claim: 1. A method of locating a break in a'fdlilted'first c6n10 ductor in a first pair of conductors from one end thereof by comparing the capacitance bet-vieen one section of said faulted first conductor',and the second conductor of said first,pair with the entire c4pacitdn6-. -bet@@@,6n @' said conductors, said'hiethod,coriiprisiiig the 'steps of;m@astif15 ing the ratio of said icapacitance of @iaid @se@tibn to the c4pacitance of the entire @faulted conductor plus the capacitan-@e between ihird and@f6urth conductors extending from one end of said first pair t.o th.e other, and measuring the ratio of said capacitance of: said 'se6tion 20 to that of said section plus that'between said third and fourth conductors. 2. The method . defined in claim I in, which said measuring ste f d 'd nd bf said ps are per orme at sai (>ne ,c@ conductors and , in w,hich the first-menti6ndd @of said 25 measuring steps is peifoi7med with.the eiid of said,f4tilt(@d first conductor remote ftom,lsaid.6ne..e'n,d c6@@e@ted fo the reirote end, of , said third; cohductbr @@d 't'@e ienio'te end of said secoiid @boiidudtor,is- c6nnq6ted to @ie@,6ie end of said fckurt-h conductor. 30 3. The method defined in claim, I in whi(,h ' said i@6asurements are accon-iplished by,connecting a oot6nti6@@eter .across the@near ends of:said fitst.,and'ttd@,a,6o@duct.or8, said,potentiometer having a,moyable@ tap, Wh6r6by the potentiometer impedances on@cither siae,'6f saia i@p a @id 35 the line ca@acitances corinected to sai,d potentidnieter form a bridge circuit,. sgd itiethod also @ i@clu@ini ifie application of excitation between - a pair, bf opposite ter@t ahd tjie @@djustnient 'of 's'@id minals of said bridge eirc@ul , tap to prov ide a null V - olt@@@ b@iwe@@; t@e @ reriiaining 40 terminals of said bridge. 4. The method defined i'n, claim 2 in which during the second-nientioned of said measuring steps the only connection between the remote,en,ds of said conductors is a connection between said second coriductor and said@fouith 45 conductor. 5. In a method for. locating. a'fault in the form -,of-, a break or the like in a first, conductor by. means @ of 'two measurements made at one end of said first coiiductor so as to conipare the car)acitance between o,,ne s6eti&n -of 50 said first coliductor and conducting mezins ihai extend with said first conductor between the ne,ar - and @ remote ends thereof with the entire capacitance between@ s'aid first conductor and said conducting Xneaiis, said ifiethod utilizing a further coriductor extending between, g@id, near 55 and remote ends -of said first conductor; the stgps of connecting the end terminals of a potentio'meter adross 'Lhe near ends of said first and further coiiductors, said potentiometer hav'mg a calibrate-d mo'vabl6 tap; cbimecting excitatioli means and a detector in such -n@anner as 60 to form a first bridge circuit two -of whose arms are the portions of said potentiometer. on either side of said tap and whose other arms include the capacitances between each o-f said first and further conductors and said. con65 ducting means; adjusting said t4p. to null- said@ first bridge circuit; detetminin,@ at the@inull position of the, eter the ratio of the resistance of said potentionieter between said tap and one end of said pgtentiometer to the total resistance of, said pot6ntiometer@, , formiiag a 70 further bridge circuit by further connectirig togo@lipr the, r-- mote ends of said first and furthcr conductors; adjusting said tap to null said second bi-,ldge circuit dete@mining at the null position of said second bridge eirc'uit the ratio of the resistance of said@ potentiometer between said tap ote iometer to th(, total rcsistanco 75 and one end of said p nt

