H" llw W l H M1 “If” W M. f H W "I, J" ‘— ~ ,_ l f ”in W I! J , I ‘1} y l W I W THESIS A 5mm 0: NOISE AND CROSS-TALK cm MLEFHONE TRANSMISSION LINES Reid L' Rayner 3924 .hnv'mu . [HESIS’ ___-____H_-H_ _ “a“ ,_ -_ V ,, H w , r .7‘ I in ST 1:JY OE T111133? 7T“. TIL'J'UI'ILIJIOI. LITIJL)’ A Tiesis Suintteu to Ihe Faculty of TICTIJAI nqfiICULSUhnL CCLLUJE LEI) L.;Lgfiifig f‘“ “3 Gaugidgte for the Degree of Pastor of Science JUIE 1924. THESXS Table of Contents. Nature of the Subject 331111011103 Balanced and Residual Components fittest of Configuration on veitage Balance w transpositions the Mbssurommnt of Reine The Hessuremsnt of Cross-Tank mhe Effect of Resistance Unbalance in Communication Circuits Series Resistance and Insulation Tests Beta taken on Transmission Lines Conclusion Bibliography 103%99 my 11 18 26 30 33 38 4O 4? Nature of the Subjnct. Inductive interference involves priLciplos uiicn have berm kLcwn for many years. The truncmiccion of clcctric power requires cn electric current and a voltLro or nrccsure to 0 LCD thic current to flow. ?hn V01tu”0 accociatod “itu any elcctric circuit is accomfianiod by an clcctzic field of force or conu itio of stress in tae surrounding space, wnocc intcusity is procortional to the voltaqc. At the cane time the corresponding'electric current in LCCOUHd‘iWlef£.ZE 1mtic field of force '1'rz;ic;1 occupies: the SL110 vrntmcir" 8312,20 and whose intnnsity is proncrtionLl to tan current. The intensity of those fields of force, in gene rul, decrcafie Vcr? rL1i1lV'Fi t1 incrn sinw dicthcc from the circuit. Conversely, Lny otr1cr cirC1it Within those 110143 of force will hLve voltages and curronts not ug or "induced" in it, when the voltage or current of the first circuit annqes. Alternating current gowcr circuits most commonly engloyed in 101.7“1 tr ns.1isci0L, uu1miy . their volt; :8 uLd currents cont i LuLlly eryiL? will continua 11;! iLc ice ViltL cs LzLj currents in a Lcichhorinc cocmunithinn circuit. Althouch the electric uni rucrctic fields dcuimisu vary rLiidly with diet non, ccwmunicufiior circuit“ LL“ vc"y scnsitivc and in tho cuqe of a 10L” rcézllol incmctivn intcrfcrcficc bacowcs a vcrr carinuq Droblcm. A vory grail amount of nowcr in trtrcnittc; cvcr L tnlcqnon1 line u"é a very S'Lll iuiuccd cur'FL will be a coufiidorafilc portion of th“ totcl Current in the line. The povor rcxuircn to livnt L FULll Plectric leo in sufficiCLt v . 831 .3. T ,.A girtt‘“. , J , y -2- to cauqe a leu» neiee in a million telewhere recciverc. It CL1 he ctLted teat the limit of inducw: currext wuicq iece Lwt cLuce objectiorLl raiae 19 apgr oxir1u1tel;,' 0110—0110 hundrci rii’LliOLtri 01' e110 ‘rictt fer a fro-1;; .'.11;c,J 0:“ 8013 C'w'cles. 011 the ot1'10r 2.111... pO'vcr Cgl'l'if? .1 by comer tz‘cz'.r.r.-1iss-tir)1- liLiir‘. if. ‘f eucn Ych bein+ increased Lnd i: oftJn rqu to thousLLcs e; KllOWgttS. The object of tmic thesis i? te study the factors vuich iLglueLce inductive interference. a lLr?e nunOcr of nchuremcnts nuve been mace on \J- actual transmiszion lines in erccr to illustrate tue tucorg precentci in this discussion. It is: the-1retic;.ll;; _1ossi'ul-3 t3 castrwct L t-ate‘unLe liLe -.'.r.‘1ich is not subject to iruuctive interference from L LeixnboriLr govar line. Likewise, it is theoretiCLllg possible t3 construct L power liLc which will not Cause inductive interference iL a nei"hberinj cowwuLicution circuit. "owever, this: 1311-1 cemlit i111 cmmet be Lttz.incd i1: j)1*a.c:ticc Lithmt'fi'i it may be aggrezicheai. It is very fortuLLtc my cat-1.01 ucei te eliflizL-te . ‘ - ." ' ‘ ‘ " I" ” . n a x‘ ‘1: ‘-‘- I x .‘ - 1‘ . ' .° ‘ interference 130‘ not 11 LLJ 4a. inteire u1t1 tafi heft ruc.101 in eituer p¢fier or teleunené trLLcfiiPcicL. Harmonics. Unuer norle operatin; GonnlthLS of tie cisturbing power circuits, interference to telephone cir uits is iue almost eLtirely to the sinner harfionics of toe 90V9T circuit voltLrec LLc currerts. The r Lsin is that the harmonics cover L considered-1e aortion of tie rue of fro ‘uencies of hurt-1.11 speech at which teleohone LooLratus is most sensitive. Except when interfer- ence is very slirdt or very revere, the LetrioeLtLl cffect of extrLLeous current in L telo hone receiver i;cresees Lonroxiectcly in direct nroportion to the MLRLituLe of toe current. IncrchiLe the frexueucy cLurer L very rs1id incresse in tufi detrineLtLl 1ffcct (rouinly as the saugrc o” toe freiuency) u» to abort 8:0 cycles beyond Wiich there is L erLiuLl cecreLre. Tue relLtive eff:ct of different frOgUQLClES th been very CLre— fully writes out by oxoeri'ient. “Due first series .11 tea s 171.3 mace by tie American Telephone Lad Teletrugh Cooany at the re;uent of tJQ CLliforLiL Joint Conuittee on Inuuctive Interferexce. IL tue first tests1mioe in 191+ a tryicsl teleqLoLe circuit st set u), usirw LrtiriciLl CLLlc Lnu erLnficd so tiLt different siLcle—fre— quencv currents couli be eegcrLtely intrOuucec in the teleqhono receiver Lt the lirtenine 9L1 of the cirC1it, s1multheeuFly with normal telepLOLe current. 3? throwinv a switch the listeLer could remove the cisturbing currert frfiu the circuit arc at the ssue time retuce tee volume of sounc .utout from the receiver, the Lmount of 9 reduction an the Loownt or tee cistnrhiLb current LeiLV set by toe rt;fier er beirr~1u.:rxn1 to the listener. The procedure “Ls for the Cullcr to reLi over tee circuit while l4“ liFtLer switcoei‘ from the noisy concitioz‘. to the cuiet corriitiot. cm rev-“arse, until t'io latter 9 Lo his (indff’rrtt or". to tr‘uicn of t to circus? .e afoul“ be {referable for Corrlg'iz- '? or a. L:1*:;a,rzt-r8cti01;. Winn-:1 the caller 121.. cfm. “P. tie settiu '1: till to? li.:tm.=er 13011111 rc wit the two circuits e“ 8111...].- I}; tuis ”5a.; tut“) relctivo il:t"21'£"‘.3'lil~’f off-cots 01' Cif3"r>:-~>.1;t fregtiezxfles " crc unriced cit. Ir. 1318 more estersi're tests wore '1110. flare wore trio other types of tests 21.10 ct t.iis tine. One was an urticulction test 11. which a. row or trial l pronoxmco c. lcrkjo nuri‘oer of sd'llu‘cles. cm, gorccntcgo re- \ cordon; correctl;r by the listener vein a. mocmirs of tile unclittx 01‘ 1.3110 transoission. 7,)ifr‘ermzt interiorritr’ fro-unzmcias were introcizced into the line cru a. test run on each :frosgioncy. The other toot cons'stcu of so . Tue cf CO H. H .J ‘ 3 H O " b *1 O .C ,4 pa 0 :3 O H. mcasurorvmts of the "noise" caused by differm; noise measurements wero "Lalo in the, sumo TILJ'I‘WI‘ as .jescribei in tile section on Foise T'easurevacnt. ”.‘his test loames con.e iderablo to tho judgment of tho ooereitor, but the I‘Q?’Ulf.“~ obtained arree very closely vi to other tests. The Howl‘s: of“ the three tf‘f‘t" a1"; soon in Fir“. 1. The telephone irate‘i’erence factor meter is or: i;;°t1‘1.l"701"t evolved fr’m this sexies of toots‘. TH": o‘njpct of tie water is to wwlyze a voltc‘c or cin‘i'ent '.‘.’;.VC, J: i L111“. the orciigstos o: 04-.ch comaonent :‘itogisncy multiplied by tonir interferrir: Isotors. She instruocnt is in prirciolc a toorro couoled voltmeter. Hie resistor clement consists of a network of resistance ii‘uiuctuico or“; cagaicitg so {Lositnm tiuit its coriittmwe at any freciuencg is grogortioncl to tie interferring: factor at tuct froc‘ueiicy. By deans of this instrument it is possible to ”recurc tne interferin” effect of any czrrent or volta‘e in a D1 er system. “meteor instrunnrt used Ior tne acne purpose is tn? wave enelyzer which consists of e fiLBJ inuuctence in series with c variable cupecity. This is singly a series reconent circuit with a telephone receiver as a aetectcr. Fortunately, hermonica are not essential in the accretion of a power line. In many cases they are untrinental chi the beat practice in power work tonic t0 the use of a pure cine Wave. The hither hurmcnicc usually reenlt from: (a) deeirn and CONPtTWCtiOn of “Wrerutcre an? flotorr, VHPTOUU cure sine waves are only ap>rcri1eted; (b) tde nee cf iron in trurcforWPrc unier concitionc awgrnachinc Inecnctic saturation, flmcryy causinr distortion of the curren' eni voltece waves; (0) tne presence 01" electric aired in the Circuit at? in sen-'3 street lirhting systems. 20am [no i a: 3 Rah 77y; nggbf 3 33‘ -u w: E. (0mg 417.5017 of fraywncg Wag/11275 Curves , Ifidd/hg Mgr/Md..._..._.. ' 1/176 (1/44 flan ---—- .. Mom filer/rad -- ------ 400 640 3m 1m (Rod mg [m ' * Frayac/Ic'g F/‘g I Balanced and Residual Couponents. In unulyzine inzmctive effects, it is conmrnieizt to divide the power circuit voltages uni currents into two General classes: (1) "balanced" with Desoect to earth us a neutral conductor or point of reference, and (S) "Recidue l“ compfletf‘lg' 111173. luncea-‘s 'it‘n res-tweet t7 tie "{,-.I‘t."