UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The echo ranger : a fault locator for power cables Naylor, Thomas Kipling 1948

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1948_A7 N2 E2.pdf [ 4.95MB ]
JSON: 831-1.0106980.json
JSON-LD: 831-1.0106980-ld.json
RDF/XML (Pretty): 831-1.0106980-rdf.xml
RDF/JSON: 831-1.0106980-rdf.json
Turtle: 831-1.0106980-turtle.txt
N-Triples: 831-1.0106980-rdf-ntriples.txt
Original Record: 831-1.0106980-source.json
Full Text

Full Text

THE ECHO RANGER A F a u l t L o c a t o r f o r Power Cables by Thomas K i p l i n g Naylor A T h e s i s Submitted i n P a r t i a l F u l f i l m e n t o f The Requirements f o r the Degree of MASTER OF APPLIED SCIENCE In the Department of MECHANICAL AND ELECTRICAL ENGINEERING Approved: In charge of major work, ^ieacr'of Department". THE UNIVERSITY OF BRITISH COLUMBIA September, 19k& 6) CONTENTS SUMMARY -Page ( i i i ) I Introduction 1 II Review of Literature 3 III Investigation 6 A. Theory of propagation b 1. General differential equation 6 2. Pulse shape 8 3. Solution for an in f i n i t e line having constant parameters 9 Examination of parameters 14 5» Approximate solution including skin effect. 19 B. Reflections 21 1 . Theory 21 2 . Terminating Impedances , Z$ 3. Effect of change of velocity 28 C. Apparatus 30 1. Description 30 2 . Accuracy - 31 "3. Range 32 4. Circuit mechanism 33 5. Operation 41 a. I n i t i a l adjustment 41 b. Trace expansion 43 c. Measurements ' 44 6. Sample calculation. 43 IV Discussion 47 A. Advantages 47 B. Limitations 47 C. Suggested improvements 48 D. Preliminary note on flash over 49 V Conclusions 52 A. Theoretical 32 B. Experimental. 53 VI Literature cited Page 54 VII Acknowledgements 58 VIII Diagrams 59 A. Apparatus 59 1. Assembly 60 2. Block diagram 61 3 . Circuit diagram 62 4. List of parts 63 B. Echoes from a power cable I 3 S 9 feet long 64 1. Panoramic viexir (compressed sweep) 64 2. I n i t i a l pulse 65 3 . Echo from a Joint 65 4. Transformer-tap echo 66 5. Transformer-tap re-echo or a second joint 66 6. Multiple echo reflected from sending end 67 7. Open-circuited far end of cable. 67 THE ECHO RANGER A F a u l t Locator f o r Power Cables, ft t3 S T PL A e~T S U M M A R Y The l o c a t i o n of f a u l t s i n low-attenuation c o a x i a l cables and open-wire l i n e s by the use of the echo-ranging techniques of radar prompted t h i s i n v e s t i g a t i o n of a method to ac c u r a t e l y l o c a t e f a u l t s i n underground power cables. As the pr o p a g a t i o n - v e l o c i t y of disturbances on a smooth l i n e o r cable i s constant, the time delay between the tra n s m i s s i o n of a pulse i n t o a cable and the r e c e p t i o n of an echo from an i n t e r n a l d i s c o n t i n u i t y i s p r o p o r t i o n a l to the distance to the d i s -c o n t i n u i t y . The low inductance and h i g h d i e l e c t r i c l o s s e s i n power cables attenuate and d i s t o r t the pulses. This d i s t o r t i o n l i m i t s the accuracy and range of equipment which must measure time i n t e r v a l s to the nearest 3 x 1 0 " " ° seconds. B a s i c a l l y , the Echo Ranger c o n s i s t s of a p o r t a b l e low-voltage impulse generator combined w i t h a timing O s c i l l a t o r and a delayed high-speed sweep on a commercial split-beam o s c i l l o s c o p e . A high-power hydrogen t h y r a t r o n d e l i v e r s 0 . 