differences ee P ~ =>) y, rT Fro. 2. ‘maintained attheterminals ofthe lamps greater than themean Ptential difference maintained between the mains without sconnecting the lamps from the mains and inserting an crdinary transformer? For example, suppose that the mean Wetential difference between the lamps be100 volts, and it quires amean potential difference of55volts tobemaintained between theterminals ofeach lamp tomake itglow properly, ‘anthisresult beattained without disconnecting thelamps from themains ? Let1111 (Fig. 3)represent thecurve ofimpressed potential difference: then thismay beregarded asbeing composed oftwo : , | coincident curves, 12,12,theordinates ofwhich areeach exactly halfthecorresponding ordinates oftheoriginalcurvel111. Now letitbepossible toretard oneofthesmaller curves andaccelerate theother sothat they take some position like 3333 and 4444 (Fig.4):then,sincethesumofthecorresponding ordinates ofthese A twocurves isequal tothecorresponding ordinates oftheoriginal curve, itfollows that themean value oftheordinates ofeach of these curves must begreater than half themean value ofthe ordinates oftheoriginal curve. Mr.Smith's view was, ifoneof thelamps, P,were shunted with ahigh resistance possessing much self-induction (Fig. 5),andtheother lamp, Q,with anon-inductive high resistance, thewave ofpotential difference attheterminals of Pwould beaccelerated, while that attheterminals ofQwould be retarded, asshown inFig. 4;hence theresult would bethat the brightness ofthetwolamps would beincreased bytheapplication ofthehigh-resistance inductive shunt and the high-resistance 2) ALTERNATE-CURBENT OMOUITS 1 non-inductive shunt—a result which experiment shows tobe correct. . — @>QS k8——>) > 7)0 ee fr Fw. & Itwill probably beobjected tothis solution that itcannot be employed without introducing alarge waste ofpower, andthere- forecanhave nopractical value. Butexactly thesame objection wasraised some years ago tothe employment ofthe ordinary transformer. And just asthe practical importance ofusing transformers has ledtosuch improvements being made inthem thatthewaste ofpower introduced bytheir useisfarmore than compensated forbytheeconomy they introduce into electric dis~ tribution, sointhesame way isitnotpossible that Mr.Smith’s solution maybesoimproved thatthegain inpower arising from vorking incandescent lamps attheir best efficiency may more ‘thancompensate forthelossofpowerinhis“shunttransformer” ? Osrae Errecr orSexr-Inpuction 1xDesrroyina Ripies. Inthe discussion onMr. Kapp’s paper, and onformer ‘occasions, wehave described tothe Society the effect ofself- ‘induction incausing thecurrent inanalternating-current circuit wo LABORATORY NOTESOX [Btarch28th, tobecome more and more asine function of the time. We assumed, ofcourse, that there was afixed self-induction ofthe cireuit, and wewere not considering the possibility ofthe presence ofiron making theself-induction vary, asweknow that itdoes. We have stated that theminor ripples orharmonic: never wholly disappear. The following isthe mathematical investigation ofthis problem :— ‘The electro-motive force being anyperiodic function, itmaybe expressed intheshape— _ ©nity E=K+>aesin(2et«)we(I) and as (2)wnB=70+E99,wehavefor0 (3) - ow . . onay>eydBein.(2171—e—tann27L4), ; /v+ Peain(poten tanctAES), 1 Wemay forany alternating current from anyexisting machine which isnot inseries with anyconstant-current machine assume that Z,=0,(Itisofcourse obvious that asselfinduction increases allthevariable part of@diminishes, leaving any constant part2ofmoreandmoreimportance, sothat,for example, self-induction inthe circuit ofany approximately constant current machine tends tomake the current more constant, diminishing theripples indefinitely asLgets greater and greater.) If,then, E,=0,and ifweusea,todenote a qeeh, Jee ee uw aril orBa/ioes, thenthevaluesofaa»oy&.,forvariousvaluesof7aregiven inthefollowing table — 1180) ALTERNATE-CURRENT CIROUITS. so f | ee ° wep ae) | ante wa}oman|orat|ago|amosa]eete|caete|aso|amosete|coset|arte|asete|toons v0|16|orents|ossx22|const wo|oie|coro’|0035.|-oo1on%s Aglance atthis table shows the way inwhich self-induction causes theovertones orripples todiminish relatively tothe fundamental tone. Forthepurpose ofseeing this more clearly, wewill nowassumethatineverycasethecoefficient ofthefirsttermis1;so that,calling . Ofv+ ink YF bythesymbol«, o 1+anoaN Pm, 9tyne=D>a SoreosatiOa a1+re (4) ofwhich, ofcourse, theamplitude ofthefirst term is1. Ifwedenote theamplitudes ofthevarious terms byBi,BnAy &e,,wehaveforthefollowing valuesof7“thefollowing values ofthese amplitudes; weshall denote a,-+«4bythesymbol y4:— 0} a n n ” naw]1|xn|oan|on|ay a 1 ‘TAL yy 49 7, Men 385 7,2]1|em fang|say|ayro|1|smo,|stsy,|ay,|0755,100 1"50007, 8337, 25007, "20007,1000 1|30007,|338%,|2600,|2000, wa LABORATORY NOTESON (atareh38, Weseethat,however great7maybe,theamplitudes ofthe overtones relatively tothe fundamental cannot bereduced more than from: Lom ®%%Seywhen yo=0to L Ldndm2m$nheywhenye=oo,Hence,iftheE.M.F.doesnotfollowtrulyasinefunctionof thetime, thecurrent, although itmay bemade more nearlys sine function ofthetime, cannot truly become so,however great theself-induction may be. Example I. Letustake what isprobably themost discontinuous periodic function which wecanimagine asoccurring inpractice. That is, A AE ve |.hs~heNSxls iVA 6| Fie, 6 letusimagine theE.M.F. attime 0tohave suddenly increased from —Vto+V, attimet=FitosuddenlydiminishfromVto—V,attime¢=7tosuddenly increasefrom—Vto+V, andsoon,remainingconstantexceptattimes0,1,1,9.7,21,de. Such avariation ofE.M.F. isshown byACDFG(Fig. 6). 1m) ALTERNATE-CURRENT CIROUITS. 808 Itiseasy bytheuseofFourier’s theorem toshow that the EMF. satisfies the law . Ln814Lyin,1 BaAE(einee+§sinTt+beinGFt+be.)(5) ‘othat the current satisfies the law (6) ad“>11on(2Aai2rily. cat Tuten (tem T) 1 re carebeing taken that iisanoddnumber. Theactual value ofVdoes notaffect thenature ofourresults, wowehave takensuchthat4%=1, And if(6)iswritten intheshape— AQin C=SasinP yea)we thefollowing table shows thevalues ofa,(the amplitude) and (the lag)ofthefundamentalandovertonesforvariousvalues at,Tr L 1.1. 02|099 72|ou 19|)167 220|107 420|ose|202|]153|20a|!061|7204|081|70-22||0625|5195|0867501|os16|eo|0163839510|o16|aio|ois|8609||0006|eae2|!096|e908100|oo16|8905|-corss|800-7||006s|8908|0006|89091900|coors]0»|one|0»|00088}so»|200026)20° 4samatter offact, wehave found itnecessary tocompute talues ofaand ¢asfarasai,and ¢,forsome ofthe smaller values ot7,butitinotnecessary togiveallsuchresults Itishereobservable that,however great#maybe,the amplitude ofthefirst harmonic cannot beless than 1-9th ofthe fundamental sine function, northesecond 1-25th, northethird Pv LABORATORY NOTESON (atareh28, 149th, We have plotted inFig. 6curves showing theway in Whichthecurrent varieswhenthevaluesof7”are0,02,1, “2,1, The distance BHTrepresents 7,theperiodic time. When7=0weknowthatthebrokenlineABC.D.EP shows how thecurrent ought tovary. ‘The dotted lineA’B’ D’FE’F’@shows the result ofanattempt tocalculate the currentwhencs=0fromtheformula,using,ifweremember rightly, axsand és Whenea=‘02wefoundthatitwasnecessarytocalculate aayand ¢before wefelt justified inneglecting higher terms. The current curve isshown inA"B” C"D"E"P'G", When7/;=1thecurrentcurve,A”BO”D’”Bas7 departed further from thebroken straight line shape. When7%=2wehave4’”B"".. ..G””. When7%=1-0wehave4,B,D,E,Ge + Tewill be found thatasgetsgreaterandgreaterthe current tends more and more tofollow thelawindicated bythe curve—or, rather, straight line—shown inFig. 7;andthisisthe Fw.7, nearest approach toasinefunctionwhichthecurrentcanattain towith such adiscontinuous E.M.F. aswehave taken, Infact, inthelimit, @tends tofollow thelaw Qn 1 6m 1 10” —Coecoe.Et+5008.I+Feconnet+Ge.(8) 18) ALTERNATE -CURRENT CIRCUITS, os Ezample I. Wegive another rather discontinuous periodic function asan example. Weassume that theE.MF. varies inthewayshown onFig.75butalthough this satisfies alawsuch asisshown in (8),Weprefertolettimebemeasured from0’,andthenwhenVisthegreatest+or—valueoftheelectro-motive force, 8 _29 1. 6m 1.) 107E=5V fein.2—5cint+35Sipt =PoinEtt hehwe (8) Hence - 1 (dim a2inhon8h>eaoiebesin.("prt—tan.a), 1 PF ‘being odd, andtheterms being alternately +and —. Foreaseofcalculation, takingay=1,andwriting C=Sasin.(257-«)s wrehave the following values ofa,and ¢,forvarious values otEeTr =ITOoT Ell d<l-l-) <1. oO1oO|un 00400 0|208 o2)9|roa|toe|a1|sos|ze|)rss|azeafae|s22|ost|coor|oe|rae|)oot|rr02-2|}625|ses||0285|75°1|)0068|sie9|0023|880510]as|sieo|-os1s|sero|)oo1ss|sse2|const|ase1w0|ore|sos||coos|aoe|)cons|asee|ons|o0+9po|ors]so»||come}2»|ons}ate|anon92 Wehave drawn thecurves representing Cforthevalues of &=0,0-01,and0-1inFig.8,thedistance AErepresenting 7,theperiodic time. 08 LABORATORY NOTESOX {afareh288, ABODE showscurrentwhen7-=0; Be L="l;ABCDE mo» »pads ABODE yy peek ‘Theordinates ofthislastcurve(for77=1)havebeenmagnified ten times, tis evident that even when Eissuch avery discontinuous 2 y Q\e A : o Ze : L/ 4 Fo. function asthis, amoderate amount ofself-induction makes the current what isvery nearly apure sine function ofthetime. Itis,however, obviousthat,however great-maybe,itis impossible tohave theovertones smaller, inproportion tothe fundamental, than wehave in _4avr Qe, 1 br, 41 107 0=STfeo.Fyre—Jycon.Ft+angcos,1¢—be Execrric Frequency Meter. Some time agooneofussuggested, during adiscussionoa transformers, thatamethod might beused foranalysing analter nate-current wave which somewhat resembled theemployment of Helmholtz’s resonators foranalysing acomplex sound vibration ws) ALTERNATE-CUBRENT CIRCUITS, sor Themethod then proposed consisted insending through oneof thecoils ofadynamometer the current tobeanalysed, and through theother analternate current ofknown frequency, and varying thefrequency ofthetestcurrent until anattraction was abverved between thetwo coils. The smallest frequency ofthe testcurrent that willproduce such anattraction gives thefunda- nental rate ofalternation ofthecurrent which isbeing analysed, aadthemagnitude oftheattraction betweenthetwocoilsofthe dyamometer the amplitude ofthefundamental rate ofalter~ tution, The frequency ofthe test current isthen gradually increased until anattraction isagain observed between thedyna— ‘mometer coils, from which the frequency and amplitude ofthe finthigher rate ofalternation isdetermined, andsoonforallthe ‘component vibrations. This method has been used with acertain amountofsuccess byMessrs. Lamb and E.W. Smith atthe Central Institution for theanalysis ofthecurrent produced bytheFerranti machine. Previousexperiments hadshownusthatthiscurrentwasnot simple sine function ofthetime, asthecurve rises less rapidly andfalls less rapidly than atrue sine curve. The difference, how- ever,between thetwoisnotvery great, and theunimportance of ‘theharmonics compared with thefundamental rate ofvibration inthecurrent curve oftheFerranti machine renders theanalysis afthiscurrent severetestofthemethod,which,however,was nuflciently sensitive toshow that the Ferranti current wave con- sistsofafundamental rate ofalternation, noappreciable com- ponents either fortwice orthree times the fundamental rate of alternation, butsensible components having respectively fourand fivetimes therate ofalternation ofthefundamental; that is,the alternation ismade upofafundamental and thethird and fourth harmonic. Quite recently two ofour students—Messrs. Healing and LeTall—have been experimenting onanother resonance method ofanalysing analternate-current vibration which wehave suggested tothem, and, considering how rough isthe apparatus they have been using, and how short atime they have been engaged onthis new method, the results areunexpectedly ‘VOL, XVII. 22 308 LABORATORY NOTESON [March28, satisfactory, andappear toshow that this method, which is extremely simple, isoneofgreat promise. Thealternate current tobeanalysed ispassed through astretched wire, VV(Fig. 9), thelength ortension ofwhich canbevaried asinamonochord, yv yi"« t 9)& Fi. 9, 18, magnet producing permanent feld. YY, wire conveying alternate ‘current, and tuned gothat itsnatural period ofvibration agreeswiththtof thealteroate current, D,alternate-current dynamo, This wire isinamagnetic field produced byapermanent magnet, V8, orbyanelectro-magnet, or,best ofall,byacoilof wire surrounding the wire VV asaflat galvanometer coil surrounds the needle, Asthe current inthe wire alternates the wire receives impulses backwards andforwards across thelines of force produced bythemagnetio field, andifthelength and tension ofthewire beadjusted bytrial until thenatural time of vibration ofthewireagreeswiththeperiodic timeofvariation ofthe alternate-current, thewire oscillates vigorously with anamplitude from 1}to2inches, emitting, ofcourse, amusical note. The sudden wayinwhich thewire starts vibrating when itslength and tension arejust right isvery striking. The frequency oanbe determined from themusical note emitted bythewire; bat38 that would require tuning-forks orsome other apparatus having 1188) ALTERNATE-OURRENT CIRCUITS, 209 fixedratesofvibration, aswellas©musicalear,wepreferto determinethefrequency byusing thewell-known formula fora vibrating string, viz, 1 /t n=Ws here nisthefrequeney, thelength ofthewireincentimetres, tthetension indynes, andmthemass ingrammes percenti- ner ofthe wire, Inthefirstexperiments wecommenced withthewirecom-Tantively slack, andongradually tightening itupitwasseento vibrate very slowly. Holding one’s finger onthewireandcount ingthevibrations, they were found tocorrespond with the seed oftheengine driving thealternate-current dynamo, D, "hichwassendingthealternate currentthroughthewire.Ontightening upthewire alittle more itresponded totwice this nnteofvibration, which. may, perhaps, have corresponded with thenumber ofpiston strokes; then thewire required agood ‘ealoftightening, when itsuddenly burst. intovibration, and acalonlating the value ofnfrom theformula just given, a resultwasobtained closelyagreeing withtheelectricfrequency, "hich, with thepartioular dynamo employed, waseight times thenumber ofrotations ofthe armature per second. On tightening thewire still more, shriller notes were heard, and muller waves seen superadded onthe fundamental vibration cfthe wire asawhole, We were atfirst afraid that the analysis ofthe harmonics ofthe alternate current would be troubled bythevibrating wire having harmonics ofitsown, butwewere reminded byMr. Lamb ofthewell-known fact— Which weought tohave remembered—that the harmonics Possessed bythevibrating string of»musical instrament are impressed onitintheactofsetting thestring invibration, cr,inother words, that a°string has noharmonics ofitsown, Hence wemay safely conclude that thehigher rates ofvibra- tionseen and heard with ourwire arereally due toharmonics intho alternate-current wave itself, To determine the frequency and amplitude ofeach ofthese harmonics the length andtension ofthewire areadjusted until itsnatural 310 LABORATORY NOTESON larchs80, period ofvibration agrees with theparticular harmonic sought forinthealternate current, andwhich agreement isevidenced bythewire vibrating vigorously when itisattained. ‘Wehave notyetgone beyond thispoint inourexperiments, but some modifications and developments ofthe method that hhave occurred tousmay bebriefly referred to, Instead of tuning thewire toaccord successively with each ofthehar monies sought for, itmay befound convenient tohave number ofwires tuned togive out, andtherefore torespond to, different notes, asinaharp, and placed inamagnetic field, then tosend thealternate current tobeanslysed through them inparallel orinseries (probably the latter would bebetter), ‘and observe which ofthewires vibrate, and what are the various amplitudes ofvibration. Or,instead ofsending thealternate current through the stretched wire orwires, itmay besent round asolenoid orelectro-magnet, and adirect current sent through the stretched wire orwires, placed relatively tothe solenoid sothat acurrent inthe wire orwires isdeflected byacurrent passing round the solenoid. Or,inplace ofa vibrating wire, amagnetic tongue the natural rate ofvibra- tionofwhichcanbevariedbyalteringitslengthoritsmoment ofinertia, orinsome other convenient way, may beplaced near ‘asolenoid orelectro-magnet round which flows analternate current. Orasetofsuch tongues tuned like thevibrating springs ofamouth organ may beemployed toanalyse an alternate current sent round asolenoid orelectromagnet. Inthefigure thealternate-current dynamo isshown near the vibrating wire, but itwill beobvious that any one ofthese arrangements may beemployed tomeasure thefrequency ofan alternate current atanydistance from thedynamo. Orthey may beused simply tomeasure thespeed ofthedynamo atanypoint ofthecircuit through which flows thealternate current produced bythedynamo,ifthenumberofalternationsperrevolation—which is,ofcourse, adefinite number foreach dynamo—be known. Or, generally, ifthere beattached toany moving mechanism some arrangement formaking and breaking «current,orvaryingthe strengthofacurrent,anyoneofthesemethodsmaybeusedatany 9) ALTERNATE .CURRENT CIROUITS, au distance from themoving mechanism tomeasure thefrequency, and, therefore, the speed atwhich the mechanism ismoving. Sothatthearrangement canbeused asasimple speed-indicator forindicating atadistance thespeed ofmoving machinery. Ifthevibrating wire consist oftwoormore wires mechanically joined together, but electrically insulated from oneanother so thatdifferentcurrents canbepassedthroughthem,then,ifthe curents beofthesame frequency, wecanuseourmethod for ‘comparing the mean value ofthe algebraic sum ofthecurrents viththemean value anyofoneofthem; andbycomparing the mean value thus found ofthealgebraic sum oftwocurrents with whatthemean value would bewere there nodifference inphase, vecandetermine thisdifference inphase.Andthesamething maybedone bysending thecurrents round different wires ona solenoid placed soastocause amagnetic tongue tovibrate. Infact, wehave ourselves hardly yetrealised alltheuses to which this “electric frequency meter” canbeput. ‘TheCatnMaN: After theexpression ofyour feelings which Mc. youhave just shown, Ineed hardly askyou toaccord ahearty voleofthanks toProfessors Ayrton and Perry forhaving given 1stheseinteresting notesatsuchashortnotice.Iwouldnowcallupon any members fortheir remarks, and would especially Point outtothestudents who have attended Professor Ayrton’s lectures atvarious times that itisnow the time toturn round onhim, Professor W.E.Ayrton: Perhaps the invitation might ‘equally apply tothose who were students aswell asto those who are. The Cuamsmax: The subject ofself-induction issuch an important oneatthepresent time thatitwould bevery desirable toheartheopinions ofmany whoarepresent. Mr. W.P.Graxviste: Although not astudent under us Professor Ayrton intheordinary sense, Ihavecertainly been“ one to-night, and forthe information gained Ivery much thank him. There isonepoint Iwish tosayaword upon, and thatisinreference tothefactpointedoutbyProfessor Ayrton, 412 LABORATORY NOTES,Bre. (Btarehne, | thattheself-induction ofacoilofwireisnotincreased bytheinsertion ofanironcoreéfthealternations ofcurrentar exceedingly rapid, and that ifthe alternations are further increased the self-induction iseven diminished. You vill probably remember that Mr. Willoughby Smith, some fiveor six years ago, gave anexperimentally illustrated paper on induction. With those experiments Iwas toacertain extent, connected ashisassistant, and onthat occasion acommutator wasusedverymuchofthesameformasthatadoptedby Professor Ayrton inhissecohmmeter, viz.,adoublecommutator mounted onone spindle, one part ofthe commutator reversing the battery, and the other part simultaneously reversing the galvanometer, soastoproduce asteady deflection. Inone experiment shownbyMr.Willoughby Smith,twoflatspiralswere placed about afoot apart, and bymeans ofthedouble com- mutator analternating current from abattery wassent into one ofthe spirals, and the other spiral was connected through the commutator toagalvanometer, sothat theinduction received by thesecond spiral wasindicated byasteady deflection. Itwas then found that byinterposing plates ofdifferent non-magnetic metals certain proportions ofthecurrent were cutoff,butthat ‘when aniron plate wasinterposed, theproportion intercepted did notincrease withthespeed,aswasthecasewithothermetals, but remained almost constant, and even slightly diminished, when therate ofalternation wasvery rapid; andIthink thisis thesame result that Professor Ayrton hassoably explained tous to-night. rrotener Professor W.E,Avxron: Iamvery much interested tohear ‘we’ ‘theremarks made byMr.Granville. Icanonlysaythat,not- withstanding IheardMr.Willoughby Smith'spaper,Iwastotally‘unaware ofsuch amethod having been employed; andIam quite sure also that Messieurs Ledeboer and Manceuvrier, who were engaged atthe same time asProfessor Perry and myself in developing easy methods fortheabsolute measurement ofthe coefficients ofself-induction twoyears ago, were equally unaware that Mr. Willoughby Smith had employed asomewhat similar commutating device. Although itisnot ofvery much conse- 1882) DISCUSSION. a8 quence,Imayeaythatwedonotreversethegalvanometer attheZrtsmor moment that wereverse thebattery; that would notanswer our porpose.Thegalvanometer is,inthisparticular apparatus, reversed midway between thereversal ofthebattery, because wewant—I need notgointo that now; but ifyouobserve these commuta- torsafterwards, youwill seethat there isthat marked difference. Idonotmean tosaythat that would make anyradical distinc- tionbetween thepieces ofapparatus. That was mentioned, I ‘suppose, inthepaper? Mr.W.P.Granviiie: Yes; and also inthepamphlet called “Induction.” Professor W.E,Avnton: Iam very glad tohear itnow, and Iamastonishedatmyownignorance. ‘Mr.W.P.GRANVILLE: There isasketch inthepamphlet. Professor W.E.Anton: That makes itmore disgraceful till. Ipresume, however, that Mr. Willoughby Smith didnot tuehisarrangement fortheabsolute measurement ofthecoefii- centsofselfandmutualinduction, whichistheessenceofour derice, Mr.W.B.Esson: May IaskProfessor Ayrton whether theMeBaoe. carve hegives oftheFerranti machine represents theelectro- motive force induced, ormerely theflux oflines through the amaturefordifferent positionsofitscoils? ProfessorW.E,Ayrton:Youarespeakingofthiscurve:frmor thisisthe actual electro-motive force ofthe machine;thatisthe telfinduction, ‘Mr.W.B.Esson: Did Iunderstand Professor Ayrton tosayMr.Emoo. ‘that intheFerranti machine hefound awave having double thefrequency superposed ontheprincipal wave ? Professor W. E.AYRT0N:No;fourorfivetimes,NotdoublePetar ortreble, butatfour orfivetimes thefrequency. ‘Mr.W.B.Esson: Isuppose theeffect inaFerranti machine **a=. ‘hich hasnoiron initsarmature isnotsufficiently marked, but, thereisawave,duetotheweakening ofthefieldbythearmature ‘earrent,whichhastwicethefrequencyoftheprincipalwave. ‘TheCaamaan:Gentlemen, youhavealreadyaccordedyour voteofthankstoProfessors AyrtonandPerry,soIwilljustsay 314‘ONTHEDISTURBANCES ARIGING FROMTHE(March8, that Mr.Preece wassoovercome bythesudden andsadnews to which Ihadtoallude atthebeginning ofourproceedings, that hefelt hecould not attend this meeting;Iwillthereforecall upon the Secretary toread hiscommunication, which isavery short one. ‘The Secretary then read thefollowing paper :— ON THE DISTURBANCES ARISING FROM THE USE OF “EARTH” FOR ELECTRIC LIGHTING PURPOSES. ByW.H.Passce, F.RS., Past-President. The magnetic field produced byacurrent flowing in straight wire, andreturning through theearth, extends tosuch distances that itisquite impossible tosay where orwhen or how atelephone circuit inthat field would bedisturbed by ‘any changes inthat current. Itisquite certain that if single conductor between Deptford and London were subject torapid alternations under apotential difference of10,000 volts— the current returning byway ofthe earth—every telephone circuit inthe metropolis would bedisturbed, and probebly rendered unworkable. The law determining the distance to which this influence extends was given bymeinapaper read before the British Association atManchester in 1887, Itis giveninthefollowing equation:— coeShey ©,being the primary currents, C,the secondary currents, Ithe length ofthe primary, dthedistance between thewires, rr,theresistance ofthesecondary, Maconstant dependent on thefrequency oftheprimary currents andtherate atwhich they rise and fall. Itfollows from this formula that currents ofrapid frequency on conductor 10miles inlength, carrying 100amperes, might beevident atadistance of19miles. Itbecomes, therefore, important todecidetowhatextentdisturbing influencescanbe eliminated. Itiswellknown that ifeither thedisturbing cireuitor 1] USE OF“EARTH” FOR ELECTRIC LIGHTING PURPOSES, 315, thatdisturbed beconstructed oftwowires carrying equal currente, 10closetoeachotherandsoerectedthatthemeanaverage dis- tance between thedisturber and the disturbed isthe same, no inductive effects are evident, forthe influence onone wire is tatirely neutralised bytheequal andopposite effect ontheother. Batifthismeanaveragedistanceisnotequal,oriftheequalityof,thetwocurrents bedeparted from,thenadisturbance willarise which will depend onthedifference between thetwocurrents aadtheir relative distances. Hence the most perfectly silent telephone circuit becomes noisy when leaks occur intheinsula tion, and hence itisthat the earth plays such animportant itintheefficiency ofthesystem. Itwas reported inthe technical papers that the London Hectrie Supply Corporation contemplated laying aconcentric conductor between Deptford and London, with the external copperconductor incontactwiththeearth,AsitseemedtothePostOfficeauthorities verydoubtful whether suchanarrange- ment would not create disturbance onthe telegraph and telephone circuits, experiments were made totest the point. ‘Theexperiments were made byme inanopen space of two orthree acres in extent. Aconcentric cable was con- structed sothat the outside and enveloping conductor could beinsulated ornot.Theinnerconductor wascomposed ofa strand ofnineteen No.15B.W.G. (72mils.) copper wires. The cater conductor completely surrounded the inner, and consisted oftwenty-four No. 16B.W.G. (65mils.) copper wires. Each conductor weighed about 1,600 Ibs.permile, and gave aresist anceof-56ohm. ‘Theoutside sheath wasoflead. The following isafullsize section ofthecable :-— MES DSo2esjoA - | 316«ONTHEDISTURBANCES ARISING FROMTHE(iarch2h, ‘Thefollowing diagramwillexplainthearrangement ofthe wwires:— : 52Yarde t 8 Yards j a a a WV, VA yy i! \ ij)27Yard 2 3 c2 68Yards ‘Twoearth-plates wereburied atthepoints AandB,and between them aNo. 7gutta-percha wire was laid along the ground, telephones being inserted ateach end. Theelectric light cable waslaid between thesame points byacircuitous route,a shown, Atthe points Cand Dtwo earth-rods were driven in, ‘and another No. 7gutta-percha wire was laid outside theares enclosed by,andabout afootfrom, thecable. Beyond each ofthegutta-percha wires, extension wires were lis) USE OF“EARTH” FOR ELECTRIO LIGHTING PURPOSES. 817 carried toenable metallic loops tobemade up,forming approxi- ratesquares (1and2ondiagram). af AWheatstone transmitter placed atBwasconnected to|theinnerconductor(I)ofthecable,theleadsheath(L)being a|usedasthereturnwire.Reversalsattopspeed(afrequency 1.of0persecond)weresentthroughthecablefromanE.M.F.of20volts, thecurrent strength being about 1-3amperes. Loud disturbance was heard inthetelephones atAandB when thetransmitter wasrunning. fEarth-bars weredriveninatthepointsFandG,and,,|Connected byawire andtelephone. Currents were picked “]spbythetelephone through thissection, theireffectbeingapparently quite asloud asbetween Aand B, a{ThepointsHandIweresimilarlyconnected,andwitha like result. . The points CandDwere next connected bythewire‘Jsebedtotheearth-bars,atelephonebeinginsertedatD.|Disturbances almostasloudasonthecircuitABwere heard. ‘The wire CDwasdisconnected from theearth-bars, andAfametalliccircuitforminganapproximate squarewasmade |up(1ondiagram). Theinterference wasstillheard,butwas only about halfasloudaswhentheearthwasusedatCD. {Theinsulated outerconductor (0)wassubstituted in 4.placeoftheleadsheathastheretitnwire,andthedisturb- \ ancesintheloopCDceasedentirely. |‘Theouterconductor (0)andtheleadsheath(L)were coupled together, forming jointly areturn wire incontact 7.with theearth. Thedisturbance inthemetallic square1 |reappeared, butwas reduced inloudness toabout half that inthefourth experiment. ThecablewasnextlaidoutstraightfromAtoB,the telephone wire being parallel toit.When thelead sheath ‘was usedasthereturnwire,louddisturbance washeardin 8.\thetelephones atAand B.When thelead sheath and outer conductor were jointly used asthe return wire, thenoiseinthetelephoneswasgreatlyreduced(toabout \one-fourth). 318 «ONTHEDISTURBANCES ARISING FROM,Ere.(March28th, Toprovethattheresultobtainedinthelastexperi- 8.jmentwasduetoearthconduction, ametallicsquare(2in °|diagram)wasformed,andthedisturbance inthetelephones disappeared. Itfollows from these experiments that thedisturbance ofthe potential oftheearth atAand B,and theestablishment between these twopoints of@potential difference, cause currents tofo through theearth between Aand Bwhich destroy theequality between those inthe inner and outer conductor ofthe concentrie cable upon’ which equilibrium depends. Itistherefore quite clear that theuseoftheexternal sheathing aspart ofthecon ductor incontact with the earth isquite out ofthe question; whereas ifthe enveloping conductor beinsulated itisquite innocent ofdisturbance, andwould notinterfere with neighbouring telephones. ‘The question hasarisen astowhether thecurrents experi mented with (viz., those produced byaWheatstone transmitter) areofthesame character asthose produced byanalternating- current machine. The frequency inthetwocases maybealike. ‘Taking the frequency at100, itwill follow that each current, whether itbeoftheWheatstone transmitter orofthealternating- current machine, will rise from zero tomaximum and fallfrom maximum tozero in-0025 second. The strength ofcurrent generally used intheWheatstone transmitter is50milli-amperes, while that ofanalternating-current machine mayamountto100 ‘amperes;andsincemostoftheWheatstone circuitsinthiscountry tarelong circuits, upon which theresistance andthecapacity are considerable, itisquite evident that inactual practice therateat which thecurrent rises andfallsisgreater inthecaseofalternating- current machines than inthe case ofthe Wheatstone transmitter. ‘The fact that itissoissufficiently proved inactual practice. Before theGrosvenor Gallery Company established their present metallic loop system, allthetelephones around andabout Bond Street were disturbed. Inorder, however, tosettle thematter,I have carefully experimented with thecurrents produced bythe alternate-current machines atWest Brompton andatEastbourne, ‘and Ifind that itisquite impossible todetect anydifference 1088 DISCUSSION. a9 Iwiween the effects ofWheatstone transmitters and alternating- current dynamos. These experiments, taken alone, aresufficiently definite to render itdecisive that the use ofthe earth insuch instances cannot beallowed; but itmay happen that inother systems of distribution where thecurrents arecontinuous andsteady, such asthesystems bywhich itisproposed todistribute bymeans ofbatteries, theearth under exceptional circumstances may be permitted. Mr.W.M.Monpey: Mr.PreecereferstothecurrentasscMone beingat@tension of10,000 volts. Iwould askwhether there is aydirect ornecessary connection between thetension and the leakage inthis case. ‘The only leakage isfrom onepart ofthe outer conductor toanother part ofitthrough the earth, and the gestest possible tension, therefore, applied totheearth must be thatduetothelossofpressure—the fallofpotential intheouter lead. This may beonly afewvolts, although theactual pressure ‘wedinthestation maybe10,000volts. Iamnotconcerned as amapologist fortheDeptford scheme, butIdonotquite seehow Me.Preece connects thevery high tension which isproposed to teused there, with earth leakage. With the same lose of Pressure involts the leakage would bethe same, even ifthe terminal potential difference were only afraction ofthat stated. Professor W.E.AvRToN: Idonotsuppose—at least,Ihoperome —that Mr.Preece does notintend byearth leakagetomeanwhat isusually understood byearth leakage. The fact, ofcourse, is,as indeed Ihappened topoint out—I think itwasin1878—that if youhave avarying current inawire, andyouuseanwninsulated tubeasthereturn,theuninsulated tubedoesnotshieldoutside ‘yece from inductive action; because itisonly when thetube surounding thewire takes back thewhole ofthecurrent that g%sbythe interior core, that theoutside space isentirely shielded from magnetic action, ‘The actions that Mr.Preece gets arprobably notdue toearth leakage intheordinary sense of theword atall,butthey areduetothefactthat when thereturn ismade bythewninsulated leadcovering ofhiscable thewhole earth ispart, or,atanyrate, theearth intheneighbourhood is 82%ONTHEDISTURBANCES ARISING FROM,Ere.(March28, demirpart,ofthereturnwire;henceaportionofthereturnelectric light current comes outside thetelephone cironit, and youhave fluctuating number oflines offorce passing through thetelephone cireuit, due tothealternating current. If,however, youare dealing with adirect current where there arenofluctuations atall,then Idoubt very much ifthere will beany difference orany disturbing effect worth speaking ofonthe telephone circuit, whether theoutside covering beput toearth ornot. ‘Therefore Ithink that thesort ofpermissive sentence that ocears tatthe end ofMr. Preece’s paper(Iamsorrythatheisnothere toanswer forhimself) issomewhat misleading. Hesays: “But “it may happen that inother systems ofdistribution where the “currents arecontinuous and steady, such asthe systems by “which itisproposed todistribute bymeans ofbatteries, the “earth, under exceptional circumstances, may bepermitted.” Ishould saythat theearth might under any circumstances be permitted ifthecurrentinthecircuitwerequitesteady; because, ‘asthe disturbance inthe telephone lines isdue toelectro- magnetic induction, andaselectro-magnetic induction cannot be produced with aperfectly steady current, itisoflittle com sequence whether theelectric light return beearthed ornot. ‘Iwasunder theimpression that that was made quite clear eleven years ago, when this very samo subject was under discussion. ‘The question arose, Ithink, atthe time the Jablochkoff Company lighted theThames Embankment, when itwaspointedout—ifImistakenot,byMr.Preece—that although the return conductor oftheir alternating-current system su rounded the going conductor, there was inductive effect on neighbouring telegraph wires buried intheground near; andI then pointed outthat thisdisturbance arose from thetube which brought back thecurrent notbeing insulated, for,inorder toget perfect screening from induction, itwasnecessary that theretam tube should beentirely insulated. Xenon, MBReE,Cxowrron: Inreference toProfessor Ayrton'’s remark that currents which arerepresented tobeperfectly steady have noeffect ontelephones—— Professor AyrToN: Isaid currents that were steady, not unsteady. 19) DIBOUSSION. sa Mr.RE.Cnorron: Imeancurrentsassteadyasarepro-Ms... duced bythe ordinary direct-current machines, driven byany form ofsteam engine orother motor that Iamacquainted with. Ihave noticed that such currents always doproduce slight sounds inthe telephone, This Iattribute totheirregularities orpulsa- tions inthe speed ofthe motor that drives the dynamo machine. Inonecase atelephone wire waslaid inthesame trench with electric light cables. These were well insulated and Iead-covered. The telephone wire was spaced afew inches distant from them. The return ofthe telephone was byearth. Inthiscase itwasfound impossible, even when several sets of dynamo machines were working inparallel, toavoid considerable disturbance inthetelephone circuit. Infact, inthis case the telephone acted asavery delicate and useful detector. When a ‘ingle dynamo wasatwork therevolutions oftheengine could be distinctly counted, Further than this, wenoticed that a telephone wire thus laid becomes amost valuable detector of leakage between thearmature winding and earth, forsuch leakage, being ofanintermittent nature, isnoticed atonce bythealternations corresponding with thenumber ofrevolu- tions, Inone case anaccidental short-circuit inthe armature Boduced aslight alternating current, which wassufficient toring thetelephone bells. Itiseasy toseethat thisapplication ofthe telephone may beofgreat useincentral station work. With regard totheinductive disturbances from theoutside conductor referred toinMr. Preece’s paper, Ithink that the tea within which those disturbances would be felt must be comparatively small inthe caseofastraightlineofconductors, mdwould, infact, only bethat which bounds the portion of eath carrying anappreciable quantity ofthecurrent. Outside thisportion ofearth there would benomagnetic disturbance,as itwould bebalanced bythepresence ofthe opposite exciting curent inthe inner conductor. Ontheother hand, ifthelinewere laidinacurved direction, thewhole space included within itwould beoutofbalance, con- sidered magnetically, andinductive disturbances would takeplace therein, 32%ONTHEDISTURBANCES ARISINGFROM,Ere,(March28th ar.eme. Mr, W.B,Esson: May Iask Mr. Crompton whether the machines hespokeofwereBrushmachines? Me ‘Mr.R,E,Crompton: No; they were machines ofourown ‘manufacture, andwere perfect machines. uc.eace, Mr. W.B.Esson: Imention this because itdoes not follow that acurrent flowing always inone direction isacontinuous carrent—it may flow inthesame direction while very discon- tinuous. Ishould think Brush machines would always produce ‘8very considerable effect onadjacent telephone circuits, Mimpeon, MRE, Cnomrrox: Wehave notyetgot steam engine that drives at continuous speed, andconsequently thecontinuous current sent isnot continuous inany machine that Iam acquainted with. Yretemoe Professor W.E.Ayron: Imayjust remark that thelast point inMr, Preece’s paper had reference tobatteries, and not tomachines atall; “distribution bymeans ofbatteries,” isthe lastsentence inthepaper. Iamperfectly prepared toadmit that there are certain machines which are very discontinuous; ‘and, indeed, itmay possibly beinthe memory ofsome people that Professor Perry and Iproposed, several years ago, adis- continuity meter, asIthink itwascalled. The apparatas was simply aninduction coil through one cireuit ofwhich theso- called continuous current wassent; inthesecondary circuit of this induction coil was placed atelephone, oradynamometer, orother suitable instrument formeasuring thealternate current induced inconsequence ofwant ofperfect constancy oftheprimary current. Experiments were made, Ithink in1879, andweshowed that with theBrush machine there was amost loud sound; in fact, the Brush machine, sending aso-called steady current through theprimary coiloftheinduction coil, gave asteady deflection with Siemens dynamometer attached tothe secondary. Ofcourse such aninduced current could only have been obtained ifthere were considerable variation ofthe current intheprimary coil. Sound could also beheard with theEdison machine, butnodecided deflection was obtained with it,ssfar asIremember, with the dynamometer ofour discontinuity meter. Butthepoint Iwasspeaking about wasthedistribution oor Ze geane pe si) eek 2Se ele rsa SG i. gBE) oaupecggs ig:Spee ipaos ZE\E gasiy22223854773 5 E5455 123: 12s gine3s1242: ala||seriteee |(82alee e)2PElSSuiASe 234f23222,22. £8 a)gigifiigui iy692028388275 E\sBEEGI<E22 E22g2ieebecgsti2 |S)|g2286428G6fai22 Z)E|Ber roe rep re osetoog f BlBR)~\g>ecescseeeng pGia Bla) |t neecneacnes 5 ase Sle geepen| ciissiiaieeemyen) ge, £2 gang £322¢ * so Bale[ Ali fgisrssteiee aBGs ne£2 pF Za552 22 geaole|FfBbessss Regge2°gy: 2|"|3822og5822£635 2°Be: é S83 BRESSES eylts 5 3I sia 33582222528 <2 eoS||ae2Bey*2hide 2aa: S68 22..8,,828s244 Ee G43 4° 82 &Sezfs 2 a** e | se | me | : | | 1889.) DISCUSSION. 338 bymeans ofbatteries. Ifthe distribution isbymeans ofFrotawor batteries, which arenot machines ofthis discontinuous nature, then Idoubt whether youmight notputthereturn toearth without disturbing telephone circuits. ‘The Cxamsay: Iwillnowaskyoutoaccord toMr.Preece ams hearty vote ofthanks forbringing this interesting subject forward;andIcanonlyhopeonmyownpart.thatweshall have anearly opportunity ofdiscussing thesubject again, onan evening whenhecanbepresent,andwhenalittlemoretimeis atthedisposal ofthemeeting than hasbeen available to-night. ‘The motion wasunanimously carried. Aballot took place, atwhich thefollowing candidates were elected :— Members: E,A.Kenyon. | Edward Manville, Associates: Richard CharlesBennett. |G.Mahon.Frank Broadbent. Allan Plucknett. J.W.Howard. Lionel Hugh Kenmore Stotherd. Students: Archibald John French. | Archer Turner. ‘The meeting then adjourned, vou. xvi. 23 su ORIGINAL COMMUNICATIONS. 17,Wesrurnsren Craunsas, Vicvonta Stazzs, 8.W,, Loxvow, 11th Apr, 1889. Dear Sin,—Referring tothediscussion onProfessor Jamieson's paper on“Insulation Resistance,” &c., and inaccordance with thesuggestion ofMr.Preece that theresult ofactual tests should ‘beforwarded toyou, Inowhand youadiagram oftests taken at Hastings. The circumstances under which the results were obtained ofcourse vary very considerably, butinthemajority of cases itmay betaken that allexposed portions ofthecircuit would have acoating ofsalt, which rendered the insulation peculiarly liable tobeaffected byslight moisture intheatmos- phere. The differently drawn lines show separate circuits, and itmay betaken that each circuit had approximately thesame liability toleakage,asoneachtherewerearelampsandincan- descents, Most. arclamps were outside, exposed totheweathers theincandescents being almost: wholly within doors. The cables were 7/16copper strand, covered withadouble coating ofgutta percha, andbraided. The total length ofcircuits atthistime ‘was, roughly, 8,900 yards, alllaidunderground in4-in. cast-iron pipes. Test boxes every 100yards, Pipes Iaidinroadway about, 15in,from surface. Anattempt wasmade tokeep allwater out, ‘| butitwasafterwards found that theliability toleakage wasnot therebyappreciablyreduced.Testsweretakentwiceadaywith|‘aWheatstone bridge andmirror galvanometer, thebattery‘giving anEMF, ofabout 90volts. The great source ofbreakdowns | ‘wasnot,asmighthavebeensupposed, thejoints,butwherethecablehadreceivedslightinjuryinpullingintothetubes.As| skilled labour was atthat time and place notobtainable, this is hardlytobewonderedat.Itwasfoundthatsomefewofthe joints didgo,butthisarose from want ofcare principally. Ido a eS o (HASTINGS INSTALLATION). TE \ l:LL vegumive=944#aestOOSputinabout100yardsaparneareroncurves.Allpipeslaidstraight. Wiree.—7)16 cables, covered withgutta-percha intwolaye llytobewondered at,JUwits10uuUtumsumeLEWULLHe tutsdidgo,butthisarose from want ofcare principally. Ido OnIoINA, VoMMUICATIONS, ts notremember, after thefirst few months’ working, any joint proving bad, except those made atthestart. The E.MLF. used ‘wasabout 2,000 volts. ‘Yours faithfully, F.B.NICHOLSON, Associate. asec, TanofBisapace 1,Werner Grae, Von Sa, Lonpor, 26th April, 1889. Dear Sta,—In farther reference tomyletter onthesubject ofinsulation tests atHastings, Ibegtohand youafewfigures relating tocost ofsame, foruseasyouthink fit. Yours faithfully, F.B,NICHOLSON. sentry, Taatin ofBesa! Hoioe NOTES ON THE COST OF LAYING PIPES FOR UNDER- GROUND MAINS. ‘Hastings,1883anv1884. Pipes.—4in.internaldiameter, and}in,thick;length,9ft. weight ofeach pipe, 156Ibs. Joints made with aclip thus, =} mpl ae —— =—————zajai== intwohalves. Inlowerhalfofclipabosscastandscréwed,{withathincastingofiron filling up;inordertotakeoffabranchtoa house,allthatisnecessary istobreakinthiswithachiseland serew inlin. gasbaniel. Each clipweighing 40Ibs, and giving anadditional 7in.topipe, sothatlength between centres ofjoints =9ft.7in,‘ustboxes putinabout 100yards apart, nearer oncarves. Allpipes laidstraight. Wires—7/16 cables, covered withgutta-percha intwolayers 318 ORIGINAL COMMUNICATIONS, andheavily braided. Pulled inpipes bywires laidinforthe purpose;wireondrum,andguidedbyrollers. Costs— Peryard.Hach Pipes, clips, andbolts wee #1 Wages wee ewe OEToolsharps, &,seeee OE Patty andredlead... we eT Carriage andcartage wee OERoadrepairs... wseGTest boxes. ee eee SE 510f ‘Total length laid, 9,400 feet. .Metal used. | Pipes... ee ae nee wee 1,085, i Chips n,n eee 1,085 Bolts ue ee ss nae nee 8,150 WireLaying—Totallength,8,870yards. ‘ Labour. wes peryard "38 Odd material... ow » 06 “44 Pipes laidintrench 1ft.6in.deepby1ft.3in.wide. Conditions ofground veryvarying. About one-balf total length ordinary macadamised road, one-quarter macadam made upof three layers where roadhadbeen raised, remaining quarter in softgravelly bedofParade. Allworkcarried outbymenquite ‘unskilled incable work, andtrouble wasexperienced atthestart frombadjoints, butafterexperience remedied thisfault.‘AreLighte—A contractwasenteredintowiththeCorporationfortheerection andlighting of15arclamps ontheParade, andthefollowing figuresrefertoitSomepostswereputinthe|groundintheordinary manner, andlittletroublewasexperienced ;| butsome three orfour hadtobeplaced inanexcavation in concrete, which materially increased thecostoflabour. ORIGINAL COMMUNICATIONS. 327 Posts about 20ft.high, castiron. Lanterns specially made incopper. Lamps, Brush, 16-hour, round type. ea Lamps. ww we each «12-50 Lanterns... se wee 5100 Posts ve nee eevee AT OB Switches... we vee 0138 Wages. oe wee 20 Tool repairs, vee nee 00 7B Materials (odd). 0GN Repairing roads... se01SGasbarrelfrom mains...+0364 Carriage. nee 02 Cable ue ene ae OS Total,perlamp..,£26144¢ F.B,NICHOLSON. COST OF RUNNING ARC LAMPS. ‘TAKEN FROM AcTUAL WoRKING. Case 1.—One hundred arelamps, each of2,000 nominal candle-power, viz., 50volts and 10amperes. Allinseries. ‘Total cost, including rent, taxes, wages, allowance fordepreci- ation, and interest oncapital, power, &c., between 4d.and 4d. perlamp perhour. Lighting forabout 3,000 hours perannum, Case 2.—Three 2,000-candle-power arcs. Time, orrunning, 383 hours perannum. Cost, between 10d.and 11d.perhour for the three arcs, illuminating anarea of4,674 square feet, and replacing 300gasjets(4feet), andcosting 28.7d.perhour, Gas 2s.2d.per1,000. Cost includes power, andinterest on£250, cost ofinstalling; wages notenough tobereckoned; coal, 5s.6d.per ton} carbons (11mm., 12in,long), 2d.each. F,B,NICHOLSON. 38 ABSTRACTS. 1H. NAGAOKA-OOMBINED EFFEOTS OF TORSION AND LONGITUDINAL STRESS ON THE MAGNETISATION OF NICKEL. : (Pht Mag, Vol. 27,February, 1889, p.1172) ‘The nickel wire tobe experimented upon, which was 1mm, indiameter and40em,long,hadtwopiecesofbrauswiresecarelyattachedtoitsends; ‘the upper piece was firmly fixed toasupport the lower piece carried & ‘eale-pan which could beloaded toany desired extent. The lower brit‘wirepassedthroughaholeinthecontreof@brassbar,theendsofwhich could aide intwo vertical Vgrooves, cut diametrically opposite each other ontheinside ofabras eylinder. This cylinder, which wasdivided onitsoutersurfaceintoacircleofdegrees,fittedonto«lowercylinder,90a8 admit ofitsbeing rotated;aactscrewinthediametralbrassbarclampel the lowor pieceofbrassattachedtothenickelwire.Anydegreeoftortion ‘ould therefore begiven tothenickel wire which passed vertically through thecentre of«magnetising solenoid, while the Vgrooves allowed ofit being stretched bytheweights inthesealepan. ‘The conditions ofthe experiment were continually varied byaltering‘oththeJongitadinal stressandthestrengthofthemagneticldproducebythesolenoid, The author states that thegoneral results may besun- aarised asfollows:—Ia allmagnetic felds with moderate loading,theeffec, foftwisting nickel wire istoincrease the magnetisation, ‘This increae depends onthe strength ofthe feld aswell asonthe longitudinal sts. Ifthe eld beweak, and the longitadinal strese auficiently great, the mag- petisation increases inone direction oftwist, and decreases fnthe othe. Eventually for aparticular stress which isapproximately proportional to ‘the Meld, thewire begins toshow opposite polarity, and the cyclic carve‘ofmagnetimtion passesgraduallyfromatwo-looped toasingleloopel form.For stronger flelds similar effects exist, but infolds higher than arita! value the increase and decrease ofmagnetisation take place forrevered Airections oftwist, and atthe eame time the course ofthe curve becomes ©versed, .DORN—DETERMINATION OF THE TRUE OHM. (Anmaten derPhysik und Chemie, Vol, 36,1888, pp.32and 896) Itisnotpossibletodomorothanglanceatthisverycompletetreatise02 ‘theaccuratemeasurement oftheohm,extendingasitdowstoovehundredpages bristling with formula andtabulated resulta ofobservations,‘Themethodofmeasurement wasthatcommonly knownaxW.Weber ABSTRACTS, 329 ‘hindmethod,modifiedbyDora.‘Thedampingofasinglepowerfalmagnetby thecoils of8galvanometar closely surrounding itwas observed,andthe ‘constantofthe galvanometer determined bycomparison with atangentflvanometer, ‘Theobservations, whichwerecarriedoutinthelatterhalfof1885 and inthebeginning of1866, were very numerous, Each observation comprised the following measurements:—(1) Measurement ofthe distance of thescales from thetwo galvanometers; (2)Verification ofdistance oftheindexmarksofthemagnet;(8)Comparison oftheresistanceofthegalvan-meterwithastandardresistance; (4)Measurement ofthedampingwithfourdiferent values ofthe galvanometer resistance; (6)Determination ofthe ‘oustant ofthegalvanometer;(6)Determination oftheperiodofoscillation; )Determination oftheair-damping; (8)Observation oftheratio Mj from thedefection ofthe magnetometer ofthe tangent galvanometer bythe ‘mlranometer magnetintwopositions;(9)Correspondingobservationatthe aceoccupiedbythegulvanometer; (10)Comparison ofthevalueofthe‘arth's horizontal intensity attheplace ofthegalvanometer with that atthe ace ofthetangent instrament,‘ThofinalmeanvalueoftheohmfoundbyDornwas—Johm=1:06243metresofmercary,1sq.mm,inectionat0°O. Thisvalue isthe mean ofthefollowing -—Six determinations insummer gave4meanof1-06943-10-00018; nineinautumn,1-06242 0-000255eightin‘inter, 1-06244 4.000082. @. H VON WYSS—RESISTANCE OF MAGNETIC IRON. (Annalen derPhysik undChemie, Vo.86,1888, p-487.) ‘The iron wire, placed inside amagnetising solenoid, formed one oftheYranchesofThomsonbridge,soarrangedthattheresistanceoftheconnec-tins and contacts coald beeliminated. Inorder toeliminate the effect ofthe beat which was necessarily produced inthe intide ofthe solenoid bythe ‘age ofthemagnetising current, asecond arm ofthebridge contained a Preisly similar iron wire inaprecisely similar solenoid sofar assize and coustraction were concerned, but with itscoils soconnected that thecurrent inonehalf neatralised that inthe other half, The two solenoids being ‘eunected inseries and tho same current passed through thom, both were ‘quilly heated, but only one iron wire was magnetised, ThefiatpointtowhichVouWyssdirectedhisattentionwasthereversal olthedirection ofthe magoetising current;hefoundittohavonoeffect, The tabulated resultsofthemessarementa andthecarvesplottedfromthem stowthattheresistancewasineverycaseincreasedbymagnetisation; thefelisotforceusedvariodinintensity. Also,itappearsthattheincreasein ‘etistanee becomes greater with more intense flea, The author carefullyAétarmined themagneticmomentoftheironwire;andhoconcludesthatthe‘angeinitselectricresistanceisnearlyproportional tothechangeinthe‘magnetic;atanyratethisistraewithinthelimitsofhisexperiments, 0 ABSTRACTS, ©.&,WEBERCONDUCTIVITY OFSOLIDMERCURY. (Annalen derPhysikundChemie,Vol,96.1889,p.587.) Mr. 1,Granthal having published some rerlts which are opps’ t ‘thoseobtained byCailletet andBouty,andbytheauthor,thelatterseekstoexplain the diferences, which are:—(1) The resistance ofliquid mercuryis nlyabontoveandahaftimesthatofoldmercury,fwtedofforHiner; (a)The temperatarecoeficientixonly0008100004intendof4percent (2)Thistemperatare cooficientinvariable, Theauthordeenotclaimabmitetractus;buthocannoteditthalheiofarwrongosGranthal'expe Inenta would goto prove; betde, hisgure agree clowly with thoeotiainel bythe two French obmerver, Both agreeanotheresitanceofmercuryjustabovethefrexingpity therefor, inorder toexplain thewide dlvergence inthe maarnremeatofibe resistancejstbelowthiepint,itastbeasamedthatonelldifeationwae ruddenalterationtakesplacewhicheithermakesWeber'sreitanceapparlw than itreally igor maker Grunthal's appear greater. There is,bower,nothinginthephesomenon ofsliieation whichwouldladustoexpectGecreae,batrathertherevere,thereforeWebervaluisthemoreprob,speciallywince,ineolidifying,beaksincontinnityoftherubatanceoftherereurymayoct,Moreover,austated,Weber'sfiguresagrewiththootCailetetandBouty;fromthevalueofthespecieresistanceatseo,vs, (2026 and the temperate coeficient, 00048, theypecfe resistanceat—59 ‘wouldbe02848forthewildmercary,andfom+1and=O00the tpecieoftheliquidmercuryat—89wouldbe09648,theratiooftheseto ‘Yalue is4108; OailetatandBoctyfindtheratiotobe£8. 3.KLEMENOIO_THE SUITABILITY OP PLATINUM-IRIDIOM WIRE AND SOME OTHER ALLOYS FOR THE CONSTRUCTION OF STANDARDS OF RESISTANCE. (Baatter, Vo39,188,» 89) at sbothe specific gravity, bthe conductivity at16:17 merery being taken asunity, «te tatmo-lectro-motive force with copper imicrovolts perdegree,pthetemperature coefficient perdegree;thenthe folowing vane have been found forvarious lloye:— i amas — ovate ans|Piatinum-tridium...2160| 878|744|7a5|1a6e|2390 BareGermansive}see|ast|ors|mar]aw|apGozered German}|gesneo-oes]123|asa|sor|a0Putioumsver |—|om|eee]—|am|oer ABSTRACTS, a ‘WithcopperandnicksinorcopperandGermansilver,thethermo-current fonsserothewarmjunctiontothecopper;inthetwolatinaalloysthe dinetionistherevere. . (Oncomparingthersittanc coilatntervaleof tenmonthswithmercaryandar, mall differences were found which were sometimes poritve andsraecinesnegativeandvariedfrom0019to0064percent.Smalldeforma.toeshavebatlittleeffectontheresistance, Thotwoplatinumalloysandcorel nickelin show asmall aecclar decree, Repeated deformations, ch tbending and unbending, hve considerable eect. Ifthe rsstance ofthe wires Dafne being heated toglowing isequal tounity, then the resistance Aermaia, when reduced tothe initial temperature (1667)Patinomtridium... .....097243|PlatinumSilver... 101898Tareickelin 100780 |BareGermanSilver 100866GoyeredNickelin 00725 |CoveredGermanBilver100604 Stretchingandtorsionincreesetherevstanceineachene;theincrease saybtransient orpermanent according totheleg oftime during which efor acta, The rate ofcooling was alto very carefully investigated, IfTisthe tenperatare ofthewire ¢that ofthe surrounding mediam, ©thespecieretanceatt¢thestrengthofthecurrentusedtoheatthewire,theradiuedlitewire,«thecoeMcientfortemperatare, andAtheconstantofcooling,‘hen SEM atao}-terar—o, soditeismal,forwiresofequaldiameterandforequalcarrents,8 3 —tcont8. Toeactualwireworenotallothemumédiameter,otherwisethevaluesofA,whichvariedfrom000081to000188,wouldhavebeenequal.‘The suthor concludes that platinem-iridlam and platioam-allver aretheBoyswhichbestmeetllthereqeirementa, Thehightermo-electromotive fersofnickelinwheninconjanctlonwithcopperivanobjectiontoitemse fora standard resistance, though ittlow temperatare coefficient isfavourabletalsuninresistancebores 2LAMDMANN—EXPERIMENTS ON BIHROMATE BATTERIES WITHOUT DIAPARAGHS, (Beibattor, Vol. 18,1889, p.94.) The cals have &height greater than their diameter and contain about6taeofasolationof12prisofbichromate ofsoditmin10)partsofvier,towhich26partofsulphuricecidisadded,Ifabout1aquaredimofsineisexpnodtothoanlation,itwillnotneedtobechangedfor see ABSTRACTS. ‘twohours,Owingtotheirverylowinternalresistance, thesecellscanbe‘conveniently used forworking glow lamps coupled inparallel. Onsocount ofthe small quantity ofmaterial required, the difficulty ofcrystallisation, ‘and the low price ofthe material, the bichromate ofsodium istobepre- ferred tothe potassium salt ortochromic acid. ‘The depolarisation and ELM, isgreater with porous carbon than with hard close-grained carbon. ‘The zincs used should befree from carbon and iron, since, ifthe cells ‘arecoupled inparallel, loeal action may besetup. M. BELLATI and 8.LUSSANA—OCOLUSION OF HYDROGEN BY NICKEL. (Beiblatter, Vol, 18,1889, p.95.) Nickelwillnotabsorbdryhydrogen butwhen®nickelwirewasused. asthe negative electrode inawater voltameter, and the quantity ofgas evolved compared with that from snother voltemeter with platinum elec- trodes, and with third with platinam anode and nickel cathode, itwas found that in200 hours the nickel had absorbed about 100 times itsvolume of Ihydrogen. The occladed hydrogen isnotgiven upifthenickel isplacedinabelljarcontaining airandsealedwithmercury,butthemercuryrises‘owing tooxidation ofthe metal, ‘When along nickel wire, hanging vertically inatube ofacidulated ‘water, served ascathode tos parallel platinum wire, the nickel wire increased by0000036 ofiteinitial length, viz., 144 m. This isnot an effect oftemperature, since comparison was made with the platinum wire, whichwouldhaveexpanded aboutequallyundertheinfluenceofheat,‘The occlusion ofhydrogen increases the resistance ofthe nickel. If thevolumeofhydrogen atO°C.and760mm,pressureis»timesthevolumeofthenickel,thentheincreaseofresistance at227°canbecalculated fromthe formals, 0.001636 +0:00005789* v. ‘With increasing temperature, the resistance ofnickel containing occluded hydrogen seoms toalter rather less than ordinary nickel; thus the tem- perature coefficient ofthe latter is000871, while that ofnickel with 100 times itsvolumeofoccludedhydrogenis0-00869. J.H. VAN UHOFF and L.T. REICHER—THE DISSOCIATION ‘THEORY OF ELECTROLYTES, (Beibtatter, Vol, 18,1889, p.98.) ‘The experiments were carried out with very dilute solutions, down to rabve gramme molecule per litre, ‘Their resistances were determined by meansofthetelephone method,uringArrhenius’ resistance vessels,Ifvis\thevolume inlitres, inwhich 1gramme molecule isdisolved;mthemole-| ABSTRACTS, 338 cular conductivity; mx themolecular conductivity foraninfinitely dilute solution;then,accordingtoOstwald, *0-%) ‘whereKisaconstant,‘The following table shows theresults oftheexperimenta:— fremp)metog.xc)+1ofremp|mx(tog.x)+10) AceticAcid... ..|1}sie)250|191|335|5337BatyrieAcid... |41|os]et|s9a|ax|2mBenzoieAcid... .|—|—|—|104]aor]soaFormicAcid........|wi]90}oe|—|—| =MonochloraceticAcid|wi}an}72|—|—|= 34 LIST OF ARTICLES aeLarixa 70 ELECTRICITY AND MAGNETISM ‘Appearing insome oftheprincipal Teohnioal Journals during theMonth of MARCH, 1889. I—BATTERIES AND ACCUMULATORS. ‘L,Donar1—Battery forPowerfalCurrents,Beiblatter, vol,18,p.169,1889. F.Exxes andJ.Toxa—Chemieal Theory oftheGalvanic Cell,—Bublater, vol. 18,p178, 1889, 1L—DYNAMOS AND MOTORS. R.Ansovx—Commercial Eficiency ofDynamos—Lum, Ely vol. 81,p.601, 1899, G,Rrcuano—Details ofDynamo Construction.—Lum. BL,vol81,p.601, 1889, Axox.—Fritache’s Wheel-Dynamo—BI. Zeit, vol. 10,p.134, 1889. TIL—ELECTROCHEMISTRY AND ELECTEO-METALLUEGY. ‘Viote and Cuussuanr—Electrolysia—C. R,vol. 108, p.248, 1889, B,Bovrr—Conduetivity and Electrolysis ofConcentrated Solutions of Sulphuric Acid.—C, R,,vol. 108, p.89%, 1889. ‘A.Porsea—Electrolysis ofMercurous NitrateasaMeasureofCurrent.—C. B, vol, 108, p.896, 1889, W,Orrwaip—Mereury-Drop Electrodes—C, R,vol. 108, p.401, 1889. A.Miner—Introdaction totheStudyofBlectro-Chemiatry.—Lam, El,vo.SI. pp.426, 477, 582, 676, 619, 1889. W,HLScuuurze—Electrolytical Properties ofMica atHigh Temperaturet— Annalen, vol. $6,p.655, 1889. X,Scunzama—Blectro.motive Force ofThin Films ofHydrated Superoride.— Annalen,vol.96,p-662,1889, 'B,Nrnxi—Disintegration ofCopper bytheElectric Current,—Beittter,vl.1 177, 1889. -Awox.—Diamonds made byElectricity.—Bl, Zeit, vol,10,p,168, 1889, Iv.-ELECTRIC LIGHT. G,Ricusny—High-Speed EnginesLum,Bt,vol.81,p.407,1889,©Ruionizx—Overall Diameters ofStranded Cables—Lum, Bi, vol, 81,PP. 425, 456, 1889.H,Duss—Counter Electromotive ForceoftheAre—Beiblatter, vol,18,p19,1889, ARTIOLES RELATING TO ELECTRIOITY, Bra 385 ‘Avon.—Schuckert's Installations inGermany.—EL Zeit, vol. 10,p.195, 1880. 0,Dimuxaxx—Calealation of«Distributing Network ofOonductors.—Bl, Zeit, ol10,p.148, 1680, &Lmarermat—Note onthe Bunsen Photometer.—Bl. Zeit., vol. 10,p.161, 1889, Axor.—Electrically LightedBuoysBY,Zeit,vol.10,p.165,1889. V.—ELECTRIC POWER. Avor,—Tree-Felling byElectricity —ET,Zeit,vol.10,p.164,1889. VL—MAGNETISM AND ELECTEO-MAGNETISM. Wx. Brown—Btoel Magneta.—Phil, May, vol. 27,p.270, 1889. ‘T.Movmaavx—Absolute Value oftheMagnetic Elements onJanuary Ist,1869. =C.B., vol, 108, p.66,1880. 6.Buasox—EflectofBlowsonthePermanent Magnetisation ofNickel—C.B, ‘ol. 108, p.94,1889. P,Jurer—Reciprocal Effect ofTwo Magnetimations atRight Angles inIron.— C.R,,vol. 108,p.398,1889.A.Porim—Relation between Magnetic Rotatory Power and theInfluence ofPonderableMatteronLaminousWaves,—C.R,vol.108,p.610,1889, H.Lorsmma—Theory ofMagneto-clectric Induction.—Annalen, vol. 6,p.671, 1880, D,Goromancax—Effect ofMagnetisation onConductivity.—Annalen, vol.8, P.804, 1889, ©Bim—Now Researches onMagnetism.—BI, Zeit, vol. 10,pp.128, 161, 1889. VIL_MEASUREMENTS AND MEASURING INSTRUMENTS. 4.Moten—Ospillary Electrometer and Mercury-Drop Electrodes.—C. Ry vol, 108, p.281, 1889, EMoumti—Measurement ofPhase Differences bymeans oftheSemicireular Klectrometer.—Lam, E,vol.31,p.630,1889. 1.Donart—New Quadrant Electrometar.—Beilatter, vol. 18,p.168, 1889, XX.—STATIC AND ATMOSPHERIC ELECTRICITY. K_Wrsuxpoxor—Difference ofPolarity inElectrostatic Dischargos—Beiblitter, ‘ol. 18,p.196, 1889,P,Nresmx—Lightning Conductors. —B1,Zit,vol.10,p.145,1889. X—TRLEGRAPEY AND TELEPHONY. ‘Yaseur—Thoory ofCurrentPropagation inaLine—Awn. Tal,vol,16,p.481,1889, 43.Moots:—Experiments onMicrophonic Contacts—Lum. 1,vol.81,p.51,1889, 336 ARTICLES RELATING TOELECTRIOITY, Ere. Anox.—Webr's Lightning Discharger.—Iat. B1,volBl,p.454,1889,‘Awox.—Binding forStlicoousBronzeWires—Lum,El,vol.81,p485,18801B.Zerscur—New Multiple Telephone Switch-Boards—Lam. El,” vol. th ,656, 1669, 45.Kannis—The Btockholm Telephone System.—Lum, Bt,vol. 81,p.608, 18. ‘Wauma—Telephony inEngland.—B, Zeit, vol.10,pp.140, 164, 1680. x1_THzZORY, E,F,Humnoux—The Divergence ofElectro-motive Forces from Electo- chemical Data—Phil. Mo., vol.27, p.208, 1889, J.W.Grms—Comparison ofthe Electric Theory ofLight and sirW. ‘Thomson's Theory of Quasi-labile Atther—Phil, Mog., Vol. 27,p38, 1889, ‘L,Porrcanni-—Conductivity ofFused Salt—C. R,vol. 108, p.188, 1890. B,L,Tovvavor—Application ofPhotography tothe Study ofInduction ‘Phenomens,—C, R,,vol. 108, p,846, 1889. ©,Dronanxcx—Difleronce between theso-called Positive and Negative Ele tricitienLun.BL,vol.31,pp.401,478,66,1889. L.A.Lonrxra—Theory ofThermo-Blectriity.—dnnalen, vol.36,p.693,168. 1M,Piaxtx—Theory ofThermo-Electricity inMetallio Oonductora—dxnale, ‘vol. 88,p624, 1889, BR,Wenonaxx and H.Horst—Electrio Discharges.—Anaalen, vol. 86, 61S 1889. 1H,Hasre—Rays ofEletricity—Annaten, vol86,p.760, 1889,43.Brnoxarx—Eihot ofHeatonConductivity —Awnalen,vol.86,p.768,168.8Tanzscuix—Specific Inductive CapacityofsomeOrganicBodies—Analem vol, 96,p-792, 1889. Hi,Bicksrasu—OonductivityofSpecalarIronOre,—Beibitte, vol13,p.17, 1889,1,Pauict—Potential ofaConductor movedinaMagneticField.—Bailie,vol. 18,p.192, 1889,‘A.Riont—New Phenomenon producevyRadiation—Doiatte, vol1%1.198, 1889. ‘XI1—VARIOUS APPLIANCES. L.Daniox—Method ofDiffusionofanBleotricCurrentintheHumanBoly— .R,vol. 108, P.881889, B,Zarscur—Profesor Kleiszner's Electric Calendar Olock.—Lum, Bt,vol.8s .419, 1889.‘Anox—Electrosmagnatio Lighter.—Lum, Bl,vol,1,P.488,1860, or uz Institution ofElectrical Engineers. Founded 1871. Incorporated 1889. Vou.XVII. 1889. No.80. TheOneHundred andNinety-first Ordinary General Meeting of theInstitution washeld attheInstitution ofCivil Engineers, 25,Great George Street, Westminster, onThursday evening, April 11th, 1889—Professor W.E.Avrrox, F.RS,, Vice- President, intheChair. Theminutes oftheOrdinary General Meeting held onMarch 28th were read andapproved. The names ofnew candidates for election into the Institution ereannounced andordered tobesuspended. The following transfers were announced ashaving been approved bytheCouncil :— From the class ofStudents tothat ofAssociates— Bernard M.Jenkin. |Leonard Newitt. Donations totheLibrary were announced ashaving been received since thelast meeting from the Director-General of Italian Telegraphs, theInstitution ofCivil Engineers, andA.R. Bennett, Member, towhom the thanks ofthemeeting were heartily accorded. ‘The Cuamaan: The paper toberead this evening ison “Underground Conduits and Electrical Conductors,”byJohnB. Verity; butIamsorry tosaythat Mr.Verity hasbeen suddenly VOL. XVIII. 24 238 UNDERGROUND CONDUITS (apea, seized with illness, and isunfortunately unable tobepresent to personally give usthepaper, which willberead bytheSecretary. ‘The following paper wasthen read bytheSECRETARY :— UNDERGROUND CONDUITS AND ELECTRICAL CONDUCTORS. ByJonw B,Verrry, Member. Myobject inbringing thissubject before you this evening is topromote aninterchange ofopinion and experience concerning oneofthefewremaining problems connected withthegeneral distribution ofelectricity inour cities. Since Ibegan towrite this paper onmyreturn from America, much information has appeared intheelevtrical press, andtheclose attention now being paid tothematter inAmerica isequalled byalike spirit ofinquiry here. The outcry against overhead wires has been gaining inforce; and although thenumerous objections urged aredevoid ofsolid grounds, there isageneral feeling that anyconsiderable extension ofoverhead cables forelectric lighting isimpossible, both onaccount ofthedifficulty, and moincon siderable expense, inprocuring way-leaves, and also the attitudetakenupbytheBoardofTradeandotherstoputastop tothem. Overhead wires inthiscountry may therefore besimply considered asatemporary measure toadmit ofthe electri supply companies getting towork, and permitted only onthe distinct pledge oftheremoval ofsuch wires within acertain period, The time allowed forthe execution ofunderground works in Provisional Orders andLicenses, although aslong aperiod as couldbeobtained, isalltooshort,andathorough discussion of the subject atthepresent time, initsnumerous aspects, is highly advisable toenable ittobeafterwards promptly and adequately dealt with. Inthree months wemay reasonably conceive that certain electzic supply companies willbeunder legal obligations toat once facetheproblem ofputting high-tension electric conductors 1660) ANDELECTRICAL CONDUCTORS, 339 underground. Itwillbenopetty attempt narrowed down toa small area, but intwo special instances the undertaking is proposed tobecarried outonalarge scale bycompanies of adequate financial resources, and with supply stations already erected orinprocess oferection. “Isfailure probable?” may seem asingular question toput here, but itisanecessary one, leading uptothe various difficulties tobeencountered, and,inviewofthealarmist state-ments attherecent Electrical Convention inChicago, onethat deserves prominence. Mr. Lynch's paper giving theresults of theworking ofunderground cables, and which waspractically theonly part ofthe meeting reported inthe English papers, showed but one side ofthe case, and from the discussion fol- lowing it(and which appeared intheElectrical World, March 2) much was tobelearned. Many ofthe companies operating underground cables had from some cause orother notgiven their opinions, and the failures alluded towere inseveral instances traced todefective conduits, insufficiency ofinsulation, poor insulation, imperfect jointing, and cheap work. Cheap work has ‘damned electrical enterprise before now, anditbehoves usallto ‘guardagainstarecurrence ofit.Anything alsoonthesubjectof conduits and underground electric conductors emanating from American electricians hastobecarefully sifted, asnoteven are oar Unionists and Home Rulers more defined than certain “overhead” and “underground” men inAmerica, The overhead ‘men accuse alltheunderground men ofbeing interested insome ‘ofthedozenformsofconduits putforth;theunderground men retaliate bysaying, “You don’t want togoundergroand,” and “You boldly saywecannot gounderground.” InEngland westand onadifferent footing altogether: if much more electric lighting work istobedone wemust go underground;andhavingrecognised that,howcanitbestbe done? The necessity and best form ofconduits arethe first, Points fordiscussion. ‘Thesuccessful working efasystemofunderground electrical cables inacity mustbelinkedtoanarrangement ofduets,asthe impossibility, without such, ofgetting atthecables forrepairs, uo UNDERGROUND CONDUITS (Apel10, inspection, tapping, andalterations isevident. Even ifthelocal authorities remained quiescent during any continual disturbance ofour streets onaccount ofelectrical conductors, commercial interest must prevent such occurring. ‘The expense ofopening upacrowded thoroughfare tolayunderground cables isinmany cases asgreat asthecost ofwhatever isbeing laid, while by using aduct this expense, once incurred, should not beagain necessary. Again, ifthecable belaid intheground without « duct, itisnecessary toarmour it,and layitdown atfirst equal to allpossible requirements, otherwise anyextension ofthelighting beyond thecapacity ofthe cable would necessitate arepetition ofthewhole original work. Inthis case thefirst. costofsuch& large cable would benearly asgreat asthesmaller cable andduct together. Finally, thelifeofthebest cable forhigh-tension currents cannot atthepresent bepredicted; and this, together with theother reasons, undoubtedly proves that aconduit with s drawing inandoutsystem iseminently desirable. ‘Anideal system forunderground work isabrick subway, well drained and ventilated, and ofsufficient size topermit ofaman working init. Failing this Utopian arrangement, what comes next best? InNew York, where thematter hasbeen placed in the hands ofaBoard ofElectrical Control, they start outby saying that aconduit orsubway forelectrical conductors is nothing more than amechanical protection forthewires within it,andaconvenience forplacing orputting them underground. Intheir experience ofthree years the Board ofElectrical Control have naturally tried andabandoned many conduits. In 1room oftheTelephone Building, New York, occupied bythe SubwayConstruction Corporation, thereisaheapofexperience— experience gained atnolittle cost—that itwould beofadvantage forengineers who lightly propose undertaking such work to see and meditate on. Wooden conduits, bitumen conduits, asphalte conduits, earthenware conduits, samples ofundergroand ‘eonductors—all are here inwhat istermed the museum, and afford acollection offailures most instructive as towhat not todo. Inmany instances, however, the conditions there would not. 1680) ANDELECTRICAL CONDUCTORS. a apply toEngland. Atpresent wehave nosystem ofsteam- heating with leaky pipes, causing trouble wherever they go. Again, here wehave notthose great extremes ofheat andcold, thefrost. compelling conduits tobelaid atanaverage depth of four feet, and theheat causing various troubles with asphalte and bitumen conduits. Atthesame time, from thenumber ofmiles ofconduits laid down there should besomething tolearn, and Itherefore pro- pose instancing some ofthemore important conduits known in America and elsewhere. Conduits are divided into two classes—the so-called solid conduit,andthehollowconduit, Solidconduits areuselesscon- sidering here, asthey practically consist ofcables bedded in asphalte orbitumen, which means ripping upthe ground and breaking the enclosing mass ifalterations arerequired. With hollow conduits, the difficulties tobeencountered seem tobeto provide @smooth channel with convenience fordrawing inand ‘oattoprevent explosions through accumulation ofgas, and to keepthem reasonably dryandwater-tight. Inthediscussion atChicago itwasstated that inoneinstance itwasmecestary tokeep ablower going toventilate aniron ‘conduit, while several authorities appear toconsider itadvisable ondrydaystotakeoffthemanholecovers.Accidents have,of course, occurred from conduit explosions, and the engineer of theUnited States Electric Lighting Company instances onewhere themanhole covers ofthe conduit were blown over theroof ofa four-storied building. But inEngland, again, thework done bythegascompanies isfarbetter, andtheleakage inour streets isnotinany way tobecompared with thequantity of illuminating gaswhich wanders, fancy free, through thesoilof many ofthecities intheStates. Itappears desirable that conduits should beventilated; and theform adopted, where possible, istoventilate them bymeans ofapipe running into thebase oruptheinterior ofthestreet lamp-posts. With regard tokeeping them water-tight, this isanother constructional question, surely capable ofbeing Properly dealt with. . Ey UNDERGROUND CONDUITS (april110, ‘Manyofthecomplaintsofleakyconduitsandultimatefailure ofcables Itrace back tothe Dorsett conduit, ‘This, oneofthe earliest forms ofconduits, isbuilt ofblocks formed ofacombination ofcoal-tar pitch and fine gravel, cast with tubular openings 2}in.indiameter running through them from endtoend. Theyre jointed bypouring soft.mastic intothe cracks between them, and the blocks made toadhere tothisby warming their ends with hotirons and allowing them tocool, after being well set. Toprevent themelted mastic from closing the ends ofthepassages, tubular pieces ofpaper areinserted, making asortofinternal sleeve coupling. ‘Theconduit soformed terminates inbrick manholes atthe street crossings, The objections tothisconduit arethat itisbrittle, porous, inelastic, andcracks with changes oftemperature, sothat itisnotlikely to bemade water-tight with average workmanship. Creosoted wood conduits have been extensively used inthe ‘States, andhave been aleading cause oftrouble and failure with lead-covered cables. Achemical action takes place between the crude creosote andthewood, setting freeacetic acid andearbonie acid gas,which, re-acting onthelead pipe, converts itssurface into afilm ofacetate oflead; theaction continuing until the entire sheathing isconverted intocarbonate oflead, orwhite lead. Creosote isalsosaidtorotrubber. There are,ofcourse, waysof overcoming such troubles; butasIdonot think weshall be disposed tousecreosoted wood conduits inthis country, the matter need notbegiven much prominence here. Thedifficulty ofobtaining aconduit that shall beanything morethanamechanical protection forthecablesiscertaintobe great, as,however good the insulating material forming the conduit may be,and however water-tight thejoints, there is» difficulty inexcluding dampness unless itispractically air-tight, asdamp airenters theconduit and condenses onit. Several conduits have been made, however, with aview to afford electrical aswell asmechanical protection tothecables. ‘The Callender bitumen conduit, forinstance, iswell known here. Itisbeing extensively laiddown bytheChelsea Electric Lightiog Company, andfivehundred yards ofitisbeing tried bythe 100) ANDELECTRICAL CONDUCTORS, 43 Metropolitan Electric Supply Company. For high-tension carrents itsuseisdoubtful, butperhaps inthediscussion some- thing may besaid onthis point. Vitrified 3-in, sewer pipes laid inconcrete, with cement joints, have been used insome instances forunderground work. The Lake conduit, shown inthe illustration (Fig. 1),ismadefromthe beststoneware clay vitrified andwell= glazed. This conduit, said tobeused with success bythe United States EPP] Electric Lighting Company atWash- CEU] ington, isconstructed withsixcom- Fro. partments, each being 2}in.by44in. Itisdelivered inshort lengths,andthejointsareprotected bystoneware coversset,in coment. TheDoulton conduit (Fig. 2)consists ofaglazed earthenware pipeinwhich thecables arelaid, separated bytoothed insulating Cr EI©6 <= SS Foo. 2 partitions. Aspace atthe bottom drains offthe condensed moisture, which isgotridofbysuitable traps, ‘The joints are ade like ordinary drain pipes. Aserious objection tosuch conduits seems tobethejoints, Which aremade with cement. Itisvery difficult tobreak a cement joint without alsobreaking thepipe; andasitmight be necessary attimes toputinanew length, this could notwell be done, Again, therigidity ofacement joint isadisadvantage, as inlaying earthenware pipes it,becomes necessary attimes toalter thealignment. TheHurlbut conduit system (Fig. 3)wasdesigned toovercome thisdifficulty bymeans ofaflexible joint. ‘Thejoint isshown in thediagram. Crepresents asbestos gaskets recessed into the coupling and resting against the pipe; Bisacavity formed Ey UNDERGROUND OONDUTTS fara, between theasbestos gaskets, and isfilled with apermanently plastic sealing material;Arepresents openings throughwhich OFgh cA :HL| Se |fad || KIC il \aD Fro, 8—Hoxzacr Fusxraus Coxocr. AAsbestonGaskets,BPlasticSealing.CHolesforlinginplasticmaterial. tthesealing material ispoured. Itisclaimed that such joints will always remain both flexible andhermetically sealed. Mr.B.Verity's conduit (Fig. 4). This isasystem ofglazed stoneware clay conduit, with separate channel foreach conductor. The joints aremade asshown ondiagram, thedouble cone-piese beinginserted beforethejointing material isrunin,Thespecial object ofthis conduit istoenable bare conductors tobeused, CeT ee )fla Fra. insulated onlyatthemanholes, orwheretheyarelikelytobe handled. Experiments have proved’ thisform tobecapable of 983 ANDELECTRICAL CONDUCTORS. a6 standing almost anything mechanically when laidonafairly good bottom. Mr.Verity hasalsopatented asystem ofglass tubes in ironpipes forthesame purpose. Pipes ofglazed earthenware with several ducts form agood mechanical protection, being capable ofresisting considerable crushing pressure; while the glazing both inside and outside provides also, under certain conditions, afirst-class electrical pro- tection.Ifitbedesirablethattheconduitshouldbeformedofa non-conductor, then nothing could bebetter than such glazed earthenware, which also isofnogreat cost. ‘The difficulty of maintaining good joints with earthenware pipes hasbeen referred tosome time agoinconnection with conduits fortelegraph pur poses, several instances having been given byMr. Preece and thers, ofroots oftrees, branches, and other vegetation having made their way through thejoints. Isthis capable ofbeing orercome byaflexible joint, which should atany rate prevent joints giving through expansion andcontraction? ‘The great disadvantage, however, totheuseofearthenwarepipesinmanyplacesinacitylikeLondonistheamountofspace required forthem;infact,thedifficultyoflayinganyconduitat allinsome ofourstreets will befound simply enormous. Asamechanical protection only, itisimmaterial whether the conduit beanon-conductor ornot. The simplest form ofsuch ‘duct is,ofcourse, aniron pipe; and theWestern Union Tele- gph Company, among others, have Inid cast-iron pipes of4in, and5in,diameter, with manholes atadistance ofabout 450 ft. apart, Inthe majority ofcases, however, theplan has been to layanumber ofseparate ironpipesinaconcrete bed,bywhich rangement thepipes can becrowded, orcurved,orkeptapart, ‘%may best overcome any difficulties orobstacles met with underground, Forinstance, four pipes may belaidonthesame levelintwolayers; orwhere only anarrow excavation canbe ade, thepipes canbecurved round ‘sothat twopipes areabreast, infourlayers, Inthis way thetrenching isoften materially diminished. Screw-jointed asphalted wrought-iron pipes of2 or3in,diameter laid inhydraulic cement concrete aresaid to Present thegreatest tightness ofduct against gas andwater, a6 UNDERGROUND CONDUITS [aprilt, together with thegreatest strength. (Illustration given—Figs 5 and 6.) Ba es nL AA AS Fro. 5. Cement -lined sheet-iron pipes, however, possess some advantages, such, forinstance, assmooth interiors, and the messspraeamgem svantage ofanironenclosure,without [REC RCCL] thecablebeingbrought intocontactVOOe withit;andonthesheetironbeingeece eatenaway,thecementstillretainsthe hanW@1@)>| pipeform.Zinetubes,again,havebeen ©oe‘]used,butarelinbletobedentedinand asphalted concrete isoften preferred to hydraulic cement concrete. - ‘Whateversystemofpipesbeused, <uitisimportant that they should bare smooth interiors andclosejoints; and Fie. 6.every conduit, ofwhatever kind, must be thoroughly well bedded, soasnottobeliable todisplacement from above. ‘Aningenious conduitcore(Fig.7)hasbeendevisedbyMr Chenoweth: awooden cylindrical rod14or20ft.inlength is ccutintwo, forming twohulf-eylinders, and thespace removed is ‘occupied byaniron rodhavingathicknessequaltotheportion sawed ont, This forms the inandrill orcore shown inthe illas- 182) ANDELECTRICAL CONDUCTORS. aur tration, Aribbon ofgalvanised iron 1in,wide andathickness of No.27gauge iswound spirally round this from end toend, securing theendstothewood toprevent > Afterpaintingtheoutsideofthecore<>)Zs withamixture ofclay, soapstone, and SSG water, thecore isplaced intheditch on SS crossed pieces ofwood. Ifmore than one Fro.7. duetistobeconstructed, other cores areplaced side byside; concrete isthen well tamped round, bringing thetoptoaneven level, ‘The iron rodisthen removed,aswellasthetwopieces ofwood, and the fastenings ofthe ribbon ends, When the cement hardens, theiron spiral canbedrawn outatanymanhole crunfinished portionofthework.Itisclaimedthatthesurfacecating ofclay andsoapstone adheres totheinterior oftheduct andproduces asmooth surface, while amonolithic structure is camred, water-tight byreason ofitsconstruction. Ithink this isanother instance ofAmerican ingenuity. TheJohnstone ironconduit, shown intheillustration (Fig. 8), issaidtobeverysuccessful, andIsawportions ofitasdesigned to belaidinNew York City. Itismade in isectionsabout6ft.long,andhassixducts. wm,Thethreeloweronesandthecentralduct mVoathetopcouldbeusedformains,andthe My twoouter topducts forhouse-to-house and “systreetlighting. Whereanyhousecircuit,Lf iarequired thetophalfofasingle section @ypy) istaken, and anew half-section with a Fro.8 toleinit,towhich anelbow canbebolted, isfixed initsplace. Aform ofconduit made byMr. W.E.Irish, ofCleveland (Fig. 9),isalsoworthy ofnotice. Intheillustration theconduit consists ofatubular pipe cast insections, with flanges which are recessed ontheinside toreceive arubber ring, formingagasket between thetwosections when bolted together. The conduit is provided with alongitudinal slot oropening along thetop,with inwardly inclined sides andflatsurfaces oneither side oftheslot, towhich thecover issecured byscrews, ‘Toprevent theentrance us UNDERGROUND CONDUITS (apc110, ‘ofmoisture through thisslotastrip ofrubberorleatherisfint laidonthetopoftheflanges, extended from oneouter edgeto theother. Onthisasolid wedge-shaped piece which fitsintothe . eae. Fr.9. slotsnugly isplaced sothatonthecover being screwed down the wedge, with thepacking surrounding it,isforced intotheslotso astoseal itperfectly. Itisclaimed that there isgreat conre- nience inlaying wires with thisconduit, andthat anywire canbe tapped atany pointbysimplyremovingthecoverandmaking thenecessary connections. The illustration representstheconduit with abranch, also ajunction box, &. Diagram Fig. 10represents Mr.J.E.H.Gordon's system, 8 ‘usedbytheMetropolitan Electric SupplyCompany. Thecablesaredrawnintoanironpipe,andsmallerironpipesareledfrom Fr. 10, ‘thehouse junction boxes tocarry thewires forhouse lighting. Testing boxesofalargersize,andbuiltofbrick,areprovided at ‘intervals, Itisgenerally admitted thatanimportant point inunder- ground work isgood manholes. Mr.W.D.Sargent, in»paper read before theNational Telephone Association, 1888, saidthst 198) ANDELEOTRIOAL CONDUCTORS. 9 within reasonable limits itisalmost impossible togetthese man- holes orworking chambers toolarge, and that they have been compelled inmany instances torebuild and enlarge them after some time. Thereisnodoubttheymustbeofgoodsizeforconvenience ofdrawing inandout,which isoften adifficult matter, especially with lead-covered cables. Inoneinstance, inChicago, theman- holes arebuilt inoctagon shape, inbrick and cement, about 4ft. square, being made air-tight with screw-head andrubber gasket; andtesting boxes placed inthem, sothat troubles canbereadily located, faulty wires drawn out, and good ones putin. Inother cases themanholes vary from 3{t. 6in.to6ft.deep, with iron curbs andtops, and cement bottoms,andthecoversmadepractically tight against water; many ofthelidsdraining thewater from the edgeandcollecting itinthecentre, thus diminishing theliability toleakage atthegasket. With theJohnstone andtheIrish con- datsitisnotnecessary toprovideforhouseconnections, asthe alteration canbemade atanytime afterwards when anapplication forlight isreceived, But, asarule, hand distributing boxes, ade ofcast iron, with screw-head and rubber gasket, arefixed, onebetween every two houses, and after jointing, theboxes are often filled with bituminous orinsulating compound. Inthe cement-lined pipe conduit system one pipe isoccasionally placed atthe topforthehouse distribution, andconnections made with themains atthemanholes, Astothefrequency ofthese latter, much depends onthealignment oftheconduit. Atangles or changes ofgrade small manholes arenecessary. ‘Thecombination curbandgutterconduit(Fig.11)showninthe illustration isworth notice. The EIT.curbandgutterismadeoffine gesconcrete,onepartofPortland GREE. acement,tenpartscleansharp2Bie|4‘and,andthreepartsofbroken \52 RRM” stone, and the exposed surfaces | coated withagranolithic mixture we 1}in,thick, such asisnow ured Fro11, forsome ofourLondon pavements, Aconduit forelectrical con- 00 UNDERGROUND CONDUITS [aprilane, ductors ofany reasonable size can beformed asshown inthe inner angle ofthecurb, with hand-holes from above closed with iron covers at intervals. The illustration istaken from the admirable report onunderground wires byMajor Raymond, the engineer commissioner fortheDistrict ofColumbia, and which contains valuable information onmany pointe. Itwassuggested some time ago that thecurb-stones should beremoved from busy thoroughfares and replaced with ahollow carb edged with cast steel, and thesuggestion may bear fruitin the future, AsIstill have todeal with electrical conductors, Icannot here refer tothegeneral cost oflaying conduits, asatonetime Iproposed doing. Tho principal cost, however, oflaying them isinmaking good, andthisdepends onthematerial oftheroad, thecheapest being macadam (varying from 5s.persquare yard, whereas asphalte would benearer 20s. persquare yard). The difficulty oflaying anyform ofconduit insome ofour London streets must, asIhave already said, bevery great. ‘The last comer will ofcourse feelthis themost; and although the Hydraulic Engineering Company may have recently had much trouble tofind room fortheir pipes, the electric supply com- panies, again, will have aneasier task than the Telephone Company, who areputting offtheevilday. Astowhat form ofconduit ismost suitable for London and ourprovincial cities there will bedifferences ofopinion, butI trust thediscussion following will narrow itdown to small issue upon which such work canbeundertaken with atolerable surety ofitslasting success. There isreally sufficient subject-matter forapaper on conduits alone, and intaking upthequestion ofunderground conductors Ipropose tochiefly restrict myself toaconsideration ofconductors forhigh-tension electric lighting circuits. With regard also tounderground conductors forlow-tensioa work, Ifeel that the Edison system,asatpresentlaiddown,is thoroughly capable offilling allthe requirements ofhouse-t- house distribution. Many companies intheStates andelse- where areusing it,andatthedown-town station inNew York 1880 ANDELECTRICAL GONDUCTORS, aL ithasbeen atwork forthepast sixyears successfully. Inthe recent discussion onProfessor Forbes’s paper several points ofinterest concerning the Edison underground conductors were explained, which itwould beunnecessary forme here torepeat. Afull description ofthe latest developments of this system, with samples ofthe improved form ofconductors, 4c,arenow ontheocean, and intheevent ofitbeing possible toprovide another evening foracontinued discussion ofthis peper, Ishall bemost happy to'show andexplain them, Everyone must acknowledge theforesight shown indevising suchacomplete systemofunderground mainssolongago,and which, in spite ofthe all-round developments ofelectrical engineering, not only holds itsown, but isstill farahead in completeness ofany system forlow-tension underground work. ‘The Westinghouse Company have, Iunderstand, acquired the rights forGreat Britain, andpresumably willuseitinconnection with large transformers for sub-centres. Itmay bealso interesting tonote that Messrs, Siemens have made concentric three-wire cables, which arebeing successfully used atElberfeld, Geneva, and Muilhausen, and they state that thetroubles from what would appear very complicated jointing have been satis- factorily overcome. Itisunderground conductors forhigh-tension currents that I propose now todeal with. Aprincipal point forconsideration is thenecessity ornotforlead sheathing where conduits orducts areemployed. Astothedurability oflead much absurdity hasbeen talked. Itisnotorious that lead pipes have been intheground under all manner ofconditions, and attheend oflong periods arestill intact. There arenodoubt certain clayey soils where lead has notmuch lasting power, butthelead may bematerially pre- served bycoating itwith coal tar, well tapeing it,and again soaking itinthecoal tarorsome such protective composition, Again, asDr.W.vonSiemens recently saidinreply toProfessor Fotbes’s paper, themanner inwhich theleadcovering isgenerally utonsuch cables, viz., byheating thelead tothepoint of melting, andforming iinthis condition round thecable, while 358 UNDERGROUND CONDUITS (apelin, passing through thepress. ‘There isnocertainty of#uniform thickness,anditisverydifficulttoobtainatubethroughwhich damp, incourse oftime, will notfind anentrance, asairbubbles andimpurities inthelead areliable tomake thecovering porous. Afarbetter result isobtained bycovering theinsulated cable with acold-drawn lead tube, asthe airwhich might still be contained inthelead block iscompressed toaminimum. Atheory astopartial wasting oflead covering oncablesis thatitmay occur from acid being present inthejute orfibrous insulation, or,again, that thepetroleum residuums may setups chemical action onthe lead; but there has not been much evidence onthese points. Manufacturers oflead-covered cables have admittedly put ontoothin acoating oflead inthe earlier days; this hasbeen gradually increased, and with proper conditions and precautions there should notbemuch fear ofits durability. Iron-armoured lead-covered cables arenot necessary ifan effective system ofconduits isused formechanical protection. InBerlin, where such cables areemployed, they aresimply laidin atrench intheground, andconsequently require tobethoroughly protected from mechanical injury. Lead alone isnotsufficient. Dr. Werner v.Siemens said inhisrecent paper that long years ofexperience hadtaught him that although covered with hemp orjute itrequires further protection torender itsecure against mechanical injury, whether bymen oranimals, aseven rats eattheir way through. Such protection isafforded by4 double spiral ofsheet iron, which, again, ismade secure ageinst oxidation byasphalting orgalvanising, orbyanother coverof tarredhemp orjute. Such armouring asthis isquite needless with awell-constructed conduit; butastowhether galvanic action takes place, Dr.Siemens hasundoubtedly shown byhisfirm's experience that with asubstantial Inyer oftarred orasphalted jute between the metals itdoes not occur. This form of ‘armoured cable waslaidatMunich forlighting thetheatre,and after nearly five years inthe ground isasperfect asatfirst. Atother central stations atElberfeld, Darmstadt, Geneva, Salz- burg, Lyons, TheHague, St.Petersburg, their iron-armoured lesd- 16803 ANDELECTRICAL CONDUCTORS. - 338 covered cables prove that, constructed inthis way, there need be nofearofgalvanicactionatanyrate. ‘Lead sheathing may bemade durable, buttheactual necessity ofitwhere aconduit isused isquite another matter, and a ‘consideration ofthetwoprincipal divisions ofinsulating material hasconsiderable bearing onthispoint. Ofthe fibrous and the homogeneous the latter seems to present some important advantages forinsulating high-tension ‘ables.Forinstance, vulcanised rubberisflexibleandtough,and affordsacontinuous andhomogeneous covering whichshouldbe superior toafibrous covering. Vuleanised rubber cables areat times sheathed with lead formechanical protection, but such is notnecessary iflaidundergoodmechanical conditions, andDr. Lant Carpenter recently said that the sulphur inthe vulcanised rubber would attack thelead inthesame way itwould copper. If,asisstatedbymanyEnglishcablemanufacturers, compound valeanised rubber willwell withstand thedeteriorating effects of ‘damp and gases, then, themechanical protection being afforded byaconduit,thenecessity ofleadasaprotector isnolongerof importance. Itistrue unprotected vulcanised rubber cables in conduits have failed intheStates, butthecompetition among dectrie light companies inAmerica isnotorious, and the questionastothemaintenance ofcablesisnotconsidered somuchastheir relative first cost. The object ofthe American cable manufacturers, naturally suiting their market, hasbeen tomake ‘ables with acheap form ofvulcanised rubber insulation to compete with thejute-insulated cables there, and which are cheaper, The vulcanised rubber evidently was not properly Prepared asitshould be topermanently counteract the deteriorating effects ofdamp andgases byreason oftheexpense. Allthose acquainted with therudiments ofrubber manufac- taremust beaware ofthe enormous difference inquality of different vuleanised rubbers. With Para rubber at3s.perIb., andthedesire tosell itatashilling, allmanner ofsubstitutes andcompounds have been introduced, While thoroughly appreciating theenergy andgenius ofAmerican electricians, we ‘anconfidently saythatcablesmanufactured inthiscountryareVou, xvi. 25 354 UNDERGROUND CONDUITS (apa, greatly superior tothose made inthe United States; andass basis for this beliefwehavethefactthatmostofthecablesfor high-tension distribution inthe States have proved faulty, according togeneral admittance, while there isundoubled ‘evidenceofthesuccessful workingofmanyvuleanised rubberand jute insulated cables inthis country and inEurope, made by Englishmanufacturers. ‘Thestatement ismadebythe Silvertowa Co., Messrs. Henley &Co., and others, that they areabsolutely prepared toguarantee foraterm ofyears their valeanised rubber cables unprotected bylead, even iflaid indamp and leaky conduits, provided there issuitable mechanical protection against injury, and that thecables arelaid under competent. supervision. With regard tofibrous insulating material, itisalleged thatit +hasthedefect ofopening cracks when bent. Ifthis bethecase, itwould become more pronounced when the material issoaked inresinous compound, and perhaps less sowhen petroleum ‘residuums areused, such as,forinstance, inozite, kerite, bitite, &c, Ifcracks arelikely tobeformed when the cable isbeat ortwisted about, airpaths would beopened upfordisruptive discharges. Intheworking directions given forthelaying ofoneofthe principal lead-covered cables intheStates, itsays: “The fibrous “material, beingsuperdried, willreadilyabsorbandretainmoisture “from the air, and cables showing aninsulation resistance of «thousands ofmegohms when firstmade mayafter exposure of8 «small partofthecoretomoist airforafewhours show agreat “falling offofinsulation resistance.” Instructions aretherefore given astocarefully sealing theends ofthecable when laying. Again, thesplicers’ hands arecautioned tobekept perfectly dry, “and free from perspiration, asalittle moisture communicated “to theinsulation may result invastly lowering theinsulation “resistance.” These remarks written bymanufacturers ofsuch ‘cables show thereadiness with which such fibrous insulation may deteriorate under conditions likely tooccur even ifthere bethe ‘most careful supervision ofthe men laying thecables. Itis acknowledged tobeabsolutely necessary tocoverjuteorcottoninsulated cables with lead, anditisinthis fact that theweakness 1680.) ANDELECTRICAL CONDUCTORS, 385, ofsuch cables seems tolie,asthemoisture- and gae-resisting portion isthelead absolutely, thejute insulation being oflittle value inthis respect. Inasmuch asthe majority ofunder- ground cables intheStates are insulated with such jute or cotton soaked insome compound, itwill beatonce understood whylead-covered cables areconsidered inAmerica astheonly ‘means forsuccessfully carrying outsuch underground cable work, expecially when thecables areliable tobelaid inleaky conduits. Itisright tosaythat apparently fewtroubles have been ex- perienced inEurope with underground conductors insulated with fibrous material. Messrs, Siemens &Halske, who useajute- covered cable, but manufactured inadifferent way both as regards theimpregnating ofthejute and themethod ofcold- drawn lead sheathing, have asuccessful record, not only forlow- tension cables, but forthevarious concentric cables supplied to Mesers. Ganz &Co., ofBuda-Pesth, for2,000-volt alternating carents. They assert that the troubles experienced with American underground cables, whether jute- orrubber-insulated, arecomparatively unknown. The question between what Ibroadly term lead-sheathed jate orwell-compounded vuleanised rubber seems tomethe ‘momentous one, and, asusual, each one has certain claims. Itiswellknown thatthecostofjute-insulated cablesisless than that ofhigh-class vulcanised rubber insulation, butthis would have little weight unless they can beproved inevery respect equal tovuleanised rubber. Inacomparison ofthe different costs ofsuch insulation, vuleanised rubber appears to beabout 25percent. more forapproximately thesame insulation resistance. Inmaking this comparison Itook the cost ofa vuleanised rubber cable, insulated with one layer ofso-called pare, andtwo coatings ofvulcanised rubber, taped, braided, and ‘costed with preservative compound, butnotcovered with lead, as thisisasuperfluity, thecable being water-tight without this Addition, Thejute-insulated cable Itook ascased with lead, and ‘Protected again byacovering oftarred juteyarn compounded, After therelative costcomes thequestion ofjointing;andas thisisoneofthemost vulnerable ‘points inanycable, itisof 386 UNDERGROUND CONDUITS [Aprilut, vital importance that thejoints should bemade asgood asany other part. With vulcanised rubber cables thejoints must be vuleanised, and first-class rubber men who would make such joints speedily should beonthe working staff, With lead- covered cables aplumber would have tobeemployed, andIam imformed bythose whohave had experience with both forms of cables that thevulcanising isproferable totheplumbing. ‘Valcanised rubber cables arecertainly easier tohandle, being more flexible, ductile, andlighter, andthisisofimportance in view ofthedifficulties occurring indrawing cables inandoutof aconduit. Again, when lead-covered cables arecoiled round adrum the part nearest the centre becomes compressed, while theouter portion isextended; onuncoiling thecable theopposite effect, isproduced, theparts before compressed being now pulled out, and the extended portion compressed. This istrying toan inelastic substance like lead, the more sowhen, unlike an empty lead pipe forwater, itisfilled with aninsulated cable. Itisquite conceivable that the surface ofthe lead may be slightly brokenup,particularly whenthemethod employed isto cover thelead hotround thecable; andalthough nocracks may bevisible totheeye, gasand moisture will find their way invery quickly, withcorresponding deterioration oftheinsulation. Next, astowhether jute orvuleanised rubber isbetter suited forinsulating high-tension alternating cables noopinion canyet begiven. Itiswell known that onsubmarine cables working with reversed currents, itisusual attimes togive thecable 8 rest. This necessity may bedue totheaction ofthereversed current ontheinsulation, and itwould beinteresting toknow howthemuchmorepowerful currentsandrapidreversalsofanelectric light circuit will affect the insulation ofacable after carrying such currents foraconsiderable period. Insome interesting papers published last year byMr. Addenbrooke, he came totheconclusion that ahigh-tension alternating current should have lesseffect onanydielectric than acontinuous current ofthesame potential. Heargued that, owing totherapid alter nations, there would notbesufficient time tofully charge the 1980] ANDELECTRICAL CONDUCTORS. 387 dielectric, as,before this could occur, the next charge would sweep out the former charge, partially recharging itinthe opposite direction. But itwould certainly appear that such an action asthisonadielectric would befarmore injurious than for ittobekept inafully charged condition aswith acontinuous carrent. FromacarefulreviewofwhatIhaveseenintheStatesandelsewhere, Imustconfessthatmyownfeelingisinfavourof& ‘ableinsulated with thoroughly good vuleanised rubber forhigh- tension work;andIwasinterestedtohearSirWilliamThomson sayonMonday, attheBoardofTradeinquiry, thatifhigh-tensionconductors wereprotected inanironpipealeadsheathing tothem ‘ras not essential. ‘Those acquainted with the present manufacture ofrubber cablesknowthatpurerubberproperisnotusedfortheirinsula~ tion, However good therubber may be,there isalways acertain amount ofimpurity and oils, which will setupdecomposition woner orlater, This isovercome byslightly vulcanising the rubber, butsothat there shall benosurplus sulphur toattack thecopper. Inthehigh-class cables nowmade bytheSilvertown Company and others thepractice istoputnext theconductor after cotton twist asubstantial coating ofpure indie-rubber, slightly vulcanised, upon thisacoating ofzine rubber, with vul- ‘anising materials, and covering this acompound vuleanised rubber ofgreat resisting qualities todamp andgases. ‘Thezinc rubber takes upthesurplus vulcanising material ofitstwoouter coverings, sothat thecopper cableisnotinjuredbythesulphur wedintheprocess, Such aninsulation asthis isasgood as could possibly bedesired foranyhigh-tension currents, andis absolutely perfect solong asthecovering iscontinuous; andas thewhole ofthematerials aretreated atatemperature of400°, ‘ayrisointemperature ofthe conductor due toaccidental over- loading willnotinjure theinsulation, With regard totheouter protection ofthecompound vulcanised rubber, there may bestill scope forfurther improvement; and, forinstance, inthe cables Tecently made bytheSilvertown Company fortheLondon Electric Supply Corporation, twobraidings ofasphalted tape have been 388 UNDERGROUND CONDUITS [apeie, ‘used, and this isfound tobesatisfactory. Among theinstances where vulcanised rubber cables have stood stringent tests has ‘been atEastbourne, where Mr.Lowrie hasused Silvertown cables, laidinironpipes,fortwoyears,andstatestheyarestillinperfect working order. Hehasalsoputdown there alength ofFowlers Tatham lead-covered cable forcontinuous-currenit high-tension work, thefibrous insulation being well wrapped round severaltimeswithtape.Thishasalsobeenworkingsuccessfully ;andin the House-to-House Company's station heproposes totryeach method ofinsulation ondifferent circuits, although theadvantages ofhandling theindia-rubber cables areintheir favour. ‘The Silvertown Company have five miles ofvulcanised indis- rubber insulated cable sheathed with iron wires atBrussels, with 1,400 volts constant current. This was laid inthe sewers to ‘years ago, andisnow quite satisfactory. ‘This company have alsosupplied twomiles ofsuch cable, bat, with asphalted tape ontheoutside,totheLondonElectricSupply Corporation, with 2,400 volts alternating current;theconductors being laid inaniron pipe. There arealso samples onthetable ofapiece ofMessrs. Henley’s vuleanised rubber cable that has been inuse overbesd fortenyears. Adifficulty with underground conductors intheStates seems tobedisruptive discharges, and Itrust the discussion may bring forth information onthissubject. Owing tothefrequent pune- turing oftheinsulation, attended with pinholes inthelesd covering, Mr.Acheson, theWaring Company's able electrician, and whom Idesire heartily tothank, turned hisattention to thematter, asthepinholes, byadmitting moisture, defeated the object ofthelead sheathing, and proved fatal tothelifeofthe cable. Mr.Acheson's theory isthat thestatic electricity generated inthecable after atime punctures thedielectric bydischarging itself between thecopper conductor andthelead sheathing; and theprotector shownonthediagram(Fig.12)isdesignedtoobviate this. This consists oftwo metallic points separated byadistance 1) ANDELECTRICAL coNDUCTORS, 3 which isless than the thickness ofthe cable insulation. The one Pointisconnected totheleadsheathing, andtheotherthrough afase wire tothecopper conductor; theidea being that the static discharge would take theshorter path between thepoints C <= ee: EE" Fr 2 instead ofrupturing theinsulation. Mr. Acheson says these dischargers should beplaced atevery section ofacable, andif thesections aremore than 200feet long there should beoneat each end. Toshow the exceedingly minute path necessary foradis- charge, itmay not beout ofplace toquote Mr. Acheson's eaperiment, Aplate ofglass about one-tenth ofaninch thick‘msbrokenintwopieces,andthetwopartsimmediately fitted totheir former position and clamped. The restored plate was placed between twodischarging points, andsolong asthepoints vereover thesolid glass nodischarge was obtained ;butupon removing it60that theline joining thepoints would lieinthe plane ofthefracture, adischarge immediately occurred. Apossible cause ofadisruptive discharge might bethat on high-tension circuits ofgreat self-induction, as,forinstance, on ‘arelighting circuits wherelampcoilsformpartoftheline,the induced current onbreaking thecircuit isvery violent, and in Preference todischarging through theline, and consequently the lampcoils, itgets away byrupturing theweak insulation, and Passing through totheouter earthed casing. ‘Theexemption from such trouble iscredited bymany entirely toathicker insulation being used, and incalculating theinsula- tionforahigh-tension cable itshould bemade sufficient to withstand adisruptive discharge ofthis nature, and tested accordingly. For instance, owing tothealmost daily occurrence ofbum-outs with some cables inChicago with y¥insulation, laid inNovember, 1887, thefourmiles laidinDecember, 1887, hadsty 360 UNDERGROUND CONDUITS (Apri1, insulation, and the two miles laid early inMarch, 1888, hare yinsulation, Careful experimental work isessential, before the sufficiency ofinsulation forsuch underground cables canbe accurately determined, and our leading cable manufacturers naturally keep theresults ofsuch experiments tothemselves. Concentric cables, although soextensively used byMess. Ganz, ofBuda-Pesth, intheir stations, donot appear asyetto have met with favour inthe States. Itis,ofcourse, undesimble touseordinary lead-covered cables foralternating currents, both ‘onaccount ofthe loss from induced currents inthe lead sheath- ingand the annoyance toworkmen handling them. If,for instance, while handling aconductor conveying such acurrent, thelead covering beinsulated toany extent oneach sideof them, theworkmen would receive the induced currents through theirbodies.Again,whenthetwolegsofanalternating current tareinseparate cables, and thecables laid ininsulating materials, anychance contact orconnection between theleads might cause the induced currents topass backwards and forwards atthese points, thus tending tocutaway thelead covering astheresult ofany sparking atpoor contacts between the two leads. The Waring anti-induction cable, with both wires enclosed inone sheathing, iswell known, and inthis thecurrents induced bybothwires,beingequalinpowerandoppositeindirection, tendto neutralise one another. Mr.Ferranti, byemploying concentric cables forconveying ‘thehigh-tension currents from Deptford Station, hasbrought such cables intoprominent notice inEngland. Aconcentric cableis obviously thebest ontheoretical grounds fortheelimination of induced currents, astheir reaction upon oneanother isnecessarily strongest inthe axis ofthecurrent. ‘The question astothe necessity ofinsulating theouter conducter oftheconcentric cable hasrecently hadconsiderable attention given toitbyMr. Preece onbehalfofthePostOfficeauthorities, and,asaresultof hisexperiments, they evidently must besoinsulated. Allthe concentric cables made byMessrs. Siemens &Halske forMessrs. Ganz &Co.areoftheusual stranded-wire form, and theoutside conductor, after being well insulated with jute, iscovered withs 1889.) ANDELECTRICAL CONDUCTORS. mL cold-drawn lead tube, having anouter iron armouring aspre- viously described. ‘These cables have been successfully used in Rome, Tivoli, Turin, andMilan, inench case carrying apotential‘of2,000volts, The difficulty ofmaking satisfactory joints with concentric cables must betoacertain extent against them, asitisim- possible toseewhat hasbeen done inside. Messrs. Siemens say that itcan bereadily accomplished, and the illustration shows themanner inwhich thejoints aremade intheir cables. The Silvertown Company have made several forms ofconcentric cables, highly insulated with vuleanised rubber, with considerable mccess, forship work, and they also have anarrangement for constructing thejoints. Astowhether concentric cables will belargely used inthis country, much depends upon thesystem ofdistribution employed. Forlong lengths ofcable used asfeeders tosub-centres, orto systems ofdistributing mains, they would beofconsiderable advantage, asnojoints would berequired enroute: but in house-to-house distribution, withfrequent joints,apairofseparate ‘ables near toeach other would appear the more easy tohandle andthemore practicable tooperate. Insumming upthismatter itwould appear that the relative advantages areasfollows:— Ist.Thegreater convenience ofmaking joints onseparate leads,asagainstthetroubleinvolved withconcentric cables.2nd.Asregards the relative cost between two separate cables and concentric cables with the same insulation resistance, vulcanised rubber unsheathed concentric cables with outside insulation are dearer, butwith jute insulation forconcentric cables and lead sheathing they arecheaper. rd. There remains theincreased efficiency onanextensive system ofalternating currents bythe useofthe concentric form, With regard tothesecond point, thefigures showing these are:— Forvulcanised rubber, twoSilvertown 4%conductors, insulated ‘to5,000 megohms at60°Fah., taped and braided over all, specially prepared forunderground work, £255 permile each = £510, 302 UNDERGROUND OONDUITS, Ere. [Apei, Aconcentric conductor ofequivalent capacity, the internal conductor insulated to5,000 megobms,asabove,buttheouter to3,000 only =£900, being adifference against, theconcentric conductor of£390 permile. Ifthe outer insulation isreduced to1,500megohms, thepricefallsto£630permile,whichis£120 above thecost ofseparate leads, Ontheother hand, forMessrs. Siemens’s specially prepared jute insulation twosingle conductors ofsame insulation, at£220 permile each =£440, ‘Aconcentric conductor ofequivalent capacity, theoutside insulated to900megobms, lead-sheathed, taped, and asphalted, £363 permile. Allcables forhigh-tension currents should bewell tested before leaving thefactory. Inother branches ofengineering the ‘engineer would notthink ofreceiving hissteam orhydraalie pipes without first seeing them tested to,say,over twice their workingpressure; itisnotsufficient forhimtoseesimplythat they will hold water orsteam, but hemust satisfy himself that they arestrong enough forthepressure they havetocarry.In the same way the electrical conductors, which have totransmit energy much asthe steam and hydraulic pipes do,should be tested totwice their working pressure. With the outside well earthed the current should beallowed totraverse the conductor aconsiderable time, and anelectrical test taken afterwards for insulation, With electrical cables forunderground work passed inthis way, and satisfactory inother respects, there should beno fear ofbreakdowns under ordinary conditions. Indealing with asubject ofsuch wide extent and great im- portance asthisitisimpossible todojusticetoalltheworkers inthe field. Ihave left out much that may beconsidered interesting and important, butthedifficulty hasbeen, with the mass ofinformation onthesubject, what toputinandwhat to leave out. Doubtless there aremany instances ofcables and conduits having been successfully atwork which have notbeen mentioned inthis paper, and Ihope such information willbe forthcoming. Ifthediscussion following bethemeans ofclearing upsome ofthedoubtful points surrounding thesubject ofunder 1682) DISCUSSION, 363 gound electrical work, and ofdemonstrating thedirection in which successful work should beundertaken, theobject ofthis peperwillbefallyaccomplished, ‘TheCuamwan: Although Mr.Verity isunfortunately absent, Pero" hisassistant, Mr.Girdlestone, ishere, and will answer any ques- tionthat heean; though, ofcourse, themain body ofquestions thatarise inthediscussion willbedealt with byMr.Verity inhis reply, which willbepublished intheJournal. Inow invite discussion orcomment upon thepaper. There arerepresentatives ofthe Silvertown Company, and probably of ‘other cable manufacturing firms, present, and nodoubt they are desirous ofsaying something onthesubjects concerning them hich have been referred tointhepaper. Mr,W.E,Gray:Ihavelistenedwithpleasure tothesew.interesting paper wehave had this evening, and agree, with regard tothecables, with Mr.Verity onmany ofthepoints he lasraised;butasthisquestionofunderground cableshasso rently been discussed, Idonot think Ihave anything toadd atpresent which might beofinterest. Mr. Verity appears toconsider ittobeaquestion asto whether lead-covered orvulcanised india-rubber cables are the best. Naturally Ibelieve invuleanised rubber, butwould prefer aottouching onthismatternow,butleavingittobediscussedbyothers. Astoconduits, Mr.Hurlbert’s idea, judging from hisflexible joint, appears tobetoprevent moisture getting in. Itseems to methatanysystemlaidunderground whichisdependent upon theconduit itself excluding moisture would beafaulty one. Onsuch amatter wemight appeal tothePost Office fortheir experience inlaying wires underground, ortothegascompanies. Ibelieve that hitherto, inspite oftheingenuity ofmany inventors andimprovers, ithasbeen found impossible tokeep moisture out, oratallevents, moisture tosuch anextent as would materially affect anyquestion ofelectrical insulation. It should beborne inmind that very little moisture inafibrous material might easily constituteaseriousfault,owingtofailurein insulation; andhowever good, mechanically, theidea maybe,still I | 308 UNDERGROUND CONDUITS, Ere,[Aprilth, AcE think that,iftheinsulation depends ontheflexible coupling ot joint remaining water-tight, itisrather leaning onabroken reed. ‘Mr. Verity refers tothe work done bythe Wester Union Company. Ihave only aslight knowledge ofwhat hasbeen done bythem, but understand that the conclusion they have arrived atisthat New York isbuilt upon anexceptional soil,in which there aremany free gases and oils, and that thetrouble they have had isinagreat measure tobeattributed tothis cause. Forthisreason ithasbeen proposed that theinsulated cable should becovered with lead, the insulation tobeused being vulcanised rubber. Inspeaking onthispoint itshould be borne inmind that when inAmerica the word “rubber” isused, inmany cases “pure rubber” ismeant, andnotalways vulcanised rubber, asMr.Verity appears tohave understood it. Reference has been made tothe “Johnstone conduit.” In the very interesting discussion atthe Chicago Convention, Mr.Johnstone spoke ofthis conduit‘as asuccess. Mr. DeCamp, Tthink itwas, didnot seem tothink that this success (atleast inPhiladelphia) wasvery marked, as,when theconduit wastaken over, itwas found tohave been broken upinmany places, and there appeared tobeadifference ofopinion between thecontractor andtheworking company. Idonotknow ifMr.Johnstone has made any recent improvements since then, but the results obtained werereported ashardlysatisfactory. ‘Thecurb andgutter conduit appears tobesubstantial andgood, iftheexpense ofrelaying thepavements could beincurred. In theCity ofLondon andother large cities this expense would be ‘enormous; anditisimprobable that tohave thecables putunder ground, theauthorities would consent tohave thestreets pulled uptoany great extent, otherwise theidea would appear tobe good and workmanlike, With regard toMr. Trish's conduit, istheidea tokeep the conduit water-tight ?Ifso,Ithink that itisamistaken one. TheDoulton pipe conduit hasanice appearance; butconduits cannot always belaidatany great depth incities, and there would beagreat risk oftheir being crushed byheavy traffic. Inaddition tothis, asMr. Verity says, they have theserious 1880.) DISCUSSION. 365 objection thatthepipesrequire tobebroken toputinabranch Me.¥. joint;thesystemtherefore appears tobehardlypractical. ° Iam very much interested inhearing ofthesuccess that Messrs, Siemens &Halske claim fortheir cold-drawn lead-covered. cables, and think that theelectric lighting community ismuch indebted toMessrs, Siemens forthewayinwhich they have tried to make electric light cables cheap, and, asthey elaim, good. The vuleanised india-rubber cables are,insome cases, notsocheap, butIbelieve them tobebetter; this is,however, amatter of opinion, which must belefttobedecided bytime. Thave tothank you, Mr.Chairman, fortheopportunity you havegiven meofmaking these fewremarks, TheCaamwan: Ofcourse youhave toremember that Mr.Rrtemor ‘Verity commenced hispaper bymaking the statement that over~ headwireswouldcertainly soonhaveseentheirlastday.Ido notthink heisquite right, and itisopen toany members to consider whether inLondon weought togoinforunderground ‘oroverhead wires. ‘The paper isnotmerely onconduits, butalso onelectrical conductors, which isavery important question inthe lighting ofnewtowns. Mr. A.P,Trorren: Ithas been objected that lead-coveredMe.Trotter, cables may, and sometimes do,deteriorate when placed under- ground, becoming carbonated, orotherwise decomposed;butitix ellknown that aconsiderable number ofspecimens oflead pipes hare been found which have been put down bythe Romans. These pipes have suffered little ornodeterioration, andthereason undoubtedly isthat they were notcomposed ofpure lead, butof analloy. Unrefined lead contains antimony, tin,andsilver, allof which tend toharden the crude metal, and such analloy would doubtless resist thecorroding action better than thelead ofcom- merce, which isoften nearly pure. The nature ofthesoil, andof thewater init,exerts powerful influencesonthecharacterand theextent. ofthe chemical changes. The extreme pliability of pare lead isnotrequired; andifasuitable alloy will better resist corrosion, aslight increase ofhardness would benodisadvantage, provided that itallows thetube tobedrawn through dies, forit isobvious that theoldmethod ofcoating theinsulated conductor ‘withmolten lead wascertain toproduce pinholes andother flaws, | 366 UNDERGROUND CONDUITS, Ere, April, Memouet. The useofthelead pipe should betokeep theinsulator thoroughly dry, and toprotect itfrom mechanical injury. The insulator need not, therefore, beexpensive, asnorubber need be used. Lead-covered cables received very severe tests before the Admiralty allowed them tobeused inship work, forwhich they arevery well suited. Ifalead-covered conductor could belaid simply intheearth, protected, perhaps,byaboard,theexpenseof thecomplicated andcostly subways andconduits might beavoided, andthiswould allow agood deal more tobeexpended inmaking ‘thoroughly sound conductor. Itwould beagreat convenience ifsuch leads could beplastered into thewalls ofnewly built houses, butpure lead israpidly corroded bycalcic sulphate inthe presence ofmoisture. Xena, MieJ..N,SHoousReD: Inthefollowing remarksonthisveryinteresting paper, Iwish totouch, notsomuch ontheelectrical merits ofthevarious conductors,asonthemechanical contrivances and theconduits which Mr.Verity haspresented tous. ‘The first question that, presents itself tothemindsofmany —and itcertainly does tomyself—is whether, inthecentral portion oftheaverage oftowns, such aswemeet with inthis : country, acontinuous conduit ispossible. Myown experience wouldleadmecertainly toanswerthequestion inthenegative, considering thenumber ofmain pipes that arerunning longite- dinally under thecanseways orpavements—gas, water, telegraph, andtelephone pipes—and then, again, thetransverse services that gointo thehouses, with their valves often atdifferent levels. ‘There arealso the lamp-posts that project inside thekerbstone, often requiring the cables togooutside under the roadway. Taking allthese things into consideration, onemay bevery for- tunateinsomecasesifalength,onanaverage,of20ft.ofany- thing likeadirect lineonthehorizontal canbegotforthecables. Thedifferences oflevel arevery considerable, andoften succeed each other very abruptly; thecables coming across large pipet and small pipes, and having tothread inbetween pipes where only 2or3in.clearance, inlevel, occur between them. A case wasbrought tomynotice, some time ago, where, incross ingabrook which wascovered in,there were eighteen cables 1880) DISCUSSION, a6 many ofthem asmuch as2to3in,indiameter, andthese Msn, were passed eventually inatrough; thewhole space available being about 2ft,6in.wide, and only 5in,deep. There are perhaps afewtowns where itmaybepossibleinthesuburbs; bat,judging from the difficulties that one sees and hears of,I fancytheconduithasyettobedevisedthatwillaffordagoodandregular protection fortheconductors, With regard tosome oftheconduits that Mr. Verity has brought before us. Inthefirst place, there ishisown conduit, which, Iunderstand, isasolid block ofearthenware with longi- tudinal circular perforations ortubes running through itfor naked copper conductors topass along. The surface ofthese tubes being ofglazed earthenware would tend tocondense moisture andbecome practically small sewers forthepassage of water: the effect ofthis moisture upon the naked conductors would certainly not conduce toimprove the state ofthe in- salation, ‘Then, again, with regard toMr. Irish’s conduit. Itmay answer itspurpose, possibly,insomeplaces;butthereis,Ithink, onenearly insuperable objection onthepart ofthelocal authori tiesthat Mr. Irish would meet with, viz., theuseofacontinuous ironcover-plate alongthepavement. Municipal bodiesobjectvery strongly toiron plates being multiplied intheir pavements; they say,andIthink very fairly, that iron plates become slippery, and cause many accidents tofoot-passengers. With regard totheDoulton conduit, Iamafraid, ashasbeen pointed outbyMr.Verity himself, and also byMr.W.E.Gray, thatitconsists ofperishable materials, liable tocause fractures in thepipe itself, orelse atthejoints. Besides, ifthejoints are good,andaremadewithgoodcement,thepipeswillhavetobe broken whenever newconnections arerequired;or,ifthejoints ‘arenotgood,thenleakagewilloccuratthem.SothatIdoubt ‘verymuch ifsuch aconduit would prove byanymeans anefficient oreconomical mechanical protection. Tomymind themain and most important question tocon~ sider—and itisonethat must betheoutcome ofexperience, from which wehave yetagreat dealtogather—is whether aconduit of 368 UNDERGROUND CONDUITS, Ere. [AprilIt, Mau 82ylength atallcanreally berelied upon. Even, forinstance, atKensington, where the streets are comparatively open, Ido notsuppose that anything likea100yards ofstraight runcanbe obtained forthebrick conduits that Mr. Crompton advocatesso strongly, with naked copper conductors. Mr. Crompton himself hhas stated that hehas had tohave recourse, atintervals, to ordinary cables. Side streets, which cause these interruptions, ‘occur much more frequently, asarule, than atevery 100yards, Another very important point hasattracted theattention of ‘many persons, andwaspointed outvery strongly inthisroom by General Webber some time back, butwhich isoften overlooked, viz,, that thecable should beaccessible throughout itsentire length with great facility. Ifitisputintoaprotecting conduit which isdificult togetinto, that conduit will have tobebroken into, atconsiderable expense, inorder togetattheconductor to make sideconnections, ortoremedy defects. ‘These points lead metothink that under most circumstances, andinthebusy parts oftowns, acable that isperfectly sheathed, and able totake care ofitself, placed underground, best meets with therequirements and difficulties oflaying, andatthesame time complies with theother question ofaccessibility forjointing, orforwhatever elseyoumay want, throughout itsentire length. NeEnkine. Mr.R,S.Ensktve: With reference toKensington Cour, it may beuseful ifsome information isgiven with regard toour conduits there.Ithasbeenremarked thatwehavedifficultyin getting aconsiderable straight length. ‘The longest runwehave actuallyatwork,withoutabreak,atthepresenttimeis800yards; wegoperfectly straight along. We have recently laid dom another length of400yards, and inthat case wecertainly came across one pipe which was rather awkward, but webad that removed toone side. The great advantage ofworking under the Electric Lighting Act isthat one gets power toaskthe water company—at our expense, ofcourse—to move their pipes asalsothegasandother companies, ifthey come intheway# 4stointerfere with theworking ofthesystem ;sothat ifitdoes notinterfere with thewater company atalltomove their pipe 18in,ononesideortheother, wecangetitdone. Iftheir 1480) DISCUSSION, 369 pipeseross ours atanawkward level, wecangetthem lowered Mtrentice ormised, asthecase may be;and soalmost always, provided there arenocellars coming uptothepavement, wecanget@ straight linedown thestreet. The only cases inwhich weget interfered with inour culvert are where the cellars have been built very nearly uptothepavement, and that isvery seldom. Insuch cases weputin2-in, ironpipes andtake cables through until wegetmore space, and then westart theculvert again. Wehave tried twoorthree different sorts ofiron pipe, and have come down tothe2-in. owing totheease with which wecan putthem allononelevel, orwecanputthem oneabove another, seording tothe arrangement ofthe pavement. We donot trouble very much about thejoints being water-tight, because we findthat condensation takes place intheiron pipes tosuch an catent that itreally does notmatter whether thewater gets in fiom theoutside ornot. Infact, under apavement you very seldom getany leakage through joints, because the pavement itself ismade water-tight, andsoyouareprotected. Inthecase ofbigpipes under theroadways, ifthey have tobeused, they verylikelyhavetodipdown,andwefindthatwater,nomatter howcareful weare,willalways collect atthelowest point;andin some cases wehave leftanopen joint toletthewater drain away intotheground. Mr. J.N,Snoousrep: May Iask aquestion before Mr. Erskine sitsdown? Ithink hehasmisunderstood myallusion toKensington Court,unlessIhavemisunderstood Mr.Cromptonhimself, Does Mr.Erskine mean tosaythat thelength of800 yards hespoke ofjust now iscontinuous bare copper conductor from end toend ? ‘Mr. R.8,Ersxeve: Yes. Mr.J.N.Suooumze: Not broken upbylengths ofCal- lender's cable? ‘Mr.R.J.Erskine: No. Wehave cables inplaces, and, we haveconsiderable lengths ofCallender’s cable through sewers, but most ofour cables are short lengths. We hardly ever put a cable down now, butnearly allbare copper. Most ofourcables werefixedbeforewehadgivenMr.Crompton’s systemafairtrial;VOL. XVIII. 26 30 UNDERGROUND CONDUITS, Bre. (Aprilia, MeEnkne.butsincewehavebeenabletoseeexactlyhowit,works,—and Mr,Crompton hasmadegreat,improvements, whichgetoveralmat alldifficulties,—the arrangement has been adopted asthestar- dard system ofmains tobelaiddown bythecompany. oir ‘TheCHAIRMAN: Wehavealittle more time forquestions ose" discussion. Wecannot takethereplythisevening, butitmust begiven afterwards, andIinvite anyfurther questions. ‘Perhaps Imight askoneortwoquestions myself. Ishould liketoknow, ifMr.Gray could tellus,why theindia-rubber cables oftheearlier days occasionally failed. Iamnotspeaking,of course, ofthe Silvertown make specially; Ihave rather inmy mind thecables which were made byMessrs. Hooper, ofvulcanised rubber. They were constructed very much inthesame waythat Mr.Verity hasdescribed, ofnearly pure rubber, putround copper. ‘The copper wastinned, asamatter offact;thencameaseparator, ‘aswecalled itinthose days, ofindia-rubber with zinc oxide; and outside was acoating ofvuleanised india-rubber. ‘The insulation obtained atfirst was extremely high—far higher than wasthe specified insulation—but, forsome reason orother, Ibelieve, Iamright insaying that occasionally, atany rate, theinso-lationusedtogodown.AgoodmanyyearsagoItested many thousands ofmiles ofcables made inthat way atMessrs Hooper's works atMitcham, and theinsulation, when thecable was tested atthe works, and also after ithad been sheathed atMillwall, wasfarhigher than hadbeen specified for. Idonot know exactly what happened,asIwenttotheFarEastjustthen; butthecable was subsequently laid ontheEast coast ofSouth America, and Ibelieve some difficulty wasexperienced onaccount oftheinsulation going down. Possibly Mr. Gray cangiveus some information onthat point. Another curious experience Ihadonaprevious occasion with acable made inthesame waybyMessrs. Hooper. There were some cables lying iniron tanks, with water inthetanks, in Calcutta, foralongperiod.‘Thosecablesweretestedfromtime totime, and found tobeextremely good. ‘Then they hadto bemoved and taken tobelaid under theHooghly;andafter being laid, although there was noevidence oftheir being 182) DISCUSSION, am damagedinanyway,theinsulation ofallofthemhadgonetriewedown very much indeed. Iamspeaking ofabout 1870, or possibly alittle earlier. Itested those cables before they left. aleutta, and Itested them while they were being laid, and directly they were laid, and theinsulation ofall—I think there were six—went down, sothat itiscertain itwas notdue toa single fault inanyonespecimen. From some reason thematerial haddeteriorated. Possibly uncoiling thecable outofthetanks ut Calcutta, removing itand laying it,developed minute cracks in theinsulating material. Probably some change hasbeen made atthepresent dayinthemethod ofconstruction toprevent the material deteriorating inthis way. ‘Thequestion ofthepuncturing ofinsulated cables used for electric lighting isanextremely important one, and onewhich I think hasnot had sufficient attention given toit, Ithas been ‘assumed thattheelectro-motive forcethattheelectric lightcable villwork at,istheelectro-motive fore produced bythedynamo, andnomore; but, asMr.Verity points out quite rightly, inhis paper, there istheprobability ofavery much higher electro Iotive force being brought tobear oncables used forelectric lighting, inconsequence ofself-induction—a subject which will come before usinanother way atour next meeting, when Professor Oliver Lodge istogive usapaper onlightning con- ductors, The surging backwards and forwards that hehasso ablydrawn attention to,that you have incertain cases inelec~ trical conductors, produces anelectro-motive force infinitely greater than you would expect. Acase was brought tomy totice the other day ofasomewhat extraordinary character. ‘Twopeople were walking together intheInventions Exhibition, long acourt where there were agood many electric wires;they were several feet away from thewires, which were overhead theywere walking onwood;andtlieybothsaythattheysimul- taneously gotasharp shock. Ofcourse, from ouroldpoint of view, weshould have said itwasquite impossible, because the insulation wasonly working with—I donotknow what itwas, whether itwas 100volts or200volts; but they could nothave received aspark ofseveral feetwith apotential difference ofeven a an UNDERGROUND CONDUITS, Ere, [Aprilllth, fretner fewthousand volts. Atthesame time, youhavetheir evidence that they independently feltashock atacertain moment, and said immediately tooneanother, Ifeltashock.” ‘They applied tothe people inauthority, butofcourse they could getno information, Now, with theillustrations that wehave had recently shown usbyDr.Lodge attheRoyal Institution theother evening, it does notseem improbablethatyoumayhavepotentialdifferences setupinwires fargreater than the potential difference that the dynamo canproduce, due tothesudden stopping ofthecurrent: forexample, whether youmay beworking at100or1,000 volts, youmay have 10,000 volts,orevenmore,produced. Ithinkthet that may explain thebreakdown ofsome ofthese cables. Cables carrying telegraph currents may lastforalongtimewherethere arenot such high voltages used, because the amount ofenergy available isnotnearly asgreat. Idonotwish todetain you any longer;butifMr.Gray,oranyotherrepresentative oftheindia- rubber and cable industry, could give ussome information asto why thecables aremore likely tolastnow with vulcanised indis- rubber cables than they lasted inthe past, Iforone should br extremely grateful. sey. Mr.W.E.Gray: Ido notknow whether there areanyother cable makers present; but you have, Mr. Chairman, touched o rather adelicate point, Ithink, byasking the Silvertown people todiscuss andcriticise thefailure ofHooper's cable. Ithink we could hardly beexpected todothis, rotor TheCnarrmax: Idonot,ofcourse,wanttoaskyoutogo“nto anysortofcommercial difficulty; butisitnotpossible, without touching oncommercial questions, tosuggest some explanation which would give confidence tothose who bave used india-rubber cables and found them fail? That iswhat Imean. Mr.Gray. ‘Mr.W.E.Gray:Thereisonethingyouhavementioned, Sir—that is,the fallimthe insulation ofthese cables. Idonot propose togovery fully into the matter, butImay saythat inordinary cables, asprobably you arewell aware, thisdepends fagooddealonthequalityoftherubberused.AsMr.Verity hasexplained, rubber canbegotat1s.perIb,and alsoat36 3) DISCUSSION, a3 perb.; thereason forthehigher price isthegreater durability cry. ofthehigher class,as,ifthelowergradewereused,although youmight even get ashigh aninitial insulation aswith rubber atthe higher price, this insulation would not last. Ithink that iswell known to rubber cable manufacturers, Asregards thequestion oftheHooghly river cable, Idonot know what thetemperature may bethere, butpresume itis greater than theordinary temperature inEngland, inwhich case thecause ofthefallmight betraced tothefact that, ifthe material were imperfectly treated, any decomposition would probably beincreased and accelerated bytheincrease intem- peratare, Ofcourse Icannot tell you why Hooper's cable did notlast, but can only saythat the explanation would apply to rubber cables ingeneral. ‘The failure may beduetoaltogether different causes, and Ican only regret that one ofMessrs, Hooper's staff isnot here toexplain ittoyou. Generally waking, there isasmuch difference between rubber and rubber 4sbetween leather and leather, or,indeed, any other material; andthisshould betaken into account inthespecification, If theconsulting engineer understands thematerial heshould be abletoget.a good thing;butifhehasnotthisknowledge,and throws the matter open tocompetition, the manufacturers would probably only give what wasasked forinthespecification, As amillustration, Icanquite suppose that when buying telegraph ‘ire, thePost Office, knowing that there arevarious qualities of this,would hardly expect togetfor£8pertonthevery best telegraph wire; and their specification would bemechanical as vellaselectrical. ‘The same thing applies torubber astoany- thing else—it ismerely aquestion astowhat people willpay for,andthedurability istoagreat extent aquestion ofprice; butthere isnodoubt whatever that agood cable, manufactured ofgood material, properly and uniformly treated, and well vuleanised, will last. ‘There are, Ibelieve, inexistence now, ‘rubber cables that have been working fortwenty tothirty years, ‘thathave lasted, andwilllastforsome time yet. Mr. Gisnert Karp: The author said that hehad not dealt we.Kew. withseveral other systems, andIthink weshould allbeglad if su UNDERGROUND CONDUITS, Ere.[April1th teKen inhisreply tothediscussion hewould amplify hispaper inthis | respect. Inparticular, Ishould beglad ifhewould add some facts upon asystem ofwhich agood deal hasbeen saidin America, viz., that ofoilinsulation. No doubt you areall familiar with theexperiment that Mr. Brooks performed some years ago. Hetwisted twoordinary cotton-covered wires together except attheends, which were kept 1in.apart; thewires were then put into ajarfilled with paraffin oil, sothat only the separate ends projected above thesurface oftheoil. Thewires werethenconnected withaninduction machine, andsparkspassedinairfromonewireendtotheother,butnosparkspassedthrough thecotton covering and oilwhere thewires were twisted together. Mr. Brooks hasemployed this discovery oftheenormous inso- lating property ofoil,intheconstruction ofanunderground system. Aniron pipe filled with some cheap mineral oilforms theduct, andinto this areplaced theconductors, which, Ibelieve, need noother covering than cotton. Ifthis system should prove really sosuccessful asitsinventor believes ittobe,itwouldbe ofthe greatest importance tocentral station lighting inthis country, and Iamrather disappointed tofind nomention ofit whatever inMr. Verity’s paper. Ifhehasseen the Brooks system inactual operation intheUnited States, Iwoukd askhim togive usinhisreply some facts about it,obtained from hisown personal observation. iethie.