‘;, 1.0., emeloyiiur the metallic mower-Circuit coriuctOrs, as a rreuq, for one side and enrth as the other side of the circuit. "Balancei" current cosponents in tie severel concuctors of a power circzit ere sucn that at every instnnt their algebraic sun is zero. The elvebreic sun of the totel currents in tue severel conductors of a power circuit at any instant is tns residual current. Similarly, the "balanced" volteqes f several conuucters are such tnut tieir algebruic sum is zero at every instent,wniile the elbe‘reic sun of tue totel voltcge to grsund at any instunt is the resiiuel voltage. An example of a residual voltece uni current is en overneeu trolley railway system with a rail return. Here tne total voltese and totcl current is residual. an unbclarceu three wire system or an unbalanced pease system with a Grounded neutral would contain a residual currert degenuin: upon the aneunt of unbalance. Sun>ose one side of a sin~le phase line is "rounded at one point. Then one side of the line is at srvuna potential ens the other is line voltuve shave creund. In this case the line voltere wouli be all residual althouvh there mivht be no residual current. Incuctive effect iron residuals are usually 0: erecter intensity tunn tnose iris melanceu volteeen are currents of exual memuitude. :38 ' ) ratio of effects free these tvo sources is exceeiirfly variable, reneins frdm about two to several tiouserd. The reletivelg creator insuction oroducine ,xmer 01‘ residuels is iue t1 tie fact tnnt tne residual cosponents associated with the several conductors ar e all "in press" enu their iniuctive effects cumulative, whereas the several balancei coegonents ere "out of nuns e" (by 120 degrees in a 3 phase systeo) end Hence their resultant inuuction is a difz‘erenti l e11ect i.e. tne in uctive effects cue to tne btluhCCJ components partially neutralize one another. Causes end Remedies for fiesiduuls- Unbslences or ineiuslities among the admittunces to firound of the sever-l conductors o: a power circuit cause residuels of the freiuencics present in the voltefies between conductors. In a system without metallic connection to certs s residuel volteee is pro~ duced. ”ith a grounded neutral system a residual current is produced are the residusl voltsee due to unbalanced line admittances is creetly reduced. anelenced admittunces ere causefi or: (1) differences of position 01‘ tle conductors Wit}1 res Prt to trovnd uri to one another, beire a function of ‘tne confieurstion, lUFl“ht above rqvnnii, locat ion or"nvnu11'.ires one otner' neishborin‘ oojects, chJ to a smell e tent, of size of COMJUCtDTF; LLJ (2) diffr1erc<2 in insuietion re,11tc1ce es be'be due to nefective insulators. Before the sub; fit of rcsicusl voltcee Wes fsttlly uncers tood by power congenies, i1; his considered soou greetive to 1¢L:1:1 sinvle )h&?0 tsp off ‘ron s hirh‘tension three guess power line. If tee sinfiie snare line is of en; {rent lenoth, the especity to ground will be uses Very much higher on two wires than on tee third. 211s Causes e lcrgc residual voltuge. If it is necesssry to nuke a nunoe r of single phase cruncnes, tney snould be so distributed as to cause sinimun unbulunce. Trunsposing the conductors, anion ten1s to couulize their relstions to {round one to one another, is an efrective remedy for unbalanced Capaci- tance. Such trunspositions must be located vita QFOJQT reward to changes in configuration and at short enouwn distances from each steer s3 tact tuere is no material eifference in the electrical c ncitions at two such points ut a given instsnt. Of connonly occurinw confivuretions the euuilu ersl trisncular is is nest nearly bulsoccd. Fence causes the least residuels due to Info-c lextced' capacit:nces, uhile the JILUQ confirurctions, especially the unsweetricsl horizontal, ere the worst in this resoect. The reuesy for untclcncel inau- lzitior. resi'itnz‘me lies in c..re1‘ul ;m-int'u'.z.nc~; for a. 1.7.1. constcucteu Lin-s msintuinei systes tgis is usually not an importcrt source of resiuunls. In a power system unvin‘ loucs‘connactea betmesn tn“ 'evercl con— ductors and grounu (as in a stcr—connecteu Egrteo tith grounded neutruls) differences alonf tee logos of toe severul onsses me; cause resisusl voltsg-ges 01-11 C‘u‘rfmts, cue to ‘onrt .71" the 10:34 bc'n." (insigoliuu 1121131153}: a. circuit consistin: 01 conuuctors witn grounc "return." Ineguelities 0: ratios or inoedsnces uuon“ the trnnsferners o: a bun: also cnuse resisuuls in such a circuit. The evident rowed; is onreful ouuslizstiun of lease, end ~— the use of lite transformers. neuQV¢l of too urounu gets fox unbalanced fl lead currents, silo-fin" only one neutra, E‘I'Tul‘ld, is the roost criective no reliable remedy for this saurce of resilusl current. Then a trsnsfo"mer bunk in a three phase system is connected in star with neutral grounded, harmonics of three times the fundamental -10- frequexcy, ard 01G multinlnn thereaf, gqfiear at reci:ugl¢, on tho Prouufiéd neutral ride. “his is bncanse of tle VariLtiON o; t10 garmeubility a: the t1‘..1:3:'ori.uzr iron it‘n Vu‘yiur *xlngimtic :3. Nisitr, Cumin” hamzaorzics in‘ tr; 15:— former excitinfi currdnts my in tneir ingwceq voltujnr. gr thfi triglw, Inuihniic co Lxxux. s grw "if; finite" in t y) thrgn trglflfl7I31“TC, tri.flgilagrmonic resiuual voltgjes gni currents gre grouicnd i: thQ neutral is groundei. DO tg-CORDOC*DQ winsin~s on nucu a trgnfornér buLC prdviun p shutt yutn for these trigle-harmnuic Componwn‘s o: the excitinr current, qu greatly lesson thfi residuals which Jipht otaarwise be cuuseu Oh the Wrounge; nautrwl siJe. U) ince the magnitude of these TOSiQuplF decrngwa very rcpinJ as the mahiuun maxnetic density is reduced, lownriz~ the volfuve imgrensed gar turn on the trgnsforvmr Rindinr, or substitutinc trgnrforzér: of Iowa: Wtjnetic density, is a very effective reLedy. Isolatirh the neutral of a tr.ns£0rwer bLnk eliwinuLO" it as a saurce 0f trigle-hgrmouio resiquble. - 11 - Effect of Cmn‘i min-Jinn on ’Iolturn Bglute. T210 L;1“1‘.;1T."0‘1‘§Ilt m": r__xu.cir" 0:.“ f n (: 111:.11.(3t’);“f‘w n.“ _‘)’T~.’J"l‘ Circuits arm 0;“ iYT”)OT‘t£IXC-‘. in \‘2-’?t“I"'1izil"." (1) the TJI‘L7.)L.1L.I.C"T‘~.~ 31' inm‘uglitihf‘. arm”: ‘ the Canaciturzcw 0;“ 13:10 COIZGIICtCI‘C to “3111218, ‘r'UiuI‘. CLILTV‘ Toriitul v1 t1;"f‘S H.113 current“, Lam (L3) tin) 1:32-11 "it; of tilt irpxzctivn nf'factr :‘)1")<.'(73_(""‘.C} ii.- enii'r‘..v.1':ica..tian Circuit? 1f tin It- 1; :1th '.'r)lt:_.'*"““ Lu). (:‘.1;‘1m:;t‘.."~ 3;“ art-..llclt, CORSiJHI'uLiOfl of t L‘: 11".."Ctiw'i‘ 731‘L'8Ct,“ 510111 in pmmu‘tl C out :31 1';.t.~‘u=r tum .. Cims‘i A'YL'LJCi’JI. 0;“ Lam; (.g. x.citg.1;cr:1xii-“.1023. i‘oi' sectians o: 111;". Witt-1.10 the li.1its 0" pgrullslc, caxuiimmtion 31' Ci._)a.cit..f:-ca ubulmicr‘: 77?;01.L\J be gjfivrm the ("Segtal‘ '..'r3i--’.-'.t, paii‘ticulgi'ly for Cii'uuitt.. 0,)»3129: ‘ it..3ut (grnum‘md noutnlu. Ii" tziere were no ii'r:)r'ul;-;'iti:’:s 01' il‘.‘)..-.uCtll‘?Sf as: to inguir tno BinCEiVGhOSS or c trtntfiasitiou sgsteu, it Youlu be gwssible tumorotiCally, negrlectiz'fi t..e effect 01' git-use (Linux!) 2.1. 1 ;.ttenua.ti)z;, to Obtain a pm'i'nct bgluice by mums 01‘ tr;.r.'31)33it 101m, irrespective 01' the i‘.I‘1‘LJ1j‘YLZ‘.‘-Lt oi“ tne conductors. Fricticnlly, hvwever, circuits even when Calefullg tranGnOScd hum a T'Ltféi‘iz.l rasultm- 111-1); 1;.1100, pgl'ticulgl‘ly a. tin, freiuencir-s 01‘ the 111321? Titrtrmics, aim; thin 111:1); 11.1.03 i=4 proymuinmd to thr‘ UZCDLJJLLCO chm: ctnrictic 0:“ the Circuit cor.fir'u;‘:.f ion. In a similzg' TRAIL-"‘1‘ tho ro— sultamt i1"; ‘uctimz (-Tun. to L '_.)d'r.rer Cii‘mzit is _ar:)10rtim.ol to the intransity 0:' tin iiimctin'n C"1{..':';.Ctm*iqtic m" tho. cari‘i "1111-11 on. (7"!111'1'11I‘QC'LOIIT” differ v'i i013 in rrvzrmct to thr-ir c-1L-_a~a..ctarirrt ic url-L 11.1mm Lit? i:.t":u~itid¢ F}; d. :-. ~. 1 -: u t . ' n ¢ -. - 1. "v1.. ' .9: . I 1‘. ‘u‘: '- an LILLLUCLIOI’I, 5.6-.» untr-‘n‘rnt ).,:I"°ulCL‘-J.;J.L- 0: t it. C110..its:, m up: 0 fully 90;) 1533? 111:1); 19131008 01" i1; :ucti or. t‘-;;..-i 0t Via-t. -.1e (iii:ect.c o- ti.e ..rit.n ‘er ext -3 .. f;).JZlI.' o.. (/31 L t‘w 7 oz: t:;n uiux,1c:xree o; t;thT c;.>uc11x4.crw:‘to "I“lu.fl LdAi on f;ie 11.;nctirxi or? muse; ii: “I~ 1 ' " ' I ~' I c A v: r ' 77 - - . x . .guicilixil (30,LnthCK.tl(HL c1.rcv11.: ;L.e k} “CTN 5,,, L’ iciuitel‘.. Jffect on CaoucituLce Vubulunce. In mnercl, the capacituhcee to rrougu or the c Louctorr or the l’iOll-tl‘éJiSm)1)?)