3,234,459 9 of said potentiometer; and combining said determined ratios to provid,- said comparison. 6. The method defined in claim 5 in which said first and further conductors are in different pairs of conductors and comprisin.- ttie further step c-f connecting in parallel 5 the conductors paired ,vith said first and further conductors thereby to form said conducting means. 7. In a method for locating a faiilt in the form of a break or the like in a first conductor that extends from either side of said break between near and -far ends, said 10 method locating said break from said near end of said first conductor by comparing a first capacitance consisting of the capacitance between the section of said first conductor between said near end and said fault and a second cor@ductor co-extensive with said first conductor 15 between sai4d near and far ends with the entire capacitance between said first and second conductors, the steps of conr,ecting said first and second conductors with a third conductor co-extensive with said first conductor between said near and far ends and with two relatively inversely- 20 variable impedance elements to form a first brid,-e circuit that can be balar@eed to measure, as a first number NI, the ratio of said first capacitance to the sum of said entire capacitance plus the capacitance between said second and third conductors, and connecting said first and 25 third conductors with said impedance elements to form a second bridge circuit that can be balanced to measure, as a s.-cond number NZ, the ratio of said first capacitance to the sum of said first capacitance plus the capacitance between said second and third conductors. 30 8. The rr@ethod defined in claim 7 comprising the furth-.r steps of determining the distance L along the path of said first conductor betweeii said near and far ends thereof, and determining the distance I from said near end to said fault by combining said numbers according 35 to the formula (NI) (N2) (NI) (N2) + (N2-NI) ][L] 9. A test set adapted for location @of -a fault in an elec40 trical conductor, said test set comprising, in combination, a potentiometer having first and second ends and a movable tap, a first tenninal connected to said first end of 10 said potentiometer and a second terminal connected to said sezoiid end thereof, indicatin.- means adapted to indicate the proportio.1 of the entire impedance of said potentiometer between said tai) and one -of said ends, a th'rd terminal, a direct-current source and a first alternatin.@-current source, means for altematively connectin@ said sources between said tap and said third terminal, voltage indicating means connected to said first and second ends of said potentiometer and adapted thereby to determine the voltage across said potentionicter, said first end of said potentiometer being connected to a rletallic chassis in said test set, and a second alternatingcurrent source adaptad to apply a compensating signal between said tap and said first endsubstantially equal in frecuenz:y and magnitude and opposite in phase to the signal appearing across said tap and first end from said first source through stray capacitance in themeasurin,@ system incorporating said test set. 10. The combination defined in claim 9 in which said second altemating-current source includes an oscillator whose output is in series lwith a variable impedance element, tliereby to facilitate adjustment of the characteristics of said compensating signal. References Cited by the Exaininer UNITED STATES PATENTS 1,034,609 8/1912 Friendly ------------- 324-52 1,871,967 8/1932 'Edwards et al --------- 324-52 2,551,942 511951 Greene -------------- 324-52 2,615,076 10/1952 Miller -------------- 324-52 OTHER REFERENCES "Laying Cable in the Forward Area," pamphlet by Signal Corps, U.S. Army, released Jan. 31, 1923, pp. 17, 18 and 19. Northrup: article entitled, "Two Simple Methods for locating Faults in Cables," pp. 278, 279 of "Electrical Review" for Aug. 29, 1903, vol. 43, No. 9. W.,&.LTER L. CARLSON, Primary Examiner. SAMUEL BERNSTEIN, LLOYD McCOLLUM, FREDERICK M. STRADER, ETaminei-s.

