‘Tho Transmission Characteristics of Toll Telephone Cables at Carrier Froquencios 1S THE desgs of a sew telepane Lawson ays a haowedge of ‘the characteristics of the asim aver wich the waves are to Pai ‘of omrse,«premusile. Whal painting experimentation is neceseary toserumfatesuch knowledge, homever whet voluminous data are iavered, shit minatise of deal, and what extremen of accracy, eve things fat Tens obvions. ‘Recon pees ave eae new Cael erie telephone system for operation aver cable pair” For mum cable lone fs aceded in ordet to decrmine the necvary repeater fi. Accurate data on the isetionloes elope versam fue ace required so that comporsasing cquslicers can be sim to give uilorm twansisson over cke fequescy band. Tn exer to design a regulating system to compensate fr the variations in ttenustion which rtult fre ‘changes im evi tempercar, procso emledge of those wagons 28 & function of frequency fe event Tei neraesry us knew the pecance ofthe eale pais in order shat the ample impodarce may be mated to ft, there avoling rections whieh wowle sgrevace croseta effete, Vor yrioaspurpmss, ea testing the cables designing the coils te balznce fo erotuale et, b nko rarmeery lv the ancamentsl parsers (estance, incline, espcaree aad conductance) or sowed primary constants of the puis. The velacy of ansmisson sks plays © pare ‘letermining the Charteris of he chanael Ie akiten inal These trantinion chaacerati, Iti, of ewnrae, coe know the roo talk couplings Dstwoon diferent pein This aubject aw been treated lkewbere’, however, a ie not conesered bare. In onde tat the cable cei systems may be apd inthe plane wit cut romuting extensive tusemision menattementt on eech.ndiideal Carer pair in ear erate section, ie ingortn ct Use diteence ia the transmission characteristics uetmeen diferent pars be known. ‘be problem theretore hocomcs one oftvistcal analysis, In noe eases (ae FoI RSG vl dn Rast na ees ee ae Pe Pat cece eg i So ctlets involve ae cummuetive with distance and the aetaruey involved fa the deerminuion of the vavins churactristics ie therefore set Dy Che ‘airs cistnge over whieh the spain is deemed to operate. For a itunes of OGD es the total loss atc top srequenry of ke. wl. be Sppresina:ely 360 ly the aenusten diffrence hetween the top fe tqerey of 6 ke, ard he bon egaeney of 12 ee, seule 000 ch the fale ab about the average temperntur, SSCP. The tabge of variation fn ioe with terpemture, suing aeial cable over the whole cistance, TL ie about 8 per cent of che total a: 60 ke, Tk is desied to correct hee Ienqueney eiferenee in lee and variations with temperstare 0 srenrtaly th oi'veual eons wll be sonar to ia 42 db ‘Prior to tie Inglnung of emperimer.tion with cee erin systems Tinted ve ad been mln coon wth cts peters opera over faerie lines frequencies te 30 Ke, of concn elatvely short (orinar end intermediste calles, Tho ovale data, however, were (eit nedequace Jor te eal atic aoblem, Acomligly, an exteasive vis of taste ms tnceraken. Reals of stardard oll able were placed fn terpmture controled rooms where th extreme temperstare variations of the mid-neet cul I sulotantillyduplicne (he actus boratery Temnpeeatares ranged from jut below UK 20 (PF) asd mczauremet fyere mace fo delermine the changes in the parameters of the cable ae Couparying thse wide tespsserce veraiens at frequencies fro 1 r {100 fe. higher sore ner Certain of te ests even stele the cl of verging the humidity content of the cable. rer measure tents were then mace on stable lengls of pais fp ctual vimonerei Cables ir which eaeier system were uo be istalled, “Thee rests cor Tkorate’ and tended the eats fom the abematory seusurements; the Sxbeeqwent operation of eine epulatons, ete, based up thew dat, howe! 10 esata dicepases The presen spor ater seer t te type el eables employed for ‘he ne eaerer lems ulins themed employe in determining tir Characters ls