(-3d multicoimuctor ci reuit are iuie.“.1ul, tile 'iagfz'iitiztze oi ta‘: percentuyre Ulib'ul'uZLCCS bcim: irate-amine; by, uni t.mref')re ch;.r. cterictic of the configurction of the circuit. Tuis churucteristic nubalurce is an innortunt factor in ueterminigz tge reeiiuul voltcye of a cirzuit isolated from ground, and iLdetermininfi tme TOSihuul current of a griunueu neutral circuit, in so fur us such curreht i: cauccu by the lite itself. Tcking as a floccure of tie cucracterietic unbtlurco, tne resin‘ul voltuee of a short, uniform, nontrsinrnnr‘nj circuit "ritiout VWtLlliC corn-fiction to around and enerrizec vith bzlurcnd three jhuce volturec between Cihductore, termed tflO characteristic residual voltcwe, the folloriuv table cffordr a countrieon of varioue confifuratiohc of "incle -circuit power linee ore" the prccticul range or erase-cectiorcl nimcu"i3ne. Characteristic leciduul VoltuHe; Percent 33 Bulcuccq ihree Puuee Voltage Between Concuctore. Configuration Percent fiesiuucl Voltume to 4 in" Equilaterul triangle ----------------------- . VertiCul ------------------------------------ 6 to 11 Horizontal Symmetriccl ------------------------ '---— 5 to 9 UnsynmetriCu1---— ---------------------- 7 to 11 Configuration Percent Residual Voltcre Isosceles triangle Buss Nor'zontsl - ----- --- ------------ -- O to 8 Base ferticsl ----- — ---------- - ------- - .o to 9 urn _______ __________ _ _____________________ _ 2 t0 6 Inverted "L" - ---------------- - ------------- 4 to 7 Triuneular circuits huie tee suallest unbulcnces uni characteristic residuhl voltuves. Symmetrical horizontcl cnc vcrticu cirCuits are about alike, the ver’ical havino slivhtly the PTQLtQT, and unsyflnetriCul horizontal circuits have the largest. The characteristic residual voltcccs of symnetrical horizontsl and vertical conficurutionr are from 2 to 8 times that of a correspondinv ecuilatcral trienvultr circuit, deoendiac ugon flie Spacinf uni },i~it of the congrctorc. Inc anorecteristic resiiual voltujes of unsymmetricel norizontul cirCiitr are about 20% greater teen those of symmetriCGl horizontal circuits. Eney may, nowevcr, be reduced to those of symmetrical Cures if toe position of the intermeoiste conuuctor is al- ternated so tnst its averure oosition is niuvuy between the two outside Cinductors. (If the circuit is trtLFo080d this conuition should be fulfilled in euch section between truispositions.) The cnaructeristic residuul voltujes of ecuilctercl triangular circuits are ClOSQlJ preportional to the conductor Spuclrj, but the concuctor specinfi has but little effect in tee cases of vertiCel and horizontal circuits. With twin-circuit lines it is Jossible to interconnect the two circuits so tiict t.1eir imbalances tend to neutrclize, ..fivin': smaller resultant unbalances amour the Caigccitmces 01‘ pair.“ or‘ interconne‘ectec conquct are then the 11111); lunces among: the cenductors of imiividuul circuits. For twin cir- cuits of 811;! tgme the "LJZiE‘TUI‘l 1u-b;.1a.11ces occur "hen conductors syvietricLlly located "'i tn respect t1 an intermediate verticcl plane are at common potential. This: girl-eminent s}.oul:‘. be avoided ir. all cases. For cirCuits e?" tn“ vertical tree, or vi tz1 to 1 LIN! lowest con- ductor: in t. vertical nlzize at”... hiwile cera'rmters displaced O'azt'LI'd a stall dist-.4- ce, 4;.1e ainimuii res1alta..r;t 111‘11».jl;i‘:cte.° a re obtuiz‘u‘éd ..11' the. to 1 0'71:- uuctors of tee t‘r") circuits ere ct co ram: pottrticl are the botto-q 311d mid'ile coxmuctors rcscctiiely 01' tne other (iiree Fi-;. 2.). For trisry‘rulsr algd morixontatl circ 1it :1 tie ninizru'i resultunt unbclm c“: are obtained ween similarly plcceq ceruuctors o‘ econ circuit are ct CO"12'.‘.311 potenticl. 2-10 result-cut un‘oe. lances i th. these L:I‘1‘;.1-&'°;;‘1(?l'.t3 are in $07.10 cases less twin 10",), am in 'écnerul less than 5),; of mosc witi tile worst Concition describeu above. 2113 arrangements irci 0;..th by Fig. 3 rive resultant urbclzmceo of the order of arismituue oi' those or“ sinrtle-circuit equilatericl triunr'fulur lines 01' correspomfiwv: C’HLGiICt‘JI‘ s )ucim', :mile tnose of Fie. 4 in reneral eive .rtrecter ui-ba.le.nces. I]: all cases the churucteristic resit'zal voltawe is taken as the nee-.t‘ure of tlie unbalance. "Were ground wires are used or in cases where unsyn-‘ictricel cir- cuits or more than to circuits are involved, 51]"!Qcial study is nccess'ery to deter‘iine t‘1e beet arr-..rwcrnert. Xitii twir. circuits 0:" wif" cor::"~1u'a.tior. if t'm iz’lt"r(.:7:.n’,‘CtiOIl ’T‘i‘lin" mxiwun unbalance be cltvz‘ 1);? tr;zi‘-“{'1O-'*izz<' the. interconnectin" '.'.'iI‘”‘,F§ the uni», lance is halveu. The two horrible interconnect 1021s re"~‘.1ltir:" from this uroce..i:_1re are choir: i1: Pic. 5. ”3‘11", elem is useful "‘1er, té’mre i"- a. doubt as to the best urreueewent. 'Jo obtain tie "‘.'I‘(_‘;.tv’?.°»t shunts re of ..rrcu‘te‘z'rxs giviue; small 1u1bclczices the twi1. circuits should be interconiectcd at ’aetf-l ends or" tae line aim at i1-t9I"."lOLil¢.t-'3 s11‘ost._tiozzs where prcctic-eblc. In cases '."l"‘.‘I'Q twin circuits sire Der: lleled on tim station nice of tr;_..11s1'or-ner beutr. but -till LQVuhtcfleous to f'x ff) '0 cunno: be intercennected on tne line side, it i the guise relst ion of the confuctors us: if true; were to Uf.) lintQI‘C'WIlILCCtQJ minimum uric-t. lune es . H) O ’1 Effect on Incuction from Buluuceu foltujes cue Currents. Tie type of power circuit pl‘Ou‘iCiLf: tir“ l: ust iruuctive effects in a. gerallel cewnmiCution circuit Liegerids 11301- the singing" of the con- ductors and tue se'perutien fro": tzie cor‘i‘nmicaxion circuit. Ir.- "gmcrcl, far (.11 tgmes of circuit, 1m ircreaxe in the S‘LCiI‘.” of the jmister conductors causes a ‘})r')nertieret." increafie in the ncrrituie of the in net ive effects. ’13:: eseive speck-" "haul: thnrfere be avoided. 01-. the otrmr Pt; I':, armle cgiacir." to erevent short-circuit!“ or "r211 rue, cue to "res, “1:11, "iruF', etc” is essent ia.l i’rou the "t; 1.2: mint ef' inguctive i3.te:':'ere::ce, as? tell w free taut of pm: 1." service. For line." segurutec. by the w‘iit'ji )1" an ordinar; Iii .1"';.;f, a. verticel t;r,;e 01' finer Circuit , in generel, cruise tie sz‘tllect i1.u-‘.cti 1:1 c‘31'1'ectz‘3, while the 11.31“i:..z;t cl t; 1- 1' 11.11213 tit: g:r«é;.tes—:t eficcts, t’iC trieruder ggws beinr i1-tv:emeci:-,te in this reflect. 2.11 rel-stir} :‘1' rite. -)_r_ r._1i1erel.t cou- I'irrurutious var" s0“102.'-u.t ”with. tie sen-..rction of tile t" o clusscs 0: lines ., d .5 v‘r’ 22"," +i‘r‘r .‘°.',».-.....-$«w~r‘ O" *711'3 ‘O"-"Y' C.'r‘!-~"f -n (“v-w. "‘ "1‘73 " H’)‘ ‘ ‘7‘ abbu Ill JIl U41. ,1 Lkl- A /¢--' ~OAQ‘J - ‘_ UAA‘-I ‘J '1‘ 1- s4 Li J ’ k‘ I:) 7 A‘ku 111- h \ ‘. L1 1 .1. bll‘l relatifiwa hcyortzrzufi of tPM? btlcmmw3i valtithi:iui Ciu1flfints induct ion. For low-vqltcfc hariznnt;l lines, 13,01J volt: or metrical arr njéncut o: :39 cancwctcrs 1. . , ~1 : '- . . . .-. agwwggiqncht. iwn‘ linzc 0: titf‘valtcfc, ‘ better tibfl a if an 11 1111 iuir;'3.V7t1‘i(x_l . r“. 1 . ‘ - ,“. ‘.'~ ‘ .-‘..‘ . ‘- . . ~ 1‘. «If; USGu, tnc 1Ltcnceuigte CnhddCCOI 5.1ulu b~ Ulfiglcflflm tTwnlu tidn circuit. ”GCCC, unsymmctrile horiantzl 90c“? circui Show}; hcvc tin itrzinrrficdiait” Cfizzglflctnr 02:00.6: .... .— 1- “ .3 . 'a a. ' . z - '..... . . “121 trv‘ r3: : * 1P1“? c0 r inictytic:: clitrt Oppocitq ridc 0? thfi rind. possible to intcrconnéct tan coniuctorr '4, L 0i ‘1 GI“. H11 tlioii‘ plus"? rthiom w!) t"; t a. partial 11cm; T‘ A ix: ;)I'O£J(Cliii f 193:, c C FjQJCtrngl arrturtfi-Lt is tne COIEULiCu— is 21101. : hiclrlgxcjxs ‘. O ‘1 . — .- tuc 71cc of fine 00i«¢ tn— VVuux ttzo OT‘T‘OIV? F:710HITM;iIV‘ cizxnzitr’ urn» cult ti? circuit: clizction thGS Ifl¢dhdo 30? tvin CjJThiitF WY t izxwérticcl 4-.. 11 1'"... f) I. . lncttci Oh *;c in one linc it is , *u-s, ...°,. - ' ")1 '3b11’1‘..lt'3 I K o: the inductivn Offcctc . ,.,- 1' LL lowest coniuctorc in a vertical glcnc and tJO niuuic cancuctorn ui?glLCQd ‘ +tr'.‘ 1 y ‘ ~ . .‘1 I " w ‘. - -.‘ oat cra u Siull Linthncc, t.fi most 19:01. “\1‘L3 L'I.~ ' coucition it i L ‘cncrcl, to have tim (ii;t‘f..‘1;. 11;; 099051’69 (301:.-.ac1nI-s ;-.t (2011.131. liott‘rtitl. Linc Ric. 6. For CirCuitS of otnwr tQJOC tic watt ‘1 4-». ‘v' 3 12.11 1 ’\ '. t, 2’? Lgoi of CDLLOCtiOn VLriOS witl tAO FQgCihfl an; Afli’flt or the power concuctoyr -Lc wit; their JOSitiOI relitive to tnc cnnmuLiCction Circwit. tn civc ; rcrercl recoumcncctian, FthQ specific case to detcnwinc the mos] LQVQ 39901ul study is alto rcguircd far lines r"l‘-. -.RLS l 't if .