United States Patent Of-fice 31234,459 3,234,459 MF,THOD AND APPARATUS FOR LOCATING FAULTS IN ELECTRICAL CABLE LINES BY COMPARING THE IMPEDANCE OF THEI ENTIRE FAULTED LINE TO THE4 liMPEDANTCE OF A SECTION OF THE LINE John A. Brazee, Wall'ngford, Conn., assiguar to Whitney Blake Coulpany, New Haven, Conn. Filed Jan. 2,1959, Ser. No. 784,709 10 Claims. (Cl. 324-52) This inveption relates to an improved method and apparatus for locating faults in electrical cables and the like. More particularly, it relates to a simple, easy-touse method and apparatus capable of accurate location of shorts and opens by comparin.- the impedance of the entire faulted line to the impedance of the section of line between the point of measurement and the fault. It is well known that the resistance of an electrical conductor varies in direct proportion to its length. Further, the capacitance between the coliductor and another conductor in the same cable varies in the same manner, s,,nce the spacing between the two conductors, indeed between all other conductors in the cable, is constant throughout its len.-th. Accordin.-ly, it is possible to locate a fault in a conductor by comparing the resistance or capacitance of the section between the fault and one end of the conductor with a standard impedance of known relationship to the length of the line. Ordinarily, capac;tance measl,.rements are used for location of breaks and resistance measurements for the location of short circuits; and the measurements are made with bridge circuits which compare th-@ impedance involved in a wellknown manner. A popular system for det--cting short circuits is the "Murray loop," in which the resistance between the fault and one end of the faulted line is compared with the resistance of another line in the same cable. The use of th@'s system involves a basic assumption, viz, that the resistance of the standard, i.e., the other line, bears the same relationship to line length as the resistance of the measured section of the faulted line. However, the cross sections of wires of the same gauge vary within relatively wide tolerance limits, ard even conductors in the same cable exhibit arelatively wide range of resistance per unit length, since they are drawn through differ-.nt dies. Furthermore, it may not be feasible to use, as a standard, a line in the same cable, and if another line paralleling the faulted conductor is used, its temperature may be different, thereby addi-@i- a ftirther element of error. Also, in many cases it'is desirable to make the n,.easur@-ments at a test station in a telephone exchange or the lilce, from which lead wires connect the faulted cable to the point of measurement. This requires additional computation to eliminate the effect of lead resistance. A more accurate method of locating shorts is the socalled Three Varley Method. Three measurements are made of various combinations of the faulted conductor the conductor to which it is shorted and ano,her con' ductor. By subtraction and division of the results of the various measurements, one may arrive at the ratio of the resistance from the point of measurement to the fault to the entire TeSistance of the faulted line. The distance to the fault is then computed by multiplying this ratio by the total length of the line. Since the resistance per unit length of a conductor does not vary appreciably along the length thereof, the inaccuracies of the "Murray loop@' are largely eliminated. However, fault location is desirably performed by relatively unskilled labor, and it has been found that numerous errors have resalted from the several measurements rePatenteci Feb. 8, 1966 2 quired and the computations which must be made foi a determination of the distance of the fault. Another troublesome factor stems from the necessity of a balance of the bridge circuit; sometimes balance cannot be obtained, requiring a change in the connections for theparticular measurement. The location of breaks has heretofore been accomplished by comparing the capacitance between a section of the faulted conductor and -another conductor with 10 the capacitance between another pair of conductors in the same cable. As with resistance, the capacitances between pairs of conductors in a cable may vary rather widely, and therefore these prior systems have been subject to large inaccuracies. It should be noted that 15 an error of as little as a hundred feet in the measured location of a fault in an armored cable suspended on poles or buried in the ground can result in considerable additional expenditure in arriving at its actual location and correction. 