1 -microsecond pulses of f i v e k i l o w a t t s (peak) to the cable. Although the range of the apparatus now constructed i s only two mil e s on power cable, f a u l t s at l e a s t f i v e m i l e s away should be v i s i b l e . The minimum r e s i s t a n c e of a detectable s e r i e s f a u l t i s about f i v e ohms and the maximum r e s i s t a n c e of a detectable shunt f a u l t i s about 2000 ohms. Without m o d i f i c a t i o n , the Echo Ranger can be used on overhead l i n e s up to four m i l e s long. On a 1044-foot p i e c e of RG-gU p o l y e t h y l e n e c a b l e , two 100-ohm shunt f a u l t s 20.2 f e e t apar t were l o c a t e d w i t h i n 0.63$. On a t h r e e - c o n d u c t o r o i l - f i l l e d l e a d - s h e a t h e d power cab le 1389 f e e t l o n g , a t ransformer tap 424 f e e t away and a J o i n t 320 f e e t away were l o c a t e d w i t h i n 1.2$. The apparatus can be r e a d i l y m o d i f i e d to d e l i v e r l 6 - k i l o v o l t 5- m egawatt (peak) p u l s e s to i n i t i a t e ^ a n Vrcf^at i n c i p i e n t f a u l t s . The power to h o l d the arc must come from a superimposed power supply such as a kenotron set o r the normal l i n e v o l t a g e . F u r t h e r re f inements w h i c h i n c r e a s e the accuracy and range w i t h o u t s a c r i f i c i n g s i m p l i c i t y of o p e r a t i o n c o u l d be a p p l i e d to advantage. T . K . N a y l o r , U . B . C. September, 1943. THE EOHO RANGER A Fault Locator for Power Cables. 1 INTRODUCTION Recent work on the location of faults i n low-attenuation coaxial cables and open-wire lines by using the echo-ranging techniques of radar has prompted this investigation of a method to accurately locate faults i n underground power cables. EchQ-ranging depends on the constancy of the velocity of propagation of disturbances on a smooth cable. The time delays between the transmission of pulses fed into one end of the cable and the arrival of reflections from Internal discontinuities are proportional to the distances between the sending end and the discontinuities. Then, i f the distance to one of the dis-continuities i s accurately known, the locations of the remainder can be easily calculated. Although much power cable i s truly coaxial, skin effect and high dielectric losses absorb most of the high-frequency components of the pulses used in echo ranging. The resulting distortion of the pulses limits the accuracy and range of the measuring equipment. , The intermittent or steady discontinuities in the smooth caMe may be junction points, open circuits, short circuits, or faults. The insulation may be carbonized and the conductor more or less burnt away at a fault. The apparatus should be portable, rugged, sensitive, accurate, simple, and inexpensive. Provision should be made to prevent extraneous disturbances such as induced voltages on the sheath, electrochemical emf's at wet faults, or accidental energizing of the core from affecting the apparatus or the operator. If possible, the measuring currents should not destroy the evidences of the cause of the fault. If the fault w i l l appear only when rated voltage i s applied, apparatus should be devised to locate the fault while i t i s arcing over. To locate a fault within 10 feet on a 10-mile cable, a time delay of 10 ' seconds must be measured to the nearest —21 3 x 10~ seconds. 3. 1 1 REVIEW OF LITERATURE • Much excellent work has been done on methods of 1 locating cable faults. Mr. Savage recently summarized many effective methods. In bridge or loop methods, the resistances or capacitances of an unknown length of faulted cable and a known p length of good cable are compared. The abrupt change in the magnetic f i e l d around the cable at a discontinuity can be detected by search..coils taken along the route of the cable. Shielding by the steel armouring n u l l i f i e s this method. Chopped or modulated currents fed into •5 the cable produce a traceable magnetic f i e l d . ^ The thump produced by Imterrupted arcing at the fault can be heard i f sensitive microphones are placed on the ground over the burled cable.' The rumbling of t r a f f i c w i l l swamp weak sounds. Although portable radio receivers are sometimes used to pick up the static from the arc, the shielding effect of the sheath 5 i s appreciable. 