‘ A not _xx‘3iulc special stuuy is reinircu in each m a (‘1') th" H. 5') “a YV?t?10;. 9‘" ”'101'9 Of L', 7‘ U..L\-A$ interconnection. t“n circuits of thc sawc of diffcrcnt vnltcwc‘, for unsymmetriccl diublc circuit limos, -17... re: arc HGQQ. w- and in Case? whfrc crwnnuca w Recowucudofi Parfi~urtiions. ”chin“ iuto account notn affectw thvc i’lf‘CHFSP d and prcctiCcl COHQidcrctinnc 0? conctrvctifin, th DuUiILtflTll trianwulcr confiflurctinn I . ’ .1 ‘ \ - ‘ 1' l 4.“‘u v. “*‘v' . i“.‘ f ‘ I',fi ‘ : "‘V‘ 19 1:1(LW'czi.-:rOCO’zvuiiai . u FlIL.1”-C1i(-Jii {)3w 1 liiw ; '1 +11 "“3 "11 u 1-; .‘i..\_,‘ ’3 '3 g P H. - i ,4 i f (’1‘ H J H A H ,3 (3 ..J J p-J b... n V .1 p .1 6 C+ ~ W 3 J “a (3 O H 3 (1* H! ,J ‘3 c'.‘ r—l. J ”T ‘ ~ “"I 4- \1, ‘\' “ 5 *‘fi ram vertical picie 0i tu‘ cu C - F A F:- S O (D 0 *4 c“ (f if “1 Q 7 C I‘ *1 Id) Circlit pvavr lin S. $Kh tho vortiCcl C' “31:13:31 0."? cr twn) for twin- 3.. athwoé§ uxtuéék. 3.31.23 Sik‘thv. nutuxvfle§ turtle». 35:3ka W .WQ \\t\\ Into FEEth‘ —_ w._ — —————~ __ ‘uthvat3 vutcfi‘uclcb £35.3wt v mT _ m 9.x _ _ TransgocitiOLs. Ono f the most thuoble means of overcomixc incuctive iLterfcrence ~J under noroel ooerctinn coniitions of power circuits is to transpose tne coneuctors o: cuch circuit, 90 cs to ciualize their relationo to cll other circuits chi to earth. Transpositiono in a pore: circvit tonc: (l) to eeuclize tie capacitcncco of itr coniuctorc to crouni, tk'roby renovirc a so*rcc of reoiuuol¢, u;o (a) to cwxon the inductivo effects from too b;lonced voltawcs znza (mirrorixz to lflflltralliflé oxrizinothxwr in rvfi.ilhorixr* lcrrfthc rif c grow llel communication lino. Tl‘--ILI’-)OSiLiOZ‘.-S‘ 01‘ a. pow-3r cirmit coco not reluco induction from rccicuelu, OLCHpt go it wcy to co indirectly h‘ a reduction in toe nefnituqe of too residuals cc Juct Lotcc. Drunsponitiono in c c0“muni0ction circuit tch; (l) to euuelize the Cepucitcnccs of its conductors to around; (3) to lessen the incuction among the severcl communiCction circuits of a line {in other ROIQS to lessen too cross-telm); arc (5) to euutlizo the incuctivc effects or the two sides of the circiitc, due t3 necr-by power circuits. Sues treLSroitionc do not protect the circuit excitet VOltLjOS injrco; between the circuit cs a woole Ltd ground. Fortunctelg, however, c FOSidqu vol'efie coes Lot cause noise in a communication line unless there is c series or rhunt im- pedance unbclcnco in the coauunicetion circIit itself. 5" ‘ ~\ It nicht at first thouéot to fiiooocec that i: two circuits were transoooed once ‘.'i thin the fur! llcl and the 01 rcuitc perfectly: belerced in ..- thewcelven, that ell indictive iitcrfcronce woulc kc neutrglizod. Thic‘xould 1 be possible i: it were not for the fact the. current Ln; voltage in a com— mmiCetion circuit are attenuatmi 2;? t? 3:! pic", Lloi "j the Tire. glee it totes a finite time for electric wave: to tre'el clorf the conductor . Thio weeis t;at current in c COKulCtOY is out o4 thco a certain auonpt with th‘ C'W'rw'rs t :“ tr.r‘ c op 1:1 .x - '4' 'r- l 9‘7 ‘ - "\I- 4*:er 1'? "x "-"n 5"- VHF: V" "' (‘- -~u J. ..‘ J...L .x. x L... I L‘-l . ta ....).. .15.... -.. I ». Jul“; u-A : 1‘4250 .. CALQ.-~ ' 1.1 “4;. ..- nitude Verieo from 1p to over.1fi per mile en; the chcrne in phese rclgticn may very from a very smell mutatity to as ni"L as 15 degree: per mile. V C Siuce it is inooesible to c whine Use vectare vaicn are not 130 de“r¢es chart and rfet zez'O, it 1? i"1.)o:iei’ole to cozt‘ei';-e tx'o incucti‘rc effect “nice A are no: 18o ce‘reos out of geese chi cat in a bulerce. Tierefore, in any trcnsoosition Hystgn there is u certcin unbtlutce brown “8 a "tgge unbalance." In general the closer transpositions are to econ other the iore comgletely is a balance obtainable. II. orcer 13;;qu tramsgvnit i015; skull '3': most Ji’i'ective tn 5:1: cant be carefully loccteu, retain section? figure the intensity of the iniuctive . effects is: uniform, ‘."'it;l respect to paints where the inuucti on chmgyx, ‘\ “ 0 I \ celled uninte o: locortinuity. also conniderutior must he token of the discontinuities in tie iUJGdLuCQ of the tele)doye circuit wtich in the 3011 sypten are loottei 7.38 mile: avert at the loefiirfi coils. The trencpositions in ecch clad: o: lire muct ulro he 1oCeted with rcccrd to the trzrcpocitions I‘ O the ”)til"? clz—mc o: lino, 1.0., the trei‘soor‘itlouc i.-. tire go'i‘ 1.111 corrauuocti 01; 1111",? Hurt of: CO-OL'QH-£.tQJ. l’O‘JC'I' Circuit ‘31“;1';S‘:‘.')()”-114101.57o .7" -. o .- .,- ' t- » . l 'u ’." ~ . blll?10 iliuse lm7101* 11;;39 at:' o”; trolkfi)3thl it all) at L, 1L,thr as a telephone circuit. A three ghe e (l1‘”i has three cifferert errenee: rte of conauctorc on the gin poeitiorr LLJ there ere two eirections or rotation. One conglote rotctiou of 2&9 pover ceneuctorc or the gin an”lt13“9 i? celled a berrel of yower. In considerinfi the severe rertriction pieced or the selection of 130109110113. trcris Jositimi tr 9 by ti re necessity of ttlwcizr' to LI‘L'nSpOf—md three-pnese power circuits, toe followin? CJLcitiOnS me; be noteu; 1. If the berrele of the you r 1iLe are very letfi, tu‘y oer be taco to coordinate witu alnomt all of tie t;)iCel errurj mzints b; having eucn non—trcnsposed section of power circuit, tlit is eccn tnird of e barrel, eiucl in leneth to one—hclf or ore-Iourth of the teleghore trensJostiou section, for meet of tie typicel erreng"meuto beluuce within this length. As the pnver Lerrel is decreased in leanh it becomes very difficult to d0“1”h e tr reoo ition 3;.te1 wriich will in e neutrel point? clove eunuch toeether to meet the reeuirewerte. 2. If tn? iflrel of mrvru'lir” ere vecrrjuwt, thfif‘lg'kVT'Ld“ to coovoiicte . iti all the t wiccl erLLeaueRte by havire complete berrel betweent ole! one t1cno>o iti01r. Tow ae tue lenfito of tie go or burr 1 it increased it ie iifficult to 00310 utie mete transposition it the teleohoue line if toe teleghone trenepooitiOL io fece'es lxnr"aie th p? or bcrrel. The M xitjun length of a flower barrel using this o“:‘te1 01 co-orcicrtion is about 1 or 2 miles. If there were only two or three com. iiccti on circuits on a 9010 line, it‘mauld be a very ecsg Lhttef to 2-1u;:e the groger trcnSpositions. L. loco circuit must no trcncpoce m ta recgcct to every 0th r circuit in the group agavell no to an outside oower line. The problem is more complicated on account of too fLCt that nearly all circni‘s are pncntaoeu fink tue pucrtom circuit must L133 be trunsgosed witn recpect to all other circuits. Previous to 1918, truLsponition o;ttets were not cusifnec to t;d0 into account 13.1".) inductive irtcri‘erorcc oz? outrioc 37.1'2’ lines. at this time the Exposed—Line TrurSpO%ition Sgstom we? desifined by tmo cnhinecrs of the American Telephone and Telegregh Compeny out a discussion of tirin GOSi’fl of this cystow will now be token up. Decivn of the fixposed—Line Truncpocition System. Before any attdnpt we? made to colect, from Leon" the typical arranqencntc of transpocitions vniciictm be uccd, a tnbl“ To? ficlc u- of the limitin“ type unbzlurcos which.wouli be p rfiittod between pair? of circuits, aosuninf tout a Niven limit of the ininccc cnrront Net eon cir— cuits Connot hm oxcceueg. Tie grococure we: on follouc: 1. A Computation ML? ”L40 of the total u:0unt of unbuloncc which could be permitteu between each combination of circuits on the lead, taken two at a time. This unbrlunce fies exgresseo in terms of the lcnnth of untrunsooseo exposure between tie circuits unich KOUlm give tue limiting Value of incuced current, no curyent being 'neuced in the rest of the cir- cult. 2. The total ullowcble unbelance for each ocir of cir'uits wee r’JUCOd by a figure representin: the unbclencei11ich curt be expected to LTiSO from irrefulerities WitJ the type of construction to to employed. -22- 3. Free data rergnrdixr' tie attenuet ion of currmts along; the teleohone circuits, taxi from the total limit in; effect of type 'LlIib;l'u.llC8, a. computation is made of the limit ing: type unbeluice per tr;.nsposition section. This sets a direct limit in t’lO relative exposure t;.r_>e between each pair of circuits. ‘J'ith the information reenrdinr' the reg-J irenents of the prepos ed transposition. syste’: prepared as indicated above, the creation of a. G’ESlffn which 810111;} meet the regairenerzts of next-13m econony ad ef‘ficienc;r was a problem reiuirine a. wrest deal of skillful and conscientious wolicntion on tzie part 031‘ a. trained expert. The problem ms 02‘ course one (if selecting: from the tygicd. wrenszr'at‘ieztts '-.‘.