20 Accordingly, it is a principal object of my invention to provide an improved method and apparatus for accurate location of faults in electrical conductors. It is another object of my invention to provide a method and apparatus of the above character capable of locating 25 both short and open circuit faults. A further object of the present invention is to provide a method and apparatus of the above character requiring a minimum number of simple measurements and computations, and thereby lending itself to expeditious operation by un30 gkilled personnel. A still further object of the invention is to provide a method and apparatus of the above character capable of fa-alt location in a line comprising serially connected conductors of different sizes. Yet another ect invention .s to provide apparatus of the 35 a ove c aracter having hi.-h precision and yet readily constructed of relatively low cost components. Other objects of the invention will in part be obvious and will in part appear hereinafter. 40 The invention accordingly comprises the several steps and the relation of one or more of such steps with respect to each of the others, and the apparatus embodying the f,-atures of construction, combinations of elements and arrangements of parts which are adapted to effect such steps, all ;as exemplified in the following detailed dis45 closure, and the scope of the invention will be indicated in the claims. For a fuller understanding of the nature and objects of the invention, reference should be had to the following detailed description takcn in connection with the ac50 companyin@ drawin-s, in which: FIG'CRE I is a schematic diagram,of a test set circuit adapted 'Lor fault location according to my invention, FIGURES 2 and 3 are simplified sch,-matic circuit diagrams show@ng the operation of the test set of FIGURE 1 55 il locatin- ,t short circuit, and FIGURES 4 aid 5 are simplified schematic circuit diagrams showing the operation of the test set of FIGURE I in locating an open circuit. In ge.neral, my invention operates to locate a fault by 60 comparing an impedance of the entire faulted line with the impedance of a section between the fault and one end of the l,ne. The ratio between these impedance is the fractional distance of the fault from the end of the line at which the measurements are made. Where the fault 65 is a short circuit, the impedance utilized is the resistance of 11- ie lin.-, and where a break or open circuit is to be located, the capacitance bet@veen the line and another conductor is u'tilized. The measurements involved require the balancing of a simple bridge circuit by moving the tap of a potentiometer forming two arms of a simple 70 bridge circuit. The faulted conductors and other conductors form the other arms@ The present invention re-

3,234,459 3 quires but two measurements. One of these determines the ratio of the imp.-dance of the section between the test set and the fatilt to the sum of the impedances of the entire favlted lin-- and another line. The oth-.r determines a ratio relating the impedance of the @.ection in qtiestion to another impea'ance including that of the other line. By simple arithmetic calculation involvin.- ttie two ratios, one may -arrive at a third ratio relatii-ig the impedance of the section in question to that of the entire f aulted Jine. 10 Tuming now to FIGURES 2 and 3, a faulted line 10 is short circuited to another conductor 12 as indcated at 14., Conductor 12 may be another line in the same cable as line 10 or in another cable, or it may be the cable armor or sheiith or even the grour.,d itsel'i', and thus the terin "conductbr," as used herein and i-@i tl-e. claims, is rot re@tricted to any particuldr forni of &Iectrically coilductin.