1 See reference 2 7 page 5 & 0 - 9 3 2 See reference 2 7 page 5 8 0 . 9 3 paragraph 2 . 2 3 See reference 2 7 page 5 & 0 - 9 3 paragraph 2 . 3 \ See reference 3 page 2 2 5 See reference 1 1 page 1 3 2 If the lead sheath i s punctured, the odour of burnt Insulation permeates the soil and cable conduits near the fault. The distance to a single fault can be determined by a method i n which the linearly varied frequency of the applied sine wave i s beaten against the echo. The frequency of the beat w i l l depend on how much the transmitter frequency changed while the echo was travelling to and from the discontinuity.^" Standing waves produced when the cable i s a multiple of a quarter wavelength of the applied sinusoidal voltage provide quite a simple and accurate means of locating gross faults on p long cables. On overhead open-wire lines and low-distortion speech or television cables, echo-ranging devices using single pulses or bursts of high-frequency oscillations have proved to be quite accurate and sensitive.^ Their rapid development during the war parallelled the evolution of sensitive radar equipment. A method of locating faults on overhead lines by using the arrival times of echoes and re-echoes of the surges produced at a flash over has been proposed by Messrs. Stevens and It Stringfield.' Messrs. Margoulies and Fourmarier have used 9 0-^v 1 See reference 26 page (22 2 See reference(20 page 46 ( 5 3 See reference(IZjpage it (2o also reference 1 page 5^1 4 See reference 29 77 1 impulses to locate high-resistance faults on overhead lines. When the faulted section of cable has been exposed, the shunt fault can be located within one inch 'In ten miles by a core-to-sheath potential-difference test developed by 2 Mr. Savage. The vacuum-tube voltmeter connected between the core and sheath at one end of the cable drops to zero and then reverses as the battery passes the fault. In a somewhat less accurate method, the battery i s connected to the end of the cable while the voltmeter probes are 3 moved along the sheath. Intermittent carbonized faults can sometimes be located i f the cable i s vibrated and the fault i s used as a 4 carbon microphone. As most cable faults appear at taps, Joints, manholes, or places where the earth has been disturbed, rough methods of fault locating often suffice. However, precise measurements must be made i f the fault i s in a length of cable buried directly i n the earth or under pavement. 1 See reference. 19 2 See reference 27 page 5^ 0-93 paragraph 2.5.4 3 See reference 27 page 5^ 0-93. paragraph 2.4 4 See reference 27 page 58O-93 paragraph 2.S.1 I l l INVESTIGATION A. Theory of Propagation. General Differential Equation . Consider a short length of line ~fr?c ^ 2 a F - ^ i , , = W * ^ " L_ : r where resistance ^t, = ohms per unit route length inductance &^ ~ henries per unit route length capacitance c = farads per unit route length conductance y = mhos per unit route length operator = 9e am In the limit as y^c—*- O and — -jir = ye + c = (y •+ c/>)e €) Take of fS) and substitute in (^) and 3L of (y) and substitute in (3) and. 1. See reference 6 page 97 Treat (s) and '(^)aa ordinary linear differential equations. Then the solutions are: Where A, B, C, and D are arbitrary functions of t_alone Substitute and (S)ln (3) and ^ a n d compare coefficients ovf £ and then (5 *«* & = -JET & " l e t U+4/> _ /ST/^ESZ. / /2-. . Then = <£, A • -t £ & '8 e . 