‘i'liCIl cit-n be * see, the best tree for each. Circuit such ttht tie type mibalcnce between no tiff.) circui‘ s "rill exceed the established li'sit. There were t‘rmnsums 01' possiole combinations and it required the merit of t'ro men for one year to arrive at the best design. Considersl-le credit is cue Tiessrs. 1.. G. (nuptial. zine it. 73. Leomiu, mic perfected tne fine thocs of systematic Pearce aim carried out tzie design of the E and L sections. Q The results 0: this research we: two ti signs :leOWIi as the 3 and L sectiOns. "Jim 23 section is designed for an 8 mile length 01' line. The L sect ion is desismed for four miles. Anotzier section imam as the I: section was also desi{'ned for :i lenrth of 1/2 mile. Any of these sections may be suppressed in length as lone as there is the required number of transposition poles. ‘This is often done where lines are subjected to excessive exposures. For instance, in the case of a Sifinile perallel, ne L section ant: four V r O I sections could be 11901 or en T“. section could. he superessed to six miles. The preperties of the 9 1. ilenent of section in 'MiiCh each circuit is transposel at lrest once 1.9 ., balances to a nu‘allel non-transposed disturbir? circuit. 2. Neutral points suitable for tne locntion of trans- lrjsitions iiilxl Mullelilgf three-phaselxmer circui s, segsrution of sgqroxinste centers of ‘1st1rbinf disturbed groups of more then 6 meter .511 U " n ‘3']. f7?» 3. Leutrel points suitable for tie location of trsns- positions in perullelin- metellic sin? circuits-—sep than 6 met era 0 punse power uretion more 4. iautrel p0 for tile occu rune continuities separa- more then six meters T’eut rilln ein ts sui.t4.ele fer tne locn We of tr‘.s- positions in Cistntbinq Dower circuits—-sederntion 01 0 less than 6 meters 6. Ventral ywixts suituole for the occurrence ’f discont'nuitics ir tclcqnone cirenits. “Xlluudulq of , ,, _ 7‘ , . , Tie . use a seetiors ' t v, . A; ‘ ..... -‘ . 1.11 9.4- )IDS)C\A “rt-3Q ‘. .511]? exposure and the drawinfis this line There are one drewinm. Combine she? four tjqes of 4'7lcu‘10 U; ~l~ J t i one. 01' and L sections are t section Sta Seetien points, Baiuwéls :1nviiq; nominal lenfth of 5 miles or 6 miles an be used. also 1/2 mile barrels cen be used if the parallel covers two Btu sections 8th secti~n points, tulSO 16 section points if pernllel cQVers two Std sections Sta section )oints id section (VI moles ". .1 .‘l'l-i 0-“. bot-rm )1 two J_—: 9°C -LOIF section or similar ;;rts71ig's. - T - . . ‘ . K‘s ,r‘ . 1 1 Li. J"? 11“.: .41 l .. .3 .. 'x‘l 'Jetreit-ZulaHLZJo line is subject to n "I 4‘ .2. 0‘ I r‘ H1 31".) 4 £1.1in .‘J SQC’elOZIm I ozzr tgr was 0;? CFNIIa :LnIit.)1 tr .11 par: ition ted below: L Section Quarter Section fiunrter section 5 Vile barrels Ccfu bf”. 1qu! cugzter section lJ‘DiILL S guerter Section points . ‘1;;s=ection 3 “9010 S énl of great u: hnve been cut in on itictizlre Efllflqll in nnnton trsieposition -24.. primarily ncie up two sections 1~*1:ned 101 Some lin es were oriv'rully tr.. 3 System known as tee J section. Tic trcrs 1holCate Wmere tic or1gl1 ”p.13;- (.1 I‘ 1.‘ ’11 J 2]. Ir'élt. -0 -ZIL': L, 1: 4‘ *1 re ' p’r‘ect " O” L "(C 101' "3 “(Pt 0‘ . .9 . - , . 35 to 50;: nErector t:1z.11 taut 0;. too stanJch Tvsten for non-e:)oeed lines. Vonever, it Teleohors ”oniary to inetall 1 and L sect i"‘13>03‘t1r_* toll li:y~r‘- “'P'mffl'ui'r 7‘ ‘1"):‘011 or not J p 15.11110115111’1023 1ul tlaLsoO? itions are ion fortgrvdrm: lines. for a XOI—OLPOSQd line au in L Circle tC) 1m} c11t oru: i1) 1‘ t1 ')l‘L. exporcu lire tr nsgositions is used by the 3011 -1’2111111016 is 110'." the 11311st ”1:" tl1e fiell trzn ooritionr on 1.11 their to tiara ;-"1.ir:"t i'i.1t‘.‘.re rugiofiir'e't. Tr;1810P1tion"cirtfincnce. I‘l’u...to1 to f'lxi'? cross—t.,l‘.: .11'111'” “1.1.." 21-1.; ".’7) l’)“ 11;. 1113??" of CI‘OSI’3-t..ll: ‘1)e?:t=.'1'-""21: 1212 11w1t )"I 11:11; 011" 01" it.’ sic”: Circuit" ;.‘I1‘1 L. 11.1;‘1‘1 v.luc between t11e": l£111.11tr31 L111; 10 ot‘: rile ci11it Eiuving 1111-11 Cross-talk, Or the onission of necessusf or iv“ inre2tion o: unrcceSSur transpositions in tc1u ‘ile circuit or to errors 1; the tgpe of phcnton transconitions. The most cougon error of this scrt it t11 rut rcnduthn'type 2 uni 3 junta”) tr 11:11-Ositi \f11'1cL1 inserts c1- and circuit course,11nb:l “ 3: both ‘0 ‘ - _ . a. I {‘1 811ect or C1oss —tulk is wuc Tectir .0: 0L, - - . . 1 -1.:- '2 .;. .- 1: "yenercll‘r true 4.31.1, 1.1 t..1e Site to .'~lL cue to transposition irregularities. . .‘ - r‘ ,‘ ‘ ‘ ‘- ‘I I n Irecuent cc1se or trcnsp0s1t1 so incomplete sections and non—transposed occur most fretwwr tlr nec termircl or eaatrm1 trc11 03:1“? Fig.0 (:1 e crovs-tclk if hihh, t on unbalance is ctiors iJLt-‘rrré1:i;.tfv tf‘st FpOMithn in One rice 'M‘:uit. julS, of o tire 4x1c3137r1. Jllt f1wetii1-1 tl H" t11e circuit tins; for the ctuir. It Luéiudxlarce the occurm; ce of cnc incomplete sections 03' O 7" ' ' -25- where thorn 2.1m short lr‘nr'thr- m” 112%? lr*;-.ii:.r" from. fish warm-t "L3" pole or 1‘aiill‘f' point to tho officr‘. or 022.7319 “931.0,. In a for Gad-"OT? tx-sn diffflr-‘mt transpofitiml systems 0:: tin 9251-“ 9019 liar" omm‘m *.';7-1>«n tho cum pole 1111-92.". 1190;} by t'm 031;)».111‘71‘. “my: (101';ziti._2f "iv" rife to considerable “" . ‘l. ' "' t - -- 1. l. "4" '\ .n ‘ 1 “ 1 . . l . ‘ ""9‘ - ‘- "y“' cram-1.2.1:. mum txaa ~,31y;,.121t:.. 1.2 e t 1». sma- 30212 212.3. it is w an} “*T‘liudl‘? ’Cnut tad”! catachrcte: so thgt. 31m t1‘¢L-;Si)-3f'}it ion 1".“E‘vt’T-QL.“ may be 0001:211thch I» L 45wN~ ~bu- szw~ Auk‘ TY/‘Y’J 0/" MOI/”0” 7' [I’M/51°05] 7 I040 >% 2 73% I % 79ch ' % 73”} % 73m" 5926 16 l6 1’? J'fC'f/O/V JECWd /V L/A/f all 1 C‘x‘M/P 60577994605779 fYP/C‘A L DETROIT ‘KflAA/W/IZUO r v ’ i _-_|J;Hf MUN-“ III! III N Fig 2 7 ®4®0®81nne LNG intollivibility of L converfction Caugad by love in the trinvwicsinn 11h? in a gupxtity vhica Caz be verv aCC'uI‘utnlj' ‘1“2=.f‘.11;"‘+:i. L‘Zw unit 733' “7"};ic7’1 thin 13‘ "1‘ Lifturnj is tsie mi 10 of rtzn 32233; cabin attic-71 11' 21023171ch as 2-. *3110 of 111m 'L’V'i‘rz" e rnz‘i“t;-3zce of 83 ohiv nfir 100p U110, ; mutual CLWLCit” a? ,054m.f. n»? wile ;LL at attfinwation canvtpnt of .139 at 833 cyclnc. The trznsflictinn lair in 1-38 01‘ ¢t21;;‘-a:‘-j (moire 01‘ 0V"??? 11:1“- ac Infill 2i? mrnz'sr 0210191000. 71‘ w- m 2;)ourLivu? in t iv‘tolryfiunuv aircuifi; i? afieluuit01;“3 Ouizaumi i? Lyuxnirou refulgrlg. It is. quit?) 1121:1112]. t.1€31‘9:‘0r0 the wit of 210189 5210111.: 13.7-.1' ST'I‘B relution t) tne mile of standuru cable. An "equivclent trgnsnirnion 1079" CbufiCu by noise in tne receiver can be measured by the trunrniqflion 1093 h9099321; to grouuce an canal effect on ilthlliifibilitjf in a. xiuiot circuit. ‘inis relax ior, .‘Lovm'mr, i= not :41 92.83 thin." to determiio. Boots whiten Livo onion Inge Soon; tneit A results Veg." consider-2.413 v'itii cifforout oboorvoro. Otuor muuititics thich I. affect this re 1.;t irm are gauntity are: fro-gammy 0:" the Loiso, volume ' is. quality of too truismitteo S‘BQCJ “1.6 a:.1oiu.t of roan noise. a larva lumber of torts more Loo in 1914 my; a. noise stun-lord "62.53 devised. It Concidtcd of or. ilif?tf11"2(‘:1t finish '52-? (mpg-bio on; groinc in: a co‘npor‘itn tom. (Tm-3 noise. unit woo than giofircu a? a. noise which would produce an emivzilont trutF‘miswion loos of .025 miles of standard Cable with no room hoist» mo 01: a line with an Pquivalont lonrrth of 25?) milos of ."tun kiwi cubic. I'ore rocont torts: indie: to. that thio factor of .025 miloe -t'rdard cable jovr noirm unit still hols for tho c-onS-itior. 01' no FOOT} (A; H) '9 O H H- .0 so, and i‘d’iJ“? tiiio condition it very satisfactorily roprer-‘onts the i .4 1t‘7llifibllit" rocvctioz; Tito zJJor2.;>:~ comiitiorf o“ trim; 0;.1ivclor.t3 bet'vqen subscri ‘D'érft 0 Too study of more recent gait; "lliCil \ ore: t2..‘.~:-'2o milvr caliqitiom of 4' “ - . . . l‘( ' - .- 1-V . 1 y o - \ ‘ I ‘0 . u r 30-3 3.0189 mo an ovoino trim. evil-nil mt o: ...O milgs our unto limit wives “ i'é-th): of .01 ilil"“S of attz: mm ccblo iLS tin bort factor non“! anilgble P1) 01' a‘re;~g,;,;o gructicgl corroitions. *3 tie I’oiso Ifousuring Instuuf'ient.- The noise ffELSIlI‘ilij instrimezit consists assent iclly of 2. Raul-co 0 501.0 91‘0" of O I ~. 