- medium. Furthermore, fhe term "short circuit" includes Iiigh resistdnee pafhs b--tme-.n conductors, sir@ce these, as well as low resistance conliections, are susc,-ptibl' 20 of aet@urate 16cation by the system to be described. Two more @-ond'ui@tors, i6 and 1-9, extend along. the length L of the faulted line 10, and a jui-np-.r 19 is iised to coiinect the 'far 6ndg of liffe 10 gnd c@onductors 16 aiid 18. 'the ptosciit invefition is adcioted t6 dbt-,rmine the distance 1 25 frc@ni cine @7rfd -of tht line 10 to the fault 14 in i manner to b,-; described. Mor@- particul@:rly, a:s sifo,@vii in FIOUR-ES 2 and 3, a test set geiierally indicati@,d :@t 20 inc.udes a potentiometer 22'withatap@24. TI@etapisme(@h@in icallylinkedtoadial 30 26,calibrated in ti@rms of the ratio, R,,IRI of the resistance R,, to the tap 24- arfd oifte @ettd 27 o-f the pote@itiometer 22 ta @the entire res;stapc@e R of the potentiometer. A sensitive current or@ vo-Itiilp-ter IS is connected across pbtenionyeter 22 b@etween a pair of terminals X and Y. The 3,5 test set is completed for the measurements of FIGURES 2- atid 3 by a battety 30 connected betv@een t@ie tap 24 and a'fhird termittal C. Otera,Lion4 of the test set 20 in the locatioji of the fawt 14-'in@line,10.is as follows: 40 A :ffrst met@surement is made, by connecting the appar;atus as shown in FIGUR-E 3 with the Y and X terminals connected to line tO@ and c6ndLictor 16 and the C terminal connect@3d to ccinductor 18. The resulting bridge circuit includes the porti6ns of'the potentiorreter,22 on eithe 45 sidb 6f'ibe tap, 24@ as t -wo of its arins; and thb line 10 and coirdl,ctor l,6 are the other two arms. The bridge is brought to bal@arlce as indi ated' by a. null readin.- of the metor -18,@ and the reading@ of thb @ dial tl-ie information, 50 R, . RlOi+-RlO2 k@RIO,+Rlo,+Rl6@ where RIOI is the resistance of the sec@tion o@.L line lO lying be- 55 tween the Y terminal and the fal-ilt 14, RIO@.is the resistance of the rerrainder of the line 10, ard R16@ is the resistance of the conductor 16. The C terminal of the test set 20 is then moved to the conductor-1-2 for. a second@measureirent, as -shon@in in FIG- 6o URE 2. With this connection, two arms of tbe brid,-e are, @ as before, the two @ portions of th-- potentiometer 22 on either side -of the tap 24. One ofthe'remaining arms consists of the resistanc-. RIO,, and the other, re,-zistances R102 and R16. When the brid.-e is balanced, dial 26 65 indicates, R@ R10 k@RlOi+RlO2+RI6 .Thus, 70 K RIO (3) P@ RIO@L +Rl@O2@z The latter fracti6n 11L is the fractional d;stance of the fault 14 from the near end of the line 10 connected to 75 4 the test set 20. Multiplication of this fraction by the known length L of the failted line gives the actual dista-@ice of the fault .14 in units of lengtli. It is seen that only two readings on the test set 20 are ired for location of the fault 14. 'I'hese readin,-s are =y tiken froni the dial 26, and between readihgs orily one connection, that of the C te:-jiiifial, is chan@ed. Ftirthei7more, the computatic-n reqtiired to convert the readings into the locat-'Lon of the fault i8 4@lor@i'entary in h@ture and easily handled by liersonnel hbving niinimumtraining. In some cases, the fdiilted lilie 10 may comprise two conductors 10,a and 10b (FIGURES 2 and 3) of different sizes or gauges. In that event, the line would, be ternicd a mixed gai-ige line having two. mixed gauges as shok-n. The locaiion of a fault iii a mixed gauge line, by i@iy inverition, involves th@- same meastirements as those cles6ribed abov& for a siiigle -aug6 lir.6. the connections aiid ieadings taken ar& e@a6fl-@ th6 sam-e. However, the c6mt@iitation is sor@icivhat d iir6i6nt. ii sectio-n 10a close.st to the te'@@t @@l-t 20 w-er6 sh6rted, ihe measured value o@' K, Ps dettrffii-.ibd ab'ove, @Ni6tild be less than (4) @vherb LI + ML2 LI is the length of the near seption 10a, L2 iS the lehgth of the far section 10b, and is the nominal resistance per unit lehgth for the gauge size of the secoiid -gection d I ivided by the noiiiinal resistance per unit I . ength of the gauge size in the first section, aiid the distatice to the faul . t from the test set 20 would be K (LI +ML, (5) (P) In the example show-@i, fai-ilt 14 is in th@ second sectioi 10b, and this wotild be iri'dicdted by d v@lue of K greater than LI +ill L, The distance to thb fault would be gi@en by )(L. +ML@) -+LI (6) p m m Again, the compiitatiohs reqtiired are simpler than those used in other fault loc,@tinsyste s c ble of r asonable accura@y. m apa e Sirftilar expressions rlay b6 d@tived for location of faults in lines having a-@iy number of inixed gauge @ For m'axi@ rnum acctiracy, how6ver, I Pt@fer to d6t@rriiine first the section containing the f@ult froni Expression 4, or similar expressions for groater number of iiiixed gauge's, then discbnnect the section ii ordet t6 mdke a direct m'easu I rernent on it.