fn die boratory and the Bald, suruines these Uherocterates for ppl IS-ginge sul a€Hegueasce yp 0 100 ke. ard Sully exienls dea to Frequencies Mie a8 700 ke for 16 end 19 ganze cable, ‘Tamas ow Chenne Toor Camus ‘oe type K coir syetanBae beer designe se tba it may be mp to-cnisting cane, thus in many crs avy the stalin of expense ev cables, Moat of the sand toll clei the Bell System contains hidy 9-gauge popes insulated conductors in “muipletw'r emus" ie, foro conduetors ave tte together fan a pelt ad wo pas tise cacacrexIStics uF POLL TeLerUONE CARLES m8 together a make 8 qué. ‘Th nomial capeciaace of «pair i 062 mf permle, ‘There are vzious wit lengths of ath pals and gua a given fable well a cables ranging x sizo fom 12-qual calle la the oversine gauge exble corsining 228 quads, Some type K is operating over “pated cabe, ie, cable whieh only the wires of each pai are twisted ete ‘Operation in two directions is acomplished by using citer @separste calle foreach diertion ar single cp wih two groupe of eoeductore rere ly layer alk Thiv avs ious nearend crest fects ‘which would veel om the lege evel diference existing betwen oppoate ucetions at «repeater pins Mons ov Measenmst As mention slave, 25C-oot sels of standacd toll cle wave plas fn e special room which could be accurately stan at any dese tempore fom about 2x0 to 130 dogres,Fabrenlt, avd easement sade for various iequences and temperaures. For tie st part, ee conssed of openciteitadmitence an ehorcccul ida ere ‘ets en pref the yen inthe euble nt femperatnres of aout ae, 30, 50, 80 ard 120 degrees Paver the frequerey range rom to 10D. From those measurements eumpulatins end tien be made to determine the ‘assance iuetane, cafe tans an corlclneee a oll ge tae ata flea, phase consent and characterise lps of tte oleae the diferent temperate and fegueasks, Desiled data Sequeney and tempertture vriatens of these quamriis ere given below. Ms: of ‘hese data ere abrned fom meastrements wade oo 16 and Uauge fins ty rel lenges of starcard rol ele, ‘The temperate Miele to ssn ata est eee and ee, restate cet ae tneremeasred a Fuentes ring the process err € Ee 3 feats temporazue readings by compring with resist erperatire fortes of these pairs. Thermocouples were ab allacel to the rie ft vatous points alg its Ingham shea temyeralores determined foom thom, After stabilizing the toom cauessture a lay st psi, the variation in able cemperatur took paar slowly enn wo be allw for in the computations ‘Alter the seston of he Palt-South Ben rae fer tial oaelton, fur masuretts were mae on certsn ofthe mr thie cble, The lest sections, extending out of Taare, Inésme, were made abeut 10 sles long in ower to obtain the averagiy eller engl. For this die lance it was nat parle ls ase pen asl hort zea mensterienss = ‘as dane it the laboratory, ada subtiation metho was devised (ig) Tm ev Be Noe wed wi be ce by He a gt m6 BELL SESPEN TECHNICAL sOUKWAE “Tis ona exeetcy fst easing the input and out cxrents At he two code of the fete pal by menns of Ueerao:ouplesiliammeter irrungemeats end then inmelintlyseading 'e, over enather pi (aed feteenee pais in Tig 1) bil cul lo 4 coavenient fed de reairsnce, the fame for all measurements and adjusting zostanee networks at both {rol che line Uti the meter reading ave the sme as for the we, Suit ‘ble charttlen enable readings of wenuation (incerton 1s) coresporsing fo the de. (end thorere sso to the ace) cmadings to be mae very api, ‘Tors, Caner Crascniasnct Boo 160 Ke, vary Lite Poratrs ‘he four primacy lve paramaters,R, Gand Cres cevitancey series jlutence shut leakage nd aunt expacitaree—are af to sme sr for "Lkinds ef tunes lines, ie te eelatiee inpartaoe of she various ements changes coneraly with fequency and che type of structure Ganseced, ‘he old une pinay “constant” obviously 6 micnamer, hd spl to speak ef them i "parameters." ence this dew nat sever py