2. Pulse Shape. If the applied voltage i s sinusoidal, the steady-state solutions of the general propagation equations result when i s replaced by ^cd . By means of Fourier series, a piecewl'se continuous function can be expressed as a sum of sines and cosines. As cd w i l l be different for each term of the sum, the phase shift and attenuation of each component w i l l not be the same along the line. As pulses are essentially discontinuous and f i n i t e , dis-tortion w i l l be excessive unless the frequency range Is kept very narrow. The ideal pulse must also be simply produced at a high enough amplitude to override background interference. Although the peaked wave shape of the discharge current from a condenser does not cover as narrow a band of frequencies as the error function, i t can be quite simply produced at high amplitude. The wavefront i s steep enough for precise measurements to be made to the toe of the pulse, yet the peak i s not so badly distorted that i t i s unrecognizable. A narrow pulse vrf.ll pass quite easily through the high-voltage capacitors which couple the apparatus to a li v e line. In practise, the wave shape of the applied voltage approximates yv, where ft? i s about 6 x 10 The factor, , i s used because the time of rise i s about 10 seconds. 3. Solution for an Infinite Line (Constant parameters) ffor f i n i t e solutions of the general differential equatio when zxi—a- ' • o Then ^ . e ^ A / Applying a voltage £ a t O Let ^Ct) - i ^ OPERATOR FORMULAE ^ S jf® - * (/-J)/®- '' (shifting) 7 where - i y i s a unit function delayed until 1. See reference 10 page 5 5 2. Bee.reference 8" page 220 /o Multiply both sides of (QJ by ^ s (y) frfc /eft- s^e of ($) ^ s ' ^ . ^ / a c e ^ ^ s Solve equation by using equation(^jto shift g^^to the l e f t . - 5 ? y » ~; - - ^  V . #• Apply formula ^2) to equation ^ J . s^<x-ir) and 3^ ^p-For a graph of JT@) ~ *J>Q@^ see reference 1^ page 22^ The following solution f o r £^is based on a method outlined i n reference 6 pages 100-2 . From equation (y)of the derivation of the general d i f f e r e n t i a l equation we. have '_ , _ -I ^_ In Equation s h i f t c5" one term to the r i g h t . r^e Substitute equation (=^v into equation The derivative of a product contains two derivatives (a)and^(b) y^ Also j Z L f & - e ^ l f s ^ ' y M-sC] *7-: Hence j k ' e ' ^ [ J f ^ * ) ^ *&>J '*S. See reference 1E> page 42 IX Then equations 2">, 2 5 , 2 6 , and 23 combine to give The f i r s t i n tegrand vanishes everywhere except at T~~/i^ The second i n t e g r a n d does riot c o n t r i b u t e , u n t i l ~£ ~ ^ hence ^ As £ /^^c) i s n o' t m u l t i p l i e d by a u n i t f u n c t i o n to cause i t to vanish f o r 7^0 , then — G ^ 33. For graph,of •^~/(^') ~ ~ ^ . ^ / ( ^ ^ ) s e e reference i l l - page 2 2 ^ 1 . Reference 1 5 pages 3 9 5-7 J3-Approximate wave shape a f t e r a r r i v a l * Express equation (33) as a f u n c t i o n of/ 7""" ^ As a f i r s t approximation ^j^~^lk~ ' J°- O^0^2-, 2. Normally the pulses are comparatively short, i . e . #7 i s l a r g e and M > ° < - ^ /• Reference 14 page 224 E x a m i n a t i o n of Parameters . Power cab le I s made up of one or more s tranded copper conductors separated from one another and from the e n c l o s i n g l e a d sheath by i n s u l a t i o n which may be o i l - o r b i tumen- impregnated paper , r u b b e r , v a r n i s h e d cambric , o r i n e r t gas . For mechanica l p r o t e c t i o n the l e a d sheath may be armoured w i t h an o u t e r l a y e r of s t e e l s t r a n d s . To a f i r s t a p p r o x i m a t i o n , the cable may be con-s i d e r e d to be c o a x i a l . For more p r e c i s e c a l c u l a t i o n s , the g e o m e t r i c a l f a c t o r s determined by Simmons 1 should be a p p l i e d . Ware and Reed g i v e formulas f o r r , 1, g , and c f o r low- f requency or d i r e c t c u r r e n t which i s assumed to be u n i f o r m l y d i s t r i b u t e d over the c r o s s - s e c t i o n of each conductor . At h i g h f r e q u e n c i e s , s k i n e f f e c t and d i e l e c t r i c a b s o r p t i o n l o s s e s change r , 1, and g . As the p e r m i t t i v i t y i s r e l a t i v e l y independent o f f requency , the capac i tance can be c o n s i d e r e d c o n s t a n t . 