-, In FIGURES 4 and 51 I have illustrated a line 32 lia@ing a fault 34@ in the form of d break or op6n circuit to be located by use o'l the test set 20. Conductois j6, 39 and 40 ext,-,nding between the ends bf the faulted line 32 are also uq.-e,d in the required mea@@tirem'ent . The test set 20 @ utilizes an alternating current geri6ratbr illusttati,@@ly indicated at 42 in @lace of the battery 30 of FI(jiURES 2 and 3 in order to make use of the capacitan@e@ between certain of the conduttors in determining the distance 1 to the fault 34. More particularly, between con,ductor 36 and the near section 32a of line 32 there is a capacitance Cl, an@d betnveen conductor 36 and the far section 32b of the faiilted lirie, there is a capacitance C2. Further, there is a capacitancd C3 b6tWeen coiiductors 38 atid, 40. The X and Y terrriinals of the test set 20 are c6hne@ted to section 32a and conduct6r 40 ' respectively' and terming C is connected to conductors,36 ahd 33. A jumper 44 con-nects the ends of conductors 36 and 33 remote from

@the test set, and for one of the measurements (FIGURE 4), a second jumper 46 connects the far ends of line 32 and tondiictot 40. Thus, in FIGURE 4, a bridge circuit is fonnea having as two ; of its arms the portions of the potentiometer 22 on 6ither side of the tap 24. One of the othei qrms comprises the capacitance Cl and the other includes capacitances C2 and C3. Accordingly, when the bridge is brought to balance as in,dicated by the meter 28, the dial 26 indicates, R@ el -k@-U-,+C2+C@@K (7) Jumper 46 is then removed and a second measarement taken with the connections shown in FIGURE 5. With the bridge at balance R. =: c I = p (8) R Cl+C3 It can be shown that PK ci (9) PK+(P-K) Cl-+-C2 The right-hand side of Equation 9 is the ratio of th-. capacitance of section 32a of the f aulted line .32 to the eiitire capacitance bet-,veen the line and conductor 36 an@d is therefore equal to the fractional distance, IIL of the fault 34 from the near end of line 32. Accordingly one may determine the actual distance to the fault 34 @y performing the arithmetic op--rat;ons on the measured values of K and P indicated in the leit-hand sid-. of Equation 9 and multiplying the resulting fraction by the total length L of the line 32. The above method of locatin.- breaks in L,'@lies is substaiatially more accurate than prior methods, since th-capacitances Cl and C2 are determined by the insulation between the line 32 and the condactor 36. The insulation around a given coti-ductor is essentially constant over the length thereof, and tberefore the capacitance betweeii two s,,ich conductors will vary almost exactly linearly with th@- distance. Accordin.- IY, the fraction ci CI+C2 of the total capacitance (CI+C2) is an accurate indication of the distance to the fault 34. On the other hand, prior methods have utilized the norninal capacitance per unit length between c6nductors or have compared various capacitances along the faulted conductor wit the capacitance between two conductors of an entirely different pair. ]3ecapse of the v,ar,'.ation in conductor insulation p--rmitted by.the tolerance ran.-es, considerable inaectiracies resulted from these pribr systems. The two measurements described above may also be used in locating a break iTi a mi,- ,ed capacitance line, i.e., a lin.e havin.- se.