—-Mr.C.J.Parttars: AsIwasinMesers,Hooper's serviceatthe time they were making theWestern Brazilian cable, which is the one referred tobythe Chairman, perhaps Imay offer one explanation. With regard tothat cable, Ithink you areaware, Sir,that thegreater part ofitwasoriginally designed forusea: 1deep-sea cable, andthat itwasafterwards laid inshallow water fasacoast cable. ‘This fact, Ithink, toagreat extent accounts for itscomparative failure, thetype being unsuited tothelocality. With regard tothequality ofthematerial which wasusedin the insulation ofthe cable, Ithink atthat time, Sir,youwere with SirWilliam Thomson, whowastheconsulting engineer of theWestern Brazilian Company, andIthinkhisdutywastosee that thematerials used were according tospecification. Ihave 1880 DISCUSSION. 35 noconnection with Messrs. Hooper now, but Icertainly know ae.ritre from my own experience that they used the best rubber that could beprocured, and that anyfailure inthe insulation isnot tobeaccounted forinthat way. Ibelieve now that arubber cable, esMr.Gray says, “properly made,” isperfectly reliable, The Cuamuax:IwasgoingtosaythatIdonotthinkMr,Rriesor Gray hashittheright nail onthehead byimplying that the failure ofthe cable was due tocheap material being used. I alsohave nothing todowith Messrs. Hooper now, orthen, and I have nointerest, therefore insupporting any particular way of making cables, butIthink that Messrs. Hooper were quite as much astonished asanyone else atthe subsequent falling offin the insulation oftheir cables. Iwould like toadd one other Point, and that is,would Mr.Gray point outwhat should bethe testoftheinsulating quality ofamaterial ifagreat number of megohms resistance means nothing? What would hepropose shouldbethetest?Shoulditbeachemical testoftherubber?or ‘hat should bethetest inpassing acable, ifthe fact that ithad afarhigher insulation resistance than could beexpected ina able, didnotshow that itwasmade ofgood india-rubber? I think, inthe case ofiron, that ifthe iron fulfil allthe mechanical testsoftension, torsion, elastic limit, &.,required ofgood iron, itisreally allthePost Office, oranybody else, would want, what- evertheiron becalled, however many “B.B.B.’s” or“Best Bests” maybeputbefore thename. Mr.W.E.Gray: Iamsorry tooccupy thetime ofthemeet- "ony ingwith what appears toberather aconversation than adis- cussion. Agentleman hasjust spoken who says hewaswith Messrs Hooper atthetime theBrazilian cable wasmade. Ihope hedidnotthink Iwasimplying anything derogatory tothat fim, Mr. C.J.Pamuirs: Not atall. Mr.W.E.Gray:I’:rtainlydidnotmeantodosointhe slightest degree. Isimp. tried togive anexplanation that might account forfailure inrubber cables; butitwould be necessary toknow alltheconditions ofthespecification before onecouldjudgeastowhythecablefailed.Idonotknowthe 316 UNDERGROUND CONDUITS, Br, [April11th, xr.omy. specification, and Icertainly could not undertake togive the reason toyou, Mr. Chairman, who tested thecable, andhads fullknowledge ofthespecification. ‘You, Sir,asked mehow tomake rubber cables. Thisisrather long process toexplain toyou, and although Iwould beglad tohave aprivate opportunity ofdoing #0,Idonotthink that thepresent occasion issuited forgoing into details ofmann- facture. ‘There isnodoubt high electrical insulation canbe easily obtained, butthepractical question isthedurability of : thecable, Itishardly sufficient tomake anelectrical test, and, because theinstruments giveagood reading, besatisfied withthe cable, The question ofthetime andtemperature would haveto come in, Idonot know whether the cable you refer towis tested atahightemperature,orwhatthedurationofthetestwas; utitseems tomethat thewhole conditions ofuseofanycable shouldbetakenintoaccountinthespecification, andsuitabletests applied. Ifpermanency ofinsulation isrequired, then themana- facturedmaterialusedshouldbesuchastogivethatpermanency, and precaution should betaken toget this; butitappears tome tobeamatter ofjudgment andknowledge ofthematerial used. Itisimpossible togointoallthese points atpresent,butI should beglad togive you anydetailed information youmight require, outside ofthis meeting. eter TheCuamnatax: Iwillaccept yourkindinvitation afterthe "meeting. Iventure tosaythatneither amonth's testnoraseveral months’ test. shows thegoing down intheinsulation. During thewhole manufacture oftheHooper's cable Ihave been referring to—it hadalltobesheathed, ofcourse, andittook many months tomake thissheathing—there wasnoevidence whatever ofthe insulation going down. Notonly didthemany tests ofeach section ofthecore, buttheconstant tests ofthesheathed cable before laying, failed toshow any suspicion offaults. Inthecase oftheHooghly india-rubber cables, certainly ayear’s testing didnotshow what thecoiling anduneoiling ofthecable atonce brought into evidence. MSoua. Mr.J.Fanqunanson: Gentlemen,—I think ourChairman wo" raised avery important question, that requires avery much | 1188.) DISCUSSION. 37 faller answer thanithashad,vizhowweshallknow india- we rubber that isworth 1s.apound from that which isworth 3s.a= pound. Now Ihave had something todowith india-rubber in nytime, and Iwill tell you the test that Iadopted. Idonot know that Mr. Gray will thank meforit,butitisatest that has been continued bytheAdmiralty forthelasttwenty-five years, andthere should beno‘harm inmytelling you here what itis. ‘The test that the Admiralty adopt asatest ofthe quality of india-rubber isthis: They take aportion ofitand put itina steam chest under apressure of60Ibs. ofsteam (about 300° Fah.), and keep itthere forthree hours. Itisthen removed and put indry heat fortwo hours at270° Fab., and then allowed tocool. Ifitisthen soft, andbends without cracking, thequality isgood. This isatestforthequality oftherubber. ‘Theproportions oftheconstituents should bespecified andtested byanalysis. Imayaddthatthesulphurshouldneverexceed Sper cent. The oxide ofzine need notbeless than 40percent. andiftobeused inasoilorforapurpose where itmay be subject. tomineral oil,theoxide should beatleast 60percent. Mr.Georce Svrrox: Idonotthink there wasanything sesiti. ‘ithregard tothecables laidontheeast const ofAmerica that ‘ould show that thematerial wasofpoor quality. Ihave met with many who were connected with that expedition, and I always understood that thefailure ofthecable wasowing toits being destroyed bymarine animals, especially thesaw-fish. A tooth ofasaw-fish wasfound embedded inthecable, having got right through thesheathing tothecore. MyCompany have made rubber cables which have been laid inmany parts oftheworld, and have given notrouble, No doubt gutta-percha haspractically superseded therubber cable. Wehavetogointothequestion ofexpense, andhitherto nothing hasbeenfoundtoserves0wellandsocheaplyasgutta-perchaforsubmarine work generally. Rubber cables have been foryears more expensive than gutta-percha, and the question ofcost isa very serious one when itisacase oflaying, say, athousand miles ofcable. Major-General Wessen [communicated]: Ihave to.thank wajgcen am UNDERGROUND CONDUITS, Ere. (aia, Yufrges Mr.B.Verity forhisinteresting paper, asitreminds methat Ireceived from Mr. Leonard F.Beckwith, chief engineer of theConsolidated Telegraph and Electrical Subway Company,of ‘NewYork, solong agoasearly inFebruary last, avery detailed account oftheelectrical subways built within thelast three years inNew York City. Mr. Beckwith gave me adescription’ ofwhat isknown in America asthe Dorsett system, and itsapplication; ofthe ereosoted wood conduits; ofthe iron and cement pipe conduits; oftheconduits ofiron pipe laidinasphaltic concrete;ofconduits ofironpipelaidinhydraulic cementconcrete ;andoftheJohn- stone cast-iron conduits. Our correspondence arose outofthefact ofMr. Beckwith becoming aware that Ihad forseveral years had myattention turned tothis subject, and that the result ofpractical experi- ence, dating since 1870, inlaying underground electrical oon- ductors, and many experiments since 1882 (when Eaisoa's system was exhibited atthe Crystal Palace), appeared inthe arrangement ofconduits known astheCallender-Webber system, exhibited twoandahalf years agoattheBorough Road Works oftheAnglo-American Brush Electrie Light Corporation. Iwasglad tohear from Mr.Beckwith that nothing precisely inthesame form had been tried inthe United States, because, with therapid extension ofunderground work owing tothe demand forelectrical distribution inthat country, itmight have been thought that ithadbeen used and proved afailure. When Isay“precisely,” Imean that nocases orconduits had been made orlaid down consisting ofthesame material ormade in thesame way. Astopriority, orastoourhaving much tolearn, Iwould once more begtobeallowed toremark that thehistory ofthis subject inconnection with telegraph conductors, inspite ofthe many warnings oftelegraph engineers, almost. exactly repeated itself inconnection with thelaying down ofconductors suitable forelectric lighting, and,Iamsorrytosay,inspiteofthose warnings, islikely torepeat itself again, Early intelegraph days the conductors were allsealed into 1889,) DISCUSSION, a0 thepipes. Bitter experience ledtothedrawing-in-and-out prow. system, in‘which even insulation issecondary tothe perfect facility forremoving andreplacing adefective cable, Allthetroublesfromsteam,heat,temperature ofthesoil, gasandwater,dangerofdamagefromotherworks,difficulty of finding room under theroad, were encountered and conquered Jong before 1880. ‘The only vital change intheprinciples and pructice then learnt, that Iknow of,hasbeen that with large cables each conductor must beprovided with aseparate “way.” ‘Thechief feature oftheCallender-Webber casing isthat this isprovided forinamore simple and less expensive manner, I believe, than with anyother material ;atanyrate, such was the carethree years ago, when theplan wasfirst thought out. Thearrangement most similar toit,amongst those referred to,IwilldescribeinMr,Beckwith's ownwords:— “This construction consists oflap-welded wrought-iron pipes, “the different sizes varying from oneand ahalf tothree inches, “according torequirements, laid intiers inhydraulic cement “concrete. ‘The concrete isfrom an inch toaninch and ahalf “thick between the pipes, and three inches thick around the “mass ofpipes. ‘The pipes arefrom eighteen totwenty feetlong, “with screw coupling joints. ‘The concrete hasdue proportions “of cement and sand.” Thework oflaying down this conduit caneasily bespecified byanyengineer, andIwillnotgive itsdetails. Itsadvantages aremanifest initsstrength, inbeing impervious, intheeasy multiplication ofcable-ways, andinitsgreat durability. But it cannot compete foronemoment with thefollowing advantages intheuseofthebitumen conerete casing Ihave alluded to, vanmelys— The occupying ofsmall space, and space ofdifferent sec- tions. ‘Thebeing able tobring thecasing tothework ready made, saving thetime during which streets have tobeopened. ‘Thelightness andcostofthematerial, which, inthecase ofthebitumen concrete casing, isabout one-half per lineal measurement oftube. 380 UNDERGROUND CONDUITS, Ere. [Aprilii&, agro Thepower ofbeing able tobend thecasing onthespotto suit theobstructions previously unforeseen. ‘The ease with which the material can becut,andahouse service-box putin, Thesmoothness thatcanbeobtained intheinteriorsurface oftheways. ‘The other day two 1,800-feet cables were drawn inat Chelsea with great ease. Each plan, nodoubt, hassome advantages over theother; Dutforthesaine cost theformer cannot compete with thelatter, ifthegreat end ofbeing able todraw inand outisessential Inview ofthe opinion expressed infavour ofthis condition by theBoard ofElectrical Control ofNew York, and bytheMunici- ality ofParis, andelsewhere, thiscanhardly nowbedoubted. ‘Thateachworkmaybecarriedoutsoastogivebadresults Ihave nodoubt; and asregards the laying down ofthe Callender-Webber casing,Ihavenonewhatever thatfailurewill infallibly follow, and facilities formaintenance inthefuturebe jeopardised,unlessthemost.evidentprecautions,angothersess apparent, are followed. Inconnection withconductors destined tobeusedwithhighpressure, Iamallwith Mr. Verity that itsmaintenance will depend onthe quality ofthe insulation, and not onthecom parative insulating properties ofthe material ofwhich its surrounding conduitiscomposed. Withperfectfacilitiesofre newal, thelatter isacondition tocause noanxiety. Oneword astoleakage. ‘Tojointing Iwill notrefer more than toremark that ifelectrical engineers studied thesubject andtaught their men aswell aswewere instructed andtaught atthePostal Telegraph Works atGloucester Road, under myold friend Mr. Andrew Bell, weshould have less tofear inthat direction. Insimple jointing andconnecting safety willcome— come intime. Butthereally serious causes forleakage lieintheaccidental, local, and varying conditions which come into existence at unforeseen points along theline, arising outoftheinsulation being attacked bygases andliquids present orpercolating inthe surrounding soil. These will bethebane ofthe electrical 19.) DISCUSSION. st engineer ofasupply system, andtheir frequency willincrease ¥ajgrcen with thelifeoftheplant. Tothe practical prevention oftheir results hemust direct untiring attention, TheCuarvan: Iamsureweareallextremely obliged toPeter Mr.Verity forhaving brought thispaper before us,and that we cannot dobetter than give him ahearty vote ofthanks forthe paper, and express our great regret that illness should have deprived usofhispresence. The motion wasunanimously carried. ‘Mr.Vertry[communicated]: SincereadingmypaperIhaveMe.Verity. received from America some samples ofEdison three-wire con- ductors andjunction boxes. Ishould have been happy toshow these tothe members had it beenpossibletodevoteanother(OISSSSSSNIO)eveningtothediscussion; butQJ WO since itisnotpossible,Iam Fw.A, onlyable toshow you illustrations inplace oftheactual things. Fig. Ashows the flexible copper connections used throughout the a system atalljunetion boxes. Fig.B p= shows asection ofthethree con- = ductors intheirontube,madein & lengths of20feet, each length a joined with theflexible connectors (A)inthejoint box, which isthen filled inwith insulating compound, The flexibility ofthe system Fw.B. ismaintained bythejoints being alloftheball-and-socket form, giving alimited movement, but quite sufficient forallrequirements. Fig. Cshows aTjunction box, and themethod ofmaking connection with theflexible connectors (A)toabranch forhouse service. The whole service isvery complete, and hasbeen invariably successful inAmerica, although, sofarasIknow,ithasonlybeenusedforcurrentsof lessthan 300volts potential. With regard toearthenware pipes, Ihave been informed that thetroubles arising from branches androots oftrees getting into es UNDERGROUND CONDUITS, Ere. (Apia Mr.Veuty: thejoints isthoroughly overcome byusing agood cement joint, and that where these troubles have arisen only clay basbeen used forfilling upthejoint space. Cyf ‘(_® (Cm °(Ss —=y)QWT BAA UAhMl Free. ‘Mr.Shoolbred hasspoken ofMr.B.Verity's conduit asoneia which alarge amount ofcondensation would take place, andhas ‘assumed that itsefficiency would beimpaired thereby, butIdonot think this isso,foreven ifallofthechannels were wetitdoes ‘notaffect theinsulation solongasthey arekept separate, which isdistinctly apart ofthesystem, ‘Thediagram, perhaps, didnotshow thedetails veryplaialy, ‘asproper means areprovided fordraining theducts intoatap andgetting ridoftheaccumulated water. Provided thejoints areproperly made andtheconduit laidonafairbottom, itwoald practically lastforever, asIdonotknow anything thatdestroys 1889) DISCUSSION. 38 ‘glazed stoneware (except main force, which would berequired in Verity ratheralargequantityforthispurpose).‘Therealobjectionto such conduits aretheir want offlexibility. Inreply toMr.Kapp, Ihad thepleasure oftwice meeting Mr.BrooksinAmericaandhearinghisviewsonthissubject,and 4description ofhisparticular system; but Iregret tosay my opinion was,atanyratesofaraselectriclightingisconcerned, that thesystem wasnotapractical one; arid that seemed the general impression there. However, asanEnglish firm have taken upthe patents forthis country doubtless they have satisfied themselves ofitspracticability, and weshall hear more ofit.‘The conduit consists ofwrought-iron pipes supplied with suitable splice boxes, hand holes, and outlets; these boxes are protected from oxidation bybeing laid inawooden trough into hich hotpitch ispoured, The wires used inthis system are ‘cotton covered, andstranded intoarope orbundle, which isthen covered with atextile weaving; the bundle ofwires isthen saked inhot mineral oildrawn into the pipe, and aheavy mineral oilisforced inforthepurpose ofexcluding moisture and maintaining the insulation. Intowns where there ismuch difference oflevel, the head ofoilinthe pipes would bea source oftrouble,asmaybeeasilyimagined, Theexperiment mentioned byMr.Kapp, asshowing the efficiency oftheoil insulation, isthus described :—Two wires were attached toaHoltz machine, theextremities being carried into anarrow jarofthe insulating oilinsuch away that when thedistance between them wasonequarter ofaninch intheoilthey were oneand s-lalf inch apart atthesurface ofthe oil; onturning the machine, thespark passed above thesurface oftheoiland notat allthrough theoil,even when thedistance between theextremities Was reduced toone-sixteenth ofaninch. Myown opinion is,after carefully considering the subject, thata5-or6-inchironpipe,withasmoothinterior,goodjoints, andsuitable distributing boxes, forins thesimplest and cheapest formofconduit possible; that islooking atitasmerely a mechanical protection forthecable. Thad hoped that thediscussion woul have better ventilated therelative advantages offibrous andhomogenous insulation. 388 ELECTIONS. (April Mr.Yesis: Ttwould certainly appear, without other proof tothecontrary, that, forhigh tension work atany rate, agood continuous insulated covering, such asvuleanised india-rubber, must posses: many advantages over afibrous material, whatever the insulating oilmay bethat itissoaked in. Itisvery encouraging tohear from Mr.Erskine ofthe Kensington Court Company's experience, andreally itmakes one think that much wehear ofthefrightful difficulties encounterel isabogey, but perhaps the gas and water companies arepat- ticularlyobliginginthisdistrict. Myownexperience isthateven ifthey arelegally obliged toremove thepipes, itisanexceedingly difficult jobtogetthem todoitunless they arequite agreeable andseesome advantage tothemselves inthearrangement. Mr. Farquharson hasgiven usagood practical receipt for testing rubber, and, asithasstood thetest ofsomany year’ experience with the Admiralty, nodoubt wecan take its perfectly reliable. After what hasbeen said about insolation tests, itisobvious that these aretotally inadequate andarerealy notest atallofquality, bad rubber showing quite ashigh insulation asgood when first made. Thave tothank themembers ofthis Society forthecourteous way inwhich they have received mypaper, andhope thatif,ia thefuture, Imay have theopportunity ofagain bringing the subject before them, itwillbewhen many tiles ofunderground work areinsuccessful operation inthiscity, andwhen awider experience anddeeper knowledge willmake itoneofthefirt importance toallelectrical engineers. Aballot took place, atwhich thefollowing were elected -— Members: FrankErskineDempster. |—HughTheodore Pinhey. Associates: John Melville Coode. Orlando V.Thomas. Henry Chevly Alexander Augustus Wolfe. Goodall. Student: Jobn Allison Muir, ‘The meeting then adjourned. 19.) DEATHOFDR,WARREN DELARUE, 385 ‘TheOneHundred andNinety-second Ordinary General Meeting of the Institution was held at the Institution of Civil Engineers, 25, Great George Street, Westminster, on ‘Thursday evening, April 25th, 1889—Professor W.CnooxEs, F.RS,, Vice-President, inthe Chair. The minutes ofthe Ordinary General Meeting held on April 11th were read andapproved. ‘The names ofnew candidates for admission into the Institu- tionwere announced andordered tobesuspended. The following transfers were announced ashaving been ap proved bytheCouncil :— . From the class ofAssociates tothat ofMembers— Sidney Dobson, From the class ofStudents tothat ofAssociates— Peter V.MeMahon. | W.H.Wickham, ‘The CuarnwaN: ‘The painful duty now devolves upon me, Gentlemen, toannounce the death ofanoldand very eminent member ofthis Institution, and avery oldpersonal friend ofmy own—Mr. Warren DelaRue, who died onGood Friday last. I ‘will ask Mr. Preece tomove aresolution, which the Council desire tosubmit toyou, expressive ofthedeep regret which Iam sureallmust feelupon theoccasion, ‘Mr.W.H.Preece, F.R.S.: Oneofthemost depressing duties attached tothose who continuetoliveinthisworld,istohaveto ‘express regret every now and then atthe loss offriends. It ‘happens that Ibecame acquainted with Dr.Warren DelaRue in theyear 1853, and from that year to1889—36 years—I have incessantly, inoneform oranother, received from him attentions andkindnesses which cannever beforgotten. Dr.Warren Deia Ruehadamagnificent laboratory attheback ofPortland Place, which wasalways open toanybody, and especially toyoung ‘VOL, XVIII. 27 386 ONLIGHTNING, LIGHTNING CONDUCTORS, [Aprist. electricians, who desired tomake experiments ortotest their ideas; and one had only toexpress tohim awish tomake an experiment inhislaboratory, touse hismagnificent battery of 15,000 cells, when hewould atonce cheerfully notonly allow the experiment tobemade, butwould invariably bepresent, and,if hecould, would himself conduct theexperiment. Ihave received during the last 36years such attentions from Dr.Warren Dels Rue that Icannever forget, and itwas asource ofextreme regret tomethat owing toill-health Iwasunable topaymylasttribute ofrespect tohisremains atKensal Green lastTuesday. Ihave therefore with great grief, butwith acertain amount ofpersonal satisfaction that such aduty devolves onmerather than onone who knew him less, topropose the resolution which basbeen already approved bythe Council, viz.—“ That the Council and “members oftheInstitution ofElectrical Engineers desire to «express their deep regret atthegreat losswhich they,aswell “as thescientific world generally, have experienced bythedeath “of Dr, Warren DelaRue, D.C.L., F-RS., and they desire to «convey toMrs. DelaRuetheexpression oftheir sincere sym- «pathy with herinhergreat bereavement.” Professor Sitvaxus P.THomrsox: Iask permission, Sir,to secondthatresolution, Wemustallfeeldeeplygrievedthatone who wasnotonly honoured byallthelearned Societies inthis country, butwho wasalso sohighly respected bythose ofother countries, hasgone away from ourmidst; andwecannot butfeel that wehave sustained, incommon with themany other Societies ofwhich Dr,Warren DelnRue was amember, amostseverelos. ‘The motion wasunanimously carried. Thefollowing paper wasthen read :— ON LIGHTNING, LIGHTNING CONDUCTORS, AND LIGHTNING PROTECTORS. ByDr.Ouiver Lovos, F.R.S., Member. Istnopecriox. About this time last year Idelivered two lectures tothe Sceiety ofArtsonthesubject of“Lightning Conductors,” inpar 1099] ANDLIGHTNING PROTECTORS. 37 manceofabequestmadetotheSocietyinmemoryofthelateDr. Mann. Inthose lectures Iabstained, asfaraspossible, from practical recipes and from anything like authoritative advice, contenting myself with calling attention tocertain aspects ofthe subject which had been overlooked. Iventured toimply that noneoftheolderelectricianshadanynotionoftherealconditions oftheproblem;thattheyall,fromFranklintoFaradayanddown tothepresent day, treated itasamuch easier matter than, in fact,itis;andthattherehadbeenverylittlerealprogress inthis particular department since thetime ofFranklin. Recent advances inelectrical theory made iteasy formeto seefurtherintothematterthanthefargreater menofthepast hadhadanychance ofdoing; and afew very simple and easy experiments soon brought theconditions oftheproblem clearly before me. Itwas these conditions upon which Ilaid emphasis inmy lectures totheSociety ofArts. Any practical outcome Ileft to alater period, and very likely toother hands, ‘The first requisite seemedtobetograsptheconditions oftheproblem asilluminated bytheory5thesecond,tocarryoutpracticalreforminthelight ofalarge experience. Such fewpractical recipes asIdidsurreptitiously introduce I fully expected tohear criticised, nordidIlayanystress onthem atthe time, (Imay say, however, that Ihave seen nooccasion since toimportantly modify anyofthem, andthat myposition in respect ofthese practical hints isstronger now than itwasthen, inasmuch asIfind that the large experience ofthe Belgian electricians, who have given more attention tothis subject than pethaps anyother school, hasledthem torecommend inrecent ‘Years almost precisely thesame methods ofprotection asthose I advocated ;theirviewsbeingacquired almostentirely bypractical experience andscarcely atallbytheory.) ‘Thetheoretical andexperimental portions ofmylectures were, however, matter ofserious consideration with me,andnothing vwassetdown under these heads butthat ofwhich Iwasabsolutely ure. These portions have byallpure physicists been, sofaras Iknow, universally accepted ;as,indeed, they contain nothing 388 ONLIGHTNING, LIGHTNING CONDUCTORS, [April2, but what any one ofthem could have easily arrived athad hehappened togive sufficient thought and attention tothe subject. They have notbeen attacked, nordoIbelieve theyare atallattackable from theside oftheory. Butfromthesideofexperience anattackhascome,and, strange tosay, ithasbeen directed, not atthe fewpractical suggestions, which might beexpected tobevulnerable omthit side, butatthetheoretical and purely scientific portions. ‘AttheBath meeting oftheBritish Association, thethea President ofSection G,speaking inthe light ofamot gigantic experience—a personal acquaintance with over lf amillion conductors —led amost good-humoured attack against thedoctrine ofmylectures, and, exercising hiswell known rhetorical and oratorical powers, succeeded probably in impressing thegeneral bulk oftheaudience with thebelief that hehad, onthewhole, thebest ofthebattle; though, atthesame time, Iexpect they were ina state ofutter bewildermentasto what the conflict had been allabout, and where the difference “twixt tweedledum andtweedledee” precisely lay. The battle was, however, admittedly adrawn one, and atitsclose Mr.Pree magnanimously offered meanother field, with anaudience ofan altogether different calibre tothegeneral public ofBath, by getting your Secretary toinvite metoread apaper before this Institution, Thus much byway ofhistorical note and introduction. Now forthescientific portion ofthepaper. Ibegin byrunning over some oftheconclusions atwhichI arrived, and proceed tore-establish and demonstrate their correet= ness eitherbytheoryorbyfreshexperiment, whichevermayseem themost simple andsatisfactory under thespecial circumstances ‘Two Main Cases oFLicntsixe Fiasi. 1,Alldischarge isvirtually that ofaLeyden jar. ‘There are always twoconductors separated bydielectric, and thedischarge isabreaking down ofthedielectric atitsthinnest orweakest place. Inathunderstorm thecharged conductors areobvious, being 192] ANDLIGHTNING PROTECTORS, 389 either two clouds, orelse acloud and earth, and the dielectric is theairbetween. 2.Ttmust sometimes happen, when onecloud discharges into another, that the potential ofthis other issuddenly raised high enough tocause ittodischarge into theearth, even though no strain previously existed inthe airbetween itand earth. The same thing may happen invarious other ways when two clouds spark into each other, asindicated bythediagrams (Fig. 1). 3.There are, therefore, two main cases—(a) When thestrain inthedielectric near theearth hasbeen ofgradual growth, in which case the path ofdischarge will beprepared inductively beforehand; (b)when thestrain arises sosuddenly that there is otime forany pre-arranged path. The first Icall “steady “strain” thesecond, “impulsive rush.” Itismost important to recognise these twocases,and tounderstand theextremely different, conditions attending the two. ‘The first case only wasever con- templated bytheolder electricians; infact, 0farasIknow, it asmyexperiments last.yearwhichfirstcalledattention tothe other case. Coxorrioxs oFPROTECTION AND OFHEING STRUCK, UNDER THE Circumstances oFBac Cast, 4.Iwillnow illustrate experimentally* theconditions under which discharge occurs ineach ofthese two main cases; andI willtake first the case ofsteady strain, orcase a.This insulated sheet oftinplate issupported horizontally afootortwoabove another plate lying onthetable, anditiselectrified byaWims- hort machine. Itrepresents acharged cloud hovering over the land,thelower plate representing theearth. Between thetwoI ‘amerectbuildings, andlightning conductors terminated invarious ‘ays, The typical terminals which Iwill here usetoillustrate theconditions arefour—viz., alarge knob, or,asIshall callit, “dome ;”asmall knob, which Ishall call“knob;” asharp points andagasflame, torepresent achimney orother furnace current ofrefiedair.Puttingtheknobandthedomebetweentheplates, *These experiments were shown with asplendid machine most Kindly ‘wooghtoveranderectedforthepurposebyMr.Wimshursthimself,whomI bereby cordially thank, 390 ONLIGHTNING, LIGHTNING CONDUCTORS, (Apelth, wefind theknob struck bypreference, even though thedome stands atamuch higher elevation. Introducing thepoint,we finditprotects both, byasilent discharge, until itislowered very considerably; and that then several points may protect when one does not. Replacing thepoint bytheflame, wefindthatitpro- tects too, but not soefficiently asthe point, and that itgets curiously beaten down and darkened inthe actofprotection. ‘The point isnotstruck byanoisy flash until itisraised pretty closetotheupperplate,whenitisstruck;butabunchofpointe iseven then noteasily struck, sometimes continuing todischarge with aconstant fizzright upalmost into thecloud, 5.Trytheeffect ofabadearth orother very high resistance interposed inthe path ofthe conductor, sayacapillary water tube orabitofwetragorawetstring. Theviolenceofthesparkisgreatlylessened,andthesound isnow gentle, butthat which was struck before isstill struck: resistance, solong asitbesomething short ofinfinite, makesno difference totheease with which agiven object isstruck under thecircumstances ofthis case 1.Insert thewet raginpath terminated bypoint, and itstill protects, practically aswellas before. Insert itinpath terminated byknob, anditgets struck ‘atthe same elevation asbefore. Thefactisthat thepath ofthedisruptive discharge isall negotiated and pre-arranged intheairabove, especiallyonthe surface ofanysmall conductor reared into this space, andthe resistance which theflash may ultimately have tomeet with in itspassage toearth isathing ofsubsequent consideration. 6.Somuch fortheconditions attending case a,thestesdy strain, Now attend tocase },theimpulsive rush. Wesball find everything very different. Alter theconnection sothat acharged Leyden jarmust, whe itdischarges, discharge direct into theupper insulated tinplate, and thence overflow totheground ifitisable toraise thepoten- tialhighenough. Iftheplateistoofarabovegroundforaflash tooccur, theLeyden jardoes notcompletely discharge; itonly produces anumberoffizzesandspits,andthegreaterpartofits charge remains init. But when theupper plate iswithin 180 AND LIGHTNING PROTECTORS. 301 sparking distance oftheground (and thesparking distance under these circumstances issurprisingly great byreason oftheimpetus ‘ith which theelectricity rushes into thetopplate), then thejar discharges completely, andwehave aviolent crack both between theknobs ofitsdischarger and intheairgapbetween thetwo plates, which isinthepath ofthedischarge; thearrangement being really two condensers inseries, but only one charged (Fig. 2). TheonlyobjectoftheLeydenjarincaseaistogivemorebody totheflash. Incase &theLeyden jar,oritsequivalent capacity, isessential. 7.Putting the dome and the knob between the plates arranged asincase b,wefindthat onegets struck aseasily asthe others theknob hasnow noadvantage: whichever isthehigher, that gets struck, without reference toother considerations, Introducing thepoint also, wefind precisely thesame istrue forititsprotective virtue, somuch insisted onbythe older electricians (among whom Isuppose Iamright inreckoning my chief antagonist), isentirely non-existent. Itgets struck no more easily, and noless easily, than the dome, and itgets truck byaflash ofprecisely the same noisy character asthe others getstruck with, Abunch ofpoints acts inexactly the same way. Acomb of24needle points protects nothing, and getsstruck just thesame asanything else. 8.Now introduce the flame, and one notices amarked differ- ace, Incase aitprotected less well than the point, but it wisnot struck noisily any more than the point was. Inthe present case itgets strack with violence, and itgets struck much more easily than anything else. Adjust dome andknob andpoint, atabout thesame level, and they getstruck, one orother, at, nndom. Adjust theflame agreat deal lower, and itprotects them all,notbysilent discharge, butbygetting struck itself instead. “Protection,” however, isinthis case not the word touse. ‘Theflame better represents achimney requiring protection, while thepoint corresponds tothepointed terminal ofalightning conductor raised agood dealhigher withtheintention ofprotect- a2 ON LIGHTNING, LIGHTNING CONDUCTORS, [Apri 260, ingit.Protect it,itdoes not,however; theflash strikes down the column ofhotairandthrough theflame, while itavoids themore lofty pointed terminal altogether. Bring apoint oraknob, oranything, info thecolumn ofbot airabovetheflame:thenitgetsstruckeasilyenough,andprotectsthe flame, but not ifitisonone side ofthe hot-air columa, ‘Theexperiment hasanobvious moral inrelation totheprotection ofchimneys: itsuggests that theContinental plan ofabaror archacross themouth ofthechimney may after allbejustified. 9,Trytheeffect ofresistance inthepath ofthedischarger ‘now, and wefind itisaltogether different towhat itwasincase Imcase bthings getstruck according totheir height, i dependent oftheshape oftheir terminals, butnotindependent oftheir resistances, Interpose awetraginthepath ofany one, and that onefails tobestruck; itisnotstruck, and itfails toprotect theothers from being struck, even though itbereared uptillittouches the topplate. ‘The topplate need nottherefore beinsulated atallcarefully forthis case bexperiment. Iurrarion oF Lionrsine. 10.Now letusmodify case bbymaking thetopplate asieve fallofwater, soastogettheflashes inashowerofrain.One cannot well tryease 1ina rain shower: the plate would be discharged toorapidly and continuously bythewater-drops for itspotential tofully rise, But incase 2thetopplate isnot necessarily charged atalluntil therush comes, and sothemin shower does noharm, Flashes intherain canbegotofsurprising length andshape, forthey make useofthewater-drops asstepping-stones. By adding salttothewater they become longer still, butthere is10 need thus toimprove itsconduetivity forwhat Iwant toshow. Notice how the flash contorts itself, taking sometime extraordinary paths asitjumps from drop todrop, butyet exhibiting itsinstantaneous character byshowing thedrops a stationary initsillumination.* 1669) ANDLIGHTNING PROTECTORS. ws 11,Remove thethings standing onearth-plate just beyond fairstriking distance, and what dowesee? Appearances pre- cisely like those which areobserved inmany lightning photo- graphs. ‘Acrowd ofviolet discharges filltherainy air—forks, and branch and multiple flashes, notvery bright orvery noisy, but extraordinarily numerous, and striking oninnumerable places at once, Arotissetupinthis airatevery attempt ofthejarto discharge, exactly ashappens inoneofthemost striking ofthe photographs belonging totheRoyal Meteorological Society. Masts and spars and deck and ocean may besimultaneously struck bythese interesting flashes;and,thoughtheydonothere appear very violent, yetIexpect that onthelarger scale of Nature they are not very safe, and may easily have aheating fect sufficienttoignitebodies.‘Theseexperimental onesareable toignite gasinthemidst oftherain. 12,While wehave this arrangement atwork wemay aswell ‘ryaninteresting little experiment onwhat happens when lightning reaches water. ‘Therain water hashere been collected inazinetraysomethreeinchesdeep,andbybringing aknob ffom thetopplate, some sixinches orsoabove thewater, aflash strikes it.Itprefers tostrike anything metallic ifitcan, butif there isnothing else within reach, itwill strike thewater. On reaching the water the flash forks outand ramifies inalldirec~ tions inacrow-foot pattern, giving thesame sort ofappearance, nlycoarser, asthat obtained byMr.J.Brown bytaking sparks ontoaphotographie dryplate andthen developing. Bring theknob nearer andnearer, thesame thing happens, until thewater istouched bytheknob, and even after itis submerged. Butassoonasthemetallicconductor issubmergedtheramifications getlessandless vigorous, and when asufficient surface isimmersed they cease. Ihave never seen these ramifica tionsspread ofthemselves below thesurface ofthewater. They appear tometokeep entirely tothesurface, But Ihave notyet finished investigating these appearances. 13,Immersing ahalf-full beaker inthewater, sparks canbe gottothewaterinsideit,thoughtheyprefertocurlroundandgo am ONLIGHTNING, LIGHTNING CONDUCTORS, (April253, outside, Thenoisethesparksmakewhentheygoinsideis«cariousone,andsoundsasiftheglasscrackedeachtime,butit does not. Iftheglass ismoist,abrushcascaderounditsedge can beseen inthedark. Ifitbedry, the water inside gets charged, and fizzes audibly back tothe knob forasecond or soafter thespark hasceased, adimple being visible inthewater below the knob. 14,Live things inthestruck water—worms, flies, fish,&, —will most certainly get struck;butsotheydounderfarles violent disturbances than what they would here besubject to Thereisnothingsurprising inthefishof#pondorlakebeing killed byaflash oflightning ;and itbasoften happened. When H.M.S. “Conway” wasstruck many years ago,andpr tected byitslightning conductors, itis related that thesea-rater ‘wasseentobeluminous onallsidesoftheship.‘Thisisexaclly the effect Inow imitate, 15,Thereisonemoreexperimentondischargeinwaterwhich Thave just tried, andwhich itisinteresting toshow.* Itakes pointed rod,and, protecting itbyaglass tube, immerse itspoint under water inabeaker containingaplateconnectedtotheother contofthejar,andpassaspark.Withthepointnegative, thereitabright glow region round itevery time, butthedischarge isquiet. With thepoint positive, theflash isofadazzling white,andis accompanied byagreat deal ofnoise andviolence, threateningto smash the Leyden jar,and throwing down the copper plate towards thebottom ofthebeaker with fury. OsciutaTory CHARACTER oFLIGHTNING. 16,Before leaving theoutdoor department ofoursubjectI mustsayafewwords ononebranch ofitconcerning which there isevidently considerable uncertainty and haziness abroad—I mean theoscillatory character ofalightning flash. ‘That aLeyden jardischarge isusually oscillatory must nowbe regarded assoextravagantly proved that any doubts that may haveexistedonthesubjectmustsurelybythistimebecleared ‘©This experiment wasnotshown atthemeeting, because intrying itoe {intheafternoon theprotecting glass tube waselectrically broken, 1880) ANDLIGHTNING PROTECTORS. 395, ‘away,atleastforthecasewherethedischarge hastoutilisea wire circuit. But perhaps itisstill doubted forthe case when 1jaroverflows itsedge, or,still more, when itmerely sparks through itsowndielectric, straight between thecoatings. Now, asIhave insisted allalong, alightning flash isaspark through thedielectric ofajarwhose twocoatings areeither two clouds, orelse cloud and earth. Hence, ifany importance is attached tothe fact (asIbelieve it)that lightning flashes are oscillatory, itisnecessary toprove itforaLeyden jarsparking direct. between itscoatings, especially when thecoatings arenot very close together. 17,The reason Idoattach importance tothe oscillatory character ofadischarge isbecause Ihave worked outthequanti- tative behaviour ofconductors onthat aspect ofthe matter; and though,asProfessorFitzgeraldsaidatBath,everything would holdjust aswell forasingle oscillation—viz., oneviolent riseand decay ofcurrent (which without any doubt must accompany a lightning stroke oranyother quick discharge whatever)—if rapid enough, yettherapidity ofsuch acharge asthis does notseem to meprobably atallsufficient toaccount forsome ofthe effects. Therapidity ofvariation ofcurrent inthat case would bedirectly connected with thetotal duration oftheflash; andthough wehave evidence that itisvery momentary, yetwehave noevidence that itissoinstantaneous (sayamillionth ofasecond)asthesemi-Periodofoneoftheoscillations maybe,adozenormoreofwhich may accompany anentire flash. However, Iadmit, ofcourse, thatallIwant isatremendously rapid variation ofcurrent; and ifcanbegiven thisbyoneoscillation, therestareunnecessary, andmay bedispensed with. 18,When Ispeak oftheoscillatory character ofaflash, letit beunderstood once forallthat Idonot mean inthe least such a thing ascanbeanalysed bywaggling thehead. Flashes analys- ablebywaggling the head must bemultiple ones, and the interval oftime between their constituents (which may be,eay, thefiftieth ofasecond orthereabouts) isalong period compared Withthatofanoscillation such asImean, bearing thesame sortofratiotoitasaquarterofacenturybearstoanhour. 206 ONLIGHTNING, LIGHTNING CONDUCTORS, [April2a, 19.Adirect experimental proof that lightning isoscillatory willbeobtained when photographs ofitaretaken onasensitive plate revolving 1,000 times asecond. Something short ofthat speed would cause theimage oftheflash toblur, butthatspeed might besufficient toanalyse outtheoscillations, when examined carefully with amagnifier, thefocussing being good. ‘Till then theeasiest proof that itisoscillatory istheoretical one, and itcanbeputinafewwords. 20.Consider anaircondenser with twocoatings, each ofarea A,separated bythedistance A,and letitburst itsdielectric. It iswell known that the discharge isoscillatory when thewhole resistance met with bythe discharge isanything less than a critical value. =o, /(hR=2/|(s) Now, attending only tothestraight part ofthedischarge, and ignoring thecurrent rushing uyinthe plates tothe spark path, theself-induction ofastraightconductor oflengthhandsectionalradiusaisveryapproximately L=2ph tog44, : ‘Thecapacity ofthedischarged condenser is ._ KAS= dak Hence the critical resistanee which must not beexceeded is given by 38p ah.Ren 2AM jogAh, or aoe,J(ly4h R=l0ph*vVJog=) 1 4h=300ohmsxS(tlog**)xh ‘Theimportant thingtonoticeinthisvalueof4//Aisthat itisapproximately proportional tothefirst power of, the distance between theplates ofthecondenser. 21,Next consider the resistance ofthedischarge path. It 1#223 ANDLIGHTNING PROTECTORS. 07 toowillbeproportional tothelength h,anditmay bewritten Rh, where R,istheresistance perunit length ofthedischarge path. The condition foroscillation, then, isthat this R,hshall be less than Ry5oFtheresistanceofunitlengthofthesparkmust delee than. k 14 300ohmsXVAGlog**). 22,The term under thesquare root will take different values according tocircumstances, and itmay begreater orless than 1 permetre. Ordinarily, however, itwill begreater than 1per metre, unless thearea ofcharged surface isconsiderable. The important thing istheway inwhich h,thedistance between the plates, enters into theexpression. Itdoes notcome invery prominently atall, but sofarasitdoes influence theresult it permits the discharge tobeoseiliatory more easily when the Plates are agood distance apart than when they are close together. 23,Take acouple oftypical examples. First, aLeyden jarbursting itsglass. Afineneedle may bejust pat through the bole usually made inthese cases,sowecantakethesectional radiusaassomething like atenth ofamillimetre. The thickness ofthe glass may be2millimetres, hence 4¥/,= 80or thereabouts; and the natural logarithm ofthis isabout4,Supposetheareaofcoatedsurfaceishalf fmetre square, then the critical resistance which a metre ofthe spark must not exceed is1,200 obms, and sothe 2millimetres ofitmust not excecd 2-4 ohms. Itisdifficult tosaywhether thisisorisnotalarge resistance forsuch ashort, spark, andhence itisdifficulttobesurethatsuch ‘spark isanything more than amere one-directional discharge. Togointoitmore fully thecurrents inthemetal coatings would have tobeconsidered. Next take asexample acloud area atanelevation ofone kilometre; andbecause alightning discharge usually 308 ONLIGHTNING, LIGHTNING CONDUCTORS, (April254, makesperforations offairdiameter, wemaysuppose to beabout amillimetre. (Awidely erroneous est- mate inthis quantity makes but little differencein theresult.) Insuch acase 44/,=4X10,andthe logarithm ofitisbetween 15and 16. Hence the charged area may be15or16square metres without bringing down the square root term below 1per metre; sothe’ resistance ofthe whole flash mayia that case beanything below 300,000 ohms withoat checking theoscillations. ‘The discharged area may indeed beasmuch as1,500 square metres without bringing thecritical resistance which the flash must notexceed below 30,000 ohms, or30ohms permetre (Understand that “resistance” here does notmean impedance. Itmeans true dissipation ofenergy resistance (sec. 25), and the current squared is+0 ‘enormous (sec. 29)that theresistance-coefficient may bequite small). Inanother place (Phil. May., August, 1888) Ihave shorn reason forbelieving that thearea ofcloud discharged atanyont fiash isusually very moderate; and hence onthewhole Iconsider itproved, sofaraselementary theory candoit,that thelightning flashusuallytakesplaceunderconditions favourable tooscillation. 24,And these oscillations areextremely rapid. ‘The rapidity dependsontheinversegeometric meanofLand8,andthisis practically almost independent ofk.Referring back totheir values we sce that LS=AKAtog4h; Qn « andsothenumber ofcomplete alternations persecond is Towaton 1& VA(2mrAlog**) which, inthesecond example ofsection 23,with Aas100square metres, becomes 3million persecond, 1882) ANDLIGHTNING PROTECTORS. 398 Ifthe discharged area were asgreat as10,000 square metres, the nate ofalternation would still beathird ofamillionpersecond. Marentat, oF tHe LicHrsixG Coxpucror. 25.Afew words may suffice toexplain the nature ofthe impedance which alternating currents meet with inpassing through aconductor, and ofthereason why iron isasgood as,or exen better than, copper forthepurpose ofconveying currents alternating with extreme rapidity. Arising current hastomagnetise thespace allaround it,and theproduction ofthis magnetisation delays andimpedes therise ofthecurrent toitsmaximum value. Afalling current permits themagnetisation ofthe space allround ittodecay, and the dying out ofthis magnetisation delays and impedes the fallof thecurrent toitsminimum. The more rapidly the current changes, themore powerfully felt isthe influence oftheaccom- panying magnetisations and demagnetisations. Now the total magnetisation produced byacurrent—its total number oflines offorce,oritstotal“magnetic induction,” asitisoftencalled—isproportional tothecurrent strength:equaltoitmultiplied by some constant, which wemay callL,and write I=L, where Iisthe total induction produced bythecurrent C. 1 isacoefficient characteristic ofthe circuit, itsvalue being defined by this equation, and iscalled the coefficient of induction excited bythe current's own self, orthe coefficient ofself-induetion.Ifthecurrent goes through p/2xcomplete alternations ina second, itcanbeshown that theimpedance itmeets with, due to thereversals andre-reversals ofitsown magnetic field, ispI. ‘This isnotthewhole obstruction itmeets with, butitisthe only part which does notdissipate energy andcause itsvibrations todecay. Itmay becalled theinertia part oftheimpedance, ‘The remaining part ofthetotal impedance isresistance, R,the dissipation ofenergy coefficient, defined by heatpersecond=RC's andthe total istheresultant ofthese twoasifthey were 400 ON LIGHTNING, LIGHTNING CONDUCTORS, (April, atright angles toeach other. Sothat, calling Pthetotal impedance, ate P=(pl+R DE NowinrespectoftheRterm,ironismuchworsethancopper, notonly 7times worse, hundreds oftimes worse;butthenfor very rapid alternations the Rterm isaltogether insignifiant ‘compared with thepLterm. 26.Inrespect ofthepL term does the material ofthe ‘conductor matter? ‘Well,insofarasthemagnetisation spokenofisthatofthe ‘space surrounding theconductor, ofcourse thesubstance ofthe conductor itself matters nothing.. But insofarastheconductor itself gets magnetised, thematerial ofwhich itismade dies matter. Now alinear current magnetises atright angles toiteell everything surrounding it—most intensely thethingsclosetoit. Ahollow cylindrical current magnetises everything outside itself, Dut nothing inside. Ifthe current were todistribute itel! uniformly through thesection ofthewire, theoutside ofthewit would get magnetised inconcentric cylinders ;butifthe current were toconfine itself tothe outer surface, and flow ass hollow cylinder, itwould escape thenecessity ofmagnetisingthe wire atall, Ineases where thepL term ismuch more important than theRterm this isprecisely what itdoes therefore. It always flows soastomeet with theleast possible total obstruction; ‘and itfinds less total obstruction bycramping itself into the periphery ofthewirethan itwould findifitutilised thewhole section. ‘The cramping into theperiphery increases R,butit decreases pLL;and onthewhole with rapidly alternating currests this isanadvantage, andgives asmaller value toP. Slowly changing currents think most about R,and usethe biggest cross-section they can find. Rapidly changing currents think most about pL,andavoid having tomagnetise more thas they need. Especially must they avoid having tomagnetise the conductor ifitconsists ofiron: hence inthat case they cramp themselves tremendously into itsouter skin, and thereby avoid 188] ANDLIGHTNING PROTECTORS, 401 having toovercome much more total impedance than they meet with inthe case ofcopper; butthough they thus keep down impedance, they increase their dissipation ofenergy term, andget their oscillations damped outfarmore quickly than when they only have topass through copper. The violence oftheflash therefore subsides more quickly inaniron, than itdoes ina copper, conductor;andsoIfinditexperimentally. - Nothing here said iainthe least hypothetical. Itisall absolutely clear and certain, and hasbeen abundantly verified. {will notgointo thehistory ofthesubject. Itiswell known.27,Letitbeclearlyunderstood onceforall,thatincomparing copper and iron conductors Inever mean comparing them asof ‘unequal thickness, i.e. ofequal conductivity. Such acomparison isridieulous inthepresent state ofknowledge. They aretobe ‘compared when oftheeame thickness; and under those circum-stancesIassertirontobeatriflebetter,certainly notawhitworse, than copper;irrespectiveofallitsothermanifestadvantages, cheapness, fusing-point, ete., ete. Want offlexibility issome- times urged against iron, butstout telegraph wire isflexible enough;and,untilexperiencedecidestothecontrary,Ifeelsure itisthick enough forlightning conductors. Theoneandonly thing onwhich anything canbesaidagainst ironisonthesubject ofitsdurability;andthisbeingachemical question, Ioffer nopositive opinion; atthesame time Jamabso lutely certain that anyslight disadvantage inthat respect isa hundredfold compensated inmost localities byitsother superlative advantages. CunnexT AND POTENTIALS DURING DISCHARGE. 28,Incases where aconductor ispretty thick, sayanything likeaquarter inch diameter, oreven atenth, andofanymoderate length, such as100yards orless(not many miles), thetwoterms ‘ofimpedance aresounequalthatitisformanypurposes needless tothink about Ratall,theimpedance ispractically pLsimply; andthis, asIhave shown inPhil. Mag., August, 1887, isunder anygiven circumstances half the critical resistance which deter- amines whether thedischarge shall beoscillatory ornotunder the Vou. XVII 28 wa ONLIGHTNING, LIGHTNING CONDUCTORS, [Aprl:3t, saine circumstances, The total impedance iscommonly tobe reckoned inhundreds oreven thousands ofohms. 29.Thestrength ofcurrent passing inthealternations canbe estimated byconsidering thatthewholequantity storedupinthe discharged body hastobetransmitted inaquarter ofoneoscilla- tion period—say, forinstance, inthe millionth ofasecond. If thequantity discharged were onecoulomb this would mean current ofamillion amperes, Inanycase thecurrent mustbe hundreds orthousands ofamperes. 30.The difference ofpotential needed todrive sostrong s current through sogreat anobstruction isenormous, being equal totheproduct PC, and may bereckoned inmillions orhundreds ofmillions ofvolts; hence itisthat lightning conductors afford noeasy path forlightning, butthat ittends tospit offioall directions, even towhat would seem, and indeed are, very inferior conductors. Itwillspitofffrom awell-earthed stout copper rod tobitsofwood and toperfectly insulated bodies. Evrects 1xtH NeiGHnoveHoop oF 4DiscuaRce. 31.Putting together theenormous electrostatic potential: existingatthedifferentpointsofaconductorconveyingwdischarge, andtheviolent oscillation towhich sostrong acurrent issubject weperceive how great must beboth the electrostatic andthe electro-magnetic induction inallspace anywhere near it. These disturbances, thus rendered certain and accounted for ‘can easily beexperienced experimentally. Iproceed torelates fewexperiments outofamultitude which Ihave made. 32.Itook aconsiderable length ofhighest conductivity No.! electrolytic copper wire, kindly lent mebyMessrs. Thos. Bolton &Sons, andarranging oneend soastobeaccessible toLeyden jars, etc., carried thewire uptoahigh gallery and then down to earth, Earth was made inseveral ways ondifferent. occasions: water pipes, gaspipes, hot-water pipes which ramified thewhole building, outside gasmains, bars buried inground;butthemost effective and indeed perfect return circuit could behadwhen wanted byconnecting theend metallically with theoutside coat ofthejar,without anyearth contact atall. When thisdirect rs) ANDLIGHTNING PROTECTORS. 403, contactwasnotmade,theoutercoatofjarhadofcoursetobe connectedtoearthalso,inordertocompletethecircuit.Very oftentheywerebothconnected toeachotherandtoearthaswell. Itmakes noessential difference;whatevermaybeconsideredthe ostsatisfactory method, that may beadopted,and thephenomena villgoonjust aswell. They may bebriefly summarised asfollows:— (1.)Iftheconductor pass within aninch ortwo ofany ‘uninsulated piece ofmetal, itgives offaviolent side fiash toit.” Ifthe far end ofthe conductor isneither earthed norconnected tojar,but isleft: insulated inair,side- flashing from itoccurs atrifle more easily, butnotvery markedly 80. (2)Iftheconductor pass within, say,half aninch ofan insulated conductor, itgives offaside flash toit;the strength ofthe flash depends onthe capacity ofthe insulated conductor, being considerable ifitbelarge3 but. some side spark occurs toanabsurdly small body perfectly insulated, eg., such athing asacoin onu stick ofsealing wax; and this when theconductor is absolutely well earthed atitsfarend. (8.)Sparks can beobtained from everything, even quite uninsulated things, connected tothe conductor: for instance, ifitbeconnected tothegaspipes, small sparks will flytothe finger ortoaninsulated body from all thegasbrackets about, and these sparks aresufficient toignite gas. (4.)Sparks canbeobtained between theends ofanylong curved conductor, be itinsulated oruninsulated, if they arebrought close enough together toform anearly closed circuit. (5.) Sparks can beobtained from orbetween insulated bodies intheneighbourhood oftheconduetor, andnot connected toitatall,every time aflash occurs. ‘This eary and striking experiment Istupidly omitted toshow atthe meeting. For instance,alargepieceofwiregauzeconnected tonothing gave offsparks toagasbracket andignited thegas. Moreover, one piece ofwire gauze sparked smallsparksintoanother, neither connected withany- thing. [Jnillustration ofthie, some giltkey-patiera high uponthewall ofthehall, 25,Great George Sire, ‘was seen tobesparking while Leyden-jar discharge ‘weregoingonthroughawirelyingonthefloor.Theywerenot20brightasthesparkings intheRoyalInaitution wall paper, butthegilding wasfurther from wy wire, and Ibelieve isquite far from any wire. At ‘illustrating theelectric currents produced incondudors during theactofreflecting electro-magnetic waves they ‘were therefore stillmore satisfactory. (6.){Sparkscanbeobtained fromquiteuninsulated bodies‘evenwhennotconnected withtheconductor: e.,from hot-water pipes, from gaspipes, from water pipes, from strips ofbrass letinto thetable, from gasbracketsin otherrooms,fromawirelyingonthefloorofadistant corridor; and, infact, allover thebuilding, with few exceptions. The sparks can betaken byapenknife held inthehand; sometimes they can betaken tothe knuckle orfinger-tip even when pressed against them. (7) Sparks can begot topass between two totally un- insulated things, neither ofwhich have anyconnection with the lightning conductor. For instance, letthe conductor bethoroughly earthed insome outdoor and distant manner, and under favourable circumstances bright short spark can beseen passing atevery discharge between agastapand awater tapofmy lecture table which happen toapproach each other closely. Let thegasescape near these sparks andit ignites. (8.)Sparks canbegotbetween twothinly insulated electric light wires ifthey lieclose enough togetber, andif« storage battery beconnected tothem anarewill be started, destroying theinsulation andburning thewires. 1989) ANDLIGHTNING PROTECTORS, = 405 (@.)Ifatanydistant place, oroutofdoors indaytime, the sparks aretoofeeble tobeseen, disturbances canstillbe easily detected bymeans ofatelephone;connectingone terminal tothething—say, theroof ofashed, orawire fence—and holding theother end inonehand. Or,of course, byconnecting thetwoterminals totwodifferent things,ortodifferentpartsofonething. (10.) Arrange Abel’s fuses between gas and water pipes, between pieces ofwire gauze andgaspipe, between hot~ water pipe andabitofsheet metal, between thelight- ning conductor anda6-inch metal sphere onlong glass stem, between two large insulated bodies, between the gasbracket ofanother room andanempty Leyden jar5 inshort, inalmost anyplace, likely orunlikely. ‘Then takeafewstrongdischargesthroughthethickcopper rodwith end well earthed, and the fuses will pop off, some atone discharge, some atanother. Very visible sparks cansometimes beseen passing into thefuses, ifthey benot closely connected, without exploding them. Acertain energyofsparkisnecessarytoignite thecomposition: much more than issufficient toexcite the retina. Ligatnine Protecrors. 33,Now letmesayaword about thepossibility ofprotecting telegraph and other instruments from damage bylightning curents which may have entered theaerial lines. ‘There isindeed noguarantee that burying wires beneath a Pavement, oreven beneath water, will effectually secure them fromlightning disturbance, axIhave nowfully illustrated, but certainly overhead wires aremore exposed. ‘Theordinary and well-known form oflightning protecting amangement istoattach apair ofplates, oradouble setof points, orapairofpointsin@vacuum, orsomeothersmallair ‘pace,asashunttotheinstrument orcoilofwiretobeprotected. Werearearrangement’ ofthekind. Now itisperfectly easy to 406 ONLIGHTNING, LIGHTNING CONDUCTORS, [Apri2th, seethattheprotection suchthingsaffordisofthemostutterly imperfect kind. Takeacoiloftangledsilk-covered wire,oranyothercolthatyoudon’tminddamaging, andattachitasashunttooneofthese protectors. Ondischarging ajarthrough it,theinsulation ofthe coilispierced inheaps ofplaces. Oronemay useashort fine wire, andseeitdeflagrated bythebranch discharge. 34.Itmay besaidthat myairspace istoowide. Very well, then, abolish italtogether. Bring theplates ofyour protector into direct metallic contact. Bysodoing, the coil isindeed shunted outofthecircuit, andifitwere atelegraph instrument nosignal could begiven, fornoappreciable fraction ofthecurrent, .takes that route; but with aLeyden-jar flash itisotherwise Although the plates ofthe protector areincontact, soldered together ifone pleases,and leduptobystout wire orrod,& branchflashstillbreaksthroughtheinsulation ofourwiretangle, ordeflagrates ourlittle bitofthin wire. 85,Itwillbesaid thejoints arebad. Well, then, dowithoat joints;takeastoutrodofelectrolyticcopperbentinanarc,sy, 2feetlong, and, bridging itacross with thewire tangle, discharge ‘ajar round it.Still aportion takes thethin wire, even thoughit offer apath yards inlength. Take astraight barofcopper aninch thick, andarrangean invisibly fine Wollaston wire ofgreater length asatapping circuit. Some ofthedischarge shall leave thebar, andspark across aminute airgapateach end, inorder tomake useofthe hair-like platinum wire. Go further still than this, Connect acoil ofthin-covered wire byoneendonly toawireconveying adischarge, standing theree upon ablock ofparaffin orother good insulator, andconnect the other endtoanylittle thing ofanycapacity atall—say, abullet lying ontheblock ofparaéin: youcanseesparkings through the insulation ofthewire onthereel atevery discharge. ‘Thoseexperiments rendermanifest thehopelessness ofany simple shunt arrangement asalightning protector. 36.The easiest mode ofexhibiting theessentials ofthese experiments—one that can betried byany one possessing 1880 ANDLIGHTNING PROTECTORS, 407 Leyden jar,apairofdischarging tongs, andayard ortwooffine silk-covered wire—istohangatangleofthewirelooselyontothe tongs, notnecessarily making anysort. ofyood contact, and then wethem todischarge ajarinthevery ordinary way. Some of theflash will take the thin wire, and will spark through the insulation atanumber ofpoints (Fig. 3). 37.Itmay bevery well objected tomethat itispretty useless ifTonlypoint outtheimperfection ofpresent methods, andoffer 1suggestion astoaproper lightning protector. Well, this ruck me too, and the result isthat Ihave devised and made what Ithink Imay callanabsolutely perfect protector—one into which great, flashes may besent, andyetthegalvanometer or instrument intended tobeprotected shall notwink, norshall the tlightest palpable orvisible disturbance bediscernible, notwith- standing that complete metallic contact ismaintained allthe ‘ime, andnotatrace ofthesignalling oruseful current wasted. ‘Thething seems sogood that Iam making itthesubject ofa ystent, and you will therefore pardon meforsaying nomore about itjust yet. 38,The next part ofthispaper islargely controversial. 1do tottakemuch pleasure inthisportion, andwish itwere unneces- sary, Butitisapurely impersonal controversy;andsolongas theold-fashioned views areinexistence, onewayofarriving atthe truth istotryand thrash them outofexistence. Ifthere isany realvitality inthem, theattack willfail. Itwillbewellbelieved thatIhavenofeelingofhostilitytotheLightning-rod Conference, when mybest scientific friend, Professor Carey Foster, wasoneof itsmembers. Had Ibeen onemyself, Inodoubt should atthat time have signed thevery documents which now, inafewplaces, Teriticise, HadIindeedsosigned,IwouldubusewhatInowsee tobeitserroneous portions with still more vigour than Inow Permit myself toemploy. Sumwany oF Pours oFDIFFERENCE AND CONTROVERSY. 39Itmay beconvenient here tosummarise afewofthe Points wherein thedoctrines which Iadvocate differ from theviews 408 ONLIGHTNING, I1GHTNING CONDUCTORS, [Aprisb. held bytheolder electricians. Andthesummary willgivemean ‘opportunity ofemphasising theincorrectness oftheolder viewsin many instances. Tquote afew statements prefixed bycapital letters froman abstractImadefortheElectricianaftertheBathmeetingofthe British Association. (See Electrician, Sept. 21and28,1888.) A,Rodsasatpresentconstructed, thoughfrequently suecessful, ‘may anddosometimes fail, even though their earth isthoroughly ‘good; thereason being that they offer toaflash amuch greater obstruction—a much worse path—than isusually supposed: an obstruction tobereckoned inhundreds orthousands ofohms, even foravery thick copper rod. N.B.—This isnotresistance proper, butimpedance. Imay bepermitted here torepudiate thedoctrine which bas several times been attributed tomesince theBath meeting, that itissafest tobewithout lightning conductors altogether. The long experience ofpersons learned inthis artisbynomeans o bedespised, anduntil anagreement astoimprovements hasbeen arrived at,the safest plan forordinary persons ixtoadhere to existing practice. Nevertheless, that conductors sometimes fil isascertain asthat they often succeed. Mystatement isthat customary arrangements arenotperfect, and aresusceptible of improvement. Thestatement oftheLightning-rod Conference is that “there isnoauthentic ease onrecord where aproperly eon «structed conductor failed todoitsduty.” Mr.Preece calls the statement “most decisive.” Itiscertainly decided, andinthe light ofother matter contained inthesame redcovers Iassert it tobeinitsnatural and intended signification decidedly fale. ‘The only signification which makes ittrue makes italso senseless; asifoneshould record thestatement that white things arewhite. InmySociety ofArts lectures, Isaidnothing against thereport oftheLightning-rod Conference, because thework done bythat body, incollecting information, abstracting papers, andrecording instances, was obviously very valuable, and much oftherepot itself iscorrect;whileasfortheoccasionalrashstatementsia that document Iimagined thesignatories would wish them t0 sinkintooblivioninsilence. ButMr,Preecehasrevivified them, 1990) ANDLIGHTNING PROTECTORS. 409 and conspicuously made himself afresh responsible forthem ;* accordingly Inowextractoneortwomoresufficiently dogmaticand unfortunate statements from the same source. “Aman may with perfect impunity clasp acopper rodan “inch indiameter, the bottom ofwhich iswell connected with “moist earth, while the top ofitreceives aviolent flash of “lightning.” “If allthese conditions befulfilled; ifthepoint behigh “enough tobethe most salient feature ofthe building, “no matter from what direction the storm may come, beof “ample dimensions, and inthoroughly perfect electrical con~ “nection with the earth, the edifice with allitcontains will “be safe, and the conductor might even besurrounded by “gunpowder inthe heaviest storm without risk ordanger.” ‘Toadhere tosuch views asthese now, with tenacity sufficient to cause them tobepromulgated asauthoritative scientific state- ments, would be,inmyopinion, little lessthan criminal. “All accidents may besaid tobedue toaneglect ofthese simple elementary principles.” Certainly this “may besaid,” weause italready has been said over and over again; but it cannot besaid with truth, Whenever anaccident happens, abeliever inthe modern exponents oftheLightning-rod Conference who could notpoint, outaflav, orabadjoint, orabadearth; orapossible flaw, or possible bad joint, orapossible bad earth;wouldfeelhimself disgraced as@practical man, Aninstance occurred inquite a recent number of Nature. ‘The writer of “Electrical Notes” records that anumber offish had been killed inapond into which theearth end ofaconductor had been led, and concludes ith theejaculation, “When will people learn tomake proper “earth connections?” Iselect this instance astypical ofthe extraordinarily contradictory advice often bestowed onthat long- suffering body theBritish Public. Before anaccident, thepond. ‘+SuyingatBath,axapreludeto#quotationfromtheReport,«The“Report itaelt ismost decisive and most important; there isatrueringabout “itheringofmenwinoknewwhattheyhaddone,andwhattheywere “writingabout.” a0 ON LIGHTNING, LIGHTNING CONDUCTORS, [Apri 2, would bepointed out asanexcellent earth, asindeed itmost likely was. After the slaughter ofthe fish, the erector of ‘eonductor isimpersonally ridiculed forhaving utilised it.So with any struck building. After anaccident defects must be forthcoming, because elsethere would be“an authentic caseon «record where aproperly constructed conductor failed todoits «duty,” which exhypothesi isabsurd and impossible; therefore itwas not constructed inaccordance with the directions ofthe Lightning-rod Conference, therefore itwasdefective. Q.E.D. ‘B.When aLeyden jarischarged itcorresponds toabent spring, anditsdischarge corresponds totherelease ofthespring. Itsdischarge current alternates, therefore, inthesame wayand formuch thesame reason asatwitched reed ortuning-fork vibrates. Thevibrations decay ineither case because offrictional heat production, and because ofthe emission ofwaves intothe surrounding medium. Asingle spark ofaLeyden jar,examined inanexceedingly fast revolving mirror, isvisibly drawn outinto nclose succession ofoppositely-directed discharges, although its whole duration issoexcessively minute, Itisvery likely that this statement will now nolonger be denied. SoIpass tothenext. CA lightning flash isaspark between cloud and earth, which aretwooppositely electrified flatsurfaces, andthefash corresponds therefore tothe internal sparking between thetwo plates ofagreat aircondenser, Alltheconditions which apply toaLeydenjarunderthesecircumstances areliabletobetrae forlightning. Sometimes theresistance met with, either in thecloud itself orinthedischarger, may besogreat thatthe sparkceasestobeoscillatory, anddegenerates intoafizzorrapidleak; butthere canbenoguarantee that itshall always take this easily manageable form; and itisnecessary inerecting protectors tobeprepared forthe worst and most dangerous form ofsudden discharge. ‘Theapparent duration ofalightning flash isdue toitsfrequently multiple character, and indicates successive discharges, notonelong-drawn-out one. Nothing that lightning hasbeen found tododisproves itsoscillatory characters ecause Leyden jardischarges, which arecertainly oscillatory, eandoprecisely thesame, wee ANDLIGHTNING PROTECTORS. an ‘This wasinanswer toMr.Preece’s contention that lightning could notbeoscillatory, because itmagnetised steel bars and deflected ships’ compasses. The next isanantidote tothe continually made statement that theonething needful foran ficient lightning conductor isconductivity. D.Although some conductivity isnecessary foralightning conductor, itsamount isoffarlessconsequence than might be expected. The obstruction met with byanalternating orrapidly varying discharge depends much more onelectro-magnetic inertia orself-induction than upon common resistance. So much obstruction isdue tothis inertia, that atrifle more or Jessoffrictional resistance, inaddition, matters practically notat all.Itisverydesirabletohaveagoodanddeepearthinorderto protect foundations and gasand water mains from damage, and inorder tokeep total impedance aslowaspossible. Ifinditsometimes thought that Ihave argued against the teed ofagood earth. This isnotso. Ihave argued against theexclusive andexaggerated attention that hasbeen paid to thisneed. Inopposition tothefollowing statements, ¢,f,g,h, i,the ‘mbstance ofwhich may beconsidered ashitherto orthodox, I make thestatements subsequent, labelled E,F,G,H, I:— «Nodanger istobefearedfromalightningconductorifonlyitbewell farthed and besuificiently massive not tobemelted byadischarge. All ‘ates ofmetal should beconnected toit,that they may beelectrically drained toearth, J.Theshapeofthesectionalareaofaconductorisquiteimmaterial; ita‘anyingpowerhasnothingtodowithextentofsurface;nothingmattersin theroditselfbutsectionalareaorweightperfootrun,andconductivity. 2.Points, ifsharp, should constitute aogreat aprotection that violent ‘ashestothemoughtnevertooecar. A.Lightaing conductors, iffrequently tested for continuity and low‘esstncebyordinarygalvaniccurrents,areboundtocarryoffanycharge‘uelytostrikethem,andareabsolutelytobedependedupon,Theeaiet ath protects allother possible paths. 4.A-certainspacecontiguoustoalightningrodiscompletelyprotected Wyitythatiftherodberaisedhighenoughabuildinginthisprotected"pion ieperfectly mate, a2 ONLIGHTNING, LIGHTNING CONDUCTORS, [April50, These statements Isay areerroneous. The followingI believe tobe correct :— E,Theobstruction offered byalightning rodtoadischarge being sogreat, and the current passing through itattheinstant ofaflash being enormous, avery high difference’ ofpotential exists between every point oftheconductor andtheearth, how- ever well the two areconnected; hence theneighbourhood ofs lightning conductor isalways dangerous during astorm, and great circumspection must beexercised astowhat metalic conductors arewittingly orunwittingly brought near orinto contact with it.When abuilding isstruck, the oscillations andsurgings allthrough itsneighbourhood aresoviolent that every piece ofmetal isliable togive offsparks, and gasmsy belighted even inneighbouring houses. Ifoneend ofarait- water gutter isattached toastruck lightning conductor, tbe other end isalmost certain tospit offalong spark, unless it also ismetallically connected. Electric charges splash about in struck mass ofmetal, asdoes theseaduring anearthquake or whenamountain topdropsintoit.*Evenasmallsparknear combustible substances istobedreaded. F.The electrical disturbance isconveyed toaconductor through the ether orspace surrounding it;expressed more simply, lightning currents make useoftheperiphery ofacom ductor only, and sothe more surface itexposes the better Better than asingle rodortape is.anumber ofseparate lengths ofwire, each thick enough nottobeeasily melted, andwel separated soasnottointerfere with ench other bymuta induction. ‘The liability ofrods tobemelted byaflash canbeeasily over-estimated. Arodusually failsbyreasonofitsinertislte obstruction, and consequent inability tocarry offthecharge without spittings and side flashes;itveryseldomfailsbyressoo ofbeing melted. Incases where athin wire hasgotmelted, the energy hasbeen largely dissipated intheeffort, and itasacted asanefficient protector ;though,ofcourse,forthattimeonl. ©SeetheReportoftheKrakatoaCommittee, | 1860) ANDLIGHTNING PROTECTORS, as Largesectional areaoffersverylittleadvantage overmoderately small sectional area, such asNo. 5B.W.G. G.Points, ifnumerous enough, serve avery useful purpose inneutralising the charge ofathunder-cloud hovering over them,andthusoftenpreventaflash;butthereareoccasions, casilyimitatedinthelaboratory, whentheyareofnoavail:for instance, when one upper cloud sparks into #lower one, which thensuddenly overflows toearth. Inthecase ofthese sudden rushes, there isnotime for@path tobeprepared byinduction, notime forpoints toexert any protective influence, and points then getstruck byaviolent flash just asifthey were knobs. Discharges ofthiskind aretheonly ones likely tooccur during violent shower, because allleisurely effects would beneutralised bytherain-drops betterthanbyaninfinitude ofpoints. H.The path chosen byagalvanic current isnosecure indication ofthe course which will betaken byalightningfash,‘Thecourseofatrickledownahill-side doesnotdeter- mine thepath ofanavalanche. Lightning will notselect the easiest path alone;itcandistributeitselfamonganynumberof posiblepaths,andcanmakepathsforitself.Ordinarytesting ofconductors istherefore noguarantee ofsafety, and may be misleading. Atthe same time itisquite right tohave some system oftesting and ofinspection, else rust and building alterations may render anyprotector useless. I.There isnospace near arod which can bedefinitely styledanareaofprotection,foritispossibletoreceiveviolent sparks orshocks from theconductor itself; nottospeak ofthe innunerable secondary discharges which, byreason ofelectro- kinetic momentum andinduction, byreason ofelectro-magnetic ‘waves, andofthecurious recently discovered effect ofthe ultra violet light ofaspark, areliable tooccur assecondary effects inthe wake ofthe main flash, Issravcuive Exrnacts Fao Reports oFDAMAGE BYLIGHTNING. 40,Reading between thelines ofexisting reports ondamage done, onecasfrequently findevidence ofmany ofthephenomena towhich Ihave now called attention. Ofcourse they arenot au ON LIGHTNING, LIGHTNING CONDUCTORS, [April 151, recorded inany prominent manner, because they aretothe observers ill-understood and puzzling facts: itisvery difficultto note what has exactly happened inanygiven case unless some elue orexpectation has been formed beforehand. The bad conductivity clue, which wastheonly oneprominently available totheskilled recorders inthefollowing cases, isavery partial one, and inmany cases isquite insufficient toaccount forthe facts without undue pressure being putupon it. Under such circumstances the record ofthe facts isofcoure farmore valuable than the comments made upon that record; and inthefollowing extracts the theoretical remarks should be eliminated orslurred over. One minor imperfection, common tomany accounts,isthat they arenotsufficiently alive tothe possibility ofallmanner of branching discharges: xothat thedischarge issaid toleave scun- ductor andgotosomething else, when thetruer statement would bethat some portion ofitbranched offatsuchandsuchapoint. Inthefollowing quotations thereferences areusually tothe pages ofthevolume oftheLightning-rod Conference, published bySpon in1882:— 41.Mlustrating Side Flashes and Surging Circuits. “LROy p.39.—J. Murgatroyd. St.Mary's, Orumpeall, near Manchester.—A lightning conductor from spire touched the eaves gutter, andagaspipe touched theendofthisgutter. The lightning passed from theconductor along thegutter tothege pipe, melted it,andsetthechurch onfirebyigniting thegas” «LBC. p.39.—Wyult Pupworth.—Tall spire struck. The church stands inanopen position with nolarge trees near. It wasprovided with aniron lightning conductor 3in,diam, fixed withironholdfasts, andcarrieddowninsidethespireandtorer into ground;thetopofitwassaidtobeattachedtoaboldcopper finialonthespireabout150fectfromtheground,and50feet above ridge ofroof; thelightning issupposed tohave firststrack thefinial, itslightly deranged some beds ofmasonry inupper partofspire,thendescended byironrodtobelfry,meltedags tube inthefloor,andsetfiretothebelfrybyignitingthegas”| 1893 ANDLIGHTNING PROTECTORS. a6. From Abstract ofReport ontheDestruction byLightning of« Gunpowder Store atBrunteligje, Yorkshire. ByMajor V.D. Majendie, RA. (LR. p.77.) “The gunpowder exploded at4.30 pan. onAugust 6,1878, during thegreatest intensity ofaviolent thunderstorm. The building was brick, with brick arched roof, length 9fect, width 5feet, height 6feet (internal dimensions). The store hadauniformthickness ofthreebricks,andwasfurnished atone endwith aniron door, atthe other end with alightning conductor. The conductor consisted ofacopper wire rope, 10gauge copper wire, therope being y';inch thick, having four Points atthetop (one large oneinthecentre, and three smaller onesround it); itextended toabout 13feet above thetopofthe building, andabout thesame length was carried intotheground and terminated inadrain. The conductor had been erected in 1876 byMr. John Bisby, ofLeeds, and was fixed toapole distant about 2inches from the end ofthe building opposite tothat inwhich the iron door was fixed (itwas not connected ‘ith theiron door inany way). Noonewasnear thestore when thepowder exploded, and itseems probable that [the earth connection oftheconductor was bad, that] themass ofiron in thedoor offered atleast anequally good path—and that the gunpowder was ignited byaflash passing between the two imperfect, conductors.” Tomake this report more completely scientific, itwould be velltoomit thewords Ihave putinsquare brackets. Extract from areply ofMr,Baldwin Latham, C.E.:— “Itisnouncommon thing forbuildings provided with what arecalled lightning conductors tobedamaged bylightning, and thecauseisduetotheinadequacy oftheconductor tocarrythe electric fluid, which will leave the conductor forabetter ora larger conductor.” (See alsosec.43.) 42.Asillustrating that @good conductor affords noabsolute security, thefollowing document isworthy ofreproduction in fall, asgiven inL.R.C., p.115:— a6 ONLIGHTNING, LIGHTNING, CONDUCTORS, [Apel8, “Usher Blitzubleiter und Blitzechliye imGebsiude welche mit Blitzableitern versehen waren. Von G.Karsten. Keil. 810. 1887, (Abstracted byR.Van derBroek.) “InthispamphletDr.Karstengivesanaccountoftwocases inwhich buildings that were provided with lightning conductors were damaged bylightning. ‘The author states that thestatistics fortheyear 1873 show that inSchleswig-Holstein twenty-six per cent. ofallthecases offirewere caused bylightning;x,part ofthese cases occurred inthe towns and the remainder inthe country. “Do lightning conductors guarantee ‘absolute protection? The author answers this question asfollows :—There is00 absolute certainty inempirical matters;eachnewcasemay direct our attention tocircumstances that had been overlooked. Iflightning conductors cannot besaid toensure perfect safety, they certainly afford avery high degree ofprotection. “The flash oflightning which struck the church atGarding, onthe 18th ofMay, 1877, fractured theconductor infifteen places, andpierced thewall ofthesteeple intwoplaces. The inefficiency oftheconductor resulted from thecareleseness with which itwasfixed; theline waslaid down the north side ofthe steeple and fastened with twenty-five wall eyes; these wall eyes were hammered toodeep into thewall, thus damaging theline and forming ashort and sharp bend ineach case, besides slo ‘unduly straining thewire. The damage tothesteeple wasthe consequence ofaneglected secondary circuit. There are1 excessively large number oftie-rods inthesteeple ;theheads of these rods arenot connected together, neither arethey, except inonecase, inclose proximity toany ofthelarger masses of metal that areabout thebuilding. ‘The conductor passed close toone ofthose heads; the south side ofthesteeple, where the opposite head is,becoming wet through therain, asecondary circuit was formed, and areturn shock followed; the damage to thesteeple wastrifling. “The rodwasprovided with aconical point, rather blunt, bat surmounted by@short platinum point. ‘The copper line-wire 1669,) ANDLIGHTNING PROTECTORS. ar ‘vasofgood material—not ofauniform thickness, butatthe weakest places not weighing less than 240 grammes per lineal metre (8oz.peryard, orrather lessthan }inch diameter ifsolid). ‘Theearth-plate wassunk into awell 10metres deep, and tested faultless after thedischarge.” 43,The following series ofnotes arebetter not separated, though they bear upon different points :— LBC, p.93ets0q.—From Abstract ofStatistice ofBuildinge andSkipsstruckbyLightning. ByF.Duprez,Memberof theAcademy. Muatrating occult dangers from secondary disturbances inthe neighbourhood ofconductors, “<The author cites three cases ofbuildings setonfire though protected bylightning rods. But theprecise cause ofthe fire was not ascertained.” Mlustrating multiple flashes. “In each oftwocases thelightning struck atonce the three rods fixed toabuilding.” Nuatrating that apoor earth ianot necessarily fatal tothe Giciency oftheconductor, exceptastotheprotection ofthe soit itself and ofthings buried init. “Out offifteen cases oflightning rods struck, inwhich theconductors were simply buried more orless inthesoil, they carried offthestrokes ineleven without thebuildings being injured oranytrace being leftofit,except that the ground wasupheaved where thelatter wastoodry.” Mustrating amatter important, iftrue: suggested byProfessor Fitagerald theoretically atBath:notyetverifiedbyme. “In twocases thestroke broke the conductor atpoints where itsdirection wasabruptly changed.” Hlustrating sideflash. “In twoother cases thelightning lefttheconductors struck, andfellupon buildings near, without causing damage tothose onwhich the rods were fixed.” ‘VOL, XVIII. 29 418 ONLIGHTNING, LIGHTNING CONDUCTORS, [April214 ustrating brush discharge from conductor, and surging circuit iteneighbourhood. “Two electrical phenomena aretobenoted assome times occurring when alightning rodisstrack, First, when aconductor isformedofmetallic platesapeculiar noiseis heard likewater pouring onafire. Second (independent!y ofthe form oftheconductor), electric sparks are emitted from bodies near. The author cites examples atBerne, 1815.” 44,The following illustrate thecarrying power offairly thin wire, and the fact that thin wires may protect although them- selvesdeflagrated. FromMr.Preece's paperon“Lightning”to Society ofTelegraph Engineers, November, 1872(L.R.C.,p.101):— “There were only two cases inthe past season where linewires (No, 8iron, 0-17 in,diam.) were absolutely fused.” From letter byAdmiral Sullivan (L.R.0., ps195):—*You will like toknow acase inwhich acopper wire acted asaperfect conductor, though fused throughout itslength, ItwasatMonte ‘Video, inthehouse oftheEnglish Consul;aflag-staffwasstruck andconducted thelightning through aflatroofnear thebellwire, ofasuite ofrooms (the wire ran insight near the comice) through ahole ineach dividing wall, andthen down tothebell inthe basement: the wire was melted into drops like shol, which burot arowofsmall holes inthecarpet ofeach room. A dark mark onthe cornice above showed where the wire had been. Atthebell there was aslight explosion and some little damage, Dut Idonot recollect whether anything acted partially as8 conductor from that point and socarried offthat part ofthe charge. “This, Ithink, shows that even anordinary bellwire willact asaconductor forarather strong stroke oflightning, asthe large flag-staffwasshattered.” 45,Asillustrating that experience hasledtotheperception ofthevalue oflarge surfuce toarod,Imay quote thestandard American work, Spang’s “Practical Treatise onLightning Con- ductors.” Philadelphia, 1887 (L.R.C., p.113):— “Aconductor oflarge surface exercises amuch greater pro- 12893 ANDLIGHTNING PROTECTORS, a9 tective action than the same quality ofmetal inthe form of# wire orsolid rod. “Not because electricity inmotion resides onthesurface, but that theexpansive action ofadischarge may have awider scope through themetal.” [The incoherence ofthereason does notdestroy thecorrect nessofthestated fact.) ‘Messrs. D.Munson &Co., ofIndianapolis, Indiana, have sent. imespecimens oftheir rods, which have flanges andsharp edges, andinvarious ways aim atlarge surface. Their rods arecom- posedofcopperandironmixed,whichiscurious, andtheirmodeofattainment oflargesurfaceisneedlessly complex. ‘Thefollowing illustrates thecontradictory views about manner ofconduction bylightning rods. Letter from Admiral Sullivan (LR.., p.199):— “Ifirmly believe inthesurface theory ofHarris, had been with him often when hemade experiments nearly fifty years since, and witnessed astrip oftinfoil ofthe thinnest kind, and about }inch wide, protect amodel mast ofabout sixinches in diameter from electric shock, that without itsplit the mast to pieces, aided byasmall hole through itscentre filled with gun- powder, And Ialways thought that the surface-conducting theory ofHarris was indisputable. But about 20years since, having toapprove aproposal oftheTrinity House foranew con- ductor ofalighthouse, which, likeprevious ones,wasaninchin diameter copper rodcalled ‘Faraday's Plan,’ Ithought Iwould goup totheRoyal Institution and askhimwhy hedidnotuse &copper tube instead, giving much greater conducting power with less copper. Idid so,and heasserted positively that the conducting power depended entirely onthevolume ofcopper in thesection oftheconductor, nomatter whether itwasinabolt, plates, ortube; and that ifHarris said differently, ‘he knows nothing whatever about it;*ofcourse Iapproved therod-con~ ductor, But singularly enough, though Ihad notseen Harris foryears, hecame totown afewdays after, and came tothe ‘BoardofTradetoseeme,andbringmeapieceofhislargetube emductor, with aconnection that hewas fitting tothe Houses 420 ONLIGHTNING, LIGHTNING CONDUCTORS, [Aprilsa, ofParliament. When Itold him what Faraday’s opinion was,he answered, ‘Then heknows nothing about it.” 46.Imay call attention tosome apparently sound doctrines inapaper byCapt. Bucknill, R.E., abstracted inZ.R.C., p.242. ‘Thatlightning canpenetrate intocollieries isprovedby ‘account ofameeting ofMining Engineers (L.R.C., p.237). ‘Thesamesortofthingisillustrated bytheaccidentatBootham Bar,York(L.R.C,,p.219),whenthelightning struckdowninto ‘cavity surrounded byhigh buildings with lead roofs, ironrain- waterpipe,andanironportcullis,togetatastreetbracket 11ft.6in,above thepavement, melting itspipe, and setting» house onfirebythelarge flame produced (L.R.C., p.219). 47.Thattheexistenceofgaspipesmakehousesmoredifficult toprotect, and infact causes dwelling-houses torequire almost ‘asmuch attention aspowder magazines, canbeillustrated bys good number ofextracts. The following may serve (L.R.C, p.289) :-— “OnJuly13,1880,duringathunderstorm, thelarge400- light gas-meter ofthis mill, though locked upinacellar, and with nolight mear it,exploded, andthegas,which issupplied through a4inch main, wasignited. ‘This wasrepaired, baton July 5,1881, during another thunderstorm, precisely thesame accident occurred.” ‘From aninteresting report, signed J.Gavey, ondamage toschurchatCardiff(L.R.C.,p.218),Imakethefollowing extract:— “On examining more closely thesurroundings ofthelightning conductor, Iobserved thatthechurchgas-pipe, anironone,about 1}inches indiameter, passed through thewall ofthebuilding about 6feet from theconductor, and was carried inadirection corresponding with theholecaused bytheexplosion (seeplan) Iimmediately concluded that this explosion was due tothe current breaking across from theconductor tothegas-pipe,and ‘onopening uptheholeIfoundthistobethefact.Theconductor crossed thegas-pipe atnearly aright angle, being aboutafoot above it. ‘The under portion oftheconductor bore evident marks offusion, and, more interesting still, thegas-pipe wasslightlycoatedwithaverythindepositofcopper,sothinthatitperished 1189.) ANDLIGHTNING PROTECTORS. an inmyattempt toremoveit;butstilltherewasanundoubted coating atonespot.” Continuing theaccount ofthis kind ofdamage, thefollowing interesting remarks byProfessor Kirchhoff have abearing onthe important practical question whether gas-pipes should orshould notbeutilised forearth;buttheirmainutilityliesintheproof affordedoftheuniquekindofdangerwhichgas-pipes introduce :— Injury toGasand Water Pipes byLightning.® Thecitygascompany ofBerlin,havingexpressed thefearthat spipes may beinjured bylightning passing down arodthatis connected with the pipes, Professor Kirchhoffhaspublishedthe following reply :-— “Astheerectionoflightning-rods isolderthanthesystemof gasandwaterpipesastheynowexistinnearlyalllargecities, ‘wefind scarcely anything inearly literature inregard to connecting theearth end oflightning-rods with these metallic Pipes,and inmodern times most manufacturers oflightning-rods, when putting them up,paynoattention topipes inornear the building that istobeprotected.” Kirchhoff isoftheopinion, supported bytheviews ofaseries ofprofessional authorities, that thefrequent recent cases ofinjury from lightning tobuildingsthathadbeenprotected foryearsbytheirrods,areduetoa neglect ofthese large masses ofmetal. The Nicolai Church, in Griefswald, has been frequently strack bylightning, but was protected from injury byitsrods. In1876, however, lightning struck the tower and setitonfire. Afew weeks before, the church had had gas-pipes put init.Noone seems tohave thought that thenew masses ofmetal which had been brought into the church could have any effect onthe course ofthe lightning, otherwise thelightning-rods would have been connected with thegas-pipes, ortheearth connection been prolonged to proximity with thepipe. Asimilar circumstance occurred inthe ‘Nicolai Church inStralsund. ‘Thelightning destroyed therodin ‘Bee alsoonthiswubjecttwoAbxtractsinJournalofInstitutionofBlcrical Bapiners,No.77,Vol.xvii,1889,ofpapersbyProf.W.Koblrausch and A.Voller repeotively. | aa ONLIGHTNING, LIGHTNING CONDUCTORS, [April25%, ‘many places, although itreceived several strokes in1856, and conducted them safely tothe earth. Here, too, the cause of injury wasintheneglect ofthegas-pipes, which were first laid ‘intheneighbourhood ofthechurch in1856, shortly before the lightning struck it. The injury done tothe school-house in Elmshorn, in1876, and tothe St.Lawrence’ Church, atItzehoo, in1877,bothbuildings beingprovided withrods,couldhavebeen avoided iftherods had been connected with theadjacent ga- pipes. “If itwere possible,” says Kirchhoff, “to make theearth connection solarge that theresistance which the electric current meetswithwhenitleavesthemetallicconducting surfaceofthe rodtoenter the moist earth, orearth water, would bezero, thea itwould beunnecessary toconnect the rods with the gasand water pipes. Wearenotable, even atimmense expense, tomake theearth connections solarge astocompete with theconducting powerofmetallicgasandwaterpipes,thetotallengthofwhich isfrequently many miles, and the surface incontact with the moist earth isthousands ofsquare miles. Hence theelectric current prefers foritsdischarge the extensive netofthesystem ofpipes tothat oftheearth connection oftherods, and thisalone isthecause ofthelightning leaving itsownconductor.” [This, likemany another theoretical remark, isnottobesup posed true.—0O. L.] Regarding thefearthat gasandwater pipes could beinjured, hefurther says: “Iknow ofnocasewhere lightning hasdestroyed ‘agasorwater pipe which wasconnected with thelightning-rod, butIdoknow cases already inwhich thepipes were destroyed by lightning because they were not connected with it, InMay, 1809, lightning struck therodonCount vonSeefeld’s castle, and sprang from ittoasmall water-pipe, which wasabout 80metres from theendoftherod,andburst it.Another case happened in Basel, July 9,1849. Inaviolent shower onestroke oflightning followed therodonahouse down into theearth, then jumped from ittoacity water-pipe, ametre distant, made ofcastiron. Itdestroyed several lengths ofpipe, which were packed atthe Joints with pitch and hemp. Athird case, which wasrelated to 169.) ANDLIGHTNING PROTECTORS. as mebyProfessor Helmholtz, occurred lastyear inGratz, Then, too,thelightning lefttherodandsprangovertothecitygas- pipes even agasexplosion issaid tohave resulted. Inallthree ‘asestherodswerenotconnected withthepipes.Iftheyhad been connected themechanical effect oflightning onthemetallic pipes would have been null inthefirst and third cases, and inthe second thedamage would have been slight. Ifthewater-pipes inBasel hadbeen joined with lead instead ofpitch, nomechanical effect could have been produced. The mechanical effect ofan clectrical discharge isgreatest where the electric fluid springs from one body toanother. The wider this jump the more powerful isthemechanical effect. ‘The electrical dischargeofa thunder-clouduponthepointofalightning-rod maymeltorbend it,while the roditself remains uninjured. Iftheconductor, however, isinsuffigient toreceive and carry offthecharge of electricity, itwill leap from the conductor toanother body. Where thelightning leaves theconduetor, itsmechanical effect is againexerted,sothattherodistorn,melted,orbent.So,too, isthat spot ofthebody onwhich itleaps. Intheexamples above given itwasaleadpipeinthefirst.place,agas-pipeinthelast place, towhich the lightning leaped when itleft therod, and which were destroyed. Such injuries towater andgaspipes near lightning-rods must certainly bequite frequent. Itwould be desirable tobring them tolight, soastoobtain proof that itis more advantageous, both fortherods and thebuilding which it, protects, aswell asforthegasand water pipes, tohave both intimately connected. Finally, Iwould mention twocases of lightning striking rods closely united with the gasand water pipes. Thefirst, happened inDiisseldorf, July 23,1878, onthe newArtAcademy;theotherAugust19,lastyear,atSteglitz, Inboth cases thelightning-rod, thebuildings, andthepipes were uninjured.”—Deutechen Bauzeitung. (Quoted inTheBuilding Nace, September 10,1880.) 48.That lightning sometimes does things which isonthe hitherto available views inexplicable, isevidenced bythefollowing record byone who, ofallothers, made lightning conductors his hobby, whoacquired animmense amount ofexperience concern- 6 ON LIGHTNING, LIGHTNING CONDUCTORS, [Apri 2th, ingthem, and towhom thecarrying outofprobably themost perfect system ofprotection inactual useislargely due—I mean the late M. Melsens :— Note ontheLightning Flash atAntwerp Railway Station, July 10th, 1865. ByM.Melsens.* “Ihave described this lightning stroke, which was very harmless, since allthedamage was confined tothebreaking ofasquare ofglass intheroof ofthecovered station ofthe railway atAntwerp; butinconnection with itIfound myself confronted bysoextraordinaryaphenomenonthatIhadtosearch through allthedescriptions oflightning strokes which might offer acertain analogy tothiswith which Ihave todo. Itwas only onlooking through mynotes again, and returning seven! timestotheplace,thatIhaveventured todescribethisvery ‘extraordinary stroke, and Ihave only published itafter aloug time, when itseemed tome that Icould establish itupon careful observations and experiments which seemed ofanature tocorroborate myconclusions. [Mention made of19lightning flashes, which present some analogytothatwhichstruckAntwerp Station.] “A fewwords aresufficient toshow thepeculiarity ofthe phenomenon. ‘The lightning crossed asquare ofglass and produced initahole similar tothat which would beproduced by@projectile thrown upwards from below ataslow rate, my from 30to50metres per second, and travelling from earth towards thesky; theedges ofthehole were melted. Ihave given adescription oftheexperiments Imade, together with Ruhmkorff, toprove that inreality lightning does travel from theearthtowardsthesky;theproofwhichIgiveofitseemsto me decisive. “The extraordinary fact, asIhave remarked, istoseethe lightning pass, bymeans of@very bad conductor, through « square ofglass 4millimetres inthickness, formingaparallelogram ofO*35x0"38,havinganglesof83°and97°, ‘©“Paratonnerres: NotesetCommentaires,” parM.Melsens,Brussel182. ‘Extracted from Reports ofBelgian delegatestotheParisExhibition, 16. 1060 ANDLIGHTNING PROTECTORS. 46 “The opening produced wasatadistance ofsome centimetres from iron and lead conductors, which were inperfect metallic communication with alltheiron ofthestation. The weight of thisIntter exceeds 120tons. But theanomaly does not stop there: totheright andleftoftheglass roofinwhich thebroken panewas,thecoveredplatform ofthestationhasaroofofzine No.13,presenting tothelightning asurface of3,000 square yards; the weight ofthis zine isnotless than 15tons; the three tall rods ofthe lightning conductor areinimmediate metallic contact with this zinc, and with the conductor, and the whole isconnected to28hollow columns, serving tocarry off main-water. The collection ofthese metals would allow one tosuppose that they were well adapted tocarry offlight- ning orany such disturbance easily; but inthis instance it despised them, and chose apath entirely unexpected. Let me add that there was ashed some 62metres from the station built onhollow columns and iron framework, and the greater part of itroofed with zine; moreover, about 40metres onthe opposite fidewere more sheds roofed with zinc; the metallic surface was notless than 2,000 square metres. Moreover, inmyinvestiga- tion, Imight mention plenty ofother metals incommunication more orless perfect with the earth—grids, sconces, gas-pipes, telegraph wires, rails, ete. Besides, thewhole oftheplatform maybeconsidered asonelarge lightning conductor inperfect communication with avery damp earth, and offering hardly any resistance tothepassage ofthecurrent from thebuilding.” 49.Ofallthebuildings intheworld notwholly made of metal, the Hotel deVille atBrussels was and isthe. most perfectly and elaborately protected. Noelectrician exists but Would ayear agohave asserted, hadhegone over it,thatitwas abeurdly and exaggeratediy safe from damage bylightning. astJuly itwasstrack andsetonfire. Thecase hasbeen investigated andpublished intheBulletin delaSocittéBelge@Electriciens forSeptember andOctober, 1888. Itseems tohave been owing tosecondary orinduced eleetrie surgings inahorizontal barofmetal totally disconnected from anything; notpretending tolead toward earth, and, being some a8 ONLIGHTNING, LIGHTNING CONDUCTORS, (Aprist, Aittle distance below astretch oflightning conductor, notoffering ‘itselfasanobjecttobestruck.Onalltheoldviewsitwasutterly insignificant. However, asamatter offact, although notstruck, although itdidnot(probably) even receive aside flash, yetthe induced surgings setupinit,induced byMaxwell and Heavisides ‘electro-magnetic waves, were soviolent astoignite some gasand cause asmall fire. Had itbeen connected tothe conductor, the sparking from itwould probably have been still stronger. Thisoccurrence, andcertainsparkings inthewallpaperofthe Royal Institution (see Electrician orNature, March, 1889)— thesparkings, too,inthegilding ofthispresent hall (sec. 32)— fareplain and straightforward intimations that theoldviews on the subject ofelectric conduction arehopelessly, and absurdly, and dangerously inadequate. Itistime that theprophets of thatoldsuperstition were slaughtered bythebrook Kishon. Practical, Questioys. 50.There remains toconsider what istobethepractical ‘outcome ofallthis. What improvements intheerection sod testing oflightning conductors are possible. This isamatter well worthy ofdiscussion, andeminently suited toit.Itis matter onwhich Ihave nottheslightest wish tobedogmatic; and ifImake afewapparently definite assertions, itisonlyby wayofexpressing such judgment asIhave been able toformat present, and they aretobetaken asintended more bywayof suggestion andquestion than anything else. With this proviso,Ishouldbedisposedtomakesomesuch statements asthefollowing :— 51.Allparts ofalightning conductor, from points toroots, should beofoneand thesame metal, toavoid voltaic action. 52.Joints should beavoided when possible, and should be made substantially when necessary. Allowance forexpansion and contraction must, notbeforgotten. 53.Sharp bends, and corners, and curves, and roundabout paths toearth should beavoided asfaraspossible. 54,Theuseofcopper forlightning conductors isaneedless extravagance, 1899) ANDLIGHTNING: PROTECTORS. 47 55.Iron hasadvantages over every other metal. 56,The shape ofcross-section isbut little matter. Flat ribbon hasaslight advantage over round rod, but not enough to override questions ofconvenience. 57.Liability tobedeflagrated byapowerful flash determines the minimum allowance for size ofcross-section, No considera- tionofconductivity andgreatereaseofpathhastheleastweight inthis connection. 58.Itishopeless topretend tobeable tomake thelightning ‘conductor somuch theeasiest path that allothers areprotected. Allpossiblepathswillsharethedischarge between them,anda lotofapparently impossible ones. 59.Agood and deep earth should ingeneral beprovided, independent ofwater and gasmains. 60.Iftheconductor atanypart ofitscourse goes near water ‘orgasmains, itisbest toconnect ittothem. 61.Ifthe place tobeprotected haswater orgaspipes inside it,theconductor shouldbeconnected totheirmainsunderground. 62,Atallplaces where water and gaspipes come near each ‘ther, and, ingeneral, wherever onemetal ramification approaches another, itisbest toconnect them metallically. 63.The neighbourhood ofsmall-bore fusible gas-pipes, and indoor gas-pipes ingeneral, should beavoided inerecting a lightning conductor. 64.Into powder magazines and such like places nogasor ‘water pipes should bepermitted toenter, unless thewhole build- ingismade ofmetal, and they areelaborately connected toitat ‘thepoint where they enter. 65.Itisnotwise toerect very tallpointed rods above theroof ofabuilding. 66.Anumber ofpoints allalong theridge ofaroofisbetter than only afew. 67.Any part ofabuilding isliable tobestruck, and, tomake quite secure, every prominent part oftheoutside should have a rodrunning along it. 68, Earth-connected aswell asinsulated bodies are liable to spitoffsparks, a8 ONLIGHTNING, LIGHTNING CONDUCTORS, (Apt2: 69.Nocomplete security can beattained unless thewhole building bemetal-lined, floor and all. 70.Inordinary houses itmay bewell totryand insulate the lightning conductor from thewalls soastolessen thechance of side flash tometal stoves andthings inside. 71.Inchimneys itmaybewelltouseinsulatorstoprotectthe bricks from concussion, 72,The cheapest way ofprotecting anordinary house isto ran common galvanised iron telegraph wire upallthecomers, along alltheridges andeaves, andover allthechimneys; taking them down totheearth inseveral places, andateach place bary- ingaload ofcoke, Rain-water spouts andother outside metal,if allwell connected together, may likewise beutilised. 73.Connecting lead roof orother such expanse with lightning conductor isnot anunmixed good, foritvirtually increases thedangerous proximity ofthelightning conductor, andmayinadvertently bring itnear tomany objects which ele might have escaped. 74.One ofthemost difficult things istoknow what to connect and what toavoid. 75.The orthodox rule, “Connect allpieces ofmetal tothe lightning conductor,” requires modification thus :—Connect all pieces ofmetal toeach other and totheearth, butnottothe lightning conductor. 76.Itmay bealways reckoned safe toearth things inde pendently. Itisoften notsafetoconnect them tothelightning conductor: ¢.g., aninside lining ofachimney should bewell earthed, butshould notbeused aslightning conductor norext nected with it. The same with rain-water pipes and gutter, ‘The came also, probably, with lead roofs. 77.Inconnecting pieces ofmetal toeach other, ifthey happentoformanearlyclosedcircuit,thecireuitshouldbe metallically completed. 78.Over thetopoftallchimneys itiswell totake aloopor arch ofthelightning conductor, made ofanystout anddurable metal. 79.Lightning conductors should bealways outside andeasily visible. 1680.) ANDLIGHTNING PROTECTORS. a9 80,Aconductor detached from the building tobeprotected issafer than one inclose contact with it. 81.Forpowder-magazinesand suchlike,anoutercagesurround- ingthebuilding, with skypoints and earth roots, and aninner ‘ageonthebuilding, withindependent earthterminals only,is thesafest plan, 82,Ifunder these circumstances there benogas-pipes nor much ramifying metal work inside thebuilding, and nometal at allgoing near either cage, theinterior may beconsidered per- fectly safe. 83,The inner cage may often beconveniently made ofcon- tinuous sheet iron. The outer cage need not then beatall ‘small-meshed ;infact, itneed belittle more than adozen vertical conductors. 84.Theresistanceofanearthmaybetestedinthecustomary waytoguardagainstactualbreaches ofcontactbyrustorworkmen; batnooverweening confidence must befelt, even though the resistance turnoutthethousandth ofanohmorless. 85.AWimshurst machine andcouple ofLeyden jarsafford a convenient mode oftesting aconductor forflagrant defects, The testing should bedone inthedusk orinmoonlight, sothat there maybelight enough towork byandyetsparks bevisible. 86.Telephones arethe handiest things todetect electric surgings inconductors inside thehouse while dischargesarebeing made totheconductor. But vacuum tubes, gasleaks, Abel's fuses, ete., ete., canalsobeemployed. 87.Another modeoftestingcanbecarriedoutonaninsulated rod,withaninduction coilandspark gapafter themanner ofa great Hertz oscillator. This isprobably themost searching plan, Sparks willthen beprobably obtained from allthegasbrackets, water taps, and picture rails inthehouse. 88,Telegraph stations, and houses supplied with electricity fromoverhead wires, should have anefficient lightning protector attheplace where thewires enter thehouse. 89.When anumber ofhouses are wired uptogether for Vighting, even byunderground wires, itmay bewell todis- connect them bymeans ofintervening lightning protectors on theprinciple offire-proof doors. 430 ELECTIONS, [April250, 90.Acentral lighting station, having atall chimney con- nected toitsboilers and dynamos, should be,byalightning protector, disconnected from the leads which carry the current from itsbecause asmall fraction ofastroke getting intothe, leads might destroy anumber oflamps, especially ifthey were already working atsomething like their fullpower. 91.Telephone arrangements, and any long length ofclose- packed insulated wires (even underground), should beprotected bylightning protectors, else theinsulation isapttobesparked through andspoiled. ‘The Cuamaan: We have hadamost interesting and instruc- tive discourse, and under ordinary circumstances, iftime allowed, itwould beatonce followed byadiscussion, which would also doubtless prove very interesting; but our permitted period this evening hasalready elapsed, and therefore the discussion willbe adjourned until ournext meeting, which willtake place onThun- day, May 9th. | Thavetoannounce thattheCouneil, incompliance with«| memorial, have extended the period during which Associates, transferred from theclass ofStudents, may attend the Students meetings, from one year totwo years after the date of transfer. q Aballot took place, atwhich thefollowing candidates elected :— Foreign Members : Camilio A,Carrizosa. | Theodore Guilleaume. | Associates: James Gemmil. Harry Neville Moody. James Walter Grimshaw. Frederick A,Nixon. Ernest deMérindol Malan. Edward William Snook. Student : « Donald Barton, * Kk ‘Tho meeting then adjourned. “ pacSO = a GURE 3, FIGURE 2. sney may occur inNature, shincase. Theothers are spark atB,theplace wherdy coatings ofjars;spaces position ofthecharging mor,inthesecond case,b2 10partinthedischarge. leak, ispermissible, but aor ABSTRACTS. 2.¥,RERROUN—DIVERGENCE OF ELECTRO-MOTIVE FORCES FROM THERMO-CHEMICAL DATA. (Phitoophicat Mogasine, Vol.27,p.200, 1689.) {thasalreadybeenremarkedbymanyobservers,thattheactualmeanured, EALF, ofseveral voltaic elements isnot the same atthe value calculated from the known reactionsoftheconatituentaof theclement.Suchdeviationsmight explained bythe non-occurrence inthe cell oftheexact reactions anticl<pated,batsomedifferentreactionmayoccurwhichevolvesheattoanextentfaffcient toaccount fortheobserved EMP, Sach action mightbeduetothe coatingofthemetalswithfilmsofsultsorgases,‘Thechiefinterestofthe suistion,however,liesintheprobabilityoftheevolutionoFabsorgtionof seasibeheatinthecell,andthelossorgaininelectric energy. ‘The author has investigated the behaviour ofseveral metals, including aereary, silver, lead, and tin, and has arrived atthe following general conclusions, ‘The primary factor indetermining the E.MLF. ofavoltaic cell isthoativeheatofformationoftheanhydrons satsofthetwometalsemployed.‘The E.M.F. may set upchemical charges ofadifferent direction andcharacter{romthosetobeexpectedfromtheheatofformationofthedisolved,ala, ‘TheE.M.P.maybesupplemented byaportionatleastoftheenergydue tothe wlution ofthe solid salts, ‘Where there islittle oFnoattraction between the water and thesalt, thenegativeheatofsolutionisderivedfromsensibleheat,andisnotsuppliedby thefree energy ofthechemical change. When metale whore salts have purely negative heats ofsolution are‘opposedtometalswhosealtstheycanreplace,theE.ACP,setupisinexcessofthetotal thermal change,HencenocellexistswhichcanfurnishanE.ME.inexcessofthefreeenergy ofthechemical change.‘Somemetalshaveatendencytoformfilmsofsub-saltsontheirsurfaces, ving rise toanunexpected thermo.chemical reaction, and therefore 40an ALP, difering from thecalculated value,Finally,theauthorconcludesthattheE.M.P.ofavoltafecellfurnishes0 more accarate measurement ofthe ‘free energs,” and therefore oftrue chemical affinity, than data derived from calorimetric observatious, a sowrnson (Comptes Rendus, Vol, 108,p.$96,1889.) on en tenarrn tec nam artes Sormaeaihcgenemnsti anmae ea gasinthefirstinstance, Afterthegashasceasedtobeevolved, theplatinum Samim aenntanece se a = ra etre oe awsoaaons 6: Ind pmtghecmetmycme in uses 1=6)therefore HY=constant Te err thepsimet that het vans wo fetes tom tn eupeins ta aa " Vere Trot (pny can rope the fein eg eneeee nara a ne Sree keer aon eerie ci Deng Soe ihe remesh es fatinam pointisdelicately auspended, anidstrongcurrentused,thecontact SNe ee ee ee i Tem ma feth rt ee wd naraaarou (Annalen derPhysik und Chemie, Vol, 36,p.655, 1889.)Apissigepsteepeanynenhesch tm edo le ee pnery iretaalmeaaiewheeikere ieomiokorparroaaneierneeenRE TET er ner TaN Siete me cous peey sung es ny wo ho nS mae Treat etme nt plo rage lant, en ee sree ree re aeCoeiaee Serre Ate renga fmm tte eer ar a eae ese SSSR Stee mos XsoNREEER—HLnOTRO.NOTIVH PORCH OP THLN PAS oF Stronsnab aoPanons (ams Ppa athv tr 8) rscierinnvainon arto ecetae tans nt tp Bip peneregriee-ery niente vrerecombaeed VOL, XVII. 30 4st ABSTRACTS. rents byother physicists have shown that forcapillary phenomens thelimitinreachedatathicknemofabout6010-*mm,forelectromotive phenoment at1to8x10-+mm,,andforopticalphenomena at4x10-*mm,The experimentawerecarriedoutwithfilmsofthesuperozidesofmax azanese ead, and bismuth;themethodbeingtoexposethethinflmsonplat ‘nam platetotheactionofelectrolysis whenopposedtocleanplatinum,andto note the alterationsintheelectro-motive force,‘Theareaoftheplatebeng scearatelyknown,anditsweightbeforeelectrolysis, thelooofweightatco radeitposibletocalculatethethicknessoftheremainingfilm.Thlimi {ng value was taken asthat atwhich the last and most considerable fll ofBMP. was noted; and itwas found tobeformanganese specie 28x 10~+ mm, and forthelead oxide 48% 10-* mm, These values areof tesameorderas,andapproximate to,thosefoundbyOberbeckforzingcadmian, ‘and copper; and tend toabow that groups ofatoms atthis pointdidmot materially differ from molecules. H MERTZ—RAYS OF ELECTRIC FORCE (Annalen derPhysik und Chemis, Yt. 86,p.769, 1889.) ‘The authorhasbeenabletoproduceactaalraysofelectricfore,and9 carry outwith them alltheelementary experiments which arecommonly pe formed with raysoflightandofradiantheat,ia,heasbeenabletoshow ‘hat the raysofelectricityproceedinstraightlines,thattheycanbepolaris, refleted, and refracted, ‘Theapparatususedconsistedoftwovorticalparabolicmirrorsofsheet ine,securedtowoodenstands.Inthefocusofonemirrorwerepace!vertically, ooabove theother, thetwo metallic bodies terminating inknobs,betweenwhichsparkscouldbemadetopassbymeansofasmallindactioncoil,‘Thelengthofthe(woconductors withinthemirrorwasapproximately‘equaltohalfthewave-length whichcorrespondstooseillationsinstraight wires,Inthefocusotthesecondmirrorwasplacedthesecondarycircuit,consistingoftwowires60cm.long,arrangedinthesameverticalstrightline,andwiththeirinnerends5cm.apart,Theseverticalwireswereom-ected bywires passing horizontally through themirror, with asparingarrangement capableofadjustment, andplacedatthebackofthemiro,»hut itcould beobeorved from behind, The whole arrangement mechresembledtwoAmericanbottlejackswiththeirrespectivereflector.‘Thofrstexperimenta weremadewithoutthereGectore, Onstartingthe owlllationsintheprimarycircuit,itwasposibletoobservebymeanso! movablecircularconductorallthephenomenadescribedinpreviouspege™. $4,upto.adistance of2metres under themost favourable conditions The‘ffctwasincreasedbythepresenceofaconducting wallparalleltotheoailla-tions;anditmaybeconcludedthatthewaveintheaircorresponding othe ‘rimaryceillationsasahalfwave-length of0ca, Ifnow theprimary circuit were fod initsposition inthepars rirror,anditseffestonthemovablecircularconductortested,itwasfound ABSTRACTS. 435, {hattheeffetwasnilbindandatthesidesofthemirror,whilstinthefocal‘ttheelectwasnowapparentatadistanceof6to6metres,whichmightvenbeincreasedto9or10metresif@plainconductingwallwereinterposed inthepathoftherays.‘This,howover,wasonlytrueinclowproximitytothewall.Atotherpointsthedirectandthereflectedwaveinterferedwith oveanother,producingnodes,Foursuchnodeswerefound:atthewall,at 33cm,at65cm.,andat96em.distancefromit,‘Thehalfwave-lengthwould AGereorebe$3cm,anditsperiodofoscillation 1-1thousandmillionthofa swcond.Thehalfwave-lengthinwireswas29em. ‘Theauthornextmadeuseofbothparabolicmirrors,andfoundthathe ‘woldobtainaffectsuptowdistanceof16metres,‘Thiswasthelimitofpace 4hisdisposal;bat,aswillbeseen,thedirvotdistanceofthetwomirrorsnightprobablyhavebeenincreased20metres,Adistanceof6to10metres ixhowever,theonebestadaptedtotheexperiments,‘Asheetofzineinterposedbetweenthetwomirrorspreventsanysparkin‘hemcondary cireuit,‘Thesamoeffectisprodacedbyasheetoftinfoilordle,oFbythepresenceofahumanbeingbutwoenthemirrors(eg.,Hertz'stmisant),Insulatorsdonotcutofftherays,whichcanpasthroughawoodendoor}astheauthorremarks,«Itisnotwithoutastonishment thatonesees ““gpurksapparinaideclosedroom.”Metalshectaplacedoneithersideof theraysandparalleltothemhavenoeffectsolongasthodistancebetweenthemisnotleesthantheapertureofthemirror,viz,1metres,Iftheyaz? ‘woughtcloserthanthisthesecondarysparksdecrease,andnallydisappear ‘henthesheetsarehalf«metreapart,Thesparksalaodisappearifthefocal sisofopemirrorberotatedthroughanangleof15°.1theprimarymirroranditcircuitbokeptvertical,whilstthesecondary cdreaitandmirrorareturnedgraduallyintoahorizontalposition,thesparks creasmoreandmote,andfinallydisappearwhenthetwomirrorsare cromod,eventhoughtheybebroaghtclosetoguther. ‘Theauthorconstructed anoctagonalwoodenframe,onwhichhewound {anllelcopperwires3em.apart,Ifthetwomirrors:wereplacedwiththeir fociparallel,andtheframe0disposedbetweenthemthatthewireswereatrightanglestothedirectionofthefoc,itproducedscarcelyanyeffectonthesecondarysparks,I,however,thesroenwereplacedsothatitswireswereFamleltothefoci,therayswereentirelyentof.Thescreenactedasa tourmalinewoulddointhecaseofaplanepolarisedrayoflight,‘Themirrora ‘werethencrowed,andsolongasthewiresontheframewereverticalorhori-zontal,2,paralleltotheoneortheothermirror,nosparksappearedinthesecondary circaitbatassoonastheframewasplacedsothatthewiresmadesnangleof45°withthehorizon,thesecondarysparksreappeared. Evidentlythescreenofwiresrewlvedtheoscillations intotwocomponents, andonly allowedthatonetopasswhichwasperpendicular tothewires.‘Thephe- ‘omenonispreciselyanalogoustothatproducedbyatourmalineplateplaced ‘betweentwocromodNicole, ,‘Withtheprimarycircuitinaverticalposition,theoscillations areinaverticalplane,andareentirelywantinginahorizontal plane;fromthe “ sons sews tr, ity attisi| ree Senrenfe beter eieebeplantea la Sone Seer SS a wn an rae are ak a obtained easily; they disappeared, however, assoon usthe xine plate wut te en onlytookplacewhenthewireswereparallel totheplaneofoscillation. ‘hadbeen heated foranhour inahotairbath atatemperatare of300°C. ‘The conductivity washigher after heating than befere inallcases ofsimple ABSTRACTS, sr etal, bot the alloy showed ©very slight falling-of, ‘Tbe percentage charge isshown inthe following table, where the two columns refer to Aidhrent pigces ofmetal -— % % Aluminiom TAD Magnesium 68S 8B German silver oe OTB eM, Inorder tomake sure thet mechanioal processes, such ashammering, rolling, eta, hadnothing todowith theresult, theexperimenta wore repeated‘ithfourdiscsofcopperelectrolytically deposited, ‘Thepercentage increasesinthe conductivity were asunder -— 3781753488 8.TERESCHIN—APECIFIC INDUCTIVE CAPACITY OF SOME ORGANIC COMPOUNDS, (AonalenderPhyeihundCherie,Vel,$6,p.782,1862) ‘The experiments were made bythe eleotrometrio method, the various ledies, which were very numerous, being wuccessively used toreplage the airasthe dielectric inanelectrometer ofspecial construction, ‘The results obtained with homologous compounds show that the specife inductive capacity diminishes asthe homologous series isascended, which is ‘<pponed toTomaazewaki's observations onthe aromatic series, ‘The specific inductive capacities ofmetameric compounds are different, The differences Tetween the specife inductive capacities ofcorresponding members ofthe formates and acetates, and ofcorreeponding members ofthe formates and ‘enzoates, are approximately equal, Ithas notbeen possible totrace any‘doverelationship betweenspecificinductivecapacityandmolecular weightorother constant. The following tables contain some of the values obtained— L ‘esta. Formate ‘Acetate, Penaate, Methyl BT THT cS s Pro e S Today ee BBO Am WO eT RE 0. Bthylformate eo on Bethylacetate eo8S Bthylpropionate ws BO Behlbutyrate os Ethyl valerate : 48 438 ABSTRACTS, D.GOLDEAMMER—INFLUENCE OF MAGNETISATION ON THE CONDUCTIVITY OF METALS. (AnmalenderPhysikwadChen,Vo.96,p.804,1889.) Apowerfulelectro-magnet wasmadeusooftoproducethemagoeticei, themagnetizationbeingmeasuredbymeansofthecurrentindacodinafxn coil,Theresistances woremeamuredbymeansofaThomsonbridge,Tometalssubmittedtoexperiment worebismuth,nickel,andcobalt.‘Thepapecontainsveryelaboratetablesoftheregulteobtained,fremwhichtheauthdraws the following conclusions:—H isthe magnetisation mearure is 068, units;sntneratio(2 arriatnoratio(52)5-0/()gto, : where¢=-0meansthatthebridgecurrentinthoplatoofmotalwasparal! tothelinesofforce,and¢=00thatitwasperpendicular.Tnthecaoofbismuththeincreaseofreitanceisproportional totquaro ofthestrengthofthefold,bothwhentheplatesareparallelandperpondicalartothedirectionofthelinesofforee,ince,however,intheex fofthismetal,themagnotistion inalsoproportional tothestrengthoftefielditmaybesaldthattheincreaseofresistanceisproportionaltothewuare ofthemagnetiaation, Ttappearsalsothatthegreatestincreaseofrestaace‘ccurs inthedirection perpendicular ofthelines offore, Tnthecaseofnickeltheresistanceincreasesinthedirectionofthelet offorce, and decreasesinthedirectionatrightanglestothemis,ower, always greaterthanunity.Axtheincreaseofresistanceisindependentofbe directionofH,thereforearisfunctionofH*,‘Thecurvescorrerpeatig tothisequation (4¥/r=f(H*))followaremarkable path,viz,,theyareall saymplotic, ie,forvery large H,Arjr approaches «maximam. Cobait behavesverysimilarlytonickel;but»isinthiecareoftenlat ‘thanunity.Themaximumchangealsoooeursmuchearlierthanischece ‘with nickel, se,forlowervaluesofH. X.BACKSTROM—OONDUCTIVITY OFSPEOULAR IRON ORE, AND ‘THERMO-ELEOTRIOITY OF ORYSTALS, (Boibiter, Vt. 18,p.178,1880) ‘Thegeneralrerltaoftheexperimentsdescribedgotoshowthattheeist similarity between electrical and thermal conductivity, though the vale obtained are ofsomewhat diferent order. Itwas found thatthe resistance ol rodsentfromspecularironorewasthesameforalldirectionsintheprincipalplaneofsymmetry, butwastwiceasgreatalongtheprincipalaxis,Thisrato ia,however,diminished byariooftemperature, althoughtherearenoinditions ofany polarietion, ABSTRACTS, 139 1%,NEREL—A REMARKABLE DISINTEGRATION OF COPPER BY ‘AN ELECTRIO CURRENT. (Beitr, Vol 18,p,177,1889.) Tieinatancereferedtoisthatof«Bornsteinglowlampofoldtype,in whichtheflamentwasclippedtothickcopperwiresbymeansofspiralmetalsprings, which were embedded inenamel, The coppor wires between tho‘upperendoftheenamelandthelamentshowedsignsofpeculiardisintegra-tion, ine copper threads, Dlackened bycarbon, which extended from the cracked layer ofsurrounding oxide, Theauthorrecallstheformationofsimilarcopperandsilvergrowthsby theelectrolysis ofsolid sulphides ofcopper and silver, Pousbly the copper vires intheglow lamp contained sulphur. 1% DUBS—THE COUNTER EMLF. OP THE VOLTAIO ARC. (Beibater, Vl.1,p.197,1889.) ‘Twocarbonplateswereplacedsidebysideandonemillimetreapart;they‘wereconnectedtoagalvanoscope, Ifthelowerplataworeexposedtothe actionofablowpipefamewhichcarriedcarbonparticlesoverontotheupper ne, there was produced aweak galvanic current, which was opposed in Ginetion totheblast, Als, ifaplate strongly heated inahydrogen blastfamewerelaidonanother,aweakcurrentflowedfromthecoldtothehot Pate,Withplatesofcopperthecurrentwasstillweaker,andnilwithiron Hates,‘ThisinanalogoustothecounterEALP.ofthevoltaicarc,andthe shorthinksthatthisdepends,atanyrateinpart,onthemechanicalforce the current, AMOy,—PAITSCHES DYNAMO WITH RADIAL-BAR ARMATURE, (Bletrctechniachs Leitchrit, Vol.10,9,184,1880.) ‘Thearrangement ofthemagnetsissimilartothatinfat-ringdynamos, orimmany types ofalternate.current machines ‘The novelty lie inthe armatare;thisiacomposedofradialbarsofiron,which,however,are arrangedatanangletotheradiosofthecircleformedbythoarmature,‘The frontareareconnectedintotwosymmetricalgroups.‘Theendsofthebars, lyingontheperipheryofthearmature,themselvesformthecommutator. ‘Thewholeofthebarsformonecontinuoasclosedcitcalt,andthetwoparallel ‘groupeareconnectedinserie,wothatthebrushesareplaced22}°apartinasizteen-polemachine(Gr):‘Theparticularmethodofcouplingmanybars 4pallows ofagiven E.MLE. being obtained with amuch lower speedof relation than indynamos ofother construction, ‘The bars being insulated froneachotherbyairspaces,thereismuchfreerradiationofheatanda eatercurentdensitymaybesafelyadmittedinthebare 0 ABSTRACTS. ¥, MEESEN—LIGHTNING CONDUCTORS. (Blattrotechnsche Zetechrit, Vol. 10,p.145, 1880.) ‘The author, who has ad much practical experienceofthetestingof lightningconductors, insistsontheabsolutenecessityofthoroughlytexting| them with agalvanometers any superficial inspection iaquite insufficient© discoverfaultyplaces,whichmaybeduetooxidationinvisibletotheee ‘Herecommends thatoneterminalofthetestingapparatusshouldbeconnectto-each rod, and theother terminal totheearth connection. InBerlin, with which city heisbest acquainted, theearth isgenenliy formed bythegas orwater mains, and their surface resistance ispracticallyai ‘Hodoesnotspeakveryhighlyofthemethodoftestingbymenosofale ‘natingcurrents;theconnectionwiresareinsuchcasesgenerallycoiledoo& ‘small dram, and theportion let cotled isliable tointroduce errors, ovisg (0, induction, Inallpossible cases theauthor strongly advocates theintroduction ‘ofamechanical joint between the conductor and the earth connection, Tbe ‘two can then beinterrupted atpleasure, and each independently tested. He ‘has frequently noticed that theextreme point of rodwhich has been strc inmoreorlesbentover,asthoughtheheatofthedischargehadsaficenty Iheatedthemetaltoallowofitsbending.‘Theauthorgivesexactdetailsofaremarkable caseinwhichlightningstruck awater-mill inHolland. From the roof ofthe house thedscharepasseddownoutsidethewal,thenbrokethroughthewallintothehouse,aspreaditselfoveralooking-giass. Fromtheglassthedischargepassedashrt distance down the wall inside, and then broke itsway through thewall ‘cond time tothe outside, completing itepath toearth through ashottt fastened back against thewall. “a LIST OF ARTICLES aeuarixe 70 ELECTRICITY AND MAGNETISM ‘AppearinginsomeoftheprincipalTechnicalJournalsduringtheMonthof‘APRIL, 1869. L-BATTERIES ANDACCUMULATORS. 4%Zaruim—Best Grouping ofCella—Buiblitter, vol. 18,p.280, 1889,L{Bouxcax—Source oftheElectricCurrentinaBattery—Beiblatter, vol.18,233, 1689, 11—DYNAMOS AND MOTORS. 1M.Drraxz—Governing the Speed ofaMotor—C. R,vol. 108, p.64, 1889;Lum,BL,vol.82,p.51,1889.W.C. Rrcusremsxt—Rate ofWorking ofDynamos,—Lum. EI, vol. 92,p.101, 1869. Gratann—Desroziers' DiseDynamo—Et. Zeit,vol.10,p.199,1889. MI.ELECTRO-CHEMISTRY AND ELBCTRO-METALLURGY. Dr.G.Gons—Lous ofVoltaic Energy ofElectrolytes byChemical Union.— Phil,Mag,vol.27,p.358,1889. 4H.Pexiat—Potential Difference atthe Point ofContact ofaMetal and aSalt ofthesameMetal.—C.R.,vol.108,p.667,1889, XN,Purscutxorr—Blectrolytic Polarisation byMetals.—C. R.,vol, 108, p.898, 1689,A.Mixer—Introdaction tothe Study ofElectro-Chemistry.—Lum, EL, vol. 32, p.126, 175, 1889. Dr.T.Enuaxo—Electro-Metallurgy ofAluminium —Et,Zeit,vol.10,p.195,1680. IV.—ELECTEIC LIGHT. B,Cuavaxnss—Electric Lighting ofthe Theatre atGenoa—Lum., El, vol. 2, .1, 1889,G.Rrewsnp—High-epeed Engines—Lam,El,vol.82,p23,1889.P,Wiuxtxe—History ofTransformers.—E1, Zeit., vol. 10,p.201,1889, V.ELECTRIC POWER. G,Ricmsnp—Electric Railways,—Lum, El,vol.82,p.161,1889, ‘You. xvm 31 “a ARTICLES RELATING TOELECTRICITY, Bea. ‘VI._MAGNETISM AND ELECTRO.MAGNETION. 8,Brows—EffectofLightuponMagnetism.—Nat., vol.89,p.572,1889. ProfessorSonveras—Diurnal Variation ofTerrestrial Magnetism.—Net, ol, | .622, 1889. A,Ertcorr—Magnetism ofGason,—Beiblitter, vol. 18,p.240, 1889. ©.Bavz—New ResearchesonMagnetism.—Et, Zeit,vol,10,p.185,1889. ‘VIL_MEASUREMENTS AND MEASURING INSTRUMENTS. Dr,J.W.W.Waononx—Method ofmeasuring Resistances.—Phil Mg, vol., P.822, 1889.@,Cuarzxox—Winding Resistance CoilstobeusedwithAlternate Carrenk—©.R.,vol. 108, p.799, 1889.A,D'Ansonvat— Universal Dead-beat Galvanometer.— Ann,Tal,volI,P.1,1889, P.H,Laprsors—GeneralTheoryofElectro-Dynamometers—Lum, Bt,vl.8% P.64,1889. A.Pastzow—Measurement ofResistance ofWires,—Britter, vol 8,p21, 1869, G.Mrot—Instrument formeasuring Magnetic Fields.—Beiblater, vol.1%P.288, 1889. Bouxs—Bdison's Meter. Zeit, vol. 10,p.209, 1889. IX-STATIC AND ATMOSPHERIC ELECTRICITY. ‘Dr.0,Loncr—Hotation ofthePlane ofPolarisation ofLight bytheDischarge ofa Leyden Jar.—Phit, May, vol. 27,p.889, 1889, ‘J,Laoaxpy—Lightning Dischargers.—Ann, Tel, vol.16,p.4,1889. E,L,Taovvator—Photographs ofDischarges.—Lum. El,vol.82,p.54,158% F,Pasomex—Potential Difference necessary toproduce Discharges inAit ‘Hydrogen, and Carbonic Ackd.—Annalen, vol. 87,p.69,1889. D,Latscutxorr—PhotographsofDischarges—Beiblitter,vol.18,p.244,188% ‘A.Rromt—Discharge ofwvery largo Battery ofJars—Beiblitr, vol1, P.246, 1889. X—TELEGRAPEY AND TELEPHONY. B,Manoanrea—Telephonograpby.—C. R.,vol. 108, p.670, 1889.,Mancapre2—Intensity ofTelephonic Currents.—C. 8,vol.108,pp.735,798,1869, E,Exavmié—Writing Telegraphs—dnn, Tal, vol.16,p.88,1889. B,Massin—Prevention ofInduction ontwo neighbouring Telephonic Wire, —Ann, Tel, vol 16,p.49,1889. ‘T,Duwrisue—Effect ofFaulty Joints onTelephonic Circuits.—dwn. Tel,vol.1, p.68, 1889, ‘Mancaptm—Graphophone and Telephonograph—dan, Tél, vol. 16,p 1860. ARTICLES RELATING TO ELECTRICITY, Ere, as E.Zeracus—Multiple Telephone Switchboards.—Lum. Ht,vol.32,p.18,1889, ,Woxscurxnosrr—Localisation of aFault common totwo Conductors.— Ine.El,vol.$8,p67,1889. A.Paxss—Use ofCommonReturnWireon‘Telephone Circuits—Lum,i,vol. $8,p.109, 1887. K,Wrrare:—Delany's Automatic Telegraph.—Bl. Zeit, vol 10,p.188, 1889. Dr.A.Tontzs—Wheatstone's Automatic Bystem_—BifZeit, vol.10,p.214,1889, J,Kannie-Stockholm Telephone System,—El, Zeit, vol.10,p.222, 1880. x—THEORY. Dr,H,Hxnre—Rays ofElectric Foree,—Phil, Mag., vol. 27,p.289, 1889. Professor G.Firzaxat>—Dimensions ofElectro-magnetic Unita.—Phil. May, vol. 27,P.822, 1889. 0.Heuvistpe—Electro-magnetio Effects due totheMotion ofElectrification through aDialectrie.—Phil. May., vol. 27,p.824, 1889. APormx—Potential Difference ofMetals inContaet—C, R.,vol. 108, p.780, 1889, 4.Bosauann—Actino-lectric Phenomena—C, R,vol. 108, p.788, 1889. —Vascur—Magnetic Rotary Polarisation.—C. R.,vol. 108, p.848, 1889.B.Brewat—Actino.electric Phenomena.—Ann, Ta,vol.16,p.12,1889.4.and P,Comtm—Electric Dilation ofQuartz—Journ. dePhys, vol. 8,p.149, 1889, ¢Roxie and P,Baxr—Induction inaMedium ofVariable Permeability.—Taam.El,,Vol.82,p.15,1889. ©.Dacusnuce—Differenco between the so-called Positive and Negative Elec- trieity,—Lum. El, vol. 8pp. 70,115, 171, 1889 F,Buavx—Electric Currents produced byElastic Deformation.—Annalen, ‘ol. 37,P.97, 1889, ‘ABioot—New Electric Figures.—Beiblattr, vol. 18,p.245, 1889. ‘X1L—VARIOUS APPLIANCES.; 6.Ricuao—Firing Cannon Electrically.—Lum, El, vol. 82,p.60,1889. Dr.Ounix—New Medical Influence Machine.—Lum. El, vol. $2,p.105,1889. NOTICE. 1.The Society's Library isopen tomembers ofallScien- tific Bodies, and (onapplication tothe Secretary) to the Public generally. 2.The Library isopen (except from the I4th August to the 16th September) daily between thehoursof11.0a.m. and 8.0 p.m., except onThursdays, and onSaturdays, when itcloses at2.0 p.m. AnIndex, compiled bythelate Librarian, tothefirst ten volumes ofthe Journal can behad onapplication to the Secretary, ortoMessrs. E.and F..N. Spon, 126, Strand, WC. Price Two Shillings and Sixpence. ov om Institution ofElectrical Engineers, Founded 1871. Incorporated 1888, Vou. XVII 1889. No. 81. The One Hundred andNinety-third Ordinary General Meeting of the Institution washeld attheInstitution ofCivil Engineers, 25,GreatGeorgeStreet,Westminster, onThursday evening, May 9th, 1889—Dr. J.Horxinsoy, F.R.S., Vice-President, in the Chair. The minutes oftheOrdinary General Meeting held onApril 25th were read and confirmed. ‘The names ofnew candidates forelection into the Institution were announced andordered tobesuspended. The Cuairnman: We will now commence the discussion onthe paper byProfessor Oliver Lodge, on“Lightning, Lightning Con- ductors, and Lightning Protectors.” Iseethat Mr, Preece is here, and, asProfessor Oliver Lodge hasdesignated him hischief antagonist, itisperhaps fitting that Ishould callupon him to take the lead inthe discussion. Mr. W.H.Preece: Ientirely disclaim, sir,thetitle, ofbeing ¥r.Preece. considered astheprincipal antagonist ofProfessor Oliver Lodge. ‘There isnobody inthis room who appreciates more highly than I dothe good work that hehasdone, orwho ismore ready atall times tosupport him when headvocates before this Institution the truth. But when hetakes the opposite course, and advo- cates that which Iamgoing toendeavour toprove toyou You. xvmt 32 “6 LIGHTNING, LIGHTNING CONDUCTORS, Ere. [May2. Mc.rrewe. tobenotthetruth, then, iftheterm antagonist isproperly applicable tome, Iamwilling toaccept it. Allantagonists aresupposed tobe,toacertain extent, inimicable toeach other, butthere never hasbeen any ill-feeling between Professor Oliver Lodge and myself, Scientific controversy isimpossible when passion enters. Iamrepresenting to-night acertain school—I amhere theadvocate ofaconference that satsome years ago, andismued ‘report, andIam thedefender ofaclass ofmen who Profesor Oliver Lodge hascalled the“older electricians.” Itisessential attheoutset that weshould clearly understand whatarethedifferences betweenthesetwoschools;broadly,they may besaid tobetheprinciples enunciated byexperience,and those deduced from pure theory. Professor Oliver Lodge bis announced himself tobethe advocate oftheory, andIpo nounce myself tobetheadvocate ofexperience. Well, Professt Oliver Lodge carefully distinguished theother night whatbe meant about theory. Indescribing experiments byKirchbef, hemade this remark: “This, like many another theoretic remark, isnottobesupposed true.” That isexactly thepoint that Iwant tourge—that whatever istheoretical must notat ‘once beassumed tobetrue. Ontheother hand, experience is theresult ofobservation, butitdoes not.necessarily follow that ‘ourdeductions from experience arealways exact. Professor Oliver Lodge hasdivided hissubject. intotwoparts, which hecalls case aand case b. ‘With respect tocase a,there isnodifference between ts whatever. Case aisexactly thecase that wastaken upbythe Lightning Rod Conference;caseaisexactlythesamecasetht Thave always taken upwhenever thesubject hasbeen brought before this Institution ;andcase@is,asIamgoingtotry'0 prove toyou, thecase ofcommon sense. There isjustof question that Professor Oliver Lodge hasraised, andwhichbe takes alittle credit tohimself forhaving shown, andthatis thatironistheproper material forlightning conductors, andis better than copper. Well, Ihave always maintained that iroaisfasgood ascopper, from aneconomical point ofview,asd 1860) DISCUSSION. “7 fromapurely£s,d.pointofview,Isaythatiron,undersomeMrPreece cireumstances, may bebetter than copper. Intheyear 1872 Iread apaper before the Society ofTelegraph-Engineers, in which Ipointed this out inthe following words: “Hitherto, “inaccordance withthedirections ofSirWilliamSnowHarris, “in hisapplication toships, heavy copper rods, ropes, orplates “of great expense hadbeen used, butIfind that anordinary “galvanised iron wire known asNo.4,which isonequarter of “aninch indiameter, and isused somuch fortelegraphic pur- «poses,isamplysufficientforanydwelling-house.” Asamatter offact,there may atthepresent moment betenmillion lightning conductors inthis world—I donot know the number, that isa mere guess,—but Idoknow, ofmy own experience, ofhalf a million, InAmerica they have used lightning conductors since thedays ofFranklin, and invariably they useiron; and Iam perfectly certain that ifyou were totake the probable ten million conductors that exist inthis world you would find 9,500,000 tobeofiron, therefore itisuseless orneedless to contest the superiority ofiron over copper. Just before Ibranch offfrom this part ofthe subject Imay state that Ihave inmyhand thereport oftheLightning Rod Conference, and there they discuss most judiciously and care- fally therelative merits ofiron and copper, and theonly reason they give infavour ofcopperasagainstironis,thatinthelong rancopper ismore durable than iron, andprobably cheaper. Ithasbeen sometimes saidthat theLightning Rod Conference knew nothing about self-induction, anddidnottreat thequestion asthough self-induetion entered into thematter atall. Itisan absurd notion. Inthis very paragraph that Iread itissaid most distinetly, “Wewillassume theconductivity ofequal lengths “and weights ofirontobeinthecase ofsteady currents ....” giving therelative resistances ofthetwo; and then, inanother part weread, “the suddenness oflightning discharge which “modifies theconductivity when dealing with iron,” and thefact that such sudden discharging ofelectricity affected thecon- ductivity was aswell known inthe year 1881 asIthink itisin tthepresent day. SoIsay,onthequestion ofcase a,there is 48 LIGHTNING, LIGHTNING CONDUCTORS, Ere. (ay e.twewe. nodifference between Professor Oliver Lodge and themembers ofthe Lightning Rod Conference. But when wecome tocase 8,wecome toavery different cae indeed; andhere wehave aninstance, notonly ofhowelectricity | Dehaves itself inthe way which Professor Oliver Lodge las defined asan“impulsive rush,” butwealso find asplendid illo tration ofthewayinwhich anelectrician candeal with quests ofthis kind inanimpulsive rush. Idonot want toapply one single idea against thebeauty ofthe experiments that. Professr Lodge hasbrought before us;Idonotwant toemploy onesingle word toqualify theskill and theability with which hehasbandled thequestion, ‘ThereisnobodywhoknowsbetterthanIdohow earnestly hehasworked atthis question fornow nearly twelve months, andIhave watched hisprogress throughout thewhole of this period. But what Idourge isthis—andIwanttopointit outstrongly toyou—that there isavery vast difference between the conduct ofexperiments and the deduction that istobe Grawn from the results ofthose experiments. Idonotthink any ofuscould have appreciated these new facts that have been brought before usifithad notbeen forthegreat advance that hasbeen made intheconstruction offrictional machines bs ‘Mr.Wimshurst. Foryears andyears Ihadasplendid influence machine that was made byCromwell Varley, but nobody cm Velieve the difference between itand the Wimshurst machie that gave those magnificent effects wesawatourlast meeting. ‘Tho firstpoint Iwant tourge upon youisthis, that there isno similarity intheaction ofaWimshurst machine intheballof theInstitution ofCivil Engineers and theproduction ofthose effects ofatmospheric electricity intheatmosphere which result inlightning. This case bofProfessor Oliver Lodge isentinls dependent upon thesudden, magnificent, rapid effects produced byaWimshurst machine; butinnature, intheproduction of atmospheric electricity, everything isbycomparison gradual, stately, and solemn. We donot know how thunder clouds are excited when they cause these brilliant discharges ofelectricity. Itisaquestion whether there isanoperation going oninthe interior ofthese clouds generating those charges ofpositive and ana) DISCUSSION. 