0 1 0 ~ '5 ' ' l- 4‘ ” tkllqard 110153 mm a spocml smut 111* Vurynm tn- '24:.0111412 Ox drc -~ . . . . . t 8“ 11: a rocciver. The sowrco of starrncord 110180 or I‘Oiso Element (as: it 18c 'tus Lilled) consists of a '«Ioiff’zitcci vibrator oFPociLtoo :ritn 0t lit-r Lipoor... Such . . . . Trio Logarzitim is so uesrvned “is? condensers, resistances and bcttories. that it will civo rut L constant compo: to noise typical of too iLtorferoLc: founi on tolophono tronsmission lines. V The st;:o rd noiso element is cornectoc to a noise shunt shown in Fin. 7, by moons of ‘f’z'llC'h current cont throw-1'; tho recoivc‘: 1-3;! ho vi. ricd. Tho ston’urd unit of noiso is tcksn LS tho round in tho tolowhoro rocwivor 'corres>oniinr to a cottinc of ore unit on the noiro shunt. Tho noiso sthisrl ro-uiros throo 1r? cells thich shoal; ho ro- plccod aftor six to ton hour? of int'roittcrt 130. Gary should always be roplccod wxon too Vibrctor goo: nOt stcrt 1173 th? switch in closou. The noise clcnont Can not be roiisd on to oyoruto for more toon 290 ho;rs. witnvut ro-uogustwont. u r)crc noiso unit Phonic always bfi ko3t on mono lor comosriSTN uro urea only as a Jri gry stgndurd. 3J8 noise snurt i: also oguivpec “ith t‘) toys as sooth in Fig. 8. The key on too TlCJt is for the purgoss of cnanwinr the receiver from too standard to tho somrce of noise boinp mocsureo. Too to; on tho loft in ono position connects the receiver oirectlg ac;oss tho lino and measures noise 1 "metallic", while in too otoor position the reccivor is connected from too J) H a {L O 1.0 t‘fi’) . of toe lino LI“. 731011116. in sorics with 103,030 01118. The mothoa of OpfirntiOL consis s singl; in first listwuinw to the imimo‘m and then to the rum-nit}. T210 cgmrzitor ao'usts the shiuzt until the standard noioo 1? equal to tho noivo to bo notsnroi. Tot volume ulono should be considorod but the ooorator shonli connurn tho relative intorforin" offsets of_the two tones. f courco, a noiro of 800 cycle Ditch “ill can?“ noro intorfnronco tnun one of lomor oitcn cs oxolcinca uanr "Varmonics." a arott cool is loft to the nuijwim oi‘tdo OiCrLtWT in nakinc thcno teats. Q) -23.. In my] masui-Wtwnt botii noise ‘rtallic an; roioo to crown} are rccordoti. The noi'm. oretsllic is a oda'wro of actual intorfr‘rnncn "will? noisn to r-rounu is an ircication of crmr‘nrc. In :vcnr'reil any noiso above 2300 Or 501') with, in tgm subscriinérs receiver is CJn‘ilV-I‘QLL Objoctiongolc. :3i1‘2co til”, noi'm if? attonuctm: fron ”,4 r . .A . ‘ .1 ._V+ . 00;: to £30}; in (-f)lI;.’f from too toll line to too FubSCI‘lDOI'S lllf‘ci‘il'lQLt, a 110188 of 500 units is sot 2~.:‘- toe comncrciul linit of noisc us :tlposurod at the toll tufit camel. .._———_——— ___ - ~ «7 01.95 7r ' ‘ ’ AM r0: Fol? xv ' ' f/W' or/mo 1‘ .V .- mmr :TANDAJD Fl} 7 "J.’ SflwVT- CONNECT/0N5 FOR NOISE Slit/IT —~———-——————-—— ———--— ——-—————-—-— lll'II'll'lllllllllll'l'llllll III fife. 32“" BA TTFlI’ 100000 .4 S TANDARO n33 -530- ~ ‘V ' 0 \ {1| ‘ "no . ounurmwnt oi (rz‘orc-iu.1.{. & Cross—talk: Iii-:0 noioz" Ccnroo 2.21 int-”era‘cr«":1cre in tcl-‘ioliono tz‘cnfi- Tnis-‘sion. i‘tltimua-n it in nor largo Lot 2;.“ tnriour a.“ noir‘n tho {Ettore ti'xat Cause cross-talk also 11:.“ lino, Subject to noiqn. "211‘? rfi-forc, the two Infec9111"""1‘3;t8 pro usually mm at Lliv same time. Crr‘mo-tcl'rc gray xiitlior bf.» 9:3)1'095‘9-‘1milrrs of FtJl rum cable or in cross-tall: uritn. If a. pot'n‘.tiul of one volt at 80;) cycle-:5 is inyrossed on the cisturbinr‘: circuit and u "mtvnxticl 01' One riillioran volt it: inuucoo in the dis turned circuit, t‘n t'."o lines are sail to have a cross—talk; or one unit. The singlificd circuits 01‘ the cross—tally: instrunent are shown in Fig. 9. T310 circuit consists of two princioul brainchcs, to cit .er 01' \miCn the source of tostinr current may be connected by means of a double polo double throw switch. The tas‘tinr‘: current is Si‘pplind by an 800 c;.'cle oscillator of tho vibratinrt raed type. ':.7hon the switch is thrown t") the loft tho tor-tin: current source is: connected to a potentirietor or, cross- tall: motor, w}- ich has an impreduzce of 600 ohms. ”hon thrown to tho I‘l'f'ht, thns'vitch 1., conncctm? to the trzmsf‘ormr "13" Emvin" a Variabl": ratio such that tho charactcrirtic inpndmicc of the; lino Hi 3* b0 ‘WG‘ij‘i up or Corn to match the inlaodanco oi‘ tho 630 031.1 poton tio.‘1v-tcr. I‘I‘LI“Sf-OI‘ETIOI' "F" in like winner is connectod to tho. disturoocfi. circuit um 15* ‘TL :0 variable 80 that J. uh e i’flpCchC o 01 D the ammrutus connected to it Ly be rtogqeu up to IiiutCh she uiiuructcristic inoedaacc 01‘ tip: line. -31.. as may be seen by the irawihq, both trcnsforwer ”E" and the cross— talk meter are connecteo to a network of resistcnces, condenser, receivers and a returcction coil arranged in the sane genertl nLrncr as the four arms of a whectstore bridse. This network is for the purpose or ceepiny the line noise in the receiver a constant for eitaer position of the switch which connects the testihg current to the meter cnc tee disturbing line. In two networz, a rctcrdcti n coil, wound with t"o carefully belanced windings "D" forms two of the cris of the bridge. The winuihgs are arranged so as to be non-inductive to currents which enter the junction of the wincires and divide egually'between then. Jhey offer a hifih impedcnce to current, which flows throu;h both windings in series. Each of the remain- ing two arse of the brides is "Lee up of a condenser, resistance and re- ceiver. The Purpose of the conicnser is to neutralize the resistance of the receiver. The ourpose of the resistcnce is to build out the imoeoance of the arms so thet variations in receiver inocCsnce go not materially affect the inpedence of the crws. With tJO ugparctuo crrcnfiei as Shown in toe craving, the current from toe cross—talk meter flows turoufh the receivers ir series while the current fnon the line flows throuqh the receivers in parallel. Very little current from the reter is wasted through the retard coil and very little current frritflnaline flows tnrrrfliiuuzsmter. Yith this urnzrfinuxt noise currents flow from the iistrubed line throurh the receiver for either position of the switch. -32.. In niacin"? a test it is sirnpl;,r necessary to throw the 800 cycle tone first on the potentiometer ‘ :6. then or. the uisturhinr line. The potentio'ieter is then varied until the tone is 01 the same intensity for both positions of the switch. Tue cross-tel}: 1.2.1} then be red airectly on the dial of the pot'3-tioneter. The tips on the transformers are g‘overned by Keys. The low in- pedance should be used on all non—loaded circuits uni extra. liiht loaded cable phentom groups. The r'iediu-n izipeduzce is used on medium loeuec “..ble nhamtom syr'iups. The hi *1 imiedun e is used on heavy loaded an mediu'rq heevy loaded cnble 'Jhsntorl cranes, uni lozi-iei Open wire nhentom strouos. The lines under test should he terninzitezi in their ("Elana-icteristic insect-.1300. A sricc‘al networ}; is “orovidee‘. for this zrirn'iso. "ams‘ir efrnts are. usually ”13:30 between the mentors and sis.n Circui ts. Tie commotion? for 'r".;.‘r:iz;-‘ this test ure me: n in Rim. 1’). The comercisl limit 01‘ cross-tell: is set cit la—Ot’) units which corresyonus to about 60 “tiles on" stenuzire Cable. s. 39: QW‘ Vuoxfia 0W2 k Ilml' x3 26 mSoxfiw‘Q QR . 0 AT 3.3 kg»? m G ESQ“; E E .356 vfii§§ $233? as} K320 BSEEG K «a SE33 Ckossm/rffr SIMP/Ifie‘i C I rel/If Arrangemenffor diffe rear B51175": of Keg: Dlsfw'bmy ——~ C I rc w 1' DISTurbed ——. 6/ re W I“ DIsfurbIIIg ——-u- C I re (/1 f Dlsflfrbea’ -———> . C Ira/I r _J To DIS-rurbi/IH Circa/'7‘ ‘ ..l. ‘ To Daft/med“ C Irc Vi?“ SIDE 7’0 SIDE E {)I 2% {)2 g {)3 LE {)4 P/M TOM TOSIDE [-2 E 0‘ g 402 E 'v———O3 L__04- PHANTO/V 70 SIDE 3"? E H‘ £3 2:09 irtgmu T’1;'1|;Ll;.z‘c~ in 00*r111ni3;.t ion Circuits. 9:19 afnct .0- (L r fwiiuul volu.“ 2.1-: carat, c;-:;;;0t 13" nm1tra.li:md by trmzspositiona It; Omar to mitt/1r z4;._)1wcL-L.tr3 0:37:01; 01' .L volt-Lin to FTI'JUJ'LLL' m; the inmr Circuit in 1)1‘0L;'L;cii.5‘ a V)lf,a.i'~': ‘m 5*‘r'011;.d on the I:éi:i:-~}L’;ni'il:." t“1"’:);LO."."‘ Circuit, 1.3: L1. ssirr:_)l~e (1.3m. Lira configure-J 0:" only 02:: pmrer ".Ji r49 at; 11¢: teirxxgns? ..irv": as; :3L'Lo'.z'zL in g‘i}. ll. L‘.L-: gamer iL', LL‘L'LLL te 99310119 wire toyetue‘ wit-1 1. 4'3 Awning? :3; 7 be con: iu‘arm w. :.L '31:..‘314': cg)”;— binatir‘n; of condom (31's: CLpacitis’es C LLLL Cr: in series, LM'LLL‘f iL; {minilel with OI between ;)”)':.N*r axl'irv": aim g-t'mmd. 4mg volt :L'": 23' , brattieflii trio pm'rar wire a.) a 7’:’)1u;;1 will ta 21'01'01’9 yrofigicn a. voltgra 3%" baht"? approxiuitely aqua-.1 to __C ‘3 vnmrn 0 mm Cg depend upon tin cross-sectiongl 0+0”, ‘) - x 0 ~ 0 - ‘ dimenslms flf tun two 111103. I: "tun. 191.*"tr1 01. um parallel 13“ varieu C and Ca. Charm-r3 in tho Fury? prijrt '10:: 2712.101: 31 um that the il-..:‘I:C.réIj’)l;.iim..z leLtdgu LL som<:-'-r."