-ments of -different capacitances. For example, if the line comprises segments of two different capacitances, and the break is in the near se,@ment connected to the test set 20, the m@ ea sured valite of K will be less than, (10) LI+NL2 where LI is the length of the near segment, L2 is the length of the remote segment, and N is the nominal capacitance per unit length between the line 32 and conductc)r 36 in the -near segment divided by the nom-'@nal capacitance between the Ene and conductor 36 in the remote segment where the hreak is in the near se-ment, the distance thereof from the test set is given by PK [L,+NL2] PK+P-K 3,23-A-,450 6 If the break is in t he remote segment, the distance is g - iven by 1 PK L x ,) (L@ +IVL@) iv +L (12) 5 For maximum accuracy, it is again advi@able to locate th" section in which tl-.e break has occurred and then isolate this section for further measurement in the manner described above for a one-gauge line. The jumper con10 nections 44 and 46 at the remote ends of the conductors may, of course, be made by latching type relays energized from the end at which the test set measurements are made. Ttirning now to FIGURE 11 shoiving the test set in 15 gr,,ater detail, the battery 30 is connected in series with a variable sonsitivity ad:usting resistor 48. Preferably, the generator 42 takes the form of an oscillator connected between the C terminal ond contact Sla of a switch SI, while battery 30 is connected to a contact Slb. The movable 20 contact or wiper SI., of switch Sl is connected to the sliding tap 24 of Dotentiometer 22, and thus the excitation of the bridge circuit of which potentiometer 22 is a part r@iay be changed from altemating to direct-current by proper s-.ttin@ of this switch. 25 The m-.tr'28 is connected between the X terminal and the movable contact S2, of a switch S2. A contact S2b iS connected to the Y terminal, and a contact S2,, is connected to the output of an amplifier 50 whose input is derived from the voltage between the X and Y terminals. 30 The X terminal is grounded to a chassis in the test set as indicated at (@2, the latter b--ing isolated from the actual earth ground. Illustratively, an oscillator 64, whose frequency is the same as that of generator 42, is connected in series with a capacitor 66 between the tap 24 and 35 chassis ground 62. The amplifier 50 preferably includes a suitable rectifier so that the meter 28 connected to its output may be a direct current meter of suitable construction. Thus, with switches SI and S2 in the positions shown in FIGURE 1, test set 20 is arranged as 40 shown in simplified form in FIGURES 4 and 5, adapted for location of a break in a conductor. When the switches wl-,ich may be ganae@d for ease of manipulation, are in the other or "b" positions, the test set circuit of FIGURES I and 2 is achieved. 45 The purpc)se of oscillator 64 and capacitor 66 is to eliminate the effect of the capacitance between the test set and actual ground. Stray capacitances always exist between the test set chassis (.and the X terminal conne,-tod thereto) and the earth,!and also between the oarth 50 and the conductors connected to the C and Y terminals. In fact, the conductor 36 or 38 (FIGURES 4 and 5) inay m some instances be the earth or cable armor in electrical contact therewith. Thus, there are resultant stray @capacitances between the X terminal and the C and 55 Y termirals. ReferTing to FIGURES 4 and 5, it is seen that the stray capacitance (not shown) between the X and Y terminals is across the entire potentiometer 22 and therefore does not affect balance of the measuring brid.-e circuit. However, the stray capacitance (not 60 shown) between the X and C terminals is in parallel with the capacitance Cl and therefore will affect bridge balance and result in inaccurate measurement of the ratio 65 ci Cl + C2 Returning to FIGURE 1, the oscillator 64 operates at the same frequency as oscillator 42 and in phase opposition therewith, as indicated by,the polarity signs. Ac70 cordin,-Iy, oscillator 64 applies a signal between the X terminal and tap 24 in phase opposition to the current from generator 42 between these points through the stray capacitances. Capacitor 66, is adjusted so that the two currents are equal in magnitude, thereby cancelling the 75 effect of such capaeitances. In practice, I have f6und