40 negative electricity that result inlightning flashes. WedonotMePuce.knowwhether theremaynotbesomeremarkable actioninthe atmosphere itself, and that thecloud, asitpasses through the atmosphere, simply actsthepart ofacollector. Wedonotknow this; butwhat wedoknow is,that throughout allthe effects of atmospheric electrification, they arenot sudden, just the reverse —they arequiet andslow. Professor Oliver Lodge has endeavoured toillustrate his case J)bythe diagram (Fig. 2) where there aretwo clouds that have been charged, the one positive and the other negative, and there isbelow this cloud another cloud inastate ofneutrality orquietude, Now, bysome unexplained action thedischarge between those twoclouds that produces neutrality iissupposed alsotoinduce acharge from thatcloud totheearth, through the unfortunate church shown inthe diagram. How thathappens wedonotknow. Inthecase oftheexperiments shown you bytheWimshurst machine, there wasatinplate ofabout twoorthree square feet, inarea, placed atashort distance above another plate, and the effects were thus produced;therewasnodoubtaboutthem,we allsawthem, andwecould alljudge forourselves oftheir sudden- ess andoftheir effects. But there atinplate was dealt with which, forthepurpose, wasaperfect conductor;acloudisavery imperfect conductor, and when Isaythat itisavery imperfect conductor Isay50,ontheauthority ofProfessor Oliver Lodge himself. [Mr. Preece here obtained the assent ofthe meeting to <ontinue hisremarks beyond theusual time allowed.] Now Aisacloud, and Cisalightning conductor. What I sayis,that alltheeffects that Professor Oliver Lodge showed us theother night areeffects that would happen ifyou were to takealightning conductor, break itintotwoat¢,andinsert two tinplates ofabout three orfourfeetarea, andthen between those . ‘woplates, ifaflash passed, youwould have theeffects that he showed usinthat space; butinnature itself itisabsolutely impossible toreproduce such aspace between cloud andearth or between cloud andcloud. ‘That which regulates thedischarge of 48 “LIGHTNING, LIGHTNING CONDUCTORS, Ere. (lay, rece. atrue lightning flash isthefield ofelectric force that exists between this point and that cloud; that which regulates the spark that flows across these two plates isresistance. Professor Oliver Lodge’s space isresistance—resistance oftheairbetween thetwo;andalltheeffectsweseetherearesimplytobeattr buted tothe difference ofthe behaviour ofafield and the difference ofthebehaviour oftheresistance ofasmallcolumnof air. Idonot speak without the book here. Experiments in A i another form were done very nearly fifty years agobyMr.Andrew Crosse. Ido notknow whether anybody inthis room hasreat the“Memorials ofAndrew Crosse :”there isnomore interesting electrical book published. Heestablished lightning conductors amile away, upon hisestate inSomersetshire, and ledthem to hisroom; hecharged jars, heformed plates, heperformed great many ofthose experiments, and hesucceeded inrepro- ducing many ofthethings wesawhere theother night, although hewasnotable tounderstand them, andhecertainly could not have explained them toanaudience inthecharming wayin which myantagonist explained them theother night. Well, gentlemen, asregards case b,Iwant topoint outthat itisacase offield versus resistance, itisacase ofsuddenness versus time, 189) DISCUSSION. 41 itisacaseofan“impulsive rush,”asProfessorOliverLodgeur.rece hascalled jit,against steady uniform growth; and what I asert most strongly and most positively isthis, that those experiments that Professor Oliver Lodge showed usthe other night have absolutely nothing whatever todowith lightning, and anydeduction from them issimply erroneous. Asregards case a,then, Iagree; but asregards case b,I utterly andabsolutely disagree. Tnow take thenext part, that oftheoscillatory character ofadischarge. Ihave been certainly amused tosee,especially intheAmerican press, statements that Ientirely distrusted or objected tothe fact that the discharge ofaLeyden jarwas oscillatory. Inever didanything ofthekind. There isno doubt whatever that under certain circumstances the discharge ofaLeydenjarmustbeoscillatory, anditwasshowntobe %,a8Professor Lodge hashimself pointed outatthe Royal Institution, and byone ofmyoldmasters, Professor Henry ofNew York. Itwasproved tobesobyProfessor Helmholtz, orrather hethrew itoutasasuggestion in1847, and itwas proved that itmust besounder certain circumstances, by SirWilliam Thomson in1853. Now, what makes mesayit mustbeso?WhyIdosoisthis,thatifinthepathofthe discharge ofaLeyden jaryou insert sensible self-induction, then there must bethis remarkable action that results inthepro- duction ofanopposing electromotive force that tends to excessively rapid vibrations ofcharges between thetwo plates. Infact,Professor Lodge himself hasgone sofarastocalculate what takes place, andhehasdone great deal more than caleu- latehere,forhehasshowntheeffectbyexperiment; andthose‘whowerepresentatthecharming lecturehegaveattheRoyalInstitution willremember well howhemade itconvincing toall ofus,thatbyinserting self-induction inthecircuit.ofdischargeofaLeyden jar,oscillation wasrendered evident toourears, for, ‘asheincreased theamount ofself-induction present, helowered thenote,fallingfromahighnotetoalownote,Isaythat‘hile itispossible tohave thisoscillation inthecaseofaLeyden ‘tr,itisimpossible inthecase oflightning. Itook upthis 452 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[Mayth, NePrest. point very strongly atBath when wediscussed this question before, and Igave various reasons then why Idid consider it possible that lightning could beoscillatory. Alightning flashis acrack between theinner and outer coatings ofaLeyden jar, where the conditions that determine self-induction do not exis ‘The current must beunidirectional. Needles were magnetised bylightning, asthey areunder certain circumstances byLeyden jars. ‘That wasthe weakest ofallmyarguments, and itisthe onethat hasbeen most seized upon;butIthinkthatProfesor Ewing hasgiven very good reasons why itmay notbestrilly true, and how, under certain circumstances, needles maybelet magnetised even with anoscillatory discharge. But wewillwaive thequestion ofneedles; wewilltake another. Wewilltakethe question ofmagnetising iron wire. Iagree absolutely with Professor Lodge inthis, that ifthere beself-induction present, there must beoscillation; and he,Iamquite sure, will agree with moifIsaythat ifthere benooscillation there isn0 self-induction. ‘Well, now, lightning can speak foritself; itdoes notrequire Professor Lodge, who hasimbued flashes ofelectricity with acer tainsentient power ofselecting their proper path, tospeak forit. Indays gone byweused towork thetelegraphic system ofthis country with abeautiful system, called Bain’s system, which was ‘one bywhich the dots and dashes were recorded onchemically ‘moistened paper inblue dots andblue lines, anditwasnecessary that apositive current should pass through this moistened papet inorder that electrolysis might take place, and soproduce the chemical changes thatresulted really intheformation ofPrussian blue. The character ofthe currents, whether constant or variable, whether positive ornegative, whether uniform ot vibratory, arerecorded withclear,unmistakeable signsThere are men inthis room—my friends Mr. Spagnoletti and Mr. Graves, and nodoubt many others—who remember the |Bain'sinstrument well,andwhoknowthatthereneverwas8|lightning storm passing overthecountry thatdidnotrecord ited | inthese bluecharacters onpaper—long lines, letters, andmarks. | Now,ifalightningflashwasoscillatoryinitscharacter,ifitwas| 160.) DISCUSSION. 439 ‘asuccession ofpositiveandnegative currentsfollowing eachotherMr.Pree. attherate ofmillions asecond, would ithave been possible to makeauniformlineonthepaper? Butmorethanthat,Ihave seenalightning conductor, where itentered theground, that was marked with acopper ofabeautiful bright colour, indicating that, that thecharge, whatever itwas, that passed through itmust have acted electrolytically upon thecopper and leftadeposit of pure copper behind. Every submarine cable throughout the ‘orld, ofanylength, isworked bytheThomson recording syphon, andthere isnever astorm, even though ittakes place ontheline ofthe cable athousand miles from the shores ofthe Atlantic at Valenti, thai isnotshown ontherecorder by“kicks” andmarks indicating theproduction ofacurrent. Would there besuch kicks andmarks, however, ifthose lightning flashes andcurrents were oscillatory? Avery careful observer, Mr.Burke ofNevin, towhom Iamvery deeply indebted fortheexperiment, andwho made itentirely “off hisownbat,” only lastSunday, wasincharge ofone ofour relay stations inNorth Wales, and, without aword fromme,heallowed, during thepassage ofasevere thunderstorm, hisWheatstone receiver torun, Now the Wheatstone receiver makes itgmarks through theaction ofapolarised electro-magnet, anditisquite impossible that any such magnet having high resistance and great self-induction could give any indication whatever ifthecurrents discharged from these lightning flashes were intheleast degree oscillatory intheir character. Ewing's explanation ofthemagnetised needle does notapply here. We aredealing with unidirectional currents ofgreat and uniform strength andofsensible duration. Now Mr.Burke allowed his receiver torunforhalf-an-hour, during the whole time that the ‘torm passed. Lightning flashes passed atanaverage ofabout, every 30seconds, and every lightning flash that passed over Nevin, within therange ofthe Snowdon mountains, recorded itsmark inclear characters, sothat “he who runs may read,” andIwill defy anyman inthis room tosaythat there isthe slightest indication onthat paper ofany oscillation inthose ‘ightning flashes. Onemore very important illustration isthis: ifthedischarge 456 LIGHTNING, LIGHTNING CONDUCTORS, Ere,(May9,| tetes.ofalightning flashbeoscillatory, iftherebeself-induction in| thelightning conductor orthewire taking thedischarge, Pro- fessor Oliver Lodge hastold us,andweallknow, that under such circumstances thecharge willnotenter thewire. What happens, then, wheneveralightningconductorisstruck?Whathappens whenever atelegraph wire isstruck? Itisfused. Ttisfused sobeautifully that you cansee—I amvery sorry Ihave not specimens here, Ihave one forthe purpose ofillustrating another point; but these fusings, these broken wires, have been foryears and years past socommon, that they fare not curiosities, and we donot care very much sboat them; but invariably when awire, and especially aniron wire that issupposed to‘be sofree from oscillatory currents inits interior,whenever suchawireisfusedbylightning, thereare distinct indications that the fusion commenced inthe middle, and not onthe exterior. Another point isthis. Ifthere isone case where self induction would resist the passage ofanoscillatory current through it,itisinthe case ofelectro-magnetic apparates Relays, Morse recorders, electro-magnets, bells ofallkinds have very high self-induction; theself-induction ofaSiemens rela, forinstance, isvery high. Iforget theexact figure, butitis,ia thelanguage ofProfessors Ayrton and Perry, fiveorsixsecohms. Well, now, ifitwere possible foraSiemens relay tospeak to anoscillating fashoflightning, itwouldsay,“Youmustnot come this way, but must gosome other way; donotcome “‘near me.” Butwhat istheresult iftheapparatus isstruck? Itisfused, notoutside, butintheinterior; andwhen youfind arelay oranelectro-magnet fused bylightning youwillfind clear indications intheinside ofitthat theelectricityhasgone through and through it,giving unmistakable evidence thst lightning isnotoscillatory initscharacter. Ithink Ihave given you sufficient evidence toshow that experience onthis particular point isagainst. Professor Oliver Lodge. Iwill take thenext point, and that istheefficiency of lightning protectors. There isnoproverb inmy experience 1589.) DISCUSSION. 438 thatissotrue asthis, that “nothing succeeds likesuccess ;”andat.Pree ifsuecess istobetaken asacriterion ofthe efficiency of lightning conductors, then lightning conductors act.according to thetheory oftheLightning Rod Conference and against the theory ofProfessor Oliver Lodge. ‘Take HerMajesty's Navy. ‘There areships inHer Majesty's navy that have been forfifty years fitted upaccording tothe principle ofSirW.Snow Harris. Since 1872 there has not been amaccident recorded inthe Admiralty, although there arealways 500 or600 ships under commission inthe navy, sailing inevery sea, subject toevery climate, atalltimes and inallseasons. Take the War Department. Wehave two admirable repre- sentatives ofthat department present to-night, and inthat department they have applied this principle totheir magazines allover theworld; and, although Idonotknow what they are going tosay, Iamperfectly certain they willsaythis—that the system inuseissufficiently successful, and represents thedoc- trines that have been promulgated here. Take the churches. Does anybody know how many churches there are inthis country? InEngland alone there are 13,500 churches pro- tected bylightning conductors, and ifone isstruck itis announced inallthepapers and creates agreat fuss; but itis oolyonein13,500, ‘Take lighthouses. Myfriend SirJames Douglass, who has allthelighthouses under hischarge, ishere, Idonotknow what heisgoing totellyou, butIfeelconvinced that there has never been alighthouse struck bylightning since they came under hischarge. Take telegraphy. Here isMr. Graves, who canspeakofoverhalfamillionofpolesthroughout thewholeofthiscountry; they areallprotected with lightning protectors, mostofwhich were notputthere originally forlightning con- ductors, butreally were found tobeso,and similar ones have consequently been ofInte years put upspecially toprotect our oles from lightning. Take apparatus. Onthis point Iwant toput myself at direct issue with Professor Oliver Lodge, whosays, “The ordinary 40 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[layXt, Me.Preece. and well-known form oflightning protecting arrangement is “to attach apair ofplates, oradouble setofpoints, orayair “ofpoints inavacuum orsome very small airspace, as shunt “to theinstrument orcoil ofwire tobeprotected. Here atr “arrangements ofthekind” there were theother night. Now hesays this—mark thewords please, gentlemen: “Now itis perfectly easy toseethat the protection such things afd “isofthemost utterly imperfect kind.” Well, there isagol deal ofcharming coolness about that statement, because Ican speak from authority and experience, and Ican give him some figures, andhecannot give meany. In1872, when wediscussed lightning conductors here, Ipointed out that inthe half-year from January toJuly, 1872, there were 9-46 percent. ofthe instruments inthiscountrydamaged bylightning. ‘Thatwa only forthe half-year, and that was not the worst part ofthe half-year, which isgenerally the autumn, and ifwedouble that itwould show that in1872 just 19per cent. ofour instruments were damaged bylightning. Ibelieve that woul bequite trae, What istheresult inthepresent day. ‘The factis—and thisisoneofthegreat benefits ofbringing apaper ofthis character tothis Institution—it often really teaches those who write papers agreat deal more than itteaches those who listen tothem;andIlearnedfromwritingthatpaperin1872 what serious damage hadbeen done toourinstruments, and atonce proceeded tourge lightning protectors and toapply them. Well, now, inthe twelve months ending March 3ls, 1889, instead ofthe damage toinstruments being 19percent, wehave applied this “utterly imperfect apparatus,” and have brought, the damage down to13per cent; sothat atthe present moment our apparatus isonly damaged tothetue of13percent. My own impression isthat even that 1% percent. isdue tothefact that wehave notapplied lightning conductors toallthe instruments, and that the instruments damaged were notprotected. Iassert this decisively before you, that there isnomore perfect instrument inthisworld than thislightning protector thatProfessor Oliver Lodge reck- lessly asserts isofthe“most utterly imperfect kind.” 1489] DISCUSSION. 45 Now, Professor Oliver Lodge waskind enough tocriticise some Xr.Pree. oftheconclusions towhich theLightning Rod Conference came, andexpressed intheir report. Critics, especially ifthey are inclined tobehostile—mind you,Idonotusethetermhostilein 4bad-bumoured sense;IknowProfessorOliverLodgemeans tocriticise good humouredly our proceedings;—but when a critic takes upapaper orabook tocriticise ithostilely, he picks allthe holes inithecan, and you can generally judge from the number and character ofthe holes the value ofthe criticism. Professor Oliver Lodge has picked three desperate holes inthis report, and what arethey? ‘The first isthat the Lightning Rod Conference hadtheimpudence tosaythat there isno authentic case onrecord where aproperly constructed con- ductor failed todoitsduty. Well, there never was atruer statement made; and when wesaya“properly constructed con- “ductor,” wemean aconductor constructed inaccordance with therules oftheLightning Rod Conference, and donotmean oneconstructed inaccordance withtherulesofProfessor Oliver Inige orany other person, But weassert, and re-assert, there isnotasignatory tothat Lightning Rod Conference Report ‘howill not stand upinthis room and assert, asIdo,distinctly anddecisively, that there isnoease onrecord where aproperly constructed conductor hasfailed todoitsduty. ‘Thenastothesecondpoint,onwhichIcannotspeakfrom experience, forIhave notyethad theopportunity. Ido notsay Iwould not doitifThad the chance, but Ihave not. “A man “may with perfect impunity clasp acopper rod aninch in “diameter, the bottom ofwhich iswell «connected with moist “earth, while thetopofitreceives aviolent flash oflightning.” Gentlemen, Iwouldnothesitateforonemoment todoit.If youwere togive measolid copper rod, one inch indiameter, fixed with anygood earth, and put mewithin therange of« good thunderstorm, Iwould stand there forthegood ofscience andforthehonour oftheInstitution ofElectrical Engineers! Then No.3isthis,that“ifalltheseconditions befulfilled; “ifthepoint behigh enough tobeinthehighest position of “abuilding, nomatter from what direction thestorm may come; 458 LIGHTNING, LIGHTNING CONDUCTORS, Ere. (May, Me.Preece, «if itbeofample dimensions and inthorough, perfect elec- “trical connection with the earth, the edifice with allitspoints «willbesafe, and theconductor might even besurrounded by “gunpowder inthe heaviest storm without risk ordanger” IfTam prepared tostand with myhands clasped round acopper rod,Ithink Ishould beequally prepared tositupon abarrel of gunpowder without fear ofpersonal results. ‘Thosearethethreepoints;andwhatistheconclusion thet my friend comes to? “To adhere,” says he, “to such a «view asthis now, with tenacity sufficient tocause ittobepro «mulgated asanauthoritative scientific statement, would be,in “myopinion, little less than criminal.” Well, like some Irish patriots,Iamcriminal,andeverysignatorytothatLightning Rod Conference Report iscriminal; and why? Because he believes inhimself, andnotinProfessor Oliver Lodge. Iwill not say much about the accidents, but ifyou read the records which hehas made, and will compare them with theprinciples oftheLightning Rod Conference, then youwill find that theaccidents may bejust aseasily explained onthe principles oftheLightning Rod Conference asthey canupoa those ofProfessor Oliver Lodge. Ishould like tohave criticised oneortwo more points,bat time fliessoquickly andIhave already overtaxed your patience. Iwill simply add that this language that Ihave referred to inthe Lightning Rod Conference Report, islanguage mt addressed tothe Institution ofElectrical Engineers, itisnot ‘addressed tothePhysicalSocietyortotheBritishAssociation, bat itisaddressed toallthehouseholders ofthis country, topeople who build houses and topeople who live inthem, people sho have interest inschools, halls—vicars, churchwardens, andpeople whoareinterested inthesafetyoftheirstructures; andwhyshoal ‘wegoand talk tothem ofthe“value ofI,”oroftheoscillatory charneter ofalightning discharge, orofthe views ofProfessor Oliver Lodge? We like tospeak tothem inplain anguage that they canunderstand, and they can perfectly understand such language asweused; infact, thewhole tenor ofthe report oftheLightning Rod Conference istoinstil inthepublic 1082} DISCUSSION. 489 mind anidea ofsafety and confidence inlightning protectors, Mr.Preece. What does Professor Oliver Lodge do? He labels every lightning protector with agreat notice, “Beware ofthis; it “bites, itfizzes, itspits, itdoes allsorts and kinds ofdangerous “things;donottrustit,itisofnouse.Sleepinyourhouse “incomfort, and donotmind those Lightning Rod Conference “people.” Ishould just like tomake one more reference, Professor Oliver Lodge hasdrawn agreat distinction between what he called the old and the new school of electricians. As I mentioned just now, Iwas brought upunder the influence of Faraday, Snow Harris, Wheatstone, and oftheoldelectricians. ‘There arevery great differences between the two schools;I cannot refer toallofthem. There is,however, one great difference, and itisthat the oldelectricians were tolerant— they were tolerant oftheir opponents, they were tolerant of ‘opposite views, they listened tothe arguments ofallsides. Butwhat ofthe present, themew school? ‘The new school, inthe language ofProfessor Oliver Lodge, says ofthe old school, that “itistime that theprophets ofthat oldsuper- “stition were slaughtered bythe brook Kishon.” We donot Raliate. ‘ThepoetCowpersaid: “+0for loge insome vast wilderness, Some boundless contiguity ofshade, ‘Where rumour ofoppression and deceit Otuneucceesal and raccesfal war, Might never reach me more!" ‘My greatest wish isthat Professor Oliver Lodge may find himself insome such desert, subjected tolightning flashes, surrounded bycoils ofwire and armed with apparatus, and I am{sure hewill speedily come tothe conclusion that itis better tocome back tothe arena ofthe Institution ofCivil Engineers insack-cloth and ashes, pronounce that hewas wrong, join the oldschool, and crypeccavi, ‘Mr.Cartes W.VINCENT: May Isayadozen words asanote ¥*Vises totheaddress which Mr. Preece has given, especially with reference totheparagraph respecting “oscillation ?”Many more 400 LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May92, Me.Vinoent. years ago than Iamquite willing now torecollect, Iassisted Professor Faraday inpreparing anumber oflecture experiments with the electric machine ofthe Royal Institution (one ofthe largest inthe country atthat time) and the Leyden batteryof fifteen large jars. The experiments made were ofthedischarge ofthis battery through various substances, wires ofdifferest thicknesses, and other conductors; but inno case was there any sign whatever of“oscillation” inthesense that thecurrent ‘wassometimes inone direction and sometimes inthe opposite. “Oscillation” inthe sense ofretardation ofthe battery dix charge, bythe substitution ofanumber ofsuccessive impultes forasinglegreatone,hehadinplenty. Asiswellknown,such “oscillations” aretheordinary means ofdischarging anybody which has electricity induced init, Imperfect conductors,cot- ductors offering resistance, converted thecharge ofthepowerfil battery into afeeble spark discharge, or,rather, continuots current, which deflected agalvanometer, precipitated iodine from iodide ofpotassium, decomposed various salts, the current from theLeyden jarbeing proved tobeatrue current. There wi no indication ofoscillation inthe sense ofreflex action, Hai itexisted, Faraday, with hiskeen perception, would have imme diately perceived the“back current.” Electricians, unless ther know, arenotvery ready torush into afray, butIimagine the majority ofthem donotaccept theidea ofoscillation inthe discharge ofaLeyden jar, ifbythat ismeant acurrent alteraste indirection, Maan ‘MrsJ.Farquitarsox: Gentlemen, Ishould liketosayaferwn" eords, notsomuch astouching onProfessor Lodge's paper,35 onthe remarks ofMr. Preece; having had considerable experience inships’ lightning conductors during aperiod of twenty years attheAdmiralty, during which itwas part ofmy duty totest and examine when accidents occurred. The system adopted bythe Admiralty isthat introduced bySirW.Snow Harris. Thetopmasts and topgallant masts being portable, the automatic connection ofthe conductor attheir junctions i# liable togetoutoforder. Generally speaking, when ships have been struck, Ihave found little difficulty inseeing howthe 198.) DISCUSSION. aot damage may have occurred, onthe assumption that the con- a. ductor was atthe time imperfect; but several cases havet™ occurred inwhich Iamquite unable tounderstand orexplain what actually happened. Asanillustration ofmydifficulty, take atypical case—H.M. gunboat “Coquette” (Ibelieve Mr.Preece hastheparticulars, asreported bymesome sixteen years ago). The main topmast wasstruck anddamaged, although fitted with anample conductor, which was uninjured, The eonductor onthe main lower mast was also uninjured, but the portion leading from themast totheship's side was torn from thebeam although ithad been well secured by1}-inch nails, buttheconductor itself wasuninjured and showed nosigns of heating. The wood hammock-berthing next the mainmast shrouds wastorn topieces foradistance offrom 15to20feet. Onexamination afault inthe main conductor on the lower mast msfound toexist afew feet below the masthead, where the conductor had been cutand imperfectly joined. The resistance belowthispoint,including theportiontornfromthebeam,was found tobenilbyabridge, only going toyyohm. What I failtounderstand iswhy thedamage was done below thefault, where theconductor wasample. Assuming that theresistance atthefault sent aportion ofthe current down bythe iron-wire shroud, itmust have leaped adistance of4feet tothewood hammock-berthing, itself abad conductor. Why this circuitous course should betaken, instead ofthe iron chain plates leading from the shrouds in#direct line tothe water, Idonot under- stand nor why itshould make such aleap atall,seeing that thepainted butt atthe fault could have been bridged bya ‘small fraction ofaninch. These facts have. come under my own notice, and Ifind itdifficult toaccount forthem onthe basis oftheaccepted theory ofatmospheric discharge. ‘Mr.J.Winsnurst: Sir,Professor OliverLodgestatesthat5Sur,there aretwo main cases oflightning flash, theoneofgradual growth, which hecalls the Aflash, the other when the strain arises 90suddenly that there isnotime foranypre-arrangement ofits path, ‘The first hecalls thesteady strain, thesecond the impulsive rush.Ithinkitisimportant torecognise thedifference VOL. XVI. 33 402LIGHTNING, LIGHTNING CONDUCTORS, Ere.[May%h,| %ssmane,betweenthetwocases,foritwasthefirst,ortheAflash,thatwas|me" contemplated bytheolder electricians, and Professor Lodge’ experiments have called attention totheother case. CertainlyI didnot know that there was anything new about theBflash, or the impulsive rush, formore than five years ago Imade some influence machines, inwhich theterminals were arranged atthe back ofthestand, from which this impulsive rush wastobetaken. Professor Lodge,Mr.Preece,andProfessor Forbes, Iknow,havethesemachines intheirpossession, butcertainly, evensofarback athat, Ididnotconsider there wasanything new about the impulsive rush. ProfessorLodgenextillustratedwhathewaspleasedtocall theolder, ortheaconditions, with the dome, theknob, andthe point, andheshowed very clearly, with theconditions hegave that the point protected everything near toit,notwithstanding that considerable resistances were introduced between itand the earth. Inthese experiments, however, hemust have made the cloud negative, and there isavery wide difference between making the cloud negative and the cloud positive. Hethen illustrated theBflash,ortheimpulsive rush,inwhichheshowed thepoint afforded nospecial protection, forthedome, theknob, and the point were each struck, one asreadily asthe other, according asitwasthehigher. Itappears tometobeclear that some additional illustration isrequired, foritisdifficult tobelieve that theolder electricians hadsuch alimited knowledge asisnow attributed tothem. Asshown byProfessor Lodge, thecloud evidently was positively electrified, hence they were allequally struck. Ipurpose now toshow you thegreat. difference inthe behaviour ofthespark when thecloud ismade negative. Inthis case youwill seethepoint again affords theprotection required ofit,foritisalways struck inpreference tothedome, although it betwice thedistance from thecloud. Infact, theonly difference which Ican find between thesteady strain and theimpulsive rush isthat intheformer thecharge isdissipated slowly,as8 brushdischarge,whileinthelattertheactualsparkoccursfrom thepoint.‘Thisarisesfromtheshortertimetoeffectthedicharge. ‘TheBflash,however,isalwaysweakerthantheAflash. 1689 DISCUSSION. 403 Thaveanarrangement heresimilar tothatusedbyProfessor tent. Lodge. The metallic cloud issupported onaglass pedestal. Here isaboard with tinfoilonthetop; here ametal dome with metal connection; here isapiece offine binding wire, which I shall press down bymeans ofthis rod, and you will seethevaria— tionofthelength ofthespark. Ipurpose showing youfirstthe conditions when there islittle preference. We will turn the machine andshow itspolarity. [Machine does notactreadily.) ‘This machine isnownearly fiveyears old; ithasnever yethesi~ tatedasatthepresentmoment. Professor OvenLovce:ThefailureofaWimsburst machinepote isofitselfastrikingexperiment. ‘TheCuarnwan: Woulditbesatisfactory toyou,Mr.Wimshurst, Pr ifsomebody else went onwith the discussion while you arepre- paring your experiment ? Mr. J.Winsnurst: Yes; quite so. TheCuatraan: Perhaps Professor Adams willgoon. Professor W.Grytis Apams: Iregret, sir,thatintheRpleer investigation ofthesubject ofLeyden jardischarges and their supposed application toflashes oflightning, Dr.Lodge should givesuch asweeping condemnation toeverything with which he maynotquite agree, orthat heshould think that theresults of previous scientific research and investigation aresoeasily tobe swept away. The discoveries made byHertz, with regard tothe discharges ofLeyden jars, areofvery great interest, and must modify our ideas onthe subject; butwhen Dr.Lodge says that lightning discharges areonly thedischarges ofLeyden jars, and asserts that alltheeffects observed insudden discharges from two plates ofmetal near together must alsobetrue oflightning, Without taking into account theeffects ofdistance, capacity, and timeofdischarge, hebegs thewhole question. Thedischarge of lightning from acloud, consisting ofdetached portions ofmatter with very slight powers ofconduction, issomething more than therapid discharge between themetal coatings ofaLeyden jar, Itwillalso appear that any effects ofresonance orside flashes withwhich Hertz orDr.Lodge have asyetmade usacquainted |Stemallcompared withtheoriginal electrical disturbance which t 404 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[Mas20, Rolewer iscompetent toproduce them;andonthisaccountthelightning discharge itself isfar moretobeguardedagainstthananyeffectof resonance orside flashes inconductors which may result from it. If,asProfessor J.J.Thomson hasshown, ahigh polish isneces sary togetthese subsidiary discharges, even inbodies ofasuitable capacity togive forth resonance discharges, then they arenot likely tododamage astheoriginal lightning flash. AsChairman oftheLightning RodConference, Imust disseatatoncefromProfessor Lodge'sstatement thattheolderelectrician, from Franklin toFaraday, knew nothing oftherealconditionsof theproblem; neither didthey northemembers oftheLightning Rod Conference treat itasatallaneasy matter. The Lightning RodConference took itupbecause there wasgreat: confusion aboat it,and this ishow they putitintheir report: “To consider the «possibility offormulating theexisting knowledge onthesujet “of the protection ofproperty from damage byelectricity.” Now, itisperfectly clear that ifthey questioned thepossibility ofathing they could not have regarded itaseasy. ‘The members oftheLightning Rod Conference didavery good pitee ofwork when they collected together andthrew into acondensed form intheir report allthat was known uptothat time with regard tolightning conductors. Their report isthecondensation ofthat knowledge asgathered from the evidence given inthe volume itself, and affords the true scientific basis forthecon- struction oflightning conductors. Nor amTas yetaware that thescientific basis, thecombined knowledge andexperience of thepast, has been overturned bythe conjectures orsurmisesa3 tolightning flashes which Dr.Lodge hasputforward, arising out ofthediscoveries ofHertz astothedischarges ofLeyden jar. Iregret that Ihave nothad more opportunity ofseeing Dr. Lodge's experiments, either attheBath mecting oftheBritish Association oratthemeeting ofthis Society afortnight since. Letmedraw attention tosome ofthepoints referred tobyDr. Lodge inhispaper, and seewhat were theviews ofthemembers oftheLightning Rod Conference upon them. Professor Lodge says that theBelgian electricians (Ishould rather sayM.Melsens) have been using foryears themethod esa] DISCUSSION. 465 ‘hich hehimself isnowrecommending fortheprotection offrlmor buildings, and thathehasseen nooccasion tomodify hisconcla- -sions ontheforms oflightning conductors. AttheBath meeting ‘oftheBritish Association, Dr.Lodge stated oftheHotel deVille, atBrussels, that “everything wascarried outinthemost approved style, regardless ofexpense,” and speaks ofitasthebest pro- ‘ected building intheworld. Theaccount which Dr.Lodge gives oftheprotection given tototheBrussels Town Hall isaltogether ‘exaggerated. Hesays: “No electrician exists butwould ayear “ago have asserted that itwas absurdly and exaggeratedly safe “from damage bylightning.” Yet within thepast year this building, 80amply protected byM.Melsens, hasbeen struck by lightning. Now, ifwerefer totheReport oftheLightning Rod Conference, wefind that thesystem ofprotection oftheTown Hall atBrussels transgresses therules laiddown bytheLightning Rod Conference. Onpage 138there isanaccount given oftheactual ‘system setupbyM.Melsens toprotect theBrussels Town Hall, anditappears asclearly aspossible that itisnotwhat theLight ning RodConference would regard asaperfect system, Inthe report, some eight distinct characteristics ofM.Melsens’ system areenumerated, andwhilst most ofthem weshould allcertainly ‘agree with, there isonewhich isdirectly contrary totheexpress instructions given intheReport oftheLightning RodConference. The account says: “The conductor consists ofnumerous thin “wires, which arevery flexible,soaseasilytobeledroundallthe “comers ofthebuilding.” Now, intheinstructions oftheLight- ningRodConference, itisstatedthatnoconductor mustbebentat agreater angle than 30°; and toobjections raised totheuseof copper tape, theCommittee say: “The objections totape willbe “found tobeobjections, nottotapeperee,buttobadpracticeon “the partofsomepersonswhohavefitteditup,andavailedthem- “selves unduly ofitsflexibility.” Now this applies farmore to M.Melsens’ finewires than tocopper tape, andthis isthesystem ‘which theCommittee condemns beforehand, butwhich Dr.Lodge -advoeates andwholly approves, and inspite ofwhich, orpossibly ‘inconsequence ofwhich, theBrussels Town Hall hasactually been ‘struck, Now, does Dr.Lodge think that when theLightning 406 LIGHTNING, LIGHTNING CONDUCTORS, Fre. (May, wxtwer RodConference object tosuch bending oftheconductor, they * have nothought oftheeffect ofself-induction orelectrie inertia? which isnosuch newthing ashesuggests, butwhich wasknown toFaraday and tothose who followed him. Let me draw attention toafewotherstatements inDr.Lodge's paper. Hesays, inparagraph 41,that theinstances there given illustrate side ashes and surging cireuits, but certainlyintro ofthethree instances given thegasissetfiretobecause thega pipe ismelted, being incontact with, and forming themain conductor. Idoubt ifeven, according toDr.Lodge’s principles, the side flashes canhave sufficient energy tomelt agaspipe when the main discharge iscarried away bythe lightning em- ductor without doing anyother damage. Asregards thethin instance, Major Majendie's account may betrusted, and quite accounts for the accident, even omitting the words which Dr Lodge encloses insquare brackets. Inparagraph 42,Dr.Lodge misrepresents thework ofthe Lightning Rod Conference. The ease cited (sev LARC, p.115) ‘cannot betaken “as illustrating that agood conductor affords n9 «absolute security,” butisdistinctly quoted by1..R.C. asshowing how afaulty conductor may bethecause ofdamage. “The fish oflightning fractured theconductor infifteen places,” asstated intheaccount given inL.R.C. “The inefficiency ofthecom “ductor resulted from the carelessness with which itwas fired: “the linewas fastened with twenty-five wall eyes, which were “hammered toodeepintothewall,thusdamaging theline,ant “forming ashort andsharp bend ineach case, besides alsounduly “straining thewire.” ‘This iswhat Dr.Lodge calls agood con- ductor. Isitatalllikely that agood conductor willbefracture! infifteen places byalightning flash ? Imparagraph 41,Dr.LodgequotesfromtheReportofthe Lightning RodConference. “Itisnouneommon thing forbuili- “ings provided with what arecalled lightning conductors tobe “damaged bylightning, and thecause isdue tothe inadequacy “oftheconductor tocarry theelectric fluid, which willleave the “conductor forabetter oralarger conductor.” Heshould have coutinued thequotation, andhewouldhaveseenthattherewas 90] DISCUSSION. 4st cause forthis statement. ‘This isthe continuation:“Thelight-Frotasee “ning descended theconductor toacertain point, Atthispoint “the iron flue enters the shaft, but some distance from the “conductor;themassofmetallocatedtherewasabettercon- “ductor thiam thewire rope, sothat inleaving therope forthe better conductor, theelectric fluid passed through thebrickwork ‘and caused thedamage.” Hemight also have cited thecase ofthe Board House atPurfleet (see LR.C., p.78). “The “lightning struck oneoftheiron cramps that hold thecoping. ~The iron cramp was situated over aplate oflead, and theends “ofitinserted inthestone came within 7inches ofthat plate, “which communicated with thegutter, and served asa fillet to “it; thisgutter waspart ofthemain conductor ofthebuilding. “The lightning struck through thestone tothecorner plate of “lead, fusing asmall portion ofit.” Now, canitforonemoment besupposed thatthisbursting through stonesandmelting of leadcanbedue tosecondary discharges orside flashes, rather than tothe diversion ofthe whole orthemain part ofthedis- charge, because ofthepresence ofamorereadyconductor? Inparagraph 47,Dr.Lodge quotes from Mr. Gavey's account ofachurch being damaged bylightningatCardiff,butdoesnot mention that theearth contact gave aresistance of115ohms. Themembers ofthe Lightning Rod Conference would scarcely havebeen satisfied with Dr.Lodge's mode ofaccounting forthe damage bybegging thequestion. Forinstance, inparagraph 43, heputs down tosecondary disturbance thesetting onfireof buildings, where there isnoevidence atallastohowthefiretook place. The Committee preferred tosay, “The precise cause of “the fire was not ascertained.” Dr.Lodge quotes @passage from ourreportas“illustrating “that apoor earth isnotnecessarily fatal totheefficiency ofthe “conductor.” ‘The quotation rather shows, and isintended to thow, thenecessity ofgood earth connection toprevent damage, “the ground wasupheaved where itwastoodry.” Inspeaking oflightning discharges, there isagreat tendency toregard itas simply adischarge ofelectricity ofonekind which starts from the cloud and proceeds through theairand theconductor tobury 408 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[May0, Roplomeritselfintheearth.Ithinkweshouldratherregarditasarush- ing together from thetwo ends ofthepositive and negative charges, andameeting ofthecharges inthemiddle, asshown in Wheatstone’s original experiments onthe velocity ofelectrial discharge. Ithinkthatthisexplanation willfitinbetterwith many ofthephenomena which areobserved inlightning dix charges. Thus,whenthecloudhasapositivechargeuponit,theearth andlead roofs and gutters, and allconductors having good earth connections, have aninduced negative charge upon then before the lightning discharge takes place—that, infact, the negative charge from theearth which istoreduce thepotential ofthepositive charge from theclouds istoagreat extent already outside thebuildings, atthetopofmetal roofs and good light- ning conductors, sothat theprincipal part oftheshock takes place intheairabove thebuildings soprotected. ‘From thispoint ofview wemay seethat agood earth ismoe important, forabad earth prevents the negative charge from spreading quickly andwith sufficient rapidity tothetops ofthe buildings and theconductors before theactual disruptive dit charge takes place. Under theheading (G), Dr.Lodge states that“intheeaseofsuddenrushesthereisnotimeforapathtobeprepared byinduction.” Ishethenprepared tosaythat thedischarge takes place without previous induction? Insuchs caseisthere nothing todetermine whither thecharge shall go? Orwould hesuppose that there isaction atadistance? Under theinduction ofthecharge inacloud, thelightning rodwiths good earth connection carries theearth's charge uptothetopof thelightning rod,andtoallmetal onthetops ofbuildings oon nected with it,orconnected toearth, sothat thebuilding belor is,asitwere, enclosed within acharged conductor, and themain effects ofthedischarge take place between thetops ofbuildings and thecloud. Starting from both ends atthe same time, the lashings andoscillations, ifany, take place forthemost partin theupper air. If,asDr.Lodge directs inhisrule 75,large ‘massesofmetalareconnectedtoearth,butarenotatthesame timeconnectedtothelightningconductor,thenthedischarge fromthecloudmaytakeplacetothelargemassesofmetal,and as88.) DISCUSSION. 469 ‘maystrikeandscatterstonesandbrickwork intheprocess,andPrise believe itismost unsafe, ashedirects inhisrule 76,toearth ‘things independently. Had Dr.Lodge known how common a fault itisforaworkman toput upalightning rodwithout making good earth contact, hewould probably have urged quite .asstrongly asthe members ofthe Lightning Rod Conference ‘thatspecial care should betaken tomake agood earth, foritis theone defect which they found tobemost common, and of ‘hich they could noturge theimportance toostrongly. Dr. Lodge now says that hedoes notnow argue against theneed of@ good earth, butIbelieve Dr.Lodge suggested atBath that points might beplaced atthebottom ofthelightning conductor just as ‘wellasatthetop. With such anarrangement, Ibelieve—and my ‘impression isderived from facts which were before theLightning Rod Conference—that the lightning rod, asfarasitwould do anything, would beacause ofmischief. Wehad many instances ‘oftheill-effects produced bybrokenorfaultyconductors. Take, forinstance, theCaterham accident (L.R.C., p.210), which is ‘worthy ofcareful study, toshow howalightning conductor should notbeconstructed. Inthis case thecopper tube was carried about 12inches into drychalk. Parts oftheconductor consisted -ofcopper tubes screwed together and giving noproper electrie -tontaet, some parts oftheconductor being fused bythedischarge. Inone case, thebottom end ofthe conductor was ledinto a bottle: perhaps theman who put ituphad some idea that a lightning flash was like aLeyden jardischarge. The kind of arth contact often met with isnot unlike the case met with at Middlesbrough (L.R.C., p.203), where agood conductor was carried underground for9ft.from thechimney and terminated atadepth ofabout 4ft.inhard, rather dryclay, theend being ‘wrapped round acommon brick buried inthe ground. The conductor making earth contact wasaflueentering theshaft, andinleavingtheproperconductor forthisfluethedischargepassed through thebrickwork andcaused thedamage. There isone other case towhich Iwill refer, which isnot recorded inthe Report ofthe Lightning Rod Conference. On May 17th, 1885, at2p.m., theLower Lighthouse attheSouth “70 LIGHTNING, LIGHTNING CONDUCTORS, Bre.(May, rptemcr Foreland wasstruck bylightning, Iwasasked bytheElder Brethren oftheTrinity House toexamine and report astothe cause ofthis lightning stroke, itbeing supposed that thelight- houseswerewellandefficiently protected. AsdescribedinAppendix IoftheReport oftheLightning Rod Conference, p.185,the instructions forprotecting lighthouses were drawn upbyFaraday, and ifhisinstructions had been fully carried out attheSouth Foreland there would probably have been nodamage done. a examination itwasfound that there wasavery excellent light- ning conductor inthe shape of@half-cylinder, 1}in.by§in, attached tothe iron-work ofthe lighthouse, passing down the inside ofthewall from toptobottom, andmaking perfect. contact with the earth plate, which was 4ft.by2ft.6in, and aboat vsofaninch inthickness, ofvery excellent copper. Faraday's instructions forthesecuring oflighthouses (asdetailed inL.RC. Report, p.186) seemed tohave been completely followed oat; butonfollowing theconductor totheearth plate itwasfound ‘that theworkman who carried outthework had placed thecopper plate atadistanceofabout4ft.outsidethebuilding, lyingfat ata depth ofonly about 1ft.,oratthe lowest part 1ft. 6in, below the surface ofagravelpathindrychalkdebris.‘Thelarge copper plate hadjust been putoutofsight inthedriest place possible, where itcould have noproper earth connection. To ‘obtain good earth contact inthe chalk isbynomeans an easy matter, hence the large sizeofthe plate used inthis case; but inthis, asinmany other cases, the object ws defeated. The Lower Lighthouse stands onthe edge ofthe cliff, at@height of290 feet above thelevel ofthe sea,and isabout 540 feet tothe east ofthe engine-house. In order togetwater fortheengines, itisnecessary togodora tosca level. The depth ofthe well bythe enginehowe is280feet. Butwhat became oftheflash oflightning which struck the conductor and ironwork of the lantern of the lighthouse? The internal ironwork, onwhich one ofthe keepers was standing atthe time, was inconnection with the lightning conductor, and inthat framework there was fabolt, the head ofwhich was about two inches below the 1980) DISCUSSION. am arleofthetrollywhichcarriestheelectriclamp.‘ThelightRptewer uingpassed from thebolt-head totheaxle ofthetrolly; then fom thetrolly byascrew, tearing away part ofthescrew-head anddepositing itonthebase oftheelectric lamp, which was injured; then totheinsulated terminal oftheelectric light mains andaway bytheunderground mains, adistance of180yards, to theengine-room, through the Holmes magneto machine, de- magnetising the magnets; and then bysome glow lamps connected tothemains and supported from themetal tie-beam oftheironroof,whichhasalightning conduetor leadingtoabigtank,whichisindirectconnection withthemetalworkofthe pump leading into thedeep well ofthe engine-house, which reaches tothesealevel. Inthe magneto machine alarge spark quarter ofaninch inlength passed between theiron and the copper conductor. One ofthekeepers was inside thelantern, Manding ontheironwork, which wasingood electrical connection with thepart ofthelantem struck, and with which thebolt which carried thedischarge totheaxle ofthetrolly was con- nected, ‘Thus thegood earth, bythemetal work ofthepump being raised from thesealevel totheroof oftheengine-house, supplies theearth charge which reduces thepotential ofthe lightning flash from thecloud; here avery powerful discharge pases along good underground conductors over long distances anddoes very little damage. We may well say, with Lord Rayleigh, that “it isonly byactual experience oflightning “conductors onavery large scale that thequestion canever “be finally settled.” Laboratory experiments may bemost important assuggestions, butnoonewould adopt. orchange any system oflightning protection without actual experience. Mr.J.Wisisuurst [proceeding withexperiment]: Iwillnowurmakethecloud‘positive, andyouwillseethatallofthem (thedome, knob, and point) now areabout equally struck. ‘The Point, asyousee,gets aslight preference, These conditions only were shown ustheother evening. But now letmemake the cloud negative, andyouwillseeavery great difference. ‘There is 125percent. indistance infavour ofthe point, and that is : ith theimpulsive rush. Ithink that shows that there isa manifest advantage infavour ofthepoint. aa LIGHTNING, LIGHTNING CONDUCTORS, Ere.[May%b, Egle Professor Oziven Lonae: Ididnotnotice whether itwas positive ornegative. Itwas sometimes one, sometimes the other. Xmane, MrJ. Wiusuunst: Indeed, Butthere isgreat differencein thebehaviour ofthespark with thepositive ascompared withthe negative cloud. [Experiment.] Letusconsider three ofthemost important points, 1.The enormous resistance, asProfessor Lodgecallsit,ofthelarge copper conductor. 2.‘The small importance which Professor ‘Lodge places upon athoroughly good earth: hesays itmaybe made asgood asheaven, butstillbeuseless, 3.The propriety of connecting allmetal inthehouse asinaring, andconnecting that ring toearth. ‘Totest thefirst ofthese, wehad better repeat theexperiment which Professor Lodge calls theby-pass: itconsists ingiving the Leyden jardischarge thechoice between theconductor tote tested and acertain length ofair. The long wire by-pes representing theconductor hecalls L,andthespark through tie airhecalls B,and heassumes that the distance between the knobsatBmeasures theresistance offeredbytheby-passorcom ductor. This appears atfirst sight tobereasonable; batit occurred tomethat itwould bewell toanalyse thedirectionof the Bspark, infact toascertain whether the spark atBiss direct current ofelectricity passing from theoutside coatingofonejartotheoutsidecoatingoftheotherjar,orwhetheritbe something else. The means Idevised forthis test was arrangement offour balls affording two paths forthesparkto select either ofthem, each path containing asmall and alage ball. When thisanalyser isplaced inthepath without theby- pass, theselection, asIwillshow you, agrees with theterminals ofthemachine, butwhen theby-pass isinserted theother path isselected. (Experiment). Inow place theanalyser incircuit. Theby-pass that Iamusing isabout 70feet ofNo.12gauge copper wire, wound round thelegs ofthetable. ‘Theby-passis now incircuit, andwewillseewhich route thedischarge takes. ‘The Cuarauan: Would itsuit you, Mr.Wimshurst, tobare themeeting adjourned, andhave theapparatus ready nexttime? 1669) DISCUSSION. a Mr,J.Wowsuunst: Ishould beverygladtofallinwiththeMe. suggestion, andwillonthenext occasion have thelarge machine ready toshow theexperiment. The CuamMaN: Before themeeting isadjourned, Iwould suggest toProfessor Fitzgerald that, ashecannot behere next meeting, hewillperhaps favour uswith hisremarks now. Professor G.F.Firzcrraup: Mr.Chairman andgentlemen, Protmor, Thave been extremely interested inMr.Wimshurst’s experiments, which aremore interesting than anything Icanpossibly tellyou. TheonlypointsthatstruckmearethatIdonotatallattribute theimportance that, Mr.Preece does tothe arguments which he brought forward against theoscillatory character ofthelightning discharge. Ithink itisquite possible, notwithstanding allthat hasbeen said, that lightning discharge isoscillatory inthesense thataLeyden jardischarge isoscillatory, butitmust bealways recollected that theoscillation preponderates inagiven direction, andisnotalways asmuch inonedirection astheother; there is ‘preponderance inonedirection, andthere isapossibility ofthis preponderance causing acurrent andproducing theeffects which Mr.Preece hasdescribed. Idonot think there isanything in theeffects hehasdescribed todisprove theoscillatory character ofthespark. ‘There aretwopoints which, inmymind, should bebrought nttomake thediscussion satisfactory. They are, first ofall,to what extent areyou hyexperience orexperiment justified in assuming that acloud acts like agood conductor? Idonot think itcanbeproved, byanything weknow of,that they do certainly actlike good conductors, Ifthere areany experiments which show that they do,Idonotknow ofthem, and Idonot think they have been mentioned either byProfessor Oliver Lodge oranyoneelse; while there isagood deal ofevidence toshow that atleast sideways and laterally they donotactlike good conductors, because occasionally clouds lieinavalley, touching hillsoneach side, and discharging lightning from their centre, hich they could hardly doifthey acted like »metal plate: so thatwhether acloud isagood conductor ornoisaveryinterest ingquestion, Ithink itisquite possible thatitmay actlikea a LIGHTNING, LIGHTNING CONDUCTORS, Ere, (las, Frist, good conductor vertically when itisdischarging rain. That seems tometobeaweak point inProfessor Lodge’s experiments illustrating lightning flashes. ‘The weak point ontheother side isthat Idonotthink that anyexplanation that Ihave yetheard, evenProfessor Adams'sexplanation, fullyaccounts forsuchresults asweheard described to-night with reference totheship andthe disruptive action onaconductor which isitself notsensibly heated. Professor Lodge similarly has called attention toseveral cases— forinstance, whether thedescription ofitisquite accurate orno ithasnotbeen disputed, that #horizontal barofmetal, totally ‘unconnected with anyother metal, gave outdisagreeable sparks during athunderstorm, That Idonot think hasbeen explained byanywho have spoken ontheother side, and Ithink itwould require some such explanation asProfessor Lodge hasgivenin order toaccount forthe injurious effects that have occurred. He says, “It has been investigated, and isreported to “have been due toasecondary charge inthe horizontal “bar totally disconnected from anything.” Hegoes onto say, “Although itdid not probably receive aside flash,yet “itgave outsparks that ignited gasand caused afire.” Now, think that thepoint that practical electricians oughttodevte themselves toistoexplain these curious and unexplained phenomena, andnottobestating that everything with reference tolightning isquite understood, forIdonotthink that everything with respect toalightning flash iscompletely understood. I would notlike, forinstance, torepeat theexperiment suggested byMr.Preece, and would bevery sorry toseehim tryit,of holding ontoalightning conductor when itwas struck by lightning, for,although hemightdoitonceortwice,Iamverymuch afraid that occasionally hewould suffer very severely. 1 would notlike todoit,and Ivalue Mr. Preece toomuch towish toseehim perform the experiment. However, Ithink thereit very little doubt, asIsaid atthe British Association meetingat Bath, that lightning protectors and lightning conductors have been and areofimmense useinprotecting mankind from the destructive effects oflightning. ‘The Cuauan:Professor OliverLodgewill,unfortunately, be 1680 pisovssion. 8 tunable tobehereattheadjourned discussion nextweek, andhePv... tellsmethatheisverywillingtoreplynowtothediscussionso farasit,hasgone,andI'aminclinedtothinkthatweshallallof usbevery glad tohear what hehastosay. Themeeting signified approval. Professor Ouiven Love: Itisvery kind indeed ofthePmiwer meeting, and ofyou, Sir,toallowmetoreplynow,butIwill beasbrief aspossible. Icannot take upallthe points, and perhaps Imay unintentionally pass over some that areimportant, Tintended tohave begun with some remarks complimentary toMr.Preece and hiswide experience, butyou must consider all that assaid. Asregards the difference between theory, experience, and observation; ofcourse there isbadtheory and good theory, just 1sthere isgood observation and bad observation, especially there isunguided observation. Now, Isaythat, bymere observation unguided bytheory, itisvery easy tooverlook agreat many facts which areplain enough inthe light oftheory. Unguided experience isbynomeans satisfactory, else would discoveries be much easier than they are. Many ofthe effects Ihave shown—sparks inunexpected places, and other things—have been observed before. Henry observed things ofthekind; and Edison noticed some curious phenomena, and said itwas not electricity but “etheric force” thatcaused these sparks; and thematter wasrather pooh-poohed. Itwasasmall part ofthisvery thing, only thetime wasnotripes theoretical knowledge was not ready forit. Professor Carey Foster says that Professor Mascart, who was present atthe last. meeting, told him that M.Melsens had observed many ofthese things, buthad notfelt sure enough topublish them, because hedidnotunderstand them. There isnothing much inwhat Ihave done about the matter; itisthe fundamental theories ofyour illustrious President and ofClerk-Maxwell that have enabled metoobserve these things now soeasily, when before they would have been almost impossible.* ©Mr.8.A.Varleyremindsme,andIcordiallyacceptthereminder,that ‘theeteoftheteexperiments isalsolargelyduetothedevelopment ofthe‘madera influence machine, 476 LIGHTNING, LIGHTNING CONDUCTORS, Bre,[May90, Egige Mr.Preece made some remarks about case}which Ididnot follow,s0Iwilldefercommenting uponthem.*Hesaysthereisno difference between uswith respect tocase a.But this difference between case @and case bwhich hasbeen somuch spokenof, partly because Iinsisted upon itsomuch inmypaper,hasonly 8very partial application, and refers towhat shall bestruck, whether point orknob; itrefers merely tothequestions of whether there shall beaflashatall,andwhatkindofterminal ismost likely tobestruck. When theflash hasonce occurred thereisnodifferencebetweencaseaandcaseb.Asregardsthe conduction ofthedischarge, they arethesame; their differences appear only inthedielectric during theerapreceding discharge. Tam glad oftheopportunity ofmaking thisexplanation, because Thave seen itstated that only under thecircumstances ofthe impulsive rush isthedischarge sudden, and that only then ae these violent shocks and side-flashes (which Ididnotshow to the full extent Imight have done last time) likely tooccur But thefact isthat with anyflash there arethese violent dis turbances, nomatter how ithas arisen;onlyitisnotsoeasy getaflashtopointsincase«asincaseb, With regard totheexperiment ofMr.Wimshurst, Imust sy that Ihad not noticed the fact hehas shown, viz., that under certain conditions itmakes adifference, even incase 6,whether thetopplate ispositive ornegative. Inmyown case itdidnot, because Iwas using larger jars and more violent rushes. Bat there isevidently aslight difference, perhaps abigdifference under certain conditions, whether the cloud ispositive ot negative, which Ihad notpreviously noticed. ‘The difference is obtrusive incasea,notincase6;butwhenIgetbackIwill repeat theexperiment, and seewhat theconditions are.t ~"eThave nothing muchtomyabout themnow,except thatthediaga Mr. Preece gives isscarcely analogous tomyease arrangement, Intht diagram the cloud inone coating, the earth another, and the tinplates aninterposed inthedielectricofacommoneasee,Anothercloudisnecesry torepresent theouter coating ofthejar inmycase 8,Itwill then becomeot like the figures inthe plate atthe end ofmy paper. ‘tndIwasinerrorinsaying{hadnotnoticedthisfact,Onmentioning themattertomysasstant,Mr,Robinson,Tfoundheknewallaboutsaat ‘remindedmethatwehadnoticedthatthemoregentletherush,themoredit 188 DISCUSSION. at Asregardsironandcopper,Mr.Preecesayseverybody knewfrvanethat iron was best. Well, the Telegraph Department certainly useiron wire toprotect their poles, but ithas not been the practice inthis country toerect iron onbuildings; and most certainly the reason why iron isasgood ascopper was not known, Moreover, the twowere always compared when theiron ‘wasseven times thesectional area ofthe copper. That isnow seen tobeallnonsense. ‘What Professor Adams says, thatthestatements oftheCon- ferencewerefairlycorrectinthelightofthenexisting knowledge, isperfectly true. Ihave nottheslightest fault tofindwith any statements made inthe past; Idofind fault with the same statements being made now. Inone sense self-induction isan oldidea, asoldasFaraday’s “extra-current,” but nevertheless itmay besafely asserted that in1881 noone, not SirWilliam Thomson, Lord Rayleigh, nor any one, knew self-induction so intimately asitisknownnow.Nodoubtsomeofthemembers ofthe Lightning Rod Conference had self-induction intheir nind, butthey didnotknow thefullextent ofitsimportance, andtheir knowledge atthat time would have ledthem to condemn ironvigorously. Still more recently itwould have been condemned inthelight ofthetelephone experiments ofProfessor Haghes, because hisoscillatory currents were notrapid enough tobring out the relative merits ofiron and copper under lightning conditions, ‘Mr. Preece said that itwasthegreat power oftheWimshurst machine that made allthe violent disturbances last time. That machine hasgreat power, but Ihave not used that machine except toshow theexperiments here. Iobserved them allwith ‘asmall single-pair-of-plates Voss machine, 18inches indiameter; andthis produces alltheeffects just aswell, only rather more slowly. Itdoes not require apowerful machine, ittakes little power. ‘Mr. Preece says that itistheWimshurst machine that makes ‘asebapprozimatetocavea,Italldependsonthetimetakenbytheplate tocharge, andthe time consequently allowed forthepath tobeprepared by static induction, (See Appeniix IV.) ‘VOL. XVIII. 34 a8 LIGHTNING, LIGHTNING CONDUCTORS, Ere. [Sy Tlethesesuddenrushes,butthat“inNatureeverything isslow,and «stately, andquiet, and gentle,”—especially lightning! Hesays henever doubted the oscillatory character ofa| Leyden jardischarge when there isselfinduction present; bat you carinot Keep itout: there isself-induction inevery circuit thateitherheoranybodyelseeversaw.Youdonotwantcvlt tomake self-induetion :there isself-induction inabit ofwire a long asone's finger, and intheperforation oftheglass of« Leyden jar,though that iscertainly themost difficult ease. I cannot admit that when oscillations are absent there isnoself- induction.* ‘Then, again, thebluing ofpaper andsuchlike prove nothing against theoscillatory character oflightning. Itisneed lessformetodwell upon this after what Professor Fitzgerald bas said. Hehasshown why itshould besotheoretically, andI iknow well that itissoinfact with many oscillatory currents You may have noticed aletter inNature aweek ortwoback, by Dr.Dragoumis, who wasusing vacuum tubes inmylaboratory todetectHertz-like oscillations :heusediodideofstarchpapera well, andgotblue marks allright, But astooscillations, they arehardlyessential tomyenso:itisLathatTwantbigsand thoughtheeasiestwaytoget.4°bigistoaveoscillations, they arenotabsolutely essential: asProfessor Fitzgerald saidatBath There isnoreal need, therefore, todiscuss them.t ‘Astothecase oflightning protectors, Mr.Preece says Tbave xnoexperience oflightning protectors connected totelegraph instruments. Itisperfectly true—I have not. Itwould be ridiculous formetopretend tofirst hand information onsich matters inthis presence, butMr,Preece has;andinatext-buok ontelegraphy byPreece and Sivewright (6th edition, 1887), itis *The reno Foriginally imagined thet Mr, Preece doubted theoil tory characterofLeytenjardischarge,waaBecausehesaldatBathfutie pprosched thequestion with some ditidence becaue Itwas supported byhitStthortin. Iwan,indeed!Butbohighenthortyhadsuphortedteiteofthe owcilatory character ofnlightning Atel, However tia masft that many perons doebt theoaclatory character ofaLeyden jedhaMr.8A.Varley,forintaneeeecurrentHlcrclReviewarticlen 4This etately true; ovelllatons are estentil tothe moat violent etl, 1889) DISCUSSION. a“ recorded what trouble lightning gives them: howitdamages thefre machines, depolarises theneedles, and spoils theinsulation;how protectors are ofsome use, but are not wholly reliable; so thatwhenever itisabsolutely necessary tokeep itout— especially from cables, where lightning might bevery destruc~ tive—then they apply agreat number oflightning protectors, inthehope that oneorother will work. That iswhere Igot nyinformation from ;*andalsofrom direct: experiment proving thatnoexisting protector caneliminate Leyden jarsparks, as M.Guillemin and Professor Hughes had found out before. Mr.Preece says there isnow inEngland only 1°3percent. of damage,—that- means destruction,Isuppose,anddoesnottake intoaccount partial depolarisation orslight damage tothe insu- lation ofwires, which must occur inmany cases without. imme- diately breaking thewires down, Butwhy isthere 1-3percent. offailure ifprotectors arealready perfect? Lightning protectors can,however, beinade absolutely perfect. ‘They aremuch easier totackle than lightning conductors attached tohouses andlarge buildings. Professor Adams insists that the Lightning Rod Conference hhasdone good work. Certainly; Iadmit it. Soalso Mr. Anderson's work on“Lightning Conductors,” revised byDr. Mann, isanexcellent book. Ofboth works Ihave made the greatest use, Iregret that Ihave seemed tobeinantagonism totheLightning Rod Conference;but,whiletheyhavedonegood work, that gives them noclaim toinfallibility. Neither doI wish tobind any ofitsmembers totheterms ofthereport. *Thefollowing arequotations frompages257-9:—«Lightning ithemostfruitfulsourceoffaultupontelegraphcircuitsinthosecountries ‘*wherethunderstorms arerife,andatmospheric electricity isundoubtedly the‘greatest enemy which those employed intheir maintenance have to encounter....Earthcurrentsareonlytroublesomeatlongintervals;but “lightning is constant source oftrouble, frequently bresking down the “citeuita, and sometimes, inspite ofallprecautions, odamaging the instru. “mente astorender their removal necemary. ... Although the dangers“liabletoarisefromlightninghavebeentoagreatextentsurmounted of “recent yeary, still itcannot beaaid that athoroughly eficient form of “lightning protector fortelegraphic apparatus hasyetbeen devised.”—Preece 4Sivewrightts «Textbook ofTelegraphy.” Longmans, 167, ExtererDoubtless manyofthat Conference feltqualms when they sigoei it, Adocument signed byanumber ofpersons isalways mae ‘orlessofacompromise. Asgenerally sound advice tobuilder and thepublic, itserved itspurpose;butifitistoberegarded asadam across the stream ofprogress, ahigh-water mark beyond which the ocean ofdiscovery may not flow, then the wonder isthat inthis ageithaslived eolong. Professor Adams made one elementary slip, which Ithink be will seetobeaslip,where,inattemptingtoexplainthefactquoted byMr.Farquharson, hestatedthatwhenacloudandaroofdx charge into oneanother thedischarge confines itselftotheair; that once the charge has gotthoroughly upinto the roofthea thedischarge occurs wholly between cloud and roof, anddoes not come down theconductor atall! Allthat, ifadhered to,is 8perfectly new theory, entirely contradictory ofwhat iskuom concerning the continuity ofelectric currents; and ifexisting facts require extraordinary theories toexplain them, they are,ia ordinary language, inexplicable. Thave taken upmytenminutes, and there aremany poiats Ihave nottouched upon, butIthink Ihad better keep strilly tothe time. ‘The Caamman: Mr. Wimshurst will continue the discussior atournext meeting, which willbeheld onMay 16th, whes,if time permits, apaper willbereadbyMr.Mordey on‘Alternating Current Working.” Aballot took place atwhich thefollowing were elected :— Foreign Member: “. Mascart. Member: Henry Sherley Price. Associates: James Dawson. I Napier Prentice. Fdward Mark Robinson, ‘The meeting then adjourned. 1099.) DISCUSSION. ro TheOne Hundred and Ninety-fourth Ordinary General Meeting of the Institution was held at the Institution of Civil Engineers, 25,Great George Street, Westminster, onThurs- dayevening, May 16th, 1889—Sir Wit1am THowsoy, F.R.S., President, inthe Chair. The minutes ofthe Ordinary General Meeting held on May 9th were read and approved. The names ofnew candidates for election into the Institution were announced andordered tobesuspended. The following transfer wasannounced ashaving been approved bytheCouneil :— From the class ofStudents tothat ofAssociates— Arthur Henry Lea. ‘The Secretary: Iwish toremind members, some ofwhom arenodoubt aware, that there arestill copies ofthe“Light- “ning Rod Conference Report” tobeobtained, either from Mr.G.J.Symons, Secretary totheMeteorological Society, who edited theReport, oronapplication tome. The published price was 7s.6d., but members ofthe Institution are allowed tohave copies at5s. ‘The Presivexr: Ithas been found necessary todefer the reading ofMr.Mordey’s paper on“Alternate Current Working,” whichwasannounced forthisevening,becausethediscussion on thepaper byDr.Lodge, on“Lightning Conductors,” was uot closed attheend oflastmeeting. Under these circumstances, theCouncil have resolved upon holding anextra meeting on the30th instant,soastogiveampletimefordiscussiononDr. Lorige’s paper this evening. Thave now tocall upon Mr. Wimshurst toresume the adjourned discussion onDr.Lodge's paper. Mr.J.Wausuunst: Sir,Iwishthisevening torepeat theMe. experiments oflastweek with thelarger influence machine, to 482 LIGHTNING, LIGHTNING CONDUCTORS, Bre.[May164, Senor, better show thattheimpalsive rush does prefer thepoint tothe dome under every condition;and,further,thatunderanycondi- tion butoneitmaintains avery decided preference forthepoint, Inmost cases itwill leap more than twice thedistance between the cloud and the dome. Iwill show that experiment atonce anditwill bewell, perhaps, totest thepolarity ofthemachine, sothat wemay seewhich terminal ispositive and which neg tive. [Experiment.” You seethelength ofthespark isaboot 13in,,with theballs arranged asatpresent. Iwill shift them, the larger forthe smaller, soastoshow the difference which results, You willobserve that this isvery great. Itisofthe first importance toknow the polarity. I’cannot with this changed arrangement getalonger xpark than about 7in.We now know thepolarity ofthemachine, and Iwill leave itthat way fortherest oftheevening;thepositiveterminalisthelet band—the smnall ball; and thenegative isattheright hand,or largeball. —y~ 4 \ps a™ Fro. Thiscloud apparatus (Fig. 1)consists ofthemetal cloud, Cs board, B,covered with tin-foil; Disametal dome; Wis piece ofinetal with ashort length offinebinding wirewhieh touches thetin-foil. Theupper endIwillpress with arod. Tie cloud, C,isnow made positive, while theboard, B,ismole 1860 DISCUSSION. 483 negative. ‘Thus arranged thedome, theknob, K,and thepoint Mt, areabout equally struck—the point having asmall advantage, [Esperiment.] . ‘ThePresipenr: The question was between the dome and this ball, Mr.J.Wrusuuxsr: Just so,orthepoint; forthere arethree things. ‘The Prestoext: Oh,three things. Mr.J.Wiusauxst: Shall Irepeat theexperiment again asI hadit?Iconnectthecloud,©,tothechainfromtheoutsideof thejar,which isinconnection with thelarge ball. ‘Mr. W. H.Preece: Could you come tothis corner, Mr. Wimshurst? Professor J.Peary: Itisdifficult todoso;these chains get incontact, itismuch better tobeclose here. Mr.J.Wiusnursr: Ishallbequitereadytovarytheexperi- ment inany way you wish. Iwill now alter the connection andmake thecloud negative. [Experiment.] Sothat there maybenomisunderstanding Iwill useonly thesmall knob, K, andthedome,D.[Experiment] Thesmallknob,K,isstruck ifabout 2}times asfarfrom thecloud, C,asisthedome, D. Professor J.Peary:Thecloudisnegative? Mr.J.Wiusnunst: Yes; when thecloud isnegative; you then getthesmall knob selected. ‘This experiment removes all doubt about thepoint being supremely useful;atleast,Ithink 0 Professor J.Penny: Atthesame time, Dr.Lodge saidhehad uever found any difference. Mr,J.Wivsnurst: Last week when theband ofmyinfluence uuachine failed, from age, Iwasabout tomake some experiments upon theremaining three chief points inDr.Lodge's paper. The firstwasthatwhich hecalls theenormous resistance ofalarge copper conductor ;the second isthe small importance which he attaches toathoroughly good earth—his observation being that he might make it“equal toheaven” still itwould notanswer; and thethird isthepropriety ofconnecting, ashenowrecommends, allpieces ofmetal within abuilding together asinaring, and oyLIGHTNING, LIGHTNINGCONDUCTORS, Ere.(May168,| Monona connecting that ring toearth. This latter seems tometobe |fataltosafety;butIwillshowyousomeexperiments bearingupma these points. ‘Totestthefirstofthem—the enormous resistance oftheem- ductors. Letusrepeat theexperiment oftheby-pass. last week Iexplained theconstruction oftheanalyser (b)(Fig. 2).1 ieoF . LH, ‘ y mansHEH Analgoer Ce) fe, Qe Q.Q 9 Past avout iteofNRCFro. 2. willhandittothePresident thathemaysee.Hereaeblocks(cd)ofwood,eachhavingtwomovablearms,onewithslarge ball and the other with »small ball atthe end; the twoarms areinmetallic contact, arranged toafford twopaths, sothit when Ibring thetwoblocks within striking distance thesparkis 1 DISCUSSION, 485 freetotake either path. Itisarough arrangement, butit6.4 answers very well. Iwill connect this analyser (})with the outsideofthejars¢andfandweseetheresult.Thenegative isontheoutsidecoatingoftheleft-hand jar(e),andthepositive ‘ontheoutside coating oftheright-hand jar(f). [Experiment.] ‘Thepositive selects thesmall ball, and thelower path (g)isused. Thiswire*by-pass” isabout70feetinlength. Ihaveplaced ‘neendofitatthis corner ofthetable; itthen runs round the tablelegsandbackagain,theotherendbeingattheothercorner‘ofthe table. Itisinsulated bythewood ofthetable andisnot ound inanyspecial manner, butmerely inaconvenient manner forkeeping itoutoftheway. Iwill connect theends ofthe “by-pass,” with thechains, totheanalyser (b). The spark is wowvery much shorter than when the “by-pass” was not connected, and wefind [experiment] theopposite path (/)is chosen, ‘The Paesipent: Are we tounderstand that the balls were connected bythelong wire? Mr. J,Wiasuunst: The connection isthis: the chain comes fromtheoutside coating ofthejar(¢),loosely round tothis (&) ‘tadoftheanalyser; themiddleofthechainisthrownuponthe (i)end ofthewire conductor. ‘The path fordischarge isalong thechain, through thewire “by-pass,” andtothejar(/)through theother chain (l). The chains areloose, and arethrown over theends (iandm)ofthe“by-pass ;”thespark isfreetopass silently bywayofthe by-pass” ortoleap thespace between the ballsoftheanalyser. ‘ThePresienr: What isthedifference between thisarrange ‘ment andthepreceding arrangement, when thelong spark wason theside next us? Mr.J.Wousunst:: Atfirst theanalyser only wasinconnse- ‘tionwiththeoutsideofthejars,the“by-pass” orconductor notbeing inuse, Professor D.E.Hucues: Ithink you aretrying torepresent theAandBspark, areyounot,which Professor Lodge spoke of inhisastlecture ? Mr,J.Wiwsnunst: Yes, 466 LIGHTNING, LIGHTNING CONDUCTORS, Exe. (Slay 16th, Xam: M+J.Wiusutunst: Hehadpreviously spoken oftheBspark ‘asameasure oftheresistance oftheconductor, and inhispaper headopted theview taken inhisformer lectures attheSociety ofArts and elsewhere. ‘The Prestpest: There isanalternative path intheprevious | experiment ? ! Mr.J.Wisuvxst: Inthefirst case Ihadsimply the whole charge leaping from oneballtotheother ballthrough thespace (y),whereas now Ihave connected thealternative path ofsome 70ft.ofNo.12gauge copper wire incircuit, andIfindthespark selects theother path (h). Ithink that that isan indication that thecharge from thecoating ofonejarpasses silently through the “by-pass” tothecoating oftheother jar. ‘TheBsparkisprobably nothingmore than the manifestation ofanoscillation. Icannot ascribe ittoanything else. Certainly itisnot the measureof resistance ;butthewholematter needsfurther investigation. Toshow the importance ofathoroughly good earth connec tion, which lately has been solightly spoken of,Iwill use the “by-pass” wire, which isofconsiderable length, and connect its two ends tothe outside coatings ofthe two Leyden jars, for that arrangement supplies the best earth and allows the charge topass through it. You will see there isvery little tendency toside-flash; itisanearly perfect ‘ state ofthings, and with such perfect earth itwould bequite | possible foraman toholdtheinch lightning-rod, while theflash | strikes, without injury, provided that heisinsulated. Let us | make theearth lessperfect, byintroducing some wetpaper atthe endofthewire—I have some inaglass here—and weshall find | that theside-flash increases asthebadness oftheearth increases;| dowhat wemay, some side-flash willoccur fortheunlike charges being suddenly released seek toequalise themselves and toavail themselves ofevery path which presents itself, Itseems tomethat this isquite consistent with theteachings ofthe entlier electricians, Iwould point out another characteristic ofthis side-flash—for every feature inithastobeconsidered— itisthat the tendency toside-flash isgreatest atthetwo ends ofthe wire, and there ishardly any tendency toside- | | 1889.) DISCUSSION. “7 flashatthemid-length ofthewire,whichisalmostneutral,Mey Atthe one end the wire “by-pass” appears tobepositive, itappears toside-flash, ifImay sayso,outwards, while atthe other end itisnegative, and itwould appear toside-flash inwards. The ends ofthe “by-pass” arecertainly oppositely i Ss *Pass™ “orne19Guage¥o% Fo. electrified. Iwill nowplace these Leyden jars (1and2,fig.3) near tothe conductor tobestruck with theside-flashes; theone atone end ofthe“by-pass,” theother attheother end, and allow several side-flashes toenter thejars; Iwill then test thejars toascertain thecharges theyhave obtained, when 48 LIGHTSING, LIGHTNING CONDUCTORS, Ere.[May16t- Minaume itwill befound that they contain nocharge whatever. This experiment appears topoint tothe fact that the side-flash isoscillatory initscharacter. Inow put several ofthese side-flashes in,still you seethejars contain nothing. Ihave nottried them with anelectroscope, butIhave apith ball, which youseeindicates nocharge; Ithen bring thejars together, westillseethat there isnocharge remaining, although wehave seen many sparks enter. Iwill now put some looee pieces ofmetal together, and you see that the side-flashing materially increases asweconnect the pieces together. Ithen connect the pieces ofmetal toearth, and the side-flash at once becomes violent; theelectricity seeks asmany paths as itcan find, andagreat part ofthe flash passes bythis second path, Iwill now connect the“by-pass” and show this effect. (Experiment.] The whole length ofthe“by-pass” isincircuit between thetwooutside coatings ofthejars. Ihold the sheet oftininmyhand and test theside-flash atthis ends then connect the two pieces together, and you will see that the side-flash increases, [experiment] forthe larger the mass of material thelarger must bethetendency toside-flash into it. Tf,onthecontrary, themass were sinall, say,apin-head insize, itwould nothave capacity toreceive aflash, Therefore, con- necting pieces ofmetal together in«house iscertainly fatal, and when they are connected toearth, which Iwill now show you, theenormous increase ismanifest. Iwillnow show youthe increase inthe side-flash with the bad earth, Itake this piece ofpaper from the glass—it has been soaking alltheevening; Tplace itbetween oneendofthe“by-pass”andthejar,thespark hhastopassthroughabout5incheslengthofthoroughly wetpaper. ! [Experiment.] You sawthat itmaterially increased theside- | spark. That, Ithink, again isonly what might beexpected. | Ifyouputanyresistance between theconductor andtheearth | theside-flash must belarger.ProfessorW.E.Ayrton:Haveyougotyourlightningcon-| ductor—that istosay, your wire—also connected with theearth? Mr.J.Wiusuvst: No,itissimplyconnected withthe| Leyden jars. | a! 180.) DISCUSSION. 480 Professor W.E.Avatox: Oh,butProfessor Lodgesaidthat¥f. your lightning conductor must beconnected alsowith theearth, Mr. J.Wrusiturst:Itisconnected withtheearth,butwith anindifferent earth, theindifferent earth being five inches of soaked paper, and also this table, Iwillshow you theexperi- ment again, firstwithout thepaper andthen with it. ‘The Prestpext: You arecomparing agood earth and abad carth? Mr. J.Wiusuuast: Yes; Iwill now put the wire indirect metallic connection with thebottoms oftheLeyden jars. The Presivext: The lightning conductor isnow inorder ? Mr. J.Wausuunst: Yes, itisnow perfect. Professor W.E.Avrtox: Wien you put anearth, doyou mean thething towhich thelightning conductor isattached, orthegas-pipe ofthis building? Mr.J.Wimsntnst: Ispeak ofaperfect earth. Professor W.E.Avntox: Have you got itconnected with thie(the jar)? Mr. J.Wiusnunst: Yes; Professor Lodge himself says that that isaperfect earth, ‘The arrangement now isthat the charge from theonecoating totheother circulates through the whole length ofthis wire, and again weget only what might have been expected, viz, lessside flash. With reference tothe oscillations,Ishouldsupposetheseto beproportional orabout proportional tothelength ofthemetal rodand the current circulating through it, Ifso,thetendency inactual practice cannot bevery great; the lightning-rod is never ofvery great length. Ihave several times tried, bythe useoflong wires and anumber ofLeyden jars, toobtain some small indication oftheoscillation onasensitive-plate rotating with great velocity, butasyet mytrials have been failures; Ishall hope todomore inthis direction. Inow hand the photographs Ihave taken toyou,sir,foryoutodistribute. Thereisnoindication whatever ofoscillation, asfarasIcansee. ‘With respect totheobservations ofProfessor Lodge last:week, that heused aVoss machine when making hisexperiments, and that any machine ofthat type would answer thepurpose, for 490 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[ay1th, Xosane, further investigation, Iwould state that, inmaking experiments with this high-tension electricity, itisofthe first. importanceto know whether the charge isofapositive ornegative kind. Professor Lodge, itwould seem, hasworked quite regardless of this knowledge; butIthink all:here will know, and must have noticed, thegrave errors which anymachines capable ofreals changing their polarity arelikely toentail ontheuser. Itwould bbefar better tofall back onthe old-fashioned frictional m: ‘ines when others ofamore reliable character cannot beobtaincs Ifmay oceupy your time forafew more minute. I<boald liketoshow you anexperiment with thespangled ‘1r. Itisnot 1avery brilliant experiment, butitistheanalogue ¢‘hesparking which wesawonthewalls. Ihave here alarge ja.;there isn0 metallic connection with theinside ofthejar,forIhave simplya sheet oftinfoil rolled upand suspended upon the rod. Now when thecharge and discharge passes, they induce analteration ofcharge ontheoutside coating, and Igetsparkling between thespangles. Itisnotsobrilliant byanymeansasitwouldbe if'adirect. charge were given totheinside coating. [Experiment shown with thegasturned down.] Itisacharge given toone body, andacorrespondingly induced charge upon thesurrounding walls. Professor J.Penny: Isthere aninside coating? Mr. J.Winsnurst: Yes, butitisquite insulated from the suspended tinfoil. ‘The sparkling isnot very brilliant,butitis enoughtoshow. The Prestext: Will you tell ushow thebottom oftheout- side coating isconnected. Mr.J.Wixsnunst: Itisconnected toonepole ofthe machine, andthesheet oftinfoil totheother pole; butthesheet oftinfoil isinsulated, theairspace being quite sufficient fr that purpose. Iwould wish toadd that Ithink Ihave shown that Professir Lodge hasstated inhisinteresting paper views which probably le may reconsider, andwhich Ihave brought under your notice by these experiments, For instance, hestates, “With reference to “the point, itsprotective virtue somuch insisted onhytheolder 1889.) DISCUSSION. sor “electricians isentirely non-existent.” Ihaveshown, Ithink, Xo. that this isinaccurate. He then states that ‘the orthodox rule “ofconnecting allpieces ofmetal tothelightning conductor “requires modification;”andhesuggests“connectallpiecesof metaltoeachotherandtotheearth,butnottotheconductor.” Here, again, the modification iscertainly worse than theorthodox rule; andonmany ofhisother suggestions grave questions may bemised, Forthese reasons, andaswevalue thesafety ofourbuildings, itseems tometobehighly desirable that thesubject should receive much more examination before the influence of this Institution isdirected tothe abolition ofthe valuable rules Inid down bythe older electricians sitting atthe Lightning Rod Conference. Tay saythat anypart ofthis apparatus isattheservice of anyother speaker. The Prestent: Ihope that asProfessor Hughes was a member oftheLigtning Rod Conference hewill letushave his tiers onthe matters under discussion. Professor D.E,HuoHes: Dr.Lodge hasbrought before Frtamor usamost interesting and remarkable paper, and itisapaper Thich invites criticism. Now, whilst Iadmire the beauty aftheexperiments whichhehasshown,Idisagreewithsomeofhisstatements and conclusions. Iquite agree with the first Portion ofparagraph 3,where hesays: “There are,therefore, two “main cases—(a) When thestraininthedielectricneartheearth “has been ofgradual growth,inwhichcasethepathofdischargewill “bepreparedinductively beforehand;(3)whenthestrainarises “sosuddenly that there isnotime forany pre-arranged path. “The first Icall¢steady strain;’thesecond,‘impulsiverush.’It “inmost.important torecognise thesetwocases,andtounderstand“the extremely different conditions attending thetwo.” ButI cannot agree with theconcluding portion ofthisparagraph, where hesays: ‘The firstcase only wasever contemplated bytheolder “electricians;infact,sofarasIknow,itwasmyexperiments “last year which first called attention totheother case.” IdonotthinkthatDr.Lodgewasthefirsttocallattention to an LIGHTNING, LIGHTNING CONDUCTORS, Ere. [May 10th, Nala theeffects ofanimpulsive rush, andthisstatement hardly does justice tothelabours ofthe older electricians, forever since the discovery oftheLeyden jarithasbeen employed forthis object. SirW.Snow Harri, inhiswork on“Lightning Conductors, 1830,” inspeaking about the Leyden jar,said: “The year 1752, which “marks animportant erainelectrical science, from thecelebrated “discovery oftheprinciple ofaccumulation just mentioned, gave “to the natural philosopher aneasy method ofconcentrating “large quantities ofelectricity produced byartificial means, soas todischarge itupon orthrough bodies with aninstantaneous “and violent explosion.” ‘Thus herecognised theimportance ofstudying theeffects of ‘animpulsive rush, theresults ofwhich ledhim tomost important conclusions, bothastothebestmaterialandformofconductor foralightning-rod—his conclusions beingthatcoppertapehad" practical andtheoretical advantages over rods ofiron, Ifsubsequent experimenters failed toverify the conclusions ofHarris, itwasprobably due totheemployment bythem ofa toofeeble electric charge where theimpulsive rush was not sufficient toreproduce theeffects that hehad noticed. Professor Guillemin and myself notonly verified and reproduced Harris’ results in1864,but,inaddition, observed anextraordinary result |duetoanextremely highandsuddendischarge through copperwires,andastheresultswereidentical withthosecitedbyDr.| Lodge inhischapter on“Lightning Protectors,” andasIcalled| hisattention lastyear tothese results, Idothink thatheought | tohave mentioned ournames andReport published in1865, | insteadof(byhissilenceasregardsourpreviouslabours)allowing | readers ofthispaper toinfer that hewasthefirst toobserve these effects. Intheyear 1864, Professor Guillemin and myself, asmembers oftheCommission dePerfectionment oftheFrench Telegraph Administration, were charged with themission oftesting the comparative merits ofthe lightning protectors then used upon their lines fortheprotection ofthetelegraph instruments from lightning. ‘These experiments were carried outatthelaboratory oftheEcole deSt.Cyr, ofwhich Professor Guillemin was the Professeur dePhysique. 90 DISCUSSION. ry Ourideawastoemployanextremely powerful Leydenjarfrtemerdischarge, orspark, brought into action assuddenly aspossible through: analternative path arrangement, onebranch ofwhich contained the electro-magnet, orafine ironwire that wewished toprotect, and thesecond branch asashunt around theelectro- magnet orfine iron wire, inwhich wecould introduce thepro- tector, whose duty wastoallow acurrent ofhigh tension, such as lightning, tochoose thispath andthus protect theelectro-magnet ofthe telegraph instrument. ‘The general idea ofourexperimental arrangement wasthe following :— .“ — 8 YS e . y ls 1 4 Inorder tohave apowerful source ofelectricity athigh tension we made use ofalargeLeydenjarbattery, A,ofsixlarge jars, having atotal condensing surface ofabout 1square metre, ‘andaswefoundittooktoolongtochargethesewiththeordinary electrical machines, weused apowerful Rhumkorff coilgiving» spark of30centimetres, Thejarswere charged toafixed degree through thewires GandJ,which were removed before discharg- ing thejar. Afixed charge wasgiven tothejarsandmeasured bythe electrometer, H,and could bedischarged through the apparatus bymeans oftheuniversal discharger, B.Aninsulated rod ofbrass, CDEF,served toconduct thecharge totheexperi- mental protector atE,andalso totheelectro-magnet orfine iron wire introduced atFtothereturn wire orplate I. With this arrangement, and using ourhighest power, weat ence found this curious result, viz, that wecould not protect our electro-magnet orfine iron wire atFfrom being burnt, notwithstanding that weintroduced the best known form of ay LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May162, Heegt protector asashunt atE.Wethenjoined Emetaliallybys copper wire, and failing still toprotect: Fwejoined Cdirect to earth, orthe return plate I,byashort thick copper wireofwe centimétrediameter.Evenintheseconditions,whereashort thick copper return wire wasinadvance oftheelectro-magut, there would still besufficient current pass through theshunted electro.magnet toentirely destroy it.Now, here wehadevident amore sudden charge orimpulsive rush than occurs generallyia practice. Sothere seems tobemuch truth inthe remarks made byMr.Preece and Professor Fitzgerald, that adischarge from cloud isvery different and much slower than that from ametallic condenser. Ontheother hand, there arenodoubt sudden light- ningdischarges observed onourtelegraph lineswhichnotonly equals butfarsurpasses anything wehave yetseen inlaboratory experiments. ‘Mr. Frank 1.Pope, inadiscussion upon mypaper upon Sel induction, in1886,% mentioned that itwas awell-ascertained fact. intheUnited States that their electro-magnets wereotea destroyed during lightning storms, notwithstanding thatthey hhad joined their wires direct toearth byashort wire inadvance oftheelectro-magnet. ‘Thus verifying inpracticetheresultsthat wehadexperimentally obtained. Inorder totest the comparative value ofdifferent pr- tectors, wediminished the charge toapoint where amarked difference intheir behaviour wasshown. The results were given inareport which Imade asreporter, and published inthe Annales Telegraphique, 1865 (tome vii., pp.290to302), inwhich werecommended (inaddition tothe use ofashunt protector) theinsertion ofashort length offineiron wire (bobine preseraa- trice) inthe main line, the fusion ofwhich would actas cut-out (similar in function to those now used in electric lighting), andthus inallcases preserve theelectro-magnet, Having been called toRussia, Iwasunable tocontinue thee experiments; butProfessor Guillemin, having taken »great interest inthem, hecontinued them, and published hisresults *TournahoftheBeceyofTelgraphBnginaers,1886,Vol.15,p41. 1880) DISCUSSION. 496 intheComptesRendus,1866.Inthispaperhedemonstrated, frteworbyexperiments, theadvantage of sheet conductor over that of 4cirenlar section, and thus verified the conclusion ofSirW. Snow Harris. And healso showed that Ohm’s lawforsteady currents does nothold good forLeyden jardischarges during thevariable period, Ibelieve that theeffects observed bySirW.Snow Harris, Professor Guillemin, myself, and, later, byDr.Lodge, can be explained byself-induetion;anditiscurioushowverynearly theearlier experimenters perceived this. Until lately Ibelieved (through reading somes abstracts only ofHarris's results) that hebelieved that theadvantage ofsheet-conductors wasdueto their greater electrostatic capacity. This isanerror, forinhis work onthe“Nature ofThunderstorms,” 1843, anabstract of hich isgiven inthe“Report oftheLightning-rod Conference,” Page 86, hesaid:—“The beneficial effect ofsuperficial con- “ductors appears todepend ontheremoval oftheelectrical “particles further outofthesphere ofeach other's influence.” Professor Guillemin, 1866, said:—“This phenomenon arises “apparently from theinductive action ofconductors upon each “other. Increasing their surface facilitates thedischarge by “increasing thedistance between mutually opposing forces.” Inmypaper upon Self-induction, 1886, Isaid :—“It iswell “known that currents inseparate portions ofthesame wire (as “inacoil) react upon each other, and Ifeltconvinced from the “preceding experiment that self-induction isentirely due to “similar electro-magnetic reactions between contiguous portions “of the current initsown wire. Let usassume that anelectric “current consists pfabundle oranalmost infinite number of parallel currents, the limit being asingle line ofconsecutive “molecules; then each line ofcurrent should byitselectro- magnetic action react oneach oftheothers similarly towires “conveying separate portions ofthe current, and the self “induction should beatitsmaximum when the lines are inthe “closest, possible proximity, asinaconductor ofcircular section, “and farlesswhen separated, asinoneofribbon form.” Dr.Lodgesaysinhispaper,paragraph F,page412:—“The 6 LIGHTNING, LIGHTNING CONDUCTORS, Ere.(Alay160, Eteal“electrical disturbance isconveyed toaconductor through the“ether orspace surrounding it;expressed more simply— “lightning currents make useoftheperiphery ofaconductor “only, and sothemore surface itexposes thebetter. Better “than asingle rodortape isanumber ofseparate lengths of “wire, each thick enough not tobeeasily melted, andwell “separated, soasnottointerfere with each other bymutul “induction.” Idonotquite agree with Dr.Lodge’s theoretical view, but Ijudge from theabove that headvocates atape form inyr- ference toasingle conductor ofcircular section; hedoes vot, however, attribute much importance tothis, ashesays farther on,paragraph 56:—“The shape ofcross-section isbutlitle “matter. Flat ribbon basaslight advantage over round ri, “butnotenough tooverride questions ofconvenience.” ‘This does not fully agree with theviews ofSirW.Soo Harris, Professor Guillemin, nor myself—the first two having demonstrated with Leyden jardischarges avery great advantage intheuseofflatribbon, and Idemonstrated the same in1886 forvoltaic currents—the difference incopper being 50percet infavour ofthe flat ribbon form;andinironstillgreater, for here aflatstrip ofiron was 1,000 percent. superior toarodof the same ohmic resistance, but ofcircular section, Asourtimeislimited Iwillnotenterintothequestion of copper versus iron. ‘The Prestoest: You have experimented sovery much upon theimpulsive currentincopperandiniron’conductors ofvariousshapes,thatIshouldmuchliketohearyourremarks uponthat part ofthepaper. Professor D,E.Huanes: Asregards thechoice ofmetal, it ‘aquestion ofrelative costandconvenience, forIregard bothins ‘andcopper axvery suitable, provided wediminish theenormos: self-induction ofaniron rodbyemploying ironintheforme tape, or,better still, innumerous separated smaller wires whoe totalohmicresistance isnogreaterthanthatof«solidrod.Leaving aside thequestion ofrelative costandconvenience, |0notagreewithDr.Logewhereherays(pararraph55)“ine 1690.) DISCUSSION, wr hasadvantages overeveryothermetal.”NoreanIagreewithfrienthetheory ofOliver Heaviside, ifitleads tothe conclusion that iron fortelephone wires isbetter than copper. The Presipent:Hewasveryguardedinthestatement,and saidthat anarrangement ofiron wassuggested which might be animprovement, buttheordinary iron wire wasrecognised as being very inferior inpractical conductivity tocopper wite in regard tomagnetisation. Professor D,E.HucHes: Iamvery glad tohear this, forI believed that Oliver Heaviside's theory agreed with Dr.Lodge's statement astotheadvantages ofiron, Franklin recommended ironashaving «higher fusing point. Ibelieve, however, that this isnotimportant, forany lightning charge that issufficient tofusecopper would bedangerously near thefusing point ofiron; besides, iron when fused byasudden electric current does sowith explosive force, sending offpieces which burn with violence, thus increasing thedanger from fire. With copper itisdifferent, and Dr.Lodge himself quotes an example from thereport oftheLightning-rod Conference (p.195), where asmall copper bell-wire protected ahouse inMonte Video, theflag-staff ofwhich was shattered, thecopper wire melted without anyfarther harm. Dr.Lodge hasmade some experiments, quoted inhislecture totheSociety ofArts, inwhich heseemed todemonstrate that iron(when tested byhispeculiar method)wasbetterthancopper. ‘Now,astheseresultswereopposedtoallourpreviousexperiments, wehad toexamine closely the conditions ofthe experiments. Many who have examined and repeated these experiments find that theconditions oftheexperiments arenot those which occur inalightning discharge. Ipointed outtoDr.Lodge that Ididnotbelieve that theB spark (upon which hedepended forhisproof) wassimultaneous with that oftheAspark. Mr. Acheson, intheUnited States, madeaseriesofbrilliantexperiments onthissubject,inwhich heproved that theBspark wasnotsimultaneous with that ofA, andthat theBspark wasdue toanextra current. Mr.Preece hasablycriticised Dr.Lodge’s method, andproves thatthecharge 498 LIGHTNING, ZIGHTNING CONDUCTORS, Ere.[Bay1b, frotwer oftheLeyden jarswere notthesame indifferent trials. Mr. ‘Wimshurst hasshown this evening, inhisvery beautiful experi ments, that the direction ofthe current influences theresult. ‘ThereforeIdonotadmitthatDr.Lodge'sexperiments onthis subject prove anything, except that byavery complicated arrangement ofLeyden jars andcireuits you canprove thatin isbetter than copper, andalso acomparatively badconductor delaying acurrent isbetter than agood one that allows the charge tobeatonce conducted toearth. Thave tried toobtain some direct. proof ofthesuperiority of copper with Leyden jardischarges, asIhave already proved by the voltaic current;butasyetIhavenomethod,exceptsome equally open toobjection asthatofDr.Lodge's; butIhave found oneresult which seems tometobehopeful. Weknowthatthecomparative highself-induction inironi duetocircular magnetism, soItook asoft ironrod, which mi previously freefrom circular magnetism, anddischarged aLeyden Jarthrough it,therodlying horizontal eastandwest. After the discharge itwasagain tested andfound tohave strong tracesof remaining circular magnetism;s0Ithinkthisahopefulproof that there issufficient time during adischarge toproduce circular magnetism inaniron wire. Perhaps itmaybeofsome interest formetodescribe briefs mymethod oftesting forcircular magnetism, asitisextremely difficulttofindifanironwirehascircularmagnetism ornot,2° itshows noevidence ofexternal magnetism, andourmagneto meters failtoindicate themagnetic condition ofapiece ofiron inwhich there isaperfect closed circuit ofinternal magnetism. Butifweseek fortheWiedemann effect, byputting theaps ently neutral wire under elastic torsion, wemay, ifthecircular magnetism issufficiently strong, observe theeffect onanontinary magnetometer. Forfeeble effects I'am obliged touseanapparstat invented bymyself, anddescribed inmypaper totheRoysl Society,* inwhich aninsulated coilofwire surrounds thewire tobetested, thecoil being connected with atelephone aud *Proc,Roy.See,1881,.526;1885,p.6. 1880 DISCUBSION. a theotome. When elastic torsion isapplied tothewire, induced ffmien” currents areproduced inthe coil, which wecan plainly hear in thetelephone; the effect, however, does not last very long,as repeated torsions destroy thecircular magnetism. ‘The effect can berendered continuous bypassing aweak voltaic current through thewire, inwhich case wehear the induced current, produced by thereaction ofthecircular magnetism produced bythepassage ofthe voltaic current. Ifwelook tothepractical results oflong-distance telephony andtelegraphy, wefind universal testimony infavour ofcopper inplace ofiron, They have proved inthe United States that they areenabled towork the telephone atadistance of1,000 miles through copper wire, when the same results cannot be obtained through more than 200miles ofiron wire. They find the same difference all over the Continent. Sir W. Thomson, inhisPresidential Address this year, cites aletter from Mr. Bennet also proving the same fact. Asregards telegraphy, Mr,W.H.Preece conclusively proved, inhisremarkable paper, that hecould more than double the speed oftheWheatstone instrument bytheuseofcopper wire, Taking allthese facts into consideration, Ishould certainly prefer copper, notonly foratelegraph and telephone linebut also asthe best material forthe conduction oflightning dis- charges. IfIhad tochoose alightning conductor formyown house andwished tofeel perfectly safe, independent ofanytheoretical views,Ishouldfeelthatbythechoiceofacoppertapeinplace ofarodtherecouldnotbeanytheoreticalorpracticalobjection toitsuse; andasregards choice between copper and iron, as ironcannever have lessself-induction than copper, Ishould feel perfectly safe inchoosing acopper tape inplace ofaniron rod, thisbeing myideal perfect lightning conductor. Mr.G.J.Symons, F.R.S.: Idonotrise, Sir,tospeak onthewr.symoos electrical matters inthis paper, asIshould behopelessly outof placeindoingso,butsimplyasthe—amanuensis, shallIputit, *Brivish Amocintion, September, 1885 800 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[lay1h, 4.6;men6 oftheConference which satsome tenyears ago upon thisquestion oflightning-rods; and theremarks Ihave tomake arerather historical and critical than electrical. Ihope, however, thatyoo will bear with mealittle, because Ithink that Dr.Lodge hasnot realisedwhatIwasgoingtoexpressastheterriblemischiefwhichhemaybedoing.Thatisastrongexpression touse,butatthe same time theamount invested inlightning conductors through- outtheworld must amount toseveral millions sterling; every public building, whether itisthe Houses ofParliament, West- minster Abbey, orwhatever itmay be,isprotected bylightning ‘conductors, andwearetold that allthese lightning conductors ar wronginprineiple andaredangerous. ‘Thatisaveryalarmingthingasregardsbothlifeandproperty, andIventuretothinkthat astatementofthatsortshouldnotbeallowedtopasswithout very severe criticisin, Ofcourse ifthere canbeanimprovement ofanykindinsuchanessential andimportant apparatus, itwoul bewrong foranyonetostand upagainst it,butontheonehand wehave therecords ofthebehaviour ofconductors during mor than acentury ;ontheother hand, laboratory experiments. Ipersonally feelvery;doubtful whether laboratory experiment: sufficiently represent reallightning ;but,bethatasitmay,Iam notgoingtosayanything whatever aboutthem,notbeingcom petent todo0. Iwillnow refer seriatin tosome points inDr.Lodge's paper. Intheearly part ofhispaper herefers tothesystem which ascarriedoutattheHoteldeVille,Brussels, andspeaksofthatinthese terms: “Imaysay,however, that Ihave seen nooccasion “since toimportantly modify any ofthem, andthat myposition “inrespecttothesepracticalhintsisstrongernowthanitwasthen,“inasmuch asIfind that thelarge experience oftheBelgian “electricians, who have given more attention tothesubject thin “perhaps anyother school, hasledthem torecommend, inrecent «years, almost precisely thesame methods ofprotection asthoseI “advocated ;theirviewsbeingacquiredalmostentirelybypractical «experience, andscarcely atallbytheory.” Inthefirstplace there was,Ibelieve,onlyoneman(andhehasunfortunately passed away) inBelgium, oftheBelgianelectricians whoheldtheseviews 1889) DISUUBSION. sot —that wasM.Melsens, who was consulted with reference totheM#-S:ma Town Hall atBrassels, andwho putupthe most wonderful bird- cage ofconductors that has ever been put up. Dr.Lodge tells usthat that building hasbeen damaged bylightning! That does not sayvery much forthe system which Dr.Lodge himself says isvery nearly carrying outhisown views. ‘Then with reference totheterminals. Iamvery much surprised tohear this question asbetween knobs, points, and balls coming upagain, because itwas allthrashed outabout 120years ago. There was alarge powder magazine down at Purfleet which wasdamaged bylightning, and acommittee of theRoyal Society wasappointed toinvestigate what had been done and why ithappened, and they examined Priestley, Hindley, andanumber ofthefirst electricians ofthat dayupon thisvery question ofknobs and points. Ofcourse, asisclearly stated inthe Lightning-rod Report, thepoint issupposed to discharge twoduties—and that isthereason why theConference recommended ablunt terminal surrounded byanumber ofsharp Points. ‘The blunt terminal wasintended toreceive adisruptive shock, ifsuch athing should happen tofallupon theconductor, andthe sharp points were supposed todissipate the electricity from theearth andsoreduce thepotential inthevicinity ofthe stroke. But surely, with respect toacloud thousands ofyards inarea,thedifference betweenapoint,sayyyofaninchinarea, andaball afoot inarea, vanishes. Professor Lodge refers inthe next place toprotection with respect tochimneys, and hehad agas-flame torepresent a achimney. There isnodoubt whatever that ahot-air current isavery excellent conductor, and forthat reason the Lightning- rodConference have always held (Iamnotquite sure whether itwasmentioned intheReport) that weshould look more tothe protection ofthe kitchen chimney than ofany other chimney inahouse, because the kitchen chimney istheonly one in summer which hasafireatthebottom ofitand hasany hot air coming from it.Butwhen hespeaks oftheContinental plan of ‘aeross-bar over thetopofafactory chimney, with aspike rising ‘upfromthat,beingbyfarthebestwayofprotection, Idonot soe LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May16 Me.Srmomagreewithhim,becauseIhavemyselfseenontheContinent & considerable number ofconductors ofthat sort which have gt outoforder inconsequence ofthehotfumes passing corroding theiron-work—they generally useiron—and ofcourse thewhole thing tumbles topieces. What theLightning-rod Conference re commended seems tome,with duerespect toProfessor Lodge, preferable system;andtheyrecommended thesystemproposed Jong agobyFaraday, that there should beaflatring onthetop ofthechimney, entirely surrounding it,but, further, that that ring should befurnished with anumber ofpoints that should do thesilent discharge business, while ifadisruptive discharge came itwould find the ring itself and have plenty ofmaterial upoa which toact. With respect tothequestion ofsize, Iwas rather sorryto hear what Mr. Preece said. Iknow that Mr. Preece’s authority isinfinitely beyond mine, and that hehas always held stroug «views astovery sinall wires being sufficient forcarrying offs flash oflightning. Ithink that possibly that isbecause the bulk ofhisexperience hasbeen with wires running down and protecting telegraph posts. Idonotknow atall,Ispeak quite subjecttocorrection, butitseemstomeaspossibléthat,inthepresence oftheverylarge mass ofwires carried bythese posts, disruptive charges, ifthey fallupon posts, areconsiderably diminished inintensity. Certainly theConference, when tbey were considering this subject, found that rods ofiron, and platinum even, ofconsiderable dimensions were melted. ‘Theove case which Iquote isthat oftheNew York packet struck by lightning intheGulf Stream inApril, 1827. Iamnowreading from Snow Harris's book,* and itsays—“ Inthiscase thedischarge “fell onapaintedironrodfourfeetlongandhalfaninchia“diameter; somefewinchesoftherodnearitspointweremelted “the linked iron chain which descended from this rod tothewater, “aboutaquarterofaninchindiameter,wasknockedinpiecesby “the expansive force oftheshock, and some ofthelinks fused. “The flash oflightning notonly melted some ofthelinks,but ‘+SnowHrs,«OntheNatureofThunderstorms,” 1843,p,108. 1880.) DISCUSSION. si “caused them toburn like ataper. ‘The melted ironfellinglow- ™-*mons “ingdropsonthedeck. ‘Theresultsofagreatnaturalexperiment “are here presented tous,andweseethat aniron rodofhalf an “inch indiameter effectually resisted aflash which fused achain “of about one half ofitsdimensions.” IntheLightning-rod Conference Report, page 223, several instances aregiven, but perhaps the most striking ofallwas that atCareassone, inthe south ofFrance, inwhich theconductor, seven-tenths ofaninch indiameter, wasfused. Assuch cases happen, Ithink that a No,5B.W.G. wire, which Ithink iswhat Professor Lodge re- commends, would very quickly gooutofthewayonbeing struck, andthen you would have the difficulty that the building would beleftentirely without protection. With respect toiron and copper, Ithink that those who will have thegoodness toperuse our Report will seethat we were bynomeans absolute astothepreference forcopper. The Conference didprefer copper, and they gave their reasons forso doing, butthey also gave iron—and ofcourse iron isvery largely used, especially inFrance, whére they goinvery freely forcon- ductors with much longer upper terminals than areused inthis country. . ‘Then wecome tothequestion ofacopper rod, andwhether a manmay ormay not grasp itwith impunity, which Mr.Preece saidhewould do, Iwas delighted tohear Mr.Wimshurst say that hethought that there would benodanger;thatismyown opinion, but, ofcourte, asIsaid before, myopinion onapurely electrical matter isworth nothing. Atthe end ofthe paper Dr. Lodge quotes anumber of instances from ourReport, butIconfess Idonotquite know why, becauseitisnotobviousthattheycontradict anyoftherecom- mendations that wehave made, Ofcourse there isthequestion ofprotected area;thatisaverydoubtfulsubject,butstill,there isthefactthat, hundreds ofinstances have been collected together ofstrokes falling mear conductors, but wehave never yetfound a ‘aseinwhich any damage hasbeen done within theprotected area asdefined byus,and, therefore, ifthat “protected area” be notabsolutely true itseems tome that there isvery strong oy LIGHTNING, LIGHTNING CONDUCTORS, Exe.[Say6, xe%ymoneevidenceinitsfavour.Atthebottomofslip13,Dr.LodgequotestheinstanceofachurchbeingslightlyinjuredonMay18tb,1877, atGarding, inGermany. Butitappears that they didnotfollos ourrules; they took the conductor down the north sideofthe steeple, not down the wet side aswe advised, Again, onthesame slipand under No. 43,Dr.Lodge raises question onastatement byProfessor Fitzgerald, atBath. He says: “Ilustrating «matter important, iftrue, euggedtel by «Profeanor Fitzyeratd, theoretically, atBath, notyetverifel by “me. Intwo cases the stroke broke the conductor atpints «where itsdirection wasabruptly changed.” ‘Thatisoneofourrules—never bendaconductor atasharp angle (eeReport, p.18). ‘The above statement isanillustration oftheresult ofnotobeying that rule, ‘Then with respect tothebrush discharges from theconductor and surging circuits inthe neighbourhood, hesays: “Two “electrical phenomena aretobenoted assometimes occuring “whenalightning-rod isstruck...peculiarnoiseisbei“likepouringwateronafire,andelectricsparksareemittedfrom “bodies near.” That isaquotation from apaper byDuprez, butIthink it would have been equally fairtothe subject tohave quoted tw orthree other words which are inthe same abstract (Repo. p-96), viz.,that M.Duprez “considers aproperly constructed “lightning-rod tobeaperfectsafeguard.” Asonehalfofthe story wasquoted, theother half may aswell beadded. Tnow come tothelastslip, and there wehave aseries ofwba! Dr.Lodgecallspracticalquestions. ‘TheyarefromNo,51onwardsandIhave gone carefully through them ascompared withou Report,andtheresultisthatnearlyallthoserules—Isuppor they might almost becalled—are inourReport, andareit ‘accordance withit,therefore Idonotquiteunderstand whatisthe ‘useofreprinting them. | Inconclusion, Ihope that insome way orother thi greatInstitution andtheverydistinguished menwhoareherr present will express something inthe way ofadistin opinion,sothatitshallnotgoforththatallthatbasbeendove 10803 Discussion. 305 ‘uptothepresenttime,ineverycountry throughout theglobe,Mr.smo,hasbeenwrong.Itiscertainly hardforthenewsystemthatat theonly large building towhich ithasbeen applied—the Town Hall atBrussels—it has failed. Lieut.-Col. R.¥.ARMSTRONG, RLE.: Itwillbereadily under- test: stood that the War Office takes akeen interest indiscussions on theefficiency andeconomy oflightning conductors, and, asIhavebadtodowiththematterinquestion formanyyears,Iwishto state afew facts connected with myexperience which may ormay notbeheld tobepertinent tothediscussion, but which Iwill state asbriefly aspossible, The Fortification Branch ofthe War Office have mainly depended forthe protection ofordinary magazines on(1)con- tinuous conductors, without joints asfaraspossible; when joints arenecessary they are ofvery low resistance tolow voltaic currents;(2)very®lowresistance earth;(3)sufficiency ofpoints; (4)andotherminordetails whichareinaccordance almost entirely with therecommendations oftheLightning-rod Conference. Whatever might have been supposed, from Professor Lodge's earlier lectures, tobehisviewson(1)and(2),hehasstatedin ‘the lecture now under discussion that he considers the measures inquestion tobevaluable, andhehasevengone¢ofarwstondmit that anordinary galvanometer may beusefully employed in testing the work. Itwill besatisfactory tosome ofustoknow that wearesupported bythe weight ofhisopinion onthese points. Itmay beinteresting tothemeeting, inconnection with thefacts I'am about tolaybefore them, ifIstate that think the War Department areprobably theonly body inthis country, andpossibly theonly people intheworld, whoreally do employ themethod oftesting theefficiency oftheir lightning conductors byvoltaic currents and galvanometers periodically. Without wishing foramoment toargue that absolute protection isguaranteed under theforegoing conditions bythe precautions weadopt, Ihave tostate that whether safety has heen attained bylow-resistance earth, jointless conductors, orby ‘Tha: i,afew ohne 06 LIGHTNING, LIGHTNING CONDUCTORS, Bre.(ass184, ye-ca other preeautions taken, ormerely byluck,tothebestofmy belief, imnocase where theabove principles have been carriedoat hasaconductor failed. Ontheother hand, Iknowofcaseswhere conductors, defective when judged bytheforegoing standard, haw failed. AtSlough Fort, ontheThames, about eight years ago«small tower, towhich aconductor wasattached, wasstruck anderickel bylightning. How theportion ofthedischarge which struckthe tower gottoearth wasnotclear. ‘Theconductor wasnotinjure. Itsearth resistance was about 200 ohms, otherwise itwisin acconl with therules oftheLightning-rod Conference. Again, about five years ago thecontents ofasmall cil magazine orshifting-room intheLake District were exploded by lightning. ‘The earth oftheconductor had been dug upbere Thad theopportunity ofhaving ittested. ‘The resistanceof1 similar earth onanadjacent magazine was, however, found tobe between100and200ohms,andthepointoftheconduetor ofthemagazine wheretheaccidentoccurreddidnottestgoodcontimits with low volts, At the Chichester Cathedral the earth resistance was over 100ohms,andpartofthelightning leftitforanotherearthof between 100and 200ohms inconnection with alarge metalic surface. Inanother case which Iexamined, where lightning west down thechimney ofadwelling-house, theearth resistance of suljacent conductor was over 100 ohms. Nodoubt many members oftheInstitution will beacquainted with cases where lightning conductors have been burnt upby lightning atriveted joints. Such acase occurred toaconductor onthecitadel atMalta, about seven years ago, butthelightning didnoother damage inthis case. Another instance ofthis sort occurred more recently, als? broad, theconductor being fused ueLow thepoint atwhich par ofthecurrent leftitthrough astaple inthewall. Theportion ofthecurrent which lefttheconductor inthis case ignited some small-arm cartridges, andthese cartridges wert inmetal cases, sothat apparently thecurrent, orelectrical action wasnotentirely onthe“differential outer skin.” 1990 DISCUSSION. 0 Iwould farther observe that, inthecases given byProfessor Lsvt-Co. Ledge, noinstance occurs offailure ofgood earth. ‘The Gardner Church Conductor, which hequotes asdamaged in1887, was fractured in15places. This does not,therefore, appear tobe« caseoffailure ofearth, butofinefficiency ofconductor, probably owing todefective joints, which would have been revealed by testing. The statement, moreover, thattheearth wasfaultless, is vague. Twould wish tosubmit, with alldeference, that although Professor Lodge's experiments are magnificent, they arenot lightning ;andthat wemust bevery cautious indrawing definite conclusions concerning lightning discharges from laboratory ex- periments ormathematical calculations until further accurate experiments havebeenmade withtherealthing. Itwould bea tigjump toattempt tosay,with anyconfidence, that lightning coming from acloud, square miles inarea, through hundreds of feetofair,with energy capable ofdestroying substantial struc tures, will follow, even approximately, thelaws governing dis charges between glass jars connected toafew feet oftin, and points orflames afew inches off,the dimensions ofthe flames being considerable incomparison with their distance from the sheet tin. Inolden times theexperiments were made with real light- ning. Nodoubt some oftheexperimenters were expended ;but there areplenty ofelectricians who could deal with lightning and take care ofthemselves atthe same time. Mr.A.J.S. Apaws: The present occasion isopportune for%*.Aisns submitting some features relative toanapparent effect ofstress inconductors. Weknow that acurrent inpassing along a conductor has tocontend with the resistance proper tothe metals and toitsself-induction; but itwould seem that such a current hasalsoaproperty ofself-relief byaradiation ofenergy. ‘The idea ofself-relief isroughly drawn from the fact that increased energy intheconductor results inincreased radiation, Observe thesimple proposition that aconductor Aunder momen- tarystress, such asthat, produced byaninduction coil, exerts an influence upon anadjacent conductor B.Now replace B,bya 08 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[May16th, ‘fr.Adams, microphone, and weobtain from itasimilar indication ofA’s effect, together with thesuggestion that inthis case, atleast, portion ofthe energy radiated byApresents the form of mechanical force, Weremove the microphonic circuit and replace itwith a single telephone held totheeat;thesamefeatureofintercepted energy isapparent, andthefact that thetelephone coilisopen, | again indicates that theenergy intercepted isoneofmechanical | motion, These instrumental aids are not necessary, however, fortheradiated energy may bemade perceptible tothe unaided ear. Itwill be understood that the lines offorce from acircular conductor, suchasA,everwiden out,and,foragiven space, love | ineffect; but that byusing aconductor having aflat surface theforee lines willberendered parallel, with corresponding | accentuation ofeffect. Iattach apiece offlatmetal toone end | ofthesecondary wire ofthis induction coil, and theflatsurface, | acting axa kind ofelectric lens, focusses the force into parallel lines, and thetone ofthecuil-vibrator becomes audibly reproduced upon itssurface. You willalso observe that byplacing this piece ofmicauponthemetalplate,andthenlayingtheearclosetothe | ‘mica, thesound becomes fullandcrisp, [Handing themetal dise | tothePresident.] The Presivent: |canhear itvery well with this mica. Mr,A.J.S. Aas: Yes, Iprefer mica between the ear and themetal toprevent chance ofsparking. Wemay arrange a —Pinay } a | Fro. 1 lens ofstillgreater power byhaving two surfaces, asshown in Fig. 1,where theplates Aand Bareshown insection, Ihave 3999.) DISCUSSION. 109 heretwosuch plates, between which Iinsert from onetohalf-n- Ain dozen pieces of}-inch plate glass. The result isthat, whilst thesound reproduced upon theinner surfaces ofAand Bis clear and audible toallaround mewhen onlyoneslipofglass isused, theeffect decreases astheremainder oftheglass slips are inserted, widening, asthey do,the distance between Aand B. This mechanical effect ofsound wasbrought under thenotice ofthis Society, andanexplanation offered, bymyself inashort poper read inDecember, 1877. Iregret that time does notnow permit metoenter more fally into the subject generally,inorderthatImightdemon- stratehowthat,iftwoormorepiecesofplateglassbeinserted between the metal surfaces AandB,Fig. 2(where, although the spaces areshown large, they should inpractice bemere films by thepressure ofallthe surfaces one upon the other), whilst sound waves are radiated from the surfaces ofAand B;Liyht, similar tothat produced inordinary vacua, isemitted between theglass plates SS. ss || Fro. 2 nd also how that, byunequally magnetising theatmosphere upon the one side ofalooped orloosely-hung conductor as against the other side, theconductor itself may becaused to bodily vibrate with each pulsation offorce. What Idowish nowtosuggest isthefar-reaching effect oftheinvisible radia- tion referred to,and the apparent focussing power offlat surfaces—of aprobably highly dangerous character intheease oflightning. Inorder toillustrate this, and aspointing toanexplanation Vou. XVI. 36 610 LIGHTNING, LIGHTNING CONDUCTORS, Ere, [May 16th, 4Adams.oftheincident mentioned inProfessor Lodge’s paperanentthe isolated iron barsparking andcausing combustion, Iquote the following experiment. Initthelens Aisconnected toone end een ») Fe.3. ofthe secondary coil. Bisplaced facing A,but isisolated. Nevertheless, the energy radiated from Aupon Bwill spark off from Btothewire W,ortotheearth—a result possessing serious significance inthecase oflightning. / i i \ Fos Soalsothisradiative aspect ofthequestion seems tooffer an explanation ofthemarine incident referred toatourlastmeeting, ‘asatFig. 4.Here wehave themetalled topmast insulated from 1899) DISCUSSION, su thelowerpartsbyacoatofpaint,atY.‘Thetopmastisstruck,Mr.Adams and,during thefierce radiation from surface tosurface atY,the flash itself will have traversed the shrouds tothe vessel's side, where one part will have gone overboard and sinother will have fallen back upon thecomparatively free metal Z,atwhich point, 1great concussion oftheradiated andtheconducted forces would seem tohave occurred. Itisnotimprobable that this radiation ofenergy will be strongly anddangerously developed intheflash itself during its pestage through theair,and afterwards inthe.conductor under stress, and this feature may perhaps gofartoexplain many of theeffects atpresent mainly accounted forbyside-flashes antl surgings. Another point, asgiving emphasis tothe possible effects of fierce electrical radiation, suggests itself inconnection with so- called “lightning photography.” Here isaninstance. “Asea- manwassitting atthefoot ofamast towhich hadbeen nailed ahorse-shoe, The man waskilled byaflash which struck the mast, and upon hisbody wasfound amark, apparently produced hytheflash, corresponding with thehorse-shoe, ‘ThePrestpext: Wasitknockedoffthemast? Mr. A.J.8.Apavs: Iunderstand not. The incident ix mentioned inoneoftheJournals ofthis Institution, and Irefer toitbecause itseems toemphasise thisidea ofradiated energy, andbecanse thefacts Ihave demonstrated inrespect ofradiation ‘andfocussing would appear torender such anoccurrence atleast 8possibility;andastothegeneralquestionoflightningmarks, which have borne resemblance tosome adjacent body, Iventure tothinkthattherecorded instances aretoonumerous andcircum- stantial forthe subject tobeeasily brushed aside asanassured absurdity.* Thave already referred tothephenomenon oflight emission intheairfilm lying between theglass slides atFig. 2,afeature Thave notseen noticed elsewhere, although itisonenotalittle remarkable from the fact that the periphery ofthe airfilm is +Nature, May 6th, 1875. ae LIGHTNING, LIGHTNING CONDUCTORS, Ere, [May 1, Mr.Adama, open tothe surrounding atmosphere. Now, what wecando artificially issurely within thecapacity ofNature, andhence it seemstomethatwehaveinthisphenomenon theapproachto anexplanation ofglobular lightning, inthat whilst theclouds andearth may constitute theinciting discs, and theatmosphere theinsulating plates, apatch ofcomparatively rarified airmy become thenucleus ofsuch anelectric cloud orglobe. Inrespect tothis feature ofradiation, there isamarked difference intheradiative eapacity ofsome metals, as,forexample, inthecase ofsamples ofNo. 11hand copper, hand brass,and hand-drawniron.Ifindthat,aizeforwizeandstressforaves, iyon shows 25percent. greater radiative capacity than copper, and brass slightly more than iron. Sothat, supposing theides ofself-relief topossess anyvalue, copper would appear tobe,size forsize, themost suitable ofthethree forlightning conduction. Yoeawn, MrC.E,SracNourrni: Tmightmentionacaseortrobearing onthis subject. Having tomaintain very many thousands ofmiles oftelegraph wires, Ihave agood opportunity oflearning the effects oflightning upon them. ‘The wiresat struck very frequently, but areseldom fused. ‘The instruments attached tothem, onthe other hand, suffer tosome exert. ‘Aninstance occurred inNorth Wales, where the wire attached to ‘abellwas struck andthecoilofthe bell was split like aeros, it four, and the whole ofthe wooden case was lined like awit brush, bythesmall pieces ofwire, showing that, although the flash must have been avery strong one, theline-wire wasnott allaffected, which was agalvanised wire ofNo. 8or11gauge These line-wires arecapable oftaking large currents. Another case occurred atShrewsbury, where aman was upamong the wires onapole one afternoon and was struck bylightaing severely. ‘The lightning struck him, andapparently entered bis body underneath onearmandpassed outofhisleg. Iinguir! why aman ofhisexperience should have been upamong the wires during athunderstorm, and the reply was that therew# nostorm atShrewsbury, itbeing amild and calm evening,bt Tlearned that there wasaheavy thunderstorm atHerefors thetime, and that thecurrent, which wasultimately thecae 1889.) DISOUSSION. os theman'sdeath,wascarriedfromHereford toShrewsbury, ¥r4distance of50miles, bythe line-wire—No. 8galvanised iron. Onanother occasion, atReading, aNo.4gauge galvanised iron wire was fused (and that istheonly case where Ihave known ofaNo.4wire having been fused), andtheNo.16copper wires, covered gutta percha, attached toitwere melted, andthe gutta percha ranover theenils and sealed them. Ishould like toaskwhether aflash oflightning hasbeen proved tobeoscillatory. From what Professor Hughes stated tonight itdoes notappear tobeso. Mr.Preece gave several examples, oneshowing thattheactionofaflashuponapolarised relay wasacontinuous lineontheprinting instrument, therefore itdoesnotlookasifitwereoscillatory. Ihaveseveraltimes tried with hand magneto machines totest thecurrents they give with agalvanometer, but could get scarcely any motion ofthe needle; and the faster Iturned theless was the motion, There- fore,Ithink, iflightning isoscillatory thepolarised relay would have either caused adotted line, or,ifthe vibrations were so pid astoprevent thetongue nottouching thecontact pins, no mark would have been made. ThePresioenr: Thetimeisratheradvanced,butTamsuresewiiam ‘rewould allsittillmidnight rather than losesome ofthestate- mentsoffactthatwehavejustheard;andIwouldinviteany. gentleman present who cancontribute some facts that would help tothrow light onthissubject todoso. ‘Mr.Sypsey Evens: Icameherethisevening intending ¥ tomake some rather deprecatory remarks onthefutile nature of thediscussion atthe last meeting, but Mr.Wimshurst’s experi- ments have putthose remarks aside, because Ifeel certain we must have alllearned agood deal from thebeautiful experiments showing that side sparks from aconductor will not charge a Leyden jar. ‘The first thing that Inoted forremark wasamost important question which hasbeen leftentirely untouched byevery speaker, namely, doBsparks, which occur inlaboratory experiments, occur innature? and isthere any evidence ofsuch sparks striking lightning conductors ? Manan, MsPreece occupied some time atthelastmeeting intrying toprove that there arenosuch cases, butIdonotthink he brought anyvery good argument tobear onthepoint. During 1877 Icarried outagreat many observations onatmospheric electricity bymeans ofaquadrant electrometer, using agold-leaf electroseope when theelectrometer proved toosensitive ;and,with your permission, Iwilloccupy thefewremaining minutes in describing what seems tomeanimportant observation thatbears ‘onthis question ofBsparks. One ofmyearliest observations was with agold-leaf electroscope during athunderstorm taking placesomemilesawayfromwhereIwas.Iwasonthesideof« hill(marked Binthesketch), and athunderstorm wastaking —— place some fivemiles away attheother endofalong cloud, which ‘wasapparently continuous between mypoint ofobservation and thethunderstorm atA.About sixfeetfrom theground Iheld gold-leaf electroscope, and, before adischarge took place at:A,the leaves would befrom }to}inch divergent; theinstantaspark tookplacetheleaveswouldjumpoutthreeorfourtimestheir former divergence. Now there youhave acaseofsudden increase ofpotential which, ifgreat enough, willcause aBspark. This wasavery striking observation, andonewhich Ihave repested sinceinsimilarcases,butneverwithsuchagreatlengthofcloud. Afterwards, when Igotmyquadrant electrometer towork, IWas often troubled by thunderclouds causing the light spot to disappear very suddenly from the scale. Agreat point made byMr,Preece atthelastmeeting was3totheoscillatory character oflightning, andheproduced areco toconvince usthat lightning intelegraph instruments could nt beoscillatory. Professor Lodge, inhispaper, does not:saythatit 180) DISCUSSION. a5 in0;hedescribes theconditions foranoscillatory discharge, but,¥«. s0farasonecansee,theseconditions donotobtaininatelegraph instrument, and, even ifthey did,weshould getthese records on theBain oranyother telegraph instrument, because itwould not beam oscillation ofequal currents inopposite directions, butone inwhich each impulse islessthan thelast—in other words, a gradual dying away. This being so,theexperiments described byMr.Preece have nobearing onthequestion atissue. Mr.Wimshurst’s experiments this evening with theballand Point,orsmallballandlargeball,remindmeofaninteresting result obtained byH.Yeates, in1876. Hehadalarge induction coilconnected with twosmall Leyden jars, each jarhaving apair ofterminals, apoint andaball, theballofonejarbeing opposite thepoint oftheother. ‘The path chosen bythedischarges from thejarsthen depended onthedirection oftheprimary current, andthemomentthiswasreversed thesparkswouldchangefromonepairofterminals totheother. That isamore striking illustration than theoneMr,Wims- hurst showed us,because points areused instead ofsmall balls. Ofcourse, itdoes notbear somuch onthequestion raised byMr. Wimshurst, where the difference was whether the cloud was Positive ornegative. BeforeIsitdownIshouldliketothankDr.Lodge,onbebalf oftheyounger members ofthis Institution, forhiswork during thelasttwoorthree years. Inthese days, when weareinundated ‘ithpapers andbooks inwhich amodicum offactisburied under ‘mountain ofmathematics, hisadmirable way ofputting ideas intoourheads isvery refreshing. Mr. W.H.Prezce: Before you close this diseussion, Sir,srrecs Perhaps you will allow metosubmit, fortheinspection ofthe member, three orfour examples indicating fusion bylightning. Thereisoneinparticular—it happenedonlyonMay11thlast, near Leeds—which gives unmistakable evidence offusion inthe interior ofthecopper wire. The outer section hasevidently been cooler, and thehotmetal inside hasburst itself through theskin, There arealso twoorthree specimens ofiron wire andgutta percha-covered copper wire that give much thesame indication, 318 LIGHTNING, LIGHTNING CONDUCTORS, Bre.[May16, urrwce. {should likejust tocorrect Mr.Evershed, who referred tomy liking forancient history, and toinform him that thecloud experiments which hemade only twelve years ago were also made byLord Mahon in1789—one hundred years ago! Sewiutan ThePuesipeNt: [think wemust now considerthediscussionasclosed.IamquitesureweallagreethatDr.Lodgehasdone exceedingly good service inhaving raised thequestion inthe manner inwhich hehasraised it,and inhaving brought intothe discussion ofthetheory andpractice oflightning conductors some very important scientific principles that had notbeen fully taken intoaccount bythose whopreceded him inthesubject. Ithink wealladmitthattheprinciple ofself-induction hadnotbeen sufficiently taken into account inconnection with the theory of lightning conductors and practical rules forsafety intheir use Tdonotknow whether Franklin had any consciousness whatever that there wassuch aquestion asthemutual influence ofcurrents inneighbouring conductors, orindifferent parts ofoneconductor, inrespect tothefacility afforded forcarrying away theelectricity bytheconductors. Itisquite clear that Snow Harris had some correct views onthesubject: wemust notstecept allhisviewsof electricity ascorrect;butmanyofusmustnowfeelthatinsome respects inwhich wethought him wrong—in whic, forty year ago, I,among many others, thought him wrong—hewasquite right. There isone thing inDr.Lodge's summary ofresalts (article 56)that Iconfess Icannot understand atall: “theshape “of thecross-section isnotofmuch importance.” ‘This seemsto bealtogether atvariance with hisown teaching onthesubjed. Snow Harris thought agreat deal ofsurface and shapeofera section. Inspeaking onthesubject attheBritish Associationa Bath, Ireferred tothelightning conductor set.upfifty yearsag? onthetower oftheoldGlasgow University buildings, under the recommendation ofSnow Harris. Itwas alarge tube ofcoppe andIwell remember being taught toconsider that that hadbees amistake, andthat thesame quantity ofcopper inasolidrodot wire rope would have been cheaper andjust equally effective. ‘Wethen thought Snow Harris wrong, andIbelieve thatFaradsy himself didnotperceive that Snow Harris was right intb! 980 DISCUSSION. sit iatter.Wenowknowthathewasright,andthatspreading sWitian thecopper over awide area iseven better than rolling itupthe same breadth intheform ofatube. Asheet ofcopper, wenow know, constitutes aconductive path forthe discharge from a lightning stroke much less impeded byself-induction than the same quantity ofcopper in@more condensed form, whether tubular orsolid. Now, astothe“practicalquestions”putforthbyDr.Lodge, [think there aresome valuable suggestions. No.72seems to meimportant: “The cheapest way ofprotecting anordinary “tiouse istoruncommon galvanised iron telegraph wire upall “the corners, along alltheridges and eaves, and over allthe “chimneys, taking them down tothe earth inseveral places, “and ateach place burying aload ofcoke.” The burying oftheloadofcokeistheheaviestpartofthebusiness, but themultiplying the mains byconnecting alarge number of comparatively small wires instead ofone close conductor docs seem tomeanimportant practical suggestion. Onthe other hand, hesays itisnouse connecting them towater pipes. That Icannot agree with atall. Onthecontrary, Iwould take these galvanised iron wirex described byDr.Lodge, andthemore ofthem the better, down allthe comers andwherever you can getthem, and connect every one ofthem toawater pipe. I Would farrather dothat than toaload ofcoke, itismore easily done; and Ithink that that isthebest wayofdoing itforthe Protection ofanordinary dwelling-house having water supplied to itinmany-branched metal pipes. Anordinary house can, I believe, bemade exceedingly safebythewater pipes. “Connectingaleadrooforothersuchexpansewithalight- “ning conductor isnotanunmixed good, foritvirtually increases “the dangerous proximity ofthelightning conductor.” Well, I ‘ould sayconnect allpieces ofmetal toeach other, and tothe earth ifyou can, butifyou cannot connect each ofthem toan ‘arth, connect them tothelightning conductor, andgive itagood earth, Ithink, onthewhole, that thespark coming from alight- tingconductor isnotoneofthemain sources ofdanger, although there isnodoubt thatDr.Lodge isperfectly right insaying that as LIGHTNING, LIGHTNING CONDUCTORS, Bre.