-mt ni‘1ilur owning}; exist? witg POT-{mt t3 tzié go my Currants, '."ELiC..'1 gr 051110"? a magnet io i‘iralu il'LC7L'ffih11114‘ in 1LT, 7115-13,; as: tlm uirtgucn 11-0.; 1;.er ,zuri'->.1;t, Caz‘ig'im- "sire il.(fl"':;.-S€‘.Slg Ii. Pi”. 1;; only on: power and one telep.am2 wire mum barn thorn :‘or :‘izmlicity. Luigi circuit such for instwncn, as a pair «)1‘ teleplmun. Ull‘fZ’f‘, or a. pair 01' telejioge wires in pamllel, \.'i ‘51 the 1? 31111;; as L—. rotu'n 9.1.1.11, ‘~.".7.'.i(;l'1 linssx Luis gig-f- netic i’ielu will IL‘IF‘: a. voltage irmuoeu in it. fine circuit is cornplnteu through 1:219 Ca-.1)£.Ci ‘53; 01' tile line- to .r‘rmmci Lin and “JD my; the current is -34.. caused to flow. Due induces voltage in this Case will be preportionul to the length 01‘ the parallel. a voltage to grounc will not Ccuse cry noise in too subscribers instrunent unless toere is either a series or shunt unbclsnce is the loop circuit. Fifi. 13 shown two telephone wires with a voltcfe incuceo to around as lhdlCLtQu by the eeneretors. Dnese voltcres will Cause currents to ground throurh tie ccoacity of the line. Suppose one side of the line has more resistsrce then the other. This is represented by Zu in Eie. 13. The drop over this resistance I Zu may be representei by u r‘ezmrctor as shown in Fig. 14, which will ounce a current to flow throu t the subscribers instruments at the termircls. A shunt unbalance due to lefective insulation will Ccuse similcr results LS? worn in Eif". 15 we Fixt. 16. These fissures are :JiSIeLLciLf' in one respect. It mig’rit be £3uA)l)0f‘;Ol txlaLt it is pmsible to neutralize an excess resistsnce in one side of toe line by placinr a resistance in toe other side of the line. Torever, toe especity o“ tne lines to around is not all concentrated in one place but is cistrihuteu uniformily along toe lire. For this reason the revistcnce or the two lires n'st be uniform throurhout their entire length. Series unbclcrces ere usuclly Caused by poor conrections in the lines. Shunt urbclunces are ecuseu by broken insulators. Series and Shunt unb;lcnces are also the princrv couse of cross- talk in phantom groups. A pountom circuit is estsblisnee by usiny the two sides of one physiCsl circuit in parallel as one side of the line and toe two wires of another physical circuit as the other side of the line. Fig. 17 illustrates a ohantom group with a poor connection in one line. The droo I L») U" I over this resistance "Ly be reulccez‘: by You when there is CLUGOd to flmr in the loop of the '_)‘.11:.r.°iCLl This is called emitters to rice show-n tout shunt un‘oslzlrces anti ll also and "aide circuits. In order to gr! t1) ugon cross-tent, or. very siniler to Pie. 1?. actual test. Toe cerfree of llNlRalilILCQ tell: at ecci point meatsm‘ei. '31'0 current florin" over the cross -tcl}~:. .Leterrzine to“: effect of :4; a u 0‘ D u "1 .‘ Get? 04'. laaulugjs a Pen erstor as 331013111 in hit". lo. pi 1am. t on the re Ti 1 l he a CI rrent circuit COI'LtLiIllllt“ the unbalance. 1:1 8. siziilcr manner it 3114.3 be 7 Cour-e cross-tel}: beers-2n tne :1E’1ti‘1toitl llliuulvilicq 111 u. iJIid1401 artiL'icill line 7.1.8 set 11) in t11e laLLorctogr A cross-tail}: meter mar-cs platcec 0:. the line cs ii. on .;,s veried on A. the smomLt of cross- were toner, tLLe 11rst with a. line resistance of 70") oims corresporoing‘ to L. nOl’l-lOLAi‘BU. line cmi secon." vi to a. line resist-once of 2000 FT . . x- ‘L A. ,. . .- ‘. .. L11e results 01 two test or: soon notel tout in L. leased line too snui'it noon cross-tel): “mile in L non—loLLgezi serious. It wcs found that a cert. in balances roulri neutralize 942. '11 other. Solution for excessive cross-tlzk, but it Tlixj’ chc ount for series urbclaome "my be re 1ove cm 021m corresgoniizg to a. loacec‘ li1.e. Figures 19 end 130. It Should be unbalance hes the greatest effect lire the series urbt lent-e is most coolinction of series cue This not to he reconoended as a '0 lg) still hove more cross-talk thnn before. 4. ~ . m \k llllli! VtSOKQ ”N ‘N PAC: {wk 4‘ N‘ a [‘t ‘ I r h llllll ukxxxxuxaok f \ {cs nukokumo §§\\ .okukufit %\ .9...‘ w m H: :H..-...--.... mm 8N”. nmn -1“ WW “qt-NF 5‘ mPk Nhiliim Jmullc #1 w» Mara l, ‘ .3 w 1 $1an R'IW. I w, E75 MICHIGAN AGRICULTUR'NL CO' I. . . -.... -- . . - . . - . .. .. . n I E. _ m - . i.-l..--. . . I a . - . . 9 . u. a ._ . - ./_. _ .0 . . _ . _ .9 _ an. . n . r. ”M . 1 - we. -. _ fl... . I1 «I! - §_ . 5 5H . . .0 1 :m I J . 5 +1 - w W. _ c 1 _ . a a _ .m _ . _ . H a m u . . 4 4. t . . .Mm. — . 1 .. _o y . — u _ . u v“ . _ .i . _ . .7]. . m. _ n . :_.: . m __.. u , fig..- _ h 2.4 ,6 .. . 1 .1. l .l...... u r. . .. _J-. - .. . __ . , ___: . ..1 ..1 f __ . 1. .H. .._._. .. h H w . . _ . . aw uh... . .-._ in...” 1 L _. h. . _ _ . .1”..._v-_ . _. 1...“..1, __ ,.,..,.._fif .. _.. ... _.. .d“ 27. Fig.2.? knurm__tt._~:_..t_.r L. .- 1.; r- _ f .. t- t r _F 5..., - .r..._. -. Gt. _ L: Hmkbbhrhhuhrr. [/fl'cfof Shy/If (MM/lace on Cross-Talk a JV " ~ f I . __._ .___-.-_ In an 3a 4o :0 to - 74 ea ‘S‘Iwn f Unfit/once m ”100- film: FA: .20 -38- "cries Resistance and Insulation Tests. The object of'mekinq routine resistcnce tests on toll lines is to locate and reecir any loose connections which meg cppesr in the svstem. ‘ Tue seriousness oi a resistance unoslnnce nos been discussed in tne previous section. The methods used in making these tests are mostly modifications of the wheststone brides. Fieureafll and 22 are simply wheatstone bridfie set up for neasurine the loop resistance. Fieure 21 illustrates a method by which it is nossible to determine tnn resistance of each wire. The 100p resistances of a+b, a+c, b+c, and c+d are QpCh senarstely determine. Now with tne four eouutitns it is possible to solve for the four unknowns a, b, c, d respectively. In most cases, howeVer, we are not interested in the actual resistance of the line but in tAe difftrence in resistance of the two sides of the line. The Furray and Verley Loop Methods provide a means of deter- mining the difference in resistance of the two sides of the line. In edch bridqe the battery is connected to tne junction of the two lines at tne distant station either throueh ground or a metallic line. In the Inersy L00p Test the two ratio arms (A.end B) may be varied to obtain a balance, while in the Vcrley Loop Test the ratio is fixed at 1:1 and the resistance R must be introduced into one side of the bricge to obtain a balance. 3y inspection of Figure 24A or 243 it can be seen that R must be equal to the difference in the resistance of the tno lines when a balance is obtained. The taximum allowable series unbalance is set at 2 ohms on side circuits, 3 ohms on Phantom circuits ani 5 ohms on non-phantoned groups. -.Sj- The most satisfsctory method of measurinr insulation resistance is by means of u voltneter as shown in Ficures 25A and 25B. ffices were first provided with 30,000 ohm voltneters and a 30 volt battery, but as lines have been kept in a better state if repair {urine the last few years, most offices are ntw provided with a 100,030 ohm voltmeter and e 100 volt battery. This, of course, is a much more sensitive instrument. Insulation resistance may be cemented from the reading of the voltmeter. Operators are provided with a curve to convert their resdirg to megohms per mile. If the curve is not available insulation resistance may be cemented from the iormuls Jar-(g, ) RL (9 ) 100,000 X = Insulation resistance in mefiolms per mile E = Voltnre of testine battery 9 == Voltneter deflection E = Resistance of Yoltneter L = Length of the circuit in nil,s. The insulations resistance should not be less than 10 nofohns per mile. Sch ema f/c Diagram: af‘ 7:: Ting Me The 4.: Three Win: Met/:04 Loop [lief/704 A x ’ _ B R J. lrf Fla 2 1 R}: g R Murragtoap Her/lad Me MM 2.? e W" n Ground Re fur/7 A fig 2 3 ’ Fig 2 3 B Varley Lo op Nomad“ M e Tall/c fie Turn 61m unlfiefar/I A A 1 4 a I? B If 1-: I - L__+__ | [7131914 ‘1‘ F/gZ’fB Vo/fmefer Me #104 ofM «wring In: (lief/2m Insular/an Bern/ca: Wire: I ma! afim To (wound kg? .. «57 A E 1“ —_—- —: _4- p.“ E 4.; F1325?! n? 258 -40- Tiz-qnlémation of Data: The circvits are listed in {groups of three, the f5.r.'=t 1.11:1 third 7min“; physical ci rcuitv EJIC’: the Canter the phantom. In listin"; cr0»‘:f‘-ta.1‘;{, phantom to Ride #1 is: listed first, side to nine second, "nd cf'umtom to 31:10, #2. third. 1 a Circuit Noise Cross-Tall: Balance Des igmut i on Retail]. ic Gr *3de Insu lat ion "fcrlcy 3 «3.3. -.Tfa.l. 200 r100 1430 350 0 1 Det. — 0.12. 