7 that @capacitbr 66 may @be set i'n the factory and remain fixed thereafter in most applications. It will be appar6nt :ffiat oscillator 64,and generator 42 may be combined, so long as ihe f6@bective outputs are isolated sufficiently from each other. Th. capAcitEince cancelling feature of the test set 20 is also-of importance in making certain m-@asurements, 6ther than -those describ@d above, wherein the test set is i@sed as a strdi ght ezip@citance,bridge for inore direct measurement of the c4pacitdnce ratios in @a broken line. This latter ty@e,of is often made where a I,m;e has tWb bteaks, ahd the distance to one from One endof,theline,is,known. ReferringtoFIGURES4.and 5,@it shbuld be,noted at this p6int that voltages Nvill be induced,by the excit@ation from,the ge-nerator 42 in other conductors (not shown) capacitively coiipled to those ii ed in the,ineasuremeiits. However, these si.Qnals have no effect bn the measurements taken with my system iiiasliiu@6h, @s . they ate, self@cancbll'mg in nature. .The fteqiieh:cy, I of the generator 42 shbijld be sufficieiiily 16w that the re@ctaaees of the measured capacitaiic@6s,are subst@ntially I ss e tb@an the resistances and self indu6taiic@--s 6f the c(jiiductors used in the measurements, but high enough to provide unbalance currents of a .magnittide.easil@ aniplified t'o dn dece@table level by the am'plifi6r 50. Further, tjie frequehcy @hould be different from that 6f Any stray spurious signals which may be @icked u@p by the iconductors used in' the measurements, inost nbtably, the siandard power line frequency of 60 cycles per second and harmi)nies thereof. I have found that .100 cycles per second is a. s itable excitation freI u quency in view of these factors. M.oreover, when such @,spurious si@na@ls are received, means should be provided to differe@tiate them fr6m the bridge unb,alance signal; ,and, @accotdingly, earphones may be siibstituted for the meter 2&@and the. amplifier 50, whereby the bp!efator may --determine by ear when an actual bxidge ilull is obtained. Thus, I have described an improved system for locat ing faults -in elect , ri , al co c nduetbrs. The faults are located by comparin@ an impedance of the section of the faulted,ae between t . he f ault and the test set to the impedance of the. entire line, thereby affording a high de.-ree of accuracy. Where the fault is a shoi-t circu'it or siniilar failure, the impedance.utilized is the ele6triIc,al resist-ance of the line, and.whete. the. fault fs a break or other failure ha-@'mg a, Eke -effect, tl-le impedance utilized is th@- capacitance @b6tween tbe conductor and other @conductors extending @arallel theieto. In either case, but two measurements are.require@d, both of these ,being from,t-he:same -end.,of the,line. From the rc@adings. obtained, the fractional distan@es of the,faults from the, measured ends. of - the faulted lines may easily be computed. T-he @ method . is thus suse6ptible of practic@, by relatively @ untrained, iinskilled @ersonnel. Moreover, th e a@pparatus, described @above is @siniple in nature and of relatively low cost.,c6nstrudtion,@vhile provi4ing accuracy I of fault locating, including compensatioli for.stray capacilances. It, is easy to. set: up. in the @ field aild @imple to operate. Further, it @ will be n6ted , fhat th6 only. high precision in,the.iest set,20 is@the p6tentioincter,22 vhereas in prior 'mstruments, precisioii resist6rs of @ both .a fixed and variable nat-ure were required. It is well known that a high precision potentiometer, providing a -voltage or resistance ratio rather than an actual resistance value, is much less - costly to cojistruct and calibrate than:a variable resistor:@of the same -precision. It @@vill- thus be @een that, the objects set forth above, aiiiongthose-made apparent from the preceding description, are dfficlently @ttained and, @ since @ certain changes may be mad-. in carrying out the above i-nethod and in the iconstruc@tion s6t forth without departing from the scope of th6 invention, it is intended that all rnatter contaiired in the . e:description or show n in the accom3,234i459 8 paidying drawings shall be inferpreted:as illustrative a-nd not in a limiting sense. It is also to be understood that the followirig,@claims are iiitended to cover @all of the generic and;ip6.cific.features of the invention herein described, aiid all statements of the scope c@i' the invention which, as a niatter -of language, @might be said to faU therebetween. I