(May163, SeWitlan there isaliability thatitmaylight gasorother combustible sub-stance.Thereisnodoubtwhatever butthatthemorecompletely thehouse canbecaged inthebetter; andforpowder magazines I believethatitisperfectlytruewhatDr.Lodgesays(andwhatIbave said myself), that theway tomakeapowdermagazineperfectly safe istocompletely enclose itiniron. Make acomplete ironhowse ofapowder magazine: line thefloors with wood orsoft:material to prevent ignition ofstray powder bypersons walking ontheflor; butletapowder magazine beaniron building with anironflor and then you donot need anearth. The powder should bekept well in,farenough from iron walls, floor, androol that noetheric spark canignite it.Whether onagranite rock orinaswamp itwould beequally safe: theneed fortheearthi absolutely done away with ifthemagazine iscompletely enclose! bymetal. Inthat case Isuppose ironwould bethebestmetal although itwould berash tosay,seeing howvery difficult isthe subject oftheimpulsive current iniron. Remembering Professor Hughes’ experiments and illustrations, and the mathematic theory worked ont somagnificently byHeaviside, wearenot allowedtooverlooktheimpedance duetothemagnetisation ofthe iron itself under the influence ofasudden current, Imayle wrong inthis, but my impression isthat this very impedance ‘would help tomake theinterior ofanironshell freer from electric disturbance than itwould bewith amass ofequal conductivity of copper, orother metal having equal conductivity. Thesubjectissotremendously interestingthatIdohopethis isonly thebeginning ofit,andthat weshall have agreat det more ofit,Colonel Armstrong spoke oftheignition ofammuni- tion completely encased inmetal. Ihope hewill experiment ia that direction. ‘The metal wasnotsoldered allround Ipresume. Lieut-Col, ARMsTRoNG: Ithink itwas, ‘The ammanition it made damp proof, andtherefore thecase must becomplete ll round. ‘ThePuestent: IhopethatColonelArmstrongwillbeable totake upthematter experimentally asascientific question; to see,forinstance, ifthinsteelinsteadofcopperwouldmakeanydifference. Besidesthat,Ithinkthatonalargerscalesomething ta: DISCUSSION. sy shouldbedone.WeallknowhowFaradaymadehimselfazi¥itian cage, six feet indiameter, hung itup inmid-air inthe theatre ofthe Royal Institution, went into it,and, ashesaid, lived initand made experiments, Itwas acage with tinfoil hanging allround it;itwasnotacomplete metallic enclosing shell. Faraday hadapowerful machine working intheneigh- bourhood, giving allvarieties ofgradual working upanddischarges by“impulsive rush;”andwhetheritwasasuddendischargeof ordinary insulated conductors, orofLeyden jarsintheneighbour- hood outside thecage, orelectrification and discharge ofthecage itself, hesawnoeffects onhismost delicate gold leafelectroscopes inthe interior. His attention was not directed to look for Hertz sparks, orprobably hemight have found them intheinterior. Edison seems tohave noticed something ofthekind inwhat he called theetheric foree, Hisname “etheric” may, thirteen years ago,have seemed tomany people ubsurd. But now weareall beginning tocallthese inductive phenomena “etheric.” Icannot sitdown without expressing inthe name ofthe Institution ourmost cordial thanks toDr.Lodge forhaving taken allthetroublehetooktobringthissubjectbeforeus,withthe beautiful experiments hehasshown, andforhaving stimulated so many minds, whether todefend oroppose hisviews. I'am sure youmust allfeel grateful toMr. Wimshurst also forthepotent asistance hegave toDr.Lodge toprove hiscase, and forthe Potent application ofhis splendid apparatus this evening to further illustrate and tocriticise some parts ofDr.Lodge's case. ‘The discussion has been sometimes warm, and hasbeen carried on with aconsiderable degree ofhumour; butIamperfectly sure thatweallfeelexceedingly obliged, notonly toDr.Lodge, butto allwho have spoken onthesubject, whether they have attacked Dr.Lodge wholly, oragreed with himwholly orinpart. Iehas Pointed outsome flaws intheLightning Rod Conference Report batIdothink that this book continues practically tohold the field, byitspractical rules andrecommendations fortherender- ingofbuildings andtelegraphic apparatus safeagainstlightning.‘Wemay admit thevalidity ofsome, orperhaps even ofallhis criticisms oftheorthodox dogma, Wemust admire thevigour Nits ofhisattack; andinthebrilliancy ofhisownexposition wecar- notbutseemuch that isinstructive andsuggestive. But, after all,theconclusions adopted bytheLightningRod Conferencedoaffordusverystrongreasontofeelthatthereisa very comfortable degree ofsecurity, ifnotofabsolute safety, given tousbylightning conductors made according tothepresent aod owrHopox rules asactually laid down inthis book. Iamquite sure that theauthors ofthis book will beexceedingly glaito modifytheirviewsinanypractical waywhatever, whencauseis shown and proof given that such modification will improve the practicalresult, wee,Major P.Caxvew, RE. [communicated]: This paper andthe discussion onitrecalls tomind the old fable ofthe wars ofthe Gods and Titans. There hasbeen plenty oflightning playing about tootohelp theanalogy, while there wascertainly aJore like serenity anddignity ofutterance about myfriend Mr.Preset. Idonotthinkitnecessarytodeclareononesideortheother, butshould like tomake afew general observations after the manneroftheGreekchorus. Iwillfollowthenumbering ofthe paper itself. 2.—This eanonly happen, ifatall,when rainisfalling heavily from theintermediate cloud. Agood deal ofthestored-up energy must bereleased inthe primary flash. ‘The thoroughly wet surface ofany object struck under these conditions will tendat least toprotect it. Ishould nottherefore expect these secondary discharges tobevery dangerous, although they may produce startling effects. Professor Hughes didnotappear tohave quite grasped themeaning ofthe“impulsive rush” discharge inbis remarks, 4.—The essential difference between atinplate andacloud bs been already pointed out. Iwillcome tothisagain later. ~ 5.—This isvery instructive, 6.— ... (“The sparking distance under these eiroumstances issurprisingly great byreason oftheimpetus with which the “electricity rushes intothetopplate.”). ..Icannot understand this, unless Dr.Lodge believes that electricity ismaterial. 8.—Here, again, theconditions oftheexperiment arewidely 1893 DIBCURSION. sat different. from those ofwhat itisintended torepresent. TheYbor,fameissuesabsolutely fromametallicpipeinmetallicconnection with ametallic representation ofearth. The case ofthereal chimney ismuch more nearly that of9with animperfect con- ductor (the soot deposit) incircuit, which isconsolatory. 10,&e.—All these experiments were very beautiful, and certainly the effects produced closely resembled those often uoticeable inthe centre of@storm atsea, with heavy rain, especially the side and multiple flashes and the violet-coloured discharges. 16,&e.—It appears tomethat Dr.Lodge utterly failstoprove theoscillatory character ofalightning flash, ‘Theweakest point inhistheory,tomymind,isthatheconsidersthepath,in tir,taken bythe discharge tohave allthe properties ofa metallic conductor, including adeterminate resistance perunit length. Now Idonot think that much real knowledge has been attained onthesubject oftheprocess known asadisruptive dis- charge, andthemolecular movements andchanges bywhich itis propagated. Taking thedissipation ofenergy inheat, which isvisibly there, Isthis due toresistance ofthesame kind asinametallic conductor?Isitconstant;or,isitnotwidelydifferentthe instant before, the instant of,and the instant after the flash? Isthere not also, probably, aconversion ofenergy chemically or byelectrolysis? WehaveverygoodreasonforthinkingthatthecounterE.M.F. ofthevoltaic areissomething considerable compared with the E.MLF, producing it.Wehave alsovery good reason forthinking thatpowerful chemical effects areproduced intheatmosphere hy lightning. ‘Then asregards theself-induction, howcanweestimate this? Weknow that, when constrained inametallic wire, thedischarge ‘will sodistribute itselfastoreducethis—will trytominimiseP, theimpedance, bydecrease ofpL, even atthecostofincrease of R.How much more freely can itdothis where there isno coustraint, where one line ofparticles ispractically asgootl ax anotherover,perhaps, yardsofsection? saa LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May16%, fone 25,&e,—Iron versuscopper.—Now Dr.Lodgeisverydogmaticabout this. Hismeaning is,apparently, that ironcertainly offers more impedance torapidly varying currents than copper, this could not, Itake it,bedisputed;butthatthistotalimpedanceis made updifferently—the true resistance factor being ofmore im- portance inthecaseofiron, ‘Thereasoning ishardly conclusiveI think, butthe conclusion isaltogether wrong Ifeel sure. The proper fanction ofalightning conductor isnottoexhaust the superfluous energy ofadischarge, buttoensure that that energy shall beallexpended inocuously intheair,and the less thetotal impedance, and especially the resistance, ofthe conductorthe better. Atthe same time Ithink that, under practical conditions, iron ofamoderatesize—sayNo.2gaugewire—isquitegood enough. 28. Idoubt this statement. Of course itwould betrueif thedischarge occupied thewhole cross section, but itdoes not; itmakes »/pL*+K¥aminimumbytraversingtheouterannali alone, thus inereasing R,atany rate until itiscomparable topI. Obviously bythis selection ofroute thevalue ofR forarapiddischargemaybeindefinitely great,eveninthecase ofathick copper rod. Idonotunderstand the limitation in hmmekets [not many miles]; pI.varies rather more rapidly thao directly asthelength, while Rvaries directly asthelength. 29.This ismere assertion. Idonot believe alightning current ever exceeds afew hundred amperes; but Idonot assert this. 30, We allknow there aremillions ofvolts;thequestionis, how produced ? 33.Lightning Protectors.—Certainlytheexperimentsdescribed byProfessor Hughes,andthoseindicated byDr.Lodge,demom strate clearly that there areconditions ofdischarge under which anyordinary protector would fail. Ontheother hand, thevery extended experience ofourown Postal Telegraphs clearly points tothepractical value oflightning protectors, The most logical conclusion would appear tobe,not that the statistics ofmany : yearsaremisleading, butthattheconditions arenotthesame. 16903 DISCUSSION, a3 Everyoneknowsthatprotectors arenotperfect,andwouldhailxsir anyrealimprovement; buttheimprovement must beproved by experience, and notbyreasoning based onfalse premises. 39,A.—The impedance offered byaconductor to#rapidly sarying current isnodoubt greater than toasteady current, but thisdoes notprove that itamounts tohundreds orthousands of ohms inaproperly constructed conductor conveying alightning discharge; orthat itismore than amere trifle compared with that offered inthe air-path ofthe flash, where, Iconsider, the energy isandshould beexpended. Asregards thestatements quoted from the Lightning-rod Conference. ‘The first statement does not, Ithink, imply that theman could not behurt, but that hemight not be; and as nothing ispremised astowhatheistostandon,Ishouldnot wyitwasuntrue, though itispossibly injudicious. Tam not prepared todefend thesecond statement, axgun- powder isanimperfect conductor;butitseemstomethatsome interesting experiments might bemade inthis direction. Asa nile, the discharge ofaLeyden jarwill not ignite, but will scatter, gunpowder;butitcanbemadetoigniteit. B—This statement isworded rather loosely. Itisnotquite apparent whether itisthejarorthe charge onthejarthat is likened to spring. Idonotseewhat isgained bythecom- parison. Noone probably would deny that the discharge ofa Leyden jar under certain conditions isvibratory, but many people would deny that the vibrations are analogous tothose ofaspring. C.—This statement begs thewhole question. Ifwhenever thespark ceases tobeoscillatory itdegenerates ofnecessity into 4fizzorrapid leak, then itfollows that every proper flash is oxillatory. But Idispute thepremise, D.—This also Tdeny. Wedon't want much commen orun- common resistance inalightning conductor;andwedon'tgetit inaproperly constructed one. EI admit there issomething inthis; but Icontend that pieces ofmetal near the conductor aremuch better connected toit, au LIGHTNING, LIGHTNING CONDUCTORS, Bre.(May162, dae F.—The first statement isrevolutionary, andseems tobe applicable tospace offour dimensions. The explanatory state mentiscontradictory toformerstatements inthis)paper.Iftrue, what becomes ofthereasons given forpreferring irunta copper. Certainly there are reasons fordividing upthe conductor. There arealso practical reasons, and Ithink mor weighty ones, against it. G.=IleavethistoMr.Wimshurst. Nothingappearsto suggested asanimprovement onasetofpoints. ‘H,I—These statements aresimply unsettling. There is certainly some truth inthem. There isalso truth inthestate ment that agood lightning conductor isasafeguard, although perhaps notanabsolute one, Inow come tothepractical outcome ofallthis asgiven by Dr.Lodge;andafterallthistrumpeting forthofanewcreed— these terrible fulminations against theprophets oftheoldsuper stition—What isthenew revelation ?Itisapparently contained inanumber ofstatements constituting, itistobepresumed,s guide tocomplete protection. Ofthese, the majority bear + suspicious resemblance tothetenets oftheoldsuperstition,© much sothat one istempted tobelieve that, after all,ox present powder magazines, &c., may escape destruction from thecoming case b;butinsome, what Mr.Preece might cal,the cloven hoof isdisclosed. Itisimpossible formetodiscus them all, but Iwill take 75asperhaps the most, dangeroos Icannot understand how Dr. Lodge can reconcile this stale ment even with his own theories. ‘Would not awell-earthed piece ofmetal inproximity toa2 elevated portion ofthelightning conductor, butnotattached te it,belikely togive risetoaspark across, establishing aderive cireuit, and thus perhaps causing thefusion oftheconductor thepoint ofderivation, and great danger offire? One instant atleast ofthis happening iswithin myknowledge. Theow ductor ofChichester Cathedral passed within afewinches ofs lead flatroof without being connected toit. ‘The roof wascot nected toearth bythedownpipes. Theconductor wasstruck andtheflashleaptacrossthisspace,theresultbeingthat600 60feetoftheconductor wasfused from thispoint up,while every Mater, downpipe was broken across atonejoint oranother, where the discharge encountered local resistance. Now afew words upon Mr.Preece’s observations. Idonot think hequite takes inDr.Lodge’s meaning intalking ofthe cscillatory character oflightning. However many oscillations there may bethere must beapreponderance oftransfer inthe true direction inorder todischarge the quantity. This will account for the fact ofitsproducing chemical traces and magnetism. ‘The whole energy which isexpended must be}QV,where Q isthequantity stored upand Vthe potential difference, and however itoscillates iteannot domore than this amount ofwork. Eren iflightning beunidirectional, there must beonegreat wave ofcarrent, giving full play totheimpedance caused byself- induction, although this impedance will notamounttoanything likewhat, Dr.Lodge considers ittobe,since therate ofvariation isnot nearly sorapid. The evidence ofcurrents intelegraph lines originating ina lightning discharge isnotvalid against theoscillatory character oflightning itself, ‘They traverse long lengths ofthin wire with considerable resistance, self-induction, and mutual induction with other wires, and Tshould saywould inany case beunidirectional andcomparatively long drawn out. But Ifully sympathise with Mr. Preece inhisdefence ofthe older electricians, How does it befitus,towhom they have leftsosplendid aheritage, torender them back mocking? Superstition! Well, ifyoucome tothat, what areallourbeliefs and theories butsuperstitions erected upon oldobservations and reasonings ? What doweknow oftheagency bywhich gravitation acts— howthesun holds theplanets,asitwere,inthehollowofhix hand? But what wedoknow ofitisnotsimply abrilliant theory evolved byNewton from hisinner consciousness and thefallof anapple, buttheresult ofcenturies ofobservation and observed accordance ofevents with calculated prediction, Thus weknow itslaws ‘although ignorant ofitsreal nature. Butinthecase oflightning,weknowverylittleindeed.Whocantellofthe vou. xv. 37 x4 LIGHTNING, LIGHTNING CONDUCTORS, Ere,[May160, ‘ot, generation ofthethundercloud, orthemeaning oftherushof wind toit? Why, thelower animals seem tohave more insight into itthan we have. Mr.Preece spoke ofthesolemnity oftheoperations ofnature, | andwas, Ibelieve, laughed atforapplying thisepithet toligh- ning. Iadopt hismeaning entirely. What canbemore solemn than athunderstorm, especially atsea. “They that godown to “theseainships,thatdobusiness ingreatwaters; thesesee “the works oftheLord, and Hiswonders inthedeep.” Andis allthistobebroughtdowntoatinplatecloudandastorminateacup?Byallmeansletusexperiment andreasononoursmallexperience, butletusavoid dogmatising, ‘atanyrate until natore hasamply justified us. Weare,Itrust, allgrateful toDr.Lodge forstirring upthe question, andforhisableexposition ofhisoriginalviews;and originality, even when inerror, helps forward theattainmentof truth, And without attempting todecide whether heisright orwrong inhisconclusions, letmerecapitulate thedirectioas inwhich Ithink hisexperiments might bebrought more nearly inaccordance with nature andwhere histheory appears weak:— First——The cloudshouldcertainlynotbemetallic. ‘Thegrt difference between ametallic and apartial conductor iswell shown inDr.Lodge’s own experiments. Wecannot makearel cloud; butatleast wecangetnearer toitwith awetspongeot handkerchief, orsome such device. ‘Second.—Farth should always bereal. The mere completion ofthe circuit inthe way weattribute toearth often leads to fallacious results, especially withhightensions andsmall quanti- * ties. Itiseasily seen that theaddition toaportionofacircuit| ‘ofaninfinite capacity islikely tomodify results—still it it; ‘often overlooked. One side ofthe generator and everything representing aconductor orprotector should bethoroughlycom neeted togood earth. Third.—The assumption that theair-path traversed byadis ruptive discharge offers thesame kinds ofimpedance asasolid conductor, andthose only, cannot, Ithink, bejustified. Fourth.—The doctrine that alightning conductor isperfect | 82) DISCUSSION. mat ‘which offers aconsiderable impedance tolightning, andistheMaes,, receptacle foravery appreciable portion ofthetotal energyof theflash, is,Ithink, pernicious,andthetruethingtoaimatis conductor offering aminimum ofanykind ofimpedance, and ‘especially ofresistance proper. This being ofcourse fulfilled when theimpedance oftheconductor isnegligible compared with that oftheair-path—a condition probably obtained withaconductorerected and earthed inaccordance with the recommendations of ‘theLightning Rod Conference. There isoneremark ofDr.Lodge's Ifind Ihave overlooked, ‘and that is84, where hetalks ofanearth resistance ofa thousandth ofanohmorless.HasDr.Lodgeevertestedareal Lightning conductor? Ifear these remarks have been tedious, and Iamfully con sciousofmyinability tothrowanylightonthesubject, but,asit isoneofdeep andabiding importance and interest tothehuman nce,Ihopetheintentionofthisprotestagainst.hastyconclusions mayatone foritsfeebleness. Lieut.-Col. J.T.BucKNtLt, late RE. [communicated]: This Lsw-Co isthe most interesting and suggestive paper Ihave ever read on thesubject oflightning and lightning-rods. Section 5isvery important, asitseems toindicate that “the Niolence ofthespark islessened” byanincrease ofohmic resist- anceintheconductor, butthat theconductor gathers thestroke aseffectively, sofarasstriking orsparking distance isconcerned, aswithamuchlowerohmicresistance. That present practice gives conductors amuch higher con ductivity than isabsolutely necessary hasbeen held before. ‘Thus, Mz.R.8.Brough, inacommunication totheAsiatic Society of Bengal, February, 1877, gave scientific reasons inharmony with themoreconvincing arguments andmathematics nowpublished byProfessor Lodge; andImyselfsuggestedin1881,toaWar Office authority, thatalargetelegraphwirewillalwayscarryoffa stroke oflightning inocuously. The clond tocloud, orcondensers inseries action, has been ablyexamined bythelecturer, butthepossibility ofsubterraneous -condensers inseries acting similarly isnotsuggested, and this aan LIGHTNING, LIGHTNING CONDUCTORS, Bre,(May1, $exco appears tometobeamore probable explanation ofthephenomena noted intheTanfield Moor Colliery than theonegiven onp.237 Lightning Rod Conference, and adopted bythe lecturer (see section 46). : ‘The coal strata separated bystrata ofvery low conducting power and connected bythe galleries, shafts, and winding-gew, and tramways ofthemine, would spark from one totheotber through these imperfect connections. And this leads toa very important matter, which Ithink bas notreceived sufficient attention from Dr.Lodge, viz.,theportion ofthemaininducedterrestrial charge. Wherewaterandgis pipes exist, [believe that they become highly charged byinduc tion before the flash, and that the flash follows the roate of minimum impedance thatexistsbetweenthecharged‘cloudand theearth system ofconductors which isinductively charged. I istherefore useless toprovide “agood earth independentofthe «water andgaspipes,” asproposed insection 59; onthecontrary, itwould evidently bepreferable toconnect thehighest portioas ofthewater andgassupply pipes totheconductors, andthusgt totheinduced charge bytheshortest route. Iam convinced that this word ehorteat isone that should never belost sight ofinlightning-rod practice. Forsimils reasons Iwould addthewords, butwhere theycannot beavided they should beconnected, tosection 63,asdisruptive isfarmow dangerous than conductive discharge ;andIamutterlysceptical ‘astoafiash melting even asmall gas-pipe, origniting thegu except bydisruptive discharge. Hence, large cast-iron gae-pipes with oakum packing atthesockets aremore dangerous conductors than small yas-pipes with threaded connections. Ishould like toask the Professor how hewould deal with ‘thegreat mass ofmetal now frequently stored inmagazines— (a)bymetal powder cases, which have replaced powder barrels; (b)byliveshell intheexpense magazines. Would heconnect them? IsayNo. Withreference tosection65.Therearenotableexceptions— tallrods being absolutely necessary over powder mills, petroleum oilwells, ete. 1689) DISCUSSION, 529 Mr.LeowanpJosera[communicated]: Availingmyselfofw+.Journ thePresident's permission tosend inwriting thefewremarks which time didnotpermit ofmy making atthemeeting, I begtoobserve, firstly—as regards Dr.Lodge's analogy ofthe Leyden jar—one objection against his theory, not already mentioned, isthat his dielectric air isthe reservoir, soto speak, ofelectricity, whereas glass isnot. Next, asregards oscillations:‘Theircharacter, aswasshownbyonespeaker, wassomething like Fig. 1. This iseasily tobeaccounted for. Letmegive youanillustration. LetFig. 2(a)represent | , tala i 6 Fra. Fro. several pieces oflinen piled ontopofeach other, and laying betweentwouprightboards. Compress theboards,andthelinen sheets willtake upawavy form similar tothat shown inb. Now, instead ofboards forour lightning, wehave the pressureofdischarge ontheonesideandtheearth'sresistanceontheother. The result isanumber ofwaves allalong the course taken, The reason thewaves getsmaller asthey near the earth isthat the earth's attraction draws the current in astraight line, Energy ofdischarge, distance from cloud to earth and earth's resistance atthat point, being known, itthen decomes comparatively easy foramathematician (which Iam tot) tocalculate the size ofthese waves oroscillations atany point. Now forthepractical side ofthequestion. Iron r.copper ‘tasbeen fought outbyabler heads than mine, soIshall content myself with asummary ofwhat xeems tometobe theadvantages ofthelatter. (1)Copper isabetter conductor; 2)iron iseasily affected byweather, and corrodes even when 30LIGHTNING,LIGHTNINGCONDUCTORS, Ere.[Mayit,| us.Jourt. galvanised; (3)magnetic defects iniron; (4)itistrue that theconductivity ofheat incopper isgreater than iniron,but continual contraction and expansion affect iron—as aoondisctor ofelectricity—sooner than copper, asthe latter, being more elastic, does notsosoon lose theoriginal construction ofits particles. Astoshape, Tpropose tubing, notrod, forseveral reasons (1)Tubing gives themetal agreater cooling surface; (2)# greater conducting surface; (3)alesser chance ofthemets! being destroyed byexpansion and contraction, asmore “play- room.” isgiven. Mr. Symons spoke oftheadvantages ofahot-air currest Asanamendment tothis, Ishould propose putting the conductor along that side ofthehouse which isthewarmest— itynear fires, ete.—and forthis reason: copper when heated isnegative, and vapour ispositively charged. Unlike elects attract each other, soweshould have moister airsurrounding theconductor, drawing theelectricity towards it,vapour being the better conductor inthe air. Ttwas proposed tometheother day(whether rightly& wrongly Isubmit toyou) that allmetal work within the building should ‘bejoined together, and these again tothe conductor, both atthe top and atthe bottom, the latter connections being made with ametal ofhigh resistance; the entire thus forming anante-induction circuit. The high resistance ofthe connections isthat the current. should give preference totheconductor. Before Iconclude, Ishould liketoaskacouple ofquestions about which Ifeeldoubtful. Firstly: Last year, attheRoyal Meteorological Society, Isaw anumber ofbeautifully-taken photographs oflightning. Innearly allofthem Inoticed twoparallel flushes, oneseemingly going up,theother dorm wards, In fact itlooked asifthe current had wanted togo toonepart oftheearth from another, but, finding theresistance high, had taken aroute through theair,viaclouds, back to earth. Ishould like toknow ifthis ever occurs. Again, something might have been said about. fire-balls or 1080.) DISCUSSION. ast balllightning—their origin,andthecauseoftheirexplosion, Mr.sos Aprobable way toaccount forthem might bethefollowing: Fire-balls areseen, asarule, only during severe storms. Now some hurricanes, like cyclones, arebutvast whirlwinds, blowing round anaxis insinuous spirals towards that axis. This was firstelaborated byMr.Redfield andCol.Reid, andverified by Professor Maury oftheWeather Burean oftheUnited States, ‘Now then, imagine aninstantaneous flash oflightning descend- ingthisspiral andsuddenly encountering theearth's resistance. ‘Thewhirling would forasecond orsocontinue above ground, thus forming thefire-ball, The bursting ofthese balls might easily beeaused bythesudden expansion ofthevapour inside, onthepressure from without (caused bythewhirling) being diminished, Another view-as tothecause ofthereport (not, however, held byme) is,that inthecentre isavacuum to which theouter aircanhave noaccess solong axthestronger spiral current ofairisblowing. Themoment, however, that the electric current passes through andinto theearth, thevacuum caused bytheflash—in addition totheother already made—is toogreat, and theairrushes infrom allsides, making the noise heard. Dr.Ouivex Lover, infurther reply [communicated]: Toape.tie. yerson unaccustomed tocontroversy, like myself,Ifindthereis anabsurd tendency toaccept anyposition which may happen to besuggested asappropriate tome,andtoreply toobjections raised against anumber ofshady views asifthey were myown, when they arereally notmine atall; being, indeed, often only ‘supposed tobemine because they arerepudiated bymycritic. Another tendency, more strenuously tobeguarded against, isastupid feeling ofsemi-irritation when one finds that state- ‘ments carefully recorded asthe result ofmuch thinking and experimenting, combined with some acquaintance with thefun- damental theory ofthemasters ofthescience, arecalled dogmatic ‘andlightly setaside, notonaccount ofanyrealobjection orvalid ‘argument, but solely and conspicuously because they donot happen tofitinwith thespeaker's preconceived opinions or hastily interpreted experience. an LIGHTNING, LIGHTNING CONDUCTORS, Bre. (Bay 1, vvteige. Inreplying tothecriticisms withwhichmypaperhasbeenthas farhonoured, Ishall endeavour toresist both these tendencies; and especially shall Itrytopick outofthecriticisms those points which seem tomewholesome and salutary, andwhich Ieanmore orlesscompletely accept. First ofall,Imust’ clear away some misapprehensions which havo arisen inconnection with questions ofhistory and priority. ‘They have arisen partly through misapprehension ofwhat Isid orintended tosay, partly onaccount ofadistinct oversight oa my part innot acknowledging some prior work ofProfess Hughes and M.Guillemin inthesame direction, Totake the last first. Iwell knew that experiments in connection with lightning protectors had been made by Gnuillemin andProfessor Hughes, because after mySocietyof Arts lectures Professor Hughes was good enough towrite mea2 account ofthese experiments, and toconsent tomyrequest tbst hisletters should bepublished intheElectrician. There they areforalltosee,anditwasamere oversight that Ididnotrefer tothem inthecorresponding portion ofmypaper. Isuppose it wasbecause Iwas trying toexplain thefacte quickly, andws notthinking about history orpriority. Tcanassure Profesor Hughes that nothing was further from my intention than to claim, orbysilence tosuggest aclaim to,any priority inoom nection with such experiments. ‘These experiments showed clearly, in1865, that nosimple shunt arrangement could entirely protect atelegraph instrument from aLeyden jardischarge;andthoughtheyhesitatedtopres their conclusion tothecase oflightning inallitscompleteness, yetIventuretothinktheirexperiments demonstrated thisals. ‘Demonstrated—not, ofcourse, that lightning protectors cannever protect, northat they may notalways partially protect, butthst inanumber ofeases they may partially, yetvery seriously, fil. That isonepoint where Ihope Ihave nowsetmyself rights theother isdue, not‘toanoversight ofmine, buttoamissppre hension bymycritics ofwhat Iintended tosay. Inemphasising theimportance ofthe“impulsive rush” case oflightning di charge (the only onevery liable tooccur, perhaps, duringahear] 1850) DISCUBSION. 33 shower) with reference totheabsence ofprotective virtue from DrLae. pointe, andthe'ease with which they, aswell aseverything else, couldbestruckbyanoisyanddestructive flash,Isaid:“Thesteady “strain case was theonly oneever contemplated bytheolder “electricians ;infact, sofarasIknow, itwas myexperiments “last year which first called attention totheother case.” These words Iused, and these words Istill repeat. But the reason Iused them was not, Ineed hardly say, toclaim a miserable figment ofpriority insotrivial amatter, but to emphasise thefact—for Ibelieve ittobeafact—that the impulsive rush case oflightning discharge had notpreviously been attended to,andthat hence many vitally important features had been overlooked. Inthe“steady strain” case, points fizzoffquietly and arenot easily struck byaflash. Intheimpulsive rush case (pace Mr. Wimshurst forthepresent) they donothing ofthekind. ‘The important thing, therefore, inthewords Ihave quoted isnotthewords “myexperiments,” butthewords “experiments “lastyear.”Itisnottoclaimtheauthorship, buttoemphasise therecentness, oftheobservations ofthis hasty and unprepared- fordischarge. Itmaybethatthecallingattentiontothepossibilityofthis caseoflightning discharge isafter allnotrecent—I only said it wasrecent “sofarasIknow” ;but, ifnoticed before, it,hasnot been emphasised oreven mentioned inany books orpapers ~authesubject that Ihave seen. Nor, indeed, doany ofmy tics really assert that itisanoldidea, Professor Hughes seems tothink that Imean, by“the im- pulsive rush,” aLeyden jardischarge ingeneral, andrefers me tuck totheyear 1752, when aProf. Muschenbroeck, orhispupil Caneus(videGanotorDeschanel passin), appearstohaveantici pated mebysomething over acentury. But thegentleman who {intnoticedanelectricspark—Dr. Gilbert, [_presume—must be creditedwithastillearlierknowledgeofan“impulsiverush”in thatsense,‘That,however, ishardlythesenseIintended. Mr.Wimshurst, inhisremarks atthe first evening ofthe discussion, seemed tothink Imeant toclaim themode ofobtain- exuoerand, iamestee ommnorons wepeo| Drtod.ingimpulsive rushesfromtheoutsidecoatingsof‘twoLeyden jars;andpointedoutthathismachineswereso:arrangedthat theoutside coats ofjars could atanytime beconvenientlycon nected tothe outer circuit. Fortunately, Idid not mean orimply this either; though, forgetfully, Imight accidentally have implied something ofthe sort, because Icannot help perceiving how much more neatly many Leyden jar experiments, recorded inbooks orshownatlectures,couldbeperformed, ifthismodeofconnecting tothe outside circuit were more generally employed, insteud ofusing «discharger workedbyhandorbypullingastring.‘Thesdra- tage ofhaving allthevariable part ofthecircuit atzero potentialuptotheveryinstantofdischarge, is20marked,andIhaveused thisarrangement sofrequently, that Imight have been tempted toforget that itwasnotreally aperfectly well-known, thoagh toofrequently forgotten, method. Iamsure, however, thatwher attending Professor Carey Foster's clasees, long ago, IsawLeyden jars thus connected uptomagnetising spirals orother such things; and Iseem toremember also having first seen itwith something ofthesame sortofpuzzlement astothewayitacted, which Idetect, orfancy Idetect, insome fewofthespeaken, especially during thefirst evening, with regard tocase b, Many remarks have been made—and some weighty ones—o0 thesubject oftheprobable absence ofconducting power inclouds andtheconsequent difficulty insatisfying theconditions ofthe Bflash (ie, the impulsive rush, orspark between bodiesinitiallyatthesamepotential). IadmitatoncethatobtainingBflashesfrommetalsheetsprovesnothingwhatever concerning thepossibility ofobtaining them from clouds. Butmyargumest israther converse tothis. Iargue (whether rightly orwrongly) ‘that flashes often occur from clouds under circumstances which, under thea,orsteady strain, orhigh potential condition, scarcslt seem natural. Thus points aresometimes struck andmelted by flashes; and itisnot easy forthem tobestruck incase « Moreover, from acloud violently raining flashes occur; wheres cone would expect rain tolower the potential gradually. 1890] DISCUBSION. 3 Again, from acloud restingonahill-topflashesoccur;andsparksPeUlve. from abadly insulated body areatonce suggestive ofimpulsive rash conditions—that is,sparks from abody ofzero potential. Similarly, Iunderstand that acloud perforated bythe Eiffel Towerhascarriedonathunderstorm topeoplebelow.Idoubt ifanyofthesethingscouldeasilyhappenundertheconditions ofease a. Bat, itwill besaid, surely some ofthe facts you adduce ‘establish thebadconducting power ofclouds. Toagreat extent theEiffelTowercaso(ifafact)does.‘Thehill-topcasemerelyprovesthattheAildwasapoorconductor. Icaneasilygetlong sparks from metals roughly uninsulated, asbywood, water, orsoil, Suppose, however, itadmitted (not asproved, butasprobable) that clouds arepoor conductors, what then? Allthat wecan usertisthatthewholeofacloud,orevenalargeportionofa cloud, isunlikely todischarge atonce. Iquite think that that isso.Acalculation ofenergy shows that aviolent flash need onlydischarge avery small portion—a few square metres—of a charged cloud, and that thesame cloud may therefore goon sparking foralong while, as,indeed, itappears todo. Now no sreat_ conducting power isneeded foradischarge from asmall aren atagreat elevation: the lateral component ofrush ix inthatcase negligible. Ofcourse itmay beasserted that clouds conduct sobadly thatno“flash” intheproper sense canever occur. Ananswer tothatassertion istheexistence oflightning. Allthat poor conductivity inclouds hastosayconcerning the impulsive rush seems tomethis: that when anypart ofthecloud receives aviolent disturbance from some Aflash initsneighbour- lod, that same partwhich receives itismost likely tospitoff theconsequentBflash.Whereas,withaperfectconductor,any therportion would bealmost equally liable. Poor conductivity g0es,indeed,tohelptheviolenceoftheimpulsive rush;forthe essence ofitisthat aconductor ofsmall capacity atzeropotential shallbesuddenly overloaded. Now, ifacharge suddenly commu- Bicated toaportion ofalarge cloud could instantaneously be shared with the whole, the potential would bereduced, and nothing very violent need occur. vetaige. ‘This may sound like special pleading, but itisonly record- ingthecircumstances that have tobeattended to, What the facta aremust bedetermined bydirect observation onflashes and clouds. But letmehere beseech meteorologists toremember that establishing thecondition forsome oneflash orclassofflashesdoes notestablish the impossibility orimprobabitity ofvery different conditions obtaining elsewhere oratother times. ‘Therear varieties ofthunderstorms, varieties ofclouds, and varieties of flashes, Every good and accurate observation will beahelpto fuller knowledge, butitwill take years ofenlightened experience and observation before allpossible varieties and circumstances ofdischarge canbesupposed exhausted. Before leaving this subject Ishould like toremind th Institution that Ihave never hesitated tocontemplate the imperfect conductivity ofclouds, whatever theconsequences of that imperfect conductivity might be;and, inproof ofthis,I enclose anextract from thereport ofmyremarks atBath, published inthescientific journals atthetime:— Extract from British Aasociation Discussion. There ms “one point where Mr. Preece might have attacked him, bat “where he did not think Mr. Preece had made out the fill “strength ofhiscase, namely, the question—What arethe “conditions ofaflash? He(Professor Lodge) hadassume! “that aflash behaves, ormay behave, like thedischarge of “condensers inalaboratory ;butitwasaquestion whether4 “clouddischarge wasofthiskind.Acloudisnotagoodom “ductor;itconsistsofglobules ofwaterseparated fromox “another byinter-spaces ofair; itmay becompared, therefore,t® “akind ofspangled jar; when aspangled jardischarges there “no guarantee that thewhole ofitdischarges, itmay discbergt “out inaslowish manner; itmaybethatyouhave firstabito ~discharge, then another bit,andsoon,sothat you mayhave* “kind ofdribbling ofthecharge outofit,andyoumaytht: ~failtoget these oscillatory and sudden rushes. Atthesam “time hedidnotthink thatthey could always guarantee doing “thiswith cloud discharges ;anditwould notbesafeinarrangitg 1288 DISCUSSION, sr «protectors toprotect foronly onecase, andthat theeasiest 5:‘ais “they must provide forthepossibility ofasudden andviolent “discharge. Still, theconditionsofactuallightningweretobe “ascertainedbyobservinglightning,andnotbyexperiments in «thelaboratory.” Proceeding now tothe remarks ofthefirst speaker atthe second evening ofthediscussion, Imust draw aclear distinction between Mr. Wimshurst’s experiments and Mr. Wimshuret’s comments thereupon. With respect totheexperiments, Ifeel obliged tohim for exhibiting andemphasising several points which, forwant oftimeorotherwise, Ihadratherslurredover;alsoforrecalling tomy memory alittle point which Ihad forgotten, though itisinmy assistant’s note book; and, lastly, fordetecting aninteresting matter which lad eseaped me. ‘Totake these successively. 1.Heseems tohave exhibited side flashes, from badly earthed conductors and towell earthed bodies, more satisfactorily than inmyhurry during thepaper Imanaged todo. Itisnot likely that heexhibited themsostronglyasIhave obtained them inthelaboratory, because theviolence ofmany of these side flashes isathing that strikes observers with astonish- ment;andwithalongconductor, asstoutasyouplease,itmakes very little difference whether itsfarend is“to earth”ornot. Itdoes undoubtedly make some. Professor Threlfall, and Mr.J. BrownofBelfast,havebothseentheseeffectsatLiverpool, and neither, Ifancy, would contemplate with equanimity theidea of bringing their knuckles near aconductor when struck, however ‘well earthed itwas. Atthesame time itisperfectly true, andImust have often recorded thefact, that from awell earthed conductor thesparks willnotcharge aLeyden jar. ‘They jump inand outagain. Sometimes, indeed, there iearesidue of‘charge, but itis ‘accidental what sign itis,anditisalways merely thetailend of 4seriesofoscillations cutoffbyresistance atsomearbitrary point. Amore striking experiment istoconnect agold-leaf electro 53LIGHTNING,LIGHTNINGCONDUCTORS,Ere.(May16, DeLode.scopetotheconductor. Iftheconnection ismetallically perfet thegold leaves arenearly quiescent. With imperfect, connection they may diverge; they always slightly kick. ‘The experiment isalittle rough ontheelectroscope, foritstrains theleaves downwards, and blows fragments offsometimes; butitiss striking thing tobeable totakeahalf-inchorevenaone-inch spark, ofconsiderable power and noise, from the capofan electroscope whose leaves hang stiffly down allthetime ani barely twinkle. ‘Thesesideflashesarenotverypainful,theylookworsethanthey feel: thecharge hops inand outofyouwithout going through you much ordisturbing the nerves seriously. Ibynomeus assert that aman would necessarily bekilled iftouching com ductor struck bylightning;butitwouldsurelybeapositionof considerable danger. 2.The fact which Mr. Wimshurst. hasobserved, but. which! had missed,isthis:thatwhensideflashesaretriedforatdifferet points inthelength ofawire juining theouter coats oftro‘equally-insulated symmetrical jars,they areobtainedmorestrongis towards either endofthewire, and arenotobtainable atalla themiddle. ‘The middle is,infacta node, There arestatiooxy ‘waves getupinthewire, whose ends arenow athigh potential andnow atlowpotential alternately, just likealong bath which has been tipped and setdown sharply. ‘Toand frothewater splashes, andtheends arenow athigh andnow atlowlevel alternately, butthemiddleisanodeandremainsofaveragelevelallthetime: itisatzero potential, andnospark isobtainable from it,Ttisquite aninteresting fact, andonethat itwoald have been apity tomiss. Iamsure weareobliged toMr Wimshurst fordiscovering it. 3.Lastly. The little point Mr.Wimshurst hasrecalled tomy memory isthis: that when acloud ortop-plate isnegative,# small terminal orpoint gets struck rather more easily—i.,at lowerelevation—than abigterminal ordome, even under the circumstance oftheimpulsive rush.‘The fact isoften so; but Mr. Wimshurst’s account ofitmay leadpersons whohave nottried theexperiment toover-estimate 1080.) DISCUSSION, sao themagnitude ofthe difference, which isfrequently quiteDr.vise. inappreciable:being,indeed,oftennon-existent. Iftheimpulsive rush isviolent—i.e., ifitproceeds from a pairoflarge jars, highly charged, intoaplate ofmoderate size— the difference isnon-existent. Careful measurement fails to show thatthethings equally struck arenotallthesame height ; andthiswhether the“clond” benegative orpositive. ‘This was thecaseIexhibited atthemeeting, andtherewasnoneedto notice ofwhat sign thetop-plate was. But iftherush bemade lessviolent, either byusing small jars orbycharging them feebly, adifference isobservable. Apoint then gets struck, as Mr.Wimshurst showed, at distinctly lower elevation than’ a knob does, whenever theoverhead flash isnégative—not when it ispositive. ‘Those arethe facts, and weareindebted toMr. Wimshurst ‘orcalling attention tothem, but,astowhat themoral andprac- tical bearing ofthem is,opinions may differ. Certainly itinnowise upholds thestatement that points always discharge silently and cannot getstruck, which iswhat basalways been meant bytheir “protective virtue.” ‘Rather it would seem that they getstruck inanimpulsive rush always as casily asanything else, and sometimes, asMr.Wimshurst shows,stillmoreeasily.Forremember thatthe“striking” isnotafizz orleak ofagentle and protective kind, bat isaviolent ani destructive flash. ‘The other experiments exhibited byMr. Wimshurst areless important. One hastodowith theexperiment ofthealternative path, or,asMr.Wimshurst. prefers tocallit,a“bye-pass.” Tdidnotmake anyreference tothis class ofexperiments in thepaper, because Iconsider Ihave gottothebottom ofthem andthoroughly understand their conditions, which Ihave pub- lished; #0that, except inavery serious memoir, they arenot really discussable. ‘This may sound adesperately arrogant thing to.say, batifso, I *Formeamrements seeAppeniis 1V. 510 LIGHTNING, LIGHTNING CONDUCTORS, Ere. (May 6, vetatee. cannot help it, One would make noprogress atallifonecould never gettothebottom ofathing and feelone’s ground secure And Iamsoprofoundly conscious ofignorance with respect to ‘avast number ofphenomena that Ihave nocare towaste time bypretending ignorance where itdoes notexist. ‘May Itherefore save time bysaying that alltheexperiments ofMr. Wimsburst onthis particular matter are completely ia accord with mytheory ;*and also that theremark ofProfesor Hughes concerning theprobable delay oftheBspark behind the Aspark iscompletely borne outbytheory. ‘The lagis,infats quarterperiodofthe oscillation. Referring toanother remark, Imay alsosaythat atLiverpal wehave recently obtained excellent photographs ofaslowly oscillating spark onarotating sensitive dise, and that theom stituent oscillations are not only conspicuous, but well spresi out and accurately measurable, and inagreement with theor within onehalf percent. Finally Mr. Wimshurst seems tohave exhibited anordinary spangled jar, inwhich acharge electrostatically induced inits outer coat exhibits sparks between thespangles inasimple and well-known way. Hethinks thisisanalogous tothesparkings02 thewall-gilding excitedwhenanoscillating currentissentthrough ‘awire circuit lying onthefloor; but,without laying unnecessiy stress onthedistinction between electrostatic andelectromagnetic induction, Ithink Iamjustified insaying that thetwoexper- ments arenotanalogous. ‘Some remarks which hemade inapparent detraction ofthe Voss machine Idonotcompletely follow. Iamvery ready to admit that aWimshurst machine isanexcellent instrument, bit itisnotnecessary tocondemn theVoss machine,ortosugget that itisworse than anoldfrictional machine, inorder tobelieve that, TheVoss machine Ihave found most excellent too,andfor small sizemachines Ipersonally prefer it;butthat isperhapsa2 idiosyneracybaseduponfindingiteasiertoturn,easiertotake topieces, thoroughly under control, and mechanically simpler. ©SeeAppendix 1890) DISCUESION, ow ‘Thereisnottheslightestdifficulty intellingwhichterminalispetaig.positive, thebrush appearance tothepoints tells usthat when- ever wechoose tolook. Certainly iteasily reverses. That isits oneobjection ;andoccasionally itisanannoying peculiarity, though Ihave gotpretty used toit. Tshould like toremark here, what after allisfairly obvious, howgreat service hasbeen done bythedevelopers ofthemodern influence machine, from Nicholson, Varley, and Thomson, to Holtz, who didsomuch, and ontoMr.Wimshurst himself. Many ofthe remarks ofProfessor Hughes Ihave already incidently answered orreferred to. Hesays,quite rightly, that Imaintain flatribbon andplenty ofsurface isanimprovement on. round rod, butastowhether itisamatter onwhich very great stress need belaid, Iwill defer discussion till Icome tothe observations ofthe President. Professor Hughes’ experiments oniron versus copper nodoubt agree with mine whenever heuses alternating currents ofthe same frequency, and disagree when heuses alternating currents ofmuchlowerfrequency. Fortelephonic frequencies ironhas much greater impedance than copper. Oliver Heaviside does not foraninstant deny this, buthesays that there may becircum- stances inwhich this extra inertia isanadvantage, inthat it helps topreserve thecharacter, orquality, orshape, ofelectric ‘waves, although itadmittedly transmits more slowly andweakens them. IfMr.Preece andobservers inAmerica find inpractice that copper wire isbetter fortelephony, asthey apparently do, thenthat means that thecharacter ofthevibrations iseuficiently preserved incopper, andthecomparative absence ofretardation is alltothegood. Itmay,however, stillhappen thatinsubmarine cables iron will shew anadvantage. Itisacurious paradox no doubt atfirst sight, and whether theadvantages outweigh the disadvantages inanyparticular case isaquantitative question of noextreme simplicity; though, nevertheless, ithasingeneral terms been worked out. Batthequestion forordinary Leyden jarfrequencies ismuch simpler,Forthemtheimpedance ofironandcopperofthesame‘VOL, XVIII. 38 a LIGHTNING, LIGHTNING CONDUCTORS, Ere, (May 16t, Ovtaige. diameter ispractically thesame, unless thewire isvery longorverythin,Theresistance oftheironismuchgreaterthanthat ofthecopper. Allthiscanbeexpressed, andhasbeen expressed quantitatively, with, inmy opinion, complete certainty, and beyond the reach ofany but revolutionary doubt towhich all scientific doctrines are liable. And onthepractical side one may saythis: The circum stance ofatelephone wire andofalightning rodarenotonly different, they are, insome respects, opposite. The object of ftelephone wire istoconvey electric waves, unaltered and unweakened, ton distance. One object ofalightning com ductoristowipethemoutanddissipate theirenergyassoonaspossible, Theveryproperties whicharedetrimental inonecae may bedesirable inthe other. Itisnodoubt aquantitative question how faritiswise towipe outenergy inthelightning conductor itself,andMajorCardewthinksitisunwisetodos atall, Possibly; but atpresent Ihold that solong astotal impedance isnotappreciably increased, andsolong asthemargin ofmelting isnottooclosely approached, solong itisdesirable to dissipate energy wherever you can, and tocheck theviolenceof theoscillations asrapidly aspossible; and hence Ihold thata moderate amount oftrue resistance isnodefect. inalightning conductor. Everyone must admire thebeautiful method bywhich Professr Hghes tests hiswires forcircular orcylindrical magnetitation. InMr. Symons’ objection tolaboratory experiments being regarded asatallanalogous tolightning, and still more clearly inCaptain Cardew’s solemn protest infavour ofthedignity of athunderstorm andtheabsence ofdignity from experimentscom ductedwithtinplate,Iseemtohearechoesofsomefineolderasted objections which were current inthetime ofFranklin, andwhich were,perhaps, somewhat moreinharmony withthattimethin with thepresent, Now that the subject has been mooted, bow ever, Imay bepermitted toassert my conviction that the intrinsiodignityandsolemnity ofnatureisaspresentinaspark one inch,asinasparkonemile,long;that,lookedatwith | 10 DISCUSSION. us insight,adropofinkhangingfromafunnel*maybeasinspiring De.‘ier. anobject. ofcontemplation asacataract; and that toexplicitly claim special dignity fortheone isimplicitly toreject itfrom theother. ‘True, one’s subjective feelings ofawearenotaroused inthe one case asinthe other, but that has todowith the relative size ofthehuman body; and sofarasanobserver is overwhelmed or liable to have his nerves shattered out of ‘existence bythephenomenon heiswitnessing, just 90farheis ‘uotinaperfectly collected andscientific frameofmind.More- ‘over, experiment under modifiable circumstances has enormous advantages over mere observation, especially observation which isonlyoccasionally possible. Hence experiments inalaboratory, andathorongh understanding ofwhatoccursonasmall’scale, areaverygoodintroduction totheenlightened studyofatmos-pherieelectricity, thoughtheyarebynomeanstoberegarded asa substitute forthat direct study. Soletmehere emphat- ically admit andinsist, infullagreement with what Isuppose wastheintention ofthesespeakers, andwiththemoredirectassertion ofColonel Armstrong, that experiments onactual lightning are highly desirable. Such experiments, asasort ofpractical out- ‘come oftheMann lectures, are, Ihope, incourse ofestablishment, tymeans ofthebright idea oftheeditor oftheElectrician, and bythe enlightened co-operation ofthe Eastern Telegraph Company andSirJames Anderson. Atforeign stations storms arefrequent, and, with suitable appliances,Itrustarecordof valuable observations may beforthcoming in,say, five orten years—without, letushope, the“expenditare ofanyobservers.” Thelarge number ofphotographie records oflightning which are nowbeing obtained allover thecountry arelikewise very valuable aidstoprogress. Mr.Symons isadetermined andconsistent. advocate oflarge ‘<roes-tection forconductors, maintaining that they areliable to befased; and asthis isaquestion ofobservation and ex- Perience ofdamage, Ishould bedisposed toallow much weight tohisopinion. Unfortunately, however, the two instances ©SirW.Thomson's"PopularLecturesandAddresses,” page48, ou LIGHTNING, LIGHTNING CONDUCTORS, Ere.(Hay1, Pr-todse.headducesinsupportofhiscontention arenotsuchaswillbar serious examination ;*since inone itisthe links ofachain which aremelted, and inthe other the conductor isnotfuse, butmerely “burnt. byuse.” ‘Mr.Symons beautifully illustrates myremark, that whenerer abuilding isdamaged, itisalways because theinfallible rulesoftheLightning RodConference hadnotbeenfollowed. Thechur’ atGarding wasdamaged because “they took theconductor dors “theNorthsideofthesteeple,notdownthewetside,asve advised"! With reference tohisdepreciatory observations ontheHotel deVille, Brussels, and theBelgian School ofElectricians, Ishall leave them tospeak forthemselves, Mr. Symons will findthst thedeath ofM.Melsens hasbynomeans removed from tht country every man skilled inlightning conductor appliances. Tdonotknow who itisthat bastold Mr.Symons that allhe lightning conductors ‘on theHouses ofParliament, Westminster “Abbey, orwherever itmay be,arewrong inprinciple andar “dangerous.” Certainly Ihave not: Ihave never either ssidot implied that well-erected lightning conductor isother than+ source ofsafety asfarasitgoes. Isaid, inmyfirstMano lecture, that theneighbourhood ofafactory chimney is“asource “ofmild danger.” And soIbelieve itis,even when possessinga good conductor. But most distinctly without aconductor it would beasource ofdanger very much other than “mild” Inever contemplated such acase, nor supposed that anyone would endeavour toincrease their safety bypulling down light- ning conductors! Colonel Armstrong quotes several cases ofdamage, wherein the earth resistance was found tobefrom 100 to200 ohms. With alldeference tohisexperience, Ifeel very doubtful ifthisamount ofresistance issufficient toaccount forthedamage. Ibare admitted allalong that thebetter the“earth” thebetter for everybody;butIhavealsopointedoutnumerousotherreasons torfailure anddamage beside abadearth. ‘=Bee Appendix IIL 1869) DISCUBSION. as Inhismost interesting observation ofthecartridges exploded DsUtes ina sealed metal case, weought tobesure that the flash did notpierce, ormelt, orignite toredness, thecase. Any violence ofthat sort might explode things inavery commonplace and unelectrical fashion. Soalso theviolent shock duetoexpansion ofair,orwhat Sir W.Thomson atBath called the sound-wave, maybeexpected tohave aneffect ondetonators, ‘Mr.Adams calls attention totheradiation ofenergy from con- ductors. Itismost true that radiation goes onfrom allalter- tating cirouits; andfrom linear oscillators likethose ofHertz, or,what isessentially the same thing, from cloud and earth joined byalightning conductor, the intensity ofradiation is prodigions. Itisamost important wayofgetting ridofenergy; itisexpended ontheether attherate ofmany horse-power. Seeapaper ofmine inthe fortheoming July Philosophical Magazine. 1am notsure, however, that theexperiments ofMr. Adams illustrate this radiation very exactly. Nor doIquite understand his last sentence. Ifthe facts beashestates, do theynotplace ironabove copper rather than below it? ‘Mr.Spagnoletti’s statements aremost. interesting. Mr. W. Groves, ofBolsover Street, tells mehehasseen thealphabetical step-by-step machine worked two orthree letters forward by atmospheric electricity ofsome kind onawire between hisplace and Sir Charles Wheatstone’s, Mr.Evershed’s observations onclouds gotosupport thecon- dlasion that thewell-known “return stroke,” and such like ob- servations, prove theconducting nature ofclouds—of some clouds atany rate. Heisquite right inpointing outthat alloscillatory character isliable tobewiped outofadischargewhichhashadtotravela great length ofthin wire, and that thefinding ofaquiet tailof current leaking away insome obscure comer ofatelegraph instru- ment isnocriterion astothevigour orcharacter ofthemain flash whence itarose. . Passing tothecriticisms communicated since themeeting, I have nospecial remark tomake onthe statements ofMajor us LIGHTNING, LIGHTNING CONDUCTORS, Bre. (May 16th. betates. Cardew except tosaythat several seem tobemere statements of personal opinion, rather lightly andcasually made. With respect toNo.30:“We allknow there aremillions of volts,” mypoint ismisapprehended. Itisfamiliar thatthere are millions ofvolts between cloud and earth; itisnot familiar that | there may bemillions ofvolts between thetopofawellearthed | and stout copper lightning conductor and the earth. When ‘Major Cardew says, ashedoes towards theend ofhisremarks, under head “Fourth,” that aconductor ofsmall impedance is desirable, everyone must agree with him; butwhen hegoes onto saythat such aconductor isobtained byfollowing therules of theLightning Rod Conference (oranyother rules forthat matter). itisnecessary todisagree with him. The impedance could not beconsidered inany sense “small,” even ifacolumn ofpure copper, afoot indiameter, wasemployed. The impedance of such acolumn, 100metres high, toacurrent offrequency one million persecond, isnearly 900 ohms. Major Cardew twits mewith having supposed that anearth resistance canbegotaslowasone-thousandth ofanohm,butif hechooses torefer tomySection 84again hewill find that 1 have supposed nothing ofthekind. IfIhadsaidthebillionth of anobm itwould have expressed precisely thesame meaning, i ‘The experience which Colonel Bucknill hashadinconnection with theWar Office conductors, and. theattention hehasfor many years given totheprotection ofpowder magazines, render hispractical remarks very weighty. Iregret they areatpresent 10brief. And now Icome tothe remarks ofthe President himself Forsuch ofthem asarepersonal, Imay bepermitted toexpress tohimmythanks, There isonepoint—that with reference to article 56—where Iwish toexplain mymeaning more fully. ‘Mystatement rans, “Flat ribbon hasaslight advantage over «round rod,butnotenough tooverride questions ofconvenience.” Now itisofcourse perfectly true that extent ofsurface diminishes impedance, that Snow Harris's hollow tubes were 1880} DISCUSSION, air better than Faraday’s solid rods, andthat ifonly onesingle stout Dt.toirs. conductor istobeused, then tape isdistinctly itsbest, asindeed itisthen also itsmost convenient, form. But Iwished toobtain small self-induction bysplitting uptheconductor into detached portions, making each portion fairly thin, For these small conductors also, nodoubt ribbon iselectrically better than wire, Bot will itlast aslong? Isiron ribbon easy toobtain? So long ascommon galvanised-iron telegraph wire is80easy to Procure, itseemed apity toinsist onany other shape ofcross- section, especially since aribbon ofcorresponding cross-section would have tobesothin astobevery liable torust away. All thisIhadinmymind inwriting section 56, Ihadsofrequently insisted onthe advantage oflarge surface inmy theoretical papers, that Ithought itpermissible tothrow itover inthe practical portion forsolely practical reasons, i.e.because toinsist onittothebitter end seemed toentail trouble and expense. But, itmay beobjected, why then did Isaythat tape had only aslight advantage over rod? Well, itisamatter of arithmetic toreckon how much better agiven tape isthan a given rod: IfImake nomistake this istheresult. The self-induction ofarodofsectional radius,r,istothatofa stripofbreadth, J,bothbeingofsamelength,l,verynearlyinthe ratio log21—logr —1 Tog27—Tog db—V? thecurrents ineach case being ofsuch rapid frequency askeep tothe outer surface. Now, unless the rods arevery short, orunless thebreadth of, thetape isenormous—its thinness being likewise excessive, ifit istoconsist ofthe same amount ofmetal asthe rod—this ratio is notmuchgreaterthanunity;andthesamewillbetheratioof theirimpedances. . Similarly thedifference between hollow tube andsolid rodis uotofanygreat practical moment inlightning-rod circumstances. With farlower frequencies, such as100persecond, when frictional ordissipation resistance istheimportant partoftotal impedance, andwhen currents penetrate acertain depth intothe visMOMTstN0,tiomTSiNoCONDUTORS,Bey| oetotee. substance ofaconductor, itisanaltogether different matter, and theadvantage oftube orplate over rodisthen enormous; asSir William Thomson hassothoroughly brought home toeverybody. ‘Suppose, asrather anextreme case, theratio ofself-inductions fortape and rodwere asgreat as2,then thetape would hare halftheimpedance oftherodforcurrentsofthesamefrequency. Such acase Ihave experimented on; but Ishould notliketo insist even then ontheuseofthetape inpreference tothered,if there were serious practical objections onthe score ofomt, unsightliness, want ofdurability, &., tobemade against it, Ifthere arenosuch objections, then tape byallmeans,and the thinner and broader the better." Itmay bejustborne inmind thatdecreasing theself-induction goestoincreasethefrequency, andhencethatifevertheeu- ductor forms alarge portion oftheentire path ofdischarge, the advantage ofreducing itsinertiaisstilllessmarked,forthe impedance depends only onthesquare root ofLinthat case. ‘The President misunderstands meinone place, where be thinks Ihave said that itisnouseconnecting conductorsto water mains. Idonotknow whence this misunderstanding cam have arisen; possibly from section 59,where Isay, “Agud “and deep earth should ingeneral beprovided, independentof “water and gasmains.” This may notbeperfectly clear, bat mymeaning wasasfollows:— Have atleast oneindependent earth, made byawellorotbet suitable means, inaddition towater main connections. Inother words, donotdepend solely onwater main connections. Probably thisisacounsel ofperfection forthecaseofontinsry <welling-houses, butforanimportant building Ithink itmaybe wise, forthese reasons. Mains arenear thesurface, and insome weathers the soil near them may have become dry. Also they ramify into thehouse andinto other people's houses, andwill therefore conduct any violent charge communicated tothen partly into these places, where, byabranch flash toagaspips damage may bedone andgasignited. *Seo Appendis 11. 18803 DISCUSSION. 50 Thave shewn that well earthed mains canthus give offun-De.tots: expected sparks atafairdistance, even when only aLeyden jar discharge isrunintothem;henceIfeelsurethatsomecasesof damage result from lightning being thus brought underground into houses, Having agood independent earth inaddition towater mains isnotindeed asecurity against thissource ofdanger, butitisa step towards it. Idonotpropose toavoid the mains altogether, because insomany places itisnotpracticable. Whether you connecttothemornot,thelightning willgotothemifitchooses, unless they arefaraway;anditisbettertogiveitaneasypath nither than letitflythrough airorsoil, andknock,ormelt,or bam ahole inthem. Itmay sound absurd totalk oflightning imocking ahole; but theconcussion ofairissogreat asto produce allthe effects ofanexplosion. Ientirely agree with Colonel Bucknill, that damage ismost usually done wherever an sirgap isjumped. Ithink compo-pipes aremostly melted where ‘fash jumps toorfrom them than where itsimply passes along them. Withreference totheloadofcoke,Iwasundertheimpressionthatitwas cheap andeasy. Itisnotnovel, and there aredozens ofotherwell-known plans,ifanyarehandier. Lastly, Icome tothe most interesting topic ofall—the cartridges exploded inmetal cases, mentioned byColonel Arm- strong (always provided that they were notmerely ignited by heat), andthePresident's remarks thereupon. Experiments ontheeffect ofscreens have gone onatintervals forsometimeinmylaboratory. Wecansuspendalittle lectrometer-like needle, charged positive atoneendandnegative attheother, inside tinfoil-coated glass box, and candeflect it bymoving towards itacharged ebonite rod. But inorder to succeed, thelidofthe box must besoput onthataLéclanché cellshall notbeable toring abell byconduction along thebox. Inother words, there must beabreach ofcontinuity, oratleast avery high resistance inthecircuit. Sosoon asaLéclanché current can pass, nopracticable motion oftheebonite rodcan disturb theneedle intheslightest degree. But there must be 530 LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May6, betetee. some limit tothis. Astronger charge moved more quickly might dosomething, sowehave taken tofiring charged bullets out ofaminiature cannon towards the box. The bulletsareno joke.Theygoslickthroughthreeinchesofwood,andittakers considerable thickness ofearth tostop them. ‘Theycanbecharged ‘and discharged, orreversed atappropriate stations, bypesing through charged metal;ortheycanbemadetopropelsuddenly apermanently charged disc. But not awink does theneoile show. That only means that thetinfoil coating istoothick. Wearegoing ontogold leaf, orasilver film, and sogradually thinning down till aneffect isobtained. Aneffect must be forthcoming with athin enough conductor, because onecango bygradual degrees tonone atall. Liquid screens can, ofcoane, alsobeemployed, andprobably quiteadecentthickness ofthee will befairly transparent, Iwould suggest, principally bymy ofquery, that theaction willbeasfollows :— Lettheresistance ofametalboxtoacurrent alongitbe8, then when asteady current (C)flows, adifference ofpotentisl (RC) will exist between itsends, whence electrostatic linesof force will radiate both inside and outside, and anelectrometer needle inside willfeelthem. Now, instead ofpassing acurest through thebox, move anelectrostatic charge, Q,with velocity, 1,towards it. Anelectric displacement occurs which results in umomentary current,proportional toQv,inthemetalwaltofthebox, andtoaslope ofpotential some specifiable fractionof RQuw which theneedle may feel. ‘When aspark strikes thebox, amomentary current similarly exists initscoating. Now, ifthemomentary current hasnotime topenetrate the entire thickness ofthemetal soastoflow initsinnermost layer, thennoneoftheslopeofpotentialduetoitcanbefeltinsidethe bor, though outside itwould bemixed upwith themuch grester direct action oftheelectrostatic charge. Butifthecovering it thin enough forsome portion ofthecurrent totravel byitt innermost layer, then anelectrostatic disturbance will occa inside,whichtheneedle,orafrog’sleg,oravacuumtube,ormicroscopic spark gap, may becompetent tofeel. Imays} 1888] DISCUSSION. on however, that frog’s legs donotappear very sensitive tothisDr.tate class ofeffects. Azine copper contact disturbs them vastly ‘more. Now, ifthemetal beiron, thedepth towhich thetransient current penetrate isvery much lessthan itisinthecase ofnon- magnetic metals; hence asuperficial layer, thick enough tomake aneffective screen ifmade ofiron, might be very imperfect sereen ifmade ofany non-magnetic metal. Ontheother hand, theresistance ofiron issoimmensely greater than that ofnon- magnetic metals tothese transient currents, that ifthelayer werethinenoughtopermitaneffecttobeappreciated atall, theslope ofpotential tobefelt might begreater than with ‘copper, oreven with tinorlead. ‘May IaskthePresident tobegood enough tosayifanyor allofthis isnonsense?and,ifitisnot(beingperhapsonlywhat heinstinctively had inhismind when hegave ahint about iron cages atBath—the hint which herepeated inreply toColonel Armstrong), then hewill probably beable tocaleulate off-hand howmuch effect istobeexpected inanygiven case, and tolay down rales whereby thedisturbance liable tobefeltthrough a given thickness ofadefinite metal can beatonce calculated. And now with respect tothe“Practical Suggestions,” which I provisionally made atthe end ofmypaper inorder that they might receive the betefit ofcriticism; and between which and themain body ofthepaper Ihave always drawn aclear distine- tion, Several have been criticised, and some have been shaken. MayIquickly runover thelist,indicating those which Istill strongly uphold and those which Iregard asdoubtful ? Nos. 51,52,and 53,Isuppose Mr. Symons would say,are “reprinted from the Lightning Rod Conference.” They have, certainly, afine ancient flavour oforthodoxy about them. But hewould nothave methrow over everything, both badandgood! They seem tomegood. Nos. 54and55Istrongly uphold. ‘No.56Ihave indicated myreasons forprovisionally main- taining. Ifthey arebadreasons Itrust thePresident will at once crush them out ofexistence. oaLIGHTNING,LIGHTNINGCONDUCTORS,Bre.[May168, br.toiee. No,57Iregardasveryimportant, especially itslattersentence.Itisjustoneofthepointswhereintherulesofthe future willdiffer from therules ofthepast. ‘Nos. 59,60,61,62arevery much open todiscussion, Nos.63and64,1think,aresound.ButverylikelyColonel Bucknill’s addition to63isanimprovement. No,65isvery doubtful. ‘There are,asColonelBucknillpoints out,very serious exceptions toit,even ifiteanever beregarded asa rule. Nos. 66and 67aresound, Ithink. No.68isafact.(This,MajorCardewwillsay,is“dogmatic.” ‘That also isafact.) No.69isacounsel ofperfection: intended forpowder magr zines, notfordwelling-houses, SirW.Thomson said it,orsome- thinglikeit,atBath.Itmustberemembered, however,that «gasometers” aredamaged when struck, according toreports in newspapers. Nos. 70and 71are very doubtful. Ithrow them out suggestions which experience must settle. No, 72is,Ithink,allright,butafterthewords“aloadof “coke” one may add, orany ofthewell-known earth contod arrangements. No.73hasbeen wholesomely criticised. Ithink Iamsafeia stillsaying “itisnotanunmixed good.” Butvery likelythe gain outweighs theloss. Infact,IhaveintheMannlectures advocated theproceeding asgood onthewhole. No.74Ishould beglad tobeable toomit, butseenopresent chance ofit. ‘Nos. 75and 76have been well criticised. Iquite feelthe force ofthecriticisms, andamglad totake refuge inNo.74.At thesame time arighteous substitute forNo.76,ifitbewrong, it verydesirable. Themiddle partofNo.76(achimney withinside metal shaft) isafrequent andvery difficult case. Itembodies theadvice which atpresent, forwant ofbetter, Igive. Boiler firemen, engine tenders, and dynamos, would beapttobe damaged, Ifear, ifcontrary advice were followed. No.77is,Ithink, generally true, forsuch things asrir 1900) DIBCUBSION. 8 water conduits under eaves, forpicture rods, &e,; not,ofcourse,Drtotes foramiscellaneous collection ofmetalobjects.No.78is,Ithink,right,ifnottootroublesome inpractice. A crown oflong points leaning well over into thesmoke may doas well. No.79probably belongs toMr,Symons andtheLightning Rod Conference. ‘Nos. 80,81,82,83areintended toapply only todesperately important places: dynamite factories, petroleum tanks, and such like, ‘They areofcourse perfectly open tocriticism, No. 84iscorrect. Nos. 85,86,87are hints towards more elaborate methods of testingthantheout-of-date planatpresentinuse,Icallitout ofdate because itisbased upon theuntruth ofNo.57,andupon entire ignorance (very natural afewyears back) ofthegreat obstruction offered bygood conductors. Itisbetter than no testing atall,butitisextremely inadequate, inthat itdetects onlyone,andthatacomparatively unimportant, kindofflaw. ‘Nos. 88,89,90,91have todowith lightning “protectors,” and,Isuppose, areorthodox andindubitable. APPENDIX I ‘Tarory orBCincurrs, of“ALTERNATIVE Pata” EXPERIMENTS, AND oF SIDE FLASH. Consider acouple ofjars connected tothe terminals ofa machine bytheir inner coats andtoawire circuit bytheir outer conts(Fig.1). ayTheyformanordinarycircuitwithaLL ~, capacity inserted equaltothesemi-harmonie =1 2 mean ofthetwojarsseparately, andanair * gapofadjustable width atA;andthemaxi- mumdifference ofpotential producible init cs isdetermined bythedistance oftheAknobs. Fro 1. When thedischarge occurs,acurrentflowsofcourseequallyround thewhole circuit, but thepeculiarity isthat uptotheinstant of 4 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[ay164, Petolee. discharge theBportion ofthe circuit isat uniform potential. IfagapexistsinBalso,asitwellmay,theterminals ofthegup may likewise beatthesame potential uptotheinstant when the rush occurs. The discharge will, asusual, beoscillatory unles theresistance ofthewhole circuit betoogreat ;and theperiod ofoscillation will beapproximately 2 «/ (LS), where 8isthe capacity ofthetwojars inseries. Now number thecoatings ofthetwojars asshewn inthe diagram, and write down their electrical condition before snd during thedischarge spark atA:— Before discharge...| +Q/+V)-Q|-v|-q@ 04+Q)0 Alter}period|}o|ofo|ofof-v]ofay Alterperiod a|-v|+e}+v} +e} o}-9} 0 After3period o|ofofo|ofsv}oter Afterawholeperioa...| +Q|+v|-@|-v]=a}0}sa}0 andsoon, with gradual damping (thedamping being omitted in thetable forsimplicity). ‘Thus, then, between theends oftheBwire exists atregular intervals almost thewhole difference ofpotential which isable tojump theairgapatA.Strictly speaking, thedifference of potential israther less than that corresponding totheAgin thas :— ‘Theequation tothecurrent atanyinstant isaccurately R cmd 2hainpt, where V,istheinitial difference ofpotential corresponding (0 theAspark, and where ® 1 :p=J/(cs- an) Now ifTyistheportion ofthewhole selftindnetion which « 8) DISCUSSION. sss corresponds totheBlength ofwire (i.e, subtracting from thetats. whole Lthe part belonging tothe Awire), and ifR,isthe resistance oftheBwire, itsimpedance is«/((pI,}+ R,'); and hile acurrent, C,isflowing through it,thedifference ofpotential botwoen itsends istherefore /((pL,)*+ Rt) C. Now thecurrent flowing through attains itsmaximum value one-quarter period after the Aspark has commenced, i.e, ina time Fo(moreexact,inatime1tan287), andinserting thisinCwegetthemaximum possible strength of current, viz. :— aR aX. [aptG=sn ¢P Hence the maximum possible difference ofpotential between theendsoftheBwireis ‘R,\* wR u+(A)t-F,veyMifolieipl; thaig,acertain fraction ofV,,thefraction being totalimpedance ofBwire a <. Aaeintiapelcnensfabaiecean,%@tmpingduring4period, Very often asufficient approximation tothis is -28 4pL andifthewires arethick and short, ornon-magnetic,andthe capacity big, thedamping during thefirst quarter of period is oftensosmallthatmerelythefractionJwilldo.sufficiently well, Sothen, ifasupplementary pairoftapping knobs becon- ected totheends oftheBwire, asshown inFig. 2,and if Aheirdistance beadjusted tobe[thsoftheAdistance, a spark isliable topass atthese knobs. This iswhat TcallaBspark, and thespark gapaffords an Alternative path totheBwire, orviceverea, 86 LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May16 - iat Cé) Fo.2 ‘There isnoneed totapoffthe whole ofthe Bwire. Any portion however small willserve, provided theappropriate rate ofL,isused. The length oftheBspark measures thedifferen ofpotential needed topropel thecurrent through theportion wirewhichisthustapped. Ofcourse,ifaBsparkactuallyoccur,itintroduces disturbance; theknobsshouldbesetsothatitjot fails. ‘There isonething nothere explicitly stated, butwhichbs tobetaken intoaccount incalculating thevalue ofR,andtht isthelossofenergy byradiation, With small jars andcieats thislossisvery great, anditincreases thevalue ofRenormous. Seeapaper ofmine intheforthcoming July Philosophical Magazine. With bigjarsandcircuits itmay besafely omitted; theexperimentally observed Bspark will agree with calculation. Butwith small jars, ifitbeomitted, theobserved Bspark wil bealways lessthan thecalculated.* Inthisway ameasure ofthedamping due toradiate dissipation ofenergy canimmediately bemade. The observation ofMr. Wimshurst about the neutral poiat, indicates atonce that thistheory alsogives thelength ofside flash obtainable from thewire, Let anypart oftheBwirebe put toearth, orletitsnatural neutral point befound, tha theVcalculated asabove foranyother point gives thelengtho side flash obtainable from that point toearth. Side flash isinfact aspecial case ofthealternative-pth experiment. With asymmetrical wire like this, everything insulated and thejars equally charged, the neutral point #| naturally themiddle. Butwith alightning conductor thelore | endistoearthmoreorlesscompletely, hence fromtheactus)| 1860] DISCUSSION. 380 bottom of the wire no side flash should be obtainable. One Dstater. always willbeobtainable, however, owing tothe impossibility of making anon-resisting earth ofinfinite capacity. Higher up,the length ofside flash obtainable must beitslength atthebottom plastheVcorresponding toheight ofpoint tried. Themaximum sideflash isobtainable from thetopofthewire.Thestrengthor energy ofthespark depends, ofcourse, onthecapacity ofthe body receiving it(ifinsulated);being}SV*,whenViscalculated ssalready said. Ifitbeanearthed body, then thewhole discharge divides itself between thetwo paths, according tothelaws of divided current appropriate tothese conditions. Intesting aconductor, aspark should begiven tothetop,and thelength ofside spark obtainable atthe bottom should be abyerved. Allelse canbecalculated, except insofarasthere maybedefects inthevisible portion oftherod. APPENDIX II. ResistaNce AND IMPEDANCE FOR FREQUENCIES COMPARAMLE TO A ‘MILLIoN PER SECOND. AfJPisthefrequency ofcurrent conveyed byawireoflength Landofordinary resistance r,madeofasubstance ofpermeability #5then itsresistance tocurrents ofexcessively high frequency is R=V(iplan), hence the resistance ofsoft iron isimmensely higher than that ofanynon-magnetic metal. ‘The self-induetion under the same circumstances is Ri+? where L,refers solely tothespace surrounding theconductor. The inertia portion oftheimpedance is* pl, +R, *Atfirm sight itmay seemaaifTweremakingamistakeinhavinganRterm fathepurely inertia part oftheobstruction,butitinquiteright.Tieterm Lappenstorepresentexactlythemagnetsation ofthesubstanceofthewir,wofar sisouterskinismaguetsed,Vol. xvin. 39 6 LIGHTNING, LIGHTNING CONDUCTORS, Ere. (May 163, tates. ofwhich thefirstterm isfarthebigger athigh frequencies, even foriron, unless thewire isvery thin. ‘The total impedance is Virbi+2pL,R+28, ofwhich, again, thefrst term usually fareclipses theothers. Numerical Ezamplez—1. Let thelength, J,ofconductingrodbe10metres,itsdiameter 1centimetre, andletitbebent into theform ofacircle (ifitbestraight, there will bebut little difference); take =1forcopper, or900 foriron; specific resistance, 1600 square centimetres per second forcopper, or7 times this foriron ;andletp=2mX10°persecond.. Then, whatever the substance oftheconductor, L,=12,000 centimetres ; while thecommon resistance, ne{-002ohmforcopper. “014 ohm for iron. Hence the effective resistance is -08ohmforcopper. Re{esohmsforiron. ‘The inertia portion oftheimpedance is, 754+-08=75°5ohmsforcoyr= Plt ReeyTos=abvcher inn ‘The total impedances arepractically thesume—viz., {55oanforcopes 82-0 ohms for iron. 2.If,instead oftaking rod10metres long, weconsider @ length 100 metres long, ofthe same thickness, these quantities become :— L,=162,000 centimetres. __§02ohm forcopper. "(0-14 ohm foriron, ne={8 ohmforcopper. {63:0 ohms foriron. « 1,003ohmsforcopper. Inertiapartofimpedance{ressShaefoe | 1,003ohmsforcopper. Totalimpedance{57ohmsforino,| 0 DIsOUSBION. wo 3,Lastly,forawire100metreslong,but1millimetre inDe.Later diameter, the values would be ‘L,=208,000. re{2ohmsforcopper. ‘14 ohms for iron. nmfake conn 1630 ohms for iron. Inertiapartofimpedance, ‘1310+8=1,318ohmsforcopper. Pl+R= {iste+630=1,940ohmsfori . 1,818ohmsforcopper. Totalimpedance{sooohmsforiron. Althis supposes the frequency tobedetermined indepen- dently ofthe conductor considered, and toremain the same; butasthe conductor increases inlength ithas atendency to decrease thefrequency;andtbatisthemeaningofmysentence insection 28,towhich Major Cardew objects, “and ofany “moderate length, such as100yards orless(not many miles).” Tought tosaythat thehere calculated values forRdonot take into account atallthelossofenergy byradiation. This willalways gotoincrease R,often very perceptibly, sometimes enormously. Iwillgointo thisfarther insome other place. ‘These examples illustrate sufficiently well the comparative behaviour ofiron and copper under well-marked and frequently occurring conditions. Ihave chosen thefrequency of«million asecond, because Ihave shewn reason forbelieving that itisnot atallunlikely toapply tothecircumstances oflightning;the capacity discharged perflash, andtheself-induction ofitspath, being neither ofthem very big. But while weareabout ititis instructive andquite easy to write down thevalues forsome considerably lower frequencies : notforslow frequencies such asalternating machines give, the theory forthem ismore complicated, butthesimple theory will, dofor,say, 10,000 complete periods persecond. ‘The result, will bedistinctly different. Nolonger does inertia constitute the hole oftheobstruction foriron, though itstilldoes forcopper andforiron itconstitutes thelargest part. eo LIGHTNING, LIGHTNING CONDUCTORS, Ere.(May166, llas oevance,Resianee, oe taal | _| jets ® onm.|onme|oim| fom|208 | sw Oometrerod 1cm, thick wen} |tae ae yer] 08|to|a0 10-metrerod1em.hick{er || Iron os wos|ast|Conner *wo|ne| 1o00-metre wire mm. thick | ‘Thedepthpenetrated bythecurrentintothesubstanceofthe wires,isdefiniteatagivenfrequency—unless thewireistoothio toleave acentral margin, and isindependent ofthediameterof thewire; atleast forthese high frequencies, Itiseasly calc lated with fair approximation, thus, the sectional radius ofthe wirebeinga:— aroda + 7e =R whence da,thedepth effectively penetrated bythe current, ot thethickness ofconductor practically made useof,is— f ini incopper yymillimetre;Forthemillionpersecondfrequency {amee {incopper y'zmm; For the ten thousand per secondfr orthetenthousandpersecondfrequency{>CPP= Itmay beafter all,therefore, that Iamwrong insaying tht rodisanything approaching asgood sstape forconductor. It ismearly asgood inrespect ofmere impedance, butwhenere there isanychance ofthewire being melted, then tapeisft etter. Rodought tobeapttohave itsskin burnt offit,unles thecentral corehastime toexert anycooling action bysbaring theheat.* Butitisbecause Idoubt whether decently substanti "*‘Thespecimens exhibited byMr.Preece, ofcopper wireincipienty fv! tylighting internally areinterenting. They tayhve been fased bytheddbeatailofwcarret;theonsidecoolingmostrapidly.TheyTookaiftheybi been bottet inside, andifgoanexplanation isneeded buttheyareootLik"pheavethefoundations ofelectromaguetinn, 1880 DISOUSSION. se conductors areinanyrealdanger from heat that Ihave asserted br.tage theadvantage ofgreater surface tobebutsmall. Iwait, however, foranexpression ofauthoritative opinion from thePresident onthese points. APPENDIX III. Ox THE MeLtiNe oFCoxpucrors. ‘The list offused conductors attheend ofthe Lightning Rod Conference Report, Appendix J,isvery short, butshortasitisitincludes thingsnotquitefreefromseriousmisleading. Overand overagain ithasbeen truly asserted that wherever there isanarc ora flash toaconductor damage islikely tobedone, Terminals which have toreceive theflash should always bethicker than the wire which has only toconduct it. ‘This must beregarded as very ancient and orthodox, aswell asvery true, Inow run through theshort listofdamage, andanalyse it.The table is headed, “List oFMerats Mere.” 1.“Copper rod, ‘35 inch diameter.” This was anupper terminal, tapering from one-third ofaninch diameter at thebase toapoint, and only 9}inches long altogether. ‘This terminal was “nearly allmelted.” 2.“Copper rope, “31inch diameter, atNantes.” Callaud, «Traité,” page 89. 3,Rope,saidtobe-7inchdiameter, atCarcassone, Callaud, «Traité,” page 89. These Iwill refer todirectly. They were notfused, but broken, oreaten into, orotherwise “burnt byuse.” 4,“Tron rod, ‘2inch diameter.” This was afew inches melted from thepoint ofanupper terminal, andsome of the links ofachain. 5,“Brass rod,2inch diameter.” This wasatapering ter minal, 10inches long, ofthe given diameter atthebase, and itwasonly melted forone-fourth ofitslength. The implied statement inthereport is,therefore, that @brass rod}ofaninch indiameter wasmelted. The fact is,that 02 LIGHTNING, LIGHTNING CONDUCTORS, Ere.[May188, tmWedge. 2}inches was melted off«sharp brass point! Fortunately in thiscase,andincase1also,thebodyofthereportitselfcontains thematerial capable ofoverthrowing thismisrepresentatioa. 6.“Copper rod, perhaps *13inch diameter.” This was common bell-wire, and itwas legitimately destroyed, bat still itprotected. ‘That isthewhole list,anditamounts tonothing more thia 1bell-wire, and tocases 2and 3,the account ofwhich [now proceed totranslate from the treatise ofM. Callaud. The Carcassone case isone ofthetwo Mr. Symons quotes inbis remarks (the other iscase No. 4,above). Itistheouly onethatsoundsimprobable, andtheevidence forit.seemstome weak; butIleave readers tojudge. The evidence fortheNante: eases, such asthey are, isperfectly good. Extract from “Traité desParatonnerres,” par A.Callaud, p.89:—“The conductor ofthe Church Sainte-Croix atNeste ‘was acable ofredcopper, acentimetre indiameter; itmas “formed ofseven strands, each consisting ofseven wires,the “wiresbeingonemillimetrethick.Iwaswitnessofastormand “ofviolent flashes which traversed it,and itshowed notrace of “deterioration, ‘Thissizecantherefore bepermitted, thoughit “seems tomeslight. The cable which existed before thatof “which Ispeak, alsoofredcopper, was found broken byafish “and damaged over apart ofitslength; ilavait8millimétres “Iknow ofconducting bars, 5millimetres, which asingle stom “has deteriorated and eaten into inaway that tenyears ofrot «would hardly accomplish. “M. Viollet-le-Duc, whose words Ihave had thehonour of “quoting, has seen atCarcassone some cables oflightning “conductors burntbyuse.Ilsavaient18millimatres. ‘Inthis “*town;' hetells me, ‘storms are frequent—daily, incertain “geasons.’” Insuchacasethesizeof18millimetres willbethea“insufficient.” ‘This lastisamost: vague account. ‘Thematerial isnotspec fied, norisitperfectly certain whether the18millimetres refer tothediameter, orwhether itmeans that itconsisted of18wires eachamillimetrethick.Evidently,however,M.Callaudsuppares| 1880.) DISCUSSION, . ‘568 ittomean the diameter,andmostlikelyitdoes,“ButwhyintheDr.tater plural? And does “burnt byuse” mean anything more than thatsome ofthethin wires were burt orfused together, orthat thecable was oxidised superficially ? Considering the exceptional character ofthe testimony, if understood intheLightning Rod Conference sense, itisapity it issecond-hand. APPENDIX Iv. Ox ConpITIONs UNDER WHICH PoINTS CAN BE PREFERENTIALLY Srauck mwCase B. Referring toMr.Wimshurst’s observation oftheeffect ofthe sign oftop-plate, the following isanextract from anApril notebook kept bymyassistant :— Large sphere (ordome), knobs, andpoint, arranged between twoplates soastobeequally struck byaBspark. The plates are connected tothe outer coats ofthe two small or pint jars, whose inner coats are connected tothe machine, between whose terminals occurs amoderate Aspark. Ist. With thetop-plate positive. Pomevesssssee245centimetres., Large knob... sn 36 Distancesof[saatknob...a3B Point... we oe BB yy 2nd. Top-plate negative, Dome ... vr sve 26 centimetres, Large knob... .30 Distances small knob... ... 37 Point ... 1. ws 80 ” Lengthening theAspark makes thedistance atwhich the point isstruck less,” The following measurements have been made quite recently, large jarsbeing used, butthevigour oftherush being diminished insome cases bymaking theAspark (i., thedistance between themachine terminals) quite short. ws LIGHTNING, LIGHTNING CONDUCTORS, Ere,[May16th, tig Two gallon jars similarly connected, instead ofthe pint jars. Objects arranged between plates tobeeasily and about equally struck, asbefore. First, with theAspark 1centimetre long. Top-plate negative. Large knob 1-2centimetres. Distances fromtop-plate ofJSmallknob14yy Point 4. 24 » Top-plate positive. Large knob .,,_1°5 centimetres. Distancesof4Smallknob.....22 | Point... 20 ” Lengthen Asparkto5centimetres— i Top-plate negative. \ Large knob... 3°4centimetres. Distances of4Smallknob... ..35 yyPoint. we oe 3D » : Top-plate positive. ' Large knob... 40centimetres. Small knob... 42 yy Point 0 we ae 39 ” 1 Repeat with Aspark ubowt 5centimetres, buttheBdistances made greater. Top-plate positive. Large knob... 5-4centimetres. Small knob... 4. 47 Point 1. se oe AT Top-plate negative. Distances unaltered, and all are struck occasionally asbefore, but the small knob gets struck rather more often than the others, and the large knob rather less often. Hence itisclear that, under circumstances when the rush is really impulsive, the difference between positive and negative top-plate, which Mr, Wimshurst called attention to,does not exist. Itonly exists insofarastherush isgradual. 1839.) ‘ELEOTIONS, 65 Aballot fornewmembers took place, atwhich thefollowing were elected :— Foreign Member: Carlos Monteiro eSouza, Associates: Edward W.Cowan. Emest B.Vignoles. ‘W. P.James Faweus. Henry Bloomfield Vorley. Richard 0..Drummond. George Wilkinson, Jesse Kemsley. Cecil Wray. Students: Alexander J.Protheroe. | Herbert Edward Starr. The meeting then adjourned. ‘The One Hundred andNinety-fifth Ordinary General Meeting of theInstitution washeldattheInstitution ofCivilEngineers, | 25,Great George Street, Westminster, onThursday, May 23rd, 1889—Sir Writ THoMson, F.R.S., President, inthe Chair. ‘The minutes ofthe Ordinary General Meeting held onMay 16th were read and confirmed. ‘The names ofcandidates for admission into the Institution ‘were announced and ordered tobesuspended. Donations tothe Library since the last statement were announced ashaving been received from Major Francis J.Day, R.E.; Major P.Cardew, R.E., Member; C.8.James, Member; towhom thethanks ofthemeeting were duly accorded. The Paesinent: Gentlemen, theorder ofthepapers has been altered from that which appears inthenotice. Mr. Mordey has specially requested that myshort communication beread first andthediscussion onittaken before thereading ofhisown paper. Itisavery kind suggestion onMr. Mordey’s part, and ashebas expressed thedesire that itshould be0,Iaccept theproporal, andbegtothankhimforit,although Ishouldbeexceedingly sorry ifany such change inthe order should inthe slightest degree interfere with the communication ofhispaper tothe Institution. Itmust not, infact, beallowed todoso. Mr. Mordey’s paper isamost important one, onasubject ofvital | interest toelectrical engineers, anditisnecessary that itshould | bethoroughly andcarefully read and listened to,and that there should beasmuchtimeallowed foritsdiscussion ascircumstances |permit. | ~Whatever discussion theremaybeonmycommunication must | therefore naturally bevery short, andIhope wemay soarrange |itthatwemaynotoccupymorethanhalfanhourofthetimeof| 9) SECURITY AGAINST DISTURBANCE, Ere. wr Ishall now, without further preface, read the following peper:— ON THE SECURITY AGAINST DISTURBANCE OF SHIPS’ COMPASSES BY ELECTRIC LIGHTING APPLIANCES. BySiWiLL Tuowson, D.C.L., F.R.S.S. (L.&E.),President. The danger tobeavoided issufficiently explained inthe following short statement byMr. William Bottomley, which appeared intheNautical Magazine forDecember, 1885:— “The following example ofacase which might occur inany “large ship, will show theamount oferror which may bepro- “duced onthecompass” [bythe electric lighting apparatus] “unless precautions aretaken toguard against it. “Suppose amain lead from theengine-room tothefore part“oftheship,tolightup100lamps,isbroughtalongthecentre “oftheship. Itmay beatadistance of10metres, or33feet, “from thestandard compass, and will ran almost underneath it. “Ifwesuppose that each lamp takes oneampere ofcurrent*there will beacurrent of100amperes altogether inthis lead. “Now, theeffect ”[ofaninfinitelylongstraightcurrent]“onthe “compass” [above it]“at adistance Dincentimetres isgiven “by theformula . 7 2X woPoa “where Cisthe current inamperes and Histhe horizontal “magnetic force. Inthis case wehave C=100amperes and “D=1,000 centimetres. Therefore v= 20 =008 1,000 HH “AtGlasgow thehorizontal force may betaken as0-15 inc.g.s. “units,therefore theeffectonthecompass willbe92=11. “Thiswillbeexpressed indegrees bymultiplying by57-3,the “number ofdegrees intheradian, orangle subtended atthe “centre ofacircle byanareequal inlength tothe radius, 83 SECURITY AGAINST DISTURBANCE OFSHIPS’ [aay sr, «Therefore, theamount oferror produced bysuch acurrent on thecompasswillbe523=746degrees. “The foregoing refers to@single wire and acontinuous “current machine, but ifanalternate current machine is «employed noeffect willbeproduced onthecompass even when “the ship's side isused forthe retum, ,When acontinuoos “current machine isused, thedanger ofproducing anerroron “thecompass canbeavoidedbyusingtwowiresclosetoone “another, but these wires should bewell insulated from the ship's side, Ifinany way one ofthewires isbrought in contact attwopoints ofitslength with theiron oftheship “there may benochange observable inthelighting, but the “current may produce asmuch error onthe compass asit «would ifthere wasonly asingle wire, “The following points should therefore beattended toinall«casesoflightingshipsbyelectricity :— «First.—With continuous current machines two wires, well «insulated, should always beemployed. “Second.—The insulation of the wires should be tested “periodically;ifanyconnectionwiththeironoftheshipis found,the fault should atonce be corrected. “Third.—When analternate current machine isused, a «single wire may beemployed and theiron oftheship used “to complete the current without producing any effect onthe «compass.“Whatmakesthisquestionofthegreatestimportance isthat «theerror may beproduced without ever being detected bythe «officers ofthe ship. Onboard ship the errors ofthecompass “are usually determined during theday, inthe morning and «afternoon, buttheelectric light isonly used atnight. The «captain may therefore carefully determine hiserrors every day, “and sethiscourse quite correctly;butatnight,whenthe electric light istured on,the ship may begoing sereral «degrees offherproper course, although she isbeing correctly «steered bythecompass, “Tn connection with the lighting ofships with electricity, see] COMPASSES BYELECTRIC LIGHTING APPLIANCES. 569 “there isanother point which should also beattended to—that “is, theposition ofthe dynamo. Ifitisplaced near aniron “balkhead, the upper end ofwhich isnear the compass, the “bulkhead may become magnetised byinduction sopowerfully “that itwillproduce aconsiderable error onthecompass.” ‘Thesubject wasalsoreferred toinMr.Bottomley’s paper on “The Magnetism ofShips and the Mariner's Compass,” read before theSociety ofArts, January 28th, 1886, and published in theJournal oftheSociety forFebruary 5,1886. Inthe discus- onwhich followed, and inwhich Captain Creak, oftheAdmiralty Compass Department, Mr.Alexander Siemens, and Dr.Hopkinson tookpart,itappeared thatinthreeships,lightedonthesingle- wiresystem with direct currents, small but not unimportant errors inthe compass, due tothe lighting currents, had been actually observed. Since that time several cases have been reported tomeoflarge passenger ships, lighted with direct carrents ontheone-wire system, inwhich asmach as4°or5°of ‘ertoronthecompass hasbeen produced bytheelectric lighting. Inthelatest ofthese cases, afewweeks ago, anerror of4°onthe North course wasfound when thelight wasput on. The light ‘asputonand offseveral times with the ship’s head North, and eery time the same error wasproduced. ‘Theprecautions forsecurity which Ihave tosuggest are— 1,The useofthetwo-wire method exclusively (unless, which isnow rarely thecase, alternate currents areused). 2.The most simple andconvenient testforfaults ofinsulation capable ofdisturbing any ofthecompasses onboard isalamp setupintheneighbourhood ofthedynamo, with oneend per- manently connected with the ship's iron, and aswitch forreadily patting itsother terminal inconnection with either ofthe dynamo mains atanytime. The switch should occasionally be moved each way bytheengineer incharge, and ifeither motion lights thelamp toany visible degree, adefect ofinsulation on thecorresponding main isproved, and ought tobeimmediately corrected. But unless thelamp islighted tofullbrilliance, the fault iscertainly notsogreat astosensibly disturb acompass. Fullbrilliance proves only one fault; andthere must bemore 30 SECURITY AGAINST DISTURBANCE, Ere. [May 220, than onesuch fault before any error ofpractical importance em beproduced inanyofthecompasses. 3.Carethatthereisnotmagnetic“leakage”fromthedynamo (aspracticalmen,guidedbyFaraday’s ideas,theory,andlanguage, have now taught thescientific world tocallit)enough toproduce ‘any compass-disturbance ofpractical moment. Capt. Creak, speaking atthebeginning of1886, intheSociety ofArtsdie cussion previously referred to,said that inone ship thedirect compass-disturbanee produced bythegenerating machinewas “felt through adistance of55ft.,” and across iron bulkheads, andthat itwasperceptible also inother ships oftheRoyal Ney electrically lighted onthetwo-wire system. ‘Myimpression isthat theimproved dynamos now made hare much lessofmagnetic leakage than those made prior to1886, butwestill want information astotheir disturbing magnetic effect atsuch distances ashave tobe considered inconnection with thecompass question. 4.Toascertain that there isnoperceptible compase-distar ance, orifthere isany totest itsamount, thecompass should be observed while thecurrent through thedynamo isstarted and stopped, either bystarting andstopping thedynamo itself, orby making and breaking thecircuit ofthe field magnets. This should always bedone before the electric light installation i taken over from the contractors. Itisbest and most easily dove when the ship isindock, orlying steadily atanchor. Onno‘account ought ittobedelayed, inanewship,tillshegoesoat forcompassadjustment. Adetermination oftheamountsofthedisturbance, ifany, forallcourses oftheship canbemade byaid ofmy deflector without moving the ship, But asufficient practical test may bemade byfirst observing the effect of starting andstopping thecurrent, onthe compass asitstands; thenadjusting asmallmagnetplacedon,orsupported alitleabove, theglassofthebowl,todeflect thecompass about45° first onone side and then onthe other side ofitsundisturbed position, andineach case observing theeffect ofstarting and stopping theelectric current. This effect ought nottobe# much as2°inany ofthethree cases. 1869.) DISCUSSION. m 5.Asmall electric lamp, with itstwo electrodes insulated andtwisted together intheusual manner, may safely (and with verygreat. advantage inmost cases ofelectrically lighted ships*) beused tolight thecompass. The effect, ifanyperceptible, of itscurrent onthe compass ought tobetested inthe manner described inNo. 4. Staff-Commander Cazax, R.N., F.R.S.: Iamverygladtoheargut.con. ‘hat Sir William Thomson has told usonthe subject ofthe effects ofsingle-wire leads and dynamos oncompasses inelectrical installations onboard ship. Icanonly speak from experience as regards ourmen-of-war, where theelectric light hasbeen adopted forsome years past. Wehave onlyoneship—the “Polyphemus”— fittedwithsinglewires;inallothervesselsthewiresareduplicated,theleadandreturnwiresbeingclosetogetherandledalongtheship's side with double wire branches tothelamps. There has never been any trouble caused bythe wires—our troubles have beencaused bythedynamos. Asaninstance, Imay mention the“Northampton,” where three dynamos oflarge external field werefixed about 37feetfrom thestandard compass onthebridge. Near this latter were twoofSirWilliam Thomson’s compasses, amdonthedeckbelowtwosteeringcompasses. Imadeexperi- nents forhorizontal and vertical force, aswell asnotations ofthe ‘angular deflection ofthecompasses. With allthree machines running, adeflection of3°to5°wasobserved atthemost distant Pasitions onthebridge, and asmuch as11°was observed atthe steering compass. These experiments distinctly pointed tothe dynamos asthecause.Thesemachines wereunfortunately placed With their redpoles uppermost. ‘The effects ofeach machine in deflecting theseveral compasces wasnext tried. First, No.1on andoff,then No,2,and lastly No.3onand off. The greatest deflection produced byanyonemachine was6°—with allthree on, 11°, Besides the“Northampton,” twoother ships have given trouble—the “Curlew” and“Landrail,”—all distinctly traceable to thedynamos, andeither from their being toonearortheir poles *Those whoknow thetrouble ofKeeping thebestofordinary compas lamps Aight, forthesteering compassonthebridgeofsomeofonrcron-Channel steamers inwinter gale. will appreciate this advantage. m SECURITY AGAINST DISTURBANCE, Bre,(ay20, qaom being placed inwrong relation tothecompasses, orfromthemachines beingofunsuitable typeforships.‘Therearemachines theexternal field ofwhich issosmall that itmay bedisregarded. Inthe“Northampton,” theofficer ofthewatch could never tll whether theengineers below were running one, two, orthre machines;whilsthehadthreedeviationtablestoconsult,aceow- ingtothenumber ofdynamos atwork. This state ofthings would beintolerable inanyweather, much lessinfogs, andthe machines are toberemoved from the vessel inconsequence Thereisnothingtopreventaninstallation oftheelectriclight being made onboard perfectly harmless asregards thecompas,if suitable dynamos and double wires areused. Itisagreat pleasure tometofind that Ihave theauthority ofSirWilliam Thomson toconfirm much that Ihave said, knowing that wemust allgladly defer tohisopinion asoneofourhighst authorities inthese matters. wre ‘Mr.ALEXANDER SIEMENS: Mr.President, Iamvery gladth you have brought this subject forward sothat wecandiscasit here, because Ineed nottellyou Iamastrong advocate ofthe single-wire system, and the experience ofmyfirm hasbeen tht hardly anytrouble hasever been experienced. ‘This isprincipallyuetoourhavingcarriedoutparagraph 4ofyourpaper.Itlisbeenourhabitalwaystotrytheeffectofthelampsandthe ynamos onthecompasses, and inmost cases there hasbees noeffectatall.Infact,inallourpractice, whichextendsoveranumberofyears,andincludesagreatnumberofships,webave really hadonly onesingle case inwhich anyeffect: wasproduced onthecompasses, and Ihave brought with meacopyofthe letters which passed between myfirm andtheshipping compsay atthetime, because Ithought itwould bebest tosubmitthe matter asitwas discussed atthat time, The shipping company wrote tousthefollowing :—With “reference tothe visit ofyour Mr. Siemens tothesteamer to “day,andtotheconversation withhimonthesubjectoftheeffect. produced onthestandard compass oftheship bythe “electric light currents inthecompanion below it,weshallbe “glad ifyou will begood enough towrite usafullletter 18803 DISCUSSION. os “detailing thecause ofthisoccurrence (the onlyonewhich hasie. yetcome toour notice), and the re-arrangement ofthewires “which you propose tomake inorder todoaway with the “disturbing effectcomplained of.Weshallbegladatthesame “time tohave anexpression ofyour views with reference to “Mr. Bottomley’s letter tothe Nautical Magazine forthe “carrent month, onthis subject.” ‘That istosay, the letter which you, sir,refer toinyour paper. Weanswered asfollows:—“ Webeg tosaythat onTuesday “last wemade experiments onthis sbip, and were unable to “find anydeflection ofthe compass needle due totheelectric “light currents,astheshipwasthenlying—i.e,,withherhead “pointing 63°tothewest ofnorth; butitispossible that when «theship isotherwise positioned there may besome disturbance, “although our experience infitting some hundreds ofships on “the single-wire system hasnotbrought any such instance to “our notice. “We believe the cause ofthe observed deflection tobethe position selected forthecompass with respect tothewires, and “we have offered tore-arrange thewires soastoneutralise their «effect;"—[Ishouldsaythatthecompass wasputupafterthe wires hadbeen fixed, andwehadnoknowledge that thecompast wastobethere]—“ butthecaptain wished usnottodosobefore “the ship sailed this time, inview ofthe disturbance tothe “ship's woodwork ourwork would involve, assuring usthat he “anticipates noinconvenience from this effect onthecompass if “it befound tohepermanent and aconstant quantity. We “have, therefore, taken nofurther steps inthematter.”—[I may sayfrom that day tothis—that is,during four years—we have notheard anyfurther complaints, although wemake itapractice tosend oneofourassistants onboard each ship which hasbeen fitted byuswith electric lights,assoonasshereturnstoharbour, toaskofthe engineers, orthose who have had charge ofthe clectric lighting apparatus, what complaints they have tomake. Andinthis way any effect onthecompasses would have been brought toournotice atonce.]—“ Wehave carefully perused “Mr. Bottomley’s letter inthe Nautical Magazine forthis vou. xvmt 40 on SECURITY AGAINST DISTURBANCE, Ere. (Mayr, MSeem ‘month, andwhilst wearenotdisposed toquestion thecoreet- «ness ofhiscalculations, wemust point outthat theexamplebe «givesisonenotlikelytooccurinpractice, forhetakesno “accountofthefactthatawirecarryingasmuchas100amperes “on board ship would belaid between decks, andthecompases “would therefore beshielded from itsinfluence bytheirondeck, “Dulkheads, and other adjacent masses ofiron, which dono “enter into Mr. Bottomley’s calculations. Moreover,thenet- “tralising effect ofother wires (branches and duplicate mains) “has been neglected byhim. “Weshallbeinterested inanyreportsyoumaybepleased “to laybefore usontheeffects onthecompass which maybe “observed during theship's first voyage, that aretraceable to “theelectric lights”—and, asIsaid, wehave had nofurther reportsaboutthisparticular shipatall.Thishasbeentheoaly casewhere aneffect onthecampass wasobserved inashipfittal byus;andifwehadknown thatthecompass wastobethere, weshould simply have putupareturn wire inthevicinity ofthe compass, s0astoneutralise theeffect ofthecurrent. ‘The second case ofacompass disturbance which wasbrought toourknowledge wasbythesame shipping company, whowished ustoinspect aship which wasfitted byanother firm, andImade 8personal report. upon itasfollows :— “Ay requested verbally, Iwent yesterday onboard your «steamship toinvestigate theeffect oftheelectric light current “on thecompuses. “1, [observed the positions ofthe compass inthesteering “house, and ofthestandard compass before theelectri¢ light was started, and took further observatious— “2, With alllights on; “3, After three lighte inthemusic saloon, immediately under «thesteering compass, hadbeen tuned out; “4, After three lights inthecompanion way lubeen turned “outs “5, After twolights inthesaloon hadbeen turned out; “6,After putting onthetwolights insaloon5 “7, After putting onthethree lights inthemusic saloon; 1889.) DISCUSSION. 875 «8,After putting onthethree lights inthecompanion way; Mi «9,After turning outalllights except those onnight circuit; «10, After turning outalllights; “11. After turning onagain theupper lights; “12, After turning alllights onagain, «The effect‘ofallthesechangesonthestandardcompasswas “hardly tobenoticed, andwascertainly lessthan halfadegree; “apretty strong wind was blowing atthe time, and this may “have had asmuch todowith thechanges asthe electric “current. “The effect ofturning onalllights wastocause adeviation “ofabout 1}°inthesteering compass, but after alllights had “been turned outagain thecompass didnotquite return tothe “firstposition.“Aftertheseobservations hadbeentaken,theship'sheadwas “swang out20°andasimilar setofobservations taken. “The effect onthe standard compass was again hardly “noticeable, and the steering compass showed not more than “half adegree variation. “The turning onand offthe lights under 4seemed tohave “some effect;butasallthevariationswereexcedinglysmall,and “not clearly traceable totheelectric light current inany par- “ticular circuit, Tamhardly inaposition tosuggest analteration “intheelectric light installation. “The captain wasgood enough toshow mesome deviation “cards obtained during therunfrom Greenock, which seemed to “indicate that the greatest, error wascaused when theship was “running onasoutherly course. “The difference inthepositions 1and 10seems toindicate “that the‘permanent’ magnetism oftheship hasnotyet “assumed itspermanent value; and, under thecircumstances, «observations during day time with thelights onand offwill be “anefficient safeguard againsterrorsofthecompass, especially “astheeffect onthestandard compass isreally insignificant.” Imay saythat aslong agoas 1879 wemade anexperiment by puttingadynamoonthedeckofourcable-ship andrunning it asamotor, andwefound that ifitwasclose tothebridge ithad a6 SECURITY AGAINST DISTURBANCE, Bre. (Mastin, Kesem, strong influenceonthecompass;thenwemoveditmoreani more away, and atadistance of50feet from thecompass¥e found there was noeffect whatever. Iwanted tocall your particular attention tothestatemest madebyCaptainCreak,thatwiththe“Polyphemus,” whichisthe only ship inthe English navy fitted onthe single-wire system, there hasbeen hardly any complaint about trouble withthe wires, whereas there have been several cases ofdisturbances— apart from compass disturbances Imean—on the other ships which arefitted onthedouble-wire system. Fete “Professor GronceFonses: Idesiretodrawattention tothe fact, that thecompasses were partially screened from thedynam bya2}inch iron deck;thatthedynamoswere,infact,actully inanironclad chamber. Onexamination ofthedynamos the polarity which caused trouble seemed tocome from imperett Joints intheyokes ofthemagnets, and Iconsider thatthe same type ofmachine could now bemade more satisfactorily,© farasdeviation ofthecompass isconcerned. fptemer Professor A.JAMIESON :Mr.President andgentlemen, shorty after theappearance ofMr.Bottomley’s letter intheNauti! Magazine, Ispoke tohim ofthecase ofthes.s.“Bombs” which wasbeing fitted with theelectric light on“the single “wire system,” under myinspection. Hecautioned meagain theprobability ofthecurrent affecting thecompasses, andkindly arranged that, when theship wasbeing ewung fortheadjustmestofcompasses, Ishouldhaveeveryopportunity ofobservingthe results produced byturning onand offthelights,asdescribed bySirWilliam Thomson inhispaper. ‘The results were tbs! thesteering compass deviated about 9°,andthestandard compsss (SirWilliam Thomson's) about 11°, intheposition most affects) bytheelectric light current. ‘The PresipeNt: The engine running allthe time?—the cironit through thefield magnets being continued allthetime? Professor A.Jamteson: Yes; and when the lights were switched off,thecompasses came back totheir natural postions Tho Prestent: Was that due tothe field ortotheelectric light currents ? 188 DIBcUSSION, on Professor A.Jamieson: Itwasdue totheelectric light frtmmor current alone. ‘The Puesipent:Thefieldmagnetshavingbeenkeptonall thetime?Professor A.Jamieson: Yes. By-the-bye, thesteering com- paswasonly about 4or5feet away from thecurrent-carrying wire, which, ifIremember rightly, hadsome 25amperes passing through itwhen allthelamps were onthat circuit, whilst the Thomson compasswasabout12or14feetawayfromit.IfI remember correctly, double wires (+and —)were run close together forthat part ofthecircuit close tothecompasses, and theevileffect wasalmost nullified, atleast itwasimperceptible onthestandardcompass,andsolittleonthesteeringonethat thecaptain said itwasnotworth minding. Ihave not met with any serious compass errors due tothe magnetic field ofdynamos alone, thefact being that thedynamos which Ihave hadtodowith were always placed lowdown inthe engine-room, and consequently inaposition farremoved from thecompasses. Mr.J.S.Rawortu: Ishouldliketocorroborate whatar.‘ Mr.Alexander Siemens said astothepractical result ofwiring ships onthesingle-wire system, having been associated with it from the commencement. Imay saythat intheearly days, when thefirst ships were fitted up,captains andowners were very much more particular than they arenow. They arebecoming quite used totheelectric light, andperhaps they arenots0much onthe look-out for defectsastheywereinthefirstelectric-lighted ships; and, except inthe one instance which Mr.Siemens has mentioned, Ihave never heard ofasingle case where anyship's compasses were affected. The onewhich hehasmentioned isone that Iwasalso connected with. Itwasavery remarkable case, andIthink Icould throw alittle more light upon that question than even Mr.Siemens did,because Ihappen toknow exactly howallthewiresinthatshipwererun.Itsohappened, which isavery common thing onboard ship, there were twoside- passages, and these, when they came totheforepart oftheship, converged intheform oftheletter V,and formed one central ore SECURITY AGAINST DISTURBANCE, Fre,(aayed, Horm, Passage, andthemain wires oneach sideoftheshipcamealo. “er these twoside passages, and then turned towards each otber, towards theapexoftheV;andIwasverymuchpuzzledforalong time tofind out what was the cause, which wire itwasthat affectedthecompass,andIfoundatlastthatonlyonewirews affecting the compass intheposition inwhich theship wasthes lying, and that turned out tobethewire which was lying vey nearly parallel with thecompass needle. The compass needle wasabout balf-way across theship, and parallel with onesideof theV:whenyouturned onthecurrent ontheonesideofthe ship the compass needle went over about 14°, and when sou turned thecurrent totheother side oftheship you scarcelygot any effect whatever. ‘Then weput some biassing magnets to throw theneedle over ontheother side, when the conditions were exactly reversed: thestarboard wire affected the needle, andthe other had noeffect whatever. Itsohappens that inthisship there was .aconcurrence ofstructural peculiarities, which is scarcely ever met with: ahuge gapwas cutout ofthetwoiron decks, and there was nointervening screen ofiron between the conductors andtheupper deck where thecompass wasplaced. It ‘wass0peculiarandsoabnormal thatoneneverexpectedtofindit, anditwasonlyafter close examination that Idiscovered thatsuch wasthecase. Thesteps Itook torectify themischief were these: Itooka3-inchwrought-iron pipeandlaiditoneachsideofthe ship along thisV-piece, andforashort distance beyond theV bothforeandaft;andtheeffect ofthese pieces ofwrought-ire tube wastosofarcure theevil that theship-owners saidthat there wasnoerror leftwhich wasworth taking anynotice what everof.Thatsimpleexpedient wassufficient, notquitetocure it,forIamquite willing toconfess thatthere wasjust trce left,—but itwassofarcured astocertainly leave only one-half s degree deflection, andtheowners said there wasnonecessity to give any further attention toit. That case and theonemet tioned byMr.Siemens are,sofarasIknow, theonly instances wehave hadinfitting uppassenger steam-ships onthesingle seTite aystem. ee MajorP.Canpew,RE,:Ishouldliketorefertothequestion 189) DISCUSSION. a9 ofusinganelectriclamptolightthebinnacle, Ihavehadex-saigrperience oftrying that, and perhaps itmight beworth just mentioning. One night Ihadputafewaccumulators onboard a little yacht: myfriends were very anxious totryanelectriclamp inthebinnacle, because inasmall yacht thebinnacle-lamps are ‘greatnuisance, andsoweriggedupalittlelamp,andIcare- fally twisted thewires together, forofcourse Ithought ofthe compass, and thelamp wasplaced just above it.Inever thought ofthefilament ofthelamp,itquiteescapedmynotice—I mean theeffect. ofthe current inthe filament ofthelamp itself, It rasmyfirst watch below. Itook them out ofDover Harbour andthen turned in;they were working away with ahead wind, andwere tacking about intheGull passage andtaking observa- tions oftheelectric lamp, more with theview oftesting how the compass worked than anything else. Iwasnotasleep, andcould hearmuch talking astowhy, when atack was made, there wasa terrible difference with thecompass. Itflashed across mymind whatitwas;Iwentondeck,turnedthelampout,andtheyhad toresort tooilagain. ThePresipext: With reference toMajor Cardew's very im-siewittam portant remark, that precaution must beused inattempting tolight compasses byanelectric lamp, Imay saythat itiseasy toshape thefilament sothat itsmagnetic moment, with the current through it,shall beinsufficient toproduce any sensible disturbance onthecompass, however thelamp isplaced out side thebowl, forconvenience oflighting; andthiswithalamp amply powerful enough tolight the compass. Idonot know what was the candle-power ofthe lamp referred tobyMajor Cardew. Major Canvew, R-E.: Itwas alow candle-power lamp, with thefilament just above thecompass, and onthecourse wewere steering thestraight portion oftheloop certainly affected the magnet. ‘The Presiwent:Wasita16ep.lamp? Major Canvew, R-E.: Obno; itwas about aS-volt 5-candle lamp.Itwasaverysmallbinnacle, andwehaditprettynear,80astogive agood light, butitdistinctly didaffect thecompass. 580 SECURITY AGAINST DISTURBANCE, Exc.(Maysi, ‘Scyuew ‘ThePuesivent: Theproper arrangement offilament for& binnacle-lampistoshapeitlike hair-pin,withitstwosides not more than halfacentimetreoracentimetreapart.Suppose, forexample, a50-volt 6-candle lamp (which isamore than amply sufficient light tosteer by), thefilament would beabout 6centi- metres long, giving area 3square centimetres, and magnetic moment, when excited by§ofanampere through it,-2¢.g.s ‘Themaximum magnetic forceofthisatadistance of10cn. (and itcould hardly beplaced nearer, even forthesmallest yacht compass!) is2x2x10-4, or1/2500, which could notdisturb the compass bymore than aboutatenthofadegreeintheselatitudes. Professor G.Forses:MightIask,SirWilliam,whetheritis quite certain that analternating current may notdemaguetise ‘8compass toapartial extent ? ‘The Presient: Itis,Ithink, quite certain that, with any practical arrangement ofthewiring, itcannot. Regarding the magnitude ofthedisturbance produced bytheone-wire systen with direct current, Imay saythat, although Mr. Bottomley illustration wasarough and ready example ofanextreme ca, you have only tovary thefigures. Take 150amperes instead of 100amperes, ortake 60amperes instead of100amperes, and take 20feet instead of30feet;varyitabout,andinstesdofa0 infinitely long wire, which isconvenient forcalculation, take any actual length ofwire concerned inany particular case, andthe ‘well-known formulas willshow youthat theeffect onthecompas ispractically very considerable. ButImustsaythattheoretical calculations ofthis kind are mere examples ofwhat at possibilities. Ifthe calculation ofsuch anexample asthi! shown inMr,Bottomley’s calculation gave only 2°or1°forthe greatest possible disturbance, then wemight restcontented tha! innopractical circumstance would itbevery serious. Allwe ‘candobytheoretical examples ofthatkindistoletusknot before wegointo iron andsteel andcompasses and ships, befor wegoout ofthelaboratory ortheworkshop—to letusknow what canbeexpected asapossible disturbance. Ifweknow thatthe greatest. possible disturbance isinsensible, wemay besatisied; but ifweknow that thedisturbance can beconsiderable, thet 199 DIScUSsiON. ss experience alonecantelluswhetherwemayneglectthethingWutan inany particular case, oragree toneglect itingeneral, ornot. . Now Imust say,asapractical matter, that inthefirst place it isbetter toavoid adisease altogether than toletadisease be produced and then tofind aremedy forit;and inthenext place Iwould say, with reference toproposed remedies, the doctor's bill forcuring the disease isliable tobemuch more expensive than adopting thearrangement inthebeginning by which thedisvase canbeprevented, and thecure isessentially imperfect atbest,afterallthatcanbedoneshortofalmost Byatroublesome andexpensive shunting ordoubling ofwires inapart oftheship, you may annul thedisturbance onone particular compass, butthen there isanother compass andanother ‘compass, allthree incessantly used inthenavigation oftheship. Itispractically notpossible toarrange themains onthe1-wire system sothat there isnosensible error onone orother ofthe compasses. Mr.Siemens referredtoanerrorof14°.Idonot Wish tocomplain toomuch of14°, but still Ithink every sailor willagree that Iamright insaying that youhad better nothave 4?error ifyou canavoid it;and any arrangement ofwiring that could produce anerror of3°ormore isnottobetolerated. Mr.ALEXANDER SIEMENS: Ithink youought totake intofull consideration that that istheonly case, outofhundreds ofships, where any compass inanyposition oftheship wasaffected. Captain Creak: May IaskMr, Siemens ifheremembers the P.and 0.ship “Oceana”? Her steering compass wasaffected totheextent ofsomething like8°, Mr. ALexanper Stemexs: You can easily arrange the wires badly, ofcourse. ‘The Paestpent: Iamperfectly aware that there aremany cases inships atpresent atsea, such asthe last referred toby Captain Creak. Ididnotcaretomention thenames ofships or Companies, butIknow many cases inwhich there areerrors of 3°,4°,and5°,undoubtedly duetotheelectric lighting. Now I would remark that itisnotatallsatisfactory tohave achanging error inthesteering compass, although thestandard compass se SECURITY AGAINST DISTURBANCE, Ere.[May280, cwuuin maybeunalfected. Theshipissteeredbythesteeringcompas. Anofficer inanother part ofthe ship looks frequently atthe standard compass, andifhefinds thecourse oftheshipiswrong hhepasses anorder oracaution tothesteersman; butitis exceedingly inconvenient iftheofficer incommand, having kno that hissteering compass waaallrightatacertaintime,shouldat some uncertain time—when thesaloon islighted upforNo.2 passengers’ dinner, forexample!—find theship offhercourse 2°. Hedoes notknow whether itiscareless steering oranerrr in thecompass, and itmay take tenminutes tofind outwhich itis that hascaused theship togooff2°, That isanintolerable state ofthings. Anything that introduces errors atalladdsto thecomplication, great enough and perplexing enough asitis that already exists, whether with theofficers, thewatch,orthe men steering, and should ifpossible beavoided;andifthe electric lighting oftheship could notbedone otherwise thanby methods which introduce errors offrom 2°to 5°at uncertait times indifferent parts oftheship, itwould beaserious question whether theelectric lighting should notbegiven upaltogetbe, ‘orthecaptain andofficers oftheship should make uptheir minis topaycareful attention toitandlool: outforthechanges. To depend upon thesteward sending amessage that heisgoing to light upacabin orpart oftheship would beavery inconveniest state ofthings. Itseems tomethat iftheone-wire system istobeused all,itought tobeobligatory touseonly alternate current wth it.Ifdirect currents areused, thetwo-wire system alone ought tobeadmitted onboard ship. Electricians must notsuppose thatifsailors donotcomplain there isnothing tocomplain o. Inthe first place, sailors donotalways know that they have suffered from theerror. Many aman hasbeen steering ft several hours, andhasnever imagined that hiscompass hadbees disturbed owing tothelighting oftheship. Even athoroughly careful man may nothave discovered that itwas thelighting ‘that hadcaused some disturbance which hemay have notice it hisreckoning. Ihave occupied your time toolong, but Iwould just,is 18893 ALTBRNATE CURRENT WORKING. sea conclusion, begtheInstitution toconsider, asfarastheinfluence Witan ofitsmembers isconcerned (and Ihope myfriends willforgive imeforbeing sourgent), whether itwould notbebetter toadopt thetwo-wire system universally inship lighting. Onthemotion ofMr. SPaGNoLETT!, seconded byMr.PREECE, ahearty vote ofthanks was unanimously accorded tothe President for his communication, ‘Mr.W.M.Monpey: AsIam afraid, SirWilliam, that Imay notgetthroughthewholeofmypaper,Ishouldlike,inthefirst, place, toexpress mythanks totheBrush Corporation forvery kindly sending over theapparatus andthediagrams that are before you. Ishould alsolike tosayhow much obliged Iamto nnybrother officials attheBrush Corporation foralltheassistance theyhave given me, Ipropose, with your permission, toread the first. portion of thepaper, and then tocondense the remainder, and miss several sections outaltogether. Iwould therefore ask those gentlemen ‘hodomethehonour ofdiscussing thepaper, that they should read the sections that are jissed out, before making their remarks. ‘Thefollowing paperwasthenread:— ALTERNATE CURRENT WORKING. By W. M. Monpey. Iwish tostate atthecommencement that this paperisnot intendedasacontribution tothecomparison oftherelativemerits ofalternate-current working andofany system ofdirect-current supply. ‘The discussion ofthis question hasalready taken place, and those who took the A.C. side have noreason tobe dis- satisfied with the result. Itmay, however, bepointed out that exclusive advocacy ofany one particular method ofworking, as being thebest forallpurposes,isnotapositionthatislikelytobe taken byanyelectrical engineer whohasanyextensive acquaint- ‘ance with thevarious requirements and conditions that aremet withinpractice. 584 ALTERNATE CURRENT WORKING. laytia, Wonkinc ALTERNATORS PARALLEL, aND Best PRINCIPLES oF CONSTRUCTION FOR ALTERNATORS. Thisisamatterofverygreatscientific interest,andofallgreaterpractical importance. Itisnottoomuchtosaythatthe complete success ofthetransformer system ofsupply depeois tongreat extent upon whether alternators can bequite easly andsuccessfully worked parallel, ‘This hasbeen strongly insisted ‘upon, especially bytheopponents ofthat system, ‘Theimportance attached toitarises partly from the fact that the most eo mical method ofsupply isthat ofusing always thesmallest plat that will dothework, andincreasing ordecreasing thenumber ofgenerating units inoperation, according tothefluctustows ofthe demand. Todothis most conveniently, the alternator: shouldbeworkedparallel,andoughttobecapableofbeingpot inandoutofcircuit easily, andwithout causing even amomeatary flicker orinterruption ofthelight. ‘The useofalarge mache andengineforthesmalldaysupplyisespecially tobeavoided. Inalmost allcases itisdesirable tohave &comparatively«mall setofplantforthiswork,evenwherelargemachinesareused fortheheavy evening work. Another argunent infavour ofparallel working,asagaiss theuseofvery large machines, isthat itreduces thecostofthespare plant. ‘Thus, ifastation isprovided with oneortwo machines andengines forthefullload, thespare plantisequsl incosttotheworking plant, orperhaps toonehalfasmuch. If,however, smaller generating units areemployed, onesetit four may beconsidered asafe allowance. ‘Thehistory ofparallel working ofalternators maybebriefly sketched. In1868 Wilde described parallel working and synchronoas action ofgenerators, and sonearly obtained synchronous motat ‘action that itisextraordinary heshould have missed it.* In1882-3 Dr.Hopkinson,t notknowing ofWilde's work +«OnaProperty oftheMagneto-electric Current tocontrol andrade‘synchronous theRotationsoftheArmatures ofanumberofBlectro-maguticInduction Machines:"—H, Wilde, Phit Meg, Jen, 1969, pp.64.62.+fProt,Ins,CB,1882-9, 1689] ALTERNATE CURRENT WORKING, 385 arrived theoretically atthe conclusion that itwas possible to work alternators parallel, butnotinseries, and subsequently demonstrated thefactbytrials made, inconjunction with Professor Adams, onthe DeMeritens machines atthe South Foreland Lighthouse, which were runparallel and asmotors. This was alllaidbefore this Society atthe time, and was very fully Aiseussed.* —-Inthe course ofthis discussion in1884, Mr. Alexander Siemens described some experiments showing that theSiemens altemate-current machine ranasamotor.t Thisis interesting as being, sofarasIknow, theonly recorded instance ofanalternator without iron inthearmature being runasamotor. Itwill be remembered that the Wilde and De Meritens machines bad iron cores. These experiments ofMr.Siemens were notvery euccess- fal,themotor frequently stopping suddenly, even when doing verylittle work;buttheywereofvalueforcomparativepurposes. Passing over thenext fewyears, wefind that theuseofalter- ators had become ofgreat importance onaccount ofthegrowth anddevelopment ofthetransformer method ofdistribution, and thattherewasandisacertainamountofdoubtandhesitation about working parallel. Itisrecognised that ithasbeen and anbeaccomplished, butthat thearrangement isnotonetobe thoroughly and completely depended upon inevery case. And there issufficient justification forthis doubtfulness. ‘The present opinionmaybefairlystatedasfollows:—Alternators maybesuccessfully run“inparalleliftheyhaveagooddealofself- induction, and tosecure this itisbetter that they should have iton-cored armatures. Inhisrecent paper on“Alternate Current Machinery,” before theInstitute ofCivil Engineers,t Mr.Kapp dealt at.considerable length with this part ofthesubject, and hisviews,Ineednot say,may beaccepted asquite correctly representing thecurrent state ofknowledge and opinion. Referring toalternators that haveself-induction negligible, Mr.Kapp says:—* Machines of *Journal, xii. (1884), pp.496.589, {Yournal, xii, p.828.4Prog,Int.C..,Ped.1°69 586 ALTERNATE CURRENT WORKING. [Mayr «this type camonly beruninparallel ifthestrength oftheir field isadjusted with almost mathematical precision,andasthis «would require more skill andattention than isavailable withthe “ordinary staffofacentralstation,suchmachinesarepractically “unfit forparallel working. ‘Tomake them fitforthismethodof “working, either thearmature resistance ortheself-induction “must beincreased. An increase ofresistance inorder tobe “effective, would have tobesoconsiderable astoseriouly «prejudice theelectrical efficiency ofthemachine, andthis “expedient may thereforebedismissedasimpracticable. The “otherplanofincreasing theself-induction isnotopentothe “same objection. Ithas the effect oflowering the plant ef “ciency, but itsinfluence upon the electrical efficiency isonly «indirect, and sosmall that itmaybe neglected. From thefore «going, itwillbereadily seen that theonly andsufficient coodi- «tion forsuccessful parallel working isasensible amount ofsl “induction inthearmature cireuit. Ifthearmature itselfdoes“notpossessthequalityinasufficient degree,achokingcoilof «suitable self-induetion must. beinserted into the cironit ofeach “machine. ‘The results here arrived at,byamere theoretial “investigation, areentirely borne out inpractice. Itiswell “known that alternators having noiron intheir armatures cannot “be run inparallel, except bythe adoption ofsome such «expedient aschoking coils; alsothat parallel running isfeasible «with those alternators which have iron-cored armatures, ani «then with different degrees ofsecurity.” Now thisisavery serious state ofaffairs, foritreally means, iftrue, that theonly machines suitable forcentral station supply urethose that, onaccount oftheir high self-induction and resistance, arebadregulators that will not bear any considerable change ofload without varying agood deal inE.M.F.,andthat farewasteful onopen cireuit. Formypart, Ihave seen enough oftheuseofiron inarms tures towish todowithout itifIcan, and Ihave repeatedly pointed out that, ifiron isreally necessary, the armature isthe very worst place toputit.For thepurpose ofincreasing the self-induetion itcan bemuch better used outside ofthe machine, 1188) ALTERNATE CURRENT WORKING, 7 andpatinsome place where itisnotsubject toallthelosses, restrietions, and disadvantages that necessarily accompany its employment inthe core ofanarmature. Ifitisnot inthe armature itcanbeentirely removed when notrequired forthe purpose forwhich itisthought tobenecessary—that is,when nlyonemachine isinuse. Toputiron inanarmature merely toincrease self-induction appears toshow awant ofcommon ingenuity andanabsence ofthesense ofproportion. Bat letusexamine the evidence onthis matter, inorder to seewhatbasisoffactthereisforthepresentviews. Inthe first place, itwill befound that the useofiron amatures isnotinalleases toberelied ontogive successful parallel working. Inthe discussion lastyear, Mr. Gordon, who tashadvery considerable experience, andwho used iron largely inhismachine, wamed usinstrong terms against working in fanllel. He said:* “We know that experiments have been “made bycoupling anumber ofsmall alternate-current machines “together, and atthe South Foreland they were successful, “but that was because they were working onarclamps.” A little elucidation, perhaps, isnecessary here—“ Many ofushave “tried them, and they will, ontrial, work together, nodoubt, “but they donotwork together tillthey have run forthree “orfour minutes; they will inthat time jump, and thatjump- “ing will take months oflifeout ofthe40,000 lamps. That “alone israther aserious difficulty incoupling machines “together, and Ithink wemay take itinpractice—I am not “speaking about the laboratory orexperiments—we donot “acouple machines.” The Zipernowski alternator will, however, work parallel, but apparently notvery well; togetittodosotheperiodicity has hadtobereduced to42,aswenow know.t Farther, Ihave recently seen acommunication from the highly experienced makers ofawell-known alternator inwhich ironisused, stating that, ifitisdesired torun parallel, the machines, engines, and arrangements must beinevery way *Journal,xvii,pp.19545,Feb.1886, +Seeunder Periodicity,” p.699, . 288 ALTERNATE CURRENT WORKING. [MaytH, identical: theimpression conveyed beingthatwiththismache thegreatest care isnecessary toensure asuccessful result. Then, again, the large experience ofthe Westinghoue Company intheUnited States hasshown that their iron-od alternators will work inparallel under some conditions, Wear toldthat,withthesemachines, “parallelcoupling alwayssucces“whentheyareloadedtoabouthalftheirmaximum outpat« “over, butthat machines working underalessloadeanzotwith “certainty besocoupled.” But the other day Professor Forbes described* succes parallel working with the iron-cored Lowrie-Parker alterslas atWest Brompton, even with very light loads. ‘Thus itwill beseen that wehave evidence ranging fon complete failure tocomplete suecess with iron-cored machines. Itisproved, then, that thepresence ofiron isnotalot sufficient toensure success, forthe use ofiron isassxiste cqually with failure and with success. That thesuccess attained iscomplete ouly asregands theft ofsynchronism, isshown bythefollowing extract from Profeset Forbes' remarks inthediscussion onMr.Kapp's paper, which!quoteatsomelength,becauseProfessor Forbeshastakenerey opportunity ofascertaining, bypersonal observation, what isbeing done, andbecause hegives uswhat isalone ofmuch service# thepresentstage,viz.,theindependent recordofactualfacts,a0! (olls usclearly what thereal difficulties are. ‘Thepassage ist:— “Phequestionofpurallelworhingwaeovewhichhabeenverymachde“-easinnetowhetheritwasdesirableoFpomsble, Tewasalwapeposite.2!‘+Mr.Kapp avi, with machines which had very high ael-indwetion—machie >‘hichthearmiatarehadagreatmasofruninit.tthesameime,theitradeci=‘of that self-induction into amachine reduced what Mr.Kapp called theps“-oficency ofthedynamo-machine; andifthey«oulworkoat,asebei!“engincers would!workout.somitermeansofmakingmachinesworkinpr“itwould bevery desimble. Another reason why theheavy sIfindaetion ea= “-jarious ete that itrequitrd such continued attention onthepartofthwinte “stationtosethatthepressurewarmaintained constant,‘Theelectricalpress Bound ofTrade Enguiry. Apri, 1x8 +t Pron Inst C8, eb 19,188 1889 ALTERNATE OUBRENT WORKING. 889 “variedsomuch,thatthequantityofcurrentbeingdeveloped,asshownbythetable ‘ofcurves,fig.11,wasseriouslyaffected,andconstantattentionwasrequiredtokeep‘suchamachineregulatedtotherightpressure;whereasamachinewhichhadno“telindaction. orverylitle,gaveoffthemmeelectricpressareornearly9.what- ‘everthecurrentmightbe,0longasthespeedwasmaintainedcoosaut.Astothe “ponibility ofworking conveniently iaparallel with those machines, hemight say“thattheexperience ofAmericahadbeencompletelyagainstit,Itwastherefound ‘thatitwasposibletoworkinparallel,butthatitenormouslyincreasedtheamount ofskilled attention required iuacentral station.” This isavery strong condemnation ofthepresent system of obtaining synchronism, andattheendofhis“Central Stations” paper* Professor Forbes returns tothesubject inthese words :— “Tventure tothink, however, that theplan now universally propowd for “making machives work inparallel will notlong betolerated. ‘This isto introduce ‘into themachine alarge amount ofinjurious self-induction, thus diminishing the“planteicleney.andrenderingtheequalisingofpressarewithxariouslotsvery*difieat.” Turning nowfrom alternators with,tothosewithoutiron-cored armatures, wearebrought face toface with the serious fact that, inspite ofthe many inconveniences connected with theuse of verylarge machines, Mr.Ferranti islaying down theDeptford station towork with engines and alternators, each ofseveral thousands ofhorse-power, andisdepending upon theemployment ofspare engines andmachines, ofcorresponding size, toprevent or tominimise theriskofwholesale extinction ofthelights. Itis understood that oneoftheprincipal reasons fordoing this isthat parallel working isnottoberelied on. The next question is,Isitclear that, inorder tobeable to runthem parallel, orassynchronising motors (for Ineed notsay thatthetwoqualities areinseparable and areinfactidentical), alternators should bebad regulators, should have large self- induction, orhigh resistance, orboth? Ithink not. Ihave mentioned Mr. Gordon's machines, which certainly hadthefirst ofthose qualities, but they would not run parallel. Other machines that Ihave referred to,and whose parallel working is wottobedepended upon, areatany rate not remarkable for absence ofself-induction. *Jounal xvi, p.195. You, xvi. 4 00 ALTERNATE CURRENT WORKING. (Alay tn, Again, istheabsence ofiron from the armature cores alone sufficient todeprive amachine ofself-induction, andtogiveita straight characteristic? Ithink not; forIhave tested machines that hadnoiron, butthat had agood deal ofresistance,very considerable self-induction, that hadvery crooked characteristic, andyetthat would notrun parallel under anycircumstances whatever. Soweseethat neithera bent characteristic, norself-inductin, nor resistance, northe use ofiron cores, nor even thesimal- taneous possession ofthe whole ofthese admirable featuresni qualities—that none ofthese things isthesecret ofsuocestl parallel working. Perhaps—and this isadreadful reflection—it istheext opposite ofallthese! Now Iamintheunenviable position ofbeing outofaccor with thetheories, thepractice, theprinciples, and theexplant tions ofthevery able men who have lately written, spoken,nd worked onthis subject. Ihave nodoubt that allthey bare said isperfectly correct, sofarastheir difficulties areconcerse, —that they have experienced considerable trouble inworking parallel, and have even insome cases met with actual failar. But this waspartly because they have been unfortunate orunwise intheapparatus they have used, and partly because theprim ciples that have been relied ontoensure success have notbeenia allrespects suited tothecase. Although theconditions ofthetwo problems arenotioall points similar, Ivery respectfully submit that weareindanger ofrepeating theoldmistake that was made regarding direet- current motors. We were taught that self-induction inthe armatures wasgood, and that aspecial form and special propor tions should begiven tosuch motors;andIwasveryunorthodor when, in1886,* Iasserted that self-induetion wasnotavirtue to becultivated inmotors anymore than indynamos, and,generally, that agood motor wasagood dynamo, andvicevered. Howeves, itissome satisfaction toknow that theviews Ithen expressed *Pht,Mog,Jen,1886,“TheDynamoas&Generatorandas8Motor.” ‘980,) ALTERNATE CURRENT WORKING. 