700 100 1800 250 4 GER. - 12-11. 9.00 600 1500 250 0 6 6.3. -. :11. 2350 400 700 0K 0 2: (Juli. - 3.0. 700 50 94:0 0?: 2 30H. " $0110 BX) 130 O 1 GQRO " u"u’&:z'1£d;d 4‘30 530 1400 0K 0 5 (Lil. - 1111. 800 500 800 OK 0 2 3.12. - Ya'ayland 25 300 12400 0?: 0 1 Charlotte - 0.12. 150 200 1700 500 1 1 (Sufi. - Jackson 750 50 600 2 Cull. - Jacmon 400 2:00 2300 14.8 0 2. Det. -,G.R. l§e00 150 1100 400 0 4 Det. - Gui. 800 100 600 3 Eat. - G.R. 4:50 530 3200 400 0 1 Cull. - Ionia 500 150 900 0?: 0 1 Flint - (Tl-.11. 1500 150 1300 O”. 13 1 Gull. - Gnocso 250 1:30 1070 1:00 4 1 (31.3. - daaizzwz 2:30 150 2500 ‘ 10 5 2 0.?2. - Carina-v 1500 15-0 500 450 6 1 Chicw’o - " 50¢) 150 500 500 31 1 Alma. - 0.11. 1000 213-0 1480!) OK 0 2 Alma. - Juli. 802) 250 -£‘.C)0 100 l 3 0.3. - Grormville 530 13.50 2000 9:0 0 1 Sarmwc - 0.19.. 800 .235") 0:: No Pxxantom 1 0.11. - Greenvilb 250 L'O 012 .3 -41- Circuit Balance Noise CTDSS—TLIK Desi rmuti on ~ 1 Cadar ijno-i}o:{o F o E’hant on 1 NNH I‘C’Cflf—l [OHIO [6'an H r; ‘1 H f-‘NUl r) N 0.1-1. - Petoslqr (ta/Jo " Jo-:.o Chr. - Potosi-gy 31011. ‘- 1’0tOF‘E’V .1. - Sgartu G011. " i;}')L-.I‘t'd. (101).. "' SQLLI'tb. Big ii. - Go}. Joli. - iiGQd U. 51g 1).. "' CroiLo :1 '7 '1 Y U031. " U011. ( '1' 1 'L ac.-. '" Joilo . r i a) no}... "' be... 1 l T? 011. " U011. I TT 1 '4; uo .0 " Uolto 1 v? ,1 '* u..‘.. "‘ god. 0.3. - I’us (er-01“. “- ' ‘.- , 0:11 . - .I'usstetton “ . " - .r r a (.111. — . u“: .cgon 0.1).. - (311.1113 CDSILO‘Jiu "' 3.1:. 0.2:. - Grunt GR. - "tickcrton 0.21. - "LmTzo'mn 0.3. - 73:1"ku "('01: 30:20 "" z': Earth F o Phat t o": 1 *«_ ‘ .I t) .Adfiore - “0.1. ' T 73, ..7 n 31" 9.. - 11004 0. .i ,. - OLE. ‘ K000. C. , 1.17.1, ‘ ~3' >1. — been 0. ' ‘ ‘ ‘ ‘-‘ 'Uoiio "' .Jxk- 1‘“. want - 0;); rte. 800 235C) 1300 500 900 800 800 600 700 600 6J0 3300 700 400 12. 00 800 50.) 700 60') 1500 3000 603 45 0 11300 90’) 1300 1300 400 400 400 3’ 0 QOO 1ij 400 13.5 3“ 0 e00 500 500 400 $300 1300 li‘rO 550 300 600 700 400 7000 2400 4000 1000 50 19.00 1300 1&00 5300 1800 3200 {32351 '7 'l cob 500 29 00 2700 300 1600 OK 77 JL 0? 7 1 90 (W Y ‘7 L Insulat i011 "furl ey 1 OO 0 COO 0010 HHH 03.410 (\3 (0 [OOO Circuit I'oise Designation Xetullic Ground 1 Plainwell-fiu". 500 6J3 q) I? o Pix-an tom 2 IJluiIr\'t'ell-. .‘GL! 0 -‘ OR. " 1191.91.14“ GnR. - Lamina Golio "’ ‘ 10111}; I‘DCfil—J Det. - 3.1.. .'0 Phantom I; 3:18;le "' 5'03. ‘r4 ._l 7 Go}. a "' Data 6 G.H. - flat. 5 G.R. - Dot. Luke 0. Luna inf: thkn O o HU‘N 'T'Q o 0 L3) 0 o ' I (.2 3 f I H H Luncinv - Luke 0 Luncin: - Lion. L¢n3.- Hulliken >4»: Lans. - Portlihd Ionic - Lgncing I onia. - La- 311.3: papaya 0.R. - grconvilb blTla - Ct. It. “1'13 "“ $0110 HNN Iciliu - LLJLSi .,: Inuit. - L'al..8111._, Icniu - RortlLLd hvh‘ H £37.11. " IOILiLL 1“]. illt " Cyclic 0.3. - OWOSSO F‘F‘h‘ 1 Balding - 10111:). H IGOR. "' 'DI‘L'I. Co Ctldo "' $0110 1 0.3. - Reed C. ._a 500 200 800 500 500 500 500 500 500 1000 600 Lax? O. - LansinPISJO .350 500 500 100 600 L00 50 1000 500 700 20) 1100 200 zoo 150 200 200 200 50 200 200 130 100 100 200 400 100 .250 100 300 100 500 300 300 200 200 03: “U' 250 400 Cross—Talk 1300 400 3&00 2900 1000 5000 600 500 1000 9000 600 2500 800 700 800 2800 400 2000 900 1300 1000 1500 1000 2600 Bulance Insulation OK OK OK OK OK OK '500 1P .13 OT o" 0? 230 123 0 7, IL OK 100 300 152 Varley 0 O to OHO UIOO Circuit T’ois o Cros s—L‘clk Balance Desirngtion Notcllic $r3und - Insulation fgrloy l Con'~:li1;- —L.‘p...1‘ a 1830 150 1 Chi. -l?oll¢.1-0. 14-00 300 52:00 1‘10 0 C 1:? 3 3.3. - Nollugd 1200 2 400 Z ‘i 3.3. -3I0110Jhi $00 500 20" 198 }.R. - Hollcnd 500 50) 5000 40 O 0.3. - H. V. 2000 100 &&0 52 — ”Ollail 600 600 1343 5; 2 H'F‘bfi W n 'J 01.; o a gllegcn - 1.“. 700 100 16 1 0.“. - Lcolumu 500 500 00 T'ollzxd — 3.”. C00 300 2700 "olland - Sun“. 1200 25\ 50 Rolland - S. V. 700 350 7000 [Oi-1H H 13 C 4 4 o I :- 2 ‘1 f‘. 55 ,p C L.) ~‘t‘t- Conclusion. ‘ . ~ - -1'- -. ',. I z n. ‘ . . ~‘ I . ‘7, ‘— . r -‘¢'..,“,3 ' » ‘ "2:1: :~3vci';..l f'..C1.0.L»1 .."iicu coteinnJ, tile liiiucuue mitts oi ”aver circuits on corrmiiccti on circuits: valrg omr Ll- 9:;c":eoi:¢_-l;,r zzio‘e l .L ridge mioer the different CDldil-iOILS wliic;i occur in gmctice. ”Jame p.) footers include not 0111;: too Ligxsiccl re 1;..ti 021:: ‘aet'w'en the tuo classes on lines, ‘u’lliCLl are 1.3121211 in city oart 10111.; case, but c130 various proyerties or characteristics: of‘ eiczi ester}, none 01‘ which .-re 1Siugll;.r not 111101.711 even aggroziimutely, e;31.)eci;.ll; in Cur-es 0: new peasex‘ limes. liesc consioeraltiom , th’et‘ier with tile cmglexity of the relati 01;.c among these various itctors, rmite it impossible to for'rmlute on accurate and single definition for general use, showin'? just What limitindf concitions of proximity between the two classes- of circuits will in all came prouiice interference. The co'r‘irmly imam foctors '-7:iich affect the .mg‘jnitude of induct ion from a (fiver. parallel are: 1. Len-3th 01‘ garcllel 3. Confi'tfurati’m 01' 11:10.3 ‘1. Formal ogerat irq: voltages ail cm'rtrmts :ai' pout-"31‘ circuits; tieir IZ’lwfllltl‘lfiELX 41;; i'uiiiutr‘LtLI i‘i'r-é.111oi:z;c;,-'. In acq- itior. to Loom, rectors aziiicli all”). i-ecqi 1;! o‘otliucple there are severcl other 11.3 321's: “wicii mast be. Jpztcitniimzd b;- rg.» {ticl test. Toms-(:9 are :‘ullv an iniroi'tzrit a: the factors linted above all therefore finite it more inocosi‘olc to r1"‘ce aim general statgtm‘ts I'*3;:‘fL..l"cii1;"'.llL.t constitutes .L an undesireble parallel. She aduitioncl factors are listed as follows; 5. Zfasgiitude of residual voltage anti current. 6. ..'ave-.-for.:1 of both balanced and residual volta ges and currer. 5:, involving: the '2; mitutio are fregiercy 0:" all harmonics. 7. Unbalances of cummziicati on circuits; their r2, cit-Lice, location, and character. 8. Abnormal conditions on IJUW’BI‘ circuits. However, it is desirable tnat soue .iiulntitz..tive cata be given to estublish an anoreximate idea of what constitutes an uroesirable parallel. as in previous discusei ors all reference to balanced and residual components will be cons iuered separately. Toner and cormuni at ion." are on *J (-0- H. O H p 1 O ..8 Rune tne f'ollo'.viur' coni oncosite sides 01' the real separated 5') feet. The power line which is not transposed parallels the co'mmicat ion lire for a cistance Of 10 "files. ‘i‘ne ayproxiriate limits 01‘ volta‘e and cmwrent which will not cause excessive irtcri'croancU are listed below. Candi ti on Limit Electra-static ii‘.-..-'uction from balanced volta'je system --———- 4000 volts liagizetic induction from balanced current system — ---------- 4 amps Residual Voltage - ---------------------------------------- - 500 volts Residual Current -—-—----—— ------------------------------- - .1 amp. All these ..iuafiitities are prI‘OIZiILILtély inversely proportional to the length. Thus, if the parallel more only 5 miles, all these quantities might be doubled. grain, 'nduction of all tyges varies inversely as too square ”of the distance betzxreen the disturbiir'; are; disturbed circuits. Ti'Lerei'ore, it“ the separat' on of the power and comumiccti on circuits be doubled tziese film'es lists-l aoeve “ray be multiplied 'oy firmu'. 11“ the power line is transposed the limit.“ of balm coil voltage i ll be greatly increased deperidinz-r upon how well the transpes it ions of the 110310.)" and. communication 33s tens are coo ‘LillLtflil. The effect 0" residual cotrm'):.e;~ts 0:" voltaee and current "i ll not be affected by transpositions. BibliOgraphy 1. Inductive Intexference between Clectric Power ant. Commie-at ion Circuits -- Iiailrnad Commission of the State of California. a. Cneineeriné Bulletin F0. 14. Yichiqan state Telephone Company. 3. Report of the Joint Committee on Inductive Interference Transactions of E, I. E. E. - Vol. 3? — p. 1491. 4. Inductive Interference as a Practical Problem by A. V. Gr’swold are R. 7. Fastick. A. I. 3. L. - 1916 - p. 1051. 5. Incuctive Effects of Alternatine Current Railroads on Communicafion Circuits by H. d. Varren. A. I. 3. D. - 1918 - p. 505. 6. The Design of Transpositions for Parallel Power and Delephone Circuits, by Karold 3. Bsborn. A. I. 3. 3. - 1918 - p. 897. 7. Review of tee Sub-committee on Wave Suspe Stancard of the atanuarcs Cowmittee, by Harold S. Osborn. in. I. 3. B. — 1918 - p. 261. 8. Ielephone Circuits with Zero Tutual Incuction, by Jill'au 5. Crawford. A. I. “a. :10 " p. 439. ' ”M Rm 935 B ’~' 0 ul \ JAILA- 1.. i .. I‘i‘i ‘O‘iIAvtvf‘J. ..o. AE'IJV 4... [iii E: g; ?. ”TH 1111151219115“! fifiiuifllflfifflfiuiflfi III? ES