anything egarling thee caustaney or inconstancy ander The true os diaibued valve of hese prasetersare usally band fom mesarenents ofthe open cri wmitaace, | jC", and the short Clu impedance, J 4 jl! ofthe ara) pals in a abort engl of ee, by means of te following formulas: AARSGTERISTICS OF TOLT TeLEPHONECKBLES =m Rvesitance) = KU — 4 SEC) + E Gaductence) = £0 = Jetbic’ 3 aR G (conductance) GUL — PEC) — 4p RU FE RAO € Gcopucitance) ~ CY = BaAhICT + § RE wi ‘These formulas give accuracies within one per cent for zeedeagt of 500 eel ress equencis up te 100 acy for Tange enbe sing ‘apaly 062 mf per se. A the eurves uf FG, € eveith ave basse on true values obtained from such computations Resitonce ‘The quansty R secesresstaoe in obwns yor ily hue a Lage vaiation ‘wth requeneyproced by the nellhimen hemrnenon caled skin elt Ad anor lrg income, reaking from the oenes of the wes in Gables, known asthe proxy effet. The magutude of the prosimity ‘fest varie withthe disses of he conductors aswell as wis hese tion. "The cues in Tig. 2 show Ce inerenent i resieteae nestling feom Shir wfect aad the toc ineeane felting proniity efface as contre for a pair of wires separated by various mails of ther diameters. The tec, B, in Fig. 2 ie srt of univer pazaicter tied ia data onan llect so tat sgl cuve wll ice for vious guises. IY is freeeney in eles per soon nd ys the de estamos for 1000 fet of he wire (Qot 8 100001 oe, he parameter ie give Ly the epantion Bo VIR + VIR a for 9 gnuge mito at f= SD cores to S200 cycles. According coche curves N= 49 (512 ke), the skin eect neenees tae ac este alt 12 percent mone than the ds. 1 For x separation of tro dames between centers of th mires of «pir (= 28) the proximity eect adde another 6 per cease the esibace ratio making Vie total a. resistance about 1.18 Himes to de. resistance A 542 ke. Tithe wires are closer together (= 4) these, resistance ie omputod we be abott 180 times te de, which i uu Ue ato sefaally reagered, "The eos causal Dy she prewence ofthe adjacent pir fn a vig ‘pice yh 0 Sin tear JOURNAL qd, the suennundingwices and he lead sheath are not incuded in these foupecatons "Those values assume s cemperatue of 20° Centgtade (68° Fuhrenlt) bt ifthe temperature varie, salen does the esstance, igure $ shows the ee. tempactore oefickat of rsitace and ite yan wi cen Seroare for tDpuye pale in ohms per alum per Geyrer,Falceabt, ig 2 Skettis est on ita be pa 1a aoR@ fo) here A ja the 0; temperature cefcon of zesstnce of copper a Uegree, Fabreneit. The av. sestance R al temperate #5 ven by the formula R= Rb AGH ol co) where Reis tho os naktance at trmperntre degrees, Pobrenheit. The Taeficent decreases wich Ineoesing frequen, but not indefnely; t Soproaches 1 the de. coeSient as ite agvpotic init ith frequency CHSRACTARISTICSOE TOFLTRLUPNONECABLES ah "The normal variation of alptomperatare in the middle western part ‘tthe eourey is from abaxt 1° Fabre, (—1 Cent) pie 109” Falreiet (4° Centigrade), Extremely hot summers ke that in 136, ‘ubich was prectle yn severe winter, ston even higher tesnperatues nd ‘thers ave occasional periods in mideweetom wines wen Ube (nsperatute ‘ves cantinoouly armnd 0" Fora week or to. Thee tempers: tes aze almost the lempertres asm by open wires, Il wires ine 5 Ita sheath ke those nan aeialeable ae sje’ o och higher tase ie temperatures in hot weather sen expe to dee sunlgat In tbe liwence nf winc, Some observations tae at Layrary Wallan, in 1086 i die rvtance of cable pir owe! that temperatures inthe ten miles fable averaged aboxt 122" Fabrenit (30° Centigrade) when the ait ‘emperature reed op taeimometcis wae ehuat 104° Falvent (40° Ceatie frie Similar ts ten az Charen, New Jere, showed temperatures [rv calew 20° tn 29° Foirenket higher than the alr temperature Tt ight day "Tae actus! obsrved cable cenperatre range ia thal ssn (1956) 8 indiented bythe de, cestance