501 have received the sanction ofgeneral practice, although Iam afraid that: intheeyes ofProfessors Ayrton andPerry those views remain asunorthodox asever. ‘Now, Iamprepared totake aprecisely similar stand with regard toalternators, and tosubmit foryour criticism theview thataperfectalternator foranyandeverypurposeshouldhave noresistance and noself-induction. Iprefertoregardthequestion ofalternate-current. parallel working asaquestion very largely depending upon what arethe bestprinciples ofconstruction forsynchronising alternate-current motors. Itisvery much better and simpler tolook atitfrom thispointofviewthanfromanyother,andIventuretoassert, {nspiteofallthathasbeensaidtothecontrary, thatifalternators aremade, amongst other things, with the least possible self- induction andthelowest practicable resistance, they willnotonly tethe best alternators, but they will best run parallel, and will ‘dosobecause they willthen bethebest. synchronising motors. Ithink that even theadvocates oflarge self-induction will admit that that quality does notimprove analternator perse, they only introduce and tolerate itonaccount ofitssupposed bearing onparallel working (Icould quote anumber ofsuch admissions); butitappears tomethat, properly regarded, self- induction isnoteven anecessary evil, itisanunmitigated evil. Itisanevil even forthis narrow andrestricted purpose. Whatdoesitdo?Whenmachinesaregettingoutofphase, itsteps inand prevents that ready transfer ofcurrent which ia required inorder tocheck theleading machine, andtoaccelerate thelagging machine. Itprevents thelatter from immediately ndunbesitatingly developing those motor properties which lie attherootofsuccessful synchronism. ‘Thefoundation ofparallel working should bethat theprime motors areunder thecontrol of the generators. The question turns, therefore, partly onthe ‘qualities oftheprime motor, butmuch more largely onthemotor ‘qualities ofthealternators. Tam prepared toadmit that ifthetendency oftheprime ‘motortoleadortolagissogreatthatundernocircumstances an itbecontrolled,—and this may arise either from excessive 102 ALTERNATE CURRENT WORKING, [Massh, power onthepart ofaprime motor,orfromdefect.ofmotor power inanalternator,—then large self-induction orresistance maybeofuseinordertopreventoneormoreofthemachines being burnt up,not, however,asusefulandnotaseffectivea1s safety-fuse;butself-induction undersuchcircumstances is notanassistance toparallel working—quite the contrary. Ifit were absent, probably themachines would runparallel allright. ‘They would bemuch less likely tobeburnt upwithout self induction than with it. Self-induction isuseful becauseit prevents themachine which ought nottohave itfrom being burnt upbecause ithasit. Now Iwill briefly describe some experiments with twoof myalternators, each made for2,000 volts and each capableof working continuously at35,000 to40,000 watts, or,say, 50EHP, output. Arvangement.—Each machine was driven bya75LHP. Fowler engine. ‘These engines were similar, their normal speed ofworking being 120 revolutions per minute. Each ofthe ‘engines, which were notcoupled orconnected inany way, ms provided with aheavy fly-wheel, and drove, besides analte- nator, aheavy and wasteful double (and inone case trebl) set.ofcountershafts provided with alarge number ofbelts, fitted toanarrangement offast-and-loose pulleys forcom venience oftesting allsorts and conditions ofdynamos. 1 mentionthisasshowingthatthemomentum ineachcasewis very considerable. Inorder tomake thetest asonerous aspossible, thepulleys ‘used were such that one engine had torun at130 revolutions, while the other ran at90revolutions, when the alterustor were attheir normal speed of650 revolutions per minute. (J.)The alternators were run uptofullspeed, andeach excited togive2,000 volts. When inphase they were switebed parallel without anyexternal load, andwithout anyimpedance coils orresistance between them. ‘They ranparallel perfectly. (2)Aconsiderable inductionless load was then put0% varied, and taken off. They ran equally well under all circumstances. 1868 ALTERNATE CURRENT WORKING. 593 (3.)They were uncoupled, and then, theload being con- nected tothe mains, they were suddenly and simultaneously switched parallel and ontothemains with perfect. success. (4.)One alternator was excited togive 1,000 volts, the other giving 2,000 volts. They were then switched parallel, andwent into step perfectly, giving aterminal P.D. ofabout 1,500 volts. Noimpedance orresistance was used inthis orin anyother case, Aload was then putonwithout affecting their behaviour. (5.) With one machine at1,000 volts and theother at2,000 voltstheywereswitched parallelwhenoutofphase,andinstantly ‘went into step. Alarge current appeared topass between them forafractionofasecond,butnotnearlylongenoughtoenableit tobe measured, ortodoanyharm. (6.) They were then leftrunning parallel while onewasdis- connected from theengine, byitsbelt being shifted from the fasttotheloose pulley. Itcontinued torunasamotor syn- chronously. Aload oflamps wasatthesame time onthecircuit (7.) The twomachines were then uncoupled and excited up to2,000 volts. They were then switched parallel when outof phase andwithout anyexternal load, andwent into step instantly. (8.) Whilst running asin(7),steam wassuddenly andentirelyshutoffoneengine, Thealternators keptinstepperfectly, one ‘acting asamotor anddriving thelarge engine andalltheheavy counter-shafting and belts. Itwasimpossible totell,except by thetopofthebelt becoming tight instead ofthebottom, which machine was the motor. Tofind thepower exerted bythealternator acting asamotor (im8),adirect current motor wasputinitsplace, and thepower required todrive theengine andshafting wasfound tobe20H.P. Itmay bepointed out that these tests were made under themost exacting and onerous conditions that could possibly be imposed, and particularly Iwould point out that onaccount ofthevery great momentum oftherevolving masses, nothing butthestrongest andmost instantaneous motor action could have kept themachines inphase. There never wasasingle case when they gotout ofstep, even momentarily, orwhen subjected aot ALTERNATE CURRENT WORKING. [laytn, tosudden and violent variations ofload. When itisconsidered that, inorder toseoure this result, itwasimperative thatthe controlofallthatmassshouldbeexertedinafractionofshyofasecond (the periodicity being 100), itwill berecognised that there wasnotime tobelost, and that the use ofanyselfinduc- tion orresistance, orofanything else that could inanyway choke, retard, check, orinterfere with the strength andis- stantaneity oftheaction wasabove allthings tobeavoided. Ishould mention that themachines apparently synchronised equally well atspeeds varying very considerably. Astothe self-induction ofthe machine itself, that isquite negligible. Itscharneteristic (fig. 1)isnearly straight, aboot Ti}ttt tttSRR CREEL TTTttt rrPtytttyPerro ig. 1Charactersti. half thedrop inthecurve being due toresistance andhalfto self-induction. Here wasamachine generally allowed tobegood when work- ingsingly, butpossessing allthequalities that have been stated asunfitting itforparallel work, tested under alltheconditiats that arerecognised asmost trying, even forthose types ofalter- nators which areuniversally, andIbelieve erroneously, regarded aasbest suited forparallel work, andbebaving throughout ins manner that simply left nothing tobedesired, and carrying, 1199) ALTERNATE CURRENT WORKING. 35 unimpaired, into parallel work those features which itpossesses when runsingly. Itrust that Ihave now said enough tojustify theviews I have expressed. Afewparticulars ofthismachine may beofinterest. Itnow takes 500watts toexcite itatfullload, and rather less than 400 watts onopen circuit with fullE.M.F. Theefficiency when working withalight load isvery high, The power required torunatfull speed, with full E.M.F, but noexternal load, is3H.P., ofwhich 1'5H.P. isordinary mechanical lossinthebearings, etc., and 1°5 HP. iselectrical waste inthearmature conductors and supports. Thisisfrom amost careful test, and shows that, sofaraseconomy isconcerned, with light loads themachine does notleave much to bedesired. This isreally ofvery great importance, forthe expense connected with central station working during thelong hoursofdaylight maybeaseriousitem. AtrerNate Current Motors. Thebearingoftheforegoingonthegreatandvitalquestion oftheconstruction ofalternate-current motors isobvious. Much ‘workhasbeendoneoflateonthissubject,andnodoubtcon- siderable success hasbeen met with, buttheuseofalternators as simple synchronising motors hasnot appeared toattract much attention, probably because itisnotknown what excellent results may beobtained inthis direction, and partly because oftheir inability tostart from rest. ‘The experiments just described, and others thatIhavecarried out,showed what perfect self-governors such motors are. Not only dothey maintain synchronism, butthey possess aninherent economy which ismost valuable, Just enough current passes through them tokeep them instep, andtodothework imposed onthem, They become generators, anddowork onthecircuit, if from any cause there isatendency forthem torunfaster than thegenerator. Thave devised avery simple means ofstarting such motors, which atasmall expense removes theonly drawback totheir employment. Theexciter, which isgeared directly insome way 398 ALTERNATE CURRENT WORKING, (ay, with thealternator, isused, inconjunction with asmall accuma- | lator,asadirect-current motortostartthealternator. Whea| synchronism isattained thelatter issimply switched into circuit, Ttwillbeobvious thatasmallbattery capable ofaheavydix| charge foraminute ortwo isallthat isrequired, Thisis recharged bytheexciter. Ineed notgointo thedetails ofthe arrangements. This isperfectly practical. Ishould bare2» hesitation inrunning these alternators inregular work asmotos, and should expect tofind their “commercial efficiency” very nearly orquite 90percent.—about thesame efficiency, whatever itis,thattheypossessasgenerators. Forinstance, themachines alluded towillwork asmotors atauseful output of50H.P.with this efficiency. ‘And here isanother advantage ingetting alternates economical when onopen circuit. Bydoing soitbecomes possible tousethevery small exciter asadirect-current motit toget them uptothe synchronising speed. Inmany situations asynchronising A.C.motor willfulflall requirements,andwilldosowithfewerdrawbacksthananyother kind ofelectric motor whatever. Itwillbeabetter regulator, more economical, less troublesome tolook after, much saferto handle,and(whatisofverygreatimportance inconnection with thetransmission oflarge powers toadistance) thedifficulty of dealing with high pressures, which issoserious with dired- ‘current motors, isvery easily overcome, fornotonly willalters- tors stand ahigher tension inthemselves, but, ifnecesars, transformers may beused toreduce the pressure between mains ‘and motor. Peniovicrry. Some confusion iscaused bythevarious ways inwhich me ofalternation areexpressed, itbeing often quite uncertain whether periodsorhalf-periods arereferredtowhen“alterot“tions” or“reversals” arespoken of. This confusion should be avoided byusing theterm “period,” which has,Ithink,alway: been taken asexpressing the complete cycle, orthechanges undergone byasimple rectangle orelementary armature rotatedfromzerothroughacomplete revolution inasimplemagneticfel. 189 ALTERNATE CURRENT WORKING. sor Tosaveconfusion Ialwayswriteitthus,~_/;andIwouldsuggest thatthissign beused, unless abetter one beforthcoming, and thatweagree tospeak ofperiods, instead ofalternations or reversals, and ofperiodicity, instead offrequency orrates of alternation orofreversal.* Professor Silvanus Thompson informs methatformerly asomewhat similar difficulty arose inregard to sound, and that confusion was avoided inthescience ofacoustics bygiving the word “vibration” thedefinite meaning ofthe double motion which itnow possesses, theFrench tuning-forks being stamped with theletters “V.D.” (vibration double) after the number. The subject ofperiodicity isbecoming avery important onc, forboth commercial andscientific reasons. Iturgently demands settlement intheinterests ofthewhole industry. The very ‘wide differences found inscientific opinion onthis matter, and in thepractice ofthe various manufacturers, must lead, and are indeed now leading, tomuch confusion and inconvenience. As thingsstandatpresent, greatuncertainty existsnotonlyasto what rate isthebest, but whether transformers, arclamps, or other apparatus made byonemaker, foragiven E.M.F. or porpose,canbeusedonacircuitfedfromanalternatorcon- structed byanother maker. Itmight besuggested that some Joint action should betaken bymanufacturers, and others con- cemed, inthedirection ofsecuring uniformity; butthere are arguments against such acourse, the principal onebeing that probably each maker believes hehas good reasons forhisown practice. Iventure tothink there isnoright orwrong periodicity. What may bebest foronesetofconditions, orwith onetype ofapparatus, maynotbebestforanother. Orperhaps itwould bemore correct tosaythat there isaright periodicity, and that there areright and wrong apparatus ;butthis israther adelicate subject. ‘The relative advantages ofhigh orlow periodicity must be ‘This sign may bewritten something like, butrather larger than theAlgebraicsign(~)for“difference.” 78 ALTERNATE CURRENT WORKING. (lay200 considered asbearing onthefivemain divisionsofapparatusused. ‘Taking these intheir natural order, they are— 1.Prime Motors. 2,Alternators. 3.Conductors. 4.Transformers. 5.Lamps, &e. Each of these divisions contains matter in relation to periodicity which inthe near future will probably form the subject ofmany papers,asthevarioussubjectsbecomebetter understood. Iam not able todomore than touch briefly 1 each, inthehope that myremarks may lead totheexpression of theviews ofothers, and tosome approach touniformity. 1.Prime Motore.—At present questions affecting prime motors donotusually bear directly onthesubject ofperiodicity. The only exception istheParsons steam turbine, with which tbere seems tobenogreat difficulty inusing asimple two-pole machine, thus making thenumber ofrevolutions andofperiods the same, There may befound some want offlexibility s regards changes ofsizeoroutput with such turbines, as,fromthe nature ofthe case, speed can only bevaried byvery large amounts. ‘Thus,forinstance, if100~_/arerequired, thereis choice except between atwo-pole machine at6,000 revolution: perminute andafour-pole machine at3,000 revolutions. Thisis, ofcourse, ontheassumption that thealternator isdriven dire. Ifindirect driving isresorted to,thedifficulty ismuch reduced. Itisunlikely, however, that thebearings ofasteam turbise would work satisfactorily except with direct driving. 2.Alternators.—It isnot allowable todoubt that the diver gencies inperiodicity found inpractice aretoagreat extent the sult ofdevelopment and survival ofwhat hasbeen foundtogive thebest results ineach case; but from anexamination ofthe | machines inuseitisvery difficult totrace any common line02 which action maybesupposedtohavetakenplace.Itistruethat| thegeneral tendency hasbeen towards reduction oftheperiod city, butthecauses which areresponsible forthat tendencyarveryobscure.Ihaveseenlargeandsmallmachines,madeatthe| 188903 ALTERNATE CURRENT WORKING. wy same time byonemaker, inwhich thenumber ofcoils was the same, although one had torun atabout 2,000 revolutions a minute, and theother atless than half that rate ofspeed. At least oneofthese was wrong. Forthepurpose forwhich these machines were made the very high periodicity was almost un- objectionable inthesmall machine, butinthelarge ones itledto- disaster. Itisacommonmistaketosupposethathighspeedofdriving, orhigh peripheral velocity, arenecessarily connected with high periodicity andwithlargeoutput. Ofcourseanygivenmachine, ifrunatahigher rateofspeed than itsnormal one, must have- 4proportionately higher periodicity, but will not necessarily be capable ofagreater output. Itisquite possible todesign a high-speed machine togive alow periodicity, andvice vered. Perhaps theimpression hasarisen because with most alternators itisnotpossible ordestrable toplace thesuccessive poles very close together onaccount ofthemagnetic leakage. Formachines with iron inthearmatures itmight beexpected thatthemagnetic andelectric losses, aswell asthevery consider~ able self-induction and consequent variations inthe effective jotential difference with changes ofload, would lead tothe reduction oftheperiodicity;butexperiencedoesnotseemto have always ledtothis result. ‘The lowest periodicity (42with theZipernowsky machine) andthehighest (132 with theWest- inghouse machine) are both found in.alternators having iron cores. What isthereason ofthisgreat diversity?.Themachines. lave many points ofresemblance, and yet one isrun atmore than three times theperiodicity oftheother. Not having seen theAmerican machine, Icanoffer noopinion based onpersonal observation; butofthe Austrian alternator Icansaythat, in nyopinion, itwould notbeadvisable toincrease therate. Inacommunication from Messrs. Ganz &Co.,published afew days ago,* Mr.Zipernowsky says that thelowperiodicity used vaschosen onprinciple toenable them tocouple their dynamos varaliel, Hevery cogently adds, injustification ofthis: “At *The Electrician, May 10th, 1849, p.16, 60 ALTERNATE CURRENT WORKING. (May38, «present nopractical electrician willhesitate toacknowledgethe “coupling ofdynamos inparallel circuit. tobeavitalquestion of “any parallel system ofcentral-station distribution.” ‘This bears out, very completely mycontention thatthe periodicity isgoverned inmost cases bysome special featureof the type ofapparatus. Zipernowsky hashad togodown to 427\ togethisiron-cored alternators tosynchronise, andfr noother reason;forhisstatement isthatsynchronous actionwis “notasecondary result,butjusttheendwewereaimingat,”and heacknowledges that there aredisadvantages connected vith this lowrate inother parts ofthesystem. ButWestinghouse with 133~L/ and Lowrie-Parker with80 —bothhavingironcores—can runparallel,theformerwithutlessthan half-load, thelatter, aswenow know, with anyload,and IthinkIcouldranparallelwithoutironandatanylosdsi periodicity, sothat Mr.Zipernowsky’s only reason fails tojustly hisaction, except asregards hisown particular apparatus. Professor Forbes recently stated that theexhaustive tests made inAmerica with theWestinghouse apparatus hadshows that there was noreason forreducing therate ofalternation, which isthehighest nowinuse. This isveryimportant,andit istobehoped that some account willbegiven ofthenature of thetrials and oftheresults obtained. Mr, Ferranti, who aton timeusedveryhighrates,hasforsometimebeenusing80-VattheGrosvenor Gallery, andhasfixed 68~~ fortheDeptfor! machines. Ishall notbesurprised toseehim increase thi before long. Inmyownmachinetheperiodicityis100,which,Ibelieve, isonthewholethebestrate;butifitberequired todosofor anyspecial purpose, there would notbemuch difficulty in changing it.Thepolefaces canbealtered innumber andinsi bymerely altering thepattern; thefield winding, being io» single coil, would beequally simple and equally efficient forany periodicity. Even with alarge number ofpoles placed close together, aswould berequired forahigh rate, magnetic leakage would not increase. The size ofthe machine would notb sensiblyaffected. ‘Thearmature wouldcertainly berathermt 1999) ALTERNATE CURRENT WORKING. cor costly forahigh rate,asagreaternumberofcoilswouldbe required; but theefficiency, output, and regulating qualities would remain practically thesame, Regarding this machine, Ithink Imay saythat periodicity isnotaquestion ofthefirst importance, AsfarasIamconcerned itisother divisions ofthe distributing apparatus that determine the rate atwhich itis desirable towork. ‘Thebearing oftheform andarrangement ofanyalternator on thebestperiodicity forthat alternator isavery close one. The question cannot besettled ongeneral principles. Thave referred elsewhere more fully totheconditions asthey affect regulation under varying loads, parallel working, andother points inconnection with alternators; and, asanappendix (see P.20), Ihave quoted some observations made revently atthe Institution ofCivil Engineers during thediscussion onMr.Kapp's Paper on“Alternate-current Machinery,” andnotyetpublished. Tnow pass ontothenext division ofthesubject. 3.Conductors.—The choiceofasuitableperiodicitydepends ‘lso tosome extent upon the conditions asthey affect the conductors. The effect ofrapid alternations invirtually increasing the resistance ofconductors has been brought before usinSir William Thomson's Presidential Address, and itbecomes neces- sarytorecognise clearly what practical limitations areimposed bythis effect. ‘Myattention wasdrawn tothis matter twoorthree years ago,when Iwascalled upon toexamine and report upon anin- stallation ofabout 1,000 100-volt lamps supplied direetly byan alternator working atabout 150~/. Itwasfound that there rasavery serious drop ofpotential inthe conductors, and, in ‘consequence ofthis, the lamps could not bemaintained atfull Power. Onexamining thecircuit Ifound thecables were of ample size, and that thelosscould not beexplained intheusual may. Asa result Icame tothe conclusion that itwas tobe secounted foronly bysome kind ofinductional action, duetothe alternating current, and Iadvised that adirect-current dynamo thould beputdown inplace ofthealternator. ‘This wasdone 02 ALTERNATE CURRENT WORKING. {May2, andwithaperfectly satisfactory result.Thelampswereforthe first time run attheir proper power, and nofurther troublewas experienced. Atthat time Iwas, like most other electricians, adisbelieer ‘inalternating currents, which Ivaguely regarded asessentially vunsatisfactory. Ithinktherewasamoresubstantial basisfir this opinion than isnow generally admitted. Alternating cur rents asthen most often used were not satisfactory ;andifgrest -eare isnottaken atthepresent time, some ofthe olddifficaltes will beencountered, with theresult ofunnecessarily diserediing thetransformer system ofdistribution, FromSirWilliamThomson's Address,andfromsomefurther ‘information hehaskindly sent me, Ihave worked outandtabi- datedsomefiguresshowingwhatrestrictions areimposedbythe virtual increase ofresistance ofsolid conductors with alternating -eurrents, asbearing onthequestion ofperiodicity. SirWilliam Thomson gives some figures foraperiodicity o 80,towhich Ihave added some columns giving diameter i» inches, sectional area, and the current which the practicable sizes will carry, taken onthe basis of450 amperes persquat inch,becausethatdensitygives1percent.lovepermilewil 2,000 volts. Italso gives about 1percent. Lossper100yanis with 100 volts,or,moreexactly,1°15percent,Ihavealsogivee thenumberofwattsat2,000volts,andat100volts,forexch size, toshow the limits forordinary primary andsecondary workingwiththisdensity. Thesefiguresaregivenfor80,or100,andfor188“Ly,thediameters ofconductors forthesue percentage increase ofvirtual resistance over ohmic resistant beingintheinverse ratioofthesquarerotsoftheperiods: | 180 ALTERNATE CURRENT WORKING, 08 Table I—Virtual Resistance, de.ofConduetors with Alternate Currents. {mame |Ame|ascanner|Cantee Tr] |atesfavesMONS| v0|a)vase|aan)setae|a[sro]20015soos}1767|274]"24%|133|26,000]13,900x0]tera]siee|487]8%|220|440,000]22,00025|842}4908|-700|1% [om|||poo|ores|ay2mlras |ose||ow||100{3987|7,864)1217 |9-8times |...oswee1900foaz|785,400{1217|astimes |.|||9]9549)ooe2|-oo8|fLewttant|45|90,000|4,500 | |18|7086)254-4|-394 8% 178|856,000] 17,800 ||{a|anode aufange|efem|e|l |¥79)ors)ara|om|{Fe )32|64,000]3,200 [es]regwo|ama]894|savezene)sao|) |1996]-7a22]24|ase]are% ow|oom| Itwillbeseenthatfor100~~,withwiresof9millimetres,or “35inchdiameter, theincrease overtheordinary resistance isalmost imperceptible;foradiameterof13°4millimetres,or“53inch,it isouly 24percent.; while what may beconsideredapractical limitisreached at18millimetres, ornearly}inch.Atgreater diameters the increase isprohibitive, and multiple conductors, tabes,orstripsmustbeused.Probablyanincreaseof8or10 Percent.willbeconsidered quitepermissible. Itsoundsagooddeal, butitmust beremembered that itisonly 8or10percent. ofthe1or2percent.lossforwhichprimaryconductorsare ‘usually calculated, Lookingatthelimitations asregardsoutput,itmaybesaid thatforprimary distribution, even atnohigher tension than 2,000volts,thevirtualresistance effectis,ifnotquiteunimportant, _608 ALTERNATE CURRENT WORKING, [Mayfv, fatany rate notvery serious. Acurrent ofabout 200 amperes may beconveyed without theeffect showing itself toanyvery objectionable extent. There may bedifferences ofopinion oo thesubject, butmost people will agree that 300,000or400,000 watts isquite asmuch asshould besupplied through anysingle conductor. Inmost cases, even atordinary tensions, subdivision oftheprimary willbenecessary oradvisable before such apower isreached, With still higher pressure, ofcourse, enormons powers canbedealt with without exceeding the permissiblesize ofsolid conductors. Itistobeobserved that there isnothing,in this virtual resistance effect, tonecessitate the reduction of current bytheemployment ofvery high pressures. Ivisonly with thelow-tension secondary conductors thatthis effect becomes inanyway serious. With 100volts themaximum load atthisdensity that canbeputonasingle solid orstranded conductor isabout 13,000 watts for133/~_/, 18,000 for100 1, and 22,000 watts for80~L’, unless much more than the8to10 percent. increase: canbeendured. This only means inlage distribution work asubdivision ofthe primary and secondary feeling points inaccordance with this limitation. ‘The inco- venience hasonly toberecognised tobeavoided. Fortunately itisnotonethat with ordinary care islikely tolead toanyprac tical inconvenience. Itexists with alternate currents, butitisa matter that thetransformer system renders itvery easy toavoid. Transformers lend themselves tosubdivision ofwork, andwith subdivision this difficulty does notexist. Itmaybeworth noting, inconnection with this partofthe subject, that theimprovement inalternators, and the great ess with which they andtheir circuits canbeworked and controlled, may lead tosome revival ofthepractice ofsupplying direct from low-tension alternators. Ifthis isdone forany butvery «mall installations, itisevident thatspecial precautions must beobserved toavoid the difficulties which have been referred to. One other remark onthissubject. Very large conductorsart tobeavoided inthe transformers aswell asoutside ofthem. ‘This isonly mentioned onaccount ofitsbearing ontransformers a ALTERNATE CURRENT WORKING, oes foranyvery lowtension work, such aselectric jointing orwelding, where oneortwoturns ofsecondary conductor isoften sufficient.* 4.Transformers.—There appears tobeageneralbeliefthata highperiodicity isbest fortransformers, and that this gives the greatest output, or—what amounts tothesame thing—that with thesame output inthe twocases ahigher efficiency isobtained when theperiodicity ishigh than when itislow. This supposed fact has often been stated. Professor Forbes—who has, Ibelieve, given much attention to thetheory ofthisvery difficult subject—contributed apaper Inst year,t consisting ofamathematical examination, which issum- marised and concluded byastatement that the periodicity may bediminished without lossofefficiency iftheironbeincreased ; thecorollaries being, ofcourse, that the periodicity may be +The follovingparalare wero condcion for primary iret sorting at2000 lly td tw iw sw for varons numbers ofipHingewprotprada cormHenlytol0apeseroarehCinroytesrof estaspernil,inshime~60-<nvmeofape te tolonng secs omy wth he save rl al owe afhemsetecedbyteiannacetae Fe ne[eeeoe|age[eae es es eeee — wo|2|Se|Se|aes|see veataintntadeay atnuihonaallveigail, contrac“SStsponstatotserPorewallpotslyStytpnanon’ Vole Xv ‘2 06 ALTERNATE OURRENT WORKING, (Say, increased without lossofefficiency ifthe iron bereduced, snd thatwithanygiventransformer anincreased periodicity relts inincreased efficiency forthesame output, oringreater ontpt with thesame efficiency. Professor Forbesconfirmed thisviewrecently, inthedix cussion onapaper byMr. Kapp before the Institutionof Civil Engineers, bypointing out that oneofthe reasons whyia America they used ahigh periodicity was-that “they wantel “to get the greatest output from their plant, and transformes “had tobeincreased insizeinorder togive thesame output “they lowered thenumber ofalternations.”* And, inhisrecest paper on“Central Stations” before this Institution,t Profesr Forbes reiterated this. Hestates that thecontrary is“oppaed “alike toexperience andtheory. Westinghouse ‘knows this. 1 “am informed Mr.Ferranti knows it,andMr. Zipernowskibs “told me that hefinds itso.” Mr.Kapp,whoworksat80“, doesnotgosofar# Professor Forbes, forhethinks that onaccount ofvisoos hysteresis there must bealimit beyond which anincrewe becomes disadvantageous, Hewisely adds that itisnotposible todetermine thislimit ontheoretical grounds. On the other hand, ithas been said that itisnot mecesary toincreasethesizeoftransformers evenforverylowperiodicities because, when thenumber ofeycles issmall, itispossible to work atahigher magnetic density. Tventure toquestion theaccuracy ofboth these extreme views. Sofarastransformers areconcerned, andspeaking onlyfrm experience, Ithink that high and lowperiodicities areboth wrong, and that amedium rate isright. For agiven outpat andefficiency, notonly must transformers bemade large forlow rates—they must also bemade large forhigh rates; bat between thetwothere willbefound @periodicity giving the smallest sizeoftransformer, orthegreatest output,orthehighest efficiency. *Proc. Inst. CE, Febraary 1,1889. +ournal, xvi, p.16(Feb. 26,1889). ve; —=—ADRENNATR GUBRRNT WoRER. an Fig. 2illustrates this. Itshows the rise oftemperature ofone ofmy transformers working respectively at75,at 100,andat125“\/persecond. Theprimary andsecondary virtual potential differences were measurably the same inall threecases,andineachcasethesameloadwasused,consisting of ‘certain number ofglow lamps. The test was continued each Bg yahERLELa Cr pith~Re a8Ceeett|Zenneeel aswe Pe Pee retle~ Ge 7 CCPPT ft|V7 Seen TTA 77 i SA Seer LV7 DasasaBaV/A CCCAl Coo 8 | time long enough forthetransformer toattain itsmaximum temperature. This was taken byasensitive thermometer thebulb ofwhich was placed directly onthe iron, and covered byapacking ofcottonwastetoensurethat,asfaras possible, atleast theactual temperatureZof theiron should be indicated. Another similar thermometer wasplaced afewfeet cos ALTERNATE CURRENT WORKING, [May away toshow thetemperature oftheroom (which wasfairl constant atabout 60°Fah.). Thecurves give theactual increments oftemperature, which may betaken assome sort ofmeasure ofthe relative efficiency attained. With anygiven output thesame riseoftemperstare is,ofcourse, always obtained with thesame efficiency, quitesyar fromhowthatefficiency ismadeup.Itwillbeobservedtht thelowesttemperature orthehighestefficiency isobtainedx 100 7; that the efficiency islower at125, and lowest at BAY Onthese lines Imay suggest amethod ofarriving atthe efficiency oftransformers, which, though lacking indirectness it simple, andrequires only ordinary appliances andfacilities, Firstrunthetransformer forsometimeintheusualwas,00 incandescence lamps, the output being measured. Itismt necessary tomake anymeasurements oftheprimary. Bymeans ofathermometer findtheriseoftemperature ofthetransformer. ‘Then, when the transformer has cooled down, send acom tinuous current through itand measure thepower inwatts thit raises thetemperature ofthewhole transformer thesame amoust aswhenworkingwithalternate currents. ‘Thistakessometinttocarryout,butitisaccurate, andhastheadvantagethatthe lost power ismeasured indirect-current quantities, about which there isnoquestion;whilethepowerexpendedinthelamps, being inaninductionless circuit, iseasily arrived atcorres. Awattmeter isthe best instrument touse for the direet-currat readings, as,onaccount oftheriseofresistance ofthecondactor (which isnecessarily required tocarry much more than theaso! current), some adjustment isnecessary inorder tokeep the power constant, and this adjustment iseffected much mow conveniently with awattmeter incircuit than ifanammeter and avoltmeter arensed. 5.Lamps, &’e—Under thishead allapparatus forutilising «t measuring alternate currents should beconsidered, sofarasthe action isaffected bythe periodicity. Imust, however, confine myself tothelamps, which arethemost important,andarethe only apparatus astowhich Icanmake any definite statements. | a ———e 1803 ALTERNATE CURRENT WORKING. wo ‘The paper ofProfessors Ayrton and Perry read last year tefore thePhysical Society® showed that theefficiency ofglow lamps isthesame forboth direct andalternating currents. The authors ofthat paper didnotdirectly allude tothequestion of relative efficiency with different. periodicities, but they evidently hadthis point intheir minds, astheperiodicities used inthe experiments appear tohave been varied from 83to226without sowing anydifference inefficiency. Itmay betaken, therefore, thatnodifference exists. Probably forvery lowperiodicities this ‘wouldnotholdgood,butforordinary ratesthequestionmaybe considered settled. Atvery low rates the temperature ofthe filament would follow tosome extent the changes ofthecurrent, and wouldbeaffectedbothinlifeandinefficiency. Isuppose thatatordinary rates there isnovariation oftemperature, the filament acting asakind ofthermal fly-wheel, maintaining its coadition unchanged throughout thewhole period. Weallknow, forinstance, that, especially with low-resistance lamps, when the carent isinterrupted the time occupied incooling down to Wackness isquite appreciable. The periodicity atwhich the temperature begins tofollow the variations ofcurrent isapoint ofsome interest, and itisone that can bevery easily investi- gated. There is,asfarasTam aware, noavailable evidence regard- ingthe lifeoflamps with different rates, oreven astothe relative life with direct and alternate currents. Atone time it ‘asconsidered thatalternate currents wouldgivebetterresultsthandirect currents, because with thelatter there wassupposed tobesometransference ofthematerial ofthefilament inthe direction ofthecurrent. Nothing hastranspired toconfirm or todisprove this supposition, and, asthe number oflamps towinuseoneach system isenormous, thevery absence of evidence may betaken asaffording astrong presumption in favour ofthe belief that thelifeoflamps isindependent of whether thecurrent isdirect oralternating, and independent alsooftheperiodicity, within practical limits. *Jour, Phys, Bee, vol. ix,p.208, oo, ALTERNATE CURRENT WORKING, [aay x, Altemate-current arelamps are notseriously affected by variations ofperiodicity, within afairly wide range. Thefeeding current of@parallel Brush alternate-current lamp remains sensiblyconstantbetween76and125~\/,butwiththelover rate aslightly higher impedance orresistance isrequired tosecuresteadiness ofthelight. Seconpary E.M.F. oF TRANSFORMERS. Thebestpotential differencetobeusedontransformer cireuits hasbeen agood deal discussed, especially with reference tothe voltage oflamps. Itisvery desirable toarrive at some conclusions onthis matter, notonly astoglow lamps,bat astothegeneral question. ‘The balance ofopinion—which is,however, notvery decisive— appears tobeinfavour ofusing 50-volt rather than 100- volt lamps, sofarasdurability ofthe lamp isconcerned. ‘Thestoutlow-tension filament shouldlastlongerthanthefiner one forhigh tension, because disintegration does notsosea reduce the cross section tothepreaking point. Itis,hor ever, not aquestion only oftife, asthedisintegration ofthe thick filament causes blackening ofthe globe; and ifiteax beshown that this blackening, andtheconsequent lowering of theeffectivelight,ismoreseriousthanin100-volt lampe,thea theadvantage ofslightly longer lifewillnotofitself besuficient torecommend 50-volt lamps inpreference toothers ofhigher voltage. Ofcoursethemostseriousobjectiontoloweringthevoltage istheincreased costofthesecondary conductors. Thisisuctimportant insmall andvery compact installations such ascanbe soreadily supplied bytransformers—for instance, insmall shops andordinary houses,where#separate transformer maybeplaced| within afewyards ofthelamps, andwhere thefallofpotentialit negligible;butforlargerworkitisadvisabletokeepthe potentialdifferencehigh,aswithdirectcurrents,andforthe additional reason that the useofvery large alternate currents | ‘withcorrespondingly largeconductors,isaccompaniedbythe| 90.) ALTERNATE CURRENT WORKING, ou inconveniences arising from thehigh virtual resistance already alluded to. Itisimpossible, however, tochoose anE.M.F. that willmeet allrequirements. For instance, high candle-power lamps are coming largely into use, and asthis practice extends, higher ELF.willbedemanded inordertokeepdownthediameter of the “stick.” Bat, ontheother hand, lamps ofsmall candle-power can be more easily made forlowthan forhigh voltage; andasthere isa growing tendency touse small candle-power lamps which in many situations give allthe light that isrequired, this point should bekept inmind, especially asitisvery probable, that beforelongademandwillspringupforlampsofthreeorfour candles. Iknow itwill besaid that itiseasy toreduce the candle-power ofordinary lamps byinserting animpedance coilin thecirenit. This isoneofthemany conveniences oftransformer ‘orking, butitshould notberesorted toforpermanent. purposes, 14,however little power such acoilmay absorb, lamps reduced in thatway, being worked atlessthan their normal candle-power, arevery inefficient. ‘Then there isanother reason forsometimes preferring 50volts to100 volts, that Ishould like tomention, and that isthat alternate-current arelamps donotrequire more than 50volts terminal pressure. Itmay beexpected that many such lamps will beused ontransformers—for instance, atsmall business places, where perhaps thebulk ofthelighting willbeinternal and byincandescence lamps, butwhere itwill beconvenient tohave one ortwo ares outside. We know that when ares are used in parallel with incandescence lamps ondirect-eurrent circuits it isnecessary, inorder togetthe best results, tohave 12to15 volts drop ofpressure inafixed resistance inseries with each arc. Thus has arisen the custom ofusing 65volts formixed arcand slow lamp circuits, where the potential difference atthearclamp terminals isreduced toabout 50volts bytheinsertion ofsuch 4resistance, theincandescence lamps,however, beingmadetowork at65volts, This isavery wasteful arrangement, but,apart from itswaste- ow ALTERNATE CURRENT WORKING, (May 2, fulness,65voltsisaveryinconvenient pressure bothfromthe manufacturer’s and the consumer's point ofview. Itdoes not lend itself tocombination asdoes apressure of50volts dynamos, lamps, ortransformers, for65volts cannot beusedio pairs on100-volt cireuits ascan such apparatus for50volts. It isanirregular andinconvenient pressure that isdemanded bythe necessities ofdirect currents, butthat Iamglad tobeabletomy need notbeimported intoalternate-current work. ‘Dhis very serious loss canbealmost entirely avoided with alternate-current arcs,asthenecessarysteadyingeffectisreadily obtained bytheuseofanimpedance coil,absorbing much ies power than does aresistance ;and assuch arcs require only 35t 40volts between the carbons, instead of45to60with direct currents, itissufficient toprovide 50volts potential difference between thesecondary conductors, But itmay besaid that, asimpedance coils absorb litte power, thearelamps may beputon100-volt circuits inparle. This may bedone without much objection sofarasthelampand impedance coil areconcerned, butthere isthedrawback thatthe «plant-efficiency”” ofthetransformer will then below. Theoutput ofatransformer islimited partly bythesizeof the conductor. The actual output ofa100-volt transformer supplying arcs through impedance coils will notbemuch more than theproduct ofthecurrent x50volts, while forthesame output inwatts onglow lamps itwould only have toyield about half thecurrent. ‘Therefore weseethat, although notineficieat, themaximum allowable output ofa100-volt transformer ot parallel arclamps isonly about one-half ofitsordinary maximum output. Itwillbesaid that twoares may beruninseries onthe 100-volt secondary. This istrue, but two ares arenotalways wanted; and, further, the working oftwo arcs (which mus forthis purpose beshunt ordifferential instead ofsimply series-wound) inseries onaconstant potential difference is theleast satisfactory wayofarranging arcs, whether for"direct ot alternate current, asinthiswaythey affect oneanother themest; having neither themargin ofcontrolling power, bycurrent, which 13995 ALTERNATE CURRENT WORKING. ous isprovided intheparallel constant. potential difference arrange- ment;norbyE.ML.F,,asintheseriesconstant-current arrange- ment. Then, again, lamps soconnected must beextinguished together, asnocut-outorswitchcanbeusedunlessitismade toinsert resistance orimpedance totake the, place ofthe extinguished lamp. Onthe whole, therefore, itispreferable tohave 50-volt secondaries where ares aretobeused, ortoadopt theinter mediate plan ofconnecting themiddle ofthe100-volt secondary (oathird terminal, using athird wire forany ares or50-volt lamps that may berequired. Itmay beworth while topoint out thattheuseofathird wire ontransformer circuits may lead to difficulty ifnotcarefully considered. With asingle transformer itisaseasy tousethree oranyother number ofterminals forthe lamps asitistodo soonasetofaccumulators, andtogetany corresponding potential difference. The various sections ofthe secondary may beunequally loaded bythis arrangement, but the primary responds asawhole tothedemand made upon it.If,how- ever,insteadofasingletransformer, twotransformers areused,with their primaries and secondaries respectively inseries, then the useofthethree-wire principle isonly possible when itisapplied toboth theprimary and secondary circuits, unless theload on each ofthetwo secondaries isalways tobealike andthesame, in which case, asscarcely need besaid, there isnoparticular advan- tageinusing three wires. Iftheloads arenotequal onthetwo, then the primary ofthe least loaded transformer acts asan impedance coiltotheother primary, andallpower ofself-regula~ tion islost. One remedy forthis istoapply the three-wire Principle totheprimary aswell astothesecondary;butthisis 4serious matter, and one that could only beundertaken where thewhole service demanded it,Another plan istohave two mains fortheprimary andthree forthesecondary, andtojoin the primaries parallel, thesecondaries beiug inseries with thethird wire atthe middle terminal, Incases where arcs have nottobeused, and intheabsence of anyvery strong evidence ofany difference oflifeinfavour of 50-volt lamps, Ithink itisbest touse50volts forsmall separate ou ALTERNATE CURRENT WORKING. (May 2, installations, and 100volts forlarger ones; butitisevident from whathasbeensaidthattherearesomanypointstoconsiderthat itisquite impossible togive any definite negative oraffirmative reply tothe question, “What isthe best secondary EMP. to “use?” Eachcase,oratleasteachclassofcases,mustbesettled ‘onitsmerits. Ifforgeneral convenience any particular pressure ‘isselected, itisobviousthatitwillhaveaboutequallystrong arguments forandagainst it. ALTERNATE-CURRENT ARCS. Reverting tothis subject, itistoberegretted that very Aittle appears tobecommonly known about A.C. arcs. Itrut that inthediscussion ofthis paper some imformation maybe furnished with regard tothem. Astotheregulating mechanism oflamps, although some difficulties were encountered inmodifying the Brash a lamp forthis work, the result has been perfectly satisfactory; thefeeding range isvery small—quite assmall aswith diret currents, and with nogreater loss ofpower inthe lamp box. Some trouble was experienced ingetting ridofnoi andvibration inthecoilsandmechanism, but.thishasbeen completely overcome, thelamp itself being quite silent. This isreadily proved byconnecting thecarbon-holdere byaflexible wireandadjusting thecurrenttothenormalvalue,whenititfound that there isnonoise. AnA.C, arcisnotsilent, asiswell known, adistinct hum being setup,bywhich itisquite possible toascertain theperiodicity. This humming noise, which canscarcely beexpected tobe -overcome, renders A.C. aresunsuited forgeneral useindoors;bat foroutside illumination, and forsuch places asrailway stations they willcertainly find avery extensive application. Iventure tothink that the opinion sooften expressed, that A.C. aresgive much less light than those supplied with direct current, isfounded on amisunderstanding ofthe proper conditions. Itisquite truethatiftwoarcsarecompared, supplied with thesame amount ofcurrent, thedirect-current lamp willprove thebetter one. This, Ithink, isbecause thebasso 1893) ALTERNATE OURRENT WORKING. as ‘comparison isnotright. Inmany cases theA.C. ares aremore efficient. ‘They should bejudged bytheenergy, andnotonlyby thecurrent. Ihave mentioned that anA.C. arerequires only 35to40voltspotential difference betweenthecarbons,whilea direct-current arerequires 45to50volts, ‘The mean may be taken at87-5 volts and47-5 volts respectively, andtoexpend the same energy inthetwo cases the current should therefore bein theinverse ratio ofthese values. ‘Thus anA.C. arc, togive the same light andtoabsorb thesame power asa 10-ampere direct- current are, should bemade for12°6 amperes. Inonerespect A.C.arcsareinferior, andthatisintheir power ofdirecting thelight downwards. They have no“crater,” and therefore donot send the light down asdodirect-current arcs. ‘Onthis account itisadvisable tousereflectors above them, unless they aresuspended below andnear agood reflecting ceiling or roof. But even this drawback iscounteracted inparallel work by their greater efficiency. Ona65-volt circuit with direct currents, asImention elsewhere, there isawaste of100 to150 watts with 410-ampere are,and inthe same proportion with other currents. This isavery serious proportion ofthewhole power, and itis only allowable towaste itbecause, asalight-producer, thearcis somuch more efficient than the incandescence lamp. Never theless, any means ofavoiding the loss istobewelcomed, andwith A.C. arcs itmay begreatly reduced bytheuse of impedance coils, which, ifproperly designed, waste very little power, andhave allthesteadying effect ofresistances. Savery AND SAFEGUARDS. Itake this opportunity ofbringing upthis question, not ‘because Ihave anything new tosaywith regard toit,butin order, ifpossible, tobring it little nearer to» rational settle- ment, Itwillbeacknowledged thatthefirstdutyofan“undertaker”‘using high tension istoprovide forthecomplete and absolute safety oftheconsumers. Ifthere arenomeans ofdoing this as ‘ALTERNATE CURRENT WORKING. [Mayto, then the system isfundamentally bad, and must. sooner orlater bbeabandoned. ‘There areseveral ways ofprotecting thehouse mains more or lesscompletely. Ionly need allude tothree typical plans, allof which have already been discussed before this Institution. Oneistheplan introduced byMr.Kent,ofentirelyseparating theprimary andsecondary conductors inatransformer byasheet ordivision ofearth-connected metal. This isopen toadverse criticism onsome points, butonthewhole itisaneffective plan. Another arrangement isthe ingenious one ofCapt. Carder, consisting ofasortofstatic mouse-trap, which immediately goe off when anunauthorised P.D. enters it. ‘The third plan isthat ofearthing thesecondary, anditwoul! bevery instructive ifthose who sostrongly oppose thisplau would state just one real objection toit. Itisaperfectly sure method, andcosts nothing. Ttrendersitabsolutely impossible fortheconsumer toget dangerous shock under any possible combination ofcircumstance: whatever. ‘The faults that may occur are— (a)Contact between primary main ormains and earth. (b)Contact between primary and secondary inatran former. (c)Contact between aperson and @portion ofthe secondary conductor. (@Contact between that person andearth. Now ifallthese contacts occur atthe same time, the «person ”willgetashock. But ifthe secondary isearthed, noshock can beobtained, except, perbaps, aslight oneduetothesecondary P.D. ‘This simple plan basbeen approved byalmost everybody, but itisreceived with opposition onthepart ofsome ofthefire insurance offices and others, Itisdifficult toascertain where thesupposed danger is.If thesecondary isearthed, theeffect ofacontact between primary andsecondary, whentheformerisaccidentally leaky,isatonttoblow outtheprimary safety fuse ofthetransformer, and 1869.) ALTERNATE CURRENT WORKING. eit cutouttheoffending house. This isthesimplest andthebest, coursetotake,andprevents thecontinuance onthecireuitof transformers that contain faults, ‘The Brush Corporation has adopted this plan with perfect, success onalarge installation abroad, and intends adhering toit ‘where local circumstances permit, Itistoberemarked that itisnot open tothe objection that has been stated against earthing ofone oftheprimary conductors, viz., that the telephones may beaffected, foritis notpermanent. The moment acontact occurs that could lead toadisturbance ofthe telephones, the fuse prevents that disturbance being continued. While effectually removing liferisks, this precaution does notincrease firerisks; infact, ittends togreater safety,asit issome sort ofcheck onthe character ofthe house wiring work, and soreduces chances offaults such asshort-circuiting ofthesecondary. ‘The rupture ofaprimary safety fuse isnot serious matter, and inany case itisonly what takes place under ordinary cireumstances when any fault occurs that remnlts inanexcess ofcurrent: Astostatic effects, Ithink itwill befound that there will always beenough general leakage toprevent such effects being a canse ofdifficulty, even ifthesecondary circuits arenotearthed. Hysrenests axp Eppy-CuRRENTS. Some explanation isnecessary byway ofpreface tothis section ofthepaper. The experiments described were made with adirect-current dynamo, and onthat account they only touch the fringe ofthis division ofthe subject. AsIhad nosuitable alternator, Iwasunable topursue thequestion asfarasIwished todo,butIhope Tneed make noapology forbringing forward such results asIhave been able toarrive at. AAsfarasTamaware,nosimpleorpracticalmethodisknown bywhich the power absorbed ineddy-currents canbedetermined andseparated from that due tohysteresis inthearmatures of dynamos. This isapoint ofsome importance both asregards os AUTERNATE CURRENT WORKING. (Saya, alternators anddirect-current machines, formany reasons; oneof theprincipal being that intheabsence ofexact knowledgeo these matters itisdifficult orimpossible todetermine towhat ‘extent the lamination ofthe iron cores should becarried. Ther arewide differences inthepractice ofthevarious designersax! makers ofdynamos inthethickness oftheiron used. Lamination may becarried tosuch adegree offinenessfr thepurpose ofreducing eddy-currents astoaddvery materials tothecost ofproduction, theiron increasing rapidly inprices the thickness isreduced ;anditmaybethatthepointissome times passed beyond which nogain inefficiency isobtainel. ‘This isaquestion wewish tobeenabled tosettle. Until Professor Ewing* and Dr.J.Hopkinsont read their RoyalSocietypapersonthesubject,verylittlewasknownoftitheating effectofchanges ofmagnetism. Professor Ewingope’ thepart ofhispaper dealing with this matter inthefollowing terms (p.552, Phil. Trans.) :— “The energy expended “inacyclic process ofmagnetisstion “can take noother form than that ofheat diffused througbost “the substance ofthemetal. Experiments have been madeby«Jouleandotherstodetermine bydirectobservation theheating “effect ofmagnetisation iniron.¢ “In most direct measurements ofthis quantity nodistinction “jsmade between the heating effect due tothe induction of “electric currents andthat duetochanges ofmagnetisation per “ge;andsoexcellent anauthority asProfessor Rowland, writing “in 1881, hasexpressed himself asdoubtful whether changesof «magnetisation, considered apart from thecurrents they induc, «giverisetoanydevelopment ofheat atall.” Further on,Professor Ewing, referring tothedissipation of energy byhysteresis, which hehad determined bylaboratory experiments, says(Phil. Trane., pp.553,554):— «ProfemorBvog, “Experimental RewarcherinMactan”Fi yeeDe.J.Hopkinton,«Maguotetion o10,Phi.Tron,Ape188.Hosie, Phe op volie 188; Oroe, Pi aps ol 5B van, Huw Chom 176, Casing tm, SeChao tePhys ns Toeile,Pothe.ofArteSoc107 | 1838) ALTERNATE CURRENT WORKING. ow “These developments ofheat aresosmall astomake it “apparent that the very considerable thermal effects which ‘reversals ofmagnetism cause intherevolving cores ofsome “dynamo-electric machines must bedue almost wholly tothe “internal induction ofcurrents, sofarastheyarenotduedirectly“tothecurrent circulating inthecoils ofthearmature. “The experiments have apractical value inshowing that cores “which are90thoroughly laminated astorender theinduction of “carrents within them unimportant, donot involve any serious “lous ofenergy;andthattheefficiencyofamachinewithasoft “ironcoreorcoreswhosemagnetismisperiodicallyreversedneed “not, onthat account, bematerially less than that ofamachine “which has no such cores. ‘The absence of iron from the “armature has been claimed, onthe score ofefficiency, asan “important advantage possessed bysome types ofmachine, but “unless theclaim has some other basis itappears tometobe “illusory. Magnetic reversal does involve some lossofenergy, “but ifthe cores are properly laminated itissosmallastobe «practically insignificant.” Iqnotethesepassagesasshowingthepositionofthequestion. Professor Ewing andDr.Hopkinson give laboratory determinations ofthemagnetic loss iniron with various inductions, andunder various conditions. Itis,however, left somewhat indoubt asto whether thelossindynamo armatures will bethesame asinthe laboratory experiments, theconditions being very different inthe two cases. Onthis point Professor Ewing shows that hysteresis ismuch lese when the iron isinastate ofmechanical vibration than when itisinastate ofrest, and states that inadynamo thelosswillbe lessthan theexperiments onstill metal might lead ustoexpect. Another way inwhich thedynamo conditions may cause some difference isthepossible variations ofthemagnetic density that may exist inthecore. Such variation may beconsiderable in armatures oflarge size. Dr.Hopkinson (Phil. Trane., p.467) takes themagnetic loss inarmatures asbeingthesameasinthesmallpiecesofiron tested byhismethod, and gives onexample calculated foran armature core “finely divided, toavoid local electric currents.” i 620ALTERNATE CURRENT WORKING. [fay v3, Itistobenoted, then, that both these authorities regarded ‘eddy-currents asunimportant inlaminated armatures butinthe later paper on“Dynamo-electrie Machinery ”*Dre. J.and E. Hopkinson published theirfamousinvestigation intotheefficiency ofdynamos, andgave very fullparticulars, from astudy ofwhich itispossible toascertain the loss ineddies, onthe assumption that themethod ofcalculating themagnetic loss iscorrect when applied toarmatures—a point which Dr.Hopkinson wasnotable todecide positively; butbyassuming that itisapplicable, then, bysubtracting thecalculated hysteresis lossfrom theascertained total power unaccounted for,theeddy-current lossisobtained; and asthis process gives thehysteresis asaccounting foronly about 40percent. ofthetotal lossinthearmature core, itshows that theeddies could notbyanymeans beconsidered unimportant, although thelamination wascarried outtotheextent ofmaking the iron about 0-02 inch thick, : Inowwish todescribe amethod bywhich theamount ofloss‘ duetoeachofthetwocausesmaybeascertained experimentally | intheworkshop onapractical scale, Itisavery simple one, although not,perhaps, capable inallcases ofthehighest scien- tific accuracy. (a.) First runthedynamo onopen circuit atvarious speeds | (two readings willsuffice), with thefields unexcited, or,better still, entirely absent, and measure thepower absorbed infriction. Plot this value toany suitable scale, asOMinFig. 3,the ordinates representing horse-power, while theabscisse represent revolutions perminute. Inthefigure themechanical friction, being aquantity with which inthis investigation wearenotdirectly concerned, ix plotted below thebase line0M’,inorder that theother losses may bemore clearly seen. ' (b.)Nextrunagainonopenarmature circuit withthebrushe- | removed, andthefieldnormally excited, andmeasure thepower | absorbed. This willbethetotal power dissipated infriction, | hysteresis, and eddy-currents, ifany, and must beplotted as Phil, Trame, part 1,1886, p.3314 mej ——Autanuars ooaaexs wontix0. a ON. Then theordinates lying between ONand0M’represent thesumofthelosses due tohysteresis and eddy-currents, and somewhere between M’and 0Nmust liealineOP,separating thehysteresis from theeddy-currents. DemonteLaninDynaAmatne FAeet || akEEEEEEEE oc ATI EERE.CCCECo Jy Ba EEEa EEE a iam Wd BH eae: AVA [| | Eat |22 sauaaietie LAAan cs Pes Pee oe Melee TT Wedesire,tofindthepositionofthisline.Thedifference vetween the rate ofincrease ofthe electric and magnetic losses ablesustoseparate themfromeachother.Itmaybetaken hat,with anygiven constant excitation, thelossfrom hysteresis ‘directly proportional tospeed, within certain limits tobehere- vou evn i s00gl on ALTERNATE CURRENT WORKING, (May 20, after stated. ‘The resistance (that oftheiron) being coutas', theeddy-currents must vary asthefirst power ofthespeed,ao} thelosstherefrom asthesquare ofthespeed.* Itwill now beseen that itiseasy toseparate therales which aredirectly proportional from those which areproportion! tothesquare ofthespeed. Professor $.P.Thompson, towhom Isubmitted thismetho, haskindly given methefollowing concise formulae forfinding the quantities -— PandP,arerespectively thecombined magnetic andelectric losses atthespeeds $and S,. k=magnetic loss; ky=electrie loss, P,\s'— PSt k=Ss ssi PS,—PS ‘Then,atanyspeed2,Hysteresis =kn; Eddies =kynt, Ifarmature cores areruninanexcited field before winding, itcan bedetermined towhat extent their lamination isslit factory, andwhether theironisofgood quality; andthen,afer winding thearmature conductors, another testshows whattbe additional increase intheloss is,and separates theeffectsof eddy-currents inthecore from those intheconductor, andthe!hysteresis fromboth. Themethod mayalsobeusefultoir| vestigate thelosses inthecores ofarmatures ofdifferent tspesordifferently builtup,as,forinstance, withironwireorive sheets. ‘This method hasauseful application inrendering itposible toseparate notonlythemagnetic lossfromtheelectric losin theiron, but todetermine thewaste byeddies inthecopper winding ofarmatures. Itisusually entirely amatter ofopinion astohowmuchof #Prot, Thompaon's“Dynamo-electric Machinery.” rddition, p10 168] ALTERNATE CURRENT WORKING, 6 thepower absorbed inrunning anarmature onopen circuit isdue toany ofthe various causes referred to; and now that itis customary towindarmatures withheavyconductors, itbecomes very desirable tobeable tofind inany particular case what proportion ofthelossisinthecopper, and what intheiron. InFig. 3areshown some results obtained with afour-pole (direct-current) Victoria dynamo run atvarious speeds upto 1,200 revolutions perminute, with constant excitation. Power readingsweretakenbymeansofaWhite'stransmission dynamo- meter, constructed bytheBrush Corporation forthepurpose of testing dynamos; andtheH.P. absorbed inmechanical friction in dynamo, belt, and dynamometer wassubtracted fromthe power spent electrically andmagnetically. Toreduce errors inthedynamometer readings thedynamo ‘wasfitted with twopulleys, onetwice thediameter ofthe other. This allowed ofthedynamometer being runatthesame speed, while two different speeds ofthe dynamo were obtained by changing thebeltfrom onepulley totheother. Readings were taken atfour speeds, viz., 450, 600, 900, and 1,200 revolutions perminute, correspondingto15,20,30,and40/~/persecond. Theposition ofOPwas readily found by“trial and error,” this being ashorter process foraconcrete instance than theapplica- tion ofthe general formule. The assumption onwhich this method oftesting isbased wasfound tobecorrect upto900 revolutions. Thereading atthehigher speed of1,200 revolutions, or40/~, showsthatdisturbing influences begantoaffect.the results byreducing therate ofincrease ofthelosses. ‘The con-tinuation ofthecurveONinafulllineshowstheactualloss; while the continuation inadotted line, which isaparabola on OP asanabscissa, shows what the losswould beifthe rate of increase holding forlowspeeds were maintained, ‘This shows that this method isperfectly applicable toordinary direet-current machines—for instance, totwo-pole dynamos upto 1,800 revolutions perminute,or30~./persecond—but thatit isless useful for alternators. ‘Thedisturbing causes areprobably rather complicated. They maybeduepartlyto’self-induction inthecircuitsoftheeddy- out ALTERNATE CURRENT WORKING. (May28, currents, tolower magnetisation (although thiseffect, ifpresent, invery slight, aswasshown bytheE.MLF. being almost perfectly proportional tospeed),andtovibration.“ Viscoushysteresis,"ifit exists, willraise themagnetic loss. There maybesuch anfeet, inspite ofthelowering ofthecurve, butitcannot berecognised, onaccount oftheunknown variation oftheeddies, atthehigher speed, Iron-cored alternators working athigh periodicities my ‘betested inthefollowing way, which willgive some approsch to correct results:—Let thetest betaken atalowspeed (not more than 30~L+), This willgivethehysteresis perperiod, andtheeddiesatthatspeed.‘Then,ifatanyhigherspeedtheEMF.is proportional tospeed, itmay beassumed that the magnetic induction isthesame asbefore, and the hysteresis perperiod will atany rate not beless than atthe lower speed. It may bemore. The difference between thecalculated hysteresis, OP, and the ascertained total loss, ON, will bethe eddies; of, ifviscous hysteresis isafact, itwillbeextra loss due tovisosty plus that due toeddies. Heating ofthearmature core, andconsequent increase ofthe resistance oftheiron, isnotresponsible forthebending down of thecurve ON inFig. 3(although itwould have that effect under ordinary circumstances), asthe experiments did not occupy sufficient time toallow ofanysensible riseoftemperature, ‘The effect isprobably very complex and obscure, but itisone that well deserves study, onaccount ofitspresent industrial importance, Itshould bestated that thedissipation ofenergy (which bas been spoken ofasbeinginthearmature)maybepartlyinthefield magnets. This was, however, notthecase inthe experiments referred to. The conditions necessary forthe prevention ot reduction ofeddies intheiron offield magnets arenowpretty generally understood,* butitshould beobserved that itisnotby ‘anymeans aseasytocheck such waste inalternatorsasinordinary dynamos. Itisboth better and easier toprevent than tocure them. ‘bee numerous practical notes onthi aubject scattered through Profexet ‘Thompeon's “Dynamo-eleetrie Machinery.” 12) aureuvate cone? wonenxo, ows If8good transmission dynamometer isnotavailable, results almost, orperhaps quite, asgood may beobtained byrunning themachine atvarious speeds bymeans ofanelectric motor, and measuring thepower required bythe motor—first with thefields ofthe machine unexcited, togetthemechanical friction, and then with the fields excited, toget thetotal loss. The motor should then berunalone, tofindwhat isabsorbed indriving it ‘atvarious speeds. ‘The increase inwatts over these last values, with the internal losses inthe motor conductors calculated out and subtracted, will give, with avery fairdegree ofaccuracy, theenergy dissipated inthemachine under test. ‘The upper curve, O'N’, wasplotted from readings taken after the armature was wound, and the difference inthe ordinates between ONandON’ shows theeddy-current lossinthecopper conductor atvarious speeds. This loss, like that oftheeddy- currents intheiron, rises asthe square ofthe speed upto900 revolutions, or30L%, above which itrises rather less quickly. The full line ON shows theactual loss, thedotted line above it showing what itwould beiftherateofincrease were maintained. The difference isnodoubt due toself-induction. aeaevan“He 24 any72aope e a a a a a fee That the effect ofself-induction isnot noticeable unless the eddy-eurrents areconsiderable isinaccordance with experiments 626 ALTERNATE CURRENT WORKING. [May280, that Ipublished some years agoinconnection with theBrish dynamo.’ The results then obtained areshown inFig. 4.The straight line connects speed and thepotential difference atthe terminals with the laminated armature;thelowerqndslightly curved lineshows thepotential difference with theoldform of cast-iron armature, which had only coarse divisions, seareely meriting theterm “lamination.” ‘The curve shows that the E.MLF. rose less quickly than the speed onaccount ofthe serious generation ofeddy currents. Inthelaminated armature, although, ofcourse, there were still eddies, their effect was not sufficient tocause any bending down oftheE.M.F. line atany practicable speed ofdriving. Coming nowtoanexamination oftheresults obtained inthe experiments, and shown inFig. 3,weseethat the hysteresislos inthis particular dynamo—which Ishould mention iscapableof working continuously with anormal load of18,000 watts—ms 0-195 HLP. at450 revolutions, and 0:39 H.P. at900 revolutions thelatter corresponding to30~L7 persecond, ‘The magnetisationt wasabout 12,300 c.g.s. lines—not byany means ahigh density, ‘The armature had aring core, nearly square incross-section, oftwelve sq.in.area, andcontained 615 cub. in.,or10,086 cub. centim. ofiron strip 0-012 in.thick (No 30B.W.G.) separated bypaper. This isadegree oflamination beyond which itisinconvenient togo. The peripheral portion ‘waswound with thisiron inparallel bands }in.wide, inorderto reduce eddies caused bylines entering theperiphery. ‘Themain inner portion ofthecore wasofwider strip. ‘Wefind from thefigures that thelossinhysteresisat900 revolutions was asfollows :— 0°39 HP. =291 watts =0-473 watts percub. in,or0029 watts percub. centim. Astheperiodicity was30,thedissipation ofenergy percomplete period percubic centim. was9,613 ergs or0-001 watts, nearly. ©The Electrician, October 2nd, 1886. 4+The useoftheword “induction” insomany senses isapttocause 02 fasion, «Magnetisation”* would perhapsbebetterthan“induction,” «magueic invduction,” “density,” “magnetic density,” ete 893 ALTERNATE CURRENT WORKING. cn Itisinteresting tocompare thisresult, obtained inwhat may seem arather crude manner, with thevalues given inthepapers referredto.Prof.Ewingsays(Phil.Trans.,p.553)“...the“double reversal ofastrong condition ofmagnetism insoft iron “involves theexpenditure ofabout 10,000 ergs percubic centim.” Hegives theenergy as9,800 ergs, with «magnetisation of13,190 invery softannealed iron. Dr.Hopkinson gives 13,356 ergs, with ‘4magnetisation of18,250. ‘Much ofcourse depends upon thequality and hardness of theiron, but myresults come out very close tothose ofProf. Ewing and Dr.Hopkinson. This may betaken asproving the method capable ofconsiderable accuracy. ‘Theeddies intheiron, itwill beseen, arehynomeans tobeneglected, even with the veryfine subdivision that wasused. ‘Theeddies intheconductor were rather greater than usual, as thearmature, beingintended foraverylowspeed,lindalarger amount ofcopper onitthan usual, The bearing ofthese curves oniron-cored alternators andon open cironit-working generally isobvious, With »magnetisa- tionof12,000, the loss inhysteresis alone, at100~L, would beover1°5wattspercub,in.ofironcore,andtheeddieswould amount toagood deal more. Iwillonly addthat theeddies, both inthe iron and inthe copper, are greatest when no externalworkisbeingdone,aswasthecaseinnyexperiments. They are reduced bythe effects ofself-induction when the circuit isclosed. Adifference exists inthis respect. between generators and motors, adifference infavour ofgenerators." Iwill not further extend this paper bydealing atgreater length with this part ofthesubject. APPENDIX. Extract from remarks byMr. Mordey inthediscussion on Mr.Kapp’s Paper on“Alternate Current Machinery."—Proc. Inst, Q.E., March 5th, 1889. “Alternators might bedivided into two classes, those which had irou inthesrnature,andthowowhichhadnot.Ashisownmachine,madebytheBrush os ALTERNATE CURRENT WORKING. [My2, Corporation, andthatofFerrantiweretheonlyexamplesofthelatterclsdecid bytheauthor,hemightbepermittedtomentionthereasonswhichhadledhinvoidtheuseofiron.Hethoughtthemakersofdynamoswouldagreethtalex alltheillsthatthedynamowasheirtowereduetothepresenceoftheiintbearmature. Itsuseindirectcurrentmachinesmightbesaidtobeaseemevil,as,althoughattemptshadbeenmadetodowithoutit,allsucceadelpadsuch machines required itforstructural purposes. Iron inarmatures when word ata high magnetic deusty, and with rapid reversals orvariations ofmagoeisns Decame heated and wasted agood deal ofpower. Inalternators this bjecie applied with very much greater force than indirect-current machine, fitthelatterthereversalsofmagnetism werecomparatively slow.‘Thusthefiteatarrivedatinquiterecentpricticeisthatthemagnoticdensitythatcanbeumdition-cored alteruators must only beabout one-half that employed incontincarrentarmatures, ‘Thisalonemeantconsiderable increaseinthesizofwm:tures, without anygain inoutput orefficiency. Although thelosspereabie intwasreducedbydeereasing themagueticdensity,thearmaturehadtobemaelre* tocompensate forit,and thetotal losewas actualy increased, notreduced ine was used toreduce themagnetic resistance, toafford mechanical support. i» introduce sel-induction into the cireait. ‘The latter, anevil initself,wasail bea moder necessity, caused bytheconvenience ofworking alternators pnb Self-induetion, forthisoranyotherpurpose.hadprobablynotenteredthemis‘ofmost designers atthetime they fist produced their machines, excep. pep:‘asanobjectionable feature,whichtheycouldnotgetridof;butnowitwabicTwroughtforwardasanadvantage. Ifironwasindispensable forthisparpse,ihhewasnotpreparedtoadmit,itcouldreadilybeinsertedinsomepartoftheciwhere ample space andcooling surface could beprovided. and from which itcw!Jvceasilyremovedwhennotrequired,thetwaswhenitwasonlynecemy1= ‘ouemachine. ‘Thearmaturewastheveryworstplacetoputtheiron,forhi[presence wasthemost objectionable, anditwasthere least under control. Releris¢ tothemagnetic resistance ofarmatures, itwould heseen that there were ta twoways ofarranging thearmature-coils and thefield-poles. The poles cold Ye arranged north andsouth, facing one another, thearmature-coils having therparalleltothelinesofforce;oFthepolescouldbearrangedinsuccemiveaitoundanironringordrumarmature-core, ouwhichthearmaturecolsweewus:forplaced.Ifiromcoreswereusedintheformercase,theoutputwouldteataboutone-halfofwhatwasobtainedwithair-cores,onaccountofthenéework with reduced magnetic density, while atthesame time magnetic andeee” disturbances would besetup, both inthe armatare and field, wasting por’ causing objectionable heating. necessitating lamination ofthefields, anddemandenninereaseintheexcitingpower.‘ThesereasonsweresufficienttoexplainwhyJadnotused ironcores; butthere were others. Magnetic resistance, betree and pole, was made uppartly ofthenecessary clearance space, partly iamotoftheepaceoccupiedbythecopperconductors, andpartlyoftheiroracore.Now,ifjnwir-coredarmaturesthecorecouldbesoarranged,asa!qiPomsible, and was done inhisown machine, that itoceupied v0moreIuanwastakenbytwicethedepthoftheconductors iniron-coredarmatar 1089,) ALTERNATE CURRENT WORKING. 9 vouldbeevidentthattheformergainedinsteadoflostbytheomissionofiron. ‘Theygainednotonlyinmagneticresistance,butinalltheotherpointetowhich thehad referred, andthey hadthevery great advantage that there was nothing to limit themagnetic density atwhich they could beworked except the magnetic ‘mturation ofthe field-poles, the only lowes to"be met and provided forbeing ‘theondinary electrical lomes inthe conductor caused hythe passage ofthe‘carrent,andaemallwastebyedddy-currents intheconductors andsupports. Sucharmatoree hadthefurther advantages thattheir oelf-induction wasnegligible, their power ofself-regulation was very great, and that onopen circuit oFlight load theyranveryeconomically ;whereasitwaswellknownthat,whenironcoreswereused, thelowesundersuchcircumstances wereoftenveryseriousindeed.Hethoughtit ‘nasquite unnecessary touseiron cores fortheparpose ofobtaining mechanicalsupportforthearmature-winding, ‘Theholdinganddriving,withaverylimitedclearance,ofanarmatureconsistingofthiniroa,separatedlysuchunmechanicalsubstancesaspaper,andoverxound withcopperwire,itselfcoveredwithcotton,and‘expanding andthrowing ontfrom thecombined effects ofheating and centrifugal tetion, was aproblem that had never been very successfully solved in firect-curreat machines, and was certainly tobeavoided, ifpossible, under the stil more onerous conditions obtaining inalternators, where the elec- trical prearare wat wrually much higher than indirect-current machines, sod in which, as the author showed, the maximum mechanical strain on thewireswasgreater.Hoventuredtothinkitwasmuchthebesttomakethe armature stationary ashehad done. Itthen had only toresist the ‘ungential drag ofthe field. Hethought agreat mistake was made insome Alternators inusing Pacinotti projections. Inallcases there should be,as early aspoasible, asteady magnetic ax inthe field. ‘This could notbedove if projections were used. ‘The Zipernowsky and Parvons machines were faulty inthisrespect.Itwasimpossibletotell,byinspectionoftheformer,whichwasthe armatureandwhichthefield.‘Theyseemedquiteinterchangeable, andthiswns ofcourte notright. Inthese two machines itwould beseen thet themagneticresistancevariedverymuch,astheironprojections passedthroughtheexele.‘Thissecemarily Ledtolames inthe field, thelamination ofwhieh recently Introduced intheZipernowsky machineshowingthatsuchlossesmustbeserious.Therewasone‘ther machine towhich hewished toallude, and that wastohimavery interesting machine indeed. Hereferred tothe alternator (Fig. 2)shown and referred to byProfessor Forbesasinsomerespectsadevelopement ofhisalternator.Inthat ‘machine Profesor Forbes used theform offield which he(Mr, Mordey) hadin hiaalternator; butheused itincombination with aform ofarmature having, like thefield, only ove coil. Heventured topoint out that, foramachine vith aniron armature, that form ofalternator had some advantages. The armature might bedescribed asaGramme ring with the copper inside and the fron outside, ‘The magnetic ux was, ormight bemade. yerfectly even, the sir-gapwasveryshort,themagueticresistanceverylov,anditfollowedthatthe excitation required wasvery small. ‘The author hadalluded to.very interesting ‘and important mater, probably new tomost people,aaitcertainlywastohim, ‘amely, that thearmature-condyctors inalternators showed atendency toheat 650 ALTERNATE OUBRENT WORKING. [May23nd, tnuch more than inthe continuous-current machines, but inthis machine described tyProfesor Forbes there washardly anyquestion ofedy-currents inthe armazare- conductor, asthe coil wasnotewept bythelinesofforce.Further,itwasevident that very lowcurrent density could beused inthearmature-oll without sensibly aifecting tho sae, cot, oroutput, and itwns also clear that the insulation ofthe rrmature-<oll wasaverysimplematter.Anothergoodpointwasthatneitherthe armature-<oil nor the feld coil need be rotated. For amachine with an iron ragnetic elreuit hethought this wasaverygoodove;andhehadonlyovether remark tomake about it,and that was that hehad himself inventedandpatentedit sometimeago!Itwas,hethought,anaturaldevelopment ofhisalternator. He ‘wasgladtofindthatProfessorForbeshadhadthesmeidea,becnasethatshowed {ittobeaoodove.Muchbadbeensaidabouttheformofthewaveyieldedby ‘alternators. He had some time ago made anexperiment with thefist ofbis‘machines,usingthemethoddescribedbyProfestorAyrton,andfoundthatthecurve (Fig. )was almost asine-curve, On the subject ofthe lagintrans- formers, towhich Professor Ayrton had alio alluded, hemight beallowed womentionthat,inthediscussionofanotherpaperbytheauthor,hehadfiststated‘thefact,*andhaddescribedaverysimpleexperiment, showingthattheprimaryandsecondarycurrentsreachedtheirmaximaandminimapractically atthesametime.” Onthemotion ofthe President, ahearty vote ofthanks was unanimously accorded toMr.Mordey forhisvery interesting and valuable paper. ‘ThePaesipent announced thatthediscussion onMr.Mordey’s paperwouldtakeplaceatanextrameetingtobeheldonMay30. Aballot took place, atwhich thefollowing were elected :— Member, George Hookham, M.A. Associates. John Boyes. Philip Peters. William Foggin. ‘M.G,Simpson. Edwin Thornton, Students, W.H.Merriman, | P.Hawkins. Themeeting then adjourned. *Journal vp 16,218. : al 1999.) SUPPLEMENTARY REMARKS. est ‘TheOne Hundred andNinety-sixth Ordinary General Meeting of theInstitution washeld attheInstitution ofCivil Engineers, 25,Great George Street, Westminster,onThursdayevening, May 30th, 1889—Sir Wutam Tuomsox, F.RS. (L.&E.), D.C.L., President, inthe chair, The minutes oftheOrdinary General Meeting held onMay 23rdwere read andapproved. The names ofnew candidates for election into the Institution ere announced. ‘The Pnesipent: Intheordinary course [should propose that these names besuspended;but,asthisisthelastmeetingbefore therecess, ithasbeen usual insuch cases topropose that the candidates beballoted forthe same evening, Ibegtomovethat thatcourse befollowed inthepresent case. ‘The proposition wasagreed tounanimously. The following transfers were announced ashaving been approved bytheCouncil:— From the class ofAssociates tothat ofMembers— Richard Lewis Cousens. FromtheclassofStudents tothatofAssociates— Arthur Henry Lea. ‘The Presipent: Before thediscussion onMr.Mordey’s paper takes place Iwill askMr. Mordey ifhedesires tomake any further observations inreference tothe subject with which it deals. Mr.Morpey: Astime didnotpermit ofmyreferring atthe lastmeeting toalltheapparatus that Ihave here, Iwish tobe allowed todosonow, Inthefirst place, Iwould saythat theseries ofexperiments described inthefirst part ofthepaper wasnotquite complete. Inorder torender itcomplete itwasnecessary that oneofthe 632 ALTERNATE CURRENT WORKING, ‘Day2, itty,»Ce eS {Bee 7,!uy pow>| < es \WF rsa \ ou on (WA a as FaNNNeBie J a 2 ee, ey > om27,Fess (antl||pre Wh 1180) SUPPLEMENTARY REMARKS, oss alternators should berun asamotor, developing mechanically thefull power that itwas designed togive electrically asa generator. Sincethelastmeeting Ihavebeenabletodothis, and have runamachine which isintended foranoutput of 37,500 watts, asamotor giving 50horse-power effective. ‘The armature now before you isoutofasimilar machine (weFig. 5). Fig. 6isuseful asshowing theexact magnetic arrangement ofthe field-magnet. Asamatter offact, this figure illustrates themagnet ofthefirstmachine that wasbuilt. Inthelater ones, such asare’shown byFigs. 8and9,there isa slight difference ofconstruction, thehorns ofthemagnet, which yN i q ss, areseenasquiteseparate inFig.6,beingjoinedbyacastweb onthe outside, forthe prevention ofair disturbance when running. The whole machine isshown inFig. 7.‘The arma- ture has 18coils, and iswound for2,000 volts. Here isone cil[shown] which youwillfindonly weighs twoorthree pounds, complete with itssupports, Each coil iscapable ofworking atabout 3horse-power. ou ALTERNATE CURRENT WORKING. (May9b, Amachine ofexactly the same type, but of»largersize,is shownbyFigs.8and9.Thismachine isforanoutputof100horse-power, asgenerator orasmotor.Itrunsat500revolations. 7 }eS i on i: | ‘EW ia: ve i—_ | a ( fe IL It } Fio, &—100 ILP. Alternator with Exciter, Fig.8isalongitudinal elevation, partlyinsection,#asto showtheconstruction ofthemagnet,andthemethodofmounting TT : ialy 4 AV)/IB)= SOY Ke a i Fro. 9. thearmature coils. Theexciter, which isdriven direct, iscuried onabracket atone end ofthe machine, wus SUPPLEMENTARY REMARKS as Fig. 9isanend elevation, partly insection, The armaturecoilsareshownattherightoftheupperpartofthefigure,whileatthelefttheyareremoved inordertoshowthepolefaces.Thavehereanautomatic regulator, whichmaybeused either for alternate or direct currents. For alternate-current (DOO AUG: ANTTa AKAN eg: RROA | NDS Ss sii \ Sa a {//||/ssist/\|HES |nV1!||\Wi M/A | 3)WNSANA Wi|AA) a a |1/1) || VA\\\ HH 8) 1)ARUN ANY |eT|| L 1/ ear\\\\\\NBHI\ \\\ et tTNS Lio io 10—Compount Rely. working itismadetoactonthefieldoftheexciter,whichit varies soastomaintain aconstant potential difference eitherat theterminals ofthe alternator,oratanydistantpointofthe ‘mains. 630 ALTERNATE CURRENT WORKING, (May04, - TTPiaItcomprises arelay(Fig.10):MING JS)|)a0))) |)and«regulator(Fig.11).Toe |Mi 7//)|4]formeristhebrain,thelatterHh HAIthemuscle,Therelay(Fig.10)HALA Hil\\|]}isacoreandsolenoidarrange AN{J)))|\|)mentverysimplymadetobe MV}E=] i{\assensitiveaspossible. Every {AMC ET||seroandaracoteeaec nll{ and, wequence, HB \\\)}_itwillworkwitharangeofonePLE) BMIi]attpercent, ANAHlH\\agToacemotthe relaymate FTE TRIG] to:moveacontact-making arm WiHi] H\\\\ which,bytwomercury contacts,Va) |Milipy\\\||controls thecircuitsoftwovan! LUNE) \\||coilsontheregulator proper 14/0 { “IJ||||Thisregulator isillustrated by WA| 2)|)Fig.11.Theconstruction i11///eebag|\\|ffversite argscopper ..\\\\[|orbrassvessel,filledwithoilor[AABRusuElecrmicionr] |\)|Water,containsafloatmadeof|)wosoersitoufSxcusry\\||thinsheet-iron. Thisiron 1)GTTTETTIGT|\\ |)vesselactsbothasafloatandSE \'||)asthemovablecoreof#A |||)])solenoid. Theoutervesselis WI —])})|||)) providedwithtwocoilsofwie Wd1))||)controtiedbytherelayand HTH —-}|\/))/}) energised bycurrentfromthe |<=}|/)//))exciter.Belowisliquidresisti —||ancecomprising twofixedlesd|)MESSE ||)platesindilutesutpharieeid |)DOT TTT|||)orothersolution, Athirdlad5151/2) |)[)plate,bentdownattheendsto(QUA 1)approach thefixedplates,it‘vei ea) ))])supported bythefloat,andby a+ itsmovement varies theliquid Fic11.—Regnlator, resistance. _ oe _—e 1990) SUPPLEMENTARY REMARKS, esr ‘Thearrangement, besidesbeingsimpleandunlikelytoget outoforder,isverysensitive, thefloat,providing analmostfric- tionless method ofsupport, without any troublesome mechanism, LAiiiy,.)ie va“Eas‘(fl 4 a,t wy } [ 1 van Wh AWW | ae ee Maa | FFi i a sal TESA U Fo,12-—Mordey‘Transformer, ‘Thefloatisbalanced toremaininanyposition,fromwhichitis onlymovedwhenanimpulseisreceivedfromtherelay. TIsaidthateverycarewastakentopreventreactioninthe relay. The reverse isdone with theregulator. Byallowing VOL. XVII, 44 s00gl 638 ALTERNATE CURRENT WORKING. Dlaysn, considerable inductive reaction inconnection with theregulstor coils, ithasbeen possible notonly tostop themotionofthe floating core very quickly, but actually toenable ittomakes slightbackward movement. onthecessation oftheimpulsefro therelay. Inthis way anytendency to“hunt” may bechecked Itshould beexplained that thefigures (10and11)arenottothe same scale, The former ismuch less reduced than the latter. Some transformers, impedance coils, and asynchroniset areonthe table. Ineed not pause todescribe them further. The transformers (Fig. 12) have already been explained at« previous meeting. (Journal, vol.xvii. p.115.) ThavesaidthatIdidnotapprove ofironinthearmatures of alternators, butIwish, nevertheless, tobeallowed todescribe 4 machine with iron, not in,but outside ofitsarmature, Ihare called ita“single-coil alternator.” Itisnotamachine thatI have made, butitisinsome respects anoutcome ofthemachine already shown toyou. Isimply describe itasshowing onewayof making analternator, Inthefirst machine (Figs. 5to9)there isonly onefield coilforthewhole excitation, instead ofthelarge numberofcoilsusuallyemployed. Thisprinciple hasbeencarried farther inthedesign shown inFigs. 13and14,andinthemodel before you, which illustrate amachine inwhich there isonly one simplestationary fieldcoil,andonesimplestationary armature coil, foranyspeed orperiodicity. Thefigures andmodel probably explain themselves. The field winding surrounds acast-in rotating cylinder, having gaps ornotches attheends 0astoallow thelines offorce topass alternately round thefield winding and round thefield and armature windings. Inthis way, with every part fixed except therotating cylinder, analternating EMF. isobtained inthearmature coil. [Model skown.] Fig. 13istaken from asetofdrawings made about eighteen months ago, and illustrating several ways ofcarrying outthe principle. Aisthe armaturecoil,Fthefieldcoil.Thefield-maguet poles areshown atmm, #e. The laminated iron masses, which carry thelinesofforeealternatelyoutaideandinsidethearmature coil,aremarkedITandKK. 18) SUPPLEMENTARY REMARKE, x ‘Imustmention,asoneofthecoincidencesthatsooftenoccur inthesematters, thatafewweeksago,inthediscussion onMr. Single Coil Alternator. Mordey. na Ee el Fie. 18, Kapp’spaperattheInstitution ofCivilEngineers, Professor Forbesdescribedanalmostidenticalmachine—amachinesimilar innearlyeveryrespecttoFig.13. ¢ 4 m_USUr iSi\ _an wig é i Section thro’ C.D. Sectionthro'AB. Pro.u, Ishould prefer thelaterformseeninFig.14,forthe‘reasons thatthefieldcoilisstationary aswellasthearmature, 10 AMTERNATE CURRENTWORKING, as, and that itisvery easy toremove the field-magnet without disturbing anyother part ofthemachine. Some ofthequite ofthis type ofmachine arefurther alluded tointheAppends (page 629). Since the last meeting, Mr.Raworth has devised «rey interesting and practical method ofexperimentally ascertsting theshape ofthe E.MLF, curves ofalternators, This enable© todecide actually byexperiment, notonly theshape, butthe relation between the curves, both ofEMF. and current; andl investigating these matters inaway that cannot failtoproduct very useful results. Ihope that during thediscussionhewi explain this method fally. Thave received agood many letters during thelastfewdays. and, asthey form part ofthediscussion, Imay beallowed to refer toone ortwo ofthem. Lord Rayleigh kindly sends meacopy ofanarticle inthe Philosophical Magazine forMay,1886,inwhichhedealswith thequestion ofvirtual resistance ofstraight conductors wihalternating currents,andarrivesatconclusions whicharepo=ticallythosegiveninthepaper.Myfigures,asIstated,wer worked outfrom thedata given bySirWilliam Thomsonii Presidential Address, and from some further figures that hems good enough togive meafterwards. Lord Rayleigh,inhi letter, gives aformula forvirtual resistance which isallthat wanted asaguide topractice. Forcopper itis 12at wer {i+io}, where aistheradius incentimetres, andrthecomplete period. ‘Thearticle seems tohave been thefirst recognition ofthe fact that the increase ofresistance due tonon-uniform distr bution ofcurrent over the section ofaconductorwasaneft thatmight beserious inpractice, Thefollowing extracts aed! interest asbearing onmore than onematter ofimportance attt present time :— Referring tothe mathematical examination :—“Toor“resultsaremerelyveryspecialcasesofagenerallay,foe“whichwemayleamthatasthefrequencyofalternative| 1880) SUPPLEMENTARY REMARKS. on «gradually increases from zero toinfinity, there isasteady rise “ofresistance and accompanying fallofself-induction, The “application ofthe general idea tothepresent case isvery “simple, Atslow rates ofalternation, the distribution of “current, being such astomake theresistance aminimum, is “uniform over the section; and this distribution, since it “involvesmagnetisation oftheouterpartsofthecylinder, leads “toconsiderable self-induction, especially iniron, Ontheother “hand,wheretherateofalternation isveryrapid,theendeavour “istomake theself.induction aminimum irrespective ofresist- “ance. This object isattained byconcentration ofthecurrent “into theouter layers. The magnetisation oftheconductor is “thus more and more avoided, but, ofcourse, attheexpense of “imereased resistance. Wemay gather from thegeneral argu- “ment ...that aspincreaseswithoutlimit,R’alsobe- “comes infinite, while the part ofL’depending upon the magnetisation oftheconductor tends tozero. “The increase ofresistance proper (not merely ofthe “«throttling’ due tothecombined effect ofresistance and self “induction) iniron wires ofmoderate diameter subjected to “varying currents isone ofthe most striking ofProfessor “Hughes's results, SofarasIamaware, neither Maxwell nor “any other theorist hadanticipated that thealteration ofresist- “ance would beimportant under such circumstances.” Lord Rayleigh objects, asamatter ofnomenclature, tomyuse ofthe word “periodicity,” and prefers theword “frequency.” Heremarks that “the great point istoinsist that vibration shall “always mean complete vibration.” ‘That wasprecisely myreason forsuggesting aword that was notlikely tolead toconfusion. “Vibration,” “frequency,” “reversal,” “alternation ”—the words ingeneral use—all earryanuncertain sound.‘Theymaymeanone ‘ortwo currents, and I’venture todisagree with Lord Rayleigh when hethinks itbetter touseaword which requires ustoinsist that itshall mean something which itdoes notnecessarily mean iflefttoitself.Asaninstanceofthelengthyexplanation whichIwish toavoid, Imay refer toLord Rayleigh’s paper, from which Ihave already quoted. Hethere says, “From analternate- oe ALTERNATE CURRENT WORKING, [May®t, “current machinewemayhavecurrentsofperiod“O1second(100 «positive and 100negative pulses persecond).” Butifmyuseoftheword “periodicity” isnotallowablelemeput inaplea forthe sign “Ly asrendering confusion impossible. Professor Ewing writes from Dundee criticising mypape. Hesays :—“ Iamsorry Icannot bepresent tohear itdiscussed; “ifIwere,therearejusttworemarksIshouldliketomale “about your plan ofdistinguishing between loss ofenergy by “magnetic hysteresis and loss byeddies inthe armatures of “dynamos. Yourplanis,ofcourse,basedonourtakingthels “byhysteresis (under constant excitation ofthefield) asproper «tional tothespeed, and thelossbyeddies asproportionalto “thesquareofthespeed.Butwecannotbesurethatthels «byhysteresis isproportional tothespeed, especially when the «speedishigh.Theprocessofassuming magnetisation, onthe «part ofsoft iron, takes some time, and wehave little or2 “information atpresent astohow farthisfactaffects thelosof “energy bymagnetic hysteresis inrapidly performed cycles. “Again, astothe eddy current, you say, ‘The resistance ¢(that oftheiron) being constant, theeddy currents must raz} «‘asthefirst power ofthespeed, andtheloss therefrom asthe “<square ofthespeed.’ But willnotacause which youreferto “further on—viz,, the impedance oftheeddy circuits—opemte “even atlowspeeds tomake thestrength ofthe eddies increse “Jess rapidly than thespeed, and thelossofenergy through «them, consequently, toincrease lessrapidly than thesquare of «thespeed? “The general effect ofthis will betogive themagnetic “hysteresis, asestimated byyour method, alower value thani “ought tohave. “TImention these points only bywayofshowing thatthe «method isopen tosome little uncertainty. ‘Thefactthatinthe “example you have taken thehysteresis hasacalculated valoe «just such asonewould expect ittohave nodoubt goes agoat “way toshow that themethod isinthemain asound one.” Now, Ithink, these objections raised byProfessor Ewing 1880 Discussion. oun “really dealt withinthepaper. Astothefirstpoint itmaybe taken that foranyrange ofspeed, solong astheE.M.F, isfound tobeproportional tospeed, theassumption ofmagnetisation must beconsidered asbeing thesame throughout (seepage 624). ‘Astothe second point, that the impedance oftheeddy cir- cuits willoperate even atlowspeeds tomake thestrength ofthe eddies increage less rapidly than the speed, itwill befound, on referring toFig. 3,andtotheexplanation given, that theincrease isonly proportional upto30“~L per second (see page 625). Below thisthelawonwhich themethod isbased apparently holds good. Athigher speeds there isafalling offshown bythediffer- ence between thefulllineONandthedotted linespringing from it. Iquite agree with Professor Ewing that these points show the method tobeopen tosome uncertainty, but anexamination of theresults obtained experimentally show clearly what limitations itisnecessary toimpose upon theapplication ofthemethod, and Within those limitations Iventure tosubmit that the method is quite applicable. Ishould mention that thecurves (Fig. 3)were from many careful andconcordant tests. Mr. Zipemowski writes that heintends taking part inthe discussion inwriting. Hisexperience, and that ofhisfirm, has been very large, andIhope that weshall have hiscontribution intheJournal, with thepaper. Dr.JounHorkixsox [readinhiswnavoidable absenceandonBe. hisbehalf byProfessor George Forbes]: Agreat deal might be said onmany ofthepoints which areably discussed inMr. Mordey’s valuable paper. Ishall confine myselftobriefremarks ‘onone ofthem. Some misapprehension may exist astotheconclusions which legitimately follow from thetheory ofrunning alternate-current generators parallel, which Igave fiveorsixyears ago, Toobtain 4greatcontrolofonemachineuponanotheritisnotofitselfdesimble tohave anylarge self-induction, asMessrs. Kapp and Forbes appear tothink, norisitdesirable tohave itassmall as Possible, asMr.Mordey appears tothink, when hesays, “Ifit(self= “induetion) were absent probably themachine would runparallel | ou ALTERNATE CURRENT WORKING. [ay, vin,“AllTight.”Themachineswillbestcontroleachotherwh,275%,1beingtheselfinduction, isequaltotheresistanceofthe armature circuit and the leads tothejunction with theleis of the other inachine. ‘This isan obvious immediate oon quence oftheequation ofparagraph 4ofmypaperreadbela this Institution inNovember, 1884, IfMr. Mordey adds aso} stantial resistance tothearmature-circuit ofeach machine, sc forexample, astheresistance of feeder, hewill find itlessexy torun parallel, He will then find the difficulty somemitdiminished byaddingalittleself-induction toeacharmatar- circuit. Inparagraph 3ofthesame paper Iconsidered connecting amachine toacircuit ofdifferent potential, and Ipointed a the limit within which such amachine would work asamot. Letusconsider theclosely analogous problem ofworking two machines ofwidely different fields connected togther asMr.Mordeyhasdone,Letthemachines beidentical,exceptingtht one shall have double the field ofthe other, and letthem be simply connected together with noexternal work, theequation of currents, modifying those ofparagraph 1ofmypaper tosuitthe case, may bewritten— Qyet2rea2Esin 24Eoin,AZ) os froin 2—2oon2} ry, —F T e+e) E in2t—7)_2ey.. 2e(t—2) T work done bymachine ofgreater electromotive force is generating electricity __ Fr Ve Qer, Qey sot =Seaytlaray romqt+tpttg} e+e) 9+GP) Nowsuppose,forthesakeofexample,that Bryaor, 18983 DISCUSSION. os vwork done be aEELMon2Hbein227} Honesa aeaeaa and this will bezero if Qari T "2 There is,therefore, nothing astonishing intwo such machines running together;itisjustwhatonewouldexpectifthe resistance isnottoogreat incomparison with the self-induction. ‘Thepotential difference asmeasured between thetwomachines =Eoin.7!—4Eoin,227), anditsmean square _ -3a +E Itthus seems that my old theory issufficient topredict allthe results obtained byMr. Mordey inregard torunning alternators inparallel. Itsimperfection isthat theco-efficient ofselfinduction isassumed constant, which is,infact, not exactly true, Totreat itasvariable renders the equations ‘unmanageable. Professor W.GryLis Apams [read inhiswnavoidable absence — andonhisbehalf byProfessor S.P.Thompson]: The principle oftheparallelworkingofalternate-current dynamomachines was fally established byDr.John Hopkinson in1883, and itwaswith theviewoftestingorproving thetruthofthisprinciple thatin July, 1884, thefirstactual successful trials inthisdirection were made bymeattheSouth Foreland Lighthouse ontherunning of theDeMeritens alternate-current magneto machines inparallel. Dr.Hopkinson wasnotpresent atthese first experiments, but onreference tomypaper, read November 13th, 1884, itwill be seen that Ithen and there established the fact that alternate- current machines worked inharmony without being rigidly con- nected, and that they mutually acted andreacted on,orgoverned, oneanother, thus giving asteadier result than when thetwo machines were both driven yoked together. Also itwillbefound that Ithen andthere established the fact stated byMr. Mordey inhis paper, that the question ofalternate-current parallel os ALTERNATE CURRENT WORKING. (May 0, Rrtewor working verylargely depends upon thequestion ofsynchronising alternate-current motors. Forassoon asIhadsucceeded ingetting thetwomachines to work inharmony when driven independently, Iatonce threw the lamp outofeireuit, and threw the belt offone ofthemachines, when itcontinued torunatthesame uniform speed, being driven ‘asa motor bytheelectric current from the other machine. In fact, thestatement that theworking ofalternate-current machines inparallel very largely depends upon their synchronisingasmotors isvery similar tothestatement that tuning forks synchronise when they areinunison. Ofcourse they'do. Onthesame occasion, inJuly, 1884,Imadetheveryinteres: ingexperiment ofloading thedriving machine with anareinthe external circuits, when beats were produced inthe arefrom the twomachines getting outofharmony ;justasbeats aregivenby twotuning forks when oneisloaded. ‘These beats, produced bytwomachines working together,ae beautifully shown inMr. Mordey’s experiments, both when the machines arenearly inunison, and also when one isrunning nearly twice asfastastheother—let merather saywhen oneis giving nearly theoctave oftheother. Ihave been very much interested inMr. Mordey’s experi ments, which Ihave seen since ourlastmeeting, andhavenoted one oFtwo facts towhich Iwould draw attention. When one machine was running atfull speed, andthe speed oftheother was gradually rising, thelamps attached tothe two secondaries were first fitful, then gave beats « first rapid and then slower, until the two were inharmony; the speeds were now intheratio of2to1,orone wasthe ‘octave oftheother; then, asthespeed increased, thebeats agait ‘became rapid andthen irregular, until thespeed ofthesecond approached nearly tothefirst, when thegetting into unism ‘wasaccompanied bybeats regular andrapid, then slower «0! slower, until they blended, justasinthetuning oftwomusical instruments. Whether Mr,Mordey hasmade theexperiment, successfully! know not, butIseenoreason why heshould notsucceed ia 60.) DISCUSSION. or running twomachines together, whenoneisrunning twoorPrtewoc three, oreven four times asfast asthe other, provided they harmonise exactly. Mr.Mordey hasgiven ustheoctave when hedrives oneat1,000 andtheother at2,000 volts, but could he notalso give usthe12th, oreven thedouble octave, andcombine together 666, oreven 500, with 2,000 volts, giving respectively anEMF. of1,330 orof1,250 volts ? Onpage 692Mr. Mordey has stated the results ofeight separate and distinct experiments. Five ofthese experiments, viz, those numbered 1,2,3,6,and 8were made inthe experi- ments atthe South Foreland in1884; also the second machine, whendrivenasamotor,wasmadetodoworkonafrictionbrake, andgave offmore than 4H.P. Each ofthese machines absorbs 4H.P. when running atfullspeed with thecircuit open. Inanother experiment attheSouth Foreland three machines were driven parallel together until they synchronised, and then the belts were thrown ontoloose pulleys from two ofthe machines, and they were both driven asmotors bythecurrent from thethird machine, and the E.M.F. atthe terminals ofthe generating machine remained the same aswhen the three were allbeing driven together inparallel circuit. Inanother experiment three machines were driven parallel with anarc inthe external circuit;thebeltwasthrownoffone ofthemachines, which continued torun asamotor, and the are became steadier when the third machine was being driven asa motor than when the three machines were alldriven directly from theengine. Also, thework done orlight given inthearc with two machines inparallel was considerably less than the work done when thetwo machines were supplying thearcandat thesame time driving thethird machine asmotor. ‘The illumination given bythearewas— (1) 13,500 candles with twomachines alone parallel. (2)16,000 candles with three machines allbeing driven parallel. (8)17,300 candles with twomachines driving thethird as ‘amotor. Itmay well besaid ofthese experiments in1884, inthe os ALTERNATE CURRENT WORKING. [ay200, Reloger words ofMr. Mordey, “These experiments show what perfect “self-governors such motors are: not only dothey maintain “gynchronism, but they possess aninherent economy which “ig most valuable. Just enough current passes through them “to keep them instep, and todothework imposed onthem. “They become generators, and dowork onthecircuit iffrom “any cause there isatendency forthem torun faster thin “the generator.” When weenquire the cause ofthissyn- chronising, towhat canweattribute itbuttotheself-induction ofthe circuit? Indeed, Mr. Mordey’s experiment, inwhich he puts two machines onthe same circuit when they arenotin step, butrunning atthesame rate, shows that theself-induction issufficient tobring them into harmony inanexceedingly shor time. This sudden coming into step isnodoubt due tothe factthat there isnotmuch iron inthearmature; butthestrong quiver and groaning ofthe motors accompanying thegetting into step isvery sufficient evidence ofvery considerable self induction, inconsequence ofwhich thespeed ofone machine issuddenlyincreasedandoftheotherisassuddenlydiminished. take, ‘MajorP.Canvew,R.E.:Mr.Mordey,onpage607,referstothe question ofthe periodicity atwhich the temperature visibly follows thevariation ofcurrent. Agood many years ago,when theSwan lamp firstcame out, wetried them forsignalling from balloons bymaking the lamps flash, andwevery soon, ofcourse, found that ifthecurrent were simply taken onand offthelamp thesignalling wasvery slow indeed. Soinorder toincrease the rapidity ofsignalling bythekey, which was operated from belo, wemerely altered thecurrent bycausing thekeytoshort-cireut certain amount ofresistance, altering thevolts onthelamp about 20percent.; they were 50-volt lamps, andwekept them burning at40volts through resistance; andwhen weputtheker down thevolts were increased upto50,bycutting outtheresist- ance. The difference intheflash wasvery visible, butnoto> visible tobequite clear at distance, andwefound wecould get about20wordsaminuteonthekey.Ihaveworkedthatout,and itissomething likewhat Mr.Mordey writes thus ~’,period- city ofabout 10per second. That ismuch below what mot 1830 DISCUSSION. a Alternators arerunat.Idonotthink wecould evergetitreally Malgr, visible and distinct over 20words aminute. Onpage 616hegives four possible contacts, a,b,¢,d,andsays ifall these occurred atthesame time aperson would getashock, That isasmuch astosaythat there istheextreme improbability ofthefourthingshappening simultaneously. Thatisnotquite thecorrect way toput it,because ofthese four things thecon- tactbetween theprimary mains and earth isathing which is always unavoidable, atany rate when circuits have been upsome time; unless thegreatest care istaken ofthem there ispretty suretobesomesortofcontactbetweentheprimarymainsandarth,andthecontactbetweentheprimaryandsecondaryina transformer isathing which may happen and nobody know any- thing about it;itmaybeonformonths.Sothatthosetwocon- tacts, aand8,arewaiting ready for@person tomake thethird andfourth himself, and ofcourse, even then, Iquite grant that it mayhappen that theperson touches, say,thesecondary conductor when standing onadrycarpet. and does notmake earth, and so might notreceive ashock;butifhedoesnotreceiveashockhe doesnotknow anything about thisfault inthetransformer, and hegoes ontouching hissecondary circuit with confidence until some occasion when hedoes make earth byperhaps touching a gas-pipe atthesame time; that isthechance ofcourse, anditis obvious that, considering theresult ofthishappening, itisworth while taking anyprecaution which isjustifiable. Mr. Mordey doesnotseeanyobjection totheplan ofearthing thesecondary. Well, asregards safety tolifefrom shock, Idonotsuppose any- body does: itwas the very first thing tooccur, Isuppose, to everybody that thedanger tolifewas gotover byearthing the secondary. Butwhen you come toconsider thefirerisks,Iam afraid that wecannot sayitissosatisfactory. ‘There isnodoubt thatputting aconnection toearth ona100-volt cirenit, which is ‘whatitcomesto,becausethissecondary circuithasgenerally 100volts,doesincreasethefirerisk,especially wherethewiresarelaidunderfloors,andperhapsinclosecontactwithmetalpipes.Nobody really recommends earthing a100-volt circuit asaregular thing. 50 ALTERNATE CURRENT WORKING, [Alay9, Mate Mr.W.M.Monpey: Ido. ‘Major Canpew, RLE.: Ibegpardon. Mr. Mordey isperfectly consistent ifheearths a100-volt circuit and does not consider that that increases thedanger orgives anydanger offire;he hasgottopersuadethefirecompaniestothateffect,andthen itwill beallright. Asfarasthe direct. shock goes, everybody issatisfied ifthecircuit isearthed. Talking ofthat, there is ageneral impression, andIhavealwaysunderstood, thatthe ‘Westinghouse Company did earth the secondary, bat their representative, who has specially come over here, distinctly denies that they do. Ishould like toknow whether any American company using transformers doearth thesecondaries ornot. Pry Dr.J.A.Fuemina: Some points which Ihadintended to deal with inMr.Mordey’s interesting paper have already been touched upon intheremarks ofprevious speakers, and nothing farther need besaid bymeupon them, but there areoneo two asyetundealt with. Atthe end ofhispaper Mr.Monier refers tothe question ofthe life oflamps used with alternate ordirect currents. We have, ofcourse, the broad experience ofEurope andAmerica onthis point. There aresome instances, however, ofthe systems running side byside, asatMilan. ‘When atMilan lately,IputthequestiontoMr.Liebwhether hethought there was any difference inthe lamp life ofthe theatres which were lighted byalternate and bydirect currents, ‘Two theatres arelitbyZipernowski machines atadistance of 1,800 metres from the station. Then there areother theatres which arenot very faraway from these, which areonthe direct. current, circuits, and inallthese theatres thelamp life isabout thesame, The regulation ofthealternating current machines isthere carried ont very carefully, forthe particular reason that the electric lighting company has toreplace all the lamps inwhich the filaments are broken bythe current; thecustomers replace lamps ifthey aremechanically broken. Then, coming toaprevious point inMr. Mordey’s paper with reference tothecurve which hehasdrawn, showing the relative amount ofwaste energy inthetransformer atdifferent 1803 DISCUSSION. oot periodicities, Ishouldliketothrowoutasuggestion forhisPras. consideration. IfIunderstand those curves aright, they meanthis,thatwithaperiodicity of100persecond,thetotalheatinthetransformer islessthanat126,andlessalsothanat75.‘Thesuggestion that Iwould make isthis, that this isdetermined bytherelative waste ofenergy produced bytheeddy currents andthat produced bytheviscous hysteresis. Wehave reason tobelieve that the waste due tohysteresis increases with the speed, andtherefore atahigh speed onewould expect increased heating due tothis cause. On the other hand, atlow frequencies theeddy currents arebetter able tofindtheir way farther into theiron and towaste more energy. Hence, ata ‘certain frequency there must beaminimum ofheating inthe transformer core. Returning tothemain subject ofinterest inMr.Mordey’s Paper, the working ofthese alternating-current machines in parallel, Ithink wemust beindebted toMr.Mordey forcallingarattention afreshtothefactthatsuccessful workinginparallelispossible. Inorder tosecureapracticalworkingofthemachines inparallel, itisnotmerely necessary that they should bring one mother into phase when one machine lags behind theother, butthat there should beconsiderable stability when instep. The force ofrestoration when disturbance ofunison occurs must begreat,Whatisreallyrequiredis,thatassoonasonemachine lagsslightly behind theother, theforces which arebrought upon thatmachine tobring itinto phase again must actstrongly and quickly. The actions which must take place inorder tosecure this must beofsuch anature that the armature ofthe machine Which islagging behind ispushed uppromptly again into its Position, That is,ofcourse, putting thematter very roughly, butitreally isdetermined bytheform oftheE.M.F. curve ofthe machine, Tdonotdesire toindulge inminute criticisms onapaper which issovery interesting asthisoneis,butIdonotliketosee Mr.Mordey usetheword‘ magnetisation”attheendofhispaper forwhatisgenerally spokenofas“induction.” ‘Thewordmag-netisation hasadefinite meaning itisused fortheintensity of sa ALTERNATE CURRENT WORKING. (ayn, Plning™&gnetisation, anditseemsundesirable totakefromthewordtheperfectly definite signification which itnowhas,touseitfor what isgenerally known asthe induction. Itisperfectlytrue that the word induction isavery hard-worked word, butitis, notwithstanding, generally well understood. weKay. Mir.Gispent Karp: Itiscustomary forthevarious speaken tocompliment theauthor onthe merits ofhispaper, and some- times their doing soisrather aformality. Inthepresent in- stance, however, Iamquite sure you will nottake ittobea mere formality when Iexpress high appreciation ofthisvery interesting paper. Itrust you will agree with methat hispaperisoneof the best contributions our Society has ever had. The experi- mental facts brought before usareofgreat value, andinesi mating thisvalue wemust notallow ourselves tobeinfluenced bytheauthor's attempt tousethese facts inupsetting awell established theory. Inthis hehasfailed, buthisfacta neverthe lessremainallright,andarecertainly ofgreatpracticalimportance. Inthebeginning ofhispaper hegives aquotation taken froma recent paper ofmine where Isaythat self-induction isnecessary toparallel working. He,onthecontrary, says thattheabsenceof self-induction andtheabsence ofresistance istheessential quality inanalternator, inorder that itmay synchronise and work as+ motor; and his reasoning isapparently sound. He shows+ characteristic ofone ofhisalternators, from which itappears that self-induction only absorbs 24percent. ofthetotal EMF. and this alternator isthen put towork asamotor, andthe experiment succeeds perfectly. From this experiment heoot cludes that amachine without self-induction would bethe bet possible motor. Hewaskind enough toshow onFriday last several ofhisfriends theexperiments quoted inthepaper, andall sueceeded perfectly. Amongst theseexperiments wasonewhichcould notpossibly beperformed ifthemachines were devoidof self-induction orsome equivalent property. Irefer tothe ‘coupling inparallel oftwomachines running atthesame speed, butdifferently excited, onetogive2,000,andtheotheronly1,000, voltsterminal pressure, Whenthesetwomachines wereputit parallel thedrop inpressure ofthe2,000-volt machine was500 1188) DIBOURBION, oss rolts, andtheriseinpressure ofthe1,000-volt machine was500MrKerr. tolts, the pressure between the coupling upleads and inthe external circuit generally being 1,500 volts. Now, itisquite evident that inthiscase thecharacteristic ofthegenerator can- tothave been thevery flatcurve shown intheauthor's diagram inwhich the small drop ofpressure (about 2}per cent.) is supposed tobedue tovery small self-induction, The fall in pressure wassomething like25percent., andthiscannot possibly bedue toohmic resistance, since the latter was small, and the author does not suggest that any very large current passed throughthetwomachines, Wearetherefore facetofacewiththisvery curious andapparently inexplicable experiment :hereis 4machine which issupposed tohave very little self-induetion, and,indeed, gives analmost constant terminal pressure when wed with aninduetionless resistance inthe external circuit, but 15soon asweusethat machine asagenerator inconnection with 4similar but weaker machine working asmotor, theterminal pressure isbynomeans constant, and there isaconsiderable drop. Now this drop cannot bedue toanything else butself- induction, orsome property equivalent toself-induetion, and I vasatfirst greatly puzzled how toaccount forthe different behaviour ofthemachine inthetwocases. AsMr.Mordey, in hispaper, has merely recorded the experiment, but bas not attempted toexplain it,Ishall, with your permission, make this attempt. Atthecutset Imust confess tohaving been inerror when Iassumed that thearmature ofthis machine hasonly an inappreciable self-induction, and Imust thank Mr. Mordey for having putthequotation referring tothis matter into hispapers andthus shown methat Iwaswrong. Tamafraidthatallofuswhohaveworkedwithalternating- current machinery have gone wrong, more orless,on thisquestion * ofselfinduction. Inthefirstplacewehavebeeninthehabitof considering thearmature asathing byitself, without taking account oftheinfluence ofthefield magnets, and, secondly, we idnotdistinguish between what isproperly called self-induction ‘andanequivalent property, whichmightbetterbedescribed asmagnetic change, This armature, asitstands onthe floor You. XVII, 45 4 ALTERNATE CURRENT WORKING. —filaym4, xrxapp. removed from itsfield magnets, has probably very little self induction. Itmust, ofcourse, have some self-induction. To expect that adynamo machine should beabsolutely without self-induction would beasunreasonable astoexpect that asteam engine, turbine, orother mechanical apparatus should have 20 inertia, Now, although thearmature may have only little self induction when byitself, when itisput into itsfield, thesel induction mustbeconsiderably increased, ascanbeeasilyseen byreference totledrawings ofthemachine, Now ifyoulook atthewall diagram, showing asection ofthe machine taken parallel tothe shaft, and consider only thecal which isshown between thetwofield poles, you canregard itas 1short solenoid, thetwo field-pole pieces forming itsiron eae Itisobviousthatinthispositionthecoilmusthavegreatself induction. The coilnext toit,which liesbetween neighbouring poles, has aminimum ofself-induetion, since there isnoina circuit through it.Thethird coilwillhave again amaximumof self-induction,thefourthcoilaminimum,andsoon.Halfthe totalnumberofcoilshavetherefore verygreat,andtheotherhalf very small, self-induction, Now consider theposition ofthecoll ‘aquarter period later. One side ofeach individual coil isnow between thepolar faces, theother side isbeyond them. Each coil cannow beconsidered asasolenoid only partially supplied with aniron core, and itsself-induction must. therefore beless than itwaswhen coinciding with thepole faces. Since, however, allthearmature coils (aud notonly one half their number)are now provided with iron cores, the self-induction ofthe armature taken asawhole must still beconsiderable. Aslong asthe machine isworking upon aninductionless resistance, thisself induction (which, although subjected toperiodic variation, cam never bezero) istheprincipal cause ofthe2}percent. dropit terminal pressure, asrecorded inthe paper. But when the machine supplies acircuit ofsensible self-induction, another cause tending tolower theterminal pressure comes intoplay. It iswhat Ihave before called magnetic change. When thereis littlelagofcurrentbehindinducedE.M.F.,themaximum carestthrough thearmature occurs verynearly atthetime when eh coilisonly half covered bypolar faces, Inthis position the 180 DISCUSSION, oss current through alternate coils tends tomagnetise thefield poles Mr.Kem more strongly, andthat through theintermediate coils tends to demagnetise them, thetwoeffects very nearly eliminating each other. If,however, there is@considerable lagofcurrent behind induced E.MLF., themaximum current occurs when each alternate coilismore nearly covered bythe polar faces, and the current through these coils tends todemagnetise thefield, that through theintermediate coils having, byvirtue oftheir position atthe time, noeffect, oronly avery small effect, instrengthening the field, Inthemachine which acts asamotor theopposite takes place; thecurrent inthecoils which face thepoles tends to strengthen their magnetisation, whilst theintermediate coils have verylittle effect, ornoeffect, inweakening them. Wehave thus superimposed upon theeffect due towhat may properly becalled selfinduction acertain effect, which isdue tomagnetic change, andwhich only comes into play ifthere isconsiderable lag. The field ofthe generating machine isweakened bythearmature current, and that ofthe receiving ormotor machine is strengthened bythe current through itsarmature, and inthis manner itispossible that theoriginal terminal pressure ofthe generator may drop by500 volts, and that ofthe motor may risebyanequal amount.‘Mr.J.Swivauaxe: Ishould liketoaddtoMr.Kapp's35... compliment toMr.Mordey’s paper, ifsuch athing were possible. Itiseminently apaper byanelectrical engineer forelectrical engineers, and istherefore the sort ofpaper wanted here. For mypart, however, Idisagree entirely with Mr. Mordey’s theory, that hismotor kept instep because ithad noself-induction. If there were noself-induction, thegenerator would have nocontrol over the motor atall. Itseems tomethat electrical engineers donot.altogether realise what iswanted inparallel running. That alternating machines tend tokeep instep isalready perfectly well known, but tomake them run parallel commercially needs large margin corrective tendency. This canonly begot bymaking themachines sothat themaximum current isnot at the sume time as the maximum EMF. If the machines are working onresistance, this means merely a | 686 ALTERNATE CURRENT WORKING. [May3, ¥Scvume, Smaller output;butifonemachine isworking anothera‘amotor, orsimply controlling itasalagging generator, the lossbyresistance inthe armatures isexcessive inproportiono thepower given out. This means that the machines must be made larger togive agiven output. They must also besome what less efficient, Mr. Mordey’s attack onthe existing theories ofalternsting current machines seems unwarranted. He makes anumberof experiments, andgets results which areperfectly inaccord with existing theories; hethen asserts that the existing theoriesae ‘wrong, without any experimental evidence, and gives notheory ofhisown totake their place. That hisown machines bare considerable self-induetion isobvious from hisexperiment with the 2,000 and 1,000 volt machines. Apart from thequestion whether self-induction kept the machines instep, itisobvious that there would have been anenormous current: through bulb armatures, ifthere were noself-induction. Idonotmean inthe east toundervalue Mr. Mordey’s experiments;thedegreeto which alternators cancontrol oneanother wasnotfully realised tillthese results were published. Indiscussing motorsIdonotwanttorepeatanything already said, andwilltherefore cutmyintended remarks down. Running amotor from asingle dynamo isnotafairtest, asinpractice motors willberunfrom supply circuits. Passing tothequestion offrequencies, orvibration frequencies, ‘Mr. Mordey’s comparisons ofdynamos seem inaccurate ;anditi very difficult todiscuss these things, owing totheloose way “self-induetion” isgenerally used byelectrical engineers. Generally speaking, ifyoumake anarmature ofagiven size,and allow agiven waste ofpower initwhich admits ofcooling by ventilation, alarger output canbegotwith asmall frequency: Itherefore prefer @very small frequency, and Ithink ‘Mr.Crompton willhere agree with me. Astotransformers. Incomparing converters under different loads and different circumstances people areoften apttobe misled bytaking thesame transformer ineach ease, Thebest transformer foreach particular use should betaken togive results ofany value. Ihave recently made alarge number 1980) DIBOUBBION. esr oftransformer calculations, making tables fordifferent circum- Syistome stances, working outthe most efficient proportion ineach case, assuming aniron induction and acopper current density which willadmit ofpractical working. Ifind, forinstance, that 450-light transformer canbemade togive 92percent. efficiency, ‘at33vibrations persecond atfullload. At83vibrations per second a50-light transformer cangive 93percent. ‘The saving incostofmaterial isfifteen shillings, orabout fourpence alight, which isinappreciable, Ahouse transformer may betaken as running atfullload foranhour aday. Such a50-light trans- formercanbemadetogiveanaverage percentage of60atthe lowfrequency, and62atthehigh. Both these lasttransformers aeprohibitive inprice, soalower efficiency would have tobe tolerated. Inthe case ofatwo-hour-a-day transformer the eficiencies arehigher, being 70°5, and alittle more with the higher frequeney. ‘Astolamination, Ithink Mr. Mordey over estimates the effect ofeddy currents. Hisexperiments donotreally distinguish between eddy currents inthediscs andvariations ofmechanical, orofairfriction, and eddy currents inthewire and inthehub ofthearmature orelsewhere. Irecently calculated thelossby eddy currents inalaminated direct-current core, buttheresults areinapaper notyetpublished ;Imay mention, however, that theyareexceedingly small,infactgenerally inappreciable. Mr.Mordey’s method seems exceedingly ingenious, buttheresults cannot begotwith any accuracy, assomany unwarrantable assumptions have tobemade. There areseveral methods ofstarting alternate current motors. Imayperhaps beallowed tomention thatstarting by secondary batteries was mentioned inapatent specification of nine in1887. ‘Theexciter may have itsfield magnets laminated, andthealternating-current canbeturned ontoittostart the motor, This isinaspecification of1886, and seems tomeastill simpler plan. Professor S.P,Tuomrson: Thereisonemostpregnant Pcwor,remark pointing tothevery remarkable experiments which Mr.Mordeyhasshowntosomeofus:onp.591,hesays,“Thefoundation andparallel working should bethat theprime motors oss ALTERNATE CURRENT WORKING, [May90, Yoiemoe, “tare under thecontrolofthegenerators.” Thoseofuswhoar “w"" ‘thewayinwhich that generating dynamo persevered initswork, when thesecond alternate current machine wassuddenly thrown upon itasamotor, could nothelp enquiring inwhat waythe steam-engine was governed which kept upthe speed. Mr. Mordey rather startled mebytelling methat hissteam-engine hadnogovernor. Thatmayaccounttosomeextent,Ithink,for some ofthe peculiarities oftheexperiments. Iwould liketo know agood deal more onthat point. ‘Mr.Mordey takes ustoarclamps, and points outthat with alternate currents50voltsarerequired, whereaswithcontinuows currents 60volts or#0arerequired. Here wehave acaseof which very little isknown. Why isitthat anordinary arelamp with acontinuous current will not work well without atleast theordinary 60volts? We know that there isineffects back E.M.F. in the are, orat least there issomething which can bemeasured assuch, Ihave lately been busying myself employing agroup ofstudents toinvestigate thequestion fat.what position inthe aredoes this back E.M.F. come it, Ihave satisfied myself that, indirect current arcs, thegrest drop ofpotential inthearc, usually some 39volts inamoust, occurs atthe positive end orcrater, and that itdoes not occur inthearcitself, nor atthe negative pole. ‘There isa definite sort ofpolarisation atthecrater end. Ihave notset examined alternate current ares, and Icannot give themeeting anyinformation upon that point, butprobably weshould find that there issome other disposition when wemake the corresponding measurement foralternate arcs. Mr.Mordey speaks ofthedrop inthecharacteristic curve of hismachine asbeing partly duetoresistance andpartly toself | induction, Idonotwanttoquarrelovertérms(andafterallit | isaquestion ofterms), butIwould rather express itthatthere is | adrop inthecurve duetoresistance,afurtherdroppossiblydue tothespurious addition totheresistance arising from self-induc tion, andafurther drop duetothedemagnetising effect of thearmature currents onthepole-pieces. ButIamnotall sure whether that does notinvolve thequestion ofselfinduc- | tion, forthelasttwothings arereally parts ofthesame 1880) DIsoUSBION. os phenomenon, theself-induction ofthearmature anditsmag-, netising effect being bound uptogether. ‘Mr.Kapp hassaidthat thearmature standing away from its field magnet could have very little self-induction, butthat all that would bechanged directly you put itinthemassive iron circuit. Idiffer from him onthat point, and will telltheSociety why thepresence orabsence oftheiron field-magnet cannot make very much difference. Inorder that theself-induction should bevery much increased bythepresence oftheiron there, thecircumstances must besuch that theiron can act. The presence ofiron canincrease the self-induction inacoilonly when andastheiron ismagnetised by,and adds to,themagnetic action ofthecoil. But inthecase ‘ofgreat masses ofironlikethese polar projections, with across- ‘sectonineachof,Isuppose, from15to20squareinches,not laminated, how much magnetism will you get into that from acurrent inacoilthat flies past attherate of1-200th part ofa second? Ithas notime tomagnetise the iron, and therefore cannot perceptibly produce self-induction. Consequently the presence ofthesolid iron field-magnets surely cannot make much difference inthe self-induction ofthe armature. Thave devised awayofelucidating apoint intheaction ofan alternating-current machine used asgenerator and motor bya graphic construction, which isasfollows :— a vl \ NN \ c /,\Z Foti Primer LetOE, andOE, represent thetwoalternating electromotive | forces inthe two armatures, agreeing nearly inphase wheo the two machines areboth acting inparallel, with thecurreat from each going inthe same direction asthe E.M.F. The current will (because ofself-induction inthe circuit) lag behind the resultant E.M.F, byanamount represented by theangle g. When thesteam issuddenly cutofffrom one ofthe machines the current through itisinstantly reversed ineign; hence, though ite E.M.F. remains inthe sume direction, electric work isdone oninstead ofbythe machine. Now, inthis diagram, amere variation oftheclock-diagrans ‘used forharmonic motions, the work done inaeycle may be represented asanarea, bythe simple device ofputting the line that represents one ofthetwo factors—say thecurrent— back byjust one quarter inphase, and then completing the parallelogram between the two factors, Let 0Crepresent the resultant current, retarded bebind the resultant E.\LF., and further retarded by90°. The area contained byOE, and0C (relating tothemachine lagging inphase, thegenerator) wil represent the electric work done bytheone machine, whilt the area contained by0E,and OC (the latter being no considered negative) will represent the electrie work dove on theother machine (the oneadvanced inphase, themotor). When you throw offthe belt and one machine becomes & motor,beforeithashadtimetochangeitsspeedtoanyappreciable extent, the E.MLF. will have altered alittle inphase, and anexcessively slight difference inthe relative position ofarmature coils andfield-magnet poles—corresponding, perhaps toone-hundredth ofaninch attheperiphery—will make allthe difference between itsaction asgenerator orasmotor. ‘The stability ofthemachines inrunning, anditsdependence upon self-induction, surely ought tobecleared up. Idonotst allagree with Mr.Kapp that you want self-induction inthe armature inorder toproduce stability;Ishouldbedisposedto agree with him that self-induction inthecircuit outside the machines would tend tomake them synchronise, because thelag ofcurrent which produces synchronism isitself duetosel 3) DISCUSSION. cn induction;butself-induction isnotneededinthearmatureforfrm, thispurpose, Dr.Fleming tells usweought nottousetheword “mag- “netisation” where wemean toexpress the intensity ofthe magnetic induction. But didnotDr.Hopkinson usethevery word inthe title ofhispaper before the Royal Society onthe magnetisation ofiron—a paper devoted tothe intensity ofthe iudaction indifferent specimens ?Mr.Mordey’s paperseemsto metobedistinguished foritsperspicuous and conspicuous common sense. Tenyears agowedidnotknow how todesign direct-current machines torun asdirect-current motors; the question was inaperfectly chaotie condition, With better designs ofgenerators, however, came abetter understanding of motors. The subject ofalternate-current machines wasalmost inaschaotic acondition until wehad Mr. Mordey’s paper to ealighten us,Hehascertainly proved tousthatgood design inthe machine isthe first essential ofsuccessful simultaneous working ofalternate-current machines. Professor W.E,Aver: Although [have notes ofmany trim remarks tomake, Iwill confine myself, owing tothelateness of thehour, tothose points ofthegreatest importance. We must, ofcourse, commence bysincerely congratulating theauthor, for, indeed, ifhehad done nothing more than show that Dr.Hopkinson’s mathematical conclusions were experi- mentally correct, there would beanenormous debt duetohimby allelectrical engineers. Anticipating thattherewouldbesomequestionto-nightas towhatwas,orwasnot,theseif-induction oftheMordeyarmature, andwhether itwas, orwasnot, greater when theiron wasina particular position, oneofmyassistants—Mr. Sumpner—and some ofmystudents—Mr. Lamb, Mr.Smith, andMr.Woods—went to theBelvedere Road, and, bythekindness ofMr.Mordey, made experiments ontheself-induction ofthe armature, with different Positions oftheiron pole-pieces, andwith different currents passing round thefield magnet. The method employed consisted in balancing theunknown self-induction ofthearmature against theself-induction ofouradjustable secohm-standard, andusing the 66a ALTERNATE CURRENT WORKING. (lay300, Tpfemer secohmmeter torapidly alternate thebattery andgalvanometer connections—an arrangement which supplies avery exact and | sensitive method ofmeasuring self-induction. Figures AandBshow therelative positions ofthearmatore coils andthepole-pieces ofthefield magnet. a . Ws U\ RAL Fro, 16 Self-InluctionoftheArmtbureoftheMorleyAlternator. [SSJeoweem| |.|ama[ee Carrent|positionofthe st je A100|10|ones|om° » roo|nao|oats '°Brom |10|oon}osny Arom|100|ossyy,|0A|1,0001,00|vaso|Fi 40 B |1000|1,000, 0-025 |008) ‘The first column inthepreceding table contains theexciting current, inamperes, passing round thefield magnets; theseeund column therelative position ofthearmature andfield magnet for each test; pand gare theresistances ofthetwoproportionalcolt oftheWheatstone’s bridge; thefifth column gives thesecobms 1sindicated bythesecohm-standard when itwasadjusted topr duce balance; and the lastcolumn contains theself-inductionof thearmature itself. Inalltheexperiments thesecohmmeteras rotated togiveafrequency ofabout100,whichisthefrequencys- tuallyemployed byMr.Mordeyintheworkingofhisdynamos, The armature hastheminimum self-induction, 0-030 secohm, whenit 160.) DISCUSSION. oot isintheposition Brelatively tothefieldmagnet, andwhenthetremorexcitingcurrentof40amperes ispassingroundthefieldmagnet; while thegreatest value, 0-038 secohm, isattained when the armature isintheposition Aandnocurrent ispassing round the fieldmagnets, Itmight have been imagined that thearmature would have hadagreater self-induction inposition Bthan inposition A,since inposition Btheiron ofthepole-pieces ofthefield magnets covers the coils better than intheAposition. The increase of self-induction ofevery other coilinposition Bis,however, more thancompensated forbythediminution ofself-induction inthe alternate coils; sothat itisinposition A,when every coil is Partially covered byiron, that the greatest self-induction ix obtained. Professor Silvanus Thompson considers that thepresence ofa large mass ofiron cannot influence thevalue oftheself-induction ofacoil forafrequency 100; butthis, Ithink, isamistake, andIwould refer him tothe experiments wedescribed when we firsthadthehonour ofbringing thesecohmmeter toyour notice, and inwhich we showed that the introduction ofasolid iron core intoasolenoidincreasedtheself-induction ofthecoil26times ‘when thefrequencywasfairlyhigh.Nodoubtthegreaterthe frequency the less effect will iron have, and forvery great frequencies thepresence ofsolid iron willprobably even diminish theeffective self-induction ofacoil; butforafrequency of100 Ithink that itwould bequite wrong toconclude that thepresence ofalarge mass ofsolid iron didnotincrease theeffective self- induction ofthecoil. Thevery remarkable experiments carried out.byMr.Mordey ofcoupling twoalternators together which areproducing avery different E.M.F., and obtaining asthe resultant E.M.F. the arithmetic mean ofthetwoE.M.F.’s, are, asMr, Kapp haspointed out,probably due totheweakening ofthefield oftheone machine and thestrengthening ofthefield oftheother bythe action ofthe currents inthe armature. Indeed, when Dr. Hopkinson brought hisadmirable paper onthe “Theory of “Alternating Currents” before thisSociety in1884, Ipointed out 6 ALTERNATE CURRENT WORKING, [Alay 0, Reteer(seepp.630and531,vol.xiii,,ofourJournal)thattheremsonly one point ofincompleteness, asitappeared tome, inhis investigations, andthatwastheneglectoftheactionofthe current inthe armature onthe field. ‘Through thekindness ofMr.Mordey inlending usoneofbis transformers, aseries ofexperiments has forsome time put been conducted bythestudents attheCentral Institution02 theefficiency ofthistransformer forvarious frequencies andwith various loads. Asfaraswehave gone atpresent—making experi ments atfrequencies 53,80, 160, 240, and with loads varyiog from alight load uptothemaximum working load intended with this transformer—we have obtained noindication whatever ofthe remarkable result described byMr.Mordey, that histransformer ismore efficient foranintermediate frequency than forahigher orlower frequency. But, onthecontrary, wehave found the efficiency tosteadily increase with thefrequency, this increse being particularly marked forlight loads. Ourexperiments were notmade bythecalorimetric method, which isavery tedious ove but bythe double method ofusing aquadrant electrometer, whichProfessor Fitzgerald andIhitonduringthemeetingof theElectrical Congress atParis in1881, fortheaccurate measure mentofthewattsgivenbyanalternate current(whichmaybeanyfunctionofthetime)toacireuitwhichmaycontainanyamountofself and mutual induction, ‘The non-inductive resistance which this method ofmeasurement requires tobeused inseries with thetransformer i,intheexperiments now being carried outat the Central Institution, composed oflong strips ofplatinoid about 2inches wide placed back toback with thin shellscced silk, the current going upone strip and down the other one. Inthiswayaresistance isobtained with alarge cooling surface andwith practically noself-induction. How theefficiency may vary with thefrequency foralosd greater than themaximum load that wehave hitherto patoo thetransformer Idonotknow, andIshall notproceedtoascertain until IgetMr.Mordey’s permission tosend more current through histransformer than itisintended tostand, butforallloads lex thanthisourexperiments havenothitherto confirmed hisresult 1089 DISOUSSION. eas ofobtainingmaximum efficieneyforsomeintermediate frequency. RrtemrThat reminds me, inpassing, that the method Mr. Mordey suggests forgetting the absolute efficiency when using the calorimetric method Idonot think isapplicable; indeed, the method hesuggests was thought ofbyourselves, and rejected, when wewere making thecalorimetric experiments which we described here ayear ago. The method weused inthat investigation was tohave astream ofwater flowing through the calorimeter until itarrived at acertain constant tem- perature;then,byknowingthetemperatures oftheingoing andofthe outgoing water, and the amount ofwater flowing perminute through the calorimeter, weobtained the total quantityofheatlost.Mr.Mordeyhassuggested inhispapera different method. Hehasproposed sending asteady current through thetransformerandraisingthetransformer bythesteady carrent tothe same temperature towhich itisraised when youareworking with analternate current;then,knowingthe watts wasted bythesteady current—which can,ofcourse,beeasily measured—he assumes youwould know thewatts wasted bythe alternate current, which aremuch more difficult tomeasure. But iththesteady current thewaste ofenergy isentirely inthe ‘ire, whereas with analternate current there is,ofcourse, waste duetotheFoucault currents intheiron, Now, ifanamount of heatthat isgenerated bythealternate current intheiron isto begenerated bythesteady current inthecoil,itwillrequire that thesteady current shall befargreater than thecoilisintended tocarry. And itwasthis fear ofburning upthecoilofthetrans- former wewere then experimenting with, and which wasalsoa borrowed transformer, that prevented ustrying this method suggested byMr.Mordey inhispaper. Professor G.Foxses: Iwould, inthefirst place, askMr.fixer" Mordey tomake quite definite what does not appear inthe paper, butwhat basbeen letout inthe discussion, that inthese experiments with parallel working there wasnogovernor onthe steam-engine. Ithasamost. important bearing, Ithink, onthe whole question whether there wasagovernor ornot,forwemust have8governor inactual work. Asamatter offactasteam- 8 ALTERNATE OURRENT WORKING. [Hay0, Brewer engine issupposed tobegoverned byaperfect governor, inorder tokeep thespeed perfectly constant, andthen theonly interrap- tion that cantake place between itand the dynamos isbythe slippingofthe belt, ifs belt isused atall,and ifnot then itis only bytheimperfection ofthe governor that’ the variationcan takeplace.Ishouldlikesomemoreinformation onthispoint, ‘asIthink itwill explain many apparent anomalies. Thave agood deal more tosayabout parallel working, butt this late hour Iwill not continue. Atthebeginning ofthepaper Mr.Mordey isperpetuating an error which agood many have been propagating oflate,.as to assuming thatthereareonlytwomeansofdistributing alternating currents, oneisbyasingle large machine, and theother bys number ofsmall machines, which must beworking inparallel. Now Ihave noticed itsocontinuously stated astheonly alternative ‘methodthatIcannotunderstand it,Ithought therehadbeen almost toomuch said about theAmerican practice, atleast many people thought so,buttheAmerican practice hasbeen invariably not towork the small machines parallel, and tohave alarge number, thedifference being that themains arenotallconnected inparallel;therearealargenumberofeach,andeachdistrict issupplied by»number ofseparate mains and feeders. One dynamo will have anumber offeeders even when the maximam current ison, and when asmall load isonallthese feeders may beputontoonemachine. ‘Astothe machine which Mr. Mordey has shown onthe paper,hehasnotactuallymadeit,buthehasdrawnit,and,ashe tells mehehad drawn itagood long time before Ipublished ‘mine, therefore thecredit allbelongs toMrMordey, forhebas priority. Ihave made oneofthose machines, andalthough atthe present moment theresults have notbeen completely satisfactory, still there aregreat hopes ofgetting agood machine oatofthat. ‘The self-induction ofthe machine inthe form inwhich itwe made was enormous, and the variations involts when the different strengths ofcurrent were taken offwere something perfectly surprising, infactyoumightalmostsaythatthecurreat 19] DISCUSSION. or vasconstant ;youcould getalmost anyvoltsoutofit,butalways Priemor nearly the same current going. Itwas avery interesting dynamo machine. ‘ThePaesiwenr: Before calling onMr.Mordey toreply ImaysiWiian perhaps beallowed tomake afewremarks onhisbeautiful experi- ent inthecase inwhich twoalternators, onewith apotential of 2,000 volts, and the other 1,000 volts, arecoupled inparallel. ‘Thesecret ofthesurprising result ofthisnovel andoriginal com- bination hasnot,Ithink, been quite touched byeither Mr.Mordey himself orany ofthose who have spoken inthediscussion except Mr.Kapp. The 500 volts difference ofpotential would, not- withstanding ohmic resistance and impedance byself-induction, }roduee current inthe circuit ofthe two armatures vastly greater than any that Mr. Mordey has found inhis experi- ments, without some equalising influence which bas not been hitherto suggested except by Mr. Kapp. Tofind what this influence is,suppose, for simplicity, the ohmic resistances ofthe armatures tobezero, and suppose the two armatures tobesimply joined inparallel ready todoexternal work butnotasyetsettodoit. Thus wehave asimple circuit ofthe two armatures. The two shafts might bemechanically constrained torun synchronously insuch relative positions that theelectromotive forces ofthe two armatures conspire inthe ireuit. The 3,500 volts would produce aprodigious current for allthat self-induction could dotoimpede it;but this current ‘ould enormously pulldown themagnetisation oftheiron claws ofthe field magnets; and might even annul and reverse that of theweaker, and thus after the first makingofthecircuitofthe twoarmatures the“prodigious” current through itwould almost instantly become much moderated, even supposing thesupposed initial phase-relation tobemechanically maintained. This is thephase-relation forseries co-operation;and,asshownbyDr. Hopkinson inhislecture onelectric lighting, delivered before theInstitution ofCivil Engineers in1885, itcould only be maintained byrigid mechanical connection between the two theft, But, inMr. Mordey’s actual experiment, the two thafts are free and independent, and they therefore “fall 65 ALTERNATE CURRENT WORKING, [May0h, Sxwulam «into step” cophasally forparallel working. Their electro motive forces become thus exactly opposed inthe circuit of the two armatures, and would yield exactly 500 volts as working E.M.F,, ifthetwofield magnets remained unchanged. ‘The ohmic resistance being still, forsimplicity, supposed ze, thephase ofthecurrent kept going invirtue oftheEMF. would beaquarter-period behind thephase oftheE.M.F. Thos the maximum ofthe current would be atthe instant when the nine magnetic fields areinthemiddle oftheaperturesofnine alternate armature coils, and the direction ofthe current would bethat tending todemagnetise theiron prongs ofthe2,000-rlt machine, and toaugment the magnetism ofthose ofthe1,000 volt machine. Thus (the period inMr. Mordey’s experiment being a100th ofasecond) twohundred times persecond, the former experiences ademagnetising, and thelatter anenhancing, magnetic force. The prongs areofcontinuous iron, andthere must therefore beample self-induction toprevent anyconsider able change ofmagnetisation inthe200th ofasecond perial ofthis varying magnetic force. Thus themagnetic fields art kept, oneofthem weaker, and the other stronger, than they were before thearmature circuits were connected ;and Ibelieve s0much weaker and stronger astoreduce the electromotive forces ofthe two tovery near equality. Ifthe iron ofthe 2,000-volt machine isnearly saturated, itsmagnetism would beless diminished than theother’s increased, and the resulting E.M.F. might benearer 2,000 than 1,000; but Mr.Mordey has found ittobeactually very near to1,500. Afew tums ofinsulated wire round one prong ofeither machine, connected toaballistic galvanometer, would give aready means oftesting thesuggested changes ofmagnetisation. ‘There isreally not much more left formetosay, except that Idowish tobeallowed tojoin inthe general andmost sincere chorus ofadmiration attheresults put before usby Mr. Mordey. We arealladmirers ofhis previously-knowa alternate-current generator. We have now before usforthe first time another form, invented also byMr. Mordey—a form which strikes measexceedingly beautiful, very admirable in many respects, and, Tampleased toadd, quite wonderful. 189 DISCUSSION, ‘660 Mr.J.8.Raworta:WhenIheardMr.Mordey’spaperreadM5, lastweek, Ithought thisvexed question ofalternate current was about totake anew departure, that wewere going togetoutof theregion oftheory and into theregion offact. Mr.Mordey described certain experiments which proved directly that previous statements made before this Institution were not correct. But tonight Iamsorry tosaywehave returned into theregion of theory again. Happily forus,however, theProfessors arenot allononeside this time, buttake opposite views; wemay, therefore, hope togetatthetruth incourse oftime. When Professor S.P.Thompson wasatourworks theother day, Ihada litle talk tohim about theamount oflagbetween the dynamo andthemotor, and hesuggested that itwould be100th ofan inch; Ithought itwould beabout 1inch, We have now succeeded inmeasuring this lag, and Idaresay Professor Thompson will beinterested tohear that itisexactly 13-16ths ofan inch, Professor S.P,Taomrson: Was itafull load, orrunning open ? Mr. J.8.Raworri: Afull load. Professor 8.P.THomrson:Thatisnotmypoint—which was thatyou had twomachines, onedriving theother open. Mr.J.S.Rawortu :Idonotdisagree with youthere. Professor 8.P. Tuowrson: The load that would convert a generator into amotor need notbemore than 100th inch. Mr. J.S.Raworti: The exact difference between the two when running with full load is13-16ths inch, or24° ofa complete cycle. Now Iwant toshow you that wehave been trying toarrive atsome more facts, and tofind out from actual experiments the curve ofE.M.F, atdifferent points oftheperiod. ‘The sketch (Fig. 17)shows aMordey-Victoria alternator with wood drum fixed onthedriving pulley. This wood drum carries asmall piece ofcopper wire (part ofahelix) sunk initssurface. ‘Twobobbinsofthealternator, marked AA,arecoupled tooneend ofthecopper wire carried bythedrum, andtheother endtoone pole ofthecontinuouscurrent dynamo, theother pole ofthe VOL, XVIII. 46 eo ALTERNATE CURRENT WORKING. aay3, Noam, Continuous-current dynamo isledtoamovable handle caryingacopperpointer,marked P.Avoltmeter, V,isputacrossthe terminals ofthecontinuous-current dynamo. The field ofthe saiddynamo iscontrolled byarheostat, marked R.Thesecondary ‘aar i \ iW New Wathod ofmeasurg thedctuol EMfofanAlternator Fro. 7 coil ofatransformer, marked T,may besubstituted forthe bobbins, AA, theprimary coilbeing excited byallthebobbins ofthealternator inseries (asusual). Intaking thecurve of EMF. during one period the dynamo, D,isregulated to give various differences ofpotential, ranging from nothing upto themaximum required, andwhile thealternator isrunning the pointer, P,istried along theseale, 8,tofindpositions ofnospark. ‘These positions onthescale correspond with known positions ot thehelical wire. From thereadings obtained itiseasy toplots curve showing theelectromotive forces developed atevery poist intheperiod. Bytraversing the pointer along the scale wecan measure exactly thepoints inthecycle, where wegetanE.M.F, equl totheopposing E.M.F. ofthecontinious-current dynamo. That point hasbeen ascertained bythefactofthere being nospk ‘atthe moment ofcontact. Wehave not had time yettocarry ‘outacomplete series ofexperiments, andIamtherefore unable toproducethediagramto-night.Wehave,however,succeeded| . 1080) DISCUSSION, on inplotting down thetransformer curve onthetopofthex... alternative curve. ‘Thecurves arevery astonishing, andeverybody who has seen them issurprised attheextraordinary results that have been obtained. The corollary isthat itisnot safe toprophecy until youare sure, We have seen agood many curves drawn onthis allthat have been very beautiful, butnot much like theactual carve, Mr,C.Zirensowskt [communicated]: Inhispaperuponte.zpee “Alternate Current Working,” read before the Institution of Hlectrieal Engineers on May 23rd, Mr. Mordey has made several statements with regard to our dynamos which wehave found tobe erroneous, Mr. Mordey says, “The “Zipernowski alternator will, however, work parallel, but “apparently notvery well; togetittodoso,theperiodicity“hashadtobereducedto42,aswenowknow.” Atanother place wefind, “In acommunication from Messrs. Ganz &Co, “published afewdays ago, Mr.Zipernowski says that thelow “periodicity used was chosen onprinciple, toenable them to “couple their dynamos parallel;”andsomelinesfurtherdown:-— “This bears out very completely my contention that the «periodicity isgoverned inmost cases bysome special feature “ofthe type ofapparatus. Zipernowski hashad togodown “to 42togethisiron-cored alternator tosynchronise, and for “no other reason;forhisstatementisthatsynchronous action “was ‘not asecondary result, butjust theendwewere aiming “aty’ and heacknowledged that there aredisadvantages con- «nected with this lowrate inother parts ofthesystem.” First ofall,wemust state that ouralternate current dynamo: arerunning inparallel circuit inathoroughly satisfactory manner, even when the load varies from zero tothe maximum output forwhich they arebuilt. This may beseen atour central stations atRome, Terin, Livorno, Frankfort-on-Main, and Marienbad. Wehave, also, not found any difficulty inrunning large dynamos parallel with small ones unto the maximum output ofthe Intter. Thus, inMay oflast year, Mr. Blathy has on ALTERNATE CURRENT WORKING, (aySuh, 4:,zp- coupled two600h.p.dynamos inparallel with twoof150hp. each, atRome, the former ofthese dynamos making 125the latter 250 revolutions per minute. Each dynamo isdirecly driven from itsown engine, thesmaller engine being ofSulze', the larger ofVan den Khethove’s construction. ‘The dynamo work parallel with greatly varying load toperfect. satisfactoo, although neither they northe engines areofequal size,the Inter being even ofunequal construction. We pretend tohave been thevery first who have made parallel coupling “an industrially proved fact” (vide Mi. Blathy's letter inThe Electrician ofJune, 1888), and thi, perhaps, may have ledMr.Mordey toconclude rather genenly, ‘and notwith reference toconcrete facts, that this first pracial application ofparallel coupling worked asmost first applications do,“but apparently notvery well.” Moreover, westate that wehave not reduced the periodicity because ofthecoupling inparallel. Since wehave commenced building alternate current dynamos wehave kept thenumber of85alternations with only slight variations. And westill remain convinced that wecan run our dynamos parallel with this number ofalternations easier and inamore simple ‘manner than with ahigher one, butcontest that this isthe only reason forwhich wedonotincrease ournumber ofalters tions, which wehave found suitable forsolong atime. IfMr.Mordey had followed ourpublications with attentioo, orhad asked usforinformation, which wewould have willingly given him, hewould have known that, amongst others, we profit from the low number ofalternations inour alternate current motor, this low number enabling ustobuild small motors without too high speeds; amotor with four poles having our periodicity, would make 1,250 revolutions per minute, but 2,950 with Mr. Mordey’s; ormust have 10pole: inorder tomake about 1,250. Ofcourse this number ofpoles israther high, andtoodear forsmall motors. Intheappendix tothereprint ofhispaper, Mr.Mordey says, “He thought agreat mistake was made insome slter- “nators inusing Pacinotti projections. Inallcases there 1980 ‘DIBOUBSION. os “should beasnearly aspossible asteady magnetic fluxinar.2per “the field. This could notbedone ifprojections were used. “The Zipernowski and Parsons machines were faulty inthis “respect. Itwasimpossible totellbyinspection oftheformer “which was the armature and which the field, They seemed “quite interchangeable, andthiswas, ofcourse, notright.” ‘AsMr. Mordey does notsayone single word toascertain hisopinion that themagnetic flux ought toremain constant, Wecannot contradict any arguments and need not contradict themere statement, But wemust say that wearevery much astonished tohear that Mr. Mordey, who, with full right, so highly appreciates the “sense ofproportion,” isnot able to distinguish the field magnet from the induction coils inour ‘dynamo. We are glad toseethat the figures wepublished about our alternate current’ motor have animated others to asimilar “success.” Our motor, which consists ofone machine only, Possesses allthequalities published some time ago, which are atpresent also claimed, and very ably pointed out, byMr. Mordey. Whether Mr. Mordey's motor, which consists ofan lectro-motor, anexciting dynamo, and anaccumulator battery, villhold good under allcircumstances and forlarge forces as ellasforsmall ones, practice will show. Mr.G.C,Fricker [communicated]: Nothaving taken partinsr.rister. theverbal discussion onMr. Mordey’s paper Ishould like tocon~ tribute afewremarks inthis Journal onsome points which appear tobe ofthe greatest practical importance, asbearing upon the inter-controlling power ofalternating machines when working parallel. Imay sayatonce that Ientirely agree with theprinciples of construction which areadvocated byMr. Mordey, and that tome, incommon, Iimagine, with the majority ofpractical workers in thiefield, hisdoctrine hascome asagreat revelation. Withregardtothisquestionofthemotoractionofalternators inparallel, itisperfectly obvious that synchronism isonly attainedbythepullwhichtheimpressed fieldexertsuponthe fieldinduced bytheimpressed current from theleading machine; os ALTERNATE OURRENT WORKING. lay a4, ar.Picker. that istosay, bytheself-induced lines offorce being drawn into, and made toform apart ofthe impressed lines. Itisevident that the greater the angle between the armature coil andthe polar faee ofthefield magnets, when theformer iscarrying its maximum current, the more difficult itwill beforthe sel induced lines tobecome interpolated into the lines oftheim pressed field, and that, inconsequence, any condition which would retard the rush ofcurrent inthearmature of lagging machine would bedetrimental toitsspeedy recovery ofphase. Besides thecondition ofrapid riseofcurrent inanalternate current motor, itisessential that the couple exerted bythe attraction ofimpressed andself-induced fields should beaslage aspossible, and therefore itisnecessary, when taking steps to reduce the one atthe same time toincrease the other fore. ‘Vhe proper conditions forwhat SirWilliam ‘Thomson termso> phased synchronism arethus anexceedingly strong impress field and anexceedingly weak self-induced field, thelatter condition being necessary toensure instantaneous riseof impressed current, andtheformer toeffect apowerful controlling couple, Itherefore agree with Mr. Mordey, inhisargument, that themost perfect alternator isonewhich hasanindefinitely small co-efficient ofself-induction and aninfinitely strong impressed field.Ontheotherhand,itisequallytruethatsome selfinduction must bepresent, otherwise nocontrolling couple could exist. Inpractice, ofcourse, theimpressed field islimited bythe receptivity ofiron, and therefore, inorder toobtaina adequate controlling couple, itmay benecessary totolerate considerable co-efficient ofself-induction. Mr. Mordey's a periments, however, prove beyond adoubt, that inamachineia which theeffective impressed field isvery large, that thesel induction may bemade very small indeed, and controling properties ofhitherto unprecedented power attained. Itseems probable that informulating the conditions maximum controlling power with anygiven strength ofimpresed field, Dr.Hopkinson, Mr.Kapp, and others, have lostsight theimportance ofthefielditselfinregardtotheproblem,snd ‘thatsofarfromsuchtheoriesbeing“wellestablished,”*| 1869.) DISCUSSION, os Mr.Kapp puts itinhisremarks, they areinreality establishedar.Foster. ona very incomplete basis. ‘ThePuesipext: This isthelastevening before thevacation, and,aswohave another interesting communication tobebrought before us,Ithinkitwouldbeagreeabletothemeetingnotto close at10o'clock, but tocontinue for atleast half-an-hour longer. IwillnowaskMr.Mordey toreply tothediscussion. ‘Mr.W.M.Monpey, inreply, said: Iwish inthefirst place Ms.sonte toexpress mythanks forthevery kind wayinwhich you have received this paper. The criticisms have, insome cases, been answered byother speakers. The remaining points inregard towhich thespeakers arenotinaccord with meImust deal with ‘ofarasIcaninthetime atmydisposal, butIneed notoccupy timebyreferring toother matters. Dr.Hopkinson’s statements onelectrical subjects arealways received, notonly with great respect, butasbeing usually correct andfinal. Ijoin inthe respect, butventuire toquestion the correctness inthis instance. IfDr.Hopkinson would bebetter satisfied bymystating that themost perfect alternator conceiv- able isone which hasanindefinitely small self-induction and resistance, Iamquiteprepared toexpressmymeaning inthat vay. Iwas acquainted with the general practical results, as regards parallel working, brought before usin1884 byhim, bat, forreasonswhichwillbeunderstood, Ididnotderiveanyassistance from histheory, nordoInow agree that that theory “issufficient “topredictalltheresultsobtained.” Icannotfullyarguethis point, butmust askDr.Hopkinson ifhewill kindly look into it, again, for Iamstrongly ofopinion that hewillfindthat itisnotas complete asappears tobesupposed. Iamthemore convinced of thecorrectness ofmyexplanation bythemeasurements oftheself induction ofmyarmature byProfessor Ayrton’s assistants, using themew Ayrton and Perry instruments, and bythestatement made byProfessor Ayrton that the time-constant, orratio ofthe ‘elf-induction totheresistance, wasabout thesame inmymachine and inthat ofanalternatorofanothermaker,whichlattermachine only exhibited very feeble synchronising properties when attempts were made toruntwoofthem inparallel. ors ALTERNATE CURRENT WORKING. (May 3, Medont, The point where mathematical treatment ofthe subject appears tohave failed isthat ithasnot taken sufficient account ofthe“stiffness” ofthefield. This isatonce apparent when this measurement oftheself-induction ofoneofmymachines isreferred to,from which itwill beseen that the self-inductio. ‘was practically thesame whether anyimpressed field existed ot not. Manifestly, with noimpressed field, thecontrolling pore: ofthe machine would bequite inconsiderable, although the values ofself-induction andresistance—upon which Dr.Hopkinsoo relies—would bethesame then aswhen themachine wasworking with astrong impressed field and under the conditions giving themaximum generating and controlling power. Hence itis evident that atheory which only takes cognisance ofself- induction andresistance must bealtogether illusory. ‘There are several reasons forthe view that histheory i: incomplete. His paper hasbeen before the world forsever! years, and has been often studied, but neither itsauthor (sbo hasbeenengaged onthesubject) noranyoneelsehaseverbased anypractical advance onit.Working parallel hasbeen recently considered, asaquestion ofgreat urgency andimportance, before this Institution and elsewhere, and there have been abundant opportunities forbringing forward any useful information, bot the whole subject was misunderstood ;andDr.Hopkinson,who hasbeen awitness ofthestruggles that onallsides were being made towards abetter understanding ofit,hasremained silent. Tam sure thatnothing would bemore unlike himthan toletthe whole industry suffer forwant ofalittle clear and definite information, ifhehad heen able togive that information. Dr. Hopkinson even took part inthe discussion onMr.Kapy's paper afewweeks ago, and threw nolight onthis question, thus tacitly subscribing tothecurrent views. Itwould beeasy tofind several other instances, besides those mentioned inthebrief enumeration inmypaper, toshow thatthe views generally held were entirely misleading andincorrect. Fot example, Mr,Swinburne alludes to*Dr.Hopkinson's mathematica *TheElectrician, JuneMth,1889,p.164. | | i 1869,3 ‘DISCUSSION. em treatment, andsaysthathehasputthematterintoEnglishinMtmordey- hisbook;*andIfindinthat(recentlywritten)bookitisstated (onp.151) that there are noalternate-current motors. But nothing less than theactual words canshow how completely misleading this theory was. Mr, Swinburne proceeded, inthe book inquestion: “It seems generally tobeassumed that the “only difficulty with alternating-current motors istostart them. “It seems probable, however, that there will bevery great “difficulty indesigning amotor which willwork with reasonable “efficiency andvarying load. Analternating-isawidelydifferent “thing from adirect-current motor.” Having seen allthese statements disproved—having seen 50-H.P. synchronising motor running with high efficiency under tarying load,anddesignedonthelinesofadirect-current motor— Mr.Swinburne now comes forward andsays: “Iseenothing in “Mr. Mordey’s interesting experiments which conflicts with the “recognised theory;infact,thepositionofsynchronous motors “ianot altered"—the italics aremine. ‘Truly the Hopkinson theory, asrendered into English, must becapable ofinfinite elasticity. Then again, only afewmonths ago, Mr.Swinburne, who had assimilated this theory, stated? that such machines asmine would notruninparallel atall,and hedescribed an“induction- “coupler” which hehad devised forthepurpose ofmaking them sorun byinserting self-induction, That device isavery good illustration oftheresults oftheoldtheories. Imay mention thatwhen Iwaslabouring intheslough ofthose oldtheories I independently devised that identical arrangement, which isone quite inaccord with therequirements ofthose theories, sofarasI vwasable tounderstand them; andhaving entirely failed togettwo ironlessalternators ofawell-known typetoworkparallelintheirsimple condition,Itriedthat“induction-coupler,” without:any improvement intheresult. That was, infact, oneoftheexperi- ‘ments that opened myeyes tothewhole fallacy. Ithelped me *«Practical Electrical Measurement.” tThe Electrician, June 1th, 1889.4Journal,vol.xvip.401. as ALTERNATE CURRENT WORKING. (fay0, MrMonks: toseethat anything that prevented theinstantaneous passage ofthe necessary correcting current must inevitably tend to prevent synchronous action. Referring toProfessor Adams's criticism, Imay saythatIwas fully aware ofhisexperiments attheSouth Foreland Lighthouse, andreferredtothem.Professor Adamsconsiders thoseexperiments asconclusively establishing Dr.Hopkinson’s theory. I need notagain consider them inthat connection, butwould only saythat the motor effects obtained were not powerful,andthat they have notbeen taken asforming abasis forpractical coo struction. Indeed, they have remained onrecord asthenotverr high tide-mark beyond which nosubstantial advance wasfound practicable, until machines having very different qualitieswee developed. Itwould cause menosurprise tofind that theSouth Foreland experiments succeeded from quite different causes than thoseimagined byProfessor Adams. Itisnotatallclearthata5 oftheadvocates ofDr.Hopkinson’s theory really understand what that theory is. Itseems tocover amultitude ofquite opposite interpretations, andresembles some oftheHebrew propheciesia itscomprehensive applicability tooccurrences—after theevent. Professor Adams,Iamafraid,rathersupposes thatImakesome sortofclaim tohaving introduced parallel working. Theobject ofthefirst portion ofmypaper wasrather tostate what linesI thought should befollowed inorder toobtain the best results, andtoshow that onthose lines complete success could beattained atatimewhen,inspiteofProfessor Adams'swell-known andvalt- able paper, considerable and well-founded doubt existed astothe commercial possibility ofparallel working. Inonerespect, bor- ever, Lamquite unable tofollow Professor Adams, and thatisin some ofhisanalogies between alternators working parallel and tuning-forks vibrating inunison. Hesays that Igave theoctave when Ihadone machine working at1,000 volts and theotherat 2,000 volts, both being atthesane speed, andhesuggests thetI should give thedouble octave, andvarious other ratios. NowI failtoseethat there isanytrueanalogy between various rates ofsound vibration produced bytuning-forks andcorresponding ratios ofelectrical pressure. Itwould bejust ascorrect tos8¥ 1869.) DISCUSSION, oro that asteam boiler with apresbure of120Ibs. onthesquare Mesontey inchwasgiving theoctave ofanother boiler working at60Ibs. Astotheverydifferentproblemofrunningalternators parallel hen thealternations areanoctave ofoneanother—a possibility thatisalso putforward byProfessor Adams, and which hasbeen mentioned tomebyMr.Sumpner andothers—I canonly saythat Thave nottried it,partly because itdoes nothold outanypromise ofusefulresultifsuccessful, andpartlybecauseIfeelpretty sureitwould notsueceed, except with machines having somuch selfinduction astorender them unsuitable forpractical purposes. Alittle consideration will, Ithink, show that this must be the case. Professor Adams refers totheloud humming noise made by nnyalternators when they were connected parallel while outof phase as“very sufficient evidence ofvery considerable self- “induction, inconsequence ofwhich thespeed ofonemachine is “suddenly increased, and oftheother assuddenly diminished.” ‘Thisnoise isreally aproof ofthecontrary. Itiseaused bythe enormous rush ofcurrent which thecomparative absence ofself- induetion renders possible, and this momentary rush puts the machines inphase with ajerk. Asbearing onthis point, Imay mention that some machines that aresaid tohave large self- induction donot make this noise under the same circumstances. ‘Their self-induction prevents any such rush ofcurrent, and, secording tomyview, also prevents their working parallel. Although dissenting from myexplanation, this rush ofcurrent isrecognised byMr. Swinburne (p.656) asaproof ofabsence of selfinduetion. Itisinteresting and important toconsider another action arising out ofthis matter. Inseveral machines—some with aud some without iron—the self-induction issoserious that it does awell toshort-circuit armature coils astocut them out when itisdesired toreduce the potential difference atthe terminals. Infact, these machines cannot beinjured by sccidental orintentional short-cireuits. ‘They obtain this ‘sdvantage, however, atthecost ofevery other good quality. ‘Therushofcurrent, theeffect ofwhich Professor Adams noticed, 680 AMTERNATE CURRENT WORKING. [ay00, Mr.Morte. but misunderstood, inmymachine, could nottake place with such high self-induction machines. They could beput into connection when out ofphase, and would simply choke back thecurrent which attempted toputthem into step. Ithink there isnoreal difference ofopinion between Major Cardew and myself onthesubject ofsafeguards. Iquite agne with himthat thetwo contacts @and bmay be“ready waiting” foraperson tomake theother two contacts and soreceive+ shock.Itistopreventthepossibility ofthisthatIsostrongly recommend earthing. Itisanabsolute safeguard, costs nothing, anddoes notcause interference with thetelegraph ortelephone service. Icannot admit that itincreases fire risks. Ithink it lessens them, because itmakes itcertain that afaulty transformer orhouse circuit cannot remain on. ‘The safety fuse onm earthed circuit instantly cuts offafaulty house, and sopre- vents that gradual accumulation ofunsound oractually defective places that, without some such safeguard, would soon render alarge supply service unworkable. Astothe supposed excessive strain ona100-volt circuit ifitisearthed, that does notexist inpractice. ‘The secondary isearthed inthe middle, sothat the strain isreally only 50volts; anda2 installation that will notstand that strain isradically badand ought nottobeallowed anywhere. The great advantage of earthing isthatitsecuressafety,protectsthecustomeragainst bad work and leakage through his meter, and protects the system against thepresence ofrotten installations. Ifthefre offices donotseebefore long that earthing does notincrease fre risks, Ishall begreatly surprised. One ortwo points inDr. Fleming's remarks require some replyfromme.Hissuggestion toacoountforthebehaviour of transformers with different periodicities is,Ithink, very pretty andveryingenious. AtfirstIcouldnotfullyacceptit,asthe greater penetration oftheeddy currents into theiron atalor periodicity seemed tobeaneffect that would notbelikely tobe felt with very thin conductors, such asthe laminated iron (see under “Conductors”); butInow seethat thecurrents inthe laminations, although very close together and insulated fom 1892) DISCUSSION. ot each other, canactonone another and produce the“virtual x.Monte “resistance ”effect. Professor Fleming’s objection tomysubstitution oftheword “magnetisation” for“induction” isnot avery strong one. I haveonly tosaythat Ithink “induction” would notbeagood word even ifithadnotalready been secured foradozen other purposes. Some limit should beplaced tothe multiplication of meanings given tothis one word, which, after all,isnot an electric ormagnetic word atall. Mr. Kapp has latterly given much attention tothis subject ofalternate-current working, and understands itsdifficulties. Onthis account, and also because Iwas obliged tostrongly oppose some ofhisviews, Ihighly appreciate hiskind remarks about thepaper. Heaccepts allmyfacts, butrejects myinter- pretation. Hewill, Ifeel sure, accept them both onfurther consideration. ‘The experiment ofwhichhesuggests anexplanation, which isinsome respects similar totheexplanation proposed bySir William Thomson, isnodoubt very interesting one. Itisa case notvery likely tooccur inpractice, but deserves attention, Ido not, however, quite agree with Mr. Kapp and SirWilliam ‘Thomson. Iwish their explanations were more convincing, and ‘were not somuch opposed toProfessor Ayrton's measurements oftheself-induction with different relative positions ofarmature and field, and with various field strengths, which showed that theactual conditions areexactly thereverse ofthose relied upon inMr.Kapp’s supposition. Ithink that inany case the actual magnetising orde~ magnetising action exerted bythearmature onthefieldmagnet must besmall, partly forthereason, pointed outbyProfessor Silvanus Thompson, that there wasnottime inahalf-period for thesolidirontoundergo anychange, andpartlybecausethe armature ispurposely made soastoexert assmall aneffect of thatkind aspossible, The machine consists ofavery powerful electro-magnet, thefield; and avery weak electro-magnet, the armature. Even ifcoils similar tothe armature coils were wound ontheextremities ofthe polar horns, and had adirect current ose ALTERNATE CURRENT WORKING. (lay, Me.Monier.sentthrough themequaltothemaximum working altemste current, they would notproduce avery powerful effect onthe field. One ofthedifficulties met with inconsidering this question isthat there isnomeans ofknowing which machine ws generator andwhich motor. This depends, ofcourse, upon the relative power oftheengines. Itisquite possible thatthe lower E.MLF. machine wasacting asthegenerator. Tecan only saythat Iwilltake thefirst opportunity ofacting onSirWilliam Thomson's suggestion astotesting thesupposed changes ofmagnetisation, and hope thus tobeable tosettle this question. ‘Mr. Swinburne challenges myconclusions onseveral points with hisusual vigour, and Ivery willingly take upthegauntlet. Ihave already alluded tosome remarks ofhiswhich place himin thehonourable light ofaninterpreter ofDr.Hopkinson’s mathe matics. The position isone that requires great powers of endurance andconsiderable courage. Mr.Swinburne says that “electrical engineers donotrealise “what iswanted inparallel running.” Iaccept this candid state ‘ment as,atanyrate, quite correctly describing hisattitude inthe matter; and amtherefore, ofcourse, prepared tofind thatbe disagrees with most ofmyviews. Hecommences bystatingthatsofarfromthemotorbeing kept instep because itsself-induction was small, that, ifthere were noself-induetion, thegenerator would have nocontrol atall over the motor. ‘This isaninstance ofavery common mistake—that ofsupp ingthat ifthere ispractically noself-induction there canbe0 generatingormotivecapability. Theexactcontraryisthecasei asItried toexplain inthepaper when Iexpressed theview tht “aperfect alternator foranyandevery purpose should have |“resistance andnoself-induction” (page591).Ofcoursethis condition isunattainable inpractice. Itisamere expression ofthedirection inwhich itappeared desirable toadvance; and, asIhavealreadysaid,Iamprepared toputitthatthemotperfect alternator conceivable isonewhich hasanindefinitely 190] DISCURSION. ors smallself-induction andresistance. ‘Thisideathatthegenerating Mt.Mowe,orcontrolling powerisconnected directly withtheself-inductionisclearlyupset.bythefact,shownbyProfessor Ayrton’s experi- rents, that theself-induction isnotsensibly altered byvariations ofthefieldstrength orposition, Ineedscarcely repeatthatthe generating orcontrolling power depends almost entirely onthe fieldstrength and position relatively tothearmature. ButMr,Swinburne sticks to'the ordinary fallacy andsays—or implies—that such amachine would not act atalleither as generator ormotor, and reproaches meforattacking existing theories without giving oneofmyown totake their place, Even thatcourse would have been quite justifiable ifIhad found the existing theory bad. ButIhave given atheory. Ihave saidwhat Tthought aperfect generator ormotor should be,and have attempted toexplain therationale oftheaction oftwo such machines when controlling each other. Letmetakeacase. The armature ofmymachine has acertain resistance and acertain self-induction. These qualities are unfortunately unavoidable, batImade themachine asgood asIpossibly could byreducing them both. This was done bytheuseofastrong and “stiff” field. Need Iexplain that, iftheproperties ofiron hadpermitted it,Ishould have been able sotoincrease the field astoallow ofa proportionate reduction oftheresistance and oftheself-induction ofthe armature? And ifIcouldhavegotanindefinitely strong field Ishould have reduced the armature resistance and self- induction also indefinitely, Then Ishould have the“perfect “alternators” that Ialluded toworkingasgeneratorsorasmotors, singly orparallel. ButMr.Swinburne, bearing outhisopening statement, implies that such machines would not work atall. ‘Then, again, speaking generally, hesays: “To make them run “parallel commercially needs alarge corrective tendency. This “can only begotbymaking themachines sothat themaximum “current isnot atthe same time asthe maximum E.M.F.” This isexactly theopposite oftheactually best conditions. Of ‘course they need alarge corrective tendency and great stability, asDr,Fleming says. Iurged this strongly inmypaper cy ALTERNATE CURRENT WORKING. [May3, tztords. (page 591). Tnverting Mr.Swinburne’s statement—which relly means that themachines must have large self-induction—I say thislarge corrective tendency canonly begottoitsfullest exteat bymaking themachines sothat themaximum current isatthe same time asthemaximum E.M.F., atanyrate when workingoo aninductionless circuit. Mr. Swinburne quite correctly interprets therestriction imposed bytheconditions hethinks best when he says “that the machines must bemade larger togive agiven “output. They must also besomewhat less efficient.” The whole object ofthefirst section ofmypaper wastoshow that such views were both practically and theoretically wrong, andI amsurprised that @literal restatement ofthem should nowte put forward. The best practicable alternator forany and every purpose should have the lowest practicable resistance and self-induction Itwillnotbelarger foragiven output orlessefficient. Itwil not have itsmaximum current atadifferent time from itsmaxi mum E.MLF,, and itwill exert large corrective tendency inthe most powerful and unhesitating manner, andwill maintain ss chronism with the least difference oftime-phase between a5 ‘two machines. Such aretheviews towhich Iamdriven bythe logic offacts. Ononepoint Iagree-with Mr.Swinburne—thai ‘large corrective margin isrequired, the internal loss must be large; buteven here wedonotfully agree. With theoldtheories thecorrective margin was out ofallproportion totheeffet required tobeproduced. Bythesudden and strong action! obtain bydeparting asfaraspossible from theoldtheories Iget thislarge corrective margin ofpower byasmall margin inthe capacity ofthe machine, Itisobviously unwise, however, to attempt torun machines parallel when they require toexert large controlling power. Myexperiments showed thattheycould exert this power;butthedrivingplantthatrequireslargecoutrl isnotsuitable forparallel working. Mr.Swinburne's calculations ontheefficiency and dimensions oftransformers are, Iamsure, very interesting; butIscarcely think weareyetinaposition tosettle these matters entirely 0» paper. There aresomany variables, andthere issolittleactu! | 1883 DISOUSSION. 3s knowledge ofthelaws governing theheating byeddies andbytori. hysteresis intransformers, that itisprobably best torely mainly onexperiment forthe present, Thus Professor Ewing says (page 642) inhisletter: “We cannot besure that thelossby “hysteresis isproportional tospeed, especially when thespeed is “highs” and myexperiments onarmature cores support this vier, Mr. Swinburne, however, issufficiently sure onthis, andalltheotherpoints,torelyuponhiscalculations fortrans- formers working under widely different conditions. Iadmire his courage, butadvise him tobeprepared fordisappointments when thetransformers areconstructed. Icannot atallagree, either, thattheeddies intheironarenegligible, inspite ofanycalcula~ tionsshowing them tobeinappreciable. Itwill befound that nytests bear outvery closely the results asregards lossthat are known tobeobtained inlow-period direct-current machines. Those experiments confirm, asregards hysteresis, theresults of Ewing andHopkinson ;andastotheeddies itisopen toanyone tocheck them byexperiment orbyreference toany reliable tests. Iwould, however, point outthat inthecelebrated paper oftheDrs,Hopkinson on“Dynamo-electric Machinery,”* one ofthetwo power readings which areavailable forthis purpose iswrong, thetotal lossofpower being given asless than the calculated hysteresis loss. Ihave pointed out this discrepancy toDr.Hopkinson, and find that heisaware ofit. Itwas no doubt anincorrect reading. Professor Silvanus Thompson and Professor Forbes ask for information astothegoverning oftheengines driving thetwo ‘lternators intheexperiments, Those engines were notgoverned ‘automatically atall; they were simply controlled byhand. Of course, ifthey hadbeen well governed automatically, thetests would have been much less severe. Asitwas, itwas lefttothe engine-driver togive each engine asmuch steam ashethought wasnecessary. Insuch acase, ifone engine ismore powerful than another, thepower ofmutual control isseverely tested;if theengines are working under similar conditions, then the *Phil, Trans, Parti. 1886 VOL. XVII. 47 eae ALTERNATE CURRENT WORKING. (as, rMoney.corrective marginisnotrequiredtobelarge.‘Theutmoststnin isputonthearrangement when, aswasdone intheexperiments, steamisentirely shutoffoneengine. Ttrust Professor Thompson will continue hisvery interetig experiments onthe seat ofthe back E.M.F. inares. The subject requires investigation. ‘The method used willprotas yield useful results asregardsacomparison between direct-ant alternate-current ares. There isonepoint towhich Iwish draw Professor Thompson's attention, and that iswhether there is,inanA.C. are,aneffect oftheopposing F.M.F. inprodacing adisplacement ofthecurrent phase, Isuppose itmay betakes asproved that thearchasanE.M.F. aswell asaresistance; abl with arapidly alternating current itseems possible that thismay produce alag,andmay actinthat respect something likeelectro magnetic self-induetion. Such aneffect may pethaps alsob found inA.C. electrolysis. Inote Professor Thompson's criticim ofmystatement that inthecharacteristic thedrop ispartly doe toresistance and partly toselfinduction. Hedoes notsaytht Tam wrong. Several effects areproduced bythecurrents—eidy anduseful—generated bymotion, and itis,Ithink, allowable to describe the net result on the characteristic nsdue tosit induction. Thave already referred toProfessor Thompson's expression of doubtastowhether thereistimeinahalf-period forthelarge mass ofiron inthepolar hors tobecome maguetised. Heis probably quite right, but Ihave had noopportunity ofsettling thepoint. Ifthere isany effect itwill probably bealmost steady one, a8pointed out bySirWilliam ‘Thomson;thatistosay,theiron will notfollow thevarying impressed magnetising force, butwill take upanaverage state, depending upon the current and therelative position ofarmature and field. Professor Ayrton, inhisopening words, appears toallyhimself with theviews that Ihave sought toupset; butIhave tothank him forhaving contributed anaccount ofexperiments which perhaps more than anything Ican say, tend tosupport mycot- tention, Hereferred tothemethod oftesting transformers th! Tsuggested. ‘That method may besafely used, except with »| veryinefficient transformer. | ‘Astothecurves oftemperature, Icanonly saythat Ihave Xe.Money noreason todoubt the accuracy ofthe tests, which were most carefully carried out. Professor Forbes pointed out that itiswrong tosuppose there areonly two means ofdistributing alternate currents, I quite agree with him;butalthough Ihavedweltatsomelengthonthesubject ofworking parallel, Idonotthink'that method is suitable forallcases, Itisagreat convenience, and ina-pro- perly designed station can always beadopted;butwhere,either from thedesign ofthemachines orfrom thecharacter ofthe engines, alarge corrective margin isnecessary, then itmay be better towork singly, putting such ofthe mains inparallel on onemachine asmay beconvenient. Ifthe alternators are suitable forparallel work, then the different mains may berunparallel from acommon oromnibus jairofterminal bars, each main being fedthrough asafety fuse, inorder toprevent more than loeal extinction incase ofashort cirouit. With regard toMr.Zipernowski’s communication, Iamsorry heshould think thestatements Imade about working hisalter- nators parallel areerroneous. ‘Those statements aresubstantially quotations from hisown writings. Heobjects toapassage inmy paper inwhich Isaid that. hisreason forworking atthevery low periodicity of42wastoenable him torunthemachines parallel. Myauthority washisownletter, published afewdays before my paper. Inthis letter—to quote itmore fully—he said: “The “relatively lownumber ofalternations (5,000 perminute), which “we have chosen onprinciple ...that webythis means are “enabled tocouple ourdynamos inparallel isnotasecondary “result, butjust theend wewere aiming at.” Mr. Zipernowski now writes (page 672): “Westate that we;have notreduced the “periodicity because ofthecoupling inparallel.” Imust leave these two opposed statements tospeak for themselves. ‘The other statement which Mr.Zipernowski questions isthat *The Electrician, May 10th, 1889. | 88 CURRENT.METER, (aya0, MeMonier. themagnetic fluxinthefield ought tobeconstant, andIamblamedfornotgivingmyreasonsforthisstatement. Ididnot gointothematter veryfully,asitseemedunnecessary ;butwhen Tsaid that want ofconstancy “necessarily ledtolosses inthe “field, the lamination ofwhich inthe Zipernowski machine “showing that such losses must beserious” (page 629),Iam sure Isufficiently indicated thenature ofthewaste which occun ‘under the conditions referred to, TamgladtohaveelicitedMr.Zipernowaki’s remarksaboot theGanz alternate-current motor, andshalllookforward toseeing “ ‘anactual description ofthat apparatus. The publications onthat subject, towhich myattention iscalled, areremarkable forthe smallamountofdefiniteinformation theycontain, Icanleannothingfromthem.Wearenottoldanything atallaboutthe form, mode ofaction, orprinciple ofthemachine. There areno doubt good andsufficient reasons forthisreticence. OfMr.Fricker’s contribution Ineed only saythat itisavery clear statement, with which Iwholly agree. ThePuesipenr: IwillnowcallonMr.Hookbam, ournewest acquisition asamember, todescribe thecurrent-meter whichhe hasbeen good enough tobring before usthisevening. Mev, _Mr.G.Hooxman: Iamvery much obliged toyouforthe indulgence which you have given methis evening, inprolong- ingthemeeting onpurpose tohear meexplain mymeter. Of course Iunderstand that thetime isextremely short, andIwill make thebest useIcanofit. Mymeter isarather small air, anditwillbeimpossible toexplain itsatisfactorily toameeting like this without diagrams; but Iknew that mytime inanyoxte would beextremely short, that Ishould notbeable tomake use ofthediagrams, and that Imust content myself entirely with just placing themeter onthetable and introducing ittoyou, doing what Ican toexplain itsworking parts from themeter itself, ‘The meter consists essentially ofanelectro-motor, andissel contained, requiring noextraneous motive power like clockwork oranything ofthatsort,thefieldbeingprovided inthiscateby 1888} DISCUSSION. os9 permanent magnets. Thebrasstubecontains 10or12bar-me.magnets oftungsten steel,extremely hard,weighing abouthalf Ib, apiece, sothat there isperhaps 5Ibs.or6Ibs.weight of tungsten steel inthebrass tube. The iron pole-pieces fitfairly ‘ellbetweenthemagnetsandthecast-iron, soastoreducethe sirspace resistance atthose points. There isasimple anti- friction train, the practical useofwhich isthat itavoids the necessity oflubrication. Adouble mercury commutator isem- ployed, theconnections between the sections onthe two halves being soarranged, that when thetwohalves aredipping with their lower edges intwo mercury cups onthesame level, the effect isthesame aswhen inanordinary dynamo onebrush presses onone side ofthe commutator and the other onthe opposite side. [Detail working ofmeter described.] ‘The proportionality isobtained inthis way. This wire is mounted onasolid copper disc, andasthat revolves inthefield, Foucault currents aregenerated inthecopper disc, atthesharp edges ofthe pole-pieces,andtheworkdoneonthoseFoucault currents will beasthe square ofthe speed; the brake-force increases asthespeed, consequently thespeed willvary directly asthe driving force. Inthis case the armature carries the current tobemetered, revolving inaconstant field, consequently thespeed willvary directlyasthecurrent.Icannotinthetime atmydisposal show you itsproportionality, butIthink every lectrician present will understand that ifwecangetridof friction—i.e, make allother work done negligible compared with thework done onFoucault currents—the thing must bepropor tional;asamatteroffact,itismostaccurate.Ofcourse,fora verysmall current, where thedriving force isalsoextremely small, friction, both static andtheliquid friction ofthemercury, enters asadisturbing cause, Thus, ina40-light meter, with onelamp there may beanerror ofperhaps 25or30percent., with two lamps 10percent., with three orfour lamps itpractically disappears, and forallnumbers above that itisabsolutely Proportional. (Meter shown inoperation.) ‘The power issupplied bycells kindly lent bytheElectrical Power Storage Company. Themeter hasbeen inuse18months, 90 CURRENT-MBTER. (ay908, Nua,andtherehasbeennofalling-offinthefield.Inourrecest patterns weomit theshort-cireuiting bar, aswefinditquite ‘unnecessary. The Presipent: You have not made itquite clear onwhat fundamental principle the meter reste—that is,thewaythe magnetic brake acts. ‘Mr.G.Hooxam: The brake ismade byFoucault currents generated inthedisc, and thebrake increases proportionally to thespeed.Tenamperes arenowrunning, andthatisthefall current forwhich wesell this meter; itisnow running atIf amperes, butitisinconvenient— ‘The PrestoeNr: Inconvenient inwhat? ‘Mr. G.Hooxwaat: Itmight throw themercury about. ‘The Presipent: Ihave seen the action ofthe mercury on thepieces ofcopper dipping into it. ‘Mr. G.Hooxwam: Inthe shunted meter we use anarmatare ofcomparatively highresistance, sothatthevariation ofresis- ance atthe mercury contacts—the only variable part—is u- important, Wetested thisvariation onameter whose armature hadaresistance of1-500ofanohm,andfoundavariationof 10percent. ‘The armature ofthis shunted meter hasaresistance of-1 ohm,s0thattheerrorisreducedfiftytimes.Iwishtocallyour attention tothis form, asanimportant point isinvolved. When showing ittoSirWilliam Thomson, hesaidIshould findanerr ‘when there wasavariation inatmospheric temperature,theshust conductorbeingofplatinoid andthearmature ofcopper;batitis ‘acurious fact that there isacompensation here forthe tempen- tureerror.Wehaveheatedthemeterupto50°F.aborethe atmosphere :theresistance ofthegermansilverremainsconstaa',consequently itgetsmorecurrent; butatthesametimethedix isheated upsothat the resistance tothe Foucault. currents is increased, andconsequently thework done less; and uptos?” wefindnotemperature error inameter shunted with platinoid. ‘ThePaesipent: Will yousayaword ortwoabout themetal contacts?‘Mr.G.Hooxnam:Thatisapointwehadtotestinpractice andwehavehadtoappealtoitastheonlytestforshowing 1009.) DISCUSSION. oot which isthebestform ofcontact. Wehave hadmeters under i... observation now formore than eighteen months; wehave tried platinum and other forms ofdrycontact, butthebest ofallisan amalgamated contact along theedge ofacopper commutator plated with nickel, the nickel being removed along afine line at theedge ofeach section. This remains quite unaltered month after month, andwenever have found anytrouble with that, and theresistance ofanamalgam contact ispractically nil. Professor J.Penny: May Iask what Mr.Hookham finds is newinthis? Isittheshape ofthemagnet hehasused, orwhat does he claim asnew ? Mr.G.Hooxnam: Iquite understand thebearing ofProfessor Perry's question, What isnew isthat this isaworking instru- ment. Iam quite aware of,and Ihave nottheslightest wish to disparage, thework done byProfessors Ayrton andPerry;Iam familiar with thespecification inwhich the principle islaid down independently. This isnotthetime todiscuss suchaquestion asthat; butIsuggest toProfessors Ayrton andPerry that they have notdescribed aworking instrument inthat specification, The Paesivent:Theprinciple ofusingFoucault currentstoSrNilam control speed foranelectric meter was,Ibelieve,firstsuggestedby Messrs. Ayrton and Perry, andthus became known tomost ofus. ‘The way ofcarrying itoutinMr.Hookham’s meter isdecidedly novel, Ibelieve; and, when weconsider the difficulties tobe overcome, wecannot but admire the result. Mr. Hookham's ‘meter fasbeen inpractical useforsome time, Ibelieve ? ‘Mr. G.Hooxam: There areabout 200inuse, andthelongest, hhas been inuse for about 20months. ‘The Paesipent: The exceedingly convenient adjustment of themagnetic field isimportant, but Iamstill more interested to learn that noreadjustment hashitherto been found necessary. ‘That shows that themagnetic part oftheinstrument isofa valuably durable character. Mr.RE, Crowrron: Ihave hadsome experience with #1, meters used oncentral station work, andhave recently, through thekindness ofMr.Hookham, hadanopportunity oftesting one ‘ofhismeters,orrather,Ishouldsay,thatithasbeentestedfor on CURRENT.MEER. (ay90, ar mebythe resident engineer oftheKensington Court Company. Fromourtestsitappears thatthisformofmeterregistersin favour oftheCompany supplying atvery small loads, although it iscorrect atordinary loads. ‘Mr.G.Hooxmam:IoughttosaythatMajorCardewhasbad the meter under test formonths, and has reported favourably on it. Mr. R.E.Cromprow:Icannotaccountforthispeculiarityin which this meter differs from allothers, other than bysupposing that thebraking action oftheFoucault currents induced inthe copper have not sufficiently compensating effects atthelover portion oftherange. Mr. G.Hooxwas:Iunderstandyoutosayitisinfavourof theCompany. ‘The Paesivent:ItisagainsttheCompany. Mr. R.E,Crompton: Istate that itisimfavour ofthe Company. ‘The Puestent: Itisagainst theCompany. ‘Mr. R.E.Crompron: Ithink you will find thefacts ae 1sIsay. ‘Mr. G.Hooxuam: Ontheoretical grounds, Mr.R.E.Crompton: Yes, ontheoretical grounds, Idont attach great importance tothis point. Iasked itmore with« view ofascertaining whether others have noticed thesame thing. Iwould desire, however, totestify myadmiration ofthemeter itself, which hasbeen beautifully thought out,andthedesign bss been well worked outfrom amechanical point ofview. ‘cNuan ‘ThePRestvexr: Inregard totheconsumer andthesupplying ‘company itisrather theother way, because atallspeeds the Foucault brake-current isinsimple proportion tothespeed, while friction isgreater relatively atlow speeds. Thus,atthe smallest rates ofconsumption, the error isfavourable tothe consumer. Mr.R,E,Cromrron:Haveyoumadeanytest,Mr.Hookham, tosoeifthatisthecase?Ithinkyouwillfindthatthecurve ‘very much flatter. ‘The Present: Every dynamo affords ademonstration that 1089) DISCUSSION, on theEMF,isproportional tothespeed.‘Thequestion,then,istwatnerely amatter ofresistance.Mr.R.E.Crowprox: Ihavenodoubtitis,andithasvery great range, butduring theearly part oftherange Idonotthink itis eo, ThePaeswext: From themost absolutely infinitesimal speed tothehighest speed, theFoucault current issimply proportionaltothespeed,exceptsofarasresistance isalteredbytemperature. ‘Theimportance ofMr. Crompton’s remark astotheaccuracy ofa meteratthesmalloutputismanifest. Inrespecttothetimesof small consumption, there aretwoviews astotheovercharging sndundercharging theuser. Itisvery cheap fortheCompanytogiveasmallcurrentalldaylong,fortheCompany isobliged tohave work kept oncontinuously during thewhole 24hours; and,provided itisnottoogreat, thedemand may benotenough tosensibly increase thework required oftheengines, orthefuel which must beused tokeep them going. Inthat case the Company might beglad togive each consumer aquarter ofan ampere ofcurrent fornothing, simply because itcost theCom- pany nothing toproduce. ‘There isnodoubt that the best neter isonethat records most accurately through allranges from thelowest tothehighest, and ifthere isafault Iagree with Mr.Crompton that itought nottobeagainst theconsumer, but nither against theCompany, forthevery smallest currents ;and Tmay say, from tests that Mr. Hookham showed meinmyown laboratory, wefound itso. Itrecorded toolittle inproportion forthesmallest currents;butfromhalfanampereoranampere upwards itgave awide range ofremarkably accurate simple Proportion tostrength ofcurrent initsrecord. Mr.§.Joyce, jun.:Itiswellknownthatinstruments ofthems so-calledpermanentmagnetclassmay,withgreatcare,bekept“!"™ fairly constant instrength, and, indeed, wehave had inMessrs. Paterson &Cooper's experience several examples ofthis. We have bad, forinstance, aninstrument ofAyrton &Perry's old style permanent magnet type come back forre-testing after being inuseforsixyears, andhave found nopractical variationinthestrengthofthemagnet. Buttheobjection toso-called 6% CURRENT-METER. (Dilayson, MeajanPermanent magnets isthattheymaybeeasilyaltered instrength “by being placed near adynamo orother powerful magnetic apparatus;orevenbyvibrationoftheirparticlesduetoknocks, especially when inamagnetic field. Iwould askMr.Hookhan whether there isany provision toprevent thecustomerorother personfromtampering withthestrength ofthemagnetofhis beautiful meter? Isthe case which covers the instrument whea inuseofstout iron, and isitdifficult ofremoval? Mr.S.Joyce:Imeanisitprotectedfrominterference onthe “part ofthe consumer? ‘Mr. G,Hooxmam: But would you lethim takethecoverof ornot? ‘Mr.8.Joyce: That israther aquestion ofwhat arrangement you make. ‘The Presiwet: Itwould beavery clever consumer that would increase thestrength ofthemagnet, andifhediminished ithewould have double topay. Naum, MrG.HooKnam: IwassorrytoseethatMajor Cardewbad left the room when Iheard Mr. Crompton’s remark, because, except myself, Ithink hehasmade most experiments onthe meter, and knows best that itmeasures infavour oftheconsumer with emall currents. We have tested hundreds ofthem, andwe find thisthecase with every single meter, andontheoretical grounds itmust beso:there isacertain strength ofcurrent whet thearmature does notmove atall,and theconsumer getsbis current fornothing. ‘The permanency ofthemagnet isofcoarse ‘aquestion ofthegreatest importance, and Ishould liketohare given amuch longer explanation than Ihave indisposingofit to-night. Ishould like toexplain how the magnets aremsde. This type ismade with agood many tums ofwire, and asatu ratingcurrentisflashedthrough them,Wethentakethespeed, and then this ishammered with aheavy weight ofcopper,st much astheconstruction ofthemagnet willstand, and.we cua rarely reduce thefield more than onepercent., owing tothelor resistance oftheairspace. Wethen byareverse current redace itabout 10per cent.;afterthatthetendencyofthemaguet isupwardsforaconsiderabletime.Whatthehistorymaybe| 2] DISCUSSION. os afterwards Icannotsay,butthetendency ofthemagnet istoms. strengthen avery little forweeks after that treatment. Iwould akMr. Joyce ifhehasever had experience ofapermanent ‘magnet thestrength ofwhose field rises after hehasperformed hisexperiment onit? Mr.8.Joyce: Yes; inthe calibration ofour magnets they areworked down, andarenever sent outwith their fullstrength, andweoften notice that after weakening amagnet ithasa tendency toriseinstrength. Mr.G.HookmaM: Itwould, nodoubt, ifitwas very much ‘weakened ;butifonly weakened 5or10percent. would that be so? Ithink Professor Hughes found that amagnet reduced to zeroby@reverse current rose instrength ontheremoval ofthe current, and continued torise. Inpractice, after 18months’ testing, wehave never found thesmallest. perceptible variation, and the severest mechanical treatment makes nodifference toour magnets, Mr.B.W.Suir: Idonotquite seehow itdepends onthe strength ofthefield. Mr. G. Hooxnam:Thestrongerthefieldthesloweritgoes. Ahearty vote ofthanks was unanimously accorded toMr. Hookham forhiscommunication. : Aballot took place, atwhich thefollowing candidates were elected -— Member: Leonard William Holmes. Associates : Roland Chambers. Allen F.Scott. - Perey B.Crowe. George Crosland Taylor. Capt. W.F.Hawkins, RE. James Taylor. Frank B.Lea. John Williamson. Dr.Henry Leipmann, F.C.S. Student: Herbert Carpmael. ‘The meeting then adjourned. . 696 ORIGINAL COMMUNICATIONS. ‘Tax Pourrecmntc, Booasxocm, ‘Slat Mey, 1689. Dean Sm,—As theinformation may beofsome interest o members, Iquote theresults oftests ofmylightning conductoratDhubri,inAssam. ItiswhatIcallmypattern,orthe “Arborial” Lightning Conductor, made ofeight No.9B.W.G. wires, worked upinto cable with topbranches, consistingof the wires themselves and along taproot with rootlets ofthe same. Icannot laymyhands onthedifferent measuremett atthepresent moment, butconsidering that myhouse stand: onadry gravel site, theresults obtained bymymethodof dealing with theearth, orroots ofthe conductor, leave litle room forimprovement ifminimum ofresistance istobeaccepted fasthetest ofagood conductor; and the fact that although standing inthemost exposed part onthebanks oftheriver Brahmaputur, and that the supports ofthe roof consist ofir posts, forsomany years, and amidst themost violent thunde- storms, the house has never been struck, though built ort thirteen years ago:— 15th May, 1878—Resistance, 11-92 B.A. units, 27th » ” ” 3564 yy ” IstJune, yy » 220 4» 12th , » » 4B yy 20th 5, ” ” 403 » 21st October, ,, ” 540 » 30th yy ” ” 235 ” 28th February, 1879 » 549 yy 4thMay, » » Bt yy FromMay,1878,toJuly,1879,alltheotherresultswere 3+,theabove figures being theonly variations ofnote inthe| fourteen months. The first result, 11-92, wasduetotheleading wirefrommyhousetothetelegraphofficetesting-roombeisg| ORIGINAL COMMUNICATIONS, oor No.12B.W.G., which waschanged toNo.1 B.W.G. Leaving outthis exceptionally high resistance explained bythe high resistance ofthelead, theaverage of29tests, giving two per month forthefourteen months, Iget3-085—even thiswasdue tothedistance—and thetemporary natureofconnections, when tested atthe house, the resistance was °35! Itwould beinteresting toknow from other members ifthey canimprove onthis, orbeat myrecord, asIlaid outover 60 rupees (£6) incompleting myearth trenches, é&e.,but Idoubt tot,under allthesurrounding circumstances, itwasmoney well laidout. Like thelate Dr.Mann, inAfrica, Ifeel some interest inmyIndian Monument toScience. Ttwould alsobeofgreatinteresttoknowtheexperienceof members indealing with linear vereue other forms ofearth for both lightning andtelegraphic purposes. Ihelped thelateMr. Schwendlerinsomeveryinteresting experiments inconnection with this important branch ofthesubject, when, bytheuseof what hewasgood enough toterm “McGregor’s exploring wire,” vesought andfound spots ofleast resistance. Some dayIwill beglad tosend you anaccount ofthese experiments. Meanwhile Ihope theEditing Committee will consider the present communication deserving ofacorner inourJournal. I onlyregret,andregretmostsincerely, thatthestateofmyhealthprevented mytaking part inthelate discussion onProfessor s st,interestin reToige'sma etingPaper. ‘W.McGREGOR, Member. ‘The Secretary, Toaitation ofElectrical Engineers. REPORT OF AFATAL ACCIDENT FROM LIGHTNING, IN INDIA. ByP.V.Luxe, Member. Axnatoon Arcax, April 12th, 1889. Isend yon anaccount ofafatal accident from lightning, which you may consider ofsufficient interest topublish inour Journal. Itisincomplete, insofarthat Iamunable, from there notbeing anyoffices oftheTelegraph Department supplied with om ORIGINAL COMMUNICATIONS. testing instruments near thespot,togivetheelectricalresistane ofthe“earth.” Itmay,however, beinferred, fromthefacttht‘thecircuitalwaysworkedwell,thatatanyratetheearthws fairly good one. ‘The accident happened ataplace called Gnatong, inSikkin, thepresent terminus ofafield-telegraph lineputuplastyer, from Darjeeling, inconnection with the Sikkim Field Fore operating against the Thibetans. The telegraphic office is probably thehighest intheworld, being situated atanaltitaie of12,200feet.ThefortofGnatong, inwhichisquartered awing of«British regiment, isinahollow, surrounded byhills at» great distance onallsides. Atthe bottom ofthe hollow iss Inrge pond, frozen over during thewinter, affording gaoi skating and sliding forthesoldiers. ‘The fortisbuilt onrising ground just above thepond,and thetelegraph office isinthefort. The building isofwood, with stone foundations and stone ends;aroofofwoodenshingle with azine-covered ridge. Half ofthebuilding isused a5 telegraph and half as8post office. The rubble masonry per tition between thetwocontains thefireplaces and chimneys, an onshelves over thefireplace inthetelegraph office thebatteries ‘areplaced tokeep them from freezing. The soilonwhich the building iserected isrocky, and thefloor oftheoffice isformed oframmed earth and stones. x___ - ____, >ae i AlBe 38. 2 Teregraph Signallers# ie s*lE3Bedroom i Stores,spare cols | u wisetthics bonesse,_{ Fo. 1. ‘Theabovesketch(Fig.1)showsroughly thearrangement ofthe ORIGINAL COMMUNICATIONS. ew rooms. Ontheinstrament table, shown enlarged inFig. 2,Ais thelightning discharger ;B,theinstrument; andC,acallbell. 5oc. é 2 . O I f€ Fro. 2 The lightning discharger isamore orless rough one, used only infield equipments, theobject inview being lightness. It consists ofthree brass plates about 1in.wide mounted ona wooden board 5in.x4in., thus— ‘The instruments are relay, sounder, and key mounted on one board, with the usual connection forworking direct orin translation. Atabout 9p.m. onthe 26th March, 1889, two military signallers, named Clarke andHewett, were intheoffice, andatthe moment theaccident occurred they were inthe position shown inFig. 2. Clarke wasstanding atD,withhiselbowonthetable,bending overtherelay. Hewett wassitting atE,withhishandonthe 10 ORIGINAL COMMUNICATIONS, key, calling upthestations onthe line toseoifthey were clear before closing forthenight. When theflash came ittook small piece outofthebrass cover oftherelay, at(a),andstrack Clarke, who waskilled instantaneously, falling onHewett, whose legs,bythesameflash,wereparalysed. Atthetime,thelight(ahurricane lantern)wasextinguished. Theonlymarkonthe man killed wasaround fing ontheleftcheek, which appeared thenext morning, when the body wasvery much swollen, with signs ofbleeding from thenose and mouth. ‘There wasnosign ofactual scorching orburning; and the medical officer wasof opinionthatdeathwascausedbyashock,andnotbytheman's being struck byaflash. The man Hewett isslowly recovering theuseofhislegs, which were temporarily paralysed. The “earth” connection was made toacoil ofwire buried in theearth just outside theoffice. ‘The same earth was used for thelightning discharger and forthe instrument. Neither the lightning discharger northekeyshowed anydistinct signs of« discharge having passed. ‘The only visible damage done wasthe small piece ofbrass taken outoftherimofthecover oftherelay. Theaction ofthelightning inthis case appears tobequite unaccountable, Admitting that theflash came along thelineintotheoffice,andthatthelightning discharger failedtosct, why should Clarke have been struck inpreference toHewett, who had hishand onthekey? ‘The line isaninsulated one, supported ontrees andwooden poles, running over great altitudes; and thunderstorms inthat district arefrequent. Ontwo previous occasions military sig- nallers working intheGnatong office had received severe shocks ‘There seems nodoubt that thelinewasstruck and thelightning conducted into theoffice, andthat thelightning discharger failed toact; buttherest seems inexplicable, The relay coils were notfused, noranydamage done beyond thepiece chipped outof therelaycover. P.V.LUKE. “The Secretary, TratitationofElectricalEngineers. ORIGINAL COMMUNICATIONS. 08 NOTES ON ASTATIC ELECTRO-MOTOR. ByCuanies ZirERNows«t, Thewell-known fact that bodies charged with electricity of thesamesignrepulse eachother wasfirstemployed byFranklin (in1780)toproduceacontinual rotation, Heusedahorizontalheel, consisting ofhorizontal glass strips provided with copper lullsattheir ends; thewheel waspivoted, andcould revolve between the ballsoftwooppositely-charged Leydenjars.Itsown lullscouldthusreceiveachargeofthesamesignasthatofthe neybeing,therefore,repulsedbythelatter.After180degrees ofrevolution each oftheballs discharged before theoppositely- charged Leyden jar,and,atthesame time, took away partofthe lctricity summed upupon thelatter. Anewrepulsion, there- fore,ensued, and soon, Franklin called this apparatus the “electric turn-spit.” Later on,Poggendorf showed that apivoted diseofglassor ebonite begins torevolve when charged with quantities of electricity ofopposite signs bymeans oftwo oppositely-placed combs ofpoints, provided that itreceives animpulse inany direction, Poggendorf also caused aninfluence machine of Holtz’ torevolve bycharging itscollecting spikes byanother influence machine. Inorder tostudy theso-called static actions ofhigh tension altemating currents, Ihave, amongst other things, constructed ©small rotatory apparatus, derived from ‘Thomson's quadrant electrometer, intheidiostatic form invented byJoubert. Ihave sueceeded inmaking this apparatus revolve speedily with ahigh tension alternating currentaswellaswithahightensioncon- tinuous current. Intheformer case, thespeed was sohigh that thenumber ofrevolutions could nolonger becounted bythe eye. Inthe latter case, Itook thecharge from abelt between theworking motor, ahigh speed engine, andthetesting room of theelectrical department ofMessrs. Ganz &Co.ofBudapest. Ofcourse, thespeed wasstillahigher onethan with 2,000 volts ofalternating eurrent, ‘VOL, XVIII, 48 ree ORIGINAL COMMUNICATIONS. ‘The figure isasketch oftheconnections inthe two abore- named cases. ey| y ~ | a, aecal axRieah5!| : | ee 4: ||“ey % | . — ~ fens 7 ‘mn Themovable partoftheapparatus consists oftwocouples of aluminium sectors, shown asquadrants AA’, BB’,forthe reason | ofsimplicity. ‘The fixed part consists offour double sectors of brass embracing themovable ones. They arelikewise repre-| sented asquadrants @C’, DD’. ‘The movable part is,besides. provided with acommutator offour parts, from which the two couples ofsectors AA’,2B’,arecharged, bymeans ofpoints SS. with electricity ofthesame sign asthat oftheopposing fixed couples ofsectors. Connecting the terminaly KK’ to.those ORIGINAL COMMUNICATIONS. 108 8S’ ofthe transformer 7,the lowtension ofthe alternate-current dynamo Mwill betransformed into high tension, and theappa ratus will start, provided that themovable sectors arenotexactly opposite tothefixed ones, Ifyouconnect oneoftheterminals tothecollecting spikes, and theother toearth, theresult willbe the same. Itisworth remarking that, with thesecond arrangement, an instantaneous charge ofthesectors willsuffice, asthefixed sectors donotchange thesign oftheir charge intherirst ease, however, acontinuous charge isrequired, astheelectricity must change sign simultaneously inthefixed andthemovable parts. Itisnot impossible that this apparatus, insome form, may be adapted forpractical purposes, such asavolt-compteur forseries working arc lamps; asrecording apparatus forearth-cireuits, or forthe variations oftheprimary tension incentral stations; as motor for very high tensions, &e, Boparzsr, June 2,1889, 708 THE LIBRARY. ACCESSIONS TO THE LIBRARY FROM APRIL 1‘TO SUNE 90, 1589, (Werksmarkedthus(%)havebeenpurchased.Ofthotenotpurchasedores“achange,sohers thedonors havesarenotventheworkshaceBeenprs bytheauthors.) men es ranniculanay DEsinAaU THAT MEMBERS SUOULD FRRAENT cortES of1m. "Wonks 70 THE LIBRARY aa SOON 43 POMIBLR AFTER PUBLICATION, Bonnett [A.R.] Electric Traction. Sm, dto. 13pp. (ast ofSasha! ‘Engineering Association, Session IV. Tenth Meeting.) Edinburgh,188 Borly J.A.) Universal Electrical Directory and Advertiser, 1889, Ta nen directly or intirectiy connectedwithBice odtheNamevand.AddreamaofManufsccarers in)Great‘Bras. metic, heContent yo A “Landon8 (esebted byMess, W.Dawson &Sons (Pebliakes)Blakesley (T.H.]PapersonAlternating CarretsofElectricity,frtheUe etelcateaud Reyincere SodHalon tvon129psLond8 "EEresentedbyMess.Whittaker&Go,(Pabiiben) BoardofTrade,[VideMasindin,MajorF.A.}Cundell [MajorJ.P.]ADictionary ofExplosives. Svo.108pp.leche, 18 (PreventedbyMajorFranciaJ,Dap,RE] *Douglass (J.C.]_ AManual of‘Teleaph Construction: theMechasctMBean ElecTelegephEngineering” todEditiones405g Tondo 27 Institution ofCivil Engineers, MinutesofProceedings. VolsXCV. ROVE.oro.todriant10pp. Linden1883 Institution ofMechanical Engineers, Proceedir 1889, No.1. ‘Santary. 80, 201 pps oeade,888(Exeter ‘Ttalian Tel Relazione Statistica miTelegrafi delRegvo dTtalisllsinoPasian18s?86)Eoccaa‘ions,1083{PreseosedbytheMinisterofBoraandTetegrapi,Rome] amos [Charles Streatfeld]. ‘The ‘More Signalle's Companion Fo“app. 3 Ext,18Marina (no: FA}, Report heBonedotTradewiheeiiryHeldandertheElectricLigtingAct,1862and188gforerat‘Provislonal OrdersaudLicenses,datedMay9,1889.Fo.28PR“ain 1858 [Prevented byMajor P.Carlow, BE) Rust [Arthur]. Electricity: theoretically and tically considered.Uythe [rorThermasEtec eraaSppsPMTs coi 705 LIST OF ARTICLES ELECTRICITY AND MAGNETISM ‘Appearing insome oftheprincipal Technical Journals during theMonths ofMAY and JUNE, 1889. 1—BATTERIES AND ACCUMULATORS. P,Canovsrcnorr andA.Srranxor®—Electro-motive ForceofBatteries—C.R,vol 108, p.987, 1889,=roxMULten—BehaviouroftheZineElectrodeinManganese Batteries— ElZeit v0L 10, 294, 1889, W,Komnaavacu and C.Ietx—Experiments onAecumalators—EI.Zeit, tol. 10,p,200, 1889, 1—DINAMOS AND MOTORS, A.Wirz—Pole-ReverslsinSeriesDynamos—C. 2,vol.108,p.1243,1889, W.C, Recuntawaxi—Alterate-Carrent Motors.—Jaun. El, vl.8%p.90, 1389, ©.Retoxten—Secondary Induction Phenomens inDyuanoe—Lum, £1,vol9% p-401, 464, 1889. 6,RicuanD—DetilsofDynamoConsruction—Lum. El,vol82,p52,188. E,Mertax—Rechniowshi's Dynamo.—Bull. Sor,Jitvol.6,p.246,1°90, UL—ELECTRO-CHEMISTRY AND ELECTRO-METALLURGY. 4H,Pecuat—Limit between Polarbation and Electrolysin—C. R, vol. 108, p-1288, 1569, ‘A.Maxxr—Introdaction tothe Study ofBlecto-Chemietry.—Lum. El, vol.92, pp.322 427, 47,90,878, 1089. PH Lapsoxn—Electo-Metallargy ofCopper-—Lem. £1, vol.82,p.851, 1889, C.E,Gunttavwe—Electrolysis prodaced byMinimam Electro-motive Force.— Lam. El, vol. 82,ps558, 1880. Dr.,Eauaxo—Electro-Metallargy ofAluminiam—E’, Zeit, vol10,pp.290, 262, 1889, IV—ELEOTRIO LIGHT. C.Canné—Comparson ofdiferent Methods ofllumination—Lum, Ei,vol.82, 286, 1482,W.peForvinuun—Electrie LightforBalloonSiguals—Lim.£2,vol.82,S44, 1869, E.Drevpomné—Electric Light atthe Paris Exhibition, —Lum,El.,vol.82, .B51, 1889, 06 ARTICLES RELATING TO ELECTRICITY, Exe, W. neFoxvisnie—The Projector onthe Eifel Tower.— fm, Kl., oh3%, p.391, Les,A.Venxia— ElectricLightingofthePalateRoyals—Hall,Sie.Intyvl6 264,182, (©.vowMitsas—Rerlin CentralStations.— ET,Zritysol,10,p.25,1982, V.BLECTRIC POWER. G.Rictanp—Electric Rallways-—Lam. El, vol. 82.p.207, 1889. E,Dincoowwi-—Electrie Lift attheParis Exhibition —Lom. El., vol.32,p.316 1849, VI-MAGNETISM AND ELECTRO-MAGNETISM. CR. Cuoss and A,S,Wrutrans—Strength ofthe Induced Current with# Magneto-Telephoue TransmitterasTufluencedbytheStrengthoftheMaguet— Phil. Mary 8ole%7y Ye392, 1889. 4.Panzer—Diamagnetism and Concentration ofEnergy.—Phil. Mag., vol 403,158 ML,Becquemer—Effect ofTerrestrial Magnetism onAtmospheric Polaristica— ©.Re,vol.108,p.997,1889. F,Mancitaxo—Necesity ofIntroducingaCorrectionforHamidityinMaguew- meter Observations—C.22vol.10%,p.1001,1889. —Cuavvin—Rotatory Magnetic Polarisation inIceland Spar—C. R,vol105, p.1097, 1889, J.Luvist—Terreatrial Magnetism andSun-Spots.—Lum. El,vol.82,p.31218F,Lannogur—Re-maguetism ofIronafterHeating.—Lum,El,vol.32,p,369,188. ‘VIL.-MEASUREMENTS AND MEASURING INSTRUMENTS. Dr. G.Gomr—Determining theStrength ofLigalds bymeans oftheVolui Balance—Nat,,vol.40,p-98,1889. B,vanArats—Nesinance ofBismuth—C:2,vol.108,p.1102,1889. G,Ricaanp—Horve-Power Dynamometers.—Lum, El,vol.82,ps260, 1889, E,MevLax—New Types ofGalvanometer.— Lum.£1,vol.82,p-268,1889, —Guiimix—Apparatus forMeasuring theResistance ofEarths.—Lum,EI,vol, p.876, 1888, F,MétorreandG.Hexnanp—Measurement ofPermeability. —Lum,Fi,vobpe415, 1889. ©.E,Geitiarax—Use ofPlatinnmdridiam and other Alloys forStendan Resistance Coils—Lam.El,vol.82,p.451,1889. K.Warrz—Method ofMeasuring HighPotentialDifferences. —dAvn.,vo.37.89)1889. K,Sraxcxen—Measurement ofSelf-Induction bymeaus oftheTelephose= El,Zeit, vol.10,p.289, 1880, ARTICLES RELATING TO ELECTRICITY, Ere. w07 ‘VIIL—RAILWAY APPLIANCES. 6.Lanwoven—Flamache's New Block System.—Jum, EI., vol.82,p-459, 1899,M.Cowwasn—Appliances ExhibitedatParis.—Lum. El,vol.32,p.556,1889, IX-STATIC AND ATMOSPHERIC ELECTRICITY. EL, Trocvaror—Duration ofLightning Flashes.—C. R.,vol. 108, p.1246, 188, ML.Wor—Disruptive Discharges inGases.—Ann., vol.87,p.206,1889.J.Eusren and H,Grirai—Method ofDetermining theElectrical Nature of Atmospheric Discharges.—-Brill,, vol.13,p.327,1889. SAunuesics—EflectoftheSun'sRayeonAtmosphericElectricity.—Beill, vol, 12,p.828, 1889. . X—TELEGRAPHY AND TELEPHONY. 4H,Wererarex—Mierophonic Contacts andTelephonic Currents.—Lum. El, vol. 82,p.272, 1889, P.H.Lepenoxn—Zigang's Electro-maguetic Telephone.—Lum. El., vol. 32, p.820, 1889, E.Zeracus—New Multiple Switch-Boants forExchanges.—Lam. El.,vol. 82,p. 458, 1889, 4.Kaneis—Stockholm Telephone System.—£l. Zeit, vol. 10.,p.244, 1889, Dr.A.Tomuen—Wheatstonc’s Automatic System.—E1, Zeit., vol. 10,p.266, 1889. ©.Gaaw1xae1—Joints inBronzeWiresEI.Zrit.vol.10,p.293,1889,C.Grawixxet and K.StaecKen—Berlin Central Telegraph Ofice—E0. Zeit, vol. 10,p.296, 1889, 4.Suck —The Hughes Apparatus arranged forUsewith Alternating Currents. — El.Zeit, vol. W,p.BL, 1889, ‘XL—THEORY. A.W,Wanb—Magnetie RotationofthePlaneofPolarisation ofLightinDoublyRefracting Hodies.—Nar., vol.40,p.117, 1889, 4.Brows—Helmboltzs Theory ofMercury-dropping Electrodes, andthePotentialDifference betweenCleanMercuryandElectrolytes. —Phil.May.,vol.27p.ast, 1889, HA. Rownaso and C.I,Hurcitissox—Electro-magnetic Effet ofConvection Currents. Phil. Moy vol. 27,p445, 1889. De.0,Lovar—Blectrostatic FieldproducedbyVaryingMaguetieInduction.— Phil. Mag., vol.2,p.469, 1889. ¥,Cunovsrcnorr—Conduetivity ofSaline Solations.—C: R., vol. 108, pp.1100, 1161, 1889, 708 ARTICLES RELATING TOELECTRICITY, Ere. A.Srozerow—Aetino-electric Phenomens.—C. f,vol.108, p.1241, 18 A,Portrn—Seat oftheElectrosnotive Force ofContaet.—Joer. Ply ap 225, 1880,—Gour—Comervaion ofElectricityJourPlyn,vol8,p-220,1880D.Koupa—Action ofLightonSelentum.—Jour. Phys,vol.8p.282108 B,Browar—Actino-lectric Phenomens.—Jour. Phys, vol.8,ps245 1882 C.Drcuanwy—Diflerences between the so-called Positive and Negative Bee teicites—Lum.£1,vo.82,ppe218,815,66,421,475,619,56,160. PP,Sancvx4-—Gaston Plant's Rheostatic Machine—Lam. I, vol 3p36, 18J.BieremandH,Gurr21—Production ofElectricityhyContactofGuwiGlowing Wires —Ann, vol.97,p.815, 1889, XIL—VARIOUS APPLIANCES. G.Ricnanp—Phonogeapbs.—Lium. 1,vol.325p.906,1859. G.Ricuan>~ Graphophones-—Lus. £1, vol 3,p.388, 1889, E,Mertan—Recording Thermometer-—Lov.I.vl.82,p.61,188. P.LeGoaxtov—Voting Apparatus—Lum. £1,vol.32p.668, 1889, ov rm Gnstitution ofElectrical Engineers, Founded 1871.Incorporated 1883. Vou.XVI 1689, No.82 ‘TheOne Hundred andNinety-seventh Ordinary General Meeting of the Institution was held at the Institution of Civil Engineers, 25, Great George Street, Westminster, on ‘Thursday evening, November 14th, 1889—Dr. J.HorgiNsox, F.R.S., Vice-President, intheChair. ‘The minutes oftheprevious Ordinary General Meeting, held onMay 30th, 1889, were read andconfirmed. The names ofcandidates for admission into the Institution were read andordered tobesuspended. The following transfers were announced ashaving been approved bytheCouncil :— From the class ofAssociates tothat ofMembers— Reginald John Jones, From the class ofStudents tothat ofAssociates— ‘Harris Henry Eley. ‘Donations totheLibrary were announced ashaving been received since the last meeting from theInstitution ofCivil Engineers; LaSociété Frangaise dePhysique; theAstronomerRoyal; Dr.Friedrich Goppelsroeder; Messrs,Buck&Hickman ; YoU. Xxvitt 49 710 LIGHTING OFCENTENNIAL INTERNATIONAL [Word ‘Messrs. Saxby &Farmer; Walter R,Skinner, Esq.; Monsiou Georges Carré, Publisher;Messrs.Dawson&Sone,Publisher; TheElectrician Publishing Co.; Messrs, Griffin &Co.,Pub lishers;Messrs.Whittaker &Co.,Publishers; Professor Dr.K. Ed. Zetzsche, Foreign Member; John Aylmer, Member; Dr. J.A.Fleming, Member; W.J.Hancock, Member; H.Gran Harris, Member; 8.H.C.Hutchinson, Member; ProfessorA Jamieson, Member;Dr.C.Lemon,Member; Professor OliverJ. Lodge, Member; and W.P.Maycock, Associate; towhom the thanks ofthemeeting were heartily accorded. Thefollowing paper wasread bytheSecretary :— THE LIGHTING OF THE CENTENNIAL INTERNATIONAL EXHIBITION, MELBOURNE, 1888 AND 1889. ByK.L.Munuay, C.E,, Telegraph Engineer totheVictorimn Railways, Member. InNovember, 1887, tho Executive Commissioners ofthe Centennial International Exhibition, which ithad been decided tohold inMelbourne inthefollowing year, consulted measto thepracticability, electrically and commercially, ofusing lec- tricity asameans oflighting theExhibition; andIfelt thatI could assure them that theelectric light wastheonly onewhich could bedepended ontoefficiently illuminate thebuildings and exhibits, andthat itscostwould notrender itsuseprohibitive. Irecommended dividing thebuildings into three sections and calling forseparate tenders forsupplying and maintaining thedynamo machines and lamps necessary tolight each part; tenderers being requested also tosubmit analternative offer fir thewhole lighting required, Mydesire wastohave each section lightedbyadifferentsystem,ifpossible, Finally, Iwas entrusted with the duty ofdrawing ups specification, andtenders were invited. ‘Thosereceivedallnamedsuchahighpricefortheworkthit Icouldnotrecommendthesoceptanceofany;but“ 19883 EXHIBITION, MELBOURNE, 1888axp1889, m requested bythe Exhibition “Commissioners toarrange with oneormoreofthefirmswhich tendered,Iwasableeventually todoso,and@contract wasentered intowiththeAustralasianHectrioLight,Power,andStorageCompany forthelightingofthegreaterportionoftheExhibition buildings; andanagreement vasmade with Messrs. Ganz &Co.tosupply transformer plant 1ndlight theArmament Court, theofficial reception and dining rooms, and offices. ‘The Australasian Company’s contract was tosupply, fitup, sndmaintain 825 Brosh arclamps, 37Brush arodynamos, 7 Brush Victoria dynamos, compound wound, and 2,000 incan- descent lamps, with allrequisite switches and fittings. The whole ofthe machinery and other apparatus was tobeofthe latest pattern andbest workmanship. Asusual insuch cases, thebuildings were ultimately extended considerably beyond thelimits originally laiddown. ‘Thelighting plant had, ofcourse, tobeincreased proportionately, sothat the apparatus actually supplied bythe Australasian Company con- sisted of,asunder :— 40Brush aredynamos, known asNo. 71,each capable of giving acurrent of10amperes tofrom 23to25arc lamps inseries. 1Brush aredynamo, known asNo.81,professedly capable ofgiving current to55arclamps inseries. 1Brush arc dynamo with cast-iron armature, known as No. 6,supplying current of20amperes for5arc lamps. 7Brush Victoria dynamos forincandescence lighting, com- pound wound for110volts atterminals, 937 sixteen-hour round-type Brush arclamps, with alabastrine globes;5Brockie-Pell 20-amperearelampsforlighting the dome;2Castlearclamps, 1,900 Swan lamps of16candle-power at100volte; 27“sun- “beam”lamps,ofsizesrangingfrom200to600candle- power each at100volts, Allnecessary switches, fuses, and leads, thetotal length of thelatter being over 60miles, 4‘LIGHTING OFCENTENNIAL INTERNATIONAL (er,uh, The71Brusharedynamoswereoftwopatterns.Ist,thit inwhich thebed-plate and field-magnet yokes arecast inove piece; thefield-magnetcores,withpole-pieces,being,asusualin thisformofmachine, ofcastiron,fastened totheyokeswith bolts, The commutator brushes and terminals are mounted on 8slabofslate,whichisfittedintoarecessleftforitinthebed-plate casting. The other pattern ofarcmachine haswhat maybetermedacomposite bed-plate, ornobed-plate atall Inthese machinesthetwoficld-maguet yokesarecastinseparte pieces, the cores, with pole-pieces, being, asinthe former pattern ofmachine, bolted totheyokes. The two horseshoe magnets thus formed areconnected bytwo pieces ofrailmy iron, which constitute whatever ofbed-plate themachines hare The commutator brushes and terminals are mounted on ashb ofwood fitted between thelongitudinal pieces ofrailway iron. The first class ofmachine will hereinafter becalled “solid- “frame,” thelatter “rail-pattern” machines, ‘Themajority ofthesolid-frame machines hadpreviously beet usedattheAdelaide Exhibition, butwerefittedwithnewlami nated armatures for use inMelbourne. The balance were seat outdirect from England. ‘The framework ofthe8dynamo ‘wasimported, butthelaminated armature wasmade and allcoils were wound intheAustralasian Electric Company's workshops Melbourne. This dynamo was ofthe solid-frame type. The No,6dynamo wasalsoofthesolid-frametype,buthadacast- ironarmature. Itscoilswerearranged foracurrentof20 amperes, ‘Two types oflaminated armature were supplied with the No,7dynamos—one having plates ‘05ofaninch inthickness theplatesoftheotherbeingonly022ofaninchthick,The former Iwillcallthick, andthelatter thin plate armatures ‘Thedynamosforincandescence lighting,seveninnumber,wert oftheBrush Victoria pattern known asNo,E2, The nominal output isstated tobe366 64-watt lamps atanaverage arms ture speed of720revolutions perminute. The machines were@ thenewest type, andhadvery massive magnetic circuits through- out,Thearmatures arewoundwithonlythreelayersofrec 9] EXHIBITION, MELBOURNE, 1888 axp 1880. ns angular wire, each section consisting ofthree convolutions. The laminatedironcoreintheVspacesbetweenthewindingsisnot orered with insulation, sothat inthese parts itisdirectly exposed tothecooling influence oftheair. ‘Thearclampswereofthewell-known Brash16-hourround type, wound with the low-resistance shunt recently adopted by themakers. Alabastrine globes were used. Inthemain annexes thearclamps were arranged inadouble rowdown each bay, one lamp being allowed forabout every 1,200 square feet offloor space. The bays were about 50feet wide, and thedistance between the tworows oflamps was about 25feet, the height ofthe globe from the floor being about 22feet. Incandescent lamps of16candle-power at100 volts, to thenumber ofabout 1,900, were used forilluminating thepicture galleries, beneath the galleries, intheaquarium, and inthe cellars. ‘The concert hall waslighted by25 sunbeam” lamps of from 200 to600nominal candle-power. The incandescent lamps inthepicture galleries weresetinasinglerow,9inchesapart,omreflecting frames,thereflecting surfacesofwhichwerepaintedadead white, These frames were about 18 inches wide, made of ©" sheet metal, with a ‘rosssectionasshown. Belowthelampan- other reflector was ump Placed, theprimary objectofwhichwasto_shadetheeyefrom thedirectglareofthe lampsandkeepthe for comparatively dark. With theabove arrangement ofre- fecting frames thewalls were theonly portions ofthegalleries illuminated bythedirect light from thelamps, and thetotal absence ofglare allowed ofavery pleasant restfortheeyes. The switch-board forthe arccircuits consisted oftwoparallel mails ofwood onwhich ordinary double brass binding screw terminals were fixed. The screws on the lower rail were con- nected tothe dynamo terminals, while those onthe upper rail formed theterminals ofthevarious lamp circuits. ‘The positive terminals ofallmachines and circuits were atone end oftheir respective rails, and thenegative terminals attheother. The 714LIGHTING OFCENTENNIAL INTERNATIONAL [Nor.us, ‘connection between thedynamo andcircuit terminals waseffected bymeans ofshort lengths ofwell-insulated cable. ‘The incandescence switch-board was mounted on'awellfinished cedarstand.Itconsistedoftwolargeslabsofblackenamelled slate, onwhich the nickelled switches and terminals were fastened. ‘Between the two slabs ofslate asuitable recess was formed inthe woodwork forthereception ofaCardew voltmeter (Goolden & ‘Trotter's make), Inthecentre ofacedar panel below thewlt- ‘meterasmallswitchwasplaced,bywhichanyoneofthemachines could beconnected tothevoltmeter byturning ahandle. Suit- able terminals were provided fortheintroduction ofsafety fuses, ‘and also forenabling themachines tobecoupled inparallel. ‘When socoupled anyoneormore machines orcircuits could be switched onoroffwithout affecting therest. The arecircuits mainly consisted of7/168 stranded cable nigrite andtape insulated, allother being ofthesame sizeof cable, butinsulated with rubber and braided. ‘Theleadsweresoarranged thatbetweeneverytwolampsgetting current from thesame dynamo, there were atleast two lamps getting their current from some other dynamos;myobject being tomake certain that thebuilding should befairly well lighted, even ifseveral dynamos broke down atthesame time, ‘Allleadsandreturnsraninonedirection fromthedynamoromtothemain buildings. Butonentering these buildings thelesds andreturns were taken intotally different directions totheter- minal lamps oftheir respective circuits, Bythis means, and careful attention tothearrangement ofthecircuits, theleads and returns never approached each other, except inthedynamo room. Consequently, short-cireuiting inthebuildings was almost im possible. The lamp circuits were runonthewoodwork which carriedtheroof.Thelampsweresuspended bycordssothat they could belowered forcarboning orother purpose. Thewires from thelamp terminals were taken upthesuspending cordsnd lashed toitatapoint afewinches from thepulley when the lamp wasatitsproper height. They were then taken tothe ‘woodwork andfastened toitatsuch aheight thatthelengthof wireformingthecord,togetherwiththatfromthepointof| We} -—=«sRXHIBITION, MELBOURNE, 1888 axp 1869. ns lshing tothesuspending rope, wasatleast equal totheheight boretheflooratwhichthewireswerefastened. Thismethod permitted thelampstobeloweredtothefloor,andatthesame timeleftnocablehanginginfestoonsaboutthebuildingwhen thelamps were intheir lighting position. From theterminal lump ofeach cironit tlewires were taken down thewooden pillars which supported theroof, through thefloor, and(bracketed totheflooring joists) tothedynamo room. ‘Theindandescence circuits were arranged foralossofpotential of10volts. ‘The installation was looked upon asbeing of@ temporary character, consequently the electrical loss was con- tiered tobeoflessmoment than incurring larger firstoutlay forheavierleads.Evenwiththeabove-named losstheleadsfor ‘portion ofthedistance consisted oftwocables, each containing 37/l4s; and fortheremainder ofthedistance tothepoint’ of distribution one cable ofthe same size was used. The distribu tionwas effected bymeans of7/16s cable, andfinally bysingle wire. Allcables were insulated to150megohms permile. ‘The steam was generated in12multitubular boilers, the pressure being maintained at120Ibs.persquare inch. Three pairs ofhigh-pressure engines were used, having gylinders 20inches indiameter, with a40-inch stroke. Normal speed, 80revolutions perminute; andthetotal indicated horse- Power with fullworking load wasabout 1,580, These engines vere made inMelbourne, and were similar tothose used for working thecable street tramways. Imay sayhere that, recog- ising thevery great importance ofhaving thevery best driving machinery, Iurged the Exhibition Commissioners tosend to England forengines specially constructed forelectric light work. Itwasfound, however, that time would notpermit ofthis, and amorder was given toafirm oflocal manufacturers who had ‘engines partly made, andcould guarantee, therefore,tohavethem ready intime. ‘Thie countershaft was inone continuous line, and measured 202feet 6inches over all. Itwas driven byropes from the ‘engine fly-wheels, intheface ofeach ofwhich 13grooves were formed forthereception oftheropes. ‘The shaft was inthree 116. LIGHTING OFCENTENNIAL INTERNATIONAL [Wer. Ua, equal sections, one foreach pair ofengines. These sections were connected bymovable clutches, sothat they could ram either separately ortogether asrequired. Toreduce tothe lowest: degree possible theinconvenience which would result if« pair ofengines broke down, thedynamos driven byeach section supplied current tolamps distributed over thewhole buildings; sothatifone-third ofthelightwastakenoff,nopartofthe Exhibition would beleft indarkness, ‘The main body ofthe shaft was 6}inches indiameter,but thesizewasincreased to7}incheswherethethreeropedrums‘were fitted. ‘The normal speed ofthecountershaft was190revo- Intions perminute. ‘The dynamos were arranged inadouble rowoneach side ofthe countershaft;thedistancesbetweenthecentreofthe countershaft andspindles ofdynamos being 28feet and 33fest respectively. ‘Atoneend ofthecountershaft anautomatic speed-recorder was placed. This recorder was essentially achronograph, and consisted ofadrum otfwhich asheetofpapercouldbefastened; the drum was driven byatangent screw working inaworm wheel, which latter could bemade fast toorreleased from the drum spindle bymeans ofafriction clutch. ‘The tangent screw ‘wasconnected byaflexiblecoupling totheendofthecounter shaft. Asmall table wasarranged having amotion parallel to thedrumaxis.Onthetableanelectro-magnet wasfixed,the movable armature ofwhich carried asmall glass pen. ‘The electro- magnet wasputincircuit with abattery and clock having» minute contact. Asthe drum was mechanically driven (without slip) from thecountershaft, itscircumferential velocity waspro- portional tothespeed oftheshaft; andastheclock marked equal intervals oftime onthepaper, thedistances between these marks gave thenumber ofrevolutions ofthecountershaft when measured onasuitable scale. The dimensions were adjusted sothat the drum made one complete revolution infive minutes when the countershaft was running atthe normal speed. Hence, ifthe speed wasnormal, andwithout variation, themarks made bythe clockwouldbeinastraightlineacrossthepaper.Ifthespeed m3 EXHIBITION, MELBOURNE, 1886axp1889. nt didnotvary, butwaseither greater orlessthan thenormal, the dock marks would still beinstraight lines, though making different angles with theedgeofthesheetofpaper.Anirregular speed would show anirregular line. (Plate I,Figs. 4and5.) Geared from thesame spindle that drove thereading drum, idmounted onthesame base, wasasmall centrifugal governor. Theopening orclosing oftheballs brought asmall contact lever against either oneortheother oftwocontact stops. Two incan- descent lamps, oneplain, theother coloured red, were hung over theengines, theposition ofthecontact lever determining which, ifeither, ofthe lamps should bealight. Atthenormal speed either would show light, butavariation ofabout half percent. oneither sidewould place oneorother ofthelamps incircuit. Themajority ofthediagrams taken bytherecorder showed a gradual increase inthespeed ofthecountershaft asthe evening advanced. This fact will bereferred tolater on. Generally speaking, thelighting throughout theterm ofthe Exhibition wassatisfactory. Onthree orfour occasions only did sything occur which would lead the public tosuspect that matters were not exactly asthey should be. This comparative freedom from inconvenience through irregularities inthelighting vastheresultofthearrangement ofthecircuitsratherthanthe constancy ofsupply toany onecircuit, The boilers, which were designed byaVictorian railway ‘engineer, and built inMelbourne, behaved admirably, though working considerably beyond their proper capacity. The steam Pressure was kept extremely regular, the maximum variation seldom exceeding two pounds per square inch forthe whole evening's run. The fuel was coke, with alittle coal added at. times when ditficulty was experienced inmaintaining thefull working pressure required. ‘Notwithstanding thevery regular steam pressure thespeed oftheengines varied considerably, even though they were fitted with mechanism outwardly resembling Hartnel’s automatic ex- Pension gear. With thefullload onthecountershaft, and the engines indicating about 1,500 horse-power, asingle dynamo thrownofforonproduced averysensiblealteration inthespeed 118° LIGHTINGOFCENTENNIAL INTERNATIONAL |[Nor.14 oftherestofthemachinery—sufficient, indeed,tonecessitatethe adjustmentoftherheostatcontrollingthepotentialdifferencest theineandescence mains, and toshow itself atthespeed-indi- cating lamps. Orieeveningsometrialsweremadetoascertain theeffectof analteration inthesteam pressure onthe speed oftheengines For this purpose the pressure was altered one ortwopounds every half-hour anditwasfound that variation inpressureof ‘onepound persquare inch produced analteration inthespeed of theengines ofabout arevolution per minute, allother conditions remaining thesame. Thecause oftrouble wasclearly enough in thegovernor, and Iexperienced agreat deal ofanxiety and ‘annoyance from afault which should nothave existed, Thefact is,themakers oftheengines could notbemade tounderstand theimmense importance inelectric lightingofabsolutesteadiness intherunning oftheengines, orthat there wasanydifference in thedriving ofdynamo machines andstreet tramways. TheBrush Victoria dynamos ranvery well through thewhole sixmonths theExhibition wasopen. Inthemain they kept cool, though occasionally thethrust bearings showed aninclina- tion toheat. ‘The commutators ranwith very little sparking when thebrushes were properly adjusted. Unfortunately, the above satisfactory remarks cannot be repeated withreference tothearcdynamos. Inthefirstplace, the insulation inmany ofthem was defective orinsufficient. Consequently, “electrical fireworks” formed anotunimportant part oftheevening's proceedings, tillsuch time asthefaulty armatures and field coils had either been rewound orweeded out. Inmany caces, themachines ranforseveral weeks without giving any indications ofweakness, when suddenlyabreakdowninthe insulation would take place. Another fault, more serious (if possible) than the last, existed insome ofthearmatures, namely, those which have been designated “thin-plate armatures.” All thelaminated armatures supplied, both thick andthin plate, were built upoftheso-called Hpieces, bound inwith astrip ofsheet iron, equal inwidth totheintended thickness ofthebobbin core. ‘There was noindication ofthe Hpieces having been passed | ae] «EXHIBITION, MELBOURNE, 1888 axp 1889. m9 between rollers,oranythingdonetogivethemthesamecurvature asthe armature. ‘Thehorns oftheHpieces were practically flat, hence their resistance tobending was simply that due tothe thickness ofthemetal, which isinsufficient inthecase ofthethin- platearmatures. Afewweeksaftertheopening oftheExhibition several ofthethin-plate armatures threw outplates. Two of theso plates struck adoor-post, andgave unmistakable evidence ofthevelocity with which they were thrown, Onsubsequent ‘oceasions theplates were notsoneatly thrown out,butwere caught bythefield magnets, andcuttheinsulation ofthearmature coils. Bythetime theExhibition closed agoodly number ofplates had been thrown out. Toprevent accidents perforated zine screens were placed over thearmatures. ‘The cause ofthese plates breaking offwasevidently thecon- tinal bending towhich they were subjected inpassing through thedifferent parts ofthemagnetic field inwhich they rotated. The force due tothe rotation ofthe armature tends tobend the plates outwards, while that duetotheattraction ofthepole-pieces hastheopposite tendency (atleast with reference totheouter plates). Clearly, then, though theforce duetorotation isconstant, asthe plates pass through amagnetic field which isnotuniform inintensity ordirection with reference toany given armature radius, theforce tending tobend theplates inwards must vary in different parteofthefield,andaconstantbendingoftheplates must result. This mechanical fault was avery serious one, and might have caused damage tolifeandlimb. ‘Beforedescribing themethods usedandtestsmadewiththe dynamo machines, itmay notbeoutofplace togive afewofthe reasons which ledtotheselection ofthemethod adopted. Inorder that thetests should beequally satisfactorytoall concerned, itwas essential that the machines should be tested under theactual working conditions, orwhile they were atwork ontheir respective circuits. Itwas notthought tobeimportant thatallmeasurements taken should beofthehighest degree of accuracy. Aspractical engineers know, theconditions ofrunning donotremain constant (especially inarcmachines) formany consecutive moments, andthefluctuations which take place affect 720LIGHTING OFCENTENNIAL INTERNATIONAL [Nor.1h, alldata,andtheresulting efficiency frominstanttoinstant, ‘Hence the desirewastomaketestswhichwouldbeofpractical rather than theoretical value, ‘The many more recently suggested methods oftesting dynamos inwhich allthemeasurements taken areelectrical, such asCardew’s andSwinburne’s, orinwhichthemechanical isthesmallerpart,suchasDr.Hopkinson’s, werecarefully consideredwithreferencetothespecialformofmachinetobetested.Of these methods, that ofCaptain Cardew seemed tobetheonly ‘nethat would give satisfactory results, forbyitthemachine ccanbetested under theactual conditions ofworking. Now that itisnolonger thought that adynamo issimplys ‘generator ofelectricity, and assuch may beused indiscriminately foranypurpose towhich electricity isapplied, makers designing special forms ofmachine tomeet thedifferent requirementsfor which they are tobeused, the importance ofconsidering the behaviour ofthe machine when working under the exact com ditions forwhich itwasdesigned cannot beover-estimated. Inconsidering the matter, the many advantages, from mathematical standpoint, ofselecting oneoftheabove-mentioned methods, orofrunning themachine onaclosed metallic circuit, could notbeoverlooked. How much more satisfactory itwould betohave the current almost absolutely steady, without those fluctuations caused byafaulty lamp orabadcarbon, which make thereadings oftheinstrument used intesting soirregular, and 80difficult toobtain with any degree ofaccuracy! Butthe machines werenotmadetorunonclosedmetalliccircuit,and, ifthey aretobetested fairly, must betested with allthe fluctuations ofcurrent,frictioncausedbypullofbeltonbearings,and other modifying influences which exist under thenormal condition ofworking. Insome classes ofmachine, where thecommutator basmany segments, andeach section ofthearmature winding hasbutfew convolutions, especially ifthe machine isintended tolight incandescent lamps orcharge secondary batteries, thechances of error inanyofthe above methods would beaminimum, Bat with machines like the Brush aredynamos, where thecommu- 10] «EXHIBITION, MELBOURNE, 1868 axp 1880. m {atorsegments arefew,andthenumberofwindings perarmature cilvery great, theerrors introduced byrunning themachine on loved metallic circuit instead oftheordinary lamp circuit for whichitwasdesigned maybeconsiderable, unlessthemetallic circuitisarrangedtohavethesameself-induction asthelamp cenit,Withsuchmachines astheBrushthecurrentshouldbemaintained asnearly aspossible atthe value forwhich the machinewasdesigned, assmallalterations inthecurrentmay produce sensible alterations inthebehaviour ofthemachine. The tests should bemade when the machines have attained their staleofsteadytemperature. Itis,ofcourse,needlesstoinsistthatthemachineshalloneachandeverydayreachitsstateof steady temperature, astheatmospheric temperature may vary considerably. Itissufficient ifthemachine attains atemperaturewhichwouldbeitssteadystatewithsomeknownatmospherictemperature withintherangeofpracticalwork, IfDr.Hopkinson’s method, where twosimilar machines have their armatures rigidly connected, were used, it,would beneedful toconsider theexact relative angular position ofthecorresponding coils inthe two armatures, Ifthis were not attended to,the armatures might besoconnected that theperiodic fluctuations of thecurrentwouldbeeitherincreased ordiminished bythecom- bination ofthemachines; hence theconditions ofpractical work might notbeattained. Inany case, with thetwo machines directly coupled (electrically), assuggested either byHopkinson orSwinburne, themodifying effect oftheself-induction ofthe cironit onthe current fluctuation would belost. For these and other reasons theelegant methods adopted byHopkinson and Swinburne were deemed unsuitable inthis porticular case,Cardew’s methodwasalsorejectedasbeing unwieldy, necessitating asitdoes theremoval ofthemachines, andsomany belt transmissions, ‘Asnootherpurelyelectrical methodwasthoughtof,itwasdetermined toadoptthemethodinvolving theuseofamechanicaldynamometer formeasuring theenergy absorbed bythemachine, andtomeasure theoutput bytheordinary electrical method. ‘Thechiefdisadvantage ofthisplansppeared tobeinthe 722 “LIGHTING OF CENTENNIAL INTERNATIONAL (Nor. 1 necessity ofhaving alltheinstruments used accurately calibrated. ‘The efficiency would notbetheratio oftwosetaofmeasurements taken bythe same instrument, oreven bytwo different setsof instruments which could bedirectly compared intera.Butas there isnogreat difficulty inthecalibration ofelectrical instr ments, atleast within adegree ofaccuracy sufficient forall practical purposes, and the mechanical dynamometer canbe directly tested, the increased facilities fortesting the machina whilst actually working ontheir respective circuits wasconsidered sufficient tooutweigh thedisadvantages ofthemethod. Inallefficiency teste but one, Siemens electro-dynamometen ‘wereusedtomeasurethecurrent. ForsomeothertestsAyrton andPerry's meters, ofboth spring andmagnet forms, were wed. ‘The values ofthereadings ofthese instruments were carefully determined bycomparison with the electro<dynamometers, The latter instrament wascalibrated byconnecting itinseries with battery, anadjustable resistance, andafixed resistance consisting ofalarge number ofsmall German silver wires arranged s0thit anynumbercouldbeputinparallel. ‘Themassiveterminals towhich the small German silver wires were fastened were cot- nected through avery sensitive reflecting galvanometer ands standard Clarkcell.Akeywas,ofcourse,placedineachcireait With agiven number ofGerman silver wires inparallel, the current through thetotal circuit wasadjusted bymeans ofthe variable resistance tillthepotential difference attheterminals ofthe German silver wires was equal tothe electro-motire foree ofthe standard cell. ‘The exact resistance ofthe German silver wires was immediately measured bythe aidofoneof Elliot'smetrebridges.Asmallstandardresistancewascom structed fromapieceofthesameGermansilverwiress *thatusedinthefixedresistance, Theexactvalueofthiscoil was found bycomparison with one ofElliot's standard BA. ‘units.3All spparatus tobeused wasplaced inposition some time before required, toallow the temperature ofallparts tobecomeequal.Thetemperature wastakenbyastandard thermometer which had been tested atthe Kew Observatory. ‘Two standard cells were used—one byL.Clark, Muirhead, &Co; 1 16] -—«-EXHIBITION, MELBOURNE, 1868 axp 1889. 128 theother made upatthetime inaccordance with Lord Ray- leigh’s instructions, whose formula forthe variation ofthe dlectro-motive force ofthe cell with change oftemperature wasused.Thetwocellswhentestedwerefoundtobealmost exctly equal. Nodifference could bedetected when they were ‘compared bythe condenser and ballistic galvanometer. But a verysmalldifference didexist,andwasshownbythedeflectiononasensitive reflecting galvanometer whenthecellswerecon-nectedzinetosinc,theremaining terminals beingconnected to thegalvanometer. In-allcalibrations somemeasurements weretakenwitheach ofthe standard cells. The results ofthe calibration tests ofthe Siemens electro- dynamometer gave thevalue oftheconstant =‘8716. ‘The con- stantasgivenbythemakerwas-87—adifference oflessthan one-fifth per cent. The same electro-dynamometer when com- pared with oneofAyrton and Perry's latest spring ammeters differed byvery little more than -O1percent. ‘Twovoltmeters were used—one ofAyrton andPerry's spring form, theotheroftheelectro-dynamometer pattern.Bothofthese instruments were wound with copper wire, and large resistance coils ofGerman silver wire, differentially wound, were used to reduce thepotential difference attheterminals oftheinstrument proper. The voltmeters were calibrated inexactly the same manner asthecurrent-meters, They were also checked by measuring thepotential difference attheterminals ofaknown resistance bymeans ofthecondenser and ballistic galvanometer. ‘These twomethods gave exactly thesame results inseveral tests, andthecondenser method wasultimately discontinued, owing to thedifficulty ofobtaining aplace where theballistic galvanometer could beused without interference from locomotives moving in * thevicinity. The dynamometer voltmeter was afterwards com- pared with oneofAyrton andPerry's spring meters: for400volts thedifference intheirreadingsdidnotexceedone-tenth ofavolt. ‘The meter bridge was calibrated byCarey Foster's well- known method. Anadditional correction had tobemade for the heatingofthegaugewirebyconduction fromthebindingscrews, 72LIGHTING OFCENTENNIAL INTERNATIONAL [YorUd, ‘Nospecial means were previously employed toprevent error from ‘this source,asitwasnotanticipatedthattheeffectwouldbe ‘appreciable. Theselection of@suitable mechanical dynamometer wasno easy. The following conditions were laid down asessential:— Ist,Itmustbesusceptible ofafairdegreeofaccuracy,and show instantly anyvariation intheload. ‘2nd. Itshould beofsuch aform thaterrors arenotlikelyto arise from thedifficulty (owing tothethicknessofthe belt) ofaccurately determining thevirtual radius thepulley; and nopower should beabsorbed between thepointatwhichthepressure ismeasured andthe dynamo spindle. 3rd.Anysprings contained mustbesoplacedthatcentri- fugal force will not introduce uncertain effectsonthe indicationsoftheinstrument. 4th. Itmust beportable, andeasily fitted up. ‘Thefirstcondition simply requires that whatever theprinciple involved itshall befaithfully executed, and that the moment of ‘inertia and number ofmoving parts shall beassmall aspossible. The second condition requires that thepower shall betransmitted tothespindle ofthedynamo atsome fixed distance from itsaris ofrotation, andthatitisthepressure atthispointwhichisshows ‘ontheindicating mechanism. ‘Thesimplestwayoffulfillingtheaboveconditionsseemedto betogivethedynamometer theformofapulleytobedirectly appliedtothespindleofthemachine tested. Itisimportantthat thedisplacement ofanyofthepartsofthemechanismshallbe verysmall,eventhoughtheindicator passesfromzerotoits maximum reading. Apulleywasaccordingly constructed theprincipleofwhich isasfollows :— ‘Theenergy supplied tothemachine istransmitted through afixed point at distance intermediate between the axisof thespindle and rim ofthe pulley. Bymeansofleversani| pistonsthepressure istransmittedtoafluidfillingthechamber inwhichthepistonmoves,andbyitthrough8glasstubewhic| 1889.) EXHIBITION, MELBOURNE, 1888axp1889, 75 lies inthe axis ofrotation to spring chamber thecapacityof whichvarieswiththepressure ontheenclosed fluid. Thepressure onthepoint through which theenergy istransmitted isshown by thedisplacement ofthefluidintheglasstube.‘Thepulley(seediagram, Fig.6)wasmadesothattherim, )ou mt tate |7 [8 | | Ais Y | hx Q Xx EX & o, mee R,could tum freely onthenave,N. ‘Two radial chambers, C, were carefully bored inthe nave atthe opposite ends ofa diameter, and pistons were accurately fitted inthem, sothat VOL. XVI. . 50 726 LIGHTING OFCENTENNIAL INTERNATIONAL [Nov.14th, though watertight they could move freely. Two bent levers, I, werehingedatthenave,eachbeingsoshapedthatwhenone arm pressed onthecentre ofthepiston, theline passing through thispointofcontact andthecentreofthehinge,wasatright angles totheradius drawn through thepoint ofcontact. Atthe same time the other arms ofthe levers stood inadirection such that thesurface ofthelever remote from thepiston coincided with theradius passing through thecentre ofthehinge. These arms were ofsuch length that they just cleared theinner surfice ofthepulley rim. The connection between therim andnave wasmade bymeans oftwo blocks ofiron, B,riveted totheinner surfaceoftheriminsuchaposition thattheywouldcomein contact with the projecting arms ofthelevers. The point of pressure wasrendered definite bytheuseofknife edges, andany shock wasrelieved bytheinterposition ofasuitable spring. Two small holes were drilled parallel totheeyeofthepulley opening into thecylinders. ‘The indicating apparatus was fixed sothat itwouldcoincidewiththeaxisofrotationwhenthepulleywasinuse, Itconsisted ofaglass tube, with acarefully fitted ebonite piston which served asamarker. This piston wasmade about 10diameters long, and had several small grooves tured in it.The glass tube wasmounted with ametallic chamber ateach end, onebeing bolted tothe pulley nave, theother closed with a circular corrugated steelspringsimilar tothoseusedinsome steam pressure gauges. The metallic chambers were mechani- cally connected byaniron tube inwhich three longitudinal slots werecuttoenabletheglasstubetobeseen.Theangular widthofeachslotwas60°.Thescalewasfastened ontheinnersideofthebarsoftheirontube;consequently eachtime‘aslotturnedtowards theeyetheglasstubeandebonitepistonwereseenprojected onthescalethatwasfastenedtotheoppocitebar,Wheninrapidmotionthescaleappeared asifdrawnround theglass tube, andtheendoftheblack ebonite piston formed awell-defined linefromwhichtoread.Everytenthdivisionwas marked inred,andthefifties were shown byabright lineonthe outer surface ofthe iron tube. Inthepulley used, thediameter ofthebore intheglass tube lim] —=«RXHIBITION, MELBOURNE, 1688 ax 1880 a vaabout1-25thofthatofeitherpiston,Hence,ifbothpistons vere inuse, the motion ofthe ebonite piston would beabout 1250 times asgreat asthat ofeither ofthemain pistons, The length ofthe scale wasabout 6°5inches, sothat amotion ofthe main pistons of1-200th ofaninch would cause theindicator to trvel over the whole length ofthe scale, ‘The pulley was aminged s0that one orboth pistons could beused. Bythis means the range ofthe instrament was doubled. Using one piston andrunning at900 revolutions perminute, themaximum reading corresponded toabout 30horse-power, and with both pistons toabout double that value. Itwill beseen that adynamometer ofthis form may be constructed tofulfil fairly well the conditions Iaid down ascasential. . ‘The sensitiveness isdetermined bythe stiffness ofthesteel spring and length ofglass tube used. Inpractice avery great degree ofsensitiveness isnotrequired, asthevariations inthe power absorbed were found toexceed two, orattimes even three percent. ofthetotal, whereas one-half percent, could have been easily read. Theammeter showed fluctuations inthecurrent at least equal inrelative magnitude tothose shown bythis mechanical dynamometer. Tocalibrate thedynamometer pulley itwaskeyed onafixed horizontal spindle. Alongbalanced armwasboltedtotherim, and atadistance from the centre ofthe spindle (measured horizontally) equalto2°feetscrewwasputintothelever.Variousweights weresuspended fromthescrew,andthecorre- sponding displacement oftheebonite piston noted. Itwas found that the displacement was directly proporticnal tothe Toad; hence thescale could besodivided that thereadings would show ataglance thetangential pull ofthebeltinpounds atany given radius. Inthepulley used theconstant was calculated foraradiusofzefeet.‘Thereadingshowedatonce thetangential pullatthisradius, and,multiplying byten,thenumberoffoot-pounds absorbed perrevolution ofthearmature. ‘Itshould benoted here that thelevers inthepulley were 79 LIGHTING OF CENTENNIAL INTERNATIONAL [Sor Md, counterbalanced forthecentrifugal force ofthepistons andfiid below them. The correctness ofthebalancing was tested by mounting thepulleyonaspindleandrotating itrapidlyinbth directions, first ofallhaving removed therim. Ifthebalancing were correct asmall deflection should beproduced, itsdirect depending onthedirection ofrotation, and thetwo deflections should beequal atthesame speed ofrotation. Noerrorms detectedowingtotheunequaldistribution ofthefuidpresse ‘onthecircular spring caused bycentrifugal force. ‘This factis notsurprising when itisremembered that theeffective diameter ofthespring wasonly 1-25 inches, andthat thefluid (glycerine) ‘used waslight. ‘Thedynamometer pulleywasplacedonthemachinewhichit ‘wasdesiredtotest,inthedaytime,andfreshglycerinepati One ofthebolts holding thefield-magnet core was taken ott, and asimilar one, but with ahole bored inittoreceive the thermometer, putinitsplace. ‘The electro-dynamometers were always carefully adjusted fr position after allmachines were infullwork, sothat nodistur ingeffect would beintroduced bytheaction ofthecurrent in lends which rannear thetesting table. Each reading noted gare themean position oftheindicator forabout 15or20seconds in allmeasurements, whether electrical ormechanical. The speed ofthe armatures was determined byoneof Harding’s counters, inallcases afullminute being taken. The constants of all electrical instruments were calculated forthemean temperature ofthedynamo room—27° centigrade. ‘The temperature correction forthedynamometer pulley wasnot determined, asonly thefirst orlasttwoorthree readings were to beused intheresults, and theaccuracy ofthese was ensured by taking thezero ofthepulley both before and after themachine wasrun, Theintermediate readings were taken only toshow | anyserious alterations took place inthegeneral workingofthe| machines. ‘These tests, theresults ofwhich aretabulated (see Table A} were made printarily toascertain thecommercial efficiency ofthe various aremachines used, By“commercial efficiency” Imest tse) «EXHIBITION, MELBOURNE, 168ano1882, glo sks 8)8)8 iregSere 3/533 éa ~355/5 BEE .| ot 5 res = e/alE)af|SeSeclalaz eliza 8 is 822 _5/5/825/888gl"!dt[EECHEgEg e328a|-——— =a 8 ZeeesizeesieeelBH|£E2RH888828 Bi}a]a* a | wl]tgsReaoclasasians¥ [fe ewnelelaaalane ile SE 15 5 V)e El 2 ]ee es 3<j2/38EEE CEEEee = a ee eee) wees |mane 7 LIGHTING OFCENTENNIAL INTERNATIONAL (Nor.14, the ratio oftheelectrical energy available attheterminal to theactual energy absorbed atthespindle ofthemachine. Itwill beseen that the current actually supplied bythe majority ofthe machines was alittle below the normal, the machines being professedly designed foracurrent of10ampere,‘Thetimesofrunasshowninthetablewerenotsulficientfr themachines toattain their state ofsteady temperature, though inthemajority ofcases thetemperature reached would beafar average forordinary working. ‘Thearmatures ofthemachines were interchangeable, sothat anygiven armature cannot beconsidered apartofanyspecfal machine; itwas important, therefore, toascertain not onlythe relative value ofthe different forms offrame, but also ofthe different armatures inthem. Tewill be seen from the table that the different formof machine having thin-plate armatures arealmost equally efficient, thesolid-frame pattern having aslight advantage, ‘Thedifference, however, between the two forms ofmachine with thick-plate armatures isstill less, and theadvantage infavour oftherail pattern, Butasonly oneofthesolid-frame machines contained8thick-plate armaturethecomparison isoflittlevalue,Ferpractical purposes itmay besaid that thecommercial eficieney ofthetwoforms ofmachine, provided thearmatures aresimilar, inthesame. Greater alterations intheefficiency ofeither claw ray becaused byslight alterations intheconditions ofworking, than the actual difference between the two classes asshown in the table. ‘The difference inthe two types ofarmatures isvery marked. ‘Taking themean from thetwoclasses ofmachines, theefficiency ofthethin-plate armatures will befound tobeabout 827per cont. greater than that ofthe thick-plate armatures. But ‘unfortunately, thethin-plate armatures were notcalculated to withstand themechanical strains towhich they aresubjected in ordinary work, 90thehigher efficiency obtained from them bas only atheoretical existence. ‘There is,however, nomechanial reason whythearmatures shouldnotbeconstructed withthis| plates,andatthesametimebefreefromthemechanicalfaults| 1880.) EXHIBITION, MELBOURNE, 1888axp1889, ‘81 showntoexistinthoseunderreview.Indeed,sucharmatures have been constructed intheVictorian Railways Telegraph work- shops which arequite freefrom thefaults named. One ofthese thusbeen running forseveral years without ever giving trouble, anddoing excellent work. @BEEEEEEEEE EEE tere ais tS amUSREERESIS""CGUaMMEGEESPanpegaBoEREEEEEEEEEEREACT esFEeset Pr LAFeootsdete4 PPEratedekaEEE AEE reer tT aimOCT tbe | Co HERE EEEEeefee fey a PEEEEEEEE EEEEEEEEE “l HEEHabaaekeateGarey b Hodersh Tachogh Phd che Fro. 1 ‘Thecurvesinthediagram(Fig.1)weretakentoascertaintherelativevalueofthemagneticcircuitsinthetwoclassesofframes.‘Themethodadoptedwasasfollows— ‘The commutator brush-holders were electrically disconnected from thefield-magnet circuit. The ordinary brushes were takenoutandreplacedbyotherssobentthatcontactcouldbemade with the commutator atthe neutral point. The armature was runatthenormal speed, and thefield magnets were separately excited byavariable current taken from anincandescent cireuit. Simultaneous readings were taken ofthecurrent through thefieldcoils,thepotentialdifference attheterminalsofthearma- ture,andthespeedofrotation. Fromthesedatacurveswereplotted asshown inthediagram. 132 ‘LIGHTING OFCENTENNIAL INTERNATIONAL [Nor.Mth, aol See all = =| 3 a W|2/*|52 |28a8RRS 22hSz cle(|[ee|S888SRR8Rae g HP|5225 22/28 s)822 -|"| 3, |isaeealalana p(e SS lege | b/2/2)4|sszeelslegelegsm)2/#| 4)28329)5) 33288 : |el) [|=~ | Google = 2 1889 EXHIBITION, MELBOURNE, 16884x01889, 138 Allthemachines tested inthis manner had similar thin-plate armatures, Theresistance ofthevoltmeter used was great, sothat the potential difference asshown byitwaspractically theelectro- motiveforceofthemachine. ‘Asthe number ofwindings and speed ofrotation ofthe armatures were the same, the electro-motive forces would be directly proportional tothetotal induction through thearmatures, The induction through thearmatures depends onthemag- netising force (H) ofthecurrent inthe field coils, and the goodness ofthemagnetic circuit. Inthe machines tested, thenumber ofwindings onthefield coilswere equal. Hence theelectro-motive forces ofthedifferent machines, when taken with thesame current inthe field, will represent therelative values oftheir magnetic circuits. The curves may betaken either assaturation curves ofthe total magnetic circuit, orascurves oftotal induction through the armature, provided nosensible current istaken from it. ‘Anexamination ofthecurves inDiagram I.willshow asmall magnetic advantage infavour ofthesolid frames. This result was expected, from the fact that the bed-plate increases the effective cross area ofthefield-magnet yokes, butisnotlikely to materially increase themagnetic leakage owing tothe great length ofcircuit. The early part ofcurve ofdynamo No,460wastaken with thevery greatest care. Only afew ofthe points aremarked for clearness, asthecurve passed evenly through them all. The ‘usual concavity atthe commencement isshown inamarked degree.Theslowriseofthiscurveforlowdegreesofexcitationseems toindicate agreater degree ofhardness oftheiron. All the curves approximate tothe same form over that portion of their length which corresponds with the ordinary range of excitation. TableBshowstheconditionsofrunningandgeneralbehaviour ofthe machines shortly after starting. Very little more time was allowed before the measurements were taken than was necessary forthelamps toassume aproper working condition, 14 LHOBTING OF CENTENNIAL INTERNATIONAL pi ge =[F/=|2eS ee '[pest?ale egel28sBSil++ee4 fete t+ % ae Heti+ +i++a:|eehBleXlnos (ss &|ii|sge8i 8Eee(283 elielsyoCeeele| a5: Fa 1/888 _zlsl8eelee8ogAPE TTTg|i {E 2/22¢8.8/8 88ales3 Vlgeigies :zSS/8]i:2/E3:8 6)82/7)228s/8sisie lg es.sealel, eles. a) a[Epee[oe ee es 7 ‘omawa 1889.) EXHIBITION, MELBOURNE, 1888axp1889. 735 Attimes, however, the tests had tobepostponed till the machinery attained itsnormal speed. Itwillbeseen that the current inmost cases wasabove 10amperes. InTable Cthemore important values given inTables Aand Barecompared, and their differences, with their proper signs, are added, toenable the alterations which take place when the machines andlamps areheated, tobemore readily detected. ‘The currentineverycasebutone,fellastheeveningadvanced. ‘Thisreduction ismainly duetotheincrease inresistance (due to heat) oftheshunt circuit inthelamps. The effect ofthis isto lessentheactionofthefinewireinthedifferential coilsandallow thecarbons tobeseparated alittle farther; thatis,toincrease the potential difference required attheterminals ofthelamps. Intenoutoftwelve ofthearcmachines tested, thecommercial ficiency wasfound toincrease with thereduction ofcurrent that tookplace from theabove cause. Intheother twomachines the current and efficiency rose orfell together. Acomparatively small increase ofcurrent above the normal isfound toproduce asensible diminution intheefficiency. Hence inpractical work itisimportant toascertain themost efficient current totakefrom themachines, and seethat itisalways maintained atthat value. ‘Theincrease intheefficiency may bepartially due toareduction inthe losses caused byFoucault currents inthe armature and fieldmagnets, brought about byanincrease inthespecific resist- ance ofthemetal byheat. ‘The lossduetomagnetic viscosity in thearmature core may also belessened bythesame agency. Itwill beseen that theaverage output ofthemachines is much about thesame whether they arehotorcold. Total output of12machines, cold 19°42 electrical horse-power. » » hot 19594 ,, » Difference... 348 oy lw Total power absorbed atpulleys by12machines, cold 279-77 H.P. » » » hot 26111 yy Difference ... 1866 ‘Thepower absorbed bythemachines ismuch lesswhen they are 736 LIGHTING OFCENTENNIAL INTERNATIONAL [NorMth, hot;thetotalreduction inthetwelve tested being 18°66,or1'5 horse-power permachine, while thediminution intheoutput was little more than aquarter horse-power permachine, This fact ‘accounts insome measure forthegradual increase inthespeedof theengines(whichhasbeenpreviously mentioned) astheevening advanced, though thesteam pressure and number ofmachine: driven remained constant. ‘Asthereisnodirectmethodofascertaining thepowerabsorbed byFoucault currents and magnetic viscosity while themachines areatwork(theusualmethod beingtofindthevalueofall directly measurable quantities and subtract their sum from the total power absorbed bythemachine, thebalance being debited totheabove causes), anattempt wasmade tofindapproximately thealteration inthemagnitude ofthese losses corresponding with given variations inthetotal induction through thearmature,by thefollowing method :— ‘The dynamometer pulley was placed onthe machine to betested, thefield magnets were arranged sothat they could beseparately excited byavariable current, and thecommutt- EEEEEE EEE|Coo gewacSoaeo22a00unae:!TseStiroeTheetetoode|TT BeeEEEEEEEee zSa ee BECOee ECCCERRSEAL eatPeHEbageSSeseaeseeccees!=:dest |Tester CrreRACE ereer APCs LgatESPsGevaupeeiueees {225701u Bs ram om Peeear | Perec eeeCurrentinAmpeesTiowphFieldCoitFie. 2 - 1690] “EXHIBITION, MELBOURNE, 1888 axo 1099. 01 torbrushes were removed from the machine. The zero ofthe pulley was noted, and the machine then run atthe normal speed, thedeflection produced onthepulley indicating thepower absorbed bymechanical friction. The field magnets were then gradually excited and simultaneous deflections ‘ontheampere- meter and pulley noted, the speed ofrotation ofthe armature being also taken. From thedata thus obtained thecurves on thediagram (Fig. 2)were drawn. ‘Ttmust notbeforgotten that thevalue obtained forthewaste power, corresponding with agiven degree offield excitation, doesnotnecessarily represent thatwhichactuallytakesplacewhenthe machine isatwork with thegiven degree offield excitation. The modifying influence ofthecurrent inthearmature cannot well beestimated. The reduction produced byitin thetotal induction through thearmature core would tend tolessen these losses. At the same time the losses inboth armaturecoreandpole-pieces ould beincreased, owing totheforward rotation ofthefield by thearmature current crowding up,asitwere, thelines offorce intoasmaller space, soincreasing theelectro-motive force ofthe induced currents, thevalueofwhichisalways3",Anothersource ofdifficulty intheBrush machine istheuncertainty oftheeffect produced bythegreat oscillation ofthearmature field (about 40°)owingtothesmallnumberofcommutator segments. Thecurves shown onDiagram II.present some striking peculiarities. ‘The points from which theseveral curves start onthevertical lineforC=Omay beregarded astheir reepective origins. ‘The power indicated bytheportion oftheordinates below thispoint isdue tomechanical friction, and isconstant foralldegrees ofexcitation ofthe field. The curves Nos. 462A and Bwere taken with themachine cold and hotrespectively. Itwill be observed that lesspower iswastefully absorbed when themachine ishotthan when cold, and that. asthe two curves start from the samepointthedifference isnotduetomechanical causes.‘These curves show intheearly part avery slow riseinthewaste powerforagivenincrement inthefieldexcitation; butastheexcita- tionisincreased, thelostpower rises with anincreasing rate. On 788LIGHTING OFCENTENNIAL INTERNATIONAL [Nov18, ‘the same diagram the curves marked Nos. 1856A and Bwer takenwithincreasing anddecreasing currentsinthefieldcallrespectively. Byreference toTableCitwillbeseenthatdynamoNo,1856isoneofthoseinwhichtheefficiency risesandfalls with the current inthe circuit. The curve taken with anincreas ingcurrent istheonly one ofinterest. The early portionis peculiar, and Icanseenosatisfactory explanation ofthepest Uiarity. From thepoint where theexciting current is4ampere ‘thecurve isnearly astraight line, which, ifproduced downwaris, would pass through the origin. Inthis machine, then, the power absorbed, driving thearmature (onopen circuit) through the magnetic field increases directly asthe current atleat between thevalues C=4and =10, But thepower required toovercome mechanical friction iscoustant. Hence, ifthe resistance ofthecircuit remains constant, asitpractically does inthis case, the efficiency ofthe machine will rise and fallwith ‘the current taken from it. Curve No. 460 was taken from the othermachine, inwhichtheefficiency andcurrentriseandfll together. The power absorbed bythearmature onopen circuit increases atfirst very rapidly with thecurrent; but when the current hasreached 3amperes the rate ofincrease isnot great, anditultimately tends towards asmall constant rateof increase, The curves given inDiagram IL.express the horse-power absorbed indriving thearmature (onopen circuit) through the variable magnetic field asafunction ofthe exciting current. This method ofrepresentation ispurely arbitrary, though com venient when itissimply desired toascertain thealteration in thepower absorbed foragiven change intheexciting current. The power required todrive thearmature onopen circuit through amagnetic field clearly willnotdepend onthevalue of theexciting current, butonthenumber oflinesofforcewhich thegiven excitation canforce through themagnetic circuit of ‘themachine. Those lines offorce only which pass through the armature produceanyusefuleffect,Consequently, ifitisdesiredtoascertain therate atwhich thelost power increases with the ‘useful induction, the curves given inDiagram II,should be 1880) EXHIBITION, MELBOURNE, 1888axp1689, 709 expressed asafunction ofthetotalinduction through the armature, | ThecurvesinDiagram I,maybetakenascurvesofinduc- tionthrough thearmature, expressedasafunctionoftheexciting current. The curve ofinduction through the armature ofdynamo No,460 isshowninthediagram(Fig.3),andismarkedNo.460A.a ar earbaesested adset|TTT eer PTTrrryer5 ee Psi eer er BAep CES AH fed$2|4TeerBpbaz tTveTrrrrrrrrr reryet|eSbt|Ar Perr| eyeWIZ TT 2OS BA(1 eee ee Beet rrrrr PerryBa2 eee Bae HHH PT beetUo[acterdesVado|TTTT Fro. 8. Theordinates ofcurve No.460B represent thepower required ©drive the armature (onopen circuit) where the induction hrough itisproportional totheabscissa, ‘AsimilarcurvefromdynamoNo,462,inwhichthepower rastedisalsoexpressedasafunctionoftheinductionthrough hearmature, isadded forthesake ofcomparison. The same ‘eculiarities whichhavebeenalreadynoticedarenowshowninaaore marked degree. Curve 460B tends towards aconstant ratio vetween thewaste power and induction through thearmature. no YORTING OFCENTENNIAL INTERNATIONAL So.ith es ee oUIgee 4 ¢ leer 3 FT eade 2 —_ 2222)2||iclees| +[EEEte3/3 35rfbjlees ss B) gceET"||lee 2H oft |gaee z Fhes| B)|che |go2 ileeee EE &Rete ElTheeela 22) leaaeel Ci cieessa)Allege| |(+[2888 alg, PilzasB[2g)Sel" ,|\2838 S|e)elye+-«|pee flaisls22 were “Joe/28e8)|EGPETE "3, 88a 3\2882/RRAR Fsae xHaj223] T| |l-deIS ee Sasleasg 22 a<™ a2[77 1889.) EXHIBITION, MELBOURNE, 1888axp1889. mH. Indynamo No.462thewaste power increases enormously, with high values ofinduction through thearmature. ‘Thedistribution ofenergy supplied tothree ofthemachines istabulated onTable D. ‘The figures given incolumn 3aretaken from theordinates corresponding tothegiven current onDiagram II. They donot necessarily represent thetrue value ofthe losswhen acurrent is taken from the armature, ‘The figures given inthe other columns aretheresultofdirectmeasurements whilethemachines were actually atwork. ‘The data forthecurves inDiagram II,were taken before the machines were started ontheir respective circuits;theywould therefore becomparatively cool. For this reason thedistribution ofthepower absorbed bythemachines when cold wasselected, so that thevalues from Diagram II.would betaken asnearly as possible under theeame conditions, Itwill be seen that the difference between the measured total and the sum oftheenergy absorbed inthecircuit and various parts ofthemachine isvery considerable, especially soin the case ofdynamo No. 460. Inallthree cases thepower absorbed atthepulley isgreater than that which canbeaccounted for. Probably this difference isdue toanincrease inthepower absorbed byuseless induction when acurrent istaken from the armature. InTable Eafewapproximate data aregiven with reference tothe general condition ofthe magnetic circuit ofthese machines, ‘Thevalues given were obtained with thearmature running atthenormal speed of930revolutions perminute onopen circuit, thefield magnets being separately excited. Incolumns 9and 10respectively thetotal induction through thearma- tures, andcorresponding values ofB,aregiven. These andthe dimensions constitute theonly useful data. Incolumns 5and6 thecorresponding values ofBandmforthefieldcores aregiven, assuming noleakage toexist. This, ofcourse, isnotthecase. VOL, XVIII, 51 i na LIGHTING OFCENTENNIAL INTERNATIONAL [SorMa Thefiguressimplyindicate'a lowusefulvalueofBinthefel cores. ‘When themachines aregenerating acurrent allthevals given willbealtered. Theactual value ofBforthearmature ‘ynamo No.460 would be23,028 when the machine isgivings mean current of101 amperes.* Theeffectofthefluctuations inthecurrentgivenbythe machines wasobserved bycarefully measuring the current give ‘bythemachine and thepotential difference attheterminalsof thefield coils ;immediately afterwards theresistance ofthefield coils was taken. ‘The product ofthemeasured resistance andcurrent givesthe potential difference requisite tomaintain asteady current ofthe given value through thefield coils. Subtracting thisquantity from thepotential difference actually measured, theremainder represents themean E.M.F. duetoself-induction, Dynamo No.460, hot. (2) eae nee ase Measured PLD, =14404 volts. C=11:09 amperes,R =820 ..CR =91-00 » Difference... 4. 53 (2) ene ae nee, Measured PLD, =1390 C=982 amperes, R=820..CR=805,, Difference... 585» Dynamo No,460, cold. (I) ese ase see,Measured P.D, =14055 volts. C=11:33amperes, R=706w..CR= 800, Difference... us 605 (2) ease sae neeMeasured P.D. =13827 C=1085 amperes,R =7060..CR= 766 5 Difference... 6167» “#(ArmatoreEMCE,=terminalPD.