mcasueaiets was Ua Crm —¥° Fabeene hit (—20" Conia to 122° Falzenbct (50° Ceatigrade). In terms of {4o, resstance Chges, this amounts to a esstace ehaoge af bout 20 om ‘BELL SYSTAM FECHICASRIOURDAT ibs per mile at $0 ke, the restunoe nl the lover temperature being hovt 96 cbs per ie and at the higher lemperatine about 116 obr=s per mile, ‘This tenons to aboot 210 per cnt vaxiton ftom the meat. “Tn addition tothe wide anmal variations, there are daily varialions of as much ap Si Farenheit et tine, Chat i, almost half es mock a8 the fovial enoon! vardten, “he protical importance of these lege 1 tance changes lies in thet lange ootelbutin to changes ia aeration {a nil bo bronght out more fly fa connoou with variations of attenut- ton with temperstore Cinlergonnd ane bari cates, ofcourse, nt subjected bo uch wide nan! taistioas tad diy variations az slmoat ence eliminated hy the attenuation of heat changes by the sof. Calle fa ducts usually Ties tne below the ceeing lw a he depth ate var Line potest fom {he sommes cat, The note range for cable ia ducts s fom slut Irezaugto sboat 70 eyes, F. Cable ured ony afoot ro underground ‘wold havea consilerily Leger sama terpezature tenge hut great desl ‘suc cable f buried two to thr eet dep, Garver in Fg. show the actal a, esstance variation with requeney and in Fig. exe shown temperature variations of resistance at typical frequenccs for P-gacge toll pei in releageh of stacard oleae CHAKACTBRISTICS OP TOLL TALMEHONRCABLES —— 01 In edition tothe variations with fequency and temperatare ther aro ‘he initia diferoncs between pins om accoune of manusctuing process, se Py Pig Rec ele empresa te, One aueh soure of vavatan jn resistance of pais Is the dierence Tn whe meters caused Ly wear of the ee nsec in drawing wie? The ye ae evel the Maclt o Cooee wedS EES ge eh ae ™ bel SSPE PECUNICAL sOURNAL permissible vinios in diameter for ordinary toil exble& + 1 per cent Iphick mopns dc. eicarce warty 0 aout == 2 pe en Stil enother cone of restance variation is the presence of sal quate tiisof impute the epee mb cow wp 28a reduction of as much a8 Sper cent in tne conductivity, "This eunies ouble ‘a carat, tee eraturevesistence carves inthe Laboratory setup. Folly, ina singe rer! leet the ovtside pai ze longer tan the tae side pei. he total pai-to-pirwraton in zesstarce from he average tt se rea caused all tee factors arate Ze about 3 ver eat mith {standard devia of about 15 pe cert. Pedactence ‘he inctance ofa ccuit frmed by two peat wits elowely quo seitive to thei length is rw onsrs[zansine22 8] ve pe oie) where the wire anstr,and D the separeton ofthe wie are mesure the same unit; the pera, aad i freqaoncy Teco: ‘Ass tall huge, the teres ut assrating cre ta coucentrate on the sursce of a wire redugee Cue magpetic fox sritia the wie aol ‘Seceaes oe inteen‘renctanne of the vere, ‘Ths inlennal inductance is given by sue nm 9 ry Raoaton (5). In like anes, the prosinity ‘efee” enlaces a conceraton of uot enmity inthe adjacent pins of he te wives of J ‘Avothes orm might well be added to formla. () to represent this owbuity cect, "The rosedore ontlined by JR. Corsoy on pyes 625 ed €26 of the Phileophias Mpisine ype of 921 bas been cased out wrth ee esl pve fa an Appendie to Us caper. ormula (a) of the Appendix gives the vio, Kyo fhe == induerce of he pie ess the Sgeomettic insur tance) to tbe tr rlsclanoe of a wire with enosntic retusa, nbich i gives ig « wellknnen forcls (7a 2 the Anpendi, Teil be cen hal the factor intwokes by proiity fel leraces with frequency lute seymttic oa ee vel, depeiting sa the spare- oma te two Witt ex the freon uareasceincedinites- Simla evr¥es tre piven in-on ovensive study ofthe matin! inductance of four parallel (een ofa quad by RS, Hoyt ane Salle Pero Mead” Theis heretical ates egies lcely wilh expesiaontal values gen By R.N- Hunter ural sit AgREagG Re Moe ati of Pa es AST