+,B+beckBALEieldcols) : = ye +12685 +8795 ‘Total =1,408-45 volts, ) Wt “RXHIBITION, MELROURNE, 188 wo 8,1 From these data itappears that thecurrent fluctuates about 15percent. oneither side ofthe mean, oratotal fluctuation of3per ent Ifthecurrent were steady thegain inthefield coils only would enable the machine tofeed oneextra lamp with thesame armature speed; but the efficiency ofthe machine would notbe much affected. OftheVictoriaBrushdynamos (No.E2usedforincandescence lighting) only four were tested toascertain their commercial efficiency. Much more complete details ofthebehaviour ofthisclass of machine aregiven inthevarious journals than canbefound of theBrush arcdynamo, For this reason itwas notconsidered worthwhiletomultiply testsoftheVictoria Brushmachines. ‘Themethod oftesting wasexactly similar tothat used for thearcmachines, excepting that both pistons inthedynamometer pulley were used, whereas intesting thearemachines only one piston was required. Vicrorta Dynamos, Hor. teevw.wfaoese|wee|scoe|seas|“haetar) mee|itees|oe|tte|oetar2)oats|“arae|seme|den|easeta) awa|moe|aes|an|ae ‘The very small current taken from dynamo No.1587 will ‘account forthelow efficiency shown, asthelosses arealmost independent oftheoutput. +Asa whole, theefficiency ofthese machines isless than was expected. Toconclude, More complete tests,toascertain themagnitude ‘ofthevarious losses which take place inthemachines;thevalues ofH,B,andy,forboth fields andarmatures; theexact magni- tude and form ofthe fluctuations inthe current from the are 4 “LIGHTING OF CENTENNIAL EXHIBITION, Ere. [Nor. Heh, machines, and itseffect inlessening the output from the ‘machines, are much tobedesired. Itwould also have been instructive tohave made along series oftests obtaining data from which curves similar tothose given inDiagrams II,andIll.might bedrawn, soascertaining practically thelawinaccordance with which the power uselessly absorbed varies. The very considerable differences inthe efficiency ofthe various dynamos, working asnearly aspossible under thesame external conditions, and having toallappearance similar arma- tures,isapparent, andthequestion is,Towhatarethesedifferences due? Itwould seem that they must becaused bydifferences in construction, bywhich thedeleterious reactions inthemachines arealtered;butifthetestswhichIhaverecordedinthispaper could have been extended, and made under various conditions, some light would probably have been thrown onthepoint. ‘The close agreement ofthe curves inDiagram I,and the {great dissimilarity ofthose inDiagram II,seem toindicate dif- ferences inthe quality orarrangement ofthe iron forming the armature cores, ‘There were nospare machines atthe Exhibition, thewhole number obtainable being inconstant use; hence the tests made had tobeconfined tothe shortest possible time, sothat the lighting ofthebuilding should notbeinterfered with. This is the reason why further tests were not made, and also for the ‘meagre and somewhat unsatisfactory nature ofmany ofthedata given. | Such asitis,however, Isubmit thispaper tothemembers of | theInstitution ofElectrical Engineers with thehope that there may besomething initwhich will draw forth remarks from ‘some ofthem whose great experience makes anything they say valuable. ‘Atthe Spencer Street Railway Station, Melbourne, Ihave had forsome years small installation ofabout 60Brush are lamps atwork lighting thestation yard, andI'am nowarranging foracentral lighting station from which about 800areand4,000 1603) DISCUSSION. 4s to5,000 incandescent lamps willbelighted. ‘Thirty oftheBrash arcdynamos used attheExhibition have been purchased and will beutilised, most ofthem being fitted with newarmatures, which Iwillhave made upon lines that Ihope, and believe, will prove satisfactory. Atany rate, Iwill avoid theevident errors in construction ofthearmatures sent out from England with these machines. Ifconsidered ofsufficient interest, Iwill beglad atsome fature time togive the result ofthe work done, and also the details ofsome experiments made with aview totest the propriety oflaminating the field-magnet cores aswell asthe armatures. ‘Teuzonarn Exorenex's Orrice, Maaocans, 300k pri, 1889. ‘The Caamman: IsMr. Mordey here? ‘Mr. W. M. Morpey:FirstImustcongratulate theauthorX*Money. ofthis paper onthegreat care that hehastaken toinvestigate amatter that perhaps hasnotbeen carefully studied before out ofthe works ofthe manufacturers. There are one ortwo little temporary faults alluded to—faults discovered byusassoon asby Mr.Murray, and remedied before weknew that they had been discovered inAustralia, Thethrowing outofsmall pieces ofthe Pacinotti projections oftheBrush armatures wasafault that did not occur foraconsiderable time inthemachines that wemade, and wasonethat might easily escape theattention ofthemanu- facturers, butwetook steps assoon aswefound itouttoremedy it, ‘The actual number ofpieces thrown outwasquite small; there were many thousands inthe machines, and only two or three dozen were thrown out, sothat perhaps Mr, Murray has dwelt with unnecessary emphasis onthis really unimportant matter. Tam sure Ineed notsayanything more about this cause of complaint Mr.Murray alludes to, Probably most ofusknow that thesame class ofmachines exactly were used attheColonial Exhibition fortwo orthree hundred lamps, atManchester for five orsixhundred lamps, atGlasgow forfiveorsixhundred 146——_“CLIGHTING OF CENTENNIAL EXHIBITION, Ere. (Nor. ld (ony. lamps; and they ranpractically without any trouble atall.I thinkattheColonial Exhibition wehadsomeofthesemachines running under avery heavy penalty clause—viz., afineof1d.for every minute forevery lamp that wasout; andoutofapossible exceeding 2,000,000 pennies which wemight have been fined, wewere fined 35s, only, and Ibelieve the fines were strictly enforced. Thepaper isavery useful one, butitsusefulness might have beenstillgreaterifithadcoveredwiderground. Itscarcelyfulilsthepromiseofitstitle,foritonlyreallydealswithoneortwo matters inconnection with thelighting oftheExhibition. The greatamount ofinformation whichMr.Murraydoubtlesspossesses astothefuelconsumption, theefficiency oftheengines, thecastof labour, andsoon,inconnection with running alarge instalation ofthat sort, would, Iamsure, bevery useful tomany memberof this Institution; itisnotoften that anopportunity occurs of running aninstallation ofathousand arelamps forasufficest time togetthenecessary figures astocost. Then there wasanalternating-current plant inusethere,¥¢ aretold, butnothing more issaid about it:many ofuswould like tohave had some useful information about that. There areagood many facts, tests, and particulars given in thepaperwhich,Ithink,arepossiblynotofgeneralinterest;but anold question israised nearly atthe end ofthe paper, where something issaid about the discontinuity ofthecurrent from the Brush machine, That question hasoften come up before this Institution. Iremember—in 1882 Ithink itwas— Professors Ayrton andPerry described what they called a‘disoon- “tinuity meter,” theobject ofwhich wastofindouthowdiscon- tinuous was the current from any dynamo. That discontinuity meter consisted ofasortofRuhmkorff induction coil,withtwo | windings, oneofwhich wastobeincircuit withthemachine,and | the other was connected tosome instrument. When Professors Ayrton andPerry described that apparatus Imade some similar tests, and Ireferred tothem before thisSociety atthetime. 1 tookthefieldmagnetofamachinethathappened tobewoundwith twowires, andputonewinding inaBrush arelight cireut, 9) DISCUSSION. ut andputaSiemens dynamometer intheotherwinding, butTas,dave could notgetany result: theSiemens dynamometer didnot takeanynoticeofit.Ievenputthetwowirestomymouth, andcould notfeel anything atall,soIcame totheconclusion thatthecurrent wasnotvery discontinuous. Professor W.E.AvRToN: Was there aniron core inthat? Mr. W. M. Money: An iron core? Yes. Professor W. E.Avaron: That makes the difference. ‘Mr.W.M.Money: Had Professor Ayrton noiron core? Professor W. E.Avaton: No, ‘Mr. W.M.Monper: Ihad one. Itwas anopen magnetic cireuit transformer. Ialways useiron cores intransformers, because Igetabetter effect inthat way. In1887,Imadeanothertestwiththesameobject,butI made itinrather adifferent way. Itook thearmature ofa Victoria dynamo andsent thecurrent ofaBrush dynamo through it,and took thedifference ofpotential. ‘Then Isent thesame current through itfrom anaccumulator, and again took the difference ofpotential,andgotbutaverysmalldifference;in fact, thecoilseemed torequire ahigher difference ofpotential with theaccumulator—a difference of0°5percent.—but that was, nodoubt, anerror inreading. Wemade another testwith trans- formers quite recently, and, infact, since Ihave seen theproof ofthis paper, that seems tometosettle thepoint, bytaking halfadozentransformers, eachmadetotransformfrom2,000volts to100volts, and putting allthe 100-volt coils inseries and all the 2,000-volt coils inseries, Isent the current from aBrush dynamo through allthe low-tension coils, and then, with « Cardew voltmeter, attempted totake thedifference ofpotential of thesecondary orfine-wire coils buttheCardew voltmeter showed nothing atall,although anequal alternating current would have given more than 50,000 volts, With thewethands against the terminals ofthe whole series noshock whatever could befelt. ‘That was atransformer with aniron core—a laminated iron core. Perhaps Professor Ayrtonwillsaythatisthereasonwhyitdidnotactasadiscontinuity meter,butitshowedzerodiscontinuity.With very small machines Ithink thediscontinuity may be 448 —-LIGHTING OFCENTENNIAL EXHIBITION, Exc. (Sor. 18, ar.Monies, Hoticeable :itwillbegreaterthanwithalargemachine running overalongcircuit.Ihaveentirelyfailedtofindanysiguof discontinuity ;butIdonotthink this discontinuity, evenifit existed, would besuch avery bad thing. Itsimply means, if ‘Mr.Murray isright astotheeffect amounting to3percent, thatanalternating currentof0°3ampereislaidontheBrash current; and some ofusthink that analternating current isnet such avery dreadful thing after all. Atpage 742,where Mr.Murray refers tothediscontinuity ofthecurrent,thereisonepointastothemethodofmeasurement » whichIwishtodrawattention. “Theeffectofthefluctuationsin “the current given bythemachines,” wearetold, “wasobserved “by carefully measuring thecurrent given bythemachine andthe potential difference attheterminals ofthefield coils ;immediately ‘afterwards theresistance ofthefieldcoilswastaken,” Thenwe have figures given ofthetests, The product(CxR)was91,.ndthe measured potential difference was144inonecase;andaboutinthe same order allthewaythrough. Then Mr.Murray says: “From “these data itappears that the current fluctuates about 15 «percent. oneither side ofthemean, oratotal fluctuation of$ “per cent.” Idonotquite seehow that value 1°5isarrived at but, inanycase, Iwould like tosaythat that method certainly cannot betaken asaccurately ornecessarily giving thedisco tinuity ofthecurrent atall. Suppose, forinstance, wehada absolutely smooth current sent through thefield magnet ofthe machine. Ifthe armature were rotated, the effect ofthe Pacinotti projections would betomake the fields asortof alternating-current generator, and they would tend tonit the potential difference slightly atthe field terminals Assume, inthissupposed case,thatthesteady electro-motive force propelling thecurrent issuch thatthecurrentisactualy absolutely steady: stillthedifference ofpotential attheterminsls ofthefield would begreater than that duetotheohmic resist- ance tothecurrent flowing through it.Idonotknow whether [an followed inthis, butIwishtopointoutthatthemethodemploye! doesnotnecessarily give anymeasure oftheunsteadiness oftht current;itmaygivesomething—it maygivesomeindication 19] DISCUSSION. no snunsteadiness ofthemagnetic field—but itdoes notgive thems.wore. insteadiness ofthe current inthe main circuit. ‘The Victoria machines are alluded to,and some tests were made which gave efficiencies varying from 77to86. Well, ‘Mr.Murray says theefficiency ofthese machines islessthan was expected. Ithink probably theefficiency wasrather higher than appears. You, Sir[referring totheChairman], have made tests ofsimilar machines, and, Ithink, have obtained ahigher result than that. Buttheefficiency ofamachine ought tobetaken, of course,atthefullload.Inthiscasethehighestresultwasgot with thelargest load, butthemachines were capable ofrunning 1good deal higher. They were sent out capable ofworking really actually higher than what was supposed tobethenormal output, Ihave here aprivate letter from Mr. Fletcher, the ‘engineer who, under Mr.Murray, carried outtheactual work, andIwilljust quote hiswords asaproof oftheefficiency ofthe machines. He says: “The Victoria machines worked very “nicely, and were allthat could bedesired. Ibadoneofthem “working forabout twomonths with 30per cent. above what is “supposed tobeitsnormal output: itranvery nicely, andonly “alittle warmer than the others.” Idonot wish toquestion theaccuracy ofthetest, which, ofcourse, rests upon theaccuracyofthedynamometer; butIthinkthatpossiblyacarefulexamina-tionofthis instrument will show that there were some causes of friction that did not appear, except asreducing the apparent efficiency ofthemachine. Professor W.E,AYRTON: Itappears tomethatifthrough Petqwor theprimary circuit ofaninduction coilthere besent astrong steady current, onwhich issuperimposed acomparatively small intermittent current, the Intter will produce but asmall effect ininducing acurrent inthesecondary circuit iftheinduction coil befurnished with aniron core themagnetic circuit ofwhich is closed, Inother words, atransformer with aclosed magnetic cireuit, which isadmirably adapted forproducing asecondary current byalternations inthedirection oftheprimary current, “Weunsuitable forgiving asecondary current bysimple varia- tionsinstrength,withoutalteration ofdirection oftheprimary 750 “LIGHTING OF CENTENNIAL EXHIBITION, Ere. (Nor. rolome current, And thiswastheexplanation Igave thelateComteda “1Moneel, in1878, when hementioned tomethat hehadfound that completing themagnetic circuit ofanordinary Ruhmkort coilactually diminished theeffect ofthecoil. Ifurther pointed out tohim that ifhecombined with his closed maguetic circuit thereversing contact-maker that Idevised forRuhmkott coils when inIndia in1870, hewould obtain farlarger effets than were usually obtained with aRuhmkorff coil. Mr.Mather, who made the experiments forProfessor Pery and myself in1882 towhich Mr. Mordey referred, haskindly looked uphisnote-book, and heinforms methat thefollowing were the details ofthearrangement :—The induction coil, “discontinuity meter,” had initecentre abundle ofsof-ina wires 7§inches long, and thebundle hadadiameter of1inch On this were coiled some 10or12layers ofdouble-cottor- covered No.16copper wire, bringing thediameter to3inches; and over this twoseparate layers ofinsulated No. 10copper wire, their ends being leftoutsothat these twolayers could either be joined inseries orparallel; the diameter ofthefinished ail being 3§inches, ‘The current from either aBrush oranEdin machine wassent through thelayers ofNo.16wire, which actel astheprimary coil, and aSiemens dynamometer wasputinthe secondary circuit.‘TheBrashmachinehadeightseparatecoltonitsarmature, andtheEdisonmachine manycoils,theemet number notbeing recorded. The following results were obtained :— Baus Macuixe, restart Inge 228|sna|; wii |lapene”|Sen”|attr|See,|Ratan |fsseee"|SR|esr|| }|aoe|0|0084|newire]208|Secondarycoilsofi || |ctionclaerin 500, | |0086 =. 22 nom om mo||OOHBehohe |Em[ee|ID]tebtfap 1689.3 DISCUSSION. aL Eotsox Macarve, Prtemer —_ iynom |nevttions| Primary |nesaneeto]C2|Secondary|suliie,|Sper|“Grea”|pile,|ampere|MAR |e010s|oome0|Secondarycolsofii ductioncoilinparallel.||0 108|0054|Fine,|0 »oe»Veaoatout| 2|00426[Thick|0 oktaete]j |oom[ioe |9 aome ‘There isnorecord ofwhat kind ofresistance was introduced into the primary ordynamo circuit tobring the current tothe given value; but, asour object was toascertain whether there wasany difference between the constancy ofthe currents pro- duced byaBrush and anEdison machine, and notwhetlier the difference, ifany, could bewiped out, itismost probable that theresistance employed wasanon~inductive one. In this same note-book ofMr. Mather’s Ifind the resulta of along series ofexperiments that wemade atthat time onthe efficiency ofaBrush machine when used asagenerator andas 4motor;anidastheseexperiments, liketheonesgivenabove, have, asfarasIremember, never been published, itmay be useful togive hereashortextractfromourtestsoftheBrush dynamo when used asagenerator, inorder that acomparison may bemade between the results now obtained byMr.Murray andthose obtained byusin1882, Itmust not, however, be forgotten thatthearmature oftheBrush machines weused was made ofmalleable cast iron, whereas inthe Brush machines tested recently byMr. Murray the armature core ismade ofthin ‘oft iron ribbon. The transmission dynamometer employed wastheonedevised byus,and which isdescribed intheJournal oftheSociety of Telegraph Engineers in1881. 32 LIGHTING OF CENTENNIAL EXHIBITION, Ere. (Sor. 1a rote September 15th,1882.Brush DyxaMo 48GENERATOR. so|Curtin|roinva|Hometoner|aemae maus|ssa|ac.sa!¢ 708at234734947h|Bm ns 2215 45 498 °729 13 aT 356 507 #708 us srt|330 su| [Nore—Ater «reference tobisvotes, Profewor Ayrton hasamend!bi cciginal remarks,andmadesomevaluableadlitionsthereto—Eo-} Xam, -_-Mr-J.8.Rawontit: Ididnotintend toriseatpresentS,‘asIthought itwould berather anunfortunate circumstance if tthecriticism onthispaper fellentirely into thehands ofrepre sentatives ofthe Brush Company. Itisnot well forscientfe subjects tobetreatedfromatradepointofview;andalthoagbweareready enough toselltheBrush machines, andtoconvince everybody that they aregood ones, Idonotthink thisInstitution would besatisfied ifwewere tospeak ofthesubject ofthispaper without regard toscientific considerations. Ithink manyof oweagreat deal toMr.Murray forbringing this subject befar theInstitution. ‘Thereisnomachine, Ithink,atpresentbef thepublic which hasbeen solargely illustrated, orgeneral . described, and, perhaps, solittle understood, astheBru} machine. Veryfewpeoplehavehadinformation astowhatits efficiency was,—what itscapabilities were,—excepting thati burned aconsiderable number ofarelamps inseries,andtbstit took very little notice whatever ofdistance. Ithink Mr.Money Doasts that, with anordinary No,8LBrush dynamo, heoot} maintain anare lamp burning inYork Station from o works atLambeth; and Ithink that another calculation i thataNo.8Lmachine willwork40lampsinseries,place! atintervals ofonemile, onsomething likea7/14 wire. Tx scientific points which areraised inthis paper aresuch the wecannot very well discuss them except generally. ‘Whatstrikesmeasbeingmostpeculiaristhatthetests 1880) DISCUSSION. 708 thevarious machines differed veryconsiderably indeed—so much we tothat Ithink Mr. Murray himself was astonished with the results hegot, Allthat Icansayisthat thetests wehave been inthehabit ofmaking attheBrash Company's works have been souniformly regular, there having been scarcely ever more than abalf oronepercent, difference, that finally weceased tomake them, We now make the two tests that arenecessary forthe mannfacture—that is,theinsulation test andthegeneral output test: theefficiency tests have been made sooften andwith such close uniformity that wehave ceased torepeat them. With practical exhibition work Ihave had aconsiderable amount ofexperience—at theColonial Exhibition, atManchester, andattheGlasgow Exhibition; and although there were one or ‘wolittle accidents there, they were clearly traceable, inatleast, twocases, toleaky roofs. Anybody who hashad electrical work todoatanexhibition knows very well that when thedynamos arefirstbrought into theshed theroof isgenerally inanincom- plete state: you have toput upwith everything that comes through, beitsnow, rain, orhail; andyouarevery clever indeed ifyou can prevent some portion reaching thearmatures ofthe dynamos. ‘The Melbourne dynamos—about 40innumber—were made inabout three months from thedate oforder; they were shipped offasrapidlyaspossible,hadalongseatrip,wereputto work intheExhibition, and thelighting was carried outsatis- factorily—so much 60that Mr.Murray has, Iunderstand, bought thedynamos, Ithink, therefore, that hehas given usvery good testimonials; and Ihave nodoubt hewill become aconstant customer. ‘With regard totheVictoria machine. Mr.Mordey haspretty fully dealt with that question, and Ireminded him when Iread theproofofthispaperthatabouttheonlyoutsidetestwehaveeverhadofourdynamos isaverycarefulonethatyou,Sir,made some eight months orayear ago, when, ifIremember correctly, ‘youputtheefficiency atsomething over 90percent. Idonot with totake upany more ofthetime ofthis meeting, asIthink there aremany members here whomay liketoaskquestions andgiveustheirviews;andifthereisanything 14 LIGHTING OFCENTENNIAL EXHIBITION, Ere.(Nor. wc. further theBrush Company canadd intheway ofexplantio“ror upon anygeneral questions, Ishall bemost. happy toreply. (i course, Mr.Murray isnot here toexplain tousallthose litle details one would like toknow, especially about thedynam metric pulley, which appears tometobethe crucial os. Mr.Mordey remarked this morning, andIthink very troly,tit ‘Mr.Murray's method oftaking thespeed was somewhat dele tive: hetook itbyaform ofcontinuous recorder. Ithink te betterplanwouldhavebeentohaveusedatachometer, sotit theexact speed might have been taken foreach measuremet made. Inotice inthediagram attheend ofthe paper givisg the full record ofspeeds, that even inthe “fairly stess” diagram there areconsiderable fluctuations, and Itake ittt with aemall load the fluctuations would have been considentis more than areshown bytheexperimental records.* Primer Professor W.E,AYRTON,F.R.S.:Inviewoftheveryvaluable‘wren’ servicesthatBrushmachines haverendered toelectriclightiog, might Ibeallowed tosayoneword more toassure Mr.Raworth ‘and Mr.Mordey that Iwasnotfinding fault with theBrash machines? Ionly spoke astoamatteroffactregandingthe measurement ofthe discontinuity, and Idid not express as opinion astotheadvantage ordisadvantage ofthisdiscontinuity. May Isuggest oreason forsuspecting that theaccuncy& ‘Mr.Murray'stestsarenotasgreatasthefiguresinhispapet ‘would entitle onetoimagine? Nooneislesslikely thanmyee! tofind fault with praise ofAyrton and Perry ammeterssnd voltmeters; butwhen aperson says that anAyrton andPerry spring ammeter iscorrectto‘01percent.—thatis,1in10,000—it shows thattheexperimenter must possess amicroscopic power of reading theammeter that Idonotunderstand atall,andit shows anamount ofaccuracy inacommercial instrament which | isequally surprising. Youmust remember thatthescaleofsch| ‘aninstrament isabout 5or6inches long, and toread to# ten-thousandth partofthedeflection when theactual deflect Some farther observations contributed byMr.Raworth willbefocal p.761.—Ep. 1882) DISCUSSION. 188 isperhaps only 3inches means reading that deflection with aProtesor microscope; andyouknow youdonotsupply witha£4or£57" electric meter a£20 microscope toread itsscale with. Professor S.P,THomrson:OnereallyoughttomakeaveryFrtcwor clear, fine distinction between machines forarc lighting and machines forincandescent lighting injudging oftheir merits. They aredesigned upon different lines;theynecessarilyhavea different commercial efficiency, because the degree ofsaturation isdifferent. Asarclighting machines aremadeatthepresent time—because they are designed with drooping characteristic curves—they cannot possibly show under testashigh anefficiency asmachines made asincandescent lighting machines aremade at, thepresent time. Iamleaving alternating-current machines catirely outofcomparison, andamdealing only with continuous- carrent machines ofthe two kinds. The existing arclighting machines are, forthemost part, oftwotypes, oneofwhich hasa very small number ofsections onthe armature, and with, con- sequently, avery small number ofparts inthecommutator. The necessary consequence ofthat willbethat theelectro-motive force generated inthe revolving armature, assuming the magnetic field tobeconstant, will necessarily have very considerable fluctuations;butitdoesnotatallfollowthatthecurrentwhich remults will fluctuate inthe same degree astheelectro-motive foree fluctuates ;foreverything thatthereisinthecircuit,thatpossesses self-induction—any coil, especially anycoilwith aniron core, most ofallsuch large coils asonegets intheseries windings onthefield magnets themselves—everything ofthat sort inthe cirenit tends tosteady the current. The fluctuations ofthe current arenecessarily, therefore, very much less inpercentage value than the fluctuations ofthe electro-motive force. Idonot doubt fora moment that Professor Ayrton sixyears ago gota very considerable discontinuity intheparticular current that he measured, butIdoubt whether hewasmeasuring exactly under the circumstances that would befairly representative ofthe current inanarelighting system. Ido notknow whether the current wasrunning through aseries ofarclamps every oneof which badanelectro-magnet init,and running round thefield 138 LIGHTING OFCENTENNIAL EXHIBITION, Bre (Nor. Zptmor magnets ofthemachine, which would greatly steady thecore And whether the arclamps were many orfew, Professor Aytsa doesnotsay.IaminthehabiteverydayofusingasmallBrat machine which lights only onearclamp. Isometimes light arclamp, which isonmylecture table forprojection purpst, from this Brush machine, sometimes from other dynama which areintended forincandescent work; and Ialways kur when the Brush machine ison,because Ihear ahumming: there issufficient fluctuation toproduce ahumming sound But ifIhadhalfadozenarelamps,or50lampsinthecirc, fedbyalarger machine with more massive field maguets1 certainly should bevery much surprised ifIheard thesam humming. Idonot think that the current from thelay Brush machines has anything like the amount offluctustiom— Icannotcallitdiscontinuity—that thecurrentfromthesmall Brush machine has. ‘The machines made on the Thoms Houston plan ought toshow more fluctuation than themachine made ontheBrush plan, because they have fewer portions inthe commutator and fewer sections on the armature. But inthe case oftheThomson-Houston currents Ihave never heard of135 complaint ofwant ofcontinuity, orofamount offluctuatice, because the field-magnet coils ofthese machines possess 12 unusualamountofself-induction. ButthisIknowtobethe case intheThomson-Houston machines: ifyouseparately excite thefield magnets, andtrytowork, say,40oF50arelamps sothst thecurrent goes through nothing but the armatureandthet Jamps, that machine refuses towork. Itrequires thesl induction ofthefieldmaghets tobethrown intothemii circuit forthat machine todoitsduty. Ishould notbesurprised ifsomething ofthesame kind occurred with theBrush machine. Ishould not beastonished ifitshowed more fluctuationthst underjordinarycircumstances. Isthere not some mistake inthe relative values ofthe magnetism andthepermeability? I'am very much surprised to findthere isanykind ofiron used inadynamo which onlygive 4,800 lines to.the square centimetre when themagnetising fort is122, There would seem tobeamistake of10times; thatis 189) DISCUSSION. 11 tosay,Ithink thenumber given forHis10times toogreat, andprotemor thepermeability, which isputat36,is10times toosmall. TesPen Mr. R.L,Cousens: There are one ortwo remarks Ishould x. liketomakeuponthepaper, Mr.Chairman, Coonan, Thave had considerablé Colonial experience with theBrush tmachine ;and although Ihave notcome here forthepurpose of praising it,I cansaythat, sofarasmyexperience hasgone, it appears tome,from aColonial point ofview, that itisthebest. high-tension machine inthemarket. There areoneortwopoints which Idonot quite agree with inthepaper. Onpage 719 Mr.Murray says that the reason theHplates ofthearmature broke offwasfrom thecontinual bending towhich they were subjected inpassing through thedifferent parts ofthemagnetic field inwhich they rotated, Forseveral years past Ihave hadunder mysupervision Brush are machines forstreet lighting, thearmatures ofwhich were in1884 replaced bythe laminated thin-plate type, and which were, I believe, thefirst, made toorder. Ihave never found anyportion ofthese flyout,and inmy opinion thefaulty ones alluded tobyMr.Murray were notbuilt upsufficiently strong, and theplates would have flown outifthe armature wassimply rotated, with nomagnetic field through the same. The insulation ofthearmatures wascertainly insufficient, and frequently flames, alluded tobyMr.Murray as“electrical fire- “works,” were tobeseen issuing from them; but this was remedied bycontinually improving the insulation whenever any ofthe machines broke down. With regard totheapparent decrease inthecurrent asthe eveningadvances, Ihavefoundthesametobethecase,theprecisereason forwhich Iamunable togive, excepting that there isa greater resistance throughout theentire circuit from riseintem- perature, From what Mr. Murray states tobethereason—viz., thelessening action ofthefinewire inthedifferential coils ofthe lamps, and therefore theincrease inthelength ofthearc—he would seem toimply that thelamps didnotthen regulate sowell. Thave found, ontheother hand, that thelamps seem toburn VOL. XVII. 52 168 LIGHTING OF CENTENNIAL EXHIBITION, Ere. (Nor. lth a, better astheevening advances, andeventwoorthreelampscan *beadded inacircuit with, say, 16lamps without practically diminishing thelight emitted from these, therevolutions ofthe machine being kept constant. This ismore particularly notice ableonacoldeveningthanononewhenthetemperature ishigThavealsofoundittobethecasewithBrushlampsoftheoldform ofwinding, theshunt resistance ofwhich isconsiderble, viz,about350ohms.Thisbeingso,Iscarcelythinkthatthe reagon given byMr.Murray canbethecorrect one. With regard tomachines ofother types, Ihave found them togive farmore trouble, andtobemore liable tobedamaged by lightning, than theBrush, although thelatter hasbeen byno means infallible, Some series machines forarclighting ofthe : drum type, with Gramme winding, which were sent outforare lighting purposes, gave anexceeding amount oftrouble from the breaking ofthewires round thecommutator; and when these portions were thickened, they broke within thecoils themselves. This fault, Ibelieve, has been alluded tobyProfessor 8.P. ‘Thompson inhispaper on“Diseases ofDynamos,” but noreason ‘wasassigned forthis, norhave Iheard ofany.* ‘The armatures ofthese machines were replaced byothers which were wound with stranded wire instead ofsolid, which to agreat extent remedied thedefects, tun, Mr,W.B.Esson: Mr. Chairman and gentlemen,—Though Ihave never made Brush machines, Ican appreciate thoroughly this paper ofMr.Murray, which isexceedingly useful andof very great service tous,since itdescribes theresults actually obtained from aninstallation ofconsiderable magnitude. Inall there areabout 900arelamps, and2,000 incandescent lamps, the total indicated horse-power being about 1,500. ‘There areseveral anomalies, however, inthefigures which *Since themeeting oftheInatitution Ihave consulted with other, andhave Dard diferent opisions exprowed axtothoreason ofthis faalt inthis(yp of snachine, afollows:—(1)Thewirbreakfromairresistancewhenthearmatarirotated. (2)Fromstraylinesofforeeleakingfrompoletopole. (8)From the shatt spindle being toacartain extent flexible, tewires ‘being therefore strained, andthus frequently breakiog.—R. L. 189] DISCUSSION. 189 ‘Mr.Murraygives.Forinstance, inTableC,theefficiency ofar.ton. themachine hotiscompared with theefficiency ofthemachine cold; andinonecasewefindthat theriseduring therunisfrom 1037 percent. when itiscold to82°56 percent. when itishot— avery great jump. We are quite prepared foradifference of ficiency like from 796 to82,orfrom 77-8 to81°9, butthe extraordinary difference between 70-37 and 82°56 requires some explanation. Itisapity that when Mr, Murray went tosuch great trouble intaking theefficiency ofthese several dynamos, hedidnotgo alittle farther andgive ussimultaneous readings oftheindicated horee-power ofthe engines, and the average electrical power appearing attheterminals ofthedynamos. Ofcourse itwould bevery difficult—one might sayimpossible—to measure simul- taneously thetotal electrical power being given by50dynamos; batwemight have been given some rough idea ofit,andwe might have been told atthesame time something about thefuel consumption. Fortunately, however, electric light engineers areable to state pretty definitely the ratio between theelectrical horse power given atthe terminals oftheir dynamos and thehorse- power indicated inthecylinders oftheir engines. Inthepresent installation, ifweassume—and theassumption israther over the mark than under—that the power actually delivered from the countershaft pulleys is80percent. oftheindicated horse-power oftheengines, theelectrical horse-power furnished bytheBrush machines comes outatabout 60percent. oftheindicated horse- power, while fortheVictoria machines itcomes outatfrom 67to68percent. Well, from actual trials which wehave made with engines driving Phonix dynamos through countershafting, without any trimming orextra precaution toensure thebest results, wehave found that wegetfrom theterminals ofour are machines 68per cent. ofthe horse-power developed inthe cylinders oftheengine; while with ourincandescence machines wegetfrom 72to73percent. That is,driving from counter- shafting. Driving with thedynamos belted directly tothefly- wheels,weshouldgetaconsiderably betterresult.IhopeMr. 100“ LIGHTING OF CENTENNIAL EXHIBITION, Ere. (Nor. 1 too. Murray will give usthefurther results which hepromises in thelast paragraph ofhis paper. Wecannot have toomany reliable figures from actual working. With reference tothegeneral question ofconstructing ae machines, there seems tobe an impression abroad tht such machines can only be made intwo ways—thet is, according tothe Brush orThomson-Houston design. ‘Thee never was agreater mistake, Speaking from experience, I may say that itisquite easy toconstruct arcmachines giving from 1,000 to1,500 volts, having ordinary Gramme wound armatures connected uptoGramme commutators the bars ofwhich are insulated with mica inthe usual way. If the machines are properly designed there isnot the slightest trouble with the commutators, and past failures have simply resulted from ignorance ofthe proper methods ofconstruction. Professor 8,P,Thompson hasjust said that inconsequenceof the commutators having sofew sections inthe Brush sad ‘Thomson-Houston machines, these are necessarily inefficient, ‘Very well; thebest: waytosecure efficiency istogive upthose machines having fewsections, andtoemploy only Gramme-wound machines, whichhavealargenumberofsectionsandahigh efficiency. kop -Mr. Gisuznt Karp: Mr. President and gentlemen,—t shall notfollow theprevious speakers intheir attemptstoshowwhat efficiencycan beobtained with different machines, We allknow that every inventor ordesigner ofdynamos thinks hisown particular machines thebest. Ishould, however, like topoint out that the efficiencies given inthe paper, apart from any interest they may have comparatively with that ofother machines, arevery instructive inthemselves. InTable Awefind forthesolid-frame machine theaverage efficiency offivetests, 78°7 percent, with thethin-plate armature. With thethick- plate armature, inthesolid-frame machine theefficiency is6975, percent., oradrop of9percent., simply duetothefactthatthe armature plates are50instead of22mils. thick. Intheotber (rail) machine wehave asimilar ratio—77-43 to69°84—say about 8percent. Now 8percent. additional inefficiency isvery well 1882] DISCUSSION. ta worthhaving,evenattheextracostandtroubleofusingthina.kar.plates forthearmature core. Most dynamo makers have already, byactual experience, come totheconclusion that theiron inthe armature cannot be laminated too much. Most makers have begun with rather thick plates—about 16totheinch—and have gradually come down tosomething like 40totheinch, There are,however, several objections totheuseofthin plates. They ‘arenotonly more expensive, butalso mechanically weaker, asis shownbythethrowing outofthePacinotti projections; andthe amount ofspace wasted byinsulating theplates isgreater. The objection ofexpense isnot avery strong one, forthe cost ofthe material inthe armature isnot very great inpro- portion tothe cost ofthe material intherest ofthemachine; andtomake upagainst this slight increase ofcost, there isthe ‘consideration that ifwehave thin plates wearesure ofgetting good iron, because bad iron cannot berolled very thin. The objection astomechanical weakness Iamglad tohear Mr. Mordey sayhasbeen gotover, but hehasnottold ushow. ‘Then astoinsulating space, thiscan, byvarnishing theplates, be reduced tosomething like 4or5mils, foreach alternate, orsay 2}mils. foreach plate; sothat only 8percent. ofthewhole space need belost byinsulation. Lately some makers have given upinsulating theplates altogether. Regarding the question ofthe fuctuation ofcurrent, there seems tobesome diversity ofopinion. Professor Ayrton says there isadistinet fluctuation; theamount ofitcanroughly be tested byaninstrument heand Professor Perry had constructed forthe purpose. Ontheother hand, Mr.Mordey hasmade tests with aninstrament supposed tobeofthesame construction, and found that there was nofluctuation inaBrash current. This result was probably due tocomplete saturation ofthecore ofhis transformer. ‘Thetheory generally advanced insupport oftheview that thecurrent from aBrash machine issteady isthat themass ofthefield magnets issufficient tosmooth down theattempts of thearmature toproduce fluctuation, The field may becompared tothe fly-wheel ofasteam engine; and although itissupposed that the fly-wheel makes the motion absolutely uniform, a 302 “LIGHTING OFCENTENNIAL EXHIBITION, Ere. (Nor. 1 4s.ayn.moment's consideration willshowthatitcannotbeso.The office ofthe fly-wheel istohelp theengine over thecentre Whatever resistance against, turning there isfrom thebelting other machinery which the engine drives, that resistancehasto beovercome bythefly-wheel whenthecrankisgoingoverthe centre. Everybody knows that abody inmotion canonly over come resistance bybeing slowed down, and there must bes certain retardation ofangular velocity inorder that thefly-whee! may give outwork and gettheengine over thecentre. The action ofthe ficld maguets ofaBrushdynamoissimilar.As long asthe current remains ofabsolutely constant strengththe field cannot push iton,and weare therefore forced tothe conclusion that there must be acertain fluctuation inorder that the field may have atendency toprevent fluctuation. Since theremustbesomefluctuation, nomatterhowheavythefield, itisobvious that the determination ofthe electrical output by simplymultiplying currentandpressure isnotabsolutely corred. Such adetermination isinthe same manner incorrect asitwould bewith analternating current, butnottothesame extent, The errortherebyintroduced maypossiblybeveryslight,butforvery accurate measurement itought tobetaken into consideration. x,Aader.Mr,@,L,ADDENB200KE: ImayeaythatIhappened tobeinMelbourne about three days after theExhibition closed, andbad ‘anopportunity ofseeing allthe machinery. Ididnotseeit running, butIsawitasitwas,andthereforeImayperhapsbe able tosupply from memory afew ofthe details which sme members have asked for. The number ofboilers was 12. They were allmade inthe Colonies; and Imay saythat nearly thewhole oftheboilers in theColonies aremade intheColonies. The type hasnodoubt been evolved partially from marine boilers and partially from locomotive boilers,Itisafire-tube boilerofthemarinetype,‘only rather different indetail, being much longer, and thetubes are returned asin@marine boiler. The flue iscarried back from thefront bythin iron coverings riveted down ontothe side faces oftheboiler. The boilers were mounted onwrought- iron supports, sothattheyrequired nobrickwork atall:they 1682.) DISCUSSION. 108 veresimplystoodintheirplace;andImaysaythattheboilersue,Asse and their arrangement made avery neat-looking job, and gave satisfactory results. Idid not hear whether any tests of evaporative power oranything likethat hadbeen tried with the boilers, butthegeneral tests ofsuch boilers arefully uptothe standard. Theengines spoken ofinthepaper were anexact copy ofthewell-known type ofMessrs. Marshall, ofGainsborough. I understand that themeasurements were taken from anengine belonging tothe Australian Electric Company, thevalve gear being carried outonidentical principles, which are,Ithink, well known tomost people. ‘The engines, asfarasworkmanship went, appeared tome tobewell turned out. They were not compound—simply paired—and therefore, ofcourse, notvery economical. Ihad noopportunity offinding outhow thevalves, &c.,worked, but toallappearances they were fairengines, and such aswould have been creditable toEnglish manufacture. Imay saythat Ithink, onthewhole, Colonial workmanship isvery good: they know thevalue ofgood work. Intheengines there were two cylinders, one oneach side, andinthe centre alarge fly-wheel, boxed inwith wooden boxing;itwasgrooved, asthepapersays,for13pulleys. These fly-wheels were, Ithink, 20feet indiameter, and they weighed 20tons; they were, ofcourse, cast and made inthe Colonies, OntheMelbourne cable tramway system 30-ton fly-wheels, nearly 30feet indiameter, areinuse. These large driving wheels drove allthepulleys, which formed themost portentous-looking line ofshafting that Iever saw; itwas, asstated inthepaper, 200feetlong, divided into three parts bypulleys with clutches, 0that itcould bedriven from either engine; and Ithink that usually twosections were driven from oneengine: thus there were something like140feetusually driven from oneengine. Theloose pulleys were onthecountershafting, sothat thebelt wasnot running atallwhen thedynamos were thrown out,andofcourse thisnecessitated alotofextra work, and made thecountershaft- inglook much heavier, ‘Thedynamos were arranged between the engines andthecountershafting intheusual way. The whole 764 LIGHTING OF CENTENNIAL EXHIBITION, Bre. [Nor Mi Me,Alte.wascertainly carried outinaship-shape andniceform,axl altogether looked neater than anything wesawattheexhibitios: inLondon. This,however,isnodoubtaccounted forbyalthemachines being ofthe same pattern, and designed towak together. Unfortunately, asIhave said, Ididnotseethe lighting, and therefore cannot say much about it. But from what| heard from outside electrical engineers itwas, onthewhole, satisfactory. ‘This installation was one ofthe largest which have been putup. Perhaps you arenotallaware ofthefut that thearea ofthecovered portion oftheMelbourne Exhibition wassomewhat larger than anyofthose lately held inLondon— the Colonial, Healtheries, and soforth; and, asfarasIcould see, thething was aswell done asawhole. Idonotmean only with regard toelectric lighting, buteverything else, The lighting ofthe picture gallery, which Mr. Murray alludes to, was very good. The shades had been nicely arranged #0that thelampsjustthrewthelightdownontotheedgeofthe floor below thepictures, and thelamps were notvisible tothe public, Idonotknow whether Ican giveanyfurfher details. I may saythat there wasatransformer plant ofMesers, Gast working intheExhibition: itdiditswork very wellduringthe Exhibition, and was, Ibelieve, afterwards bought byaMelbourne firm foratheatre installation, The Thomson-Houston people also hadatransformer plant atwork, which also didwell. OfMr.Murray himself, Imay saythat hehasbeen, Ibelieve, electric engineer oftheVictorian Railways fromthefirst,andtele- graphy inVictoria maybesaidtohave risen under hiscare. He has the electrical connectionstolookafterforsomethinglike2,200 miles ofrailway; and while hehasareputationforcarryingout everything satisfactorily andefficiently, heisatthesame time 8 most impartial man, Iamsorry tosaythatwhen Ilastsawbim hewasinpoor health. aetreen Mr,W.H.Preece, F.RS.: Iwould like merely toremark, Sir,that Mr.Murray sent hispaper over toEngland tomycare ‘and desired metolook after itforhim; and Ishould have bees perfectly preparedtohavesupported anyofhisviewsiftheyhal been severely handled orseverely criticised. 188) DISCUSSION. 185 TamquitecertainthatMr.Murrayhimselfwillfeelveryae.Preece pleased that hispaper has been sowell received, and weasan Institution arevery much indebted tohim. We are, indeed, fgledwhen ourmembers indistant parts supply uswith papers ofthis description, and communicate touswhat hasbeen done inother lands, Tam sorry Mr.Murray hasomitted one important point: there islittle ornothing about the bebaviour ofthe arclamps, andlittle ornothing about the amount oftheillumination, What wereally dowant toknow istheexperience ofmanagers ofexhibitions, andofthelighting oflarge spaces, oftheamount ofillumination that isgiven bythedistribution ofarclamps at various distances apart. You willpretty well know that Ihave hadagreat deal todo with exhibitions—the Inventions, Colinderies, Liverpool, Man- chester, and others—and Ihave always laid itdown asalaw to totally disregard the nominal light-power given byarcs, orany other indication ofcandle-power, and have taken only thepower consumed inthearelamp. The power consumed inthe Brush arelamp is500watts, andbyallowing awatt to2square feetof surface area, Ihave found that you getabrilliant illumination; bygiving3squarefeettoeachwattyougetalittleless;andso, byregulating thesurface tobeilluminated tothenumber of watts, you areable toregulate your illumination asyouplease. This rule hasbeen frequently followed, and itismost desirable that weshould have information astotheworking ofsome such rule, sothat wereally allmay beguided intheillumination, say, ofrailway stations. InParis, where 13acres were covered with glass andilluminated, Ifound, intheGaleries desMachines, that the amount allowed was less than that Ihave named: itwas something like 5square feet per watt; and Ithink that if those who have charge ofarclights will bear that inmind, and give ussome information onthat point, they will help us very much. ‘OnbebalfofMr.Murray,IamsureIamonlyexpressinghis views when Isaythat heisextremely indebted totheInstitution forreceiving hispaper sokindly. 768 “LIGHTING OFCENTENNIAL EXHIBITION, Bre.(Nor.1h be ‘TheCarma: Mr.Raworth, Ithink,referred toatest1 Nowe"made ofthe Victoria machine some little time ago. Asthe results ofthetest were satisfactory, Ithink that theBrub Company areentitledtothem.Ishouldstatethemachines were| specified tohaveanefficiency of92percent.,andtheycames near inthetestthat Ididnotfeel myself justified intakingthe slightest exception tothem. ‘Those machines had anefficiency«f about 89percent, atthemaximum load. Onthemotion oftheCuamRMan, ahearty vote ofthankswi unanimously accorded toMr, Murray forhisinteresting paper. ‘Themeeting then adjourned. | | | | | ' tor MELBOURNE EXHIBITION. Notes onMr. Murray's Papen. Dynamos Nos. 454 to467 were not made bythe Brush Company, and asIhave nomeans ofascertaining their exact construction and windings, Ishall deal only with the second series, covering dynamos 1846, 1840, and 1849 with thin plates, and1843, 1861, and 1856 with thick plates. These dynamos, excepting inthethickness ofthearmature plates, areknown to bepractically identical inconstruction. Dynamos Test. Highest efficiency—thin plates ..,828 percent. Lowest, » we HT y Difference... 81 Highest efficiency—thick plates... 73°31, Lowest, » oo OTB Difference ... 5°75 Average efficiency—thin plates... 7-43 » » thick... 6984 Difference ... 7:59 ‘The variations inthe above results ineach setarevery cemarkable. Inthecase ofdynamos 1840 and 1849, their efficiencies are jiven (hot) at828 and 74°81: difference =999 percent.;but mmreferring toFig. 1,wefind that thevalues ofthemagnetic ‘reuits forsimilar currents arepractically identical. Any actual lifference inefficiency must therefore besought forinthe amature. Unfortunately, noexperiments bearing onthis point 1reference tothesetwodynamos arerecorded; butonecannot esist the conclusion that aloss ofpower inone armature ceeding that intheother by7-99 per cent., equivalent to ‘bout 2horse-power, would have produced such amarked neease oftemperature astohave excited special attention. One ofthemost remarkable recorded results oftheexperi- nents istheimprovement inefficiency asthemachine increased 768 MELBOURNE EXHIBITION. intemperature. Thiswas,ofcourse,tobeexpected toaslight extent, owing totheincreased resistance tothegenerationof eddy-currents inthe iron ofthe armature, assuggestedbythe authors butonreferring tothetables thefigures given appar quite inexplicable. Referring toTable C,wefindtheaverage increase ineffciens, when hot, isgiven as— ‘Thin plates... ase60 percent. ‘The maximum increase isgiven— No.1840,thinplates......11‘1percent.; whereas there was anactual decrease inthe case of No,1856, thick plates, of.., 1°3percent. Assuming thatthedifference wasduetothevariationofthe eddy-currents inthearmature, onewould have expected tofini thegreatest variation inthethick-plate armatures, inwhich the eddy-currents would undoubtedly bemore serious than inthee with thin plates; but the facts asdisclosed bythe experiment: ‘arequite inthecontrary direction. The thin plates shor, theaverage, animprovement of6percent., while thethit plates show only 2-44 percent. ‘The maximum improvement inthethin plates was111p= cent., andoneofthethick-plate armatures actually deterioate! 1°3per cent. Itisclear, therefore, that thevariation cannot beaccounted forinthe armature. The only other place where the variate ‘canbelooked forisinthetipsofthepole-pieces, butthemas these issosmall that itispractically certain that any excesire lossofenergy concentrated inthem would have raised thei temperature within the30minutes ofthe“cold” experimen’ and thereby have nullified thevariation. From thelast series ofexperiments with theBrush dypim: ofwhich details aregiven, Mr.Murray concludes that thecur was not steady, and that the total fluctuation was about 37 cent.5butitisevidentthatthedifference betweentheobser! lossofpotential andcurrent-resistance may have been dueei MELBOURNE EXHIBITION, 169 tofluctuation inthe current, orfluctuation intheresistance of themagnetic circuit. Mr.Mordey's experiments show that the fluctuation ofthecurrent from No.7Lworking onafullload of lamps isextremely small, ifnotentirely inappreciable byany instrument less sensitive than atelephone; but weknow, from theconstruction ofthe machine, that the resistance ofthe magnetic circuit isbeing varied constantly bytherevolutions of ‘hearmature, andthat thedistribution ofmagnetic density in thefield magnets varies inastill greater degree. ‘These varia- sions inthefield magnets arereflected onthearmatures, where heyproduce opposing forces,sothatitdoesnotatallfollowthat ‘ariations inthefield produce equal orcorresponding variations ncurrent. atNoe.1889. JOHN8.RAWORTH. m0 ELECTRICAL ENGINEERING IN AMERICA. [Nor.28, ‘The One Hundred and Ninety-eighth Ordinary General Meeting of the Institution was held atthe Institution ofCir Engineers, 25, Great George Street, Westminster, a ‘Thursday evening, November 28th, 1889—Sir Wmiux ‘Taomsox, F.R.S,, President, inthe Chair. The minutes ofthe Ordinary General Meeting held November 14th, 1889, were read andapproved. The names ofcandidates for admission into the Institution were announced and ordered tobesuspended. ‘The following transfer wasannounced ashaving been approved bytheCouncil :— From the class ofAssociates tothat ofMembers— Clement Joachim. ‘The following paper wasthen read :— ELECTRICAL ENGINEERING IN AMERICA. ByG.L.ADDENBROOKE, Associate. ‘The lasteighteen months ortwo years inAmerica have been much more characterised bycommercial andindustrial progress with already existing types ofplant and apparatus, than by« number ofnew inventions. Inventive ability seems tohave been more concentrated onimproving and elaborating already existing inventions, than ontheevolution ofnew forms and ideas. The leading types ofmachines andapparatus areallwell known toyou, and have been sofully described and illustrated that toenter into aminute account ofthem now would only be totraverse oldground. Itherefore propose todevote myself this evening tocentral station; construction, and internal work, andtheapplication ofelectricity totramway andother industrial Purposes, Finally, Ipropose tomakeafewremarks onthetone,| policy, and aims ofAmerican electrical engineers, andthose commercially interested with them inadvancing theapplication ofelectricity. | 182] ELECTRICAL ENGINEERING INAMERICA. m MyvisittoAmericawasnotquiteoftheordinarycharacter, asIcamethroughonmywayfromAustralia,andconsequently hadtotraversethewholebreadthofthecontinent beforereaching thebetter known towns onthe Eastern Coast. Iwas thus able to makeageneral,ifrapid,surveyofwhatisgoingoninthefar West andintowns remote from themanufacturing andengineer- ingcentres, aswell asinNew York andsome ofthelarger cities. Ovennean Work. Throughout thecontinent, from shore toshore, themain thing that forces itself ontheobserver istheprevalence ofare lighting. Hardly anywhat weshould callmoderate-sized village ontside theoilregion appears tobewithout itsarclamps, ‘Arelamps, when used forpublic lighting, areusually fixed on polesfrom20to40feethigh,according tothecapriceofthepartiesinterested intheirerection; ortheyareslungfromthe comers ofintersecting streets, inwhich case they usually hang lowdown;orthetowersystemisadopted.Whenpostsareused, theyareofrough pine, never painted, often outofthestraight, andwarped orbent. Ifthelamps arefixed, iron spikes are usually driven atintervals oneach side ofthepole, starting about 8feet from theground, and forming asort ofladder by whichtoreachthelamp.DuringmywholejourneyfromSan Francisco toNew York, Icanhardly recollect seeing oneship- shape, neat, and smart-looking post, whether forcarrying arc lamps orforany other electrical purpose. Everything hasa temporary andexpedient look about it,which isvery offensive to English eyes. Creosoting does notseem tobepractised, andthe onlyprotection applied tothepoles which Isawanywhere, con- sisted ofboards fixed round thebase and reaching about 6feet high; butthisprotection was,Ithink, only applied tolarge poles carrying circuits ortelephone wires. Thelamps themselves arealways entirely without ornamenta- tion, and are usually very roughly fixed onthepoles. Ifthe globes arenotclear glass they areusually only lightly frosted, 0that thelight glares inone’s eyes inallitsnakedness. ‘When thelamps arehung over the centre ofintersecting m ELECTRICAL ENGINEERING INAMERICA. [Nor2k, streets, four wires arebrought down from thecorners ofthe neighbouring houses, thelamp hangs from the centre, andis lowered and raised byapulley and cord. The wires from which thelamp hangs, thecord, and the leads allfestoon about in accordance with the chapter ofaccidents, and look, asweshould think, both slipshod and untidy. The only relief tothis state ofthings isafforded bythe towers. These towers areusually erected atthe intersection of streets, with thefour legs atthefour angles ofthepavement. ‘They areconstructed ofexceedingly light ironwork. Tapering gracefully upwards toagreat height, they areatany rate neat ‘andnotuninteresting-looking objects inthemselves, andcertaials form very definite landmarks. Three orfour arclamps hung round thetopofoneofthese towers shed asufficient light over ‘agreatareawherethehousesarenothigh.Withthismethod ofdistribution a20-light machine can beemployed tograt effect;especially asincountrytownsandvillages,wherethesetowers arechiefly employed, theobject israther tomake darkness generally visible, ifImay soexpress it,than toprovide bright illumination inwell-defined places. Byusing thetower method one’s eyes arenot. dazzled bythe brightness ofthelight,whichissomuchthecasewithlampsat alow level. Inmyown case Iused sometimes atnight togetquite an irritated feeling whén walking about, ifatalltired, from the incessant glaring inone’s eyes ofnaked (arc lamps, andthe impossibility offinding rest anywhere. Imentioned thiseffect totwoorthree Americans, butthey didnotseem tonotice it. Important, however, asthepublic lighting is,inthelarge towns itoccupies acomparatively inconspicuous place incot- parison with thenumbers oflights used forilluminating stores, saloons, hotels, restaurants, andallsorts ofpublic buildings. Are lamps aresolavishly used inthese places, both inside andout, that theprincipal streets areoften literally ablaze oflight. There were, asfar Icould ascertain, about 2,000 arclamps innightly useinSan Francisco, 3,000 inChicago, andI| ‘understoodthat3,500wereshutdowninNewYorkwhileIws| 182] “ELECTRICAL ENGINEERING INAMERICA. mm there, owing totheaccidents which hadtaken place, andofwhich #0many accounts have been published. ‘Themajorityofthearclampsusedoutsidearefixedbymeans oflight iron rods projecting from thebuildings, though some, ofcourse, arehung.Thecircuitsfeedinglampshungeitherinsideoroutsideof building, either rundirect from thenearest poles oneach sideofit, thusmaking anangle over andacross thepavement which isvery unsightly, orasortofbarrel-shaped glass insulator isused. This insulator hasahole through thecentre andagroove round the middle oftheoutside. The wire coming inone direction is passed through thecentre oftheinsulator, turned back onitself, andmade off; thewire inthe other direction ispassed round theoutside, and also made offonitself; thetwoends oftheloop arethen carried tothelamp, through which thecircuit iscom- pleted. Nearly allthe wire Isaw was insulated with “under ‘writers’ insulation,” Where lamps areused internally, this is often carried through wooden window frames andsuch other like places without anyfurther protection orcare. Once inside, the wireiscarried along ceilings andwalls quite exposed, butsepa- tated about 3inches bysmall wooden cleats. ,Sometimes instead ofcleats small white porcelain insulators areused, ‘AsfarasIcould learn, itistheinvariable custom forallcom- panies torun, fixup, and entirely look after thelamps they operate. The customer simply pays arent fortheuseofthe lighted lamp ofsomuch aweek, month, oryear, asthecase may be,and has nofurther trouble ofany sort. ‘The rates differ according tothetime thelamps aretobeinusoeach night. To seethat thelamps burn properly, linemen promenade thestreets allnight, andaresupposed toseeevery lamp ontheir circuits at certain intervals. Thecircuits themselves arealmost invariably carried onpoles erected along theedge ofthekerbstones. Afairestimate ofthe average height ofthese posts is,Ithink, about 30feet ofcourse someare much more, andmany less. Long wooden cross arms are fixednearthetopofthepoles,carryingfromtwotosix,andeven eight, insulators oneach arm, ‘These arms arenotsostrong aswe Vou, xvii. 53 74 ELECTRICAL ENGINEERING IN AMERICA. [Sor 2h should make them forthepurpose. Owing tothelength ofthe arms, andtotheir notbeing very securely fixed tothepoles, the weight ofthecircuits often pulls them outofthehorizontal. To seeawarped andbent pole, setcrookedly intheground,sndwith thearms atvarious angles with each other, and thewires onit allhangingindifferent curves,seemsaboutasdismalandwor begone apiece ofengineering ascanwell beimagined. Imustmakesomeexception, however, inthecaseofthe‘Thomson-Houston Company inNewYork,who,atanyrate,had ‘gotlight iron stays attached tothepoles andarms, atanangle fastogive more support tothearms and keep them intheir places. ‘The insulators aresupported onthearms bywooden pegs, which have ascrew-thread cutonthepart that projects. Onthis one oftheordinary green glass insulators isscrewed. Green glass insulators areused forevery purpose inAmerica, whether forelectric light, telegraphs, ortelephones. They arevery cheap, and answer thepurpose well. Still, the effect ofgreen ghs insulators onarms painted adull red, fixed onbare poles from which thebark hasjust been stripped, isnotesthetic. ‘Themajority ofthepoles forcircuits have nosteps provided forascending them. Onthe frequent occasions when itis necessary togouptothetoptomake alterations oraddner circuits, thelinemen useclimbing irons. ‘The tearing ofthe wood bythespikes ofthese climbing irons gives thepoles* jagged and frayed-out appearance from top tobottom. One could fancy they lead tothe habitation ofsome immense squirrel, ‘TheAmericans nearly always usesolid wire forrunning their circuits, covered, asIhave mentioned already, with “anderwriten’ “insulation.” Little trouble istaken tokeep thevarious circuits ononepole inregulation with each other,aslongastheydonot touch, Indeed, thewhole construction ofthepoles andarmsis 0flimsy that nogreat strain could beputonthewires, The poles themselves arefixed atshort intervals, often notmore thia 30yards apart, andseldom above 60. ‘The cireuits arerather hung uponthepoles than ranit| 180) ‘ELECTRICAL, ENGINEERING IN AMERICA. 73 thesense anEnglish lineman attaches totheword. Separate suspenders Inever saw used, except fortelephone cables. For lternating incandescent, circuits of1,000 volts several separate cables ofsmall sizearealsoused, instead ofrunning alarge one withsuspenders, ashasbeen thecustom here. What Ihave just said about short spans forelectric light ‘work also applies totelephone and telegraphic work: thespans used inthis work aregenerally also short. Inthelatter case, since the wires are light, they arerun pretty taut without patting much strain onthe arms, sothat there isnotmuch sag. Wires runinthiswayareofcourse notsoliable tosway about as longspansareinhighwinds,andmakeintermittent contactorget overlapped. Further, thewires, being runalong thestreets, are protected from high winds inagreat measure, and arenotpar- ticularly exposed tothem, asour over-house work is, This method oferection, therefore, rough asitis,and notwithstanding theenormous number ofwires which areoften suspended from onepole, gives animmunity from contacts and accidents that an English engineer atfirstcanhardly understand orappreciate. Forleading inwires into houses, orforputting onextra wires onapole—in fact,forgeneralpurposes ofthesort—ablockofwood isused the lower end ofwhich iscut off atan acute angle, theupper being cutaway and ZX) turned down, andascrew-thread cutonitto YA. carryaninsulator. When thefaceofthe KYf\.) angle isfixedagainst thesideofahouse or © apole,andaspikeorscrewputthrough it,the f insulator projects somewhat, andforms asup-MiA,portformakingoffthewire.‘Thewholething Hiisofthecheapestdescription but,nevertheless,.\\seemstoansweritspurpose.Wires are led infrom the outside with much \\\} less care than isgenerally used inEngland; \\_ andthisapplies alsototelephone andtelegraph work. Itoften puzzled mehow they worked at allafter they hadbeen putupafewmonths. But thefact that the climateismuchdrier,andfinerainnotsofrequent ashere, 1 ELECTRICAL ENGINEERING INAMERICA. [No.2 hasnodoubtagreatdealtodowithit.Inthecaseofarclight circuits entering andcoming outofhouses, andworked insere— asthey always are—it is,ofcourse,a material point that although thedifference ofpotential between thecircuit andtheearth may be3,000 volts, yet the difference between theoutgoing and incoming wire isonly some 50volts; sothat thetendencyofte current tobreak across the leads isreally small. Again, the chance ofthecurrent going toearth ondryceilings andwalls iy when one comes tolook into it,not great. Thus, notwith- standing therough character ofmost American work, andthe poor insulation ofthe wires, inthis class ofwork there iss greater freedom from faults andahigher margin ofsafety thn onewould atfirstimagine. The only place where themarginof safety isreally lowiswhere thewires enter buildings. Thave now, Ithink, gone over allthemain points inar lighting construction work. Construction work foralternating currents isgenerally carried outonmuchthesamelinesasfor arelighting; the only difference isthat inthe latter ave converters arestuckagainsttheoutsideofbuildings insteadofat lamps,which,ofcourse,areinparalle!insteadofinseries.Arough hole isknocked through thebrickwork justunder theplace where theconverter isfixed, forthesecondary leads, which areslipped through acouple ofrubber tubes andcarried inside. Butlittle attempt isusually made toexecute thispartofthework inwhatweshouldcallaneatorworkmanlike manner. Let usnow make aresuméofthisbranchofthesubject. ‘There areinAmerica now, Ibelieve, more than aquarter of million arclamps inconstant use, Inthelargo towns these art lampsarefedbymachines equaltokeepinggoing50or60lamps each. Outside acentral station, supposing ittobefor1,000 lights weshall findtallanduntidy-looking poles 30or40feethigh. At thetopthese poles willhave fourorfivecross arms, each carrying sixoreight insulators. ‘These poles willbecontinued along the neighbouring streets atintervals of30to60yards, accordingto circumstances. Everyhereandthereacircuitwillleavetherestforthesideofaneighbouring house,supplyoneormorelamps 189] ‘ELECTRICAL ENGINEERING IN AMERICA. a being run inthe latter case along theface ofthe houses, and then return again atthenext pole. Oracircuit may jump across thestreet, supply alamp ortwo, and then return again to thenearest pole onthesideitfirststarted from. Here and there one ortwo circuits will leave the main line of poles altogether, andturn upaside street: inthiscase thepoles intheside streets aregenerally smaller. Itissurprising the lengthsomeofthearecircuitsextendinthisway—notunfrequentlyover adozen miles. Asallthe machines inastation areusually ofthe same size, and arerun atthesame speed, when the circuit islong the number oflamps fedoffanominal60-lightmachinemayperhaps have tobereduced to45toallow for the resistance ofthe leads, ‘Asyou areaware, some ofthese aremachines arenow used on power circuits inthedaytime, which forms avery welcome source ofextra revenue. Itrust Ihave notwearied youbyatoominute description of this, anot very interesting part ofelectrical engineering in America; but Iwas anxious toconvey asclearly asIcouldon what aremarkably simple andcrude basis almost thewhole fabric ofAmerican electrical engineering inthisparticular atpresentreste,ForIthinkIamnotoverstating factsinsayingthatatleast nine-tenths ofalltheelectrical work inAmerica—putting aside isolated plants—is carried out, asfarastheconstruction work is concerned, inthemanner Ihave described. Inallthis portion of thework Isawnext tonothing which showed particular ingenuity, skill orbrillianey ofidea, Simple andcrude means were every- where used toeffect inarough andcrude manner theenddesired. Inthemeans employed there islittle toexcite either admiration orinterest. Theinterestliesinthefactthatinsuchasimple andelementary manner somuch hasbeen done. Uxpenorouxn Worx, ‘Turning nowfrom thisbranch ofthesubject tounderground work, Imust confess that Ihad agood deal ofdifficulty in getting really accurate andreliable information onthissubject— ‘atanyrate, ofacomparative character. The Edison Company ms ELECTRICAL ENGINEERING INAMERICA. [Nor2h have, ofcourse, buried their wires allalong, andtheir three conductors enclosed inanirontube,ransolid,withinsulating material inside, arewell known. ‘AsfarasIcould ascertain, the various systems ofconduits have notproved very successful #0fa, ‘and, since theproof ofthepudding isintheeating, thefactthat theSubway Company inNew York are, Ibelieve, laying down ironpipes entirely—at least, they were inevery place Isawwork going on—points tothisconclusion. (~The system ofconstruction forunderground work atpresent being carried outinNew York isasfollows :— Manholes atconvenient distances aremade intherosdway— not,asarule, inthepavement, butjust ontheriseoftheroad, andclear ofthegutter. These manholes arecircular, andare ‘covered byaheavy iron lid. Below themanhole isabell-shaped pit8or9feetdeep,andbrickedround,andofsufficientsizeatthebottom foracoupleofmentostandfairlycomfortably towork. Ironpipes such asareused forwater orgasarelaid inthe ground from onemanhole tothenext. Iunderstand thateach electriccompany will,asarule,have#separate pipe,sothatifthereareseveralcompanies runningalongthestreet,theywill each require aseparate lineofpiping opening into themanhole. There will, therefore, often beseveral pipes entering ateachside ofthemanhole. These pipes enter themanhole atsuch height as,roughly,tocomebetweenthehipsandbreastofamanstandingonthebottom—at aheight, infact, tobring them inafairly convenient position forworking. The cables aredrawn through theiron pipes intheordinary way. Branches aretaken from the manholes under theroadway andpavement toanyhouse between themanholes requiring current, Asregards theclass ofcable which isbeing largely laid down inthese conduits, theconductor consists ofseven strands ofNo.16copper. These areinsulated byalayer ofahard but somewhat flexible material, containing, Ibelieve, some rubber inconjunction with other materials, but the complete material ischeaper than rubber used alone. The thicknessof thismaterialsurrounding theconductors isone-twelfth ofaninch. ‘Itsomewhat resembles ebonite, andisofaconsistency tostand 18] “ELECTRICAL, ENGINEERING INAMERICA. 10 rough usage andhard wear. Itwillbeinteresting toseehowit lustsinpractice. Imust sayitseems tomeavery suitable materialforthepurpose forwhichitisintended; andasfaras insulatingqualitiesgo,Iwasabletoseetherecordoftestsofa gooddealthathasbeenlaid,andtheywereexcellent,goingup toseveral thousands ofmegohms per mile, The insulating material iscovered with alead pipe about one-sixteenth of aninch thick, and containing, Ibelieve, apercentage oftin, ‘These lead-covered cables are usually jointed ateach manhole, andalittle slack leftonthem there. Thejointing isdone bya specialsystem,andwhenitiscomplete electrically thetwolead tubes aredrawn together over asmall bitofbrass tube, which protects the point ofjunction; they are then “wiped” together with ahotiron like anordinary plumber’s joint. No doubt this work isheavy and expensive, but itlooks asifit should lastwhen once putdown, ‘InrernaL Work. Coming now foramoment totheinternal work inbuildings, most ofwhat Isaw had arough, temporary look about it, evidently being put upascheaply aspossible, Idonotsay thatnogood,solidworkisdone,buttheproportion ofittothe remainder is,Ifancy, agood deal smaller than inthis country. Gas fittings arefreely utilised, and Ithink Iamcorrect in stating that inthearrangementoflightsthegas-fittingwayof managing things isadhered tomuch more closely than inthis country, where wehave rather tried toavoid imitating the ordinarywaysofarranging gaslights.‘Turning aside foramoment, itappears tomethat oneofthe great wants allover theworld now is,ifImay soterm it,a “science ofillumination.” Uptoandasfarasproviding lightwe areexceedingly scientific. Butinutilising itwhen wehave got it,weproceed inanelementary and haphazard manner. ‘Mr.Preece has, Iknow, from thefirst paid agreat deal of attention tothis point, and appreciated itsvalue, Itseems to methat itwould really beofgreat practical benefit ifacom- mitteeofthisInstitution, orsomeotherresponsible body,would 180 ELECTRICAL ENGINEERING INAMERICA. [Nor284 accumulate some accurate and authoritative data onthesubject ‘When once started properly, these data would increase ofthen- selves, andwould soon form aguide which would beexceedingly useful indrawing upspecifications and arranging tothebet advantage thelighting ofeither open spaces, public building, Private houses. Weallhaveourownindefinite notionsonthesubjectasthe result ofexperience, butifthey could only beput into definite shape—and Ithink the subject does admit ofbeing patinte definite shape—it would, Ifeelsure, beagreat advantage, Ofcourse there areinnumerable patterns ofswitches andcat- outs andfittings inAmerica,ashere,butnonecametomynotice showinganyverybrilliantconceptionindesign.Lampsocketsand fittings aregenerally somewhat larger andclumsier than withus The chief difference Inoticed was that aswitch isalmat invariably included ineach lamp socket. Americans,asarule, donotgotothesame amount oftrouble and expense thatis general here tofixswitches inhandy places near doors. Ifyou wantthelightonoroff,youturnitonoroffattheburnerasyou woulddogas. Cenrea, Stations, Comingnowtocentralstations,thoseIwentoverwerethe newAreLightStationinSanFrancisco, theEdisonStationinChicago, theWestinghouse Station inPittsburg,andtheThomssa- ‘HoustonStationinNewYork;andImayaddthatsincemyretom toLondonIhavehadanopportunity ofgoingcarefully overthe new Westinghouse Station inSardinia Street, Lincoln's Inn Fields,theplantinwhichisentirelyofAmerican manufactureandpattern, andarranged intheAmerican manner. Imighteasilyhaveseenmanymorestations,insomecasesperhapslargerandbetterarranged thanthoseIdidsee.Butthose Ididseeare, Ithink,alltypicalofthelatestAmerican work,andallwereintheprocessofmakinglargeadditionsto their plant, sothat they were quite uptodate. ‘The stationinSanFranciscoisneartherailwaystation.It isalarge brick building, erected entirely foritspresent purpose, about 250feet long by80fectbroad, with aground andtwo 19] «ELECTRICAL, ENGINEERING INAMERICA. Ta upper floors, and hasbeen finished about eight months. Itis intended toprovide accommodation forabout 2,000 are lights when filled up. Atpresent itisfurnishing current forabout 600. Itisarranged intheusual—I wasgoingtosayinvariable— American plan ofplacing theengines and boilers ontheground floor, and arranging thedynamos, switch-boards, and electrical partionofthestationonthefirstfloor.Inthisinstancethe secondfloorisusedasastorehouse, andforgeneralpurposes. ‘Theboiler-house forms one end ofthebuilding, and there ‘wasnothing above it,Itisarranged fortwolines ofboilers, oneoneach side, one side only being filled upatpresent. They face towards the centre ofthe building, with »large space between forcoals and general working purposes. The boilers were ofthe well-known water-tube pattern. This type, varied slightly indetail byeach maker, seems tobeingeneral tseeverywhere intheStates. From theboilers thesteam isled to 650-horse-power single-cylinder Corliss engine, which, by ropegearing, drives asingle lineofcountershafting. Imust say Iwassurprised tofind that,inanowstationsolargeasthis, and one inwhich expense evidently had not been spared, the engine was neither compound norcondensing, particularly as there waswater within amoderate distance. ‘There wasonly this oneengine inthestation driving allthedynamos, though space ‘rasprovided forthree more astheamount oflighting increased. ‘Theworking ofthestationwastherefore dependent onthe single engine andlineofshafting keeping ingood order. The engine drove thislineofcountershafting asIhave said, theloose pulleys being onthe shafting, Atthestations Ivisited where countershafting was inuseIwasstruck with thecare and com- pleteness with which means were provided foradjusting the plummer blocks and bearings inany direction, The general ideawassomewhat asfollows:—The plummer block, orupright, stood inacast-iron shoe with aflatplaned bottom, through which theholding-down bolts projected. The holes fortheholding- down bolts intheplummer block were bored large #0astopermit ‘certainmovement intheblock.‘Theedgesofthecast-iron shoe‘wereturnedupallround,andthroughtheseedgesprojected screw 188 ELECTRICAL ENGINEERING INAMERICA. (Nor.2, bolts,thepointsofwhichtouchedtheplummer block.Byloon- ing, then, the holding-down bolts atany time, and screwing ot unscrewing thescrew bolts asrequired, theblock could beslewed slightly oradjusted tolinewith thegreatest nicety andee. Besides this, theunder bearing worked inguides, being kept initsplacebyastrongscrewprojecting upfromunderneath. Bywork- ingonthisscrew, andonthecapoftheupper bearing, theheight ofthebearing could, ofcourse, beadjusted with great exactness, Withthesemeansthebrassesarealsoquicklyandeasilyreplaced while ifthey getwarm when theshaft isrunning, thebottom bearingcanbeslackeddown.‘TheThomson-Honston Companyin NewYorktoldmetheydidthis,andtheirdynamosareonthesume floorasthelineofshafting. Itis,ofcourse,astillsimplerand safermattertodothiswhenthedynamos aresituatedalmostright above theshafting, asisthegeneral practiceinAmericanstations. Leather belting oftheordinary description wasinuse, exceptfor driving, where ropes are used,asIhavementioned. Ifound leather belting employed ateach ofthestations Ivisited. ‘These long lines ofcountershafting must, ofcourse, bewaste fulofpower; still, itisworth pointing out that such shafting runs under thebest conditions when doing work right overhead, since both thedead strain and-the running strain onthebelts to thedynamos tend tolifttheline ofcountershafting, which, therefore, ifthe weight isproperly adjusted, must ramvery lightly initsbearings. ‘The method ofdriving from underneath isalso economical in belting, since theweight ofthebelt hanging onthedynamo pulley causes ittohave agood grip, andpermitsofashortlength being used. The belts pass uptothedynamos through holesin thefloor, theholes being elongated topermit the beltbeing tightened, ifnecessary, byscrewing thedynamo along onitsbed rails. Ineach ofthestations Ivisited inAmerica, theupper flow carrying thedynamos was entirely ofwood, strongly made, andsupported from underneath atintervals bystout wooden uprights, butstillofwood. The weight ofthe floor inthe San Francisco station is 169] ELECTRICAL ENGINEERING IN AMERICA. 188 entirely supported bythewooden uprights, being only steadied atthe sides, Iwas informed that atthe old station inSan Francisco, worked bysame company, and which, Iunder- stood, was supplying about 1,200 arelights, agood deal of trouble wasexperienced from thefloor, which was fixed tothe vralls, andwhich, vibrating with themotion ofthemachinery, shook the brickwork. Ifound the vibration on these wooden floors less than Ianticipated, considering the amount of machinery onthem;but,ofcourse,themotionofagood dynamo isvery even, and ifthejoints onthe belt areneatly made there isnojerking. Inthis San Francisco station all the machines were Brush 60-lighters. From 50to60lights inseries ononecircuit seems tobeabout thegeneral rule forlighting inAmerica when any number oflightsisrequired. Asallthemachines arerunfrom thesame countershafting, and atthe same speed, whatever work they may bedoing, itisnecessary toprovide anautomatic regulator toadjust thecurrent through thefield magnets for each machine when doing lessthan itsnormal output. Onthe otherhand,ifthecircuitisaverylongone—and Iwastoldsome ofthe circuits atSan Francisco are asmuch as13or14miles— thenumber oflamps onthecircuit must bereduced tocom- ensatefortheextraresistance. TheseBrushdynamos wereall mounted onshort iron columns about 9inches high, sothat they were lifted entirely from thefloor. Iwastold that thisarrange- mentwasfoundveryhandyforkeepingthemachines clearand freefrom dirt, andforinspecting them. Ididnotnotice much difference between themachines and those turned outbyour friends onthis sideofthewater, andthestyle oftheworkman- ship seemed tobeabout thesame. Alight overhead trolley, with lifting gear, ranoverhead sothat itcould quickly be brought over any machine, and Iwas informed that should a machine gowrong they could replace anarmature inhalf an hour. Intheolder station inSanFrancisco alarge number of ‘Tuomson-Houston machines areused, butBrush machines only areemployed inthenewstation, asthey findtheBrush armature ismore easily replaced, andissomuch theeasier ofthetwoto 788 ELECTRICAL ENGINEERING INAMERICA. (Nor.S8h, repair ifanything goes wrong. Itisworthy ofnote inthis connection that the latest advices from America inform usthat theThomson-Houston Company have bought theBrush Works Cleveland, Ohio, and therefore presumably theThomson-Houston Company will inthefuture control this interest aswell their own. From thedynamos themains areledtoalong narrow shel, wherearefixedtheresistances andindicating instruments foreach machine. The mains next proceed totheswitch-board, whichis oftheplug pattern—brass plugs with boxwood handles united byaflexible insulated cord. Alltheflooring being ofwood, asitisineach ofthestations Iséw, andbeing kept clean and dry, therisk ofgetting abed shock isnot great, except byshunting the current through oneself, orgetting inseries with amain, which arenotordiny accidents. AtSan Francisco themotor business issteadily increasing, and forms awelcome addition tothe revenue ofthe station, with- outaverygreatadditional expenditure. ! IwillnowgoontotheEdisonStationinChicago. ‘Thereisno necessitytodescribeitatlength,astheplansonwhichEdian stations are constructed have been published over and orer again. ‘Theboilers,whicharesituatedonthegroundfloor,areofthe water-tube pattern—Heine arrangement. Inthis pattern, which seemstobeagooddealused,thetubesaremuchmorenearly horizontal than inthewell-known Babcock &Wilcox type. All thetubes come into alarge flat box ateach end, and there are small manholes intheboxes opposite theends ofthetubes, forgetting abthem. These flatboxes areprevented from ex panding with thepressure bytubular stays, which pass right through them, and areexpanded intheir places. These small tubes serve asblow-holes fordirecting ajetofsteam onthe | uppersideofthetubestoclearthemofashes.Theyarekept| plugged upwith stopper, under ordinary circumstances, prevent thegases coming through. ‘These boilers badadamper projecting downwards inthefurnace opening, fordirecting the 188] “ELECTRICAL ENGINEERING IN AMERICA. 73 draught over thesurface ofthecoals when firing, tocause more complete combustion and prevent smoke. This damper may have been ofsome useforthepurpose, batitcertainly made ita dificult operation tofireproperly. Theengines were alsointhebasement, inthenext room to theboilers. ‘They were oftheordinary Armington-Sims pattern, single cylinder, and neither compound nor condensing. There wasonetoeach dynamo, thedynamos being situated overhead on thefloor above. The dynamos were ofthe ordinary Edison pattern, for,Ithink, 1,500 lights each. Everything was very neatly arranged inthedynamo room, thecurrent being conveyed about bylarge copper bars, The walls were simply lined with resistances and indicators forthe different circuits. Infact, the place gave onerather theimpression ofawholesale clock store, though the attendant informed me they did not have much difficulty inkeeping their resistances properly adjusted. Allthe mains areunderground, onthethree-wire system. Indeed, there isagood deal ofunderground work inChicago, theTelephone Company also having alarge number ofcables buried. The next station Ivisited wasthat oftheWestinghonse Company atPittsburg. Itissituated inalarge building ofthe warehouse type, quite inthecentre ofthetown. ‘The boilers here areofthe water-tube type, but heated by natural gasinstead ofcoal. This natural gasisbrought inpipes tothetown from adistance of35miles. The company who own theline ofpiping—of which, bytheway, Mr.George Westing- house isalsoPresident—sell this gastomanufactories, &.,ata Tate which isnearly equivalent tothe cost ofcoal; sothat, although theconvenience isgreat inusing it,theeconomy isnot solarge asmight atfirst beanticipated. Work atthe Pittsburg station was inthe first instance begun with two Corliss single-cylinder engines oftheordinary type, driving ontoaline ofcountershafting; but this methodofworkinghasnotbeenextended, andoneoftheseenginesbas since been removed. The next additions tothe station con- ‘sistedofWestinghouse Standard enginesof250indicated horse- power, driving 2,500-light alternating dynamos. ‘These engines ELECTRICAL ENGINEERING IN AMERICA, (Sw. stand onsquare blocks ofconcrete and masonry; theblocks forthis sized engine are about 7feet by5feet. The lat additions are ofthe new type ofWestinghouse Compouni engine. From apower point ofview, aswell asfrom anelectis lighting point ofview, thestation hastherefore something of historical interest. While inPittsburg Ihad anopportunity ofgoing carefully through the Westinghouse engine factory with Mr. Ralph Bagaley, thevice-president and manager, andofseeing allthe detailsofconstruction ofeachofthedifferent typesofengines. ‘The Westinghouse Company make three styles ofengine, eachworkedbyapistonvalveandsingleeccentric.‘TheStandard typeissowellknownthatIwillnotsayaxy- thing about it. Init,asyou are probably aware, thestean valve isbetween the cylinders, and setataslight angle to them, while thecylinders aresetrather over their work, andnot directly inavertical line above thecrank shaft. ‘TheJunior andthenewCompound engines areonsomewhst different lines totheStandard, while themain principles involved Fa dhPES i meme.Apo:— | { i} Sse —penises Pee AT ‘Wesrixonotse Courouxp Exome. 16] ELECTRICAL ENGINEERING IN AMERICA. er arenearly thesame. Inthese twolater types thepinton valve lieshorizontally along the topofthe cylinder heads. Itis operated byasingle eccentric working outside theengine case in vertical direction, themotion being changed toahorizontal one bymeans ofanLpiece attached tothe engine case byahinge. Thewhole movement isofsimple character, and works exceed- ingly well. Infact, there arepractically nomore working parts intheCompound engine than intheJunior orthe Standard two-cylinder patterns, andtheoneisassimple astheother. The Standard type hasnow stood thetest ofseveral years’ vwork very satisfactorily, andthere isnoroom todoubt that the Compound engine, constructed inthesame shops andbythe same method astheStandard engine, willwear atleast. aswell. Themakers’ tests show exceedingly good results forthis Com- pound engine—viz., from 18}Ibs.to194Ibs.ofwater perbrake horse-power hour with condenser, and from 24Ibs. to25} Ibs. without condenser. Isaw thetesting apparatus inthemakers’ shops, andengines undergoing their testpreparatory todelivery, andfeelperfectly confident that thefigures given bythemakers arethoroughly reliable. Iwill notdescribe this engine more fully, although todo s©would probably elicit aninteresting discussion, asitwould be wandering from thesubject ofmypaper. Ontheother band, theremarks Ihave made are not intended tocompare the Westinghouse Compound engine with anyother engine ofatall similar type, but rather todraw the attention ofelectrical engineers tothefactthat there arenow, atanyrate, twolarge firms ofthehighest repute turning outthis class ofengine, and ready toguarantee economy and efficiency onaparwith Corliss engines. Returning tothePittsburg station, fiveWestinghouse engines ‘of250 indicated horse-power drive five alternating-current dynamosof2,50016-candle-power lightscapacityeach,besides themachine driven from theCorliss engine. The exciters are ran from smaller engines. The engines driving thealternat- ingdynamos make 250revolutions per minute, the dynamos making 1,050 revolutions per minute, The power isconveyed 188 ELECTRICAL ENGINEERING INAMERICA. (North overhead bybelts tothe first floor, where thedynamos ar placed.Eachofthesedynamos isonaslidingbed,sothatthe beltcanbetightened, and,asIhavealreadysaid,theexcites arerunseparately from themachines. ‘The machines aresowell known itishardly necessary todescribe them. Imay, however, perhaps with advantage sayafewwords ontheconstructionof thearmature, which isthe least easy part tounderstand from written descriptions ordiagrams, InProfessor Silvanus Thomp- son’s book ondynamo machines the diameter ofthe armature is given atnearly twice itsbreadth. While this istrue ofthe smaller machines, inthe 2,500 machine—which isthestandard sizeforlarge stations, andisthesizeemployed inthenewstation inSardinia StreetinLondon—the lengthofthedrumiscom- siderably greater than the diameter. The drum itself isformed ofthin iron plates, just like anEdison armature, only oflarger diameter; andastheironnearthecentreisuseless,holesawpunchedinthesheets,whichlightens thearmature andsecuresventilation, Roundtheoutsideofthisironcore,andoverthetwoends, insulation islaid, Asregards thecoils ofwire inwhich thecurrent isgenerated, they aremade inthis way :—Beginning attheinner side ofthe coil, the wire iswound onanobloog frame tocorrespond with theshape ofthefield magnets, out turn onthetopoftheother. ‘Thecoils areonly onewirethick; —that is,there isonly onelayer ofwire between the dram aod thefieldmagnets. Whenthesethinflatcoilsarelaidonthe drum, theround ends ofthecoils areturned down atright angles over theedge ateach endofthedrum, andtheconnections from coiltocoilaremade.Asmallpieceofwoodscrewedintothe sideofthedrum, andprojecting over theendofeach coil,keeps itinitsplace. When allthecoils arefixed intheir places, sheets ofmica arelaidover them allround thearmature, andbandsof fine wire with alittle thicker ineulation underneath them sre ‘bound round thewhole armature atshort intervals, inexactly the same way asisusual with the armatures ofcontinuous current machines, Anarmature madeinthiswayhasperhapsratheranamateurish lookaboutit,butthereisnodoubtthatthey work well inpractice andgive excellent results, Itis 1882] “ELECTRICAL ENGINEERING INAMERICA. 189 farther worth noting that astheficld magnets arelong and narrow—a ratio ofatleast 10to1inthelarge machines—they cannot beveryeconomical asfarasexciting current isconcerned. Stil, since theexciting current, anyhow, bears such asmall ratio totheenergy which the machine will transform into electric carrent,thisisnotaverymaterialpoint. The mains areran from thedynamos totheswitch-board in troughs cutinthejoists supporting thefloor, which ismade to takeupover these troughs. ‘Theswitch-board atPittsburg consisted ofawooden frame boardedover,about9feethighand12or14feetlong,andstanding about3feetfromthewall,soastogiveroomtogobehindand make the connections. This,Imaysay,istheusualwayof arrangingswitch-boards inAmerica. ‘Tho general type ofswitch used inAmerica foralternating- current work—and itisnot confined tothe Westinghouse Company—is exceedingly simple inconstruction andoperation,as well asefficient. Ingeneral principle itislike anordinary tobacco chopper. Tothetopofasmalluprightaleverishinged, terminating inaninsulated handle. ‘The metal partofthislever hhasahatchet-shaped projection cast onit,projecting inthe same planeasthemotionofthelever.Whentheleverispulled downthehatchet-shaped piecesticksinbetweentwospringpieces setuptoreceive it,andsomakes contact. Ifthelever hastwo hatchet-shaped pieces attached toitopposite each other, and engaging inspring pieces oneither side, according asthehandle isthrown over inonedirection ortheother, wehave atwo-way switch. Unite two ofthese levers together byacross bar of insulating material, with ahandle inthemiddle, andwehave a double-pole two-way switch. Placeaspringinthecolumnsholding thehingesonwhichtheleversturn,tothrowdownthelevers when nearing thespring contact-pieces oneither side, and we have thecomplete article. ‘These switches were allmounted on wooden bases. Infact, Isaw noslate used inAmerica anywhere forsuch purposes, except attheElectric Club inNew York. In thenew Westinghouse Station inSardinia Street, however,the wholeoftheswitchesaremountedonenamelledslate,andvery nicetheylook.Vou. XVII, 54 700 ELECTRICAL ENGINEERING INAMERICA. (Sor.234, Oneexciter isusedforseveral machines; eachexciter has resistance initefield magnets, which canbevaried, There iy moreover, aseparate resistance inthefield-magnet circuitofexch alternating machine, sothat the E.M.F. ofthemachine canbe varied while itsspeed remains constant. Double throw-over switches are connectedsothateachater nating-current machine can beconnected toeither oftwo exciters, ‘andsothat anymain circuit canberunfrom anymachine. The {general principle followed istokeep themachines atworkas fully loaded aspossible. Thus inthedaymany circuits willbe runoffonemachine, butasevening comesontheywilloneby onebetaken offand putonother machines. ‘The throwing over iss0quickly done astoonly just. cause aperceptible flicker inthe lights, Allthese switches were bare andunprotected, Ididnot hear orseemuch ofthe method ofrunning two separate machines parallel onthesame circuit, Whatevermay bedoneinisolatedinstances, itdoesnotappeartobepractied generally ;indeed, with theAmerican system ofrunning anumber ofsmall circuits instead offewer heavy ones, thedesirabilityof doingsodoes not often come in, Ishould like toendorse here what Isee Professor Geo. Forbes saidinhispaper onforeign central stations with regard tofeeder. The area which canbeeconomically supplied from acentn! station hascertainly been greatly increased bytheadoptionof 1,000or2,000voltsinstead of100,Butatthesametimeour notions ofthearea which canbesupplied from onecentre have experienced acorresponding growth. If,then, such anincreased area issupplied onahigh-tension aystem from onecentre, wehave asgreat, ornearly asgreat, fit ofpotential todealwithasifthelow-tension systemwereused over acorrespondingly smaller area. Tosecure thehighest classoflighting, therefore, together with theflexibility desirable inaninstallation lighting alarge ares thewhole oftheoutside work should bebased onascomplete snd carefully designed asetoffeeders asfrom ourpresent knowledge ‘weshould employ with alow-tension system. Onmany ofthe Westinghouse feeders lighting atgres 18] ELECTRICAL, ENGINEERING IN AMERICA. mm distances, there isafallofpotential ofasmuch as10percent. ‘Toavoid pilot wires anindicator isused, which, Ithink, Professor Forbes hasalready described here. There isnoregular voltmeter fordriving themachines by,butanindicator isused with but onesingle mark onthedial, When themachine isgiving thevoltagewhichisrequiredatthefurtherendofthe feeder tokeep thelamps attheir proper brightness, thehand oftheindicator willberight over themark onthedial. The lossinthefeeder will,ofcourse, differ according tothecurrent flowing through it; +0,toallow forthis, acompensator isattached totheindicator. ‘The main current flows through the compensator, which is arranged tocause aslight back-pull onthe needle ofthe indieator 5this pull becomes greater asthecurrent increases. Asthe current increases, therefore, aslightly higher voltage onthemachine willbenecessary tokeep theneedle initsplace. ‘Theinstruments have, ofcourse, tobemadesothattheycanbe adjusted tothecalculated requirements ofeach feeder. When once setright, however, their operation issimple enough. ‘The Westinghouse Company also useapiece ofapparatus containing anadjustable converter, which enables them tosupply current from thesame machine totwofeeders having adifferent fallofpotential. . The Thomson-Houston Station inNew York issituated in ‘Twenty-Fourth Street East, near theriver, andoccupies what was formerly alarge sugar refinery. Itis,therefore, abuildingofthe warehouse class, but very strongly built. Iwas told that this station wasneither solarge norsowell-arranged astheir station inBrooklyn, buttheIntter, Iregret, Ihadnottime tosee. Two boilers struck mehere asoutoftheordinary patterns. The simplest way todescribe them is,Ithink, toconsider them a large locomotive fire-tube boilers setonend. ‘The tubes are about 15feet long, and the'water line isabout 10feet upthe tubes, This gives aconsiderable steam space above, through which thefiretubes pass, superheating thesteam generated, and thoroughly drying it.Atfirst sight onemight suppose this an uneconomical pattern, astheheated gases pass 20quickly into theflue. But thegases appear togive outtheir heat well, 10) —RLECTRICAL, ENGINEERING INAMERICA. [Ho1h forMr.Foster, thestation superintendent, told methetem- perature ofthegasesgoingintothefluewasnotabove59 degrees Fah,Thelowerportion ofthisboileris,however, the 4 !Tubes: i i } 1 WW fff gage hele slatadil|i \ Pore ae Bonen wr Temes interesting part.Icannot verywelldesoribe itwithout theaidofadiagram. .Theoutershelloftheboilerisexpandedoutand prolonged aboutacoupleoffeetbelowthetubes.Itisturned ‘upatthebottom, andunitedallroundtothecrownplateofthe boiler, intowhich thelower ends ofthetubes areexpanded. This ‘prolongation oftheshellthenformsacircularhollow,orchamber, about 2feetdeep. Theinternal space enclosed bythishollow chamber forms thefire-box orfurnace, andisabout 6feetin diameter. Anysediment collects inthehollow chamber, andcad ‘betakenoutatmanholes. Thereweretwooftheseboilersin ‘use,anditwasincontemplation toincrease their number. ‘The engine—and there isonly one—which furnishes powertothisstationisalargecompound condensing Corlissengine;thetwocylinders being intandem. Itisaveryfinepieceof workofitsclass.Itdrivesontoalargefly-wheel with»smoolb __ a _— 180] “ELECTRICAL, ENGINEERING IN AMERICA. 103 face ofabout 60inches breadth. This fly-wheel transmits itspower toalineofcountershafting, which isarranged onthefloor above, bymeansofalargeleatherbelt58inchesbroadandofgreatthickness. Ithink, inall,seventeen machines arerun from this Tine ofcountershafting—fifteen 55-light arcmachines and two alternating-current machines;inthiscasethemachinesareon the same floorasthecountershafting. Thegeneralconstruction oftheThomson-Houston alternating-current machines isonthe same lines asthose oftheWestinghouse Company. The main dif- ference isthat instead ofthe ends ofthe coils onthe armatures being turned over theedge ofthecore,asintheWestinghouse machine, thearmature has asort ofprolonged lipallround the outside ateach end, and the ends ofthe coils lie fiat onthis instead ofbeing turned over. Totheir smaller machines, also, the Thomson-Honston Company frequently add acommutator, and commute part ofthe current toexcite thefield magnets,as Mesers, Zipernowski, Deri, and Blathy areinthe habit. ofdoing. ‘The switch-board here isonmuch thesame lines asIhave already described. Itwasallquite open andexposed ;butIwastoldthat, inview ofanaccident which had happened there ashort time before, they proposed tocover theexposed parts up. ‘This way ofleaving the working parts ofhigh-tension alternating-current switeh-boards exposed isnot confined to American manufacturers:agooddealoftheContinental workis even worse. Yet allthedifferent patterns ofswitches andcon- verterscouldbeprotected atamoderate cost,andbytheexerciseofaverymoderate amountofskillandingenuity. Icannothelpthinking that nottodos0,tosave money inthis direction, is falseeconomy. IsoLaTeD PLants. Let usnow turn forafewmomentsfromcentralstationsto what arecalled “private installations” inthis country and“iso- “lated plants” inAmerica. Ihave already spoken ofinternal construction work,orwiringandfittings,inconnectionwithcentral stations. With isolated plants there isnotmuch difference. Of course thecharacter ofthework differs greatly indifferent places, asitdoeshere;but,takingitasawhole,Idon'tthinkitisequal 04 ELECTRICAL ENGINEERING IN AMERICA. [Nor. 2b, toourstandard, whether from anengineering orartistic point of view. Inoticed, bytheway, that there isamuch greater ten- dencytoturnincandescent lampsupwardsinfittings,likegajets, instead ofdownwards,aswithusthemajorityoflampsareturned. Teolated. plants aredriven bysteam and gasengines, located incollars andbasements, exactly like they arehere. Asyouaw aware, accumulators arenots0generally used;Ionlysawtheset atthe New York Electric Club, With respect todynamos, the Americans have been fully alive tothe developments, both theoretical and practical, inour knowledge ofthe maguetic cireuit which have taken place during thelastfour orfiveyear, and alltheir modern machines appeared tomefairly uptodate inthis particular. What Imay bepermitted tocalltheEdison- Hopkinson type, the Kapp type, the Manchester type, andthe Weston type—the latter with twomagnets, placed horizontally— appear tobethefavourites, just asthey arehere, ‘The construc tion and workmanship ofthemachines appeared tobemuch on& parwith ourown;there islittle tochoose between inoneorthe other. Astothewaytheinstallations arerun,judging bythet crucial practical test, the look ofthe commutator and brushes itismuch thesame ashere;certainly, Ithink,onthewhole,not better. ‘The lamps, asarule, Ididnotfind any steadier, norare they kept nearer their proper voltage than they arehere. This applies toarclamps aswell asincandescent:thereisnomagic about arolamps over ontheother side ofthewater; theyare subject tothe same vicissitudes and uncertainties asthey are here,anddonotburnsteadierthanthoseinanymoderate-sizedstation here where theappliances aresuch that thelamps get afair chance. The switch-boards inthese private installations arenearly always mounted onwood, andthere iseven adisposition tocat things fine inthe strength and solidity ofthese wooden base- boards, somewhat, asitseemed tomeinoneortwoinstances,st theexpense ofefficiency andlasting qualities. ExecrricaL TRaMways, Such very full accounts have been published ofeverything thathasbeendoneinthisdirection, thatthereisreallylittleI 1W,) ELECTRICAL ENGINEERING INAMERICA. 705 can add tothem, Inthe Electrical World for October 19th there ispublished atable ofthe electrical tramways inactual operation and under contract intheUnited Stater. ‘The total nileage ofboth is1,260 miles, with 1,884 cars, on179roads. Of this, 716 miles were, itappears, actually being operated by electricity atthe date ofthe return, and 544 miles were under contract, Ofthistotal, 21miles were worked bycarswith storage tutteries, andalltherest, with theexception oftwoorthree miles, byoverhead conductors. Nothing canbesimpler than theway these overhead con- ductors areputup. Two lines ofposts, about 20feethigh, areset upopposite each other, oneoneach sideofthestreet; light steel wires arerunacross from onetotheother; asmall iron holding- piece isattached tothecentre ofthesteel suspender, and this holding-piece carries abare hard-drawn copper wire, along which thecurrent flows, returning bythe rails. Where feeders are used, they run-along one sideofthestreet, andconnections are made atintervals with the central wire. The way theconductor iscarried round corners isonaparwith therest ofthework. Posts aresotatthefurther corners ofthestreet, andfrom each ofthese two, three, orfour steel wires arerunatanangle with each other, meeting corresponding wires runfrom posts onthe inner side ofthecurve. Inthiswaythewire ismade totake theform ofpart ofarough polygon, while thehorizontal and vertical movements ofthefishing-rod connector on.the carare sufficiently ample tokeep thecontact-wheel onthewire when thecarisgoing round acorner. ‘This method oferection does notlook very nice, certainly, butitappears toanswer thepurpose well. Such work could, Ithink, bemore solidly carried outhere forabout £70 mile, Inafewplacesstandards withprojecting armsplacedinthecentre ofthestreet have been used forcarrying theoverhead conductor. Ifneatly constructed, there isnoreason why these should beunsightly, andthey canbeused forthefurther pur- pote oflighting thestreets, Agreat point about these electrical tram-cars isthewayin ‘which their speed canbevaried. ‘They canberunatthreeor 196 ELECTRICAL ENGINEERING INAMERICA. (Nor.28, four miles anhour incrowded streets andwhile turning comers; orinbroad roads insuburbs where there isnot much carriage traffic, the pace can beincreased to16miles anhour. The cars arestopped very quickly and reversed, and will, ofcours, runeither wayonthesame track. ‘Thecompaniesclaimareductionof50percent.inthecos ofworking over the employment ofhorses, putting aside allthe extra facilities which theelectrically propelled cargives. Avery noticeable point about allelectrically propelled carsisthequist ‘andgradual manner inwhichtheystartandcometorest,and the evenness oftheir motion while running. Inthese poins they greatly surpass both cables and horses. Itook theoppor tunity oftalking with agood many members ofthegenenl publicinthetrainandhotelsabouttheworking ofthesecar. ‘There isnodoubt that theoutside public thoroughly appreciate them and understand their value. Atpresent electrical tramway work isthe most “live” branch ofthe electrical industry inthe States, and everyone connected with itwas fullof buoyancy andhope atitsfuture prospects. Itisreally sadto think that, after alltheskill and care which hasbeen lavishedon clectrical tramway work here, itshould bedragging along insuch ahalf-hearted andpottering fashion. Coxciusion, WhileinAmerica,throughthekindnessofMr.LockwoodI hadanopportunity ofgoing over thetelephone exchange system ofNew York with him, and ofseeing theCourtland Street Exchange, which contains more than 3,500 subscribers, andis capable ofextension todouble that number. Ialso visitedthe WesternElectricCompany'sfactoriesbothinChicagoandNew York, where allthetelephonic apparatus foruseintheStater ismade, Although adescription ofthepresent state ofthis industry wouldproperly comeundertheheadingofmypape,seeing thatthepaper hasalready extended tosuch length,and that, owing tocircumstances, telephony isamatter ofcompa tively limited interest here, however important initself,Iwillnot saymore about it,butwillconclude with afewgeneral remarks 16] “ELECTRICAL ENGINEERING INAMERICA. wr Intheearly portion ofthispaper Iendeavoured toconvey an ideaoftheroughnessandcrudenessofAmericanoutsideconstrac- tionwork. After reading over theproofs ofthepaper, Idonot think Ihave overdrawn thepicture; rather itisunderdrawn, if anything. Atthesametime,Idonotwishitinanywaytobe understood that American engineers arenecessarily rough and untidyintheirideas,andincapable ofdifferent work. Putting aside their electrical plant, thenumbers ofbeautiful tools, crowned by thelatest pattern ofthephonograph, which come overhere from America,pointtoanentirelydifferent conclusion. Inaccuracyofworkmanship and careful finish, when they consider itsuits their purpose, theAmericans areourequals, ifnotoursuperiors: the difference lies inthe fact that our ideas often differ from theirs astowhere thisaccuracy andcareful finish should come in, Englishengineering wouldhardlydaretoappearbeforeitsfellow- citizens astheresponsible owner and designer ofsuch aterrible eyesore asAmerican outside work is;buttheAmerican engineer appears tolook atitfrom adifferent point ofview. Itisnothis business tothink ofhis fellow-citizens. Each ofhis fellow- citizens, heconsiders, isquite competent and capable oftaking care ofhimself. Itisthebusiness ofthe electrical engineer to forward and extend theapplications ofelectricity. Ifindoing thishedoes anything disagreeable tohisfellow-citizens, itisfor them toprotest when itbecomes intolerable tothem, notforhim tothink ofitbeforehand. And Iwould like you totryand think foryourselves, and really feeltheenormous facilities which American engineers have inthepast enjoyed, andstill enjoy, inextending their business, bysimply being allowed toerect open wires everywhere through thestreets. Even where wires arenow being put underground, they arelargely put underground inorder tocarry onexisting work and toextend it,notmerely totake advantage later of what may chance toturn upontheroute. Besidesthis,wehaveallhaditdrilledintousbythistimethat itdoes notmatter how perfect, how desirable, how useful ‘thingislikelytobe,beforeitcancomeintogeneralusethe public have gottobeeducated uptoit,toseeit,and feel its 798 ELECTRICAL ENGINEERING INAMERICA. [NorSib desirability and utility, and itspracticability—to understand it each man forhimself. Now theAmerican system ofoverhead wires and cheap line construction, badasitmay beinitself bs from thevery first brought electricity home toeverybody, right totheir very doors, notsomuch asaluxury, butasapractial desirable factor ofeveryday life,anattraction topublic resorts and agood advertisement forbusiness. Consequently electricity isinAmerica nomore awonder, butafact ofeveryday lifeto ‘anextent that itisnotyethere. Looking attheir plant, looking attheir apparatus, andlooking atthefacilities which Americans have enjoyed forextended pne- ticalexperiments onalargescale,itdoesnotappeartomethat their actual achievements, simply from anengineering ad scientific pointofview,havebeengreaterthanourshave,insite oftheapathy onthepart ofthe public and investors with which wehave had tocontend. Butthefactremains that they have done more business than wehave, even considering thedifference inpopulation, Ofwhat work wehave done sofar,Englishmen may feel justlyproud,andwithbettertimesItrustwemaystillbeable infriendly rivalry toshow that wearecompetitors worthy of ‘America's best mettle. Petuat ‘ThePuesiverr: Thegreat extent ofthesubjects toucheduponinthispaperissuchthatIamsureweallfeelthatmorethan what remains ofthis evening will berequired forits discussion, and itisproposed tocontinue thediscussion atour next (the Annual) Meeting, onDecember 12th. Before thstdate ‘complete print ofMr.Addenbrooke's paper willbeinthehands oftheSecretary, ready fordistribution, upon application, tothose desiring copies. We have some time still fordiscussion this evening, andIhope themembers oftheInstitution willuseitby giving their views upon thecommunication which hasbeen read. Fotemer Professor Gi.Forses: ‘Thefeeling which Ihave badduring thewhole time oflistening tothis interesting paper hasbeen ‘oneofadmiration oftherealistic wayinwhich Mr.Addenbrooke hasbrought before usathoroughly good idea ofthewayir 14683 DISCUSSION. 799 whichelectriclightworkisdoneintheUnitedStates,Imustprs saythat one can see inthe face ofitthat itisavery exact representation—anyone can seethat—and Ican bear witness to itasrecording exactlythecondition ofaffairswhenIwasthere last,which isnearly eighteen months ago; andapparently there hasbeen butvery little change inthegeneral way ofworking from the time when Iwas there. Itseems tometobevery much thesame; andinthat part ofthepaper which deals with the general modes ofconstruction and working ofelectric systems, whether itbeareorincandescent, Imust saythat I thoroughly endorse theopinions that Mr.Addenbrooke expresses. Itisavery clear and good account ofthework, Itisun- doubtedly true what hehassaid inhisconcluding remarks—that their general modes ofconstruction arerough-and-ready, andare ameyesore, and may sometimes lead todangerous results. But although they arerough-and-ready inmanner, thecountry isone which feels that they want toextend these modern developments asquickly aspossible, and that ifthey waited todoitinthe expensive wayinwhich weareaccustomed toconsider necessary, itwould bealong time before thesame extension would bemade tooutlying districts;itexistsatthepresentmomenteveninthe farWest ofAmerica, AsMr. Addenbrooke haspointed out, itis justthesame asregards theAmerican railways: when wetravel onthem weareaccustomed tosaythat anEnglish engineer would not build such arailway—such light work, and generally such bad permanent way, comparatively;butanAmerican immediately answers you:“Lookatthethousandsofmilesthat “we have tocover, theenormous distances that wehave togo; “if wehad put down the whole ofour railways inthe same “substantial and expensive manner asyouhave inEngland, we “should not have one-fourth ofthe facilities, and the trade of “the country would not have reached thepresent position that “it as” Inthewhole ofthis paper Idonotseethat there isanything tocriticise, ortoaddto,inthematter;andIcanonlycongratu- late theSociety onhaving thecondition ofthings soforcibly broughttotheirmind.Ithinkweareontherighttrackcertainly, 200 ELECTRICAL ENGINEERING INAMERICA. (Nor.2h Peter butwedowantouroutdoor workcertainly moresubstantial, securely, and permanently erected, and weshall beallthe betterintheendforthereminder. Wehavebeenveryslor inbeginning ourcentral station work onalarge scale, batnow that wehave begun it,wearedoing itpretty thoroughly, andI donotthinkitwillbelongthattheUnitedStatescansaythattheyarefaraheadofusinthisbranchofthepracticalapplications ofelectricity. Thave said that inthedescriptions which Mr.Addenbrooke hasgiven usIhave seen very little difference from thecondition that things were inwhen Iwas last inAmerica, and yet, inthe correspondence Ihave with friends over there, Iamcontinually hearing ofvery great strides that arebeing made; andIbai hoped that Ishould have seen inthe paper some account some further advances, though, however, Imust say Ididot ‘expect much new information inthe allusion toelectrical tram- ways. When Iwasover there, there were very fewelectrical tramways, and, curious enough, Iwasmore often inanelectri tramway driven byaccumulators than Iwasonelectric tramways driven otherwise. Onthree separate occasions—in Boston, Phils delphia, and New York—I travelled ontram-cars driven by ‘accumulators, andIwasonly ononedriven otherwise. Those were allmostly experimental, whereas now itistotally the other way. Thadhoped that Mr.Addenbrooke would have been ableto give ussome information onthe advances inAmerica incertain details which are interesting usatthe present moment vey much indeed. Iwould specially mention the question of motors, and particularly alternating-current motors. Webare allheard avery great deal ofwhat isbeing done, butthe accounts that come tousaregenerally filtered through vario media, and wewould have liked tohave had some dirt description from theUnited States ofwhat isbeing donebsmeans ofthealternating-current motor. Iseethat late! ‘M.Zipernowski hasstarted hislarge station formotor work, andthatreally seems tobeamore practical thing thansoya the alternating. current motor work that wehave accurate it formation offrom America. 1860) DISCUSSION. oor Also,Ishould havelikedtohaveheard something moreposmor about thepractical useofmeters intheUnited States, especially foralternating-current work.TheuseoftheEdisonmeterwe areperfectly familiar with, andweknow how very largely itis used inAmerica; butthealternating-current meters have only lately been used, and Iwould have been glad ifMr. Adden- brookewouldhavetoldustowhatextenttheyarebeingused, ‘andhow thedifficulties ofusing meters arebeing met. One ofthechief difficulties inusing meters which wealways come across when wearethinking about putting them in,is their small range; and ifthemeter hasonly arange of10,or 12,or15,asisgenerally thecase, there isadifficulty. One meter willregister from onelamp to12lamps, butwhat areyou todobetween that and 100lamps? Are you tobegin with registering only eightornine,andleavethefirsteightornine without record? orwhat istobedone? These arethepoints onwhich Ithink itwould beinteresting, ifthere isanyinforma tion from America, tohave itgiven us. ‘There isone feature inAmerican factories which Inoticed when Iwasover there, and from what Ihave heard from friends inAmerica since, Iknow has been extended still more, about which also Ithink some information might well have been given us,andthatisabouttheenormous amountoftimeandmoneythathasbeenspentinthefactories upontests. The laboratories ofallthelarge manufacturing companies aresplendid establishments, and Iknow that since Iwas there inmost ofthese thesystem ofaccurate testing has been in- creased very much indeed; and itwould have been useful to draw attention tothis fact, because inthis country itistoo often the case that itisnot done—there isnot sufficient test- ing. Iwill mention one point inwhich Ihave noticed the difference very much indeed, and that isinthetesting ofcon verters, Idonot think any converters inthis country have been putthrough areally thorough setoftests forallthe different sizes ofconverters that aremade, and fordifferent out- pats, and Iknow that itisdue tothecareful tests that have beenmade upon theother side that theconverters which arein 02 ELECTRICAL ENGINEERING INAMERICA. (or$88, Frtmer useoverthere aresomuch more eficient thanmany ofthow which are inuse on this side. Tam glad that Mr.Addenbrooke hasdrawn attention tothe importance, which hasbeen overlooked somuch, ofhaving ow mains, even onhigh-tension alternating-current systems, sup plied byfeeders. Ibelieve there isnot asingle central station inthis country ofhigh-tension alternating-current, distribution inwhich feeders areemployed. InAmerica, Ithink that with every large central station feeders areemployed asuniformly with them asthey areonthe low-tension system. Thereisonlyoneotherquestion Iwouldliketohavebad some information from Mr.Addenbrooke about, and that iswith regard tooneclass ofunderground wiring, which Iamnotquite able torecognise from the description hegave. Ishould be glad ifhewould telluswhat isthetype ofcable which hebas described:probablyIshallknowitbysomeothername, Mr. G.L.ADDENnRooKE: The Calder. Professor G.Fornes: Idonotrecognise it. Moun, MrH.C. Dosovan: Ihoped that Mr.Addenbrooke would have dwelt more fully onthe atmospheric conditions ofthe American climate. ‘Theextraordinary dryness oftheairnaturally renders certain precautions astoinsulation utterly unnecessary io thenorthern parts oftheUnited States and Canada. Acommoa blue glass insulator stuck ontoa piece ofdeal, and that nailed ontoapinepost,would,inthiscountry,andmoreespecially in London, besoon asource ofvery great lenkage forbeitremem- ered that the conduction toearth would not bethrough the glass andwood, butover thesurface when such surface isdamp and dirty. Therefore, inadry, clean atmosphere, blue glassot wood isquite asserviceable asthe finest porcelain and ebonite. Itisquite natural toseethat adryclimate isapremium on carelessness and rough-and-ready work. What Mr. Addenbrooke has sographically described with regard toelectric lighting work, Ihave myself observed with regard totelegraphic intallations some years agointheStates, andhave heard many others remark onthesame subject. The accepted opinion isthat the superior brain-power ofthe 18.) DISCUSSION. 08 Americans enables them todisregard insulation, &c.; but, from y.. @process ofobservation and reflection, Ihave come tothe™”™"™ conclusion that itisnot superior brain-power that favours clectrical enterprise inAmerica, but climate. When onthe introduction oftelephonic and electric lighting enterprises in thiscountry, most ofthe installations put upbyAmerican electricians soon proved, onaccount offaulty insulation, in- efficient, thefaultwasnotinthebrain-power oftheAmericans, batdue totheEnglish climate, which, oneneed hardly say, is attimes damp. Mr.W.P.Graxvitte: Although, perhaps, alittle apart from sr, theimmediate subject under discussion, Ishould liketocall°™"™* attention totheunsatisfactory way inwhich theinsulation resist~ ance ofelectric light mains isusually expressed—that is,by simply stating thenumber ofmegohms perknot, irrespective of thediameter oftheconductor. Nowthatwehaveinsulated wires ofsuch widely different dimensions, itseems necessary that the dielectric resistance should beexpressed interms ofthediameter aswell asofthe length ofthe conductor, because itsinsulation varies notonly with itslength, butalso with itsdiameter. Professor S.P,Tnowrson: Though itisfour orfiveYears prtcuor sinceIwasintheStates, Mr.Addenbrooke's paper brings outso™“"""™™ forcibly thestyle ofwork outofdoors there that IfeltasifIwere buck again intheStates. Iam disappointed inthat fact, because Thad hoped that fiveyears would have made agreat difference inthestyle ofoutdoor work that wasinvogue then inAmerica. One thing, ofcourse, isnew since, and that istheuseofalter nating currents with transformers. In1884, when Professor Forbes lectured inPhiladelphia onalternating currents, hecould notfind inthewhole oftheUnited States analternating-current machine wherewith toillustrate his lecture. Ishould like torefer totheglass insulators. Ibrought one back with mefrom the States, and will have ithere next meet- ing. Though alike inother respects, mine differs from Mr. Addenbrooke's inbeing even more primitive:thepieceofwoodis simply chopped toarough point with ahatchet, and theglass insulator isforced upon it;thereisnoscrewonthewood.The 808 ELECTRICAL ENGINEERING INAMERICA. [Nov.280, Prater back ofthepiece ofwood ishollowed alittle voastoftaguine ‘Beers soyconvenient treetrank. Incidentally itfarnishes anintered- ingillustration ofthewayinwhichthepatentlawsoftheUnited States encourage small inventions;fortherewasaspecialpate taken out inthe States for the manufacture ofawooden amts holdaninsulator, thebackofwhicharmishollowed outsoasto fitagainst atreotrunk! Mr, Addenbrookehasreferredtothelongcirenitsonwhichse lights areused intheevening asbeing applied now intheday- time forthedistribution ofpower. Iwish hehad been ablete sayalittlemore,becauseIwanttoknowwhatkindofmotorsxe uted running onlong circuite ofthis kind. Are they consaa current motors? How have they been made towork asconsta- current motors? Orare they simply machines applied dynamos—Thomson-Houston orother dynamos—and used a motors?Mr.Addenbrooke makes theremark that thelamps aremade roughly as,ormore #0than, theaverage arclampsinthiscoon, andyetonehears somuch about thesteadiness ofelectric lights onthe other side. Iam inclined tothink that the difference is, afterall, notinthemachinery—either intheengines ordynamoe— orintherough-and-ready ways ofwiring, orinthedesign and construction ofthe mechanism ofthelamps;Iaminclivedto think that the main difference lies rather inthe men—thaia Americatheelectrical engineer whohasthechargeofastatin, orwhoisperhapsonlyalineman,takesagreatdealmorepersoalpride inkeeping hismachines, hiswires, andhislamps ingoo! order than theaverage engineer whom wecanmanage togrovit this country. The working engineers inthis country net seem tometohave exactly thesame amount ofpersonal pridetheAmerican, ‘TheAmerican engineerswilltellyouandenfort upon you that their system isthebest intheworld, andthat, there isnoequal toit, Every lineman, every fitter, belies | devoutly inthesupreme excellence ofthesystem upon which be isatwork, andwillnothear aword against it,orin favour of}- bodyelse's;hehasapersonalprideinhisown.Itisadifferen? oftemperament, perhaps, andmay beduetotheir climate—tht aplendid climate bytheside ofwhich ours issovery dismal. 1980) DISCUSSION. 205 Mr.J.Swavaonxe: Onepoint Mr.Addenbrooke hasnotue mentioned isthedifference inthecommercial management, or theorganisation ofAmerican electric light businesses. In Fogland afrm isusually run with one man asahead, who ispartly anelectrician, partly abusiness man, partly anengineer, andpartly acommercial traveller. InAmerica they separate all these, ‘Theelectrical man hasnothing todobuttodesign appara tus: hehasassistants under him;andallthecostsoftesting, andofeverything made experimentally, arekept separate, and areallowed forinaseparate account. When heisdesigning machines, they arenot forspecial orders, butsimply forwork thatisgoingtobeputinhandandmadewholesale.Hehasto perfect hisdynamo, and when itisdesigned hehas nothing more todowith it;itgoes into thehands ofthesuperintendent ofthefactory, whose business itistokeep tothedesign exactly, andmake large numbers ofthemachines without any variation whatever. Whenever .work can bedone inthat way the Americans can beat us,and the reason isobvious. Ishould like very much toask what sort ofthing the Zipernowski motor isthat wehear about. We have seen accounts and pictures ofit,but thepictures arealways care- fally arranged sothat you cannot seeanything. Referring totheinsulation ofcables, Iwish—and Idaresay great many other people wish—that the way ofmeasuring insulation and conductivity ofcables inmegohms, permile or perknot, wasgiven up. Ifresistance istobeused atall,itmust beinmegohm-miles; butIdonotseehow thediametercanbe taken into account,astheleakagedoesnotvaryasthesurface. Itwould make itrather complicated, and Ithink itwould be better tospecify themegohm-miles foreach particular conductor and insulation. ‘Asregards central lighting generally intheStates, Ithink wearea good deal misled inEngland bycontinually forgetting theprice ofgasinAmerica; Idonotthink itreally isthat the Americansaresoverymuchaheadofusinelectricity,butthat they arelamentably behind usincoal gas. ‘The Pauswent: Ifthey cannot get their coals ata Vol. xvin, 55 608 ELECTRICAL ENGINEERING INAMERICA. [Nor.%, moderate expense, they cannot get their gas atamodente expense, Professor G.Forses: There aretwo kinds—gas coalwi steam coal, Pisvam, ‘ThePuusipenr: Iamsorry thatIshallnotbeabletobe present this day fortnight, Ishould like very much tohare heard Mr.Addenbrooke's reply, but Itrust Ishall seeitallia print. For myself, Imust saythat Ihave listened tohispaper with exceedingly great interest. Iamsure itwill domuch good tousinthiscountry; itcertainly isofahighly stimalating character. The very description ofwhat hecalled “rough-and-reais” American methods should becarefully weighed inour minds, astogetwhatever good wecan outofthe“rough-and-ready" method. The business oftheworld—in railway engineering and agreat many other things incommerce and colonising— hhas been done byrough-and-ready methods, and electrial engineering hascertainly been promoted very largely intht way. The very beginnings inelectrical engineering were roug- and-ready, and ifwestop toattain toomuch perfection is common details weshall lose inthe race, orbefurther bebis! intherace than weneed be. But when danger isinvolved it rough-and-ready methods they must berestrained, and instes, every practicable precaution, andevery attainable perfection of appliances toensure safety, must bedemanded oftheat structor. Iquite agree with Professor Forbes astothedifficulty ia understanding how these currents could beused formotors. ‘Astothe question which Mr. Swinburne put toMr.Adder brooke, Ithink itaveryimportant one,Wecertainlyarever" anxious toknow, andtoknow assoon aspossible, what practical methods there arefordistributing power byelectricity. We have hung back agreat deal toolong; weought tohave sewing machines, lathes, andmachines inworkshops, alloverevery plat thatissuppliedbytheelectriclight,nowdrivenbyelectricity. +toelectric propulsion forcarriages, theoverhead wires thatMi Addenbrooke hasdescribed willbeaverygreat difficulty inde 1888.) DISCUSSION. 807 inthiscountry inthewayofelectric traction. Iwould like toPt... askhim thequestion, How isitpossible, inthe towns, tohave thewires lowenough togive thecontact totheelectric car,and yethigh enough tobeoutoftheway ofthings that areoften carried through thestreets, such ashighly loaded waggons, fire escapes, and many other things that reach upmuch further than would be convenient for the current-catcher of the electric tram-car? . Mr.G.L,ADDENsHOOKE: Iwould havepreferred answeringws,Ati these numerous questions atthe end ofthe discussion next meeting. Ihope you willnotexpect metobeasortofwalking dictionary onAmerican engineering ofallsorts; but asfaras farther advances go,Iratherputthequestionononeside.It hasbeen really, perhaps, Professor Forbes's special linetohave told ushitherto allabout American advances, and Iwasnotmyself so anxious tohear thevery latest thing that wasdone aswhat was really the practical position ofaffairs altogether. We know perfectly well that, however good aninvention is,ittakes twoor three years asarule before itworks out tobeofpublic use, and therefore Ididnotgointo those parts ofthesubjects perhaps quite somuch asImight otherwise have done. ‘With regard tomotors onarelight circuits, Icannot tellyou very much, asIdidnot actually seeany ofthemotors atwork, butIwill trytosupply some information later. Isaw other motors atwork driving printing machines andsoforth, andgiving great satisfaction. ‘Asfarastesting andexperimenting go,Iquite bear outwhat Professor Forbes has said. There isnodoubt that the Americans arebeginning tobecarefulonthispoint,andtoappreciate the value oftesting. But you must understand that different localities differ very much. America isnot,oneplace anymore thanEuropeis;andbecausetheydosoandsoinoneplace,itdoesnotfollow that thesame isdone everywhere else. Ofcourse the dryness ofthe climate has anenormous influence ontheir work; andnotonly thedryness oftheclimate, batwhen they have rainitcomes down heavily, andthey donot get thedrizzling days which weget here, and which ruin all 08ELECTRICALENGINEERINGINAMERICA.[Nor.28, ticAdd insulation, Asmostpeople know, agood heavy shower will improve insulation after itisover, but itisthe constant damp and drizzle that affect ushere. ‘Asregards theinsulation ofwires. Itisperhaps alittleoutof the subject, but Imust sayIthink that intalking about the insulation ofelectric light wires the absolute thickness ofthe insulation should alwaysbementioned. Ofcourse,iftheinsalation isexpressed interms ofthediameters ofthecore andinsulation, wecangetatit,butitisvery important weshould know the absolute thickness oftheinsulating material: ifitisafiret-class | material, andeothick,weknowitislikelytoanswerourpurpose. Itwasremarked byProfessor S,P.Thompeon that there does notappear, from mypaper, tobeanyvery great difference inthe overhead work since hewasinAmerica fiveyears ago. Well, asI have pointed out inmypaper, there isnodoubt: that from an engineering point ofview thething answers itspurpose fairly well, ‘andcould notpossibly becheaper; consequently, until thepublic protest, thecharacter oftheoverhead work isnotlikely tobe greatly altered. ‘The interest which men take intheir work was also referred to byProfessor 8.P.Thompson, ‘Thereisnodoubtthatthatisvery much so. For one thing, their wages are ofcourse better, employment perhaps more constant, and thechances ofgetting onbetter; and these arethings which always make men take a greater interest intheir work. With regard tothePresident’s question astothecrossing of streets forthe electric tram-car connection, the fishing-rod connection extendsabouteightorninefeetabovethetopofthe tram-car. This leaves agood deal ofroom fortraffic topass under. Ifa cartofhay,forinstance,werepiledupveryhigh,andthewireswerenotatasufficient heightforittopassunder,Isuppose it would have togoround. ‘The Puestext: How about the short-circuit between the two wires ? Mr,G.L.ApveNsrooxe: Ididnothear anything about that,butinlargetownstheyusethis[drawing sketch].Postsarefixed with double cross-arms upon them, and thewires arerunso, It 1880.) DISCUSSION. 9 isanexceedingly convenient wayofdoingit,andlampscanbeasAdier putonthetop. ‘ThePaestvent: Allother carriages keep outoftheway? Mr.G.L,AppeNsrooxe: They must take care also tokeep outofthe lineofposts. ‘They cangoontheline oftramway as longasthey keep clear ofthat part oftheroad where theposts areplaced. ‘TheParsient: The discussion will now beadjourned until thisday fortnight, when theAnnual General Meeting will be held. Aballot took place, atwhich thefollowing were elected :— Foreign Members: Edward J,Hall, jun, | Axel Hultman, Masayuki Otagawa, Members: William Henry Allen, |_—_‘William Alexander Bryson. Associates : GeorgeCarnegie Alexander. |George8.Hooker.William John Alexander. ©.Frewen Jenkin, Arthur J,Arnot. Hugh Haweis Paynter. Francis Gibson Baily. Peter Theodor Raaschou. William St.John Beale. William John Harrison Ryder. Henry W.Bowden, Charles Sheldon Thomson. Vernon Keble Cornish. Hubert Edwin Tutte. Herbert John Dowsing. ‘Thomas Savile Watney. HenryLlewellyn T.Foster. |WilliamWynnWilliams. Students: Sidney Hopwood Blake. | William Rowland. ‘The meeting then adjourned. 10 ABSTRAOTS. W. XOMLRAUSOH and 0, HEIM—EXPERIMENTS WITH ‘ACCUMULATORS FOR STATION WORKING. (Eletrotchrtche Zeitwchrift, Vel. 10,p.308, 1889.) ‘Thetwocellsonwhichtheexperiments werecariedouthadformedpartof Abattery which badbeen indally usefrom October, 1882, ullDecember, 1887 Each cell contained three poritive and four negative plates; thesurface ofeach plate was 25square dm, and the liquid was dilute sulphuric acid ofspecie gravity 1115, thequantity being 34 litren, ‘The cell, whea ready forworking. weighed about20kg.‘Thecapacityofeachcellwas45ampere-boen, thenormal charging current was 5amperes, and thedicharge current 6°SamperesWiththenormalcurrenttheE.M-F.rienouchargingfrom20902voltsat {heeadofhalfanhour,theafallsagainto2055,rsingatheendoffourhoursto52, andthen increases weaily tilltheendoftheeighth hour, whes itrise rapid,finallytouching2°84volts,Ondicharging withanormalcarrenttheE.MF.Sirs ies lightly from 1-022 to1928, and only again reaches itaformer value ater shoat one and ahalf hours; itthen falls offvery regulary til the end ofthe discharge, The remarkable riso ofboth curves should probably beattributed to some hind ofsecondary action. ‘The resitance was very carefully messared, a4 ‘eas found tobeproportional tothe charge and tothe dcharge respectively.‘Thedensityoftheacwastakenregulary,andfoundtobedirectlyproportionaltothesate ofcharge ofthecell; itvaried from 1-115 to1147. This value agrees ‘extremely well withthe value arrived atfrom theoretical calculations these givirg 1146, from which theobsorved value difered only bythetied decimal place. ‘Aconsiderable number offurther experiments were made with current: exceetng theuormal, thedetailed resuts ofwhich areallgivea fathepaper. W. WEDDING—ARC LIGHT rioTOMETRY. (Electrotechnsche Zeitachrift, Vol. 1,p.887, 1880.) ‘The experiments were mostly carried out with aSiemens differential lamp. worked with acontinuous current of14amperes, and takingapotentialdifereace of4710.52 volts, The eandard unit oflight employed wat the German normstcandle,burningwithafameof«Leightof45min.Thismaybetakenasequtothe English aperm candle, Asanintermediate wandard ofcomparison, were sued twoAlbert burners of25t080candle-porer. Itwould, perhaps hare bee: preferable touse the amylacetato normal lamp ofvou Hefuer-Alteneck. The frelamp was suspended above the photometer bank,40atoadmitreadilyofte light being measured inalldirections, and under various angles. _ a ABSTRACTS. an ‘Thefollowingtablecontainssomeoftheresliswithanakedlight:— Angie, tat, Angie Batt. - 109 - 18640°20" 1,908 4-28 2,009or.8868 e042" so ‘TheEMF.variedbotwoen4677and47-1voltz,From aninspection ofthecarves, itappears that thelight meamured inahorizontaldirectioniaveryamall;itincreasesrapidlyupto20degrees,andthenmore slowly, tillitreaches amaximum atabout 42degres, and then once morefallsof,Soonafterpassingbelow60degresthelightalmostdisappears, owingtoteshadowcastbythelowercarbon,whilestillfurtherbelowthelampsomelightinthrown down from thecrater ofthe upper carbon, The author gives «formula forcalculating themean spherical illumination from thelight measured under the several angles allround thelamp. Experiments were also carried outwith three dierent kinds ofglobes, and itwas found thatthe mean epherial illumination lost respectively 41,40,and 58percent. ofthat given bythenaked light. Bytheure ofreflectors the diminution ofthe illumination was reduced toabout82percent. ‘The author finally treta ofthemathematical calculations forarriving atthe bestAispositionoflampsforilluminating anygivenarea. (W, MEGBAUR—EXPERIMENTS ON THE PERMEABILITY OF DIFFERENT SORTS OF IRON AND STEEL. (Electrotechnitche Zeitachrift, Vol.10,p.48, 1889.) ‘Theexperiments wero cartied outonrods ofthevarious metals 60mm.log ‘nd5mm. indiameter, twenty-two diferent kinds being experimented oninall.‘Theobservations weremadobyGaussmethod,withtherodsinthefitposition.The action ofthemagnetising collonthemagnetometer wasaullifed byplacing a second coil ontheopposite side ‘LotMequal magnetism ineg unite vvHy horizontal intensityoftheearth'smagnetiom,wvRy distance ofthrodfrom themage inem, jvNyx numberofturnsonthecoil, 1tho sectional weightoftherodistakenasabout18gr.,andweassumea ‘pecific magneton of180 c.g. unite forthe best kinds ofiro, themaximem ‘moment oftherodwillbeabout 2,400 c.g. units From Muller's formela 5300 ‘sndwith Waltenhofen’s constants themagnetising force fortheabove value of ‘moment comes outas* ni=7,000gs,units oa ansraacrs, Te anitrs rol show thtongeeheBsroaa comentadnetontherslidpce”Anlxnemoofhichhepermaaagraatof smstte,teply ohcnarconofgnel utrAnongerrseelebnappcet sa.TheratemarlfreofnetaopermeSieh le Magi hafoa a wih sagt fo oe sm homage oneness Te Caprinents Iav oan tore vale ofsont gain» Therne br {io torn, win hrsn te tard byBeg ns heume Car fre magnon dnt nh ey ee De. ¥, WERFLISRAON-TIIE INDUCTION COILS OF MICROPHONES. (Ech Zalchi,V387388) "Teexpewrecaela0ollhepalofAe waacs blow l= = — Foaxar| ar|ea|os[am|om]our[ogCoitNo.2... 10|os|125|190|912]osm|4a00 ‘Ingachcoilsthemostinteresting thingisthecourseofthemagneticindaction imtheprimarycoil.Inscollwithanironcoretheinduction maybeconsideredfrmls opoftincemodyuni te ndton ft cea Seni ntecon thon cao inwot ae omer ne ‘nan conpron nih sear, yt tog. Inthe cperiontprimaryowrepicsi haranonsetemeTaescndaryclaitenanosvarcameophspivotdean rcunce aay sertn intherss IntwrayCielcrestingcnnmaga teinheteda theinduced current inthe secondary coil. Although theprimary windingoftail No Thedmor anal ray fr ansie Pan oe cal ot, yet frsql nino tt cert faa iatha hat os cal Na 2Bach spent hcl No? cs Shey abe nce ch ta amperes alHe yt bag ‘accountofthemorefavourable arrangement oftheironcore.Again,thegreaterTeaistanceofthecoilNo.1makesthemicrophone lesssensitive. Henceasl ‘oathcoldpaobhelle ABSTRACTS. as ‘Dr, LHONHARD WHBER—ATMOSPHERIC ELECTRICITY. (Electrotechninche Zeitechrift, Vol. 10,p.881, 1880.) ‘Tho article deals with tho rosulu obtained daring ayear's observations More generallytheobservationsweremadobymeansofkites,butoccasionally& Yalloon was sent up. Agreat nomber oftables and curves are given ofthe rmeararementsofthecurrentsoberved. ©.GRAWINKEL—CONNECTION OF ACCUMULATORS FOR ‘TELEGRAPH WORK. (Electrotechniche Zeitechrift,Vol10,p.44,1889.) ‘Thearticledealswiththebestarrangementwhichcanbeadoptedforsap- Piyingtheveryvariuswaatsof«telegraphlinebymeansofsecondarybatterie, 10that both open aad clowed clreuit working can becarried onsaccemfally.‘Withoutthediagramwhichaccompanies thearticle,itwouldbedicalttomake thearrangement clearly understood .BOUTY aad Z,POINCARS—CONDUCTIVITY OF MOLTEN SALTS. (JournaldePhysique,Vol.8,p.968,1889.) %,Porwcang—IDEM. (CournaldePhysique,Vol.8,p.813;ComptesRendus,Vol.108,p.17,1889.)Taworkingateachhightemperatures at850°10500°therearetwogreatAificatiestobecontendedwith:oneis,tatifglamtubeisusedtocontainthe saltunderexperiment atmuchtemperatures itswallsbecoixemoreorlewcon-Anctory;theother,thatthermo-currenta aresetupatthejunctionsofthesaltand theelectrodes used. ‘The firet imore particularly troublesome when analloy‘ithaapecidemeltingpointiousedasthesourceofheat.‘Thisdefectwasgot idofbythe simple expedient ofusing anair-bathastheheatingmedium.A ‘pecial arrangement ofelectrode sered toeliminate thesecond: thisconsistedof ‘2ingalated vowel containing asolation ofthesaltunder experiment, thebottom ofthevowel being stopped byaplugofSbrousasbestosinsuch@mannerthat ‘hiletheupperpartoftheplagwasinthesolution,thelowerpartwasimmersed inthe molten alt. ‘Aconsiderable number ofexperiments were mule with potamiam nitrate. Taconductivity may becalculatedfromtheformula = 07241 [1+01005(¢=3509), todthecalenlated values agreo very well with those obtained from experiment. Working with mixtaresofpotamiumnltrtoandsodiamnitrate,itwasfoundthat thejoint condactvity asmearured agreed closely with thevalve calelated from theconductivities ofeachsalttaken separately. Having observed that silver which bad been exposed totheaction of«bathofturednitrateofsilverremainedunpolarsed infrednitrateofpotasiam,it au ABSTRACTS. camesimpletoreplacetheashenlacrodebyothersofserwinise ducinganycutrbinginduence, ‘Thespecieconductivityoffusedmimeofrer isfound fom the formsla = 1288[1+0.005(—2809); ‘hatis,fortemperatures between200°andB70.ItSssetnthisformaht thetemperature coeficentlediferestfromthatforpotbiteadfosodinm nitrate. Experimenting onammoniumnitratetheformalawasoat c= O4[1+00078(¢—200), Thich once more gives adileeat coeficient. If, howorer, the tempentn coeficient bemultiplied fneach case byihedensityofthefedmit«conta number isobtained, asshown inthefollowing tble:— a Potassium Nitrate... ..|ss |et |0005 90 Soliam yg 0H || |Silver ne ee ed Aemonion yy“Z| awe|tae ons|a Tabing into account thegreat dificult there lsinearning exeily tt ooniin atthe igh temperntares, stwill befound that thecoefcens we Infavere ratio tothe densitier ofthecorresponding walla Itsali ©remarkedthatthemolecslardenier~0'0807forotasiamnitrateand0420fo ‘immonium nitrate, eappeing them loth broughtioaaimilartemperate 380 are very clon toeachother; them two quantiles difer, however, comidentsfromthemoleclarconductivities ~0-0597andO02—ofivernitrateadwlanltrate reapectvely, which donotdifer much. Teiknown that iathecaesolutionsthetwoformereltarenormal,whilethewootherareabnor. ‘There is limit tothen experiments, when thetemperatures have rain! that ofthefsin ofgam; this may begotover bythe wseoftuber ofpores Experimenting oncorde, theauthor made wieofwich tubes, aad theaverelectrodebeforealladedto;theseaidnotpoletoabyextent,and,morothe amonatofpolarisaquiregular.‘Thetemperaturaweremeal reans of«platinam-rhodiam thermocouple, ‘The following formule repre thevations forpotassom and sodiam chlorides reapectioly— ¢=180(1+00066(¢~750]between700°and800; = 815(A+C0004(¢—760)between718°and80. | 1will bonoticed thatthe cofclents ofvariation are the same for both ali ‘The densities being memared atabout 70° {twas found that they were pracathemame,vie,163.‘Theproduct—1,070—te notvorydiferentfromthaohenitrates given above, vin, 120, ‘The molecular conducivitin are revpeil (0818 and 04112 forpotareum chloride and sodinm chloride, and itSsiter toremark thatthe ratio ofthe conductivities ialmostthemameastherato06° ofthe molecalar conductivities ofthe nirvter ofpotasstm and sodam. ‘When eoldieaton takes place intheelectrolyte, the reatance inswwe ABSTRACTS. 815 regularly. Atthomomentwhenthewholeofthesaltbecomessolidtheconductivity fallsto1-500thofthatofthefusedsaltatthesametemperatare;themitdecreases very rapidly, thecoeficient ofvariation being 100times what itwasbefore. In fonder togain some approximate ideaoftheeffectoftheporcelaintube,some rmearurements were taken ofitsconductivity, theresults being given bythe formala €=10(00578 &+0-0000125¢-1680), 1M. DEPREZ—ELECTRIC TRANSMISSION OF POWER AT BOURGANEUF. (Comptes Rendus, Vol.109,p.455, 1889.) ‘The sourceofpowerisawaterfallsituatedat“LesJarranda,” ontheMaalde, 8distance of14kilometres from Bourganeuf. The water actuates @horizontalturbinewhichiscapableofgivingoutamaximum powerof180IL.P.ataspeed‘of150revolutions per minute, From thetarbine thepower istransmitted tothe dynamo onthefloor above bymeans oftwobelts, there being twopulleysonthe ayoamo.‘Thegeneratorhastwoarmatures,bothononeaxle;theyarewound ‘with wire 22mm. indiameter, and have each aresistanceof2ohms,Thesection ofeach wire is3'8aquare mm., thas giving asection of7°6squaremm.forthe current, which may be85amperes without excessive heating. ‘The twoarma-‘turesarejoinedupinseries,andcangiveanEMP.of5to65voltsforeachrerolution perminute, ‘The electro-maguets areseparately excited byacurrent of20amperes and 90volts. ‘Thelinoismetallic,andconsiataofasiliceousbronzewire6mm.indiameter,carried onporcelain insulators. ‘The total resistance oftheline is28ohms; the Ingulation iafound tobevery perfect, even inwot weather. ‘Themotorisofexactlythesametypeasthegenerator,andisalsoseparatelyexcited, current being taken either from oneofthelighting dynamosorfrom battery ofaccumtlators. ‘The motor isused todrive thedynamos which supply current forlighting thetown; these areGramme dynamos giving anoutput of 250aunperes and 120volts ‘Atsome preliminary experiments made with thegenerator and motor, beforetheywereinstalledatBourganeuf, withanartificiallineof30ohms,thelightingAynamos gave collectively anoutpat of385 amperes and 112 volts when the ‘generator was developing 22amperes and 8,750 volts, Inasecond experiment, ‘with 25obms resistance inthe line, the lighting dynamos gave 876 amperes and 115volts when thegenerator was giving anoatputof20amperesand3,550volts. ‘The lighting dynamos used having commercial efficiency of80percent., whilst the Deprex dynamos are said bytheauthor tohave acommercial efficiencyof 90percent, theresult actoally obtained might beimproved upon ;andthere is nodoubt tbat «power of60H.P, might beobtained atBourganeuf with 100H.P. ‘atthe turbine at Zee Jarrande.” a6 LIST OF ARTICLES aeuarixe 70 ELECTRICITY AND MAGNETISM ‘Appearing insomeoftheprincipal Technical Journals during theMonth ot JULY, AUGUST, SEPTEMBER, andOCTOBER, 1889. L—BATTERIES AWD ACCUMULATORS. E,Drxovoxnt—Gendron’s NewBattery—Lam.El,vol.89,p.27,188.B.Kotnié—Potential DifferenceofGalvanicCella—Beiblatter, vol.13,p39618% C.Hxme—Effect oftheDensity oftheAcidontheCapacity ofAccumslaen— Beiblatter, vol. 18,p.409, 1889. W,Kouznavsom andC.Hrtx—Researches onAccumalators forStation Work — El.Zeit,vol.10,p.927,1889. ©,Grawnnxt—Conpling upAccumulators forTelegraph Work.—FI.Zeit,tal10,p.446, 1889. 1L—DYNAMOS AND MOTORS. R.V,Proov—Varions Dynamo Armatares.—Lam, El.,vol.88,ps7, 1889. G,Ricuanp—Details ofDynamoConstruction. —Lam.£1,vol.83,p.519,188%©.Rarontan—Secondary Phenomena ofInduction inDynamos—Lem, Elyv3. 38,p-605, 1889. E,Mertax—Rechoiewski’s Dynamo.—Bull, Soe. Int vol. 6,p.285, 1889. IL.—ELECTRO.CHEMISTRY AND ELECTRO-METALLUEGT. 8.Auaummive—Electrolytic Dissociation e.Hydration. —PhilMog.,vl.2880,1889, E.Doren—ElectrolyasofDistilledWater—C.R,vol.109,p.108,1889. N.Purscaixorr—Varistions inCurrent Strength daring Electrolysis—C.R ‘ol. 109, p.195, 1889. L.Potxcanné—Condnctivity ofElectrolytes atHighTemperatures —C.2,vl 108, p.174, 1889, G,Ricuano—Electro-Metallargy ofAlominiom.—Lam. £1,vol.8p.151,188. AL,Poxtmitas—Energy Absorbed inElectro-metallurgical Procemes—Lex.El, vol. 88,p.251,1889, 8.Aunaexivs—Modern Theory ofElectrolytey—Lum. El,vol.88,pp40,458 518, 568, 1889. E.Coux—Absorption ofElectric Omcillations inElectrolytes—Ana, vol.3& 217, 1889, W,Osrwan andW.Nuaxst—Free Tons-—Deibliter, vol.18,p.898,188%. ARTICLES RELATING TO ELECTRICITY, Ere. ar 1M,LoxnandW,Nanxsr—Conductivity ofsomoSilverSalts—Beibliter, vol.18, 1.395, 188. ASoxovorr—Electrieal Oscillations inElectrolytex—Beiblater, vol.18,p.402, 1869, B.Nenei—Blectro-Crystalliation ofCopper.—Beillattr, vol.18,p.686, 1889. 1V.BLECTRIC LIGHT. E,W,Surru—Shant Transformer.—Phit. Mag, vol28,p.182, 1889. G,Rroanp—Detals ofGlow Lamps.—Lum. El, vol. 88,.10,1889. G.Ricmaxp—Are Lamps.—Lum, El, vol.88,p.61,1889. B.Dixvpoxi-—Clerc’sElectrle Metor—Lam. 51, vol. 88,p-155, 1889. ‘A.Pataz—Photometry—Lum. El,vol.88,p.255, 1889. E,Divvoxné—The Electric Light attheSt,Lazare Railway Station.—Lum. £2, ‘ol. 94,p7,1889. A. Paraz—New Form ofViolle's StandardofLight—Lum. Elyvol.34,p.81, 1889. ‘H.Kavaen and C.Roxox—Channel Spectrum ofCarbon inthe Electric Are— “Ann, vol8,p.80,1889.E.v,Gorman—Photographing theElectricSpark—Dilitter, vol.18,p.422,1888, —Storme—Dependence ofCandle-Power ofGlow Lamps ontheCurrent Strength. —Baiblatter, vo.18,p.821, 1880, ‘A.Husarmat—Ilumination produced byAre Lamps inaStraight Line.—Bull. ‘Soe. Int, vol.6,.827, 1889. W.Wappino—Photometry ofAreLampt.—El. Zit, vol10,p.987, 1860. K.Wrrsxan—Laying Underground Ceblo—£t. Zeit vol.10,p.957, 1880. Dr.'T,Exnanp—Banson Photometer.—El. Zeit, vol. 10,p.877, 1889. R,Remuaanx—Alteroate Carrents r.Continuous Currents forCeatral Stations.— El.Zeit vol 10,p.897, 1869, V.—ELECTRIC FOWER. M.Lescano—Tranamimion ofPower byAlternate Carrents—C, 2,vol.109,p. 172, 1889. M,Darnxz—Resolts obtained attheBoarganeut Iostallation.—C. R., vol. 109, DP.894, 456, 1889. M. Lxntanc—DistribationofEnergybyElectrllty.—Lum. Ei,vol.88,pp.101, 220, 263, 1889. BP,Mancrti.sc—Electric Despatch Boat.—Lum. £1, vol. 88,p.268, 1869, M.Luntaxc—Tranamission ofPower byAlternate Carrents.—Lum. £1. rol. 38 pP.858,61. #,Dirovoxné—ElectricHoistingGear.—Lum. El,vol.33,p.431,1869, G,Ricmano—Electric RailwaysandTramways.—Lum. El,vol.38,p.465,1889. 818 ARTICLES RELATING TO ELECTRICITY, Era. ‘Vi—MAGNETISM AND ELECTRO-MAGNETISH. 5.J.Tuomsox—Magnetic Effects produced byMotion inanElectric Fidi— Phil. Mog, vol.28,p.1,1889. ‘A.Taxanapari—Thermal Effects produced byReversals ofMagutintinio Soft Iron.—Piil. Mog., vol8,p.207, 1889, L, Kemanexs-Lenocuowaxt—Actionof«UniformFieldonaMageeteBot. J. dePh, vo. 8,p319, 1889,P,Janzr—Heat ofCombution ofIromiaMagneticFeld,andThermo-mauesePhevomena.—J. dePh, vol. 8p. 812, 1869. L,Kvtr—Magoete Coercive Foree.—Bilite, vol. 18,p.551, 1880. P.Jounin—Dispemion ofthe Magnetic Hotation ofthePlane ofPolariso Light.—Beibltter, vol 18,p.654, 1889. W.Nronaun—Permesbility ofvarious Kinds ofIron andSteel.—ET. Zi, 10,p.348, 1889, VIL._MEASUREMENTS AND MEASURING INSTRUMENTS. ©.V.Bors—Quarts asanInglator.—Phil. Mag, vol.28,p.14,1889. C.T,Horcutmsox andG.Witxxs—Comparlaon oftheMereary Unit withte B.A. Unlt.—Phil Mag. vol.28,p.11,1880 L,Doxcax, G.Wiuxes, and C.7.Hercwmsox—Determination oftheB.A Unit in’Absolute Measure byLoreas's Method.—Piit. Mag, vol.28,p.9% 1889, E,vaxAvsni—Eleetrical ResistanceofBismuth.—Phil, Bag,vol.2,p38,1889. ‘T,Mocanavx—Cantes ofcertainErrorofMaguetograpbs.—C. R.,vo.1p272, 1889. E,Dovrr and L,Poroanai—New Method ofMeasuring theResistance f Molten Salts—J. dePh, vol 8,p.368, 189. —Pruuissten—Electromoters.—Lum, El, vol. 38,pp.16,78,11,178, 188W.pxFonvietie—Registering Instraments onthoEifelTower.—Lam. El,wl28,p.422, 1880, ©.Drcuanste—New Galvanometer.—Lum, F1,vol.$8,p.456, 1889. E,Disepouni—Miot’s Magnetic Tnductometer.—Lam. El, vo. 88,p-£1,188 B.Kotax—Simple Form ofFlectrometer.—Beititter, vol. 18,p.21, 189. ‘VIII—RAILWAY APPLIANCES. M.Cossuaxx—Railway Appliaces attheParis Exhibition—Zam.£2,vl3 PP. 201, 265, £01, 189.G.Dowoxtand—Posra.-Viwar—Motor forWorkingRailwaySigaale—BalSoc. Int, vol.6,p.382, 1889. Dr.A.Tonten—Siemens &Halske's Improved Block Systom.—EU, Za, vol.1% vp.405, 428, 1869, ARTICLES RELATING TO ELECTRICITY, Ere. a9 TE-STATIC AND ATMOSPHERIC ELECTRICITY. 6G.Foutan—Water-Spray InfluenceMachine—PhilMag.,vol28p.42,1889.H.W, Horrmar—lntcrmittent Lightaiog Flashen—Phil. Mag, vol. 2,p.106, 1882, D.Cortapox—Daration ofLightning Flashes—C. 2,vol.109, p.12,1889.E,L,Trouvetor—Daration ofLightoingFlashes.—Lum. £7,vol.88,p.70,1889.4.Fazrntna—Determination ofthePotential Difference necessary toProduce Sparks inAir.—Ann, vol. 38,p.281, 188.6.Faoxaxx—Silent DischargeinAiratNormalPressure.—Beidliter,vl.18,p.417, 1889. F,Exusn—Atmorpheric Electricity.—Beibliter, vol 1p.427, 1889. K.A.Branpen—Earth Carrenta—Beitlttr, vol. 13,.734,1889. 1,Somxcxx—Theory ofStorms.—Beilater, vol.18,p.739, 1889. Dr.L, Wenen—Experimentson Atmospheric Electriclty BI,Zit,vol10,p.381, 1889,Dr.5.Kottant—Atmorpherie Electricity.—El.Zeit,vol.10,pp.429,407,1689.Max—Storma onGerman Tand Linea in1888.— El.Zeit, vol 1,pp.45, 482, 1889, X.TELEGRAPHY AND TELEPHONY. E,Mencapten—Intensty ofTelephonie Effect—Ann,Tel,vole16,p,115,1880,=Lacaxpe—Wires ofanAlloyofCopperandMagnesium. —AunTel,vol.16,124, 1889.=Vascny—Propagation of«CurrentinaTelegraphLine.—Ann. Tel,vol.16,135, 189. P.H,Lupenonn—Zignog's New Telophonic Apparataa—Lum. El,vol.38,pp. 24,122, 1889, A.Gouurovx—Telegraphic and Telephoaic Tntercommunications,—Lam.El, vol. 88,p.107, 1889. P.Samurt—New Exchange formany Lines—Lam. £1,vol.8,p.216, 1889 E.Zarsoux—Enzmana's Telephone Relay.—Lum. El,vol.3,p-826, 1889, P.Saqoxi—New Multiple Printing Telegraph.—Lun, £1,vol83,pp.658,611, 1860, .Sacquin—Acceleration ofTelegraphic Transmission byUsing «Condenser— Lam, El, vol84,pp.27,68, 1889. J.Moosex—Microphonie Contacts.—Beidatter, ol 18,p.584, 1889. —Hisnoxtucs—Syachronous Printing Telegraph—EI. Zeit vol. 10,p.835, 1889. Dr.V.Winrtisaacu—The Induction Coils ofMicrophones.—E2Zeit,vol.10,p. 378, 1880, RPerscu—Telephonie CablesEl.Zeit,vol10,p.881,1889,K,Winmwan—v. Ryrelbgrghe's Maltiple.Telegrph—£1. Zeit,vol.1,p.490,1889, H,Disoumn—Theory ofDuplexTelegraphy —El.Zitvol.10,p.448,1889, 80 ARTICLES RELATING TOELECTRICITY, Ere. x1—TEEORY. 0.J.Lonor andJ.L.Howanp—Electric Radiation anditaConcentration by | reaneofLenses—Phil. May,vol.8,p.48,1889. | 45.7, Borromver—Expansion with RiseofTemperatare ofWiresunderPelling| Strom—PhiMoprl284190 | 1H.Henrz—Propagation ofEloctrie Waves throogh Wiren.—Pil. Meg, vol2 | e117, 1888. J.T, Borrouter and A.Taxaxaparé—Thermo-electric PostonofPasinoid— | Phil, Mag, vol. 28,p.168, 1889. H.A.Rowzaxb—Relation ofElectrostatic toElectromagnetic System—Pi ‘Mag.,vol.88,p.904,1889. E,B.Rowe—Ralation ofElectrostatic toElectro-magnetic System.—Phil. Ma.vol. 28,p.815, 1889. N.Prurtcuneorr—Elestrommotive Force ofContact.—C. Ryvol.108,p.105, |1889, |=Krovewnoti—Electro-apillary PhenomenaJ.dePh.yxol.8,p.472,1889.| ¥,Laanogor—Residatl Charge.—Lum, £1, vo.88,p.67,1889. 1M,Luntaxc andP.H.Lepenorn—General Equations oftheMotionofElectricity. Lum, El, vol. 38,pp.187, 204,278, 417, 604, 614, 1880. 1,Raom—Electrical Phenomena produced byRadiation.—Lum. £2,vol.33,p. | 100,1888. | J.Bisex and HlGurrei—Motion ofElectricity iaRarefed Gaser—Ann, vol. *| 88,p27, 1889. E,Conn—Specie Inductive Capacity ofWater—Ann, vol. 38.42, 1889, ¥,Braox—Carrenta ofDeformation.—Ann, vol: 88p-5%,1889. ‘A.Wixuniwaxx—Determination ofSpecific Inductive Capacity bymeans ofa | ‘Telephone.—Ann., vol.38,p.161,1889. | "T,Hoxrx—Conductivty ofGases.—Avn, vol.88,p.172,1889, | K.Wesrxooxcx—The Differences oftheTwo Kinds ofElectricity.—na, val. | 38,p.222,1889, ©.Souexaxx—Cyelie Change ofElectrical Conductivity.—Amn, vol.38,p.256, 1889, 45.Demuvtx—Potential Difference ofDielectric Media along theLinesof Forca.—Beibltter, vol. 18,p.890, 1882. K,Sremmante—Apparent Resistance ofConductor Traversed by«Carrent— Beiter, vl. 18,p.891, 1889. UL.Gotz andA.Konz—Volta's Fundamental Experiment.—Beibittr, vol.13, 309. P.H,Doves—The EquationforthoElectricCurrent—Beilitter, vol.18,p.399 1889, V.Dronax—Action ofSelf-Induction inElectro-magnetic Carrent Interruption. —Baiblater, vole18,p.408, 1889. —Nacont—Action oftheElectric Spark onConduetorn—Beilittr, vol. 18,p. 421, 1860, M.Tuoxs—Dependence oftheElectro-motive Position ofPalladiam onthe ‘AmoantofHydrogenOceluded.—Beiblatter, vol18,p-589,1889, ARTICLES RELATING TO ELECTRICITY, Exc. eat 0,Tnose—Analysis ofSurfaceResistance —Beiblitter, vol.18,p-648,1889.G.Leox—Eaquivalence ofanInfinitely Small Carrent and Small Magnet— Beiblitter, vol. 18,p.548, 1889, L,Bourzuaxx—Theory ofHall's Electro-magnetic Phenomenn.—Beillattr, vol. 18,p.548, 1889, A.¥,Wartaxnorex—Meaning ofvarious FormuleofMagnetiaation.—Beiblétter, vol. 18,p.651, 1889. P.Cotmann—Sparks onBreaking Cirenit—Beillatter, vol.18,p.662, 1889.K,Asexns—Discharges fromFlamesandPoints—Beiblatter, vol.18,p.669,1889.A.Riaut—Discharges produced byRadlation.—Beiblitter, vol. 18,p-666, 1889.F.Urrexnomx—Speciic InductiveCapacityofsomeKindsofPaper.—Beitlittar,vol. 18,p.711,1889. G.Liznex—Experiments onElectrical Figures onPlates Rendered Sensitive to Light. —Beiblaver, vol. 18,p.870, 1880. M. Hoon—EffectofUltra-Violet Rays.—Beiblatter, vo.18,p.731,1889.—Joouear—Hertz's Experiments.—Beiblatter, vol.6,p.918. * C.Gnawrexce:—Geometric Solutions ofsome ProblemsonCouplingupBatteries. El. Zeit vol. 18,p.888,1889. (0.Fnétice—New Method ofObserving Curves ofOscillation.AU.Zeit,vol.18, pp. 845, 369, 1889. BR, Saven—The Kelation between the Resistance ofCromed Electric Waves ia Conducting Surfacer—EI. Zrit., vol.10,p.351, 1889, ‘XIL—VARIOUS APPLIANCES. E,Dievpoxxé—Electrically Driven Tools, £e.—Lan. El, vol. 88,p.801, 1849. G.DomoxrandA.Lxravre—Synchronising Clocks.—Bull.Soc.Int,vol.6,p.810, 1889, ,Muncaptzn—The Phonograph and Grapbophone.—Bull. Soe. Int, vol.6, 856, 1889. Vou. xvi. 36 NOTICE. 1,The Society's Library isopen tomembers ofallScien- tific Bodies, and (on application tothe Secretary) to the Public generally. 2.The Library isopen (except from the 14th August to the 16th September) daily between thehours of11.0 a.m. and 8.0 p.m., except onThursdays, and onSaturdays, when itcloses at2.0 p.m. An Index, compiled bythe late Librarian, tothefirst ten volumes ofthe Journal can behad onapplicatum to the Secretary, ortoMessrs. EB.and F..N. Spon, 125, Strand, W.C. Price Two Shillings and Sixpence. —d ov ue Institution ofElectrical Engineers, Founded 1871. Incorporated 1889. “You XVI. S«wBD, 7 No.83. ‘The Eighteenth Annual General Meeting oftheInstitution was held atthe Institution ofCivil Engineers, 25, Great George Street, Westminster, onThursday evening, December 12th, 1889—Dr, Jonx Horktxsox, M.A., F.R.S., Vice-Presi- ‘dent, inthe Chair. The minutes oftheOrdinary General Meeting ofNovember ‘28th were read and confirmed. Donations tothe Library were announced ashaving been received since the last meeting from Messrs. Macmillan &Co.; Dr.J.Hopkinson, Member, V.P.; Mr. G.W.deTunzelmann ; and Sir William Thomson, President; towhom the thanks of themeeting were unanimously accorded. ‘The Secretary stated that, inaddition tothebooks that had been presented totheLibrary, hehad toannounce that Lady Bright hadbeen good enough topresent totheInstitution acast ofthebust ofthelate SirCharles Bright, executed byCount Gleichen (now Prince Victor Hohenlohe), R.A., and which was exhibited intheAcademy some years ago. Uponthemotion oftheCuainan, thethanks ofthemeeting were unanimously voted toLady Bright forherkind donation, VOL. XVII. 57 ory REPORTOFTHECOUNCIL. (Dee.12a, ‘The CHaiRMAN announced that the ballot boxes would remain open till8.30 p.m. Mr. T.Buckney, Mr.G.Driver, Mr. C.‘T.Fleetwood, and Dr.R.M.Walmsley were appointed Serutineers. The Secretary then read the following Report ofthe Couneil:-— REPORT OF THE COUNCIL. ‘The Council have the satisfaction ofreporting that the number ofnew members elected into theInstitution during thepresent year exceeds that oflastyear, which wasalready much beyond theaverage. 12Foreign Members, 31Members, 109 Associates, and 31 Students, making atotal of183, have been added totheregister, and 39candidates have been approved forballot atthe first meeting inJanuary next. 19Associates have been transferred tothe class ofMembers, and24Students totheclass ofAssociates during theyear. Our losses bydeath have fortunately been fewer than those oflast year, comprising 1Foreign Member—Mr. Holst, ofthe Great Northern Telegraph Company ;10Members, amongwhom areincluded Mr. C,H.B.Patey, one ofthe secretaries tothe Post Office; Dr.Warren DelaRue, whose valuable experimental researches inelectrical science aresowell known; Major-General Murray, R.A.; Mr. Walter Hall; and Mr. Henry Sach, Super- intendent ofTelegraphs totheGreat Eastern Railway—the two Inst named having been among the oldest members ofthe Institution ;and 5Associates. The resignations, which have also been ‘considerably fewer than those oflastyear, include 3Foreign Members, 5Members, 10Associates, and 1Student. ‘The Institution continues toenjoy thegreat privilege of holding itsGeneral Meetings inthelecture hall oftheInstitu- tion ofCivil Engineers, anadvantage which isall themore valuable inthattheattendance atthese meetings continues to increase very considerably, averaging nearly 200, and insome cases having reached over 300persons, 1869.) REPORT OFTHECOUNCIL. 235 The papers read during thesession have been mostly ofa practical nature, and have extended over different branches of electrical engineering, aswill beseen bythefollowing list. Several ofthese have given rise toimportant andvaluable dis cussions, LIST OF PAPERS READ BEFORE THE INSTITUTION DURING THE YEAR1669, Dare.Tm. Action. Jun, 24.—The Insulation Resistance of Electric LightTestallations aeomeProfAJamisox, FRS.E‘M.lnst.C.E., Member. Feb, 21.—On certain Phenomens connected with Imperfect Earth in‘Telegraph Circuits A.B.Kenwmutr, Amociate, »28.—Some Electric Lighting Cental StationsinEurope,andtheirLewons... Prof.Gxo.Founrs,M.A.,F.RSS,(LAE.Member, Mar, 26.—Laboratory Notes onAlternate-CarrentCitetitseee aePROW.ELAvarox,F.RS., V.P., and Prof. Jone Pranr, P.RS., Member. 4280m the Disturbances arising from the ‘we of Earth"forHleetrieLighting Porpowetne ew oeWellPace,F.RS,Past.President. April11.—Undergroani Conduits and Electrical Conductors ase awww SOUNB, Vanity, Member. »25.—On Lightaing, Lightning Conductor,andLightningProtectors... DrOuvanJ.LovenF.RS,Member,May28.—OntheSecartyagunstDisturbances of ‘Ships Compasses byBlectric Lighting Appliances anaes Sit Wattax Tuomsox, DCL, LLD., BRS, (L&E), Provident, +»23.—On Alternate-Current Working... W.M.Moapxr, Member, Nor. 4—On theLightingoftheMelbourneCen- teonial International Exhibition... K.L. Munnar, Member, 28—Blectrical Engineering inAmerica... G.L.ADDEXEROOKE, Awo- ‘ite, ‘The Council have awarded theannual premiums inrespect topapers read during the twelve months ending theSlst_of May last, asfollows :— ‘TheInstitution Premium, value £10, toW.M.Mordey, ‘Member, forhispaper on“Alternate-Current Working.” 226 REPORTOFTHECOUNCIL. (De12h, The Paris Electrical Exhibition Premium, value £5,to Dr.Oliver Lodge, F.R.S., Member, forhispaper on“Lightning, “Lightning Conductors, andLightning Protectors.” TheFahie Premium, value £5,toA.E.Kennelly, Associate, forhispaperon“CertainPhenomena connectedwithImperfect “Earth inTelegraph Circuits.” ‘Tar Boaro orTRave. Atthecommencement oftheyear the Council were consulted bytheBoardofTradeonseveraltechnical pointsinconnectionwith theregulation ofthesupply ofelectric energy. ‘ACommittee wasatonce appointed, whocarefully considered the several matters inquestion, and their recommendations, whichwerefullyapproved bytheCouncil,weredulyforwarded totheBoard ofTrade, and have substantially been adopted by that department oftheGovernment. The most important matter discussed was thequestion of regulations foraerial conductors, which alone occupied noless than nine meetings oftheCommittee. | Your Council being convinced that the establishment ofa standardising laboratory bytheBoard ofTrade would befound indispensable fortheproper administration oftheElectric Light- ingAct, andhaving received acommunication from theElectrical ‘Trades Section oftheLondon Chamber ofCommerce, calling attention tothe importance, from anindustrial point ofview, of the standardising ofelectric measurements, reappointed the Standardising Committee, forthepurpose ofconsidering the matter, and ofdrawing upascheme forsuch alaboratory, with estimate ofthe cost. ‘After thishadbeen done, twojoint meetings oftheCommittee and aconsultative Committee ofthe Electrical Trades Section of the London Chamber ofCommerce were held, and the draft sclieme, asthen finally approved, wasadopted bytheCouncil, and forwarded bythem totheBoard ofTrade. ‘Ajoint deputation ofthe Institution and ofthe London Chamber ofCommerce subsequently had aninterview with the President oftheBoard,tourgetheadoption ofthescheme. d 1869) REPORTOFTHECOUNCIL. ear ‘There isevery reason tohope that this recommendation, which hasalready been partially adopted, will eventually befully carried out,andthus render unnecessary anyother standardising establishment, which, however well organised, and under what- ever influential auspices itmight becreated, could scarcely expect tomeet with thegeneral recognition which aGovernment establishment must necessarily command, Paris INTERNATIONAL EXHIBITION, The Council having been urgently requested byColonel Laussedat, President oftheOrganising Committee oftheExhibi- tionofHistorical Apparatus attheParis Exhibition (Exposition Retrospective duTravail), toobtain forthe electrical section theloan ofapparatus representative oftheprogress ofapplied lectrical science inGreat Britain, took measures accordingly, andbythekindness ofthePostmaster-General, theManagers of theRoyal Institution, SirWilliam Thomson (President), Mr. Latimer Clark (Past-President), and the School ofElectrical Engineering, were enabled tosend forexhibition, inthename of theInstitution, asmall but interesting collection ofinstruments andapparatus. Your President has received from Colonel Laussedat aletter warmly thanking theInstitution forthevaluable addition made bythem totheExhibition ofwhich hehadcharge. Pans Meerixe. Knowing that considerable number ofthemembers would bevisiting Paris during thetime appointed forthesitting of theElectrical Congress, theCouncil gladly accepted thecourteous invitation oftheSociété Tiiternationale desElectriciens, communi- cated bytheir President, Colonel Sabert, tohold ajoint meeting with the Société, ‘The meeting took place onthe23rd August, your President being, bytherequest oftheSociété, inthechair, and addresses ofwarm welcome toour members were delivered byColonel Sebert, thePresident oftheSociété, andbyMonsieur E.Mascart, thePresident oftheElectrical Congress, towhich SirWilliam Thomson replied inaninteresting speech. oe REPORT OFTHECOUNCIL (Dee1, Monsiour DeChardonnet subsequently explained hismode of manufacturing artificial silk, and Monsieur J.Carpentier his electrical apparatus, called respectively the“Mélographe,” the “Mélotrope,” and the“Batteur deMésure.” The phonograph andavariety ofinteresting electrical exhibits were onview. During theensuing week, themembers were invited tovisit thefollowing establishments :— ‘The Works ofMessrs. Sautter Lemonier, The Works ofLaSociété Générale des Téléphones, TheLaboratoire Centrale deIlectricité, Electrie Lighting Stationa—Central Station Palais Royal, Grand Opera, ‘The following highly-interesting lectures were also most kindly given, viz.:—On,andrepetition ofHertz’s Experiments,” byProfessor Joubert, intheLaboratoire Centrale deI'Electricité; “On, and repetition ofElihu Thomson’s Experiments,” by Monsieur Abdank Abakanowicz, inthe Industrial Court ofthe American Section ofthe Exhibition. Bythe great kindness ofMonsieur E,Mascart, special ascent oftheEiffel Tower wasarranged forthemembers. ‘The members were also most hospitably entertained atan evening reception bythemembers then inParis oftheBritish Executive Commission, viz., Messrs. Dredge, Preece, and True- man Wood, given attheelegant pavilion oftheCommission in theExhibition grounds. The success oftheParis meeting waslargely due tothe kind exertions ofMr.John Aylmer, ourLocal Honorary Secretary. Epmsuron Exnierrion, 1890. The Council lately received adeputation ofgentlemen officially connected with theExhibition ofElectrical Engineering ‘andInventions tobeheld inEdinburgh next year; and, asthere appears tobeevery probability ofthere being much tointerest ourmembers, ithasbeen decided that ameeting ofthe Institu- tion shall beheld inEdinburgh, either shortly before orshortly after themeeting oftheBritish Association atLeeds. 1990.) REPORT OF THE COUNCIL. a ‘The Council have, attherequest oftheExecutive Committee oftheExhibition, given their views onthesubject ofthesystem ofpresenting awards, andonother points affecting thepractical advantages tobegained bysuch Exhibitions. ReporTiNe or Discussions. Inthebelief that theverbatim reporting oftheproceedings oftheInstitution’s Ordinary General Meetings wastending to encourage discursiveness onthepart ofmany ofthespeakers— which was neither desirable initself nor inaccordance with the wishes ofthemajority, asevidenced bytheready adoption of the ten minutes rule—the Council decided toprohibit the reporting ofthediscussions inthetechnical journals, and to furnish tothelatter anofficial abstract ofeach evening’s dis- ‘eussion, prepared byagentleman specially engaged for the purpose, This arrangement hasbeen loyally accepted and adopted by the Technical Press. ANNUAL CONVERSAZIONE, ‘The Conversazione ofthePresident, onthe24th May, took place inthe galleries ofthe Royal Institute ofPainters in Waters Colours, andwasvery largely attended, Annual, Drvwer. ‘The success which attended the Institution Dinner held on the4th November, when themembers were honoured bythe presence ofthePrime Minister andother distinguished guests, hasdecided the Council toconsider itasthe first annual Dinner oftheInstitution ofElectrical Engineers. Move orEtecrine Tae Couscr.. Asalready made known toyou bytheSecretary's circular of the 3rd instant, the result ofthe Council’s invitation toall Members and Associates, both athome and abroad, hasbeen such astolead them torevert totheoriginal system ofannually nominating forelection onthe Council only asufficient number ofMembers andAssociates tofilltheoccurring vacancies, ‘They 0 THE Liprany, [Dee 12a, trust that this decision, being that ofthe majority, will be +accepted without further demur. With theview ofrendering the rule which requires the annual retirement of three members of the Council mor effective inits intended purpose, viz, acomplete althoug: gradual change inthe composition ofthat body, ithas beer resolved, upon theinitiation ofProfessor George Forbes, tha twoatleast ofsuch retiring members each year shall bethose who have served during the longest time onthe Council. Professor Forbes and Dr. Fleming have accordingly retired. Mr.Stroh, having thesame object inview, had already pressed hisresignation ontheCouncil, which hasbeen accepted. ' Fivaxciat, Posrriox.‘ ‘The finances ofthe Institution continue inasound condition, theincrease inthenumber ofmembers more than compensating fortheincreased expenditure incurred inseveral respects bythe wider field ofitsoperations. THELIBRARY. | REPORT OF THE SECRETARY. beg toreport that theaccessions totheLibrary during the year, asreported from time totime inthe Journal ofthe Institution, number 81, byfarthe greater majority ofwhich have been presentations, forIhave again thepleasure ofrecording the liberality ofauthors and publishers inresponding tomy applications; infact, Ihave notyetinany onecase been met with arefusal. TheInstitution continues toreceive, bythekindness ofH.M. Commissioners ofPatents, the specifications ofall electrical patents, The number ofpatents applied forthisyear, uptothe 30th November, was19,295, ofwhich 1,109, ornearly 5:75 per cent., were electrical. ‘The number ofperiodicals and transactions received regularly bytheInstitution isslightly increased since lastyear. Alistof these isappended. _«< 1889.) TRANSACTIONS, PROCEEDINGS Ere, RECEIVED. 831 The number ofvisitors totheLibrary during theyear has been 238, ofwhom 34were non-members. . ¥.H.WEBB, 12th Decenber, 1889. Secretary andLibrarian. APPENDIX 10 SECRETARY'S REPORT. TRANSACTIONS, PROCEEDINGS, &c., RECEIVED BYTHE SOCIETY. ENGLISH. “Asiatic Society ofBengal, Journal and Proceedings. Cambridge Philosophical Society, Proceedings Greenwich Magnetical andMeteorological Observations Institute ofPatent Agents, Tranmctions Tostitation ofCivil Engineers, Proceedings. Tastitation ofMechanical Engineers, Proceedings. Iron andSteel Iastitute, Proceedings, Liverpool Engineering Society, Proceedings Physical Society, Proceedings. Royal Dublin Society, Transactions andProceedings Royal Engineer’ Fontitate, Proceedings. Royal Intitation, Proceedings. Royal Meteorological Society, Proceedings. ‘*Royal Society, Philosophical Transactions of. ‘Royal United Service Inttation, Proceedings. Society ofArts, Journal. Society ofChemical Industry, Journal. Society ofEngineers, Proceedings. University College Calendar. AMERICAN. ‘American Academy ofScience and Arts, Proceedings. ‘American Insitute ofElectrical Engincers, Transactions. Canadian Society ofCivil Engineers, Transactions. ‘Franklin Tomitute, Joaraal of. Joha Hopkins Univerity Circulars Library Balletin ofCornell University. ‘Ordnance Department oftheUnited Stater, Notes. Smithoonlan Institution Reports. FRENCH. Académie desSciences, Comptes Rendas Heblomadaires deaSéances de. Société Belge dElectricins, Balletin dole, Société Frangaise dePhysique, Séances deIa. Société desIngénieurs Civile, Mémoires Société Iaterationale desElectriciens, Bulletin dela Société Scientifique Industrielle deMarscille, Bulletin dela ©PresentedbyPrfemorDEHaghe,ACS(PastPreside) 2 LISTOFPERIODICALS RECEIVED. —(De.Im LIST OF PERIODICALS RECEIVED BY THE SOCIETY. ENGLISH. Electrical Engineer. Electrical Plant, Electrician, Engineer. Engineering. English Mechanic and World ofScience. Milustrated Official Journal, Patents. Indian Engineer. Industries, Tavention. ‘Mechanical Progress. Military Telegraph Balletin. Nature, Philosophical Magazine. ‘Telegraphic Journal and Electrical Review. ‘Telephone, The. AMERICAN. Electrical Engineer. Electrical Review. Electrical World. GasandWater Review andJournal ofElectric Lighting. Journal oftheTelegraph. Science. Scientific American. United States Patent Office, Official Gazette of. FRENCH. 7 Annales ‘Télégraphiques. LElectricité. Journal dePhysique. Journal Télégraphique. LaLumicre Electrique. Liétectricien, Revue Internationale deI'Electricité etdesesApplications GERMAN, “Annalen derPhysik und Chemie. Beiblitter zuden Annalen derPhywik and Chemie. ‘Centralblatt fir Elektrotechnik. Electrotechnischer Anzeiger. Electrotechnische Zeitachrift. ‘Verhandlungen desVereins zurBefirderung desGewerbseises Zeitschrift fiir Elektrotechnik. Zeitschrift fr Tnstramentkunde, 1889.) SUGGESTIONS BYMAJOR FLOOD PAGE. eas FRALIAM. Giornale det Genio Civile. TTelegraiate srarisn TngenieroyFerreteroExpafolyaAmericano. TheCuarwaN:Gentlemen,—I thinkyouwillallagreethat—« thepast year hasbeen asuccessful one—the first year ofthe “Institution ofElectrical Engineers.” Ithasbeen remarkably so inoneortworespects. Thepast year hasbeen thefirst inwhich webave hadanAnnual Dinner, andthesuccess ofthatDinner was mach that nodoubt itwill berepeated infuture years. ‘Then, again, theBoard ofTrade have consulted your Council with regard tothe establishment ofanelectrical laboratory, andother matters, andagood deal ofthetime oftheCouncil hasbeen devoted to the consideration ofthose matters. AAsregards papers, wehave hadapretty busy session, and have hadmany interesting discussions. Itherefore have great pleasure inmoving—That the ReportoftheCouncil,justnowread,be “received and adopted, and that itbeprinted intheJournal of “the Institution.” ‘The motion, having been duly seconded, was carried unanimously. Major S.FLoop Pace: Iamsure, Sir,that theReport ofthe Council which wehave just heard isvery satisfactory, but it seems tome that steps maybetakenwhichmaystillfurther improve the position ofthe Institution, Itisonly fair to members tostate that Ihad, inignorance ofthe rules ofthe Institution, given notice oftwo resolutions tobeproposed to- night. But Ihave been informed, inamost courteous manner, bytheCouncil, that thebusiness to-night isstrictly confined to the following, viz.:—“To receive and deliberate upon the “Report ofthe Council onthe state oftheInstitution, and to “elect officers fortheensuing year;” andithasbeen pointed out tomethat resolutions ofthekind ofwhich Ihadgiven notice should bebrought before aspecial general meeting convened by theCouncil, orcalled forby10members signing arequisition according totherules.IdonotwishtoputtheCouneiltothe et SUGGESTIONS BYMAJORFLOODPAGE.—(Dee18 trouble, expense, and inconvenience ofcalling aspecial meeting, neither doIwish totrouble 10members tosign arequisition calling upon theCouncil tocalla meeting; butIwish tosuggest totheCouncil that they should take into consideration whether thetime hasnotcome when itwould bevery greatly forthe advantage ofthe engineering profession ifthemain Institutions connected withengineering vouldbebroughttogether intoclose relations than exist now. Imean especially the Institution of Civil Engineers, that ofNaval Architects, the Mechanic Engineers, andtheIron andSteel Institute, with ourselves. Now Idonotthink there isasingle member ofthis Institutionbut willsaythat weshould gain bysuch anarrangement. Idonotfr‘amoment suggestthatthereshouldbeanyamalgamation orany absorption: itisclear that such anInstitution astheCivil Engineers could neither absorb usnoramalgamate with us,the conditions ofmembership are sodifferent; but ifwehad General Council onsomething like thelines oftheBritish Assoc- ation, each oftheInstitutions retaining itsautonomy, Ithisk that thereport oftheyear following such analliance would tell usthat theInstitution ofElectrical Engineers was more succes falthan ever. Itwould result, amongst other things, perhaps,in some such arrangement asnow exists atBurlington House, where, you areaware, the Royal Society, the Royal Academy, the Chemical Society, and other Societies arelodged under onetwo Ithink that ifwewere todraw closer together weshould find that allbranches ofthe engineering profession would guin, while weastheyoungest would certainly not gain theleat. Itherefore beg toask the Council whether they willule this into their consideration, with the view ofincreasingthe value and usefulness ofthis Institution even more than itbas been improved, asweareallglad tohear, inthelast twoyea. Itisonly natural itshould improve: theindustry andthescience with which weareconnected isthemost engrossing branch of science, andthecoming industry oftheage,anditisbutnatural that theInstitution should improve. Thaveanothersuggestion tomake,anditisthis:Thegreat development ofelectricity, although weallhopethatitwill| 1809.) SUGGESTIONS BY MAJOR FLOOD PAGE. 35 toon show itself inmotive power, andinother branches, still, at thispresent moment, thechief development isinthedirection of electric lighting; and Iamsure that aswewalk about London andseethelarge number ofhouses that have notices onthem that theelectric lighting isbeing done bythisorthat firm, we aregratified when weseethe name ofwell-known electrical engineers onthat board :wefeelthat theelectricity willprobably beconducted into that house safely, andthat theworkmanship willbesound andgood. But Ithink thetime hascome when itisforthe interest ofthe Institution toexercise some influence inthesubject ofwiring houses, specially when weread the names ofsome non-electrical firms that areputting theelectric light into houses. Isuppose weallagree that itwould bea great blow tothis Institution, and toelectricity ingeneral, ifa considerable number offires were totake place inconnection withelectricity ;andIsuppose wealsoallagreethatif electricity isinstalled inahouse safely and well, itdiminishes immensely thechance offire, butthat ifitisdone carelessly andonunsound principles, and byinexperienced persons, it Positively increases thechance offire. Weknow what hastaken inNew York, andthelarge amount ofexcitement that hasbeen aroused whenever any accident isascribed toelectricity. There wasaremark theother day inoneofthe New York papers bearing very forcibly onthis—viz., that the papers were so entirely taken upwith theoneaccident that hadhappened in connection with electricity, that nonewspaper hadasingle line togive tothe 35accidents that had, during thesame time, taken place inconsequence ofthe use ofgas. Ithink that itisimmensely forthe interest ofallwho areconnected with electricity that wetake some step towards ensuring that the Wiring ofhouses isdone onsound principles;anditseemstome that this Institution isthenatural place forustocome to,toask theCouncil toconsider whether they cannot introduce something intheshape ofanexamination. Weknow that lately examina- tions have been introduced intheSurveyors’ Institution, andin thePlumbers’ Company;andtheInstitutionofCivilEngineers itself has lately introduced asystem ofexamination before men 896 SUGGESTIONS BYMAJORFLOODPAGE. (Dein ‘can beelected members ofthat Institution. Ithink that inthe interests ofthe science and industry ofelectricity itisver important thattheseshouldbearranged; andIasktheComdl toconsider whether itwillnotgive agreat fillip tothewott ofthis Institution ifthey can commence some system o ‘examination, sothat men—wiremen and others—shall beableto saywhen they arecalled upon, “Ihave the certificate ofthe «Institution ofElectrical Engineers.” Iamquite sure tht ‘men who had such acertificate would beable togetmuch more work than those who had not. Iwould therefore ask the Council whether they will take into their consideration the two questions that Ihave ventured tobring before you, with theview ofimproving still further theposition ofthis Institi- tion—the one, whether they can place themselves into com munication with the Institution ofCivil Engineers, and the other Institutes Ihave named, toseeifwecannot draw the whole ofthe fiveInstitutions closer together, without inany case giving upour autonomy;and,also,whethertheycanse their way toinstitute something inthe shape ofexamination, with theview ofpreventing any prejudice being created agains electric lighting inconsequence offires taking place byhonses being badly wired. Thave very great pleasure insupporting theReport which hhasbeen brought before us: itismost encouraging. Myoaly object inaddressing youhere to-night isthat Ithink weought tomarch with thetimes; andsurely itisaverygood opportunity totake some such steps asIhave suggested. We areonthe verge ofanelectrical session ofParliament: since thegrest railway session of1846 there hasbeen noParliamentary session ‘occupied soentirely and monopolised byany one branch of industry, asthesession which isnow about tocommence willbe ‘occupied byelectric Bills. You areallaware that there are10 Jess than from 430 to450 different electric lighting schemes that will bebefore Parliament during next session; andIean conceive no moment inwhich the Institution ofElectrica Engineers candomore good tothe general public, and an more clearly show itsright tospeak fortheelectrical industry, 1880.) VOTESOFTHANKS, est than at the commencement of the “electrical session of “Parliament,” which isjust about tocommence. The ballot-boxes were withdrawn. TheCxamMan: Gentlemen,—The Council,Iamsure,will beexceedingly glad togive very careful consideration tothe important proposals which Major Flood Page hasmade. They villtake hissuggestion anddosoatonce, andthus avoid the necessity ofcalling aspecial general meeting oftheInstitution. Professor Si:vanus P.THOMPSON: Weare, indebted, Sir,to theInstitution ofCivil Engineers forthecourtesy which they extend tous,year byyear, inallowing ustheuseoftheir rooms andpremises forourmeetings, and wecertainly cannot doless than give them avote ofthanks forthefacilities which they give tous, They saveus,nodoubt, what would beaverymuch larger expense ifwehadtoengage rooms ofourown, and anestablish- ment ascommodious asthat inwhich wehold ourmeetings. I havetherefore very much pleasure inmoving— That thecordial “thanks oftheInstitution bepresented tothePresident, Council, “and Members oftheInstitution ofCivil Engineers forthe “continuance oftheir kindness and liberality inpermitting this “Institution tohold itsgeneral meetings intheir lecture hall.” Themotion,havingbeenseconded byMr.W.LaxtCanPexrer, wascarried byacclamation. ‘Mr. ALEXANDER SigMeNs: Ihave very much pleasure in moving—“That the thanks ofthe Institution are due tothe “members who soably represent usabroad asLocal Honorary “Secretaries and Treasurers, fortheir continued kind services in “that capacity.” You allknow that there arequite anumber of gentlemen invarious parts oftheworld whoundertake thevery invidious task ofgetting thesubscriptions offoreign members together andsendingthemonheretoourSecretary. Youwill also have observed that atnearly every meeting anumber of foreign membersareproposed,andtheyhavemostlybeeninduced tosend intheir names byourLocal Honorary Secretaries. Inmaking thismotion itisperhaps invidious toparticularise, butinthisyearIthinkitisnaturalthatIshouldmentionthe nameofMr.John Aylmer specially, because weareallvery 838 VOTESOFTHANKS. (Dee188 ‘grateful forthevery kind wayinwhich hemanaged theaffairs at themeeting inParis. . ‘The motion, seconded byMr. SyoNey EVERSHED, wascarrie! ‘unanimously. Sir James N,Dovotsss: Ihave very great pleasure in proposing—“That thethanksoftheInstitution areduetoMt “Edward Graves, Past-President, forhiskind andvaluable service “asHonorary Treasurer.” ThenameofMr.Gravesissowell known that itrequires nowords ofmine topresent thismotion beforeyouforyourheartyacceptance; butyouwillallbesory tolearn that Mr.Graves isprevented byindisposition from being here to-night. The motion, having been seconded byProfessor Jon Pextt, washeartily carried. Mr. C.E.SracNoernt: Inall Institutions and Societies the financial position isanimportant one, andIam sure you willall agree with meinmoving—“ That ourbest thanks areduetoMr. «J, Wagstaff Blundell and Mr.Fred. C.Danvers fortheir kind «services asHonorary Auditors during thecurrent year.” Themotion, seconded byProfessor S.P.THOMPSON, waseared unanimously. Sir Davip Satowoxs: We all know that with civilisstio wedonot gettheadvantages that one would expect. Ont necessary consequence ofcivilisation isthe employment of solicitors, followed bytheir bills. Inourcase weoccupy apostion half-way between themost civilised and thesavage state, is having asolicitor, butnosolicitors’ bills; Ithink, therefore, that the least wecan doistoreturn our sincere thanks toour Honorary Solicitors fortheir services. Asarule, these gentlemen havebs! nothing todo,but onarecent occasion, when wechanged our name,theygavetheirservicesmostwillingly, andforthatalooeourthanksaredue,aswellasforthedutieswhichwillnodoubt devolve upon them inthefuture, Itherefore begtomore— “That the thanks ofthe Institution are due toMessrs. Wilso, “Bristows, &Carpmael fortheir kind services asHonora; Solicitors.” ‘Themotion,seconded byMr.Gissent Karr,washeartilycared 189] ELECTRICAT ENGINEERING IN AMERICA. 899 ‘The Cuatnuan: The next business before usistocontinue the discussion onMr.Addenbrooke's paper on“Electrical Engineer- “ing inAmerica.” Mr.W,Lant Canrenter: Iwasprevented byanengagement Misese, outofLondon from being present tohear Mr. Addenbrooke's paper and thediscussion upon itlast meeting, butIhave had an opportunity ofreading itsince; andasitfelltomylottomake a veryhurried journey toVancouver's Island and back during August andSeptember last, Iwasable toseesome ofthepoints thatMr.Addenbrooke hasalluded toinhispaper, and hence I should like totake theopportunity ofsaying how heartily I concur inallthat hesaid with regard tothetemporary and tmesthetic character ofthefittings forelectric lighting inthe Western towns, both intheUnited States and Canada, Ihad not theopportunity ofgoing over anyofthecentral stations atall, andmyjourney being exceedingly hurried, Iwas only able mainly toobserve what onecanseeinpassing rapidly through these cities, and just toconverse with various people onthe general position ofthings. One ortwopoints struck mewhich appear, perhaps, tohave escaped Mr.Addenbrooke's notice, or possibly may have escaped myattention when reading hispaper, orinthe interpretation ofit.One wasthevery extensive use made ofwater-power inthetowns inthemountainous districts, especially onthePacific slope oftheRocky Mountains; andit certainly wasvery remarkable toseehow inmushroom towns three orfiveyears old,every oneofthem waslighted electrically, inmostcasesbythewater-power obtainedfromthestreamsandfallsfromtheRockyMountains andtheirtributary ranges,andusednot onlyforarelighting, butforelectric traction work. There were some towns that Isawwhere they had notyetproperly formed thestreets, buthad levelled theplace simply, andonthat they hadput down rough rails and carried out their locomotion entirely byelectricity, using overhead conductors, thepower being almost entirely derived from water intheneighbourhood. That wasvery notably the case atthe town ofSpokane Falls, in Washington Territory, which was burnt down last July: the whole place was lighted byarelighting inafewhours after the VoL. XVI: 58 . 0 ELECTRICAL ENGINEERING INAMERICA. (DieI8h, Apener, cMtastrophe. Thesame thing applies toatown between Lake Huron and Lake Superior—Sault Ste. Marie—you seeexactly the same thing;thepowertherebeingtakenfromtherapidsranning between thetwo lakes, thetotal vertical fallbeing about 50fet. Butthepoint that specially forced itself upon mynotice waste very small amount ofincandescence -lamp lighting which m: ‘employed inanyofthese Western towns ;andasfarasIood make out from conversation with many people, the one grea reason foritwasthat thething was sobadly managed that, asx rule, thelifeofthelamp wasexceedingly short, and infactthey ‘considered that 300hours wasquite aphenomenal lifeforalamp. ‘The reason, tofarasIcouldascertain, appeared tobethatthere ‘was not that skilled supervision with regard both tothelaying down ofthe installations and their management that wear accustomed to;andsoitwasthatinmanytowns—I amspeaking ofthese extreme Western towns, both intheUnited States andin Canada, towhich the same remark applies—glow lamps were scarcely employed, andwhere they hadbeen employed they were frequently givenup,thereasonbeing,asfarasIcouldmakeout, thewant ofanyproper arrangements whatever inthebeginning andalsointherunning oftheinstallation afterwards. With regard totheuseofwater-power,—it isprobably withia the knowledge ofmost present that three large towns ia Canada—Montreal, Ottawa, and Quebec—are lighted bywater power,inthecaseofQuebecfromtheMontmorency Falls,eight miles off; andIheard agreat deal about theprobability of electric traction being derived from thesame source. Inconver sation with people atOttawa Iwastold, although Ididnotseeit, that thealternate currents which were used sovery much there forlighting, whichwerederivedfromtheChaudidre Falls,wert ‘also used forpower inagood many instances bytheTesla motor; Tdonotstatethisofmyownauthority, becauseIhadnottimete seeformyself, butIwastoldthat alarge number ofTesla motors were employed commercially inOttawa with these alternating currents. The only other point onwhich Iwould trouble themeeting inthelargenumberofwiresusedforlong-distance telephooe 1886) DISCUSSION. on working which were running outofNew York—my attention wasms. specially called tothem—running toBuffalo andalarge number ~ ofothertowns. Long-distance telephony seemsnowtobea very easy thing, and largely used. There hasbeen agreat reduction intheoverhead wires in New York, Chicago, andother large towns during thelastfour or fiveyears, compared totheWestern towns ofwhich Ihave been speaking, and there seems tobevery much more underground work done, Iam here contrasting what Isaw inthe streets of large cities in1884 and 1889. Mr.G.S.Ram: Ihave lived ayear and ahalf intheStates, Me.Ram and can therefore endorse agood many ofthe things that Mr. Addenbrooke hassaid. Idonotaltogether agree with hisremarks about theglare ofthearelighting when done onthepoles inthe ordinary way. InNew York there isno arclighting onthetower system;thereweretwotowersofgreatheightinUnionSquare about twoyears ago, butthey were abandoned infavour oftheare lights onordinary poles;anditmaybethatonegetsusedtoit, batIcertainlyverymuchpreferthe“blazinglight”inNewYork tothecomparative “utter darkness” inLondon. ‘They have inAmerica many ways ofdoing things which appearstrangetotheBritisher. Forinstance,Ihaveseeninone ofthelargest hotels inNew York, lighted byitsown Edison plant, thelarge Edison dynamo being runwith large blocks of iceonthebearings to,keep them cool. This, theengineer told me,wastheway healways ran themachine. Itwasprobably notdone atthedirection oftheEdison Company, butonthesole responsibility oftheengineer, whohaddeveloped atheorythat1ce wascheaper than oil. Mr.Addenbrooke's remarks ontheoverhead work might be thought byanyone who hasnotseen itforhimself tobepossibly somewhat overdrawn and exaggerated; butMr. Addenbrooke,I think, wasstrictly speaking within thelimits when hesaidthat everything hasatemporary and half-done look about itwhich ix very offensive toEnglish eyes. Inorder that those who have not seen thework may beable tomore fully appreciate Mr.Adden- \brooke’s description, Ihave brought here @photograph ofabit ory ELECTRICAL ENGINEERING INAMERICA. (Dre. ur.nam ofBroadway, New York, which Itook inthe beginning ofthis year, When Itook thephotograph itwas myintention toget picture ofthebuilding, notthewires; buttheresult issuch that onemay sayitisapicture ofwires, with abuilding intheback- ground. Only thethicker and nearer wires canbedistinguishe! inthephotograph, butinatransparency Ihave made, bytheuse ofamagnifying glasshundreds ofwirescanbedistinguished, while with amore powerful glass thedetails ofthewayin whichtheyarefixedtothepolescanbeseen,Thereisonepat with 17cross-bars, each ofwhich carries 10green gla: insulators, besides oneonthetopofthepole, andevery insulator appears tocarry awire. Anumber ofwires seem torunupand downandroundthispole,Idonotseeforwhatpurpose, andsnumber ofendsarecoiledupandlefthanging. Therearealso anumber ofthickish insulated wires, which areprobably are light wires covered with “underwriters’ insulation,” whichar fastened tothepole itself, andnottotheinsulators atall; ands number ofthese areseen torisetoa much higher pole, with it cross-bars atright angles tothe first, about 10feet off,and which they areapparently tied below theeross-bar. Onthispole oneofthe cross-bars istwisted round, andappears tobewaiting foragust ofwind toblow itdown, Between these twopoles there isathird pole, from which wires radiate inalldirections. With theaidofaglass theinsulation ofmany ofthewires may beseen tobehanging inshreds. This spot wasnotselected for thepurpose ofshowing the wires, but Ithink itgives afair average sample ofelectrical engineering out ofdoors on Broadway, New York. [The photograph and transparency wer handed round forezamination.} ‘Mr.Spagno- Mr.C.E.SpagNoLetti: The necessities forinsulation inthe fee twocountries, England andAmerica, aresowidely different that that may account forthedifference intheappearance ofthe styleofworkineachtoaverygreatextent. Itwill,perhaps, beremembered bymany here that when Mr.Cromwell Varley went toAmerica, many years ago,hewasvery much struck with thecareless wayinwhich theinsulation was attended tothere, andtherewasratherananimatedcorrespondence inthepaper:| 1868.) DISCUSSION. as onthatsubject. Here,inEngland, wehave,fromourownMr,eae. experience,aknowledgeastohowtheinsulationofthewires varies. On the Great Western Railway, inthe South Wales district, one week ourwires will test exceedingly high—some 8,000,000 or10,000,000—and perhaps the next week, with continuous showery ordrizzling rain, wemay get insulation results onthesame wires below 200,000 ohms. Some years ago ‘Mr. Edison was over here with anew instrument—a modification ofBain’s chemical printing telegraph—and hewas trying some experiments atthe Post Office. The instrument wasavery sensitive one, and the induction inthe underground wires through London was such astovery much interfere with it— infact, torender working impossible. Upon the request of Mr. Culley, who wasthen the Engineer-in-Chief ofthePost Office, wearranged togive Mr. Edison aroom atour Faling Station, into which one ofthe Post Office wires from Liverpool wasled. Mr. Edison wai inthat room making experiments forabout afortnight, and hetold methat inNew York, without theslightest difficulty, hecould get2,000 words ‘minute out ofthis machine, but the most hecould dounder thebest conditions that hecould getover here wasonly atthe tte of 614 words aminute. This will show the great difference ofthe two atmospheres; and, asMr. Varley then saidafter hisexperience there, hethought that inAmerica they mnight work avery long line with very inferior insulation,asthe atmosphere there wastotally different from that ofthis country. Mr.W.H,Paresce, F.RS.: Ishould just liketoaddarather srrecs. interesting facttowhat Mr.Spagnoletti hasjust communicated about Mr. Edison. When hecame over here with the electro- magnetic shunt business, hethought hewas going todogreat things, and before hecame tothe Post Office hemade some experiments oncircuits belonging totheEastern Telegraph Com- Pany, who gave him theuseofacable that waslying inthe tanks ofthe Telegraph Construction Company atGreenwich, Mr,Edison wanted tostudy theeffect ofhisautomatic system on ‘ables, andthey gave himtheuseofthiscable forawhole night. Hetoldmethishimself, andhesaid, “Now thefirstthing on ELECTRICAL ENGINEERING INAMERICA. (Dec.ith, ur.Pree“thatIwantedtoknowwasthelengthofadot.”Thelength ofthecable hewas experimenting with was some 800 or1,000 miles—I forget theexact length—and hesaid, “When Ipats “dot into that cable, what doyou think was the length ofi “whenitcameoutattheotherend?” Isaid,“Idonotknor; “perhaps sixinches.” Hesaid, “Six inches! itwas 28feet:| “thought that dot was never going toend.” That isavey interesting anecdote, because itshowed that atthat momentfer ofusreallyknewwhattheeffectofself-induction inacule! cable was. Onthis point itisalsointeresting toknow—andI think Ihave mentioned itinthis room before—that the fr Atlantic cable, when coiled inthe tanks atGreenwich, gave speed ofabout. 14words per minute; and SirWilliam Thoms, Varley, and Fleeming Jenkin brought out:their curb key,which wasgoingtoincreasetherateofworkingoftheAtlanticcablefrom 14words perminute toabout 4or6words perminute. It wasin1866.Thecablewaslaid;thekeywasbroughtdoato Valentia. ‘Totheir surprise they found that, with theondioary single key, they were able toget15words aminute through the Atlantic cable; the reason being simply that they bad experimented with acoiled cable intanks inthefirst instancy, andthenwithacablelaidstraightalongthebottomofthe Atlantic, when theeffects ofself-induction disappeared, andthere remained only the effect ofelectrostatic induction toretanl working. Togoback toanother question referred tobyMr.Spagnolett, Ihave previously pointed outhere that thegreat reason why there issuch adifference between the insulation inthe United States andtheinsulation inthiscountry isthatintheUnited States theprevailing winds from theocean arefrom acol region toawarm one: theshores ofthenorthern straits of America arewashed bytheArctic current, and theprevaiixg winds arecold and drywinds. Inthis country itistherevere: wehave ourshores washed bytheGulf Stream, and thepre- vailing winds aretheS.W. winds, which come upon ourshor colder than theocean, laden with moisture, and theresult istht wehave dampness; and thus there aredifficulties thatrender insulation with usextremely difficult. 1888) DISCUSSION. as Ido notfeelatallable tonight tosayanything about theax.reese poper ofMr. Addenbrooke, nor totake part inthis discussion, Thave twice visited the United States, and oneach occasion I have given theresult ofmyvisit there tothis Institution. But Thave never come froin theStates without feeling benefited ;and Thave nohesitation whatever insaying this—that ifanymarked improvement has taken place inthe management orinthe arrangement oftheworking ofthetelegraphs ofthis country through myexertions,Ibelievethatithasbeengreatlydueto thecontact that Ihave made with the energy and the go-a- headedness ofourfriends ontheother sideofthewater. They have difficulties tosurmount that wedonot meet with here; ‘wehave difficulties tosurmount that they donotknow ofthere ; andthetendency isthat wearerather induced tocriticise their doings from our point ofview, while they criticise our doings from their point ofview. Ihave never yetmet anAmerican electrician who has come over tothis side ofthe water who has notadmired something that wedo,and gone back feeling a wiser man; and Iamquite sure that there isnot asingle member ofthis Institution who would gototheother side of theAtlantic who would notcome back again andsaythat he hadlearned alot,andfeltafter itamuch better man. Mr. R,E,Cromrrow:ThepointsInoticeasspeciallyinterestingms arethedevice which theauthor mentions ashaving beenadopted“ inAmerica forpreveriting thevibration ofmoving machinery being communicated toadjoining premises. This appears to consist inplacing thedynamos onawooden platform supported onwooden posts, and theplatform notbeing connected atthe dynamo leveltothewallsofthebuilding.‘Theexperience that Thave had inEngland hasshown methat such anarrangement insome cases may actually increase instead ofdiminishing the vibration, with theresult that theincreased vibration communi- cated down through the ground tothe adjoining walls is actually greater than ifthemoving machinery had been placed onmore rigid foundations. Inote that attheEdison station in Chicago aconsiderable number ofresistances and indicators are ‘wed forthevarious feciler circuits. This appears tomeasif | 6 ELECTRICAL ENGINEERING IN AMERICA, (Dee. 1b.| Ycanion, theywerethereignorant oflaterpractice, asinthebest-manage! modern systems which employ feeders ithas been found quite possible, bycareful design ofthenetwork andfeeder system,to avoid theuseofsuch variable resistances. ‘The hatchet-shapel switch described byMr. Addenbrooke isvery largely used is England and onthe Continent, aswell asinAmerica, Heis quite right insaying that itisone ofthe cheapest andbest switches that have been hitherto designed. Inote alsotht Mr.Addenbrooke re-states thefact which Ihave always insstel upon, viz., that uptothe present date alternating-current machines arenotasamatter ofpractice run inparallel inthe American central stations, This has been over and over contradicted inthis Institution, and itisinteresting tose ‘Mr.Addenbrooke confirm myviews. ‘The vertical boiler described inthepaper isinteresting toa *mechanical engineer from thefact that apparently, inthiseas, itgives notrouble, The useofsuch boilers hasbeen frequenty attempted inthis country, but there aresomany disadvantage: anddangers connected with itthat ithasinmost cases been gives up. Theupper part ofthetubes which pass through thesteam space above thesurface ofthewater isliable toget90hotthat they consequently collapse and give great trouble. Ian also state from personal experience that itisvery difficult to make theupper ends ofthese tubes tight intheupper tube plate. Iregret much that wehave little chance ofavailing ourselves oftheinteresting information Mr.Addenbrooke gives usofthe variousmethods ofoverhead connections forelectric tramways. have nodoubt that thismethod ofsupplying theelectrical energy totramways isthebest andcheapest ofall; butwehave toover~ come theesthetic prejudice ofthemunicipal authorities, andI amsure that thiswillbesostrong that wehave hardly achance ofusingtheoverheadsystem. ‘The whole ofMr.Addenbrooke's paper isvery interesting,ss itshows what very great advantages American electrical engineer- inghasenjoyed inthelicense that hasbeen given them tocary out pioneer work intheroughest possible manner. Idonot think itright toseverely criticise such work, butconsider thstit 1880] DISCUSSION. a hasbeen agreat service inpopularising thedemand forMe electricity. Mr.A.Reckexzauy: IquiteagreewithMr.Preece’sremarks, Mr.A,which were tothe effect that those ofuswho visit the United States cannot return without having learnt something, andvice verea forAmerican electrical engineers visiting thiscountry. Iamvery sorry tonotice that Mr. Addenbrooke usessome hard words with reference toelectrical engineering practice in America, Heseems greatly troubled about posts, and says that “when posts areused they areofrough pine, never painted, “often outofstraight andwarped andbent.” Ihave seenanumber ofthese posts: they areunsightly certainly, whatever shape they may be; but there aresome posts inthe States, especially in larger cities, which arestraight, smooth, and painted, andthere arealso posts which arequite ornamental, made ofcastorwrought iron, Mr.Addenbrooke's paperseemstobeacriticism uponthebad work done inAmerica, and hesays solittle about thegood work that isdone there ;therefore onehardly knows where tocommence discussing hiscomments, andwhere toleave off. Hesayslittle or nothing about central stations and isolated plants inwhich secondary batteries are utilised. There hasbeen aconsiderable development inthis direction, and some mention oughttohave beenmade, ifonly superficially, inapaper ofsuch pretentious title. There arenumerous central stations andisolated plants which areworked with storage batteries, as,forinstance, at Detroit,feeding1,600lamps;JerseyCity,1,000lamps;Hones-dale,500lamps;Haverford College,500lamps;Cherryfield, 500lamps; Watsontown, 500 lamps; Allentown, 500 lamps; the Metropolitan Telegraph andTelephone Company, New York, 1,100 lamps; Mill's Buildings, New York, 1,000 lamps; and the American Express Company, 250 lamps. These areonlyafew which mybrother, who ishere from San Francisco, mentioned to me atrandom. There isone very important subject—a great and thriving industry inthe United States—and that iselectric railway engineering. Concerning this Mr. Addenbrooke says nothing as ELECTRICAL ENGINEERING INAMERICA. [Dee12 wr.A._,,, beyond afewgeneral statistics which havebeenpublished inthe newspapers, andoneisinclined tothink that hehas eitherntseenanyelectricrailways, orifhehasseensomehehasnottakentheslightest notice ofthem, Such matters would have been far more interesting tousthan alistoffaults which hiscritical eye happenedtodiscover. M5Bomon Mr. Frepenick Reckenzavn: Ithink mybrother baswell covered theground onwhich Ishould raise adiscussion, Itis true, electrical engineering practice inAmerica differs somewhat from European practice. While overhead wiring isdone tosome extent ina“slipshod” way, touseMr,Addenbrooke's expression, there is,onthe other hand, aconsiderable amount ofskill dit played inthis class ofwork bythe larger companies, such asthe Brush, Thomson-Houston, United States, and others, who treat itasaspecial department. Overhead wiring, properly done, combines economy with efficiency and simplicity. Astothe construction ofarelamps, Imight saythat wecertainly donot seesuch ornamental lamps inAmerica asweseehere and onthe Continent. The American are lampsareplainandsimple,bat answer thepurpose;whatiswantedisnotsomuchornamentsas illumination. Itiscertain that cheap construction reduces the cost oftheelectric light and facilitates itsintroduction. InNew York, forinstance, thearelight prices aresolowthat many small storekeepers couldaffordtouseit,bothinsideandout.‘Thishasaddedtotheillumination ofthestreets,whichotherwise wasin many places defective. Inregard toincandescent lighting, Mr. Lant Carpenter gave asthe reason why itisnotextensively introduced io theWest, that such plants have been badly managed. This is true, perhaps, tosome extent. Onthe other hand, Westen people—who, asarule, jump atnewthings, andwant thebiggest —often have intheir towns anarclighting plant before theyhave ‘agasworks; they find thearclight notonly thebiggest, butthe most economical fortheir purposes. Incandescent: lighting is rather looked upon asarefinement tobeadded Inter on. Itcosts fagreat deal more inconstruction aswell asforoperation, per ‘unit oflight, than arelighting; and inmany places where in 1689.) DISCUSSION. oo Europe theywould notusearclights, suchasrestaurants ormi.¥.offices,theyuseitthere,anddonotmindthelittleflickering and theglare which itbrings with it. ‘[email protected],ADDENBROOKE, inreply, said:IthinkthatassuchMr,Addeo- ‘large number ofpoints have been raised bythevarious speakers, Thad better rather take them together than attempt toreply to every one separately. Referring, firstofall,towhatMr.F.Reckenzaun saidabout theuseofarelights, Imust saythat Iwasstruck with theway inwhich they were employed inlittle saloons and places 12or 14feetsquare,andnotmuchmorethan10feet.high;anarelamp often being suspended inthe middle, when,ofcourse,the light. seems really brighter than daylight. Iwould saythat Idomention inmypaper thegood con- struction ofthe Thomson-Houston Company—at any rate, in ‘New York. Ispecially mention that, and Ialso may state that they rantheir lighting circuits during thetime that theother companies suspended their operations inNew York owing tothe extent ofthe difficulties which had arisen with the Town Council. ‘There isnodoubt that thetramway companies arebringing in theuseofabetter class ofposts. First ofall,they appear to have started with aneven poorer class ofposts than were being ‘used forother things. Most oftheir posts were arranged one on each sideofthestreet, with across-wire carrying theconductor inthemiddle; butthey arenowlargely giving upthat class of structure forneat wrought-iron posts. These posts arearranged inthecentre ofthe two tracks, with anarm oneach side, ifthe trackisadoubleone;or,ifthetrackissingle,thepostsare atthesideoftheroad, with one arm projecting. Itiseasyto putlamps onthese posts, which areuseful toprevent vehicles running against them, The electro-motive force is,asarule, 500or546volts.Fivelampsareputinseriesoneveryother,or third, pole, bridging across themains: thelamps burn very well, and,ofcourse, doforlighting the streets. Imay saythat the tame method ofusing five lamps inseries isused forlighting the cars. With respect towhat Mr.Crompton said about theconstruc- 850 ELECTRICAL ENGINEERING INAMERICA. [Dec.124 Az,Adien tionoftheflooring ofcentral stations, which Idescribed as separated from thebuilding, andonly steadied from thesides: I didnotintend toimply inmypaper that thiswastoreducethe vibration ofthe machinery, but itwas toprevent thevibration ofthe machinery affecting the building. The upright boiler that Ideseribed asused bythe Thomson-Houston Compaos may besubject tothedefects which Mr.Crompton mentioned;allthatIcansayis,thattherewerethreeoftheminuseinthe ‘Thomson-Houston station inNew York, and that Mr.Foster,the superintendent there, told methat they proposed increasingthe number. ‘Aremark wasmade byMr.Lant Carpenter astolong-distance telephony. Ihad thepleasure ofspeaking from New Yorkto Boston, through onetranslator coil, and Imust saythat the talking wasexcellent. ‘The distance is250miles, Iwasinthe central office atNew York, and Ispoke toone oftheir officials who wasinanoffice situated atsome distance from theExchange inBoston. Atthe last meeting SirWilliam Thomson alluded tothe rough-and-ready methods which theAmericans employ. ‘There isnodoubt that they have enjoyed, and doenjoy, enormous facilities from this way ofgoing towork; and although Ihave strongly criticised it,yetIdonotwish tocriticise itentirely in aspirit ofcondemnation, ‘Turning now tothedifference between theAmericans and ourselves, weareaccustomed totalk ofEnglish engineers and American engineers. Iwasrather surprised when Iwasthere at thenumber ofengineers Imet inAmerica who were born in England, They might have been outthere twenty years orless butreally theproportion Imet wasquite large. From allI observed Idonotthink that there isreally very much difference between American engineers andEnglishengineers. Itisthe conditions ofthecountry that aredifferent, and theengineers over there have adapted themselves toit. When Iwasin Australia Isaw very much the same class ofthing there. Englishmen outinAustralia adapt themselves tothecountry. Electric lighting hasnot developed inAustralia tothesame | 1889.) DISCUSSION. ss extent that ithashere; butstill,outthere, inengineering Mr.,adten matters they doadapt themselves tothe necessities ofthe country inthesame way that they have done inAmerica. ‘Then there isanother point Inoticed, not exactly incon- nection with electric lighting. Itisthat inAmerica theTown Councils, Mayors, and the authorities ofthetowns seem tobe stronger than they arehere, Ofcourse itisademocratic com- munity; butwhen they put#man inaplace, heseems tohave a great deal ofauthority inthat places and Ithink that the American corporations ofthetowns perhaps have allowed all this overhead work togoahead inasort ofgood-humoured spirit toletpeople have achance, with the feeling that they could put their foot down, and that they would puttheir foot down, and stop itwhenever they thought fittodoso. You see frequent references andcases inthepapers oftheMayors speak- ingonallsorts ofsubjects inanauthoritative manner which Ido uotthink would beused inthiscountry. Iwasatthelastmeetingaskedsomequestionsaboutmotors, 1did notseemuch ofmotors onarelight circuits—or, rather, run from arelight machines—because, ofcourse, there would besome difficulty inrunning motors inthedayonarecircuits without, thelamps burning. Ithink they aremostly series motors, with arrangements forcutting outmore orless ofthefield-magnet, coils, and with aswitch forshort-cireuiting the motor when not required tobeused. Nodoubt theclimate inAmerica isvery different from what itishere. Totakean instance, Ithink that therainfall inthe region about Salt Lake City isnotmuch above 9or10inchesa year, andagreat part ofitoccurs inthewinter when there is mow, Even when the rain does come, itcomes quickly;they donotgetthelong, drizzling rains that weget here: ‘Thequestion ofalternating-current motors liasbeen alluded to,IwastoldthattheWestinghouse Company haveexpended ‘nenormous sum onexperiments inthat direction, and afair numberareinuse,butIwasunabletogetanyoriginalstatisticsthatIcould bring forward astotheir efficiency orthekind of Satisfaction they were giving. (852 ELECTRICAL ENGINEERING INAMERICA. [DeeInk, weteten With regard tometers itwasmuch thesame thing, Ims toldthat theWestinghouse Company—and they areadvertising it—have got8,000metersinuse,TheAmericans arenotlonginsayingthattheyhavearesultwhentheyhavegotit,and, nodoubt, experimenting onthis scale with anapparatus which inmany ways fulfils the requisite conditions, the natunl progress and improvements which inevitably follow when any- thing once gets into extended usewill enable good working meter tobecompleted. Idonotknow that there isanything more that Icanusefully add, except thatImightperhapssayhowmuchthehospitality which was extended tothe American electrical engineers orer here during lastsummer wasappreciated. Ifound that everyone inAmericawasinaverygoodtemperwitheverybody here,and disposed tobeexceedingly courteous, while the opinion was universally expressed ofhow much kindness they hadmetfrom their English brethren, Reinam, TheCuatnway: Tam sureyouwillalljoiningivingahearts mmmvote ofthanks toMr. Addenbrooke forhispaper. Itisa interesting paper initself, and notonly that, butithasgives risetoaninteresting discussion, ‘Ahearty vote ofthanks was unanimously accorded toMr. Addenbrooke forhispaper. niggy, Professor 8.P,Tnowrson: May Ibeallowed todrawthe attention ofthemeeting totheblue glass insulators which have been placed onthetable? Ibrought them from America five ‘years ago, and Ihave now brought them down here toillustmte Mr,Addenbrooke's paper. Ihappened tocatch sight thisweek ofabook newly published byaChicago engineer ofgreat ability, which isnowbeing circulated inthiscountry asaguide tothose who have todeal with dynamo machines. Inthat book theble glass insulator with thewooden pinisspecially recommended3 theproperthingforrunningarelarapcircuits! BsAddeMr,G@,L,ADDENBROOKE: Iwouldmention thatIforgottoalludetothefactthatIhavelaidonthetablespecimens ofthe class ofcable which isnow very largely being putinto thecom duits inNew York. The insulation isvery hard, andismore like | ebonite than indin-rubber. | 1889.] ELECTION OF OFFICERS AND COUNCIL FOR 1800. 638 The Scrutineers handed intheir report ofthe result, ofthe ballot for Council and Officers forthe year 1890, which the SECRETARY announced tobeasfollows:-— President: J.Hopxinsox, M.A. D.Sc. F.RS. Vice-President: Wri. Crookes, F.R.S. ALEXANDER StEMENS, Assoc. M.ProfessorW,E,AYRTON,rns.|Inst.CE. R.E,Cromrroy, M.Inst. C-E. Ordinary Members ofCouncil : Sir James ANDERSON. Gissert Karr, Assoc.M.Inst.C.E. SirAtsert J.L.Carpet, K.C.LE. SirHenryMance,C.LE.,M.Inst. Major Pair Carvew, RE. CE. W.LantCanrenrer, B.A.,B.Sc.|Professor JouxPenny,ME, SirJames Dovotass, RS. |D.Sc, ERS.Captain SirDovetas Gatrox,|SirDavinSaomos, Bart,M.A,K.C.B.,D.C.L., LL.D.,F.RS.|Professor SitvanusP.THOMPSON,Colonel R,Ravxsronp Jacnsox. |B.A, D.Sc. F.RAS. Acsociate Members ofCouncil : Avausrus Evex. i Francis H.NaLbER. Stpvey Suarp. Honorary Auditors : ‘Faepenice C.Danvers, | Avacstus Strow, Honorary Treasurer : Epwarp Graves, Past-President, . Honorary Solicitors : ‘Messrs. Witsow, Bristows, &Carr@att, 1,Copthall Buildings, E.C, ‘The Paesipent proposed avote ofthanks totheScrutineers fortheir labour intheexamination ofthe ballot lists, which was carried unanimously, st ELECTION OFASSOCIATE. ‘Dee.1h, ‘The Caiman: Gentlemen,—I thank you very heartily forthe great honour you have done meinelecting measyour President forthecomingyear.Itakeitasagreathonourtobeelected topreside over such anInstitution asthis—an Institution which hhasalready attained togreat celebrity and importance inthe industries with which itisconnected, and which,Itrust,inthe future will attain tostill greater importance, and still greater utility. Itake ittobeanhonour, too, tosucceed inthischairso distinguished @man asSir William Thomson, the most dis- tinguished electrician living. Icanonly say,gentlemen, thatI will domyvery best tomaintain thehonour oftheposition to which youhave been sovery kind astoelect me. Aballot took place, atwhich the following candidate ws elected:— Associate: Septimus Felix Beevor. The proceedings then terminated. ws . . THE LIBRARY. ACCESSIONS TO TE LIBRARY FROM JULY 170 DECEMMEN 3, 166)(lentemartthus(2hattonpurchaseOfthsenolparasorrete sechang hr dora araen she bc pe Feared Astronomer Royal (WIM. Cre, Reyrt tieDearofWy Buck and Hickman. Ilusirated andPricedCatalogue ofPatentandImproved Pat)Charton [M.J.)ApercuGénéraldesdispositions etinstallations deExposition tkiaSFpeteagesieeeae Baga in| ‘puneCapininEanes Transformation—Mcnsigue-Chaat ColsoneiritreChimieEleuiee: Maguetiome: Sra.SqPisne1689tireentl byGeeges Cartlaiieg) ‘DeTunselmann. [Vide Tonzelmaun.)*Dumont (6.1,Leblanc(M.].and DelaBédoyare (.]Dictiounairethéoriqueiaaso Farmer, {Vide Sexy&Farmer be ayn ‘asd ——FieciealPaper, 1874-189, S10, London, 1880. Conanofthe ‘olenSeeRESconusTheoryoftheGalvanicCel12pp.(Pro,Ply‘seyWali ate oonhe PleinofBlctroesinWatefefromAit.Typ(Phi Meg te fate ‘aie Op CD ngeh oy oon the Charunine’ Carer” andSuraes ofInenndewenee Lampe2 COEDS tage} es bos the Us-of Danie Cel as Suntan ofHlecto-mative Foe Sp it Hage As) ‘688 Aug.) ‘1883 ProbiemtontheDstiatonofeteCorentinNetworkofConte von" getet bythe Sel afSas” Boype COMA lo, ai ‘aks oothe Necemity fraNational Stndartsing Labortory forElectret eaten "Ss peElerneloteSeayaTernhEos aaaBlceinns CLRWWNovsany YY Teoh Reais FeSEine 2onealof rat ‘i VOL. XVII. 59 856 ACCESSIONS TOTHELIBRARY. J.A}andGimingham (C.1L]OnomeIndranenttorthe FeeriotafMsuootve WoreandElctealFonteif {fengteSeyefTearaphBiersedBrinn,VoLXe oesis Gray {Andrew}, Absloto Mearrements inElectriityandMogae.od Fay(onSanasetop Londo,189 {ProtntedbyMessrsMacmillan&Co,(Pebtisber}] arg:-Graban).Connonra,to,hpPay,(Pron Sreyof incor) POccn ‘on VURIV,PapeeFW,sey Chatham,1888Cantor YectaresonHeataneserthaySteam,(DeliveredbeforettetyofhrsMay19,30andShans).S00”GEppEdn,Hickman.{VideBockanMiktan)Hopkinson (John), Magnetic and ether Physical Propertieofon.te eeeks,POMFensVoLCENpedon188 ‘Houston [FalwinJ.]ADictionaryofElectricalWords,Terms,andPhra Sreo6 ‘SeeFarkEan{Hsnted byFt btectrcan Printing ant Vablahing Corse) Hotchinson (Sl, C)_ Bret SketchoftheMeteorologyoftheTomi attsinasian."fo.Sp Pater) Indian Government Telegraphs. Administration Report oftheIndoLeerTapDementor sareesharToppCataInstitution ofCivilEngineers. MinutesofProceedings, withotherSelected ‘anclAtatracted Papers. Wol,XCVIL. 1ee0, PartIIL.Bro.an ince“axdow —— Migute ofProceoings, with other Selected and. Absit PuenValReMiWooSeppPies “Loanie—— Bret Subjects. Vole LIX. toXCVIIL, Sewions 7800 188 oe Spe “nda Jamieson [Andrew]. Onthe Derigning ofContingons-Current DynanMachines. "Bro.Upp.Plates:(Prone.oftheFnaivationofEnginemephateic8Blands ‘lego i —"Blectra Fingering. Stemi aunged QuetonsonTeegaps ‘Tank Clogor —— Blementary Manasl ofMagnestom andBleticity. Part -—Mageetin one Tonle "Tie Meare anJanice.)Kennelly(A.%)andWilkingon(Hl,D.)_PracticalNotesforEletit saci. VolekVare UutoandSiopleMeasrig laraent feS08ppsCreblcedaTe) "roa(Prete byHD,Witkinn Moar) Kleismner,(VideZetzsche.] Ledge [Oliver3}MolarViewofElec.(NatureSein)Sra,4 * 1aan18 Maycock(WV. Pere}. Praccl Becta!NotasandDetwiiny,frthe FiAincering tenant Precteal Monger wih esegaltons boOhersedinBecame Warsle Tatton ofBecta EagetersundtePawlyBisOfcWh Binge,‘Rey.1505 in188 Meyer (Neruard). Systeme «Appareil ‘Telégraphique Automatique Usi-(eran, Soins Appr Tepamigve AntoneUs Ton Men Muaro (oo)andJamieson (Anirn), AYukattokofHil Bewalt eee eat asic Suge ty "nae Accession 10 THE Linear. sr National ElectricLight,Association. Proceedings atitsNinthConvention. aa feLighAppearing aotGove. aepipne ‘NewZealand. Re;onthePostOfficeandTele; yentforthe fod Bronte rhDeere fos(teen ye 2a Paris Exhibiti 1889, Pt hic Group ofElectrical Jury, ParisExision, 188Phomempbic Grey i,as(eae y3Ayan We Patent Agents, Institute of, Transactions. Vol. VIJ.—Session 1888-89. ‘bro. 237pp. London, 1889 (Exchange,) RayalObseratory, Greenwich, seofChoenauersonThalbr ‘PurchasebytheBoardofAdmiralty, attheRoyal‘Observatory, Green-Tae hap iesGases at"SFR Oe Ts [Presented bytheAstronomer Royal)—saeDkWancoTeforPecbeerfh eelDehaogeefeeeaeata nore pe (PresentedbytheAstronomerRoyal.} ‘Sax! t.Railway SafetyAppliances. 4thEdition. 56 cby&Parmer, aySafetyAppli PP.1089Ainge) aengMga”tt,Conninogtl neTe gtMogMpafren,Cga He ofig Copan al ere ode ‘Smithsonian Institution. Annual Report ofthe Board ofRegents ofthegonianInitio,fovealTeperofeasofRagefhe Enunyasninrsfew lyase ibePeees des Sse oanvange.] foal [Athen 5) Kneisl Tasman fPore teMaing Cpe,(atin8, rama eeiersoftheNe“Association ofCollieryManagers,Wigan,Se,2s,18|foals,Prangase eaten atinSooTranedePnigne,, ColinteMecoiespulse Poe on am on sinh b1 (Postersang sae (ewes), Begins otPorn on Kncromi sad MaguLm, TygaofFee ps——_and Tait[P.a‘Treatise onNatural Philosophy. NewEdition.‘Vol.1.PartL.,608pp.PartII.,527pp.8v0. Cambridge, 1888 unsslnana (6, ie) Bectiy loMon Uefa Str eonssean(0,tla 7ts ‘University Colleges London. Calendar. SessionMDCCCLXXXIX- BET allagy Hendon, pee ([Exchange.} +Uyponbern 73 teryofcheTaatimae,com.soy,naeoe +WeenWilaaBerveryeoejTansememetmae RelyandMagaedta,VOUL~Hagutien endEleste-Dyoame. yo, ae Paes ‘Wilkinson (H.D.) [Vide Kennelly andWilkinson.) Zatasche [Prof. Dr.K.Ed.] Die Elektrische Kalenderahr desProf.Kleismer, en CD BdDeBeaneRabebatieee Saati a9 se LIST OF ARTICLES ELECTRICITY AND MAGNETISM ‘Appearing insome ofthe principal Technical Journals during the Monthsof NOVEMBER and DECEMBER, 1889. L—BATTERIES AND ACCUMULATORS. =Krovctixoti—Polarisation ofMetals byImmersion inaLiquid, byMotionis the Liquid, and byWithdrawal from the Liquid—Jour. dePhys, vol8 p-519, 1989, J.Gax—Note ontheTheoryandHistoryofBatteries—Jour,dePhys,vol& p.527, 1889, E.Warnono—Theory ofthe Voltaic Cell and ofPolarisation.—Ann, vol$8, S21, 1489, F,Srmerrz—Theory ofAccumulators—Ann., vol.$8,p.34,1889.¥,Sraziwrz—Silver- MercuryCell.—Ann, vol.88,p.514,1889. 1L_D¥NAMOS AND MOTORS. A.Witz—Rerersal ofthePoles ofaSeries Dynamo.—Jour, dePhys. vol.§ .581, 1889, G.Riemany—Detals ofDynamo Construction.—Lam, Elyvol.94,pp.167, 89. Axon.—Dulait's Dynamo,—Lum, EL, vol.8,p.278, 1889. P,Tanx—RelationofE.M.F,toSpeed.—Zum. Elyvol.38,pp.374,428,1889 F,Lannoque—Electro-magoetic InductioninDynamos.—Lam.Et,vol.4,p.40.1889, Awox.—Use ofAlternate CurrentsasContinuousCurrentswithoutaCommntater. Lan, Ely vole34,poOA2, 1889, ‘J,Srerax—Thermo-magnetic Motors.—Ann, vol. 88,p.427, 1889, IL—ELECTRO-CHEMISTRY AND ELECTRO-METALLURGY. P.H. Lepenoen—Electro-Metallurgy ofAluminium.—Zum, £4, vol.$4,p.138, 1889. P.IL,Lapsnoen—Electro-MetallurgyofIron,—Lum,l.,vol.Hp.965,1889. A.Mixet—Metallargy andElectro-Metallargy.—Lum, El, vol$4,p.$01, 188% H,Rrax—Tho SpiralVoltameter Zam,El.,vol.4,p.$90,1889.A.Miser—Electro-Chemistry—Lum.El,vol.84,p»965,1889, A.Mixat—Sizo ofthe Electrodes and Choice ofMaterial toobtain the Be Results inElectrolysis.—Lam. El,vol.34,p.612, 1889. ARTICLES RELATING TO ELECTRICITY, Fre. 859 ©.Ricmand—Electric Welding.—Lam, EL,vol.34,p.575,1889.C.Frowa—Maximam Polarisation ofPlatinum Plates inSulphuric Acid.— Anne, vol. 88,p-262, 1889. 0,Lauacasx—Motion oftheTonsiuFusedandSolidSilverTotide—Ann., vol.38, .296, 1889, IV.-RLECTRIC LIGHT. D.Larscunxorr—Modification ofKrus's Photometer—Jour. dePhys.. vol.8, .548, 1889, G,Fraeanis—Phase-Difference, Lag,andWasteinTranaformers.—Jour, dePhytyvol.8,p.648,1889,—Henznznc—Absorption ofLightbyWindowGlass.—Laue, El.vol.&,p.178, 1889, F,Urrrxnonx—Lumninosity ofanAreLamp working with Alternate Currents— Laon,El,vol.84,p.179,1889. L,Wetssexonvcn—Relative Value ofLarge and Small Are Lamps forLighting Open Spaces.—Lum. El, vol. 84,pp.187, 294, 1889. €.Jacqotx—The Characteristic ofTransformert.—Lum, EI., vol.34,p.201,1889. G.Rienanp—DetailsofConstructionofGlowLamps—Lum. El,vol.$4,pp.907, 878, 1889. L.Pasquatse—Regulating Mechanism ofAreLamps.—Lum. £1, vol.4,p.812, 1889,H.puRorue—ListofCentralStationsinFrance—Lum. El,vol.8,p.820,1889. C.Jacqvx—The Central Station ofthe Halles Centrales.—Lum, El, vol. 34, p.S51, 1889, G.Ricuaxp—Are Lamps.—Lum. El, vol. 4,p.406, 1889.E.Dropoxxsi—The CentralStationsattheParisExhibition —Zum,El,vol,34,Pp.501,562,1889. =Vanrour—Electric LightingofTrains.—Zum. El,vol.4,p.519,1889. A.Brnxerzix—Constant CurrentDistribution —£l.Zrit.,vol.10,p.506,1889,Asox,—Lammer-Brodhun't Photometer.—E1, Zeit, vol,10,p.644,1889, V.-ELECTRIC POWER. PHLLepesomn—Electric Railwaysand‘Tramways.—Lam, £1.vol.84,pp.418,473, 1889, VI—MAGNETISM AND ELECTRO.MAGNETISM, W.pxForvietiz—A Turing Magnetic Field produced byTwo Induction Colle. =C.R,, vol.109, p.782, 1889. —Gour—Magnotic Potential Energy, and MeasurementofCoeficienteofMag- netisation.—C. R.,vol.109,p.985, 1889. G.,Fa—EffectofMagnetismonResistance.—Jour. dePlys.,vol.8,p.545,1889, F.Macurxt—Magnetisation ofIron.—Jour. dePhys, vol.8,p-852, 1889, 4.Sravax—MethodsofProducingPowerfulFields.—Ann., vol.88,p.440,1889, ‘K.Kaute—Maguetic LinesofForce.—El, Zeit,vol.10,pp.466,627,1889. 00 ARTICLES RELATING TOELECTRICITY, Ere. ‘VIl—_MEASURING INSTRUMENTS AND MEASUREMENTS. ‘BTumztrats and A,Poutock—The Clark Cell asaSource ofSmall Constant Carrent—Phil. Mag. vol. 28,p.858, 1889. R.Tametrati—Application oftheClark Cell totheConstructionofaStandart Galvanometer—Phil, May., vol.28,p.416, 1889.H.8,Carmcanr—Improved ClarkCellwithLowTemperatare Coefficient —PhilMag., vol,8,p.420, 1889. R.TumzirAti—MeasarementsofHighSpecliicResistances—Phil. Mag.,ol8, p.452, 1889. RTunzurats and A.Porvocx—Meararement oftheResistance ofImperfects Pariied Salphur—Phil. Mag., vol.28,p.469, 1889. J.J. Boavaxi—Alterations inthe Resistance ofHyponitric Acid due toChange ofTemperature. —C. R.,vol. 109,p.804,1889. A.Tenquex—Resistance ofthe Biffel Tower and itaEarths.—C. R, vol 10 p.839, 1880, J.Beururvor—Conductivity ofAspartic Acid—C. R,vol.108, p.864, 1889, G.P,Guiwatpi—Resistance ofAmalgume ofPotassium and Sodium.—Jor. & Phys, vol8,p.46, 1889, W.E.Ararox—Practical UnitofInduction—Lam,El,vol.4,p-107,1889.A.Pataz—Fischinger's Ammeters andVoltmeters.—Lam. £1,vo.34,p.108,18. E,Rooxn—Interruptor forRuhmkorff Coils.—Lum, Ei. vol. 34,p.880, 1889. G,Ricuanp—Shallenberger's Ammeter.—Lum, BL, vol.$4,p.426, 1889.YF,M,Ricuanp— Application ofRichard'sRegistering SystemtovariousElectal‘Monruring Instruments.—Bull.Sor.Iut.,vol.6,p.885,1889. PF,Urrexnonx—New Method for the Measurement ofSmall Reslstances.—Beil vol. 18,p.$85, 1889, K.Frongmn and St.Lurpecx—Alloys forResistance Coils—Beibl, vol1% p.957, 1889, L,Wanen—A New Galvanometer.—Beibl, vol. 18,p-962, 1889. ‘Axox.—Angatrom's Apparatus forMeasuring theStrengthofMagneticFielis— El.Zeit, vol.10,p.548, 1889.‘P,Meren—Mensurements ofMagneticFieldsbymeansofTransversal Presa.‘andtheConstantofPermeability forManganese Steel—EU. Zeit,vol.101682, 1889, ‘VIIL_RAILWAY APPLIANCES. M.Cossuaxx—Appliances attheParisExhibition, —Lum.El,vol.$4,pp.18h 67, 457, 655, 1889. Axox.—Delfion's Automatic Siguals-—Lam. El., vol. $4,p.181, 1889. —Komtronsr—Zetsche's Applisncer—Lum. El, vol. 84,p.222, 1889.E.Zerscur—Seoemann's andSchilling’s Appliances. —Lam.El.,vol.34,p46,1889, R,Pevact—Bachmann’s Signals—EU. Zeit, vol.10,p.518,1889, ARTICLES RELATING TO ELECTRICITY, Ere. est IX.-STATIC AND ATMOSPHERIC ELECTRICITY. ©.Tomtmsox—Lightning andGunpowder Magazines —Phil.Mag.,vol.28,.368,1889,D,Larscmxorr—Photography ofDiseharges—Jour, dePhys. vol. &p.531, 1889, A.Barreut1—Earth Current,—Jour.dePhys,vol.8,p.610,1889.©.C—Lightaing Discharges andLightning Conductors —Lum.£/.,vol.34,p.192,1889,L,Patarami—Experiments ontheElectricity oftheSoil—Lum. £., vol. 34, 266, 1889.A.Herownniiex—Pasage ofElectricity throughGases—Anu.,vol.88,p.534,1889. K,NarrenensPamage ofElectricity throughGases-—Ann. vol.38,p.653.1889.E,Saumasty andL.peLaRive—Repetition ofMerta'sExperiments. —Beibl.,vol.18,p.971, 1889, LL,Wanen—Atmospherie Electricity EY.Zeit,vol.10,pp.321,571,1889. XTELEGRAPHY AND TELEPHONY. 1.peRorux—The World's Telegraph Cablex—Lum. El,vol.84,p-109, 1882. —Masstx—EliminationofInductionouTwoParallelTelephoneWires.—Lim. El, vol. 84,p.190, 1889, W. peFoxvistix—The Australian ‘Trans-Continental Telegraph—Lum Ely, vol.34,p.143,1889, A.Paraz—Duilding Telephone Lines.—Lam, Fl, vol.34,p.185, 1880. C.Jacquix—Acceleration ofTelegraph Transmission bymeansofaCondenser.— Lam. Fl. vol34,p.173. 1889, A.Pataz—Decamp's Telephonie Souniler.—Luw, £7, vol. 84.p.212, 1869.A.Pavaz—Mever's Universal Automatic Telegraph Apparatus —Lue,£1,vol.34,p.279, 1889. =peNaxsourr—Manufactory ofIndia-Rubber atGrenelle—Lum El, vol 34, 1p.279, 1889, P,H,Lepxnosa—D'Argy’s Microphone.—Zam.El.vole34,p.819,1889. E,Zerscue—Duplex Hughes Apparatus with Two Relays,—um. £1, vol. 34, p.418,1880,A.peSexues—Control Apparatus for‘Telephone Exchanges. —Lam.Zl,vol.34,p.415,1899,J.Revvar—Telegraphic Communications between England and Australia.— Lam, El, vol. 34,p.464, 188%. ©.Guawrexet—Acceleration ofSpeed onUnderground Cubles—AL Zeit, vol, 10,p.353,1889. =Mirxoxrotus—Mallett's Antomatic ‘Type-Printing ‘Telegeaph.—£1.Zeit. vol. 10,p.391, 1882. ws ARTICLES RELATING TOELECTRICITY, Bre. xL—THEORY. ©,Cuworsox—Simplitication ofKirehhof's Second Axiom.—Jour. dePlys, vol. 8,p.536, 1889, A,Sreraxorr—Apparstas forshowing theEffects ofaDielectric Medium onthe IntensityofMagneticForces—Jour. dlePie,vol.8,p-543,1889. G.Fenzanix—Electro-dynamie Rotatious produced hyAlternate Cirrent.— Tour dePhys, vol. p.608, 1889. A.Riam1—Electrical Phenomena produced byRadiatious —Jour dePhys, ol8, p.G13; Beibl, vol.18,p.976, 1889, ©.Decase—Lag inPhysical Phenomena.—Lum. E7., vol.$4,pp.128,16, 1849, HHunrz—Ientity ofFlectriity and Light—Lam, El, vol, 34,p.240, 1882. V. Droxax—ActionofSelf-Induction inElectro-maguetie Current,Tateraption Lam, El, vol. 3,p.384, 1889.. H,RowLaxp—Modern Views onElectric Currents.—Zum. El. vol. 84,p3: 1869, —Cannan—Theories ofElectricAction.—Lum,£1.,vol.3,p.489,1882.E,Remaxowsren—Stoletow's Actino-clectrie Experimenta.—Lum. Ei, vl.3. .5I6, 1880, =Reoxacir—Condactivity ofSaline Solutions—Lum, £1. vol.2.p.527,188. “W,Grese—Conductivty ofGas Flames—Aun., vol.88p.408. 1889. J,Eater and H,Gurrei—Discharge ofNegatively Electrified Bodies byLight— Ana, vol. 88,p.560, 1849. ©Laxcun—Infiuence ofElectricity onthe Flow ofLiquids from Orities— Beibl, vol. 18,j.935, 1889. XI2—VARIOUS APPLIANCES. G,Rscuan—Some Mechanical Appliances—Lum, El, vol. 34p.213, 188 —Pauter—Applications ofElectricity toMining—Lum. EY,vol.84,p.2% 1380, La,Wenttzn—Electric Telescope.—Lrm,Bl,vol.3)pe334.188(i,Ricnano—Electrie OreSeparators.—Ln. By vol.34,p.428. 1882,K,Dinevoxssi—Mine Exploders.—Jaon, EU,vol.36,p.431,1889.FyZarscuss—Holler's Electric Level Indientor.—Toom, Blk, vole, p.481.188 N,Sexcusew—Applicatious ofElectricity toAgricultare.—Lum. B.v0h3 P.558, 1889, sss APPENDIX. COMMUNICATION. Tax Unrvensrry, Guatcow, February 11th, 1890, With regard tothe remarks upon submarine telegraphy byMr.Preece, atthemeeting ofthe 12th December, 1889,* which have just come under mynotice, thefollowing may beof interest tomembers:— The first Atlantic cable was never coiled intanks atGreenwich. Itwas made, half atBirkenhead, and half atGreenwich. The two parts were joined inone circuit forthe first time onboard the “Niagara” and “Agamemnon,” atQueenstown, atthe end of July, 1857, and Ithen gotsignals through itatsomething less than one word perminute with thespecial instruments which had been prepared byMr.Whitehouse. When thecable waslaid, in 1858, weobtained about three words per minute byhand- signalling, with mymirror galvanometer asreceiver, during the three weeks ofsuccessful working ofthecable, Inthisconnection Imay quote thefollowing extract from aletter ofmine to the Athenawm, dated October 24th, 1856, and published November Ist, 1856:—“Amodeofoperatingsoastoclearthe “wire rapidly ofresidual electricity, which Ihave worked out “from theory, andaplan fortelegraphic receiving instruments to “take themost fulladvantage ofit,which hasrecently occurred “to me, allow menow tofeel confident ofthepossibility of “sonding adistinct letter every 34seconds bysuch acable,” or about 34words perminute, +Journal,p.B44,above.+Collected Mathsmatical andPhysicalPaper”artic'elxxvi.,vo.ip.101.VoL. XVII 60 864 COMMUNICATION. Itwas the second (1865), and notthefirst (1857), Atlantic cable that was coiled intanks atGreenwich, and itwas is connection with this cable that Mr.Varley and Mr. Jenkin scted along with me. Varley and Jenkin and Iwere perfectly aware ofelectro magnetic induction,® and wewere more pleased than surprised when wefound that, after having promised tothecompany eight words perminute, weactually obtained, byhand-signalling, with mirror-galvanometer asreceiver, 15words per minute through each ofthe two cables (the second and third Atlantic cables), when the laying ofboth across the Atlantic was completed in 1866. WILLIAM THOMSON. ‘*See“Remarks ontheDischarge ofsCoiledElectric Cable,” BrithAmoctation Report, 1859, part 2;ofmy“Collected Mathematical and Phyl ‘“Papers,” vol. ii,article Ixxx. 865 INDEX TOVOL. XVIII. 1889. AccomiostotheLibrary.(SeeLibrary.) men‘Accident(Fatal)fromLightainginIna,Reportofa,byP-V.Takex697‘Adams,Mr.A.J.8,RemarksonMr.A.E.Kenselly’s Paperoncertain‘Phenomena connectedwithImperfectEarthinTelegraphCircuite ..158—— Remarks onProfesor Oliver Lodge's PaperonLightning,Lightning Conductor, andLightniog Protectory ena BOT Adams, ProfeaoeW.Grylls,RemarksonProfesorOliverLodge'sPaperon Lightaing,LightningConductor,andLightning Protectors.=468 —RenarksonMr.W.M.Mordey'sPaperon Alieraate-Carent Working(43Addenbrooke, Mr. G.L., Remarks onMr. K.L.Murray's Paperonthe ‘LightingoftheCentennialTaternational Exhibition,Melbourne... 762 ——FlectrialBngloeeringinAmerion neon TTD ——ReplytoRA@sarksontheabovePaperneieeo BOTSBABAddress(Iasogura!) ofthePresident(SirWilliam‘Thomsoa)— Ether,Electricity, andPonderable Matterweneame eam‘Alvernate-Current Circuits,LaboratoryNoteson,byProfemorsW.E.Ayrton findJobaPerryneaeeee ee 8H —Working,Oo,byW.8Mondeyoeeee ae8B Supplementary RemarksprevioustotheDiscumion wo68L‘RemarksoatheshorePaperby— Dr.JohnHopkinson(commonicatad) .s.svoeeaeABProfettorW.GryllsAdams(communicated) svoneoeAB Dred.A.Flemingneneeeae 80 Mr.Gisbert Kappovene Professor .P.'Thompeon oeeons BTWB Ayrtona Bee eeBLGeorgeForbeneeeae eaeaeOOB: SirWilliamThomson eee OT Mr. J.S.RAWOHD oe ee eee ee »G.Zipersowakt (communicated) eo AE7Gc.Fricker(communicated) ene OTB2WeM.Mordey(50reply)aeeat OTB America,ElectricalEngineeringia,byG.L.Addenbrooke 2.wo.710 ‘Andersen, Mr. F,V,,Remarks onProfeaor G.Forbu's Paper oosome Electric Lightiog Central StationsinBurope,Men.seae221,2T4 ‘Armstrong,Lieut-Col.R.Y.,RemarksonProfemorOliverLodge'sPaperon Lightning, LightalngCondactors andLightalngProtectors x.se605 866 INDEX. Ayrton, ProfemorW.E.,onSecondingVotoofThankstothePresidentfor —— RemarksonProfessorJamieson'sPaperontheInsalationResistance ofElectric Light Installations ae ve G5,77102) 117, 118 —— Remarks onMr. A.E.Kenneliy's Paper oncertain Phenomens connected with Imperfect Earth inTelegraph Cirenits wu. we vw 158 —— Remarks on ProfessorG.Forber'sPaperonsomeElectricLighting Central Stations inEurope,Be.aseoneoseeaeBL ——nd Professor John Perry, Laboratory Notes onAlternate-Carrent ——ReplytoRemarksontheabovePaperbyhimselfandProfessorJoho —— Remarks onMr.W.Hl.Preece's PaperontheDisturbances arising from the Use of“Earth”forElectricLightingParposes... ...$20,322—— Remarks onMr. John B.Verity's Paper onUnderground Condaita fand Electrical Conductors ease ose vne ‘870, 875, 378. —Remarkson Mr.W.M.Mordey'sPaperonAlternate-Carrent Working661-++-— Remarks onMr.K.L.Murray's Paper ontheLighting oftheCea- fennial International Exhibition, Melbourne =... ge 149,7H Balanco-SheetforYear1888 eee ena BOM—— Presentation and Adoption of n,n ne eae BH Bate, Mr. D.C.,Remarks oaMr. A.E. Kennelly's Paper oncertain Phe- ‘nomena connected with Imperfect Earth inTelegraph Cireaits ... 135, 157 -—— Remarks onProfessor G.Forbes's Paper onsome Electric Lighting Central StationsiuEurope,&e,seeoteceeneeanes8B Beckingule, Mr. E.W., Communication inreference toProfessor Jamiesoa's Paper ontheInsulation Resistance ofElectric Light Installations... 12?Bright,SirCharles,Bustof,Presentation byLadyBrightof... 823Bright, Mr. Charley, Commanication inreference toProfessor Jamiewn's ‘Paper ontheInsulation Resistance ofElectric Light Installations... 123 Backnill, Lieut.-Col. J.T.,Remarks onProfesor Oliver Lodge's Paper oa Lightning, Lightulng Conductors, and Lightning Protectors... 527 ‘Cappel, SirAlbert, RemarksonProfessorJamieson'sPaperontheInsolation ResistanceofElectricLightInstallations... wea weB Cardew, Major, Remarks onProfesor Oliver Lodge's Paper onLightning, Lightning Conductor and Lightning Protectors wa. =. au ms 50 —— Remarks onSirWilliam Thomson's Paper onthe Secarity against DisturbancesofShipe’Compasses byElectrLightingAppliances... 578—— RemarksonMr.W.M.Mordey'sPaperonAlternate-Current Working648 Carpenter, Mr. W,Lant, Romarks onProfewor Jamieson's Paper onthe Tasulation Resistance ofElectric Light Installations... 68,116 —— Remarks on ProfessorG.Forbes'sPaperonsomeElectricLighting Central Stations inEurope, G6. oe see ane weewea BB INDEX. esr ron Carpenter, Mr.W.Lant, Remarks onMr,Addenbrooke's Paper onElee-‘rlealKogineering iuAmerica. eee wean 8D Central Suions (Some Electric Lighting)inEurope,andtheirLemons,by CertainPhenomenaconnectedwithImperfectEarthinTelegraphCirca,On, ‘Remarks ontheabove Paper by— Mr W.P. Grave eae ae 1B SCORPitman oe 8 ATS Adam on18 SyDespointes ace ee oe eee USTfyGeT.Fleetwood oe BTfGW. SCrwley cee 188 (Chamen, Mr.W.A.,RemarksonProfessorJamiesou's PaperontheTasulationResistanceofElectricLightTostallationsweeomaneveTL Clock(Electric)fortheLibrary,Donationof,byMr,Preece ewCompasses (Ships), OutheSecurity against Disturbances of,byElectricLightingAppliances bySirWilliamThomsonws eaeOTConductors (Electrical), Underground Conduit and, byJoba B.Verity... 838Condatte(Underground) andElectricalConductor,byJobaB.Verity838Connell,AnnualReportofthewemeaes aes ——tadOcersfor1890,Electionof.eee ee88Cousens, Mr,R.L.,Remarks onMr.K.L.Murray's Paper ontheLightingoftheCentennialInternational Exhibition, Melboarme wo.au.weeTBCcrawiey, Mr. C.W.S.,RemarksonMr.A.E.KennellysPaperoncertain ‘PheuomensconneciodwithImperfectEarthinTelegraph Circats..158 Creak, Staf-Commander, RL., Remarks ooSirWilliam Thomson's Paper 02theSecarityaginatDistarbunces ofShiga’CompassesbyElectricLight- IngAppliances eee ae BTCrompton,Mf,RE,RemarksonProfesorJamiewoa'sPaperoatheFamaation Resistance ofElectricLightIostallations ay same——RemarksonProemorG.Forbes'sPaperonsomeElectrLighting —— RemarksonMr,W.H.Preeee'sPaperontheDisturbancesarising fromtheUseof“Earth”forElectricLightingParpowesww.820 ——RemarksonMr.GeorgeHookham’sElectricityMeter... 2.69——- Remarks onMr. Addenbrooke's PaperonElectricalEngineering in Crookes,ProfeuorW.,Announcement oftheDeathofMr.WarrenDelaRue$85Carrent Meter. (See Eletricity Meter) os rapRx. DeaBae,Mr.Warren,Announcement oftheDeathofvvwxow388 Doepointes,Mr.,RemarksonMr.A.E.Kennelly'sPaperoncartinPhe- ‘omensconnectedwithImperfectHarthinTelegraph Circuits...=.157 ‘Disturbances arising from theUse of“Earth” forElectric Lighting Parpors, On the,byW.H.Preecee eee aneee | ‘RemarksontheabovePaperby— ProfeworAyrtonneeee eB,BBMe.RE,Crompton nee ee BD|iyWB, Bead we ee ee BE | —— ofShip’ Compaase byElectric Lighting Appliances,OntheSecurity | gaint,bySirWiliamThomson nsweve nemo OTDonorstoLibrary.(SeeLibrary.)Donovan, Mr. H.C., Remasks onProfesor Jamieson's Paper ontheInsala-tionResistanceofBletricLightTawallations ve 8,18—— Mamarks oaMr. A.E.Kennelly’ Paper onceriain Phenomena ‘connected withImperfect Earth inTelegraph Circaita ww wm MB | —— RenartonMr.G.1,Addenbrooke's PaperonElectricalEngineering Drake, Mr. Bornard, RemarksonProfessorJamieson’sPaperootheTaselation Teaisance ofElectric Light Ynsallations swe ae ae oe arth(Imperfect) inTelegraphCireats,OacertainPhenomena connectedvith,byALE.Renselly neon ee “Earth” (Use of), On the Disturbances arising from the, for Electric Lighting ParpoosbyWH.ProvesoneveawoemSI Eddison, Mr. R.W., Remarks onProfowor G.Forber's Paper onsomeElectricLightingCentralStationsinEurope,86.wrmeaoeam254 Election ofNew Members 49,77,120, 160, 197, 240, 81,328, 84, 499, 40,‘65,630,685,80,854 ElectrClockfortheLibrary,Donationof,byMr.Preeseuve me | —Light Installations, TheInsulation Resistance of,byProfeaor A. | —— Lighting Appliances, OntheSecurity against DisturbancesofShips Compassby,bySieWilliThomsonaeneon ——— Central Stations inEurope, and their Lemons, Some, byProfeworGeorgeForbes eeonsen ateeae |———Porposes, OntheDisturbances arisingfromtheUsof |“Barthfor,byWePreecewieaoe eee ElectricalConductors, Underground Condaits and, byJohn B.Verity. 8 — EngineeringinAmerica,byG.1.Addenbrooke a) ‘RemarksoatheabovePaperby—ProfeworGeorgeForbes eee Me HLC.Dosovan. : ca aoe Profewor8.P.Thompson rer INDEX. eo oy ‘Remarks onPaperbyMr.G.L.Addeubrooke (continued)— Mr. W.Lant Carpenter oi caer nee oe oes80oeGSRam anne aneneeneeneeA beWeHRPreeceveeene ee ABbeRE,Crompton nese ABfeAReckenaan, ccna nen ST ty-Addenbrooke(imreply)vaso veBT,AD Profesor8.P.Thompeon—Exhibition ofBlueGlassInsulatorsfrom Electricity (Ether), and Ponderable Matter (Inangural Address ofthe President, SirWilliamThomson). nemee mee——Meter,Exhibition andDeseription ofhis,byGeorgeHookbam 688 ‘Discusionthereon—‘SirWilliamThoms... sssssssssse_61,692,698 Me.RE,Crompton neon neat aeeONbySTOxe ae ce cents 8 Electro-Motor, NotesonaStatic, byCharles Zipernowski ww. ave ane 701 Erskine, Mr. R,S.,RemarksonMr.JobnB.Verity'sPaperonUnderground Conduite and Electrical Conductors ae se ose nese ne BOB Eason, Mr. W.B.,RemarksonProfessorJamieson's PaperontheInsalation Resistance ofElectricLightInstallations ws.wieseveeaee—— RemarksonthePaperbyProfessorsAyrtonandPerry,Laborstory ‘Notes onAlternate-Current Circuits. wv sue ve wees BIS —— Remarks onMr.Preece's Paper ontheDisturbances arising from theUseof“Earth”forElectricLightingPurposes... ssousneB22—— Remaris onMr. K.L. Murray's Paper onthe LightingoftheCen- tennial International Exhibition, Melbourne we me one ne 788 Ether,Electricity, andPonderable Matter(InauguralAddremofthePresident, SirWilliam Thomwoo) se ne sete tee Commanication inreference tothe above Paper, by Sir William Everahed, Mr. Sydney, Remarks onProfessor Jamieson's Paper ontheInsulation ResistanceofElectricLightInstallations. sue aus 6B. —— Remarks onProfessor Oliver Lodge's PaperonLightning,Lightning Conductors, andLightning Protectors aes seeneeveeBID Farquharson, Mr. J.,Remarks onMr.John B.Verity's Paper onUnder-‘groundConduitsandElectrical Conductors 1.wisomeaeeoweTO.—— Remarks onProfesor Oliver Lodge's PaperonLightning,Lightning Conductor, andLightningProtectors seema neon 00 870 INDEX. i race: PalAccidenttomLightingiaTada,ReportotabyP.Veake... tat|Fiugrald, Profewor G.FRemarks onProfawor Olver Lodge's Peperos| Tighniog,LightningConductorandLightingProtectorsee43 FleetwoodMt.CT,emarksonProfemorJaniePaperonthelel tionResistanceofElectricLightInstallations. ees SL “Remarks onMr.A.E.KennellyPaperoncertainPhenomen connecedwithImperfectEarthinTelegraphCreal)seeT FlemingDr.J.A.,ewarkaoaProtemorO.ForbesPapeonsomeHecticTightingCentalSutoninBarop,Ae.esereewoI——“TemisonMe.W.3MorieyPaperonAlteraat-CarrentWorking62 Fortes,PrfaorGeorge,onMoringVolsofThanktothePresidestfor — "Some Electr Lighting Cental Satins iBarope and thir Lemons 161—nplytoomarteontheabovePageteee ewoe268= nenartsoeSirWillanToomon'PaperontheSecutyagua; DistarbancesofShip’CompanesbyElectricLightingAppliances 8| ——Remaria onMr, W.MMorley Paper on-Alterate-Cerreat | —— ‘RemaronBr...Addenbrooke's PaperonBectrclEngineering Fricker, Mr.Guy C,Remaria onProfemor Jamia's Paper onthe TnetomRestateofElectricLightInallatione eneeeeTD—"pemarts oneuWeMeMordeysTaperonAlteraCurent | Geipt, M., Telegram tnreference toProfewor Jamieson’ Pape, a8to Insulation Resistance oftheEdinburgh Conservative ClubInstallation... 61 ! Granville, Mr.W.P., Remarks onMrA.E. Keanely's Paper onceria | ‘Phesomena connected with Imperfect Earth Telegraph Cicate M7, 188——“Temarhs onthePaperbyProfore Ayrton antPers,Laterory | ‘NotesonAlternate-Current Circuits... 0.sesee -~su | ——"TemartsonMf.©.L.Addeobrooke'sPapercaElectricaRopieeriog Graven,MrE(RetiringPeden),VolsofTousksto)ah sepretationofPremiumsbYmewsee ——RemarinActoowledgmentof VoteofThanks"sea Gray, Mr.W.B,emarka onBt. Jamie's PaperonthToelatonRe TianofBecreLightTotalalne ee weeneTBre nemartson Mf.debn B.Verity Paper onUnderground Cialisna creaCondeconvyaeaes 88,87,805 aatiogs Toston, Consmsication giving Detallof the by.B.Niehlaon(inreferencetoPrfemoramiewn'Paperswe at HonoraAvfitryVoteofTahato er) INDEX. an ——Trowrarer (Mr.E.Graves),VotoofThanksto0soos 838 Hookbam,Mr.George,Exhibition andDescription ofhisElectricity Meter688—— Remarks inreply toDiscussion ontheabove as. ae sae oneOOH Hopkinson, Dr.Joho, RemarksonProfessorJamieson'sPaperontheInsula- ‘tion ReristanceofElectricLightInstallations 0exaveMD ——Remarks onMr,W.M.Mordey'sPaperonAlternate.CurrentWorking oases cece anette ae AB. —— RemarksonMr,K.I.Murray'sPaperontheLightingoftheCenten- nial International Exhibition, Melbourne x. we snes one 766 —— Remarks inAcknowledgment ofhisElectionasPresident... ue854 Maghes, Profesor D. E., RemarksonProfessorOliverLodge'sPaperon Lightning, Lightning Condactors, andLightning Protectors... 491 Haman, Mr. H., Remarks onProfesor Jamieson's Paper ontheTosulation ResistanceofElectricLightInstallations. veevo OB Imperfect Earth inTelegraph Circuits, Oncertain Phenomena connectedwith,byALE,Kennelly oeoecece neeaeawe29Tnaagural AddressofthePresident (SirWilliam‘Thomson)— Ether,Electricity, and Pondersble Matter tue meee Communication inreference tothe above Paper, bySir William Installation (Hastings), Communication giving Detale of,byF.B.Nicholson (inreference toProfessor Jamieson's Paper) x eve we 2H Installations (Electric Light), The Tasulation Resistance of,byProfessor Institution ofCivil Engineers, Vote ofThanks toPresident, Council, and MembersofeeeaseseaneneeneaeoesO9T Insalation Resistance ofElectric Light Installations, The, byProfesor A. ‘Telegram from Mr.Geipel inreference totheabore Paper ww. awe 6LetterfromM.PlcouinreferencetohisFormulaquotedbyProfessorJamiesonimtheabovePapervueeevee aeOL‘RemarksontheabovePaperby— Mr.ROE, Crompton sweeten Professor W.E.Ayrton vee 65,77,102, 117, 118 tySpdoeyEversbed on. ween 8wyW.LaatCarpenter... mw 68,116byWeBBwon ae ane a) feGuyC.Fricker eee we weWeGraycoon TS an INDEX, ‘RemarksonPaperbyProfesorA.Jamies(continued)— Mr.RD. Smollie communicated). eee ieee feSORWalkeroeeee eae nee04SProfesor8.P.Thompson wvneceaneneve OT myCBSpaguoletth eae aeSTB beBWyle ee aes BB ‘CommunicationsinreferencetotheabovePaperby— ProfeworJamin: ase oeae Mr, William MeWhirter 0 vse se se ane ee weBW.Beckingmle nance Profesor Jamieson(inreplytoDiscussion). veoneowe18MrFB.Nicholwa eee eaten ‘Jamieson,ProfessorA.,TheInsulationResistance ofElectricLightInstalla- ——LatterinreferencetotheabovePaper nea ee——ReplytoDiscussion ontheabovePaper. cee ne——RemarisonSirWilllamThomson's PaperontheSecurityagainstDisturbances ofShips!CompassesbyElectricLighting Appliances...576‘Joveph, Mr, Leonard, RemarksonProfestorOliverLodge'sPaperonLight- ning, Lightning Conductors, and Lightning Protectors... ws ane 529 Joyee, Mr.S.,jan., Remarks onMr. George Hookham’s Electricity Meter 683 Kapp, Mr.Gisbert, RemarksonProfessorG.Forbes'sPaperonsomeElectric LightingCentralStationsinEurope,&e.sayawceaesweHT——RemarksonMr.JohnB.Verity'sPaperonUnderground Conduitstnd Electrical Conductors ais ese cena ne SB —— RemarksonMr.W.MMordey'sPaperonAlternate-Current Working632—— Remarks onMr. K.L.Murray's Paper ontheLightingoftheCen- tennlalInternationalExhibition,Melbourne ss.amesvs00 Kempe,Mr.,RemarksonProfessorJameson'sPaperontheInsulation Resist~‘anceofElectricLightTnatallations a.usmeeee ‘Kennelly,Mr.A.E.,OncertainPhenomenaconnectedwithImperfectEarth fimTelogeaph Circuits ase ase season I ~ INDEX. 3 Laboratory Noteron\lterate-Carrent Circuits,byProfessorsW.E,AyrtonfandJobaPerryaneese eee ame‘RemarksontheaborePaperby— Professor Ayrton (inreply) eons 1 BIB Library, NamesofDonors to...44,69,161,211,41,282,887, 686,709,823 — aecestonstOkeene ee198,TO,B58——ReportoftheSecretaryantothenena 880 “LightingoftheCentennialInteroational Exhibition, Melbosrae, 1888and1889,"The,byKL.Mary yee veTHO RemarksontheaborePaperby— ProfeworW.E.Ayrion eee, TK fGisbertKappnee ee80 Dr.JoboHopkingon nonsense ae6BLightning Conductor (+Arborial) atDhbubr, Assam, TeslaofTetaof,by —— Lightning Conductors, andLightning Protectors, On, byProfeworOliverLodge neem vee eee 888RemarksontheaboreFaperby— syCharlesW.Vincent ene ne89 Ted.Wimabart “tenaaan ProfeworW.GryllsAdamsyesswy GRLRitegera aa TD So opietogher LO at Lieat-Col, RY, Armstrong eeMe.ASSAde anaes OTCESpagnotets non oo aeSydneyEvebed nae BBSirWilliamToomon 2a BBMejorCardew(communicated)... er)Lioat-Col,J.Backillcommunicated) meBRT Me.LeonardJoseph(communicated). ne 2D ProfeworOliverLodge(inreply)neeee ATSBB ——NoportofaFatalAccidentfrom,inIndia,byP.V.Luke. auCOT om INDEX, ‘LocalHonorarySecretaries andTreasurers, VotoofThanksto=.ou887‘Lodge, Profesor Oliver, OnLightaing, Lightning Conductor, and Lightaing —— ReplytoRemarksontheabovePaperinsvsevemeAT6,SSI Lake,Mr.P.V.,ReportofaFatalAccidentfromLightaing inIndia... 687 ‘Mance, SirHenry, Remarks onProfessor Jamieson's Paper ontheInsula- tion Resistance ofElectric Light Installations wuz sue veo 6 McGregor, Mr. W., Results ofTestsofhis“Arborial””LightningConductor atDhabri,Assam oneangneeaneaes atea88‘MeWhirter, Mr.William,Commanication inreferencetoProfessorJamieson's ‘PaperontheInsalation ResistanceofElectricLightInstallations 121 Melbourne, The Lighting oftheCentennial International Exhibition, 1888 nd1889, byKe Murray ose sews ae tease THO ‘Meter (Electricity), Exhibition and Description byMr.George Hookhamof ‘Mordey, Mr.W.M.,RemarksonProfesorG.Forbea'sPaperonsomeElectric LightingCentralStationsinEurope,8c.secece seem288—— RemarksonMr.W.H.Preece'sPaperontheDisturbances arisingfrom theUseofEarth"forElectricLightingPurposes. weswS19 ——OnAlternate-Current Working se eons ove eSB —— Supplementary RemarksinreferencetotheaborePaper... =.63—— ReplytoRemarksontheaborePaper nememes OTS—— RemarksonMr.K..Murray'sPaperontheLightingoftheCen- tennial International Exhibition, Melbourne. wee TAS;Motor.(SeeElectro-Motor.)‘Murray, Mr.K.L., The Lighting oftheCentennial International Exhibition, Melbourne, 1888 and 18890 aes TIO NameoftheSociety,Changeof2mw eae Nicholson, Mr. F,B.,Communication (inreference toProfessor Jamiesoa's Paper omInsulation Resistance, &e.) giving Details ofthe Hastings Notes (Laboratory) onAlternateCurrentCircuits,byProfesworsW.E.Ayrton fandJohnDerryice aeons neeateneenear ae Original Communications:— Details and ‘TestsoftheHastingsInstallation, byF.B.Nicholaoe(inreferencetoProfeasorJamieson'sPaper)veween 8H Revults ofTests ofhis“Arborial” Lightning Conductor atDhubri, ‘Assam, byW.McGregor ee ewe 6 Report ofaFatal Accident from Lightning inIndia, byP.V. Luke... 09? ‘Notas onaStatic Electro-Motor, byCharles Zipernowaki... x. TOL INDEX. a5 Page,MajorFlood,Suggestionsbynone aeeaee_ 88 Patey, Mr. C,H. B,CB. Announcement oftheDeathofi.avn282 Perry,ProfesorJohn(Profemor W.E.Ayrtonand),Laboratory Noteson Phenomeas (Oncertain) connected with Imperfect Earth inTelegraph Cir- celts,byALE,Kennelly... oeeenstetra neORD Phillips, Mr.C.J.,Remarks onMr.John B,Verity's Paper onUnderground Conduits and Electrical Conductors ws. ns see one ones OTH loon, M., Letter from,inreferencetohisFormalaquotedbyProfessor ‘Jamieson inhisPaper ontho Insulation Resistance ofElectric Light ‘Pitman, Mr. C.E.,Remarks onMr. A.E.Kennelly's Paper oncertainPhenomens connectedwithImperfectEarthinTelegraphCircuits... 148,Ponderable Matter, Ether, Electricity,and(InauguralAddressofthePres- ent,SirWilliamThomson)nsswe,snes wetsae Preece,Mr,W.H.,DonationofanElectricClockfortheLibraryuuaw4—— Remarks onProfeent Jamieaon's Paper onthe Insulation ResistancefofElectricLightInstallations. sesieseeasa OD—— RemarisonMr.A.E.Keanelly'sPaperoncertainPhenomenacon- ‘ectedwithImperfect EarthinTelegraph Circuits... ...149,187—— Remarks onProfessor G.Forbes's Paper onsome Electric Lighting Central StationsinEurope,Beweeweeoewes288,241,261——OntheDisturbances arisingfromtheUseofEarth”forElectricLighting Parposes weceeeeoecease ame——ontheDeathofMr.WarrenDelaRueos.seeseas ay888—— RemarksonProfesorOliverLodge'sPaperonLightning,Lightning Conductors, and Lightning Protectors... ous ae aes 45, 515 —— Remarks onMr. K.L.Marray's Paper ontheLighting oftheCea- tennial International Exhibition, Melbourne... ie veo ous 704 —— Remarks onMr. Addenbrooke's Paper onElectrical Engineering in ‘Promiums, Presnntation of,toMr. A.C.Cockburs, Mr.E.Stallibrass, and ‘Ms. BO, Walker wv ee ant aaeB President (Retiring), VoteofThanks to.semene ee AL——(SirWilliamThomson),InauguralAddresofthe,—Ether, Electricity, —— RemarksinAcknowledging VoteofThanks. aeve 40——Communication inreferencetohisInauguralAddrem.. ... 128—— Remarks onProfessor Oliver Lodge's Paper onLightning, Lightning Conductors, and Lightning Protectors vow tase BIG ——0theSecurityagainstDisturbances ofShips’CompaasesbyEleetric Lighting Appliances ase cee ve nee ae tenes BT——ReplytoRemarksontheabovePapetweeSosaeoeoesTD—— Remurs on Mr, W. M. Mordey’s Paper onAlternate-Current Working eae see canteens teat OUT are INDEX. President (Sir Willa Thomon), RemarksonMr.GeorgeHookham’s Blc- —— RemarksonMr.G...Addenbrooke's PaperoaElectricalEmginocing — Commanteatoa inreference toMr. Protces Remarks onSubasrioe Cables, inthe Diseumion onME.Addenbrooke's Paper x... Append—Connell,andOf'cersfor1890,ElectionofoeBS ‘Ram,Mr.G.8.,RemarkaonMr.Addenbrooke'sPaperonElectricalEngineer- Raworth,Me,J.8., RemarksonSirWillamThomson'sPaperoatheSeeariy squint DitarbancesofShipsCompassesbyElectricLightingAppliancs57 — RemarksonMr.W.M.Mordey'sPaperonAlterate-Current Working.68 ——RemarksonMr,K.I,Murray'sPaperontheLightingoftheCenteonial International Exhibition, Melbourne we vz e646 Reckeasaun, Mr, A., Remarks ooMr. Addenbrooke's Paper onElectra! EnginoeringimAmerica eeeeweewees Reckenmus, Mr, F.,Remarks onMr. Addenbrooke's Paper onElectrical EngineeringinAmericaeee ean [ReportofaFatalAccdentfromLightaingiaIndia,byP.V.uke. @f —— oftheConnell, The Anoeal ane nee ame en ——oftheSecretaryatotheLibrarynwoveae enBD ‘Resitance (Tho Tngalation) ofElectric Light Installations byProfenor A. Results ofTouts ofhisLightning Conductor atDhsbri, Amam, byW. McGregor. ae ea mee eee Robingoa, Mr. Mark, Remarks onProfessor G.Forbes's Paper onsome Electrlo Lighting Central StationsinEarope,86...wesw 18 ‘Remell,Me.Stuart,RemarksonProfesorG.Forbes'sPaperonsomeElecticLightingCootralStationsinEurope,86.ese naeaeMB SeratincersoftheBallotforCouncilandOflcer,Appointment of... Security against Distartances ofShipe’ Compames byElectr Lighting Appliances, Onthe, bySirWilliam Thomson eae ow Remarks onthe above Paper by— Stal-CommanderCreak,BN.nseaeae Me.AlexanderSiemens oeProfemorGeorgeForbesmeeee Mel. 8Raworthne see see TE SirWiliamThomson(Ineply)oe aeeeeSiemens, Mr.Alexander, Remarks onProfeuor Jamionon's Paper onthe Insulation ResistanceofElectricLightInstallations =... ms© INDEX. err Siemens, Mr, Alexander, Remarks onProfessor G.Forbes's Paper onsome ‘ElectricLightingCentralStationsinEurope, de...224,247,256,279—— Announcement oftheDeath ofMr. C.H.B.Patey oe se am 282——RemarksonSirWilliamThomson's PaperontheSecurityagainstDistarbancesofShips’CompassesbyElectricLightingAppliances... 573 Smellie, Mr, Communication inreference toProfessor Jamieson's Paper (nthe Insalation ResistanceofElectricLightInstallations... msOL Smith, Mr. Holroyd, RemarksonProfesorJamieson'sPaperontheTnsala- tion ResistanceofElectricLightInstallations... oveewe1A Ships? Compasses, OntheSecarity against Disturbances of,byElectric Light- ‘ingAppliances, bySirWilliain Thomson eve ue veo OUT Shoolbred, Mr., Remarks onProfessor Jamieson's Paper ontheInsulation ResistanceofElectricLightInatallation® ..eve weweMIL—— Remarks onProfessor @.Forbea's Paper onsome Electric Lighting —— Remarks onMr. Joha B.Verity's Paper onUnderground Conduits fand Electrical Conductors ose nee ceases 986, 869 Some Electric Lighting Central Stations inEarope, and their Leasons, by ‘RemarksontheabovePaperby— Mr.Gisbort Kapp oe ose see see wee nee ne oneID Dr.Fleming. ie ne eee meme IT Mr. J.Swinbarme ne ee eee BV, Andemon sone eee BB, BT fyAlexander Siemens... vse vse sue 984, 47, 256, 279 syMark Robinson.c. ose see see see nee one one 288 meRESCrompton oe aeoeaeneeweeBE myWoHLPreece onsen seas 286, 24M, 26vyStuartRowellseeceeseeseeneeneeneo Major-General C.E,Webber (communicated)... sv swe208. ProfessorW.EsAyrtonsueevesteweeweeoneoneBOL Professor George Forbes (inreply) ve weve mene 2068 Spagnoletti, Mr.C.E,,onSeconding Vote ofThanks toMr.E.Graves (WheRetiting President) eave nena AB —— RemarksonProfessorJamieson'sPaperontheInsalationResistance ofElectric Light Installations. me vee me QB onthe Death ofMr.CHB. Patey se sue ae oes ame282 oe INDEX. Spagnolet, Mr. C.B.,Remarks onProfeaor Oliver Lodge's Paper onLightning,LightaingConductors, andLightningProtectors ws.SIE—— Remarks onMr. Addenbrooke's PaperonElectrical Engineering in Special Resolutions ireferenestoChangeofNameeeeae ‘Static Electro-Motor, Notesona,byCharlesZipernowaki ...=...=.701 Students’Meetings,PeriodduringwhichAuociatestranaferred fromthe CameofStadentamayAttend.ceeeweae 0 Satton, Mr.George, Remarks onMr.John B,Vetty’s PaperonUndergrownd ConduiteandElectrialCondaetoreoeeweaeST Swan, Mr. Howard, Remarks onMr. A.E.Kennellys Paper onceria‘Phenomena connectedwithImperfectEarthiTelegraphCircuits... 156Swinburne, Mr, J,RemarksonProfesorG.Forbes'sPaperonsomeElectric “LightingCentral StationsinEurope86.weeee aI9— RemarksonMr.W.M.Mordey'sPaperonAlteraateCurrentWorking.653 —— RemarksonMr.G.L.Addenbrooke's PaperonElectricalEngincer- IngimAmerich neeet aeeeame S Symons, Mr. G.J,RemarksonProfesorOliverLodge'sPaperonLightning, ‘Lightning Condactor, andLightningProtectory wexeone499 “Telegraph Circuits,OncertainPhenomenaconnectedwithImperfectEarthin, DyALE.Renvelly aneween ea“Thompaoe, Profesor Silvanus P,LaterinreferenceotheAwardtohimof —— RemarksonProfemorJamieson’PagerontheTngslationResigance ofElectricLightInstallations aiewreea UG ——ontheDeathofMe.WarrenDelaRae ss 88 ——RemarksooMr.W.M.Mordey’sPaperonAltermate-Carrent —— RemarksonMr.K.L.Matray'sPaperontheLightingoftheCen. tennlal International Exhibition, MeIbOarm® we ome east TES —— [email protected],Addenbrooke'sPaperonElectricalEngincer~ ——Exhibition ofAmericanBlueGlasTomlators st“Thomwoo, SirWiliam. (See President.) ‘TransferofAmociatestotheClamofMembers 2,44,129,289,385,681,709,779—— ofStudentstotheClausofAsociaten 244,80,161,282,897,85481, 631, 708 ‘Trott, Mr. A.P»,RemarksonMr,JobaB.Verity'sPaperonUnderground GoodenandHclContacwewemows | UUndérground Conduits andElectrical Condsetory byJuba B.Verity. 388 RemarksontheabovePaperby— | _ . 4 INDEX. 879 roe ‘RemarksonPaperbyMr.JohnB.Verity(continued)— jeRS Brine eee eae B88ProfeworAyrtonsowesenese 810,875,818 Major-GeneralWebber(communicated) svsam 87T Me.Verity,fareply(commanicated) wveee BBL ‘Uwe of Earth”forElecticLightingPurpose,OntheDisturbancesarising from the,byW.H.Preecesee iene eae ‘Verity,Mr.JohaB.,Underground ConduitsandElectricalConductors... 888— Reply (communicated) toDiscusion ontheabove Paper ve. 38 Vincent, Mr. Charles W.,Remarks onProfesor Oliver Lodge's Paper oa“Lightning,LightningConductor, endLightningProtectors. wwe459 Walker, Mr.8.F,,Remarks onProfessor Jamieson's PaperontheTaulaton ResistanceofElectricLightTotallations eam 100,118. Webber, Major-General C.E.,on Moving Voto ofThanks toMr. E.Graves — Remarks (communicated) onProfewor G.Forbels Paper onsome Electric Lighting Central Stations iuBurope, £600 eves 288—Remarks(commanicated) onMr.JohnB.Verity'sPaperonUndertgoundConduitsandElectricalConductors vewae eneITWinaburst, Mr. J,RemarksonProfesorOliverLodge'sPaperonLigbtaing, Lightalng Condactor, andLightaing Protectors... wu 461,471, 481 Wright, Mr.Arthur, Remarks onProfesor G.Forbes's Paper onsomeElectricLightingCentralStationsinEurope, 8.eve 280Wyles, Mr.F.,Remarks onProfessor Jamieson's Paper ontheTngslation TRevitanceofElecticLightInstallations aoeaeeveBIB Wynne, Mr. Frank, RemarksonProfesorG.Forba'sPaperonsomeElectric LightingCentralStationsinEurope,£6.messswee ne205 Zipermowaki, Mr.Charles, Remarks onMr,W.M.Mordey's Paper on Alteroie-Correat Working, ve sue we ene sa OTL NotesonaStaticElectrMot0r eee oeTOL ABSTRACTS, ‘Accamolators forStation Working, Experiments with, byW.Koblrauch and — forTelegraph Work, Connection of,byC.Grawinkeln. oe1B—=oFartakyandSebonek,Experiments withthe,byDr.A.¥- Vou. xvi. 61 “880 INDEX. Acoumulators,TheUseof,inTelegraphy, byC.Heim... sea ‘AcidalatedWater,Spontaneous ExplosionsoccurringdaringtheElectors of,bymeans ofanAlteroatiog Current, byG.Mancarrer andJ. Action ofaMagoetonChemicalAction,byH.A.RowlandandL-Bell.© Alterating Current, Spontaneous Explosions occurring duriag the lectrolyais ofAcidalated Water bymeansofan,byG.Mancarrier and — Garren, Electrolysis bymeansof,byE.Drechuel =o8H ——— The Mechaniam ofElectroysis bymeans of,byJ.Chappuis ‘Anon,Friteche'sDynamowithRadil-Bar Armatares ow‘AreLightPhotometry, byW.Wellingwae esos——TheCounterEMF.ofthe,byH.Dube Armature, Radial-Bar, Fritsche's Dynamo with—Anoo. = ‘Arsonval. (See D'Arsonval.)‘AniclesrelatingtoElectricityandMagnetinnappearinginsomeoftbeprincipal English aod Foreign Techoleal Journals, Clasifed Lits of 15,208, 834, 441, 705,86 Atmospheric Electricity, byDr.Leonard Weber vse as ee a ‘Ayrton, W.E,,and J.Perry, E@lcency ofGlow Lamps with Direct and Dickstrim, TE,ConductivityofSpecularTronOreeae DadContacts,PasmageofElectricCurrentsthrough,byBllatiandLass.2Batteries, Bichromate, withot Diaphragms, byE.Landmann w= 31 Bebaviour(The Electolytic)of Mica ataHigh Temperature, by W.H.Sebalue 4 Bell, L.(H. A.Rowland and), Action of MagnetonChemicalAction. Bella, M.,and8.Laswoa, Occlasin ofHydrogen byNickel... 3 — —— PeamageofElectricCurrentsthoughHadContacte... Bergmana, J.,Observations ontheChange intheConductivity ofMetals rodacedbyHeating eeenee a Bichromate Batteries without Diaphragms, Experiments on,byE,Tandnaaa Boarganeuf, Electric Traominion ofPower at,byM.Depres —.. ~85Boaty,E.,andL.Poincaré,Conductivity ofMoltenSalts... 8 Change inthe ConductivityofMetalsproducedbyHeating,Obserratiosea the,ByJDergmaneae Changes(Permanent) prodacedinCopperWiresbythePasageofaCurren DyFeLaroque eee meCChaperon,G.,andE.Mercadier,Electro-chomical Radiophoor |,1.#Chappals, J,andG.Manceurier,TheMechanismofElectrolysisbymess fofAlteroating Currents ee eee ete — G.Manourrier and), Spontaneous Explosions occurring dating te Electrolysis ofAcidalotetWaterbymeansofanAlteuating Curent& INDEX. 8st ner ChemicalAction,ActionofaMagneton,byH.A.RowlandsandIsBell..88 Chloride, Electrolysis ofCaprous, byF.Quincke ane ve ame 208 Cusifed Lists ofArticles relating toElectricity and Maguetiom appearing insome ofthe principal English and Foreign Technical Journals£5,208,284,441,705,816,858, CombinedEffectsofTorsionandLongitndinal StrestontheMaguetisation ofNickel,byH.Nagaokaseeeeovane 828 Caapounds (Organte), Specific Inductive Capacity ofsome, byS.Tereschin 437 ConductivityofMatas,IntluenceofMagnetistion onthe,byD.Goldhammer 438 ——— Observations onthe Change inthe, produced byHeating, ——ofMoltenSalts,byE.BoatyandI.Poinearsn. Soeoe81.—ofSolidMerenry,byC.EnWeber ease ase880——ofSpecularIronOre,byH.Bickstrima.ne ea488Condectors, Lightning, byW.Koblransch wee ee TB iy Neetemee MO —— — Eurthing, bymeans ofGasandWater Pipes, byA.Voller... 84 Consection ofAccumalators forTelegraph Work, byC.Grawinkel 818Contacts(Bad),PassageofElectricCurrentsthrough,byBellatiandLnsana202——(Microphonic), Experiments ov,byJ.Momserveone we482Copper, Disintegration of,bythoElectric Current, byB,Nebel ve awe499 Copper Wires, Permanent Changes produced in,bythePamageof«Current, CounterE.MLF.oftheArc,The,byH.Dabevueeave 489 Caproas Chloride, Electrolysis of,byF.Quincko see aes OO Carrent, Electrolysis ofMercurous Nitrate asaMeasure of,byA.Potier ...432 —— Permanent Changes produced inCopper Wires bythePassageofa, Carrenta (Alternating), Blectro-dynamic Rotation produced bymeans of,by —— Cirect andAlternating) EfficiencyofGlowLampswith,byW.E. —— Paseage ofElectric, through Bad Contacts, byBellati and Lussana... 202 D'Arwonval, AUniveral Dead-Beat Galvanometer wv me vows 208 Dead-Beat Galvanometer, AUniversal, byD'Aroaval oa 208Depres,Mc,ElectricTrantmission ofPoweratBourganeaf mo.swe815 DesignforaStandardofElectricalResistance, A,byDr.J.A.Fleming... 201Determinationofthe‘TrueObmbyEDoraseseeneve BBDinintegrationofCopperbytheElectricCarrent,byB.Nebelwww499 Dimociation TheoryofElectrolyte,byJ.H.vanHoffandL.'T,Recher..$82 Divergence ofElecro-motive Forces from Thermo-chemical Data, byE.F. Dorn, E,,DeterminationoftheTrueObmonenee BRB ‘Drecbeel, E.,Electrolysis bymeans ofAlternating Current... vw sw208 82 INDEX. om Dube,H.,TheCounterEMP.oftheAronesee aeDynamo with Radial-Bar Armatures, Fritache's—Anon. aw Earthing Lightalng Conductors bymeansofGasandWaterPipes,byA.Valet EficiencyofGlowLampewithDirectandAlternating Currents,byW.E. ‘AyrtonandJohPerryseneanonenae ElectricalResistance,ADesignforaStandardof,byDr.J.A.Fleming..2 ElectricCurrent,Disintegration ofCopperbythe,byB.Nebelww(9—— Carrents, Passage of,through Bad Contacts,byBellatiandLassam32 —Force,Baysof,byH.Hertzseanenea ——TranemissionofPoweratBourganent, byM.Deprez.. a Electricty, Atmoepheric, byDr.LeonardWeber wvxeaneMDEleciro-chemical Radiopbony, byG.ChaperonandE.Mereadier 8 Eleetro-dynamic Rotation produced bymeans ofAlternating Currents by Electrolysis bymeansofAlteroating Currents,byE:Drechiel —ofAckdalated WaterbymeansofanAlternatingCurrent,Spontaneoss Explosions occurring daring the,byG.Mancravrier andJ.Chappuis.. —ofCaprousChloride,byF.Quincks neaeeta BE——ofMerearous NitrateasaMeasureofCurreat,byA.Potiera»——TheMechanism of,bymeansofAlteroatingCurrents,byJ.Chappris aod G,Manceartier eae seam ‘Electrolytes, Dissociation Theory of,byJ.H.vantHoff andLT. Reicher 3 Electrolytic BehaviourofMicaat«HighTemperature, The,byW.H.Scbalue Electro-motiveForceofThinFilmsofHydratedSaperoxides,byK.Schreber1 ——Forces,Divergence of,fromThermo-chemical Data,byE.F. E.M.F.,TheCounter,oftheArc,byH.Dube sa Ewing, Profewor J.A.,and W.Low, Induence of«Plane ofTransrereSectionontheMagneticPermeability ofanTrou Bar.ayeExperiments onMicrophonic Contact,byJ.Moomerneseveam— with theAccumalators ofFarbaky and Schenck, byDr. A.¥. Waltenboten ee eae ae eae Explosions (Spootancous) occurring daring the Electrolysia ofAcidalaed Water bymeans ofanAlternating Current, byG.Manauvrier and J. Farbaky and Schenck, Experimenta with theAccumslators of,byDr. A.¥. Waltemhoten onsen eaten aw Ferrari,G.,Electro-dynamic RotationproducedbymeansofAlternating —| ‘Fleming,Dr.J.A.,ADesigaforaStandardofElectricalResistance. Fritche'sDynamowithRadial-Bar Armatare—Anos. awn Gatvaiometer,AUniversalDead-Beat, byD'AMoavA. seweSH INDEX. 8s usandWaterPipes,EarthingLightningConduetom bymeansof,byA. VoURE ane ee enna low Lamps, Eficioney of,with Direct and Alternating Currents, byW.E. Goldhammer, D.,Influence ofMagnetiation ontheConductivity ofMetals 488 Grawinkel, C.,Connection ofAccumelators forTelographWorkawa818 Heating, Observations ontheChangeintheConductivityofMatasproduced Heim,C.,"TheUseofAccamalator inTelegraphy ievay wee208—=W. Kohiransch and), Experiments with Accamulators forStation Herroun, E. F., Divergence ofElectro-motive Forces from ‘Thermo- chemical Data ee eens 8Herts,H.,RayeofElectricForcesenee eee AE‘Hof, J.H.van, and L.'. Reicher, DissociationTheoryofElectrolytes... 383 Hydrated Saperoxides, Electro-motive Force ofThin Films of,byK. BebEObee eee nee eee ABS Hydrogen, Ucclason of,byNickel,byM.BellatiandS.Lusunawu$82 TreandescentLamps,(SeeGlowLampe.) InductionCoilsofMicrophones, The,byDr.V.Wietlibach wnsas812 Inductive Capacity (Specific)ofsomeOrganicCompounds,by8.Tereachin487 InfaencoofPlaneofTranrverve SectionontheMagoeticPermeabilityof anIronBar,byProfeworJ.A.EwingandW.Low wow 88— ofMagnetintion ontheConductivity ofMetals, byD.Goldhammer 438Iridiom(Platinum) forStandardResistance byJ.Klemencle «x.880TronandSteel,Experiments onthePermeability ofdiferentSortsof,byTT —— Bar, Influence of«Plane ofTransverse Section ontheMagnetic Permeability ofan,byProfewor J.A.Ewing andW.Low wn 88 —— re, Conductivity ofSpecular,byH.Bickstrim n,n 3B — Resinance ofMagnetic,byG.HevonWyaeanne B88 Klamencie, J,,Platinam-Iridiam Wire andsome otber Alloys forStandard Koblrausch, W.,Lightning Conductors eae ae ae eB——andC,Heim,Experiments withAccumalator forStationWorking... 810 Tamps (Glow), Effcloney of,with Direct and Alterating Currents, by WE.AyrtonandJ.Perryieweeaeeae Landman, E.,Bichromate Batteries without Disphmgms S81 Larrogae, F,Permanent Changes produced inCopper Wires bythePasmage Lightaing Conductors,byW.Koblrausch o.oo tyNomen ee MO ——— Ennthing,bymeansotGasandWaterPipesbyA.Voller... 84 884 INDEX. Lists (Classiod)ofArticlesrelatingtoElectricityandMagnetismappearing {Insome oftheprincipal English and Foreign Journals 85,208, $34,441, 1 816, Longitadinal Stress,CombinedEffectsofTorsionand,ontheMagnetisatioafofNickel,byH.Nagaoka semeas neeeaeLow, W.(ProfeworJ.A.Ewingand),InfluenceofaPlaneofTransverse Section ontheMagnetic PermeabilityofanTronBar...es Tussane, 8.(M. Bella and), Occlusion ofHydrogen byNickel... 8 —— —— Passage ofElectric Currents through Bai Contacts... 22 ‘Magnet, Action ofa,onChemical Action, byH.A.Rowland andL.Bell... Magnetic Iron, Resistance of,byG.H.vouWyss eae weSM—— Permeability ofanIron Bar, Influence ofaPlane ofTransvene Sectiononthe,byProfessor J..A-EwingandW,Low a‘Magnetiaation, Influence of,ontheConductivity ofMetals,byD.Goldhammar 433——ofNickel,Combined EffectsofTorsionandLongitadinal Stressoo ‘Manauvrier, G.,and J.Chappuis, Spontaneous Explosions occurring during theElectrolysis ofAcidulated Water bymeans ofanAlternating J. Chappais and), The Mechanism ofElectrolysis bymeans of Alternating Currents eee ete ae MeasareofCurrent,ElectrolysisofMercarousNitrateasa,byA.Potier..4! MechanismofElectrolysis bymeansofAlternating Currents,The,byJ.Chapprieand G.Manwurrier s,s awesome ‘Mercadior,E.(G.Chaperonand),Electrochemical Radiopbony...8 ‘Mercurous Nitrate,Electrolysisof,asaMeasureofCurrent,byA.Potier...43? ‘Mercury(Solid),Conductivity of,byC.1.Weberwaetease ows Motals, Inftuence ofMaguetisation ontheConductivity of,byD.Goldhammer 488— Observations onthoChange intheConductivity of,produced by Heating,byJ.Bergmann eaeenero AE ‘MicaataHighTemperatare, TheElectrolytic Behaviourof,byW.H.Schulte483Microphones, The Induction Coils of,byDr.V.Wletlisbach ... ws v=‘SIP‘Microphonie Contacts,Experimentson, byJ,Moosersn somA ‘Molten Salts, Conductivityof,byE.BoutyandL.Poincaré... ow.aw818 ‘Moover,J,,Experimenta onMicrophonic Contacte aw owHE ‘Nagaoka,H.,Combined EffectsofTorsionandLongitudinal StressontheMagaetisation ofNickel oe eee neta Nebel, ByDisintegration ofCopper bytheElectric Current vay ve a489 Neeson, FLightning Conductors. nena HO Negbaur,W.,Experiments onthePermeability ofdifferentSortsofIronand ‘Nickel, Combined Effects ofTorsion andLongitadinal StressontheMagoeti- |sationof,byH.Nagaoka seweeeae eee |—— Ocelasion ofHydrogenby,byM.Bellatiand$.Losana SHE INDEX. 885 ‘Nitente (Merearous), Electrolysis of,axaMeasureofCarrent,byA.Potier..482 Occlusion ofHydrogen byNickel, byM.Bellati and $.Tusuna vs 892 ‘Ohm, DeterminationoftheTrae,byB.Dorasewean BBB Ore,SpecularTron,Conductivity of,byH.Bicksttima, on. ABOrgunic Compounds, Specife Inductive Capacityofsome,byS.Tereachin 437 Parker, J.Thermo-electrie Phenomena wv ms ease 2OL ‘PangoofElectricCurrentsthroaghBadContacts,byBellatiandLassana202PermanentChangesprodacedinCopperWiresbythePassageofaCurrent,YyB.Larroque wv sae sn one nt one anne 0Permeability (Magnetic)ofanIronBar,TatluenceofaPlaneofTraneverseSectiononthe,byProfesorJ.A.EwingandW.Low.vwwe88 oF differoat Sorts ofTron and Steel, Experiments onthe, byW. Negbane ee nee eee tee tne HL Perry, J.(W. E.Ayrton and), Bficiency ofGlow Lamps with Direct and ‘Alternating CURED! sn ok knee wees BL Phenomena, Thermo-slectrie,byJ.Barker eee 20D Photometry, Are Light,byW.Weng... sueuvsuewweawe810) Pipes (Gas and Water), Earthing Lightning Conductors bymeans of,by Voller ee tee eee eae PlanoofTransverseSection,Influcuceofa,ontheMagueticPermeabilltyof anTron Bar,byProfenorJ.A.EwingandW.Lowwvwww Platioum-ttidium Wiro and some other Alloss forStandard Revistances, Poincaré,L.(E.Boutyand),Conductivity ofMoltenSalts. uevu818 Potier, A.,ElectrolysisofMercurousNitrateasaMeasureofCurrent... 482 Power, Electric Transmiaion of,atBourganeaf, LyM.Deprez ww vu816 Quincke, F.,Electrolysis ofCuprous Chloride we as we nee 20 Radial-Bar Armature,Fritece'sDynamowith—Anon..: weosae499Raciophony, Electro-chemical, byG.Chaperon and E.Mercadier 8D RaysofElecticForce,byH.WerteeeweeoesABH Reicher,L.'T.(J.H.ran€Hoffand),Dissociation TheoryofElectrolytes .902Resistance (Electrical), ADesign for Standard of,byDr.J.A.Fleming... 201—ofMagneticLron,[email protected] Resitances,Platinum-Iridiam WireandsomeotherAlloysforStandard,byJ.Riemenci eee 880 Rotation (Blectr-dynamic), produced bymeans ofAlternating Current, by G.Ferrarini eae seek Rowland, H.A.,andL,Bell,ActionofaMagetonChemicalAction 83 ‘Salus,Conductivity ofMolten,byE.BoatyandL,PoincaréwvsuseO18 06 INDEX. “ Schenck(Farbakyand),Experimenta withtheAccamalatory of,byDrA Schreber, K.,Electomotive Force ofThin Films ofHydrated Superorder «2 Schulte, W.H.,The Electrolytic Behavionrof Mica at«High TempersareSolidMereary,Conductivity ofbyC.Le WebravwnmBYSpecie Inductive CapacityofromeOrpunieCompounds,by§."Terexhio-. SpecalarTronOre,Conductivity of,byH.Bickarim eno 8Spontancous Explosions occarring during the Electrolysis ofAcialind Water bymeans ofanAlterting Carent, byG.Mancrrer StandardofElectricalResistance,DesignforabyDr.J.A.Fleming. 3 — Resistances,Pltioum-Iridiam WireandwomeotherAllofa, byTeKlomeneieane eaeeeeeaB® Sution Working, Experimenta with Accumulator for,byW.Koblrasichnd ‘Steel (Iron and), Experiments onthe Permeability ofdiferent Ses byWENO ae eee eeeStrom(Longttdinal, CombinedEifectofTorsionand,outheMagnetiton Saperoxides, Thin Films ofHydrated, Electro.motive Force of,byK ‘TelegraphWork,ConnectionofAccumslatorfor, byC.Grawinkel...-##‘Telegraphy, TheUseofAccumulators in,byC.Heim. sueaeoe ‘Temperature, TheElectrolytic BehavioarofMicaat«High,byW. ‘Terechin,8.SpecificInductiveCapacityofomeOrgnnieCompounds. &‘Thermo-chemical Data, Divergence ofElectro-motive Forces from, byEF | TuemoslevePhoneJ.Puterosoe | TTHot(SeeHof.)‘TorsionandLoogitadial Strer,CombinedEfectaof,ontheMagoetimtie ‘Transniaion ofPower(Electric),atBowrganen, byMDeprer =»8‘Tranevene Section, Infleenco of«Plane of,ontheMagnetic Permabilito ‘anIroaBar,byProfemorJ.A.EwingandW.Low ww& ‘TraeOhm, Determinationofthe,byE.Dornwee UnivermlDead-BontGalvanometerAyhyD'Atmonval eoBF UseofAccamslatori Telegraphy, The,byC.Helmavmte VanHof,(SeeHf)Voller,A.,EarthingLightning Conductors bymeansofGasandWaterPipes VouWym(seeWyas) 4 INDEX. 8st race Wattenhofen, Dr.A.¥., Experiments with theAccumulator ofFarbaky and ‘Water Pipes (Gas and), Earthing Lightning Conductors bymeans of,byA. Weber,C.12,CondugtvityofSolidMereary esse 800 Weber,Dr.Leonard,Atmorpherie Electiity sue eeBIBWedding,WoyAreLightPhotometry neeos meB10Wietlibach,Dr.V.,TheInductionCoilsofMicrophonesnsmsos818 Wires(Copper),PermanentChangeproducedi,bythePasageof Carrent, byFLLarrogue ae ee ene anes 2Wy,GL.von,ResistanceofMagneticTronee cee BD | | Gr?@®°at.e&ee, x ~9Bd, Seig9°?ay re 5VS 2g ® °@%.bsheaft 2 ae\ees@..: ._2°@ ge: e*s a! €> be @ » “fee ¢ @:-o@ te<< Use2£.e.°' “= ~.@°. «Je*_@F+" Petey Ne “i yeas Hu24. @ . j @ a Vaan “2 eo te. © ‘¥ NS{ r ee NeBoe 3S:‘Se °@ BNe,. e s&s * &— ee Ad @)@‘Se ¢@°. one Wee ae ee S Ce®. LoWee. 6e@q @e, ne.aS ay @° oH: -@ \ & e: Oe:6,e. 2es Sy EN Baresos Yo, e : P h Pa |2. i _de hy r Es . =| ): |. < 4 J vdated| < 4 x 9 “>? is . J* ~ | f i+ 4 tt . a) ou ~ j ‘= ~ ~