2 ~~~ ^ Jit^ — _yC"" f a r a d s per meter A/ where x i s the v e l o c i t y of l i g h t , 3 x 10 m/sec i s the r e l a t i v e p e r m i t t i v i t y of the i n s u l a t i o n . 1 See r e f e r e n c e 2.8 2 See r e f e r e n c e 30 pages /Oj //" As skin effect w i l l concentrate most of the current at the outer surface of the core and the inner surface of the sheath, the self-inductance due to flux linkages in the metal conductors w i l l be reduced. However, the flux in the insulation w i l l s t i l l link the core. This interconductor inductance can be calculated resistance by decreasing the effective cross-section of the con-ductors. The variations i n the resistance and self-Inductance w i l l be evaluated according to the method outlined i n "Surge Phenomena"."'" -7 henries per meter where /*f i s the relative permeability of the dielectric. For most dielectrics ^ i s unity. At high frequencies, skin effect w i l l increase the From solution of the equations: Ramo and Whinnery show that, for a single conductor, where i s the radius of the conductor in meters radians per second i s the absolute permeability of the conductor or henries per meter 1 See reference 6 page Z-2-2-2 See reference -2-5" page -2_/2-y £ i s the d-c resistance in ohms of a one-meter length of the conductor. ^ i s the re s i s t i v i t y of the conductor i n meter-ohms la the effective resistance i n ohms per meter length of the conductor ^7. i s the Internal self-inductance of the conductor i n henries per meter. For the very large values of Cd encountered i n surges, the Be3sel functions can be approximated by formulas l i s t e d on page 157 of reference /O . <z<£> ' ST A j - i ) = , f T—— f t ' * . + 2-As for sinusoids and ^ are equivalent e Dielectric losses in the insulation of power cables account foremost of the apparent conductivity. Ordinary conduction losses increase with temperature; but at the / 7 temperatures normally encountered, they are s t i l l very small. 1 Mr.. H.H. Race has found that the total loss per cycle in the o i l used in "solid-type" cables f i r s t increases and then decreases with frequency. The frequency at which maximum loss occurs increases with temperature. The apparent gain i n accuracy does not warrant the added complication of approximating an equivalent shunt conductance, for the operational equations w i l l be solved only approximately. If the variation i n g i s neglible, then Insulation Velocity of propagation gas polyethylene o i l - f i l l e d paper compound varnished cambric rubber 1.0 2. 25 3-5 3-7 4.5 6 9&3 f t . peryte-Bec 655 525 510 460 400 •3. See reference g^L page^S' /8 For eccentric single-core cables, Brown^gives the surge impedance as: 1 See r e f e r e n c e /3 5. Approximate Solution Including Skin Effect, ^1 (Jacottet Method) 2 In.general, for an Infinite line where From section \ a, /—T,— / y? As a f i r s t approximation, assume (f i s negligible. / Let iU~r 1 •2-•4. 6-Expand hy the binomial theorem For a large rate of change of 6 i.e.,- large ^  , ignore a l l but the f i r s t two terms. Let ^ a ~TA^^~7t~ 7 Then ^ - £ £ -ff jcrg) 8. 1. Compare solution in reference 6, page 122. 2. See equation 3 of Section 3. £0-For =• ^  £ 3-lo. If. /A-4-ex - /rs-fc sT**^ (As before, «=• O ) NOW e'^-yij -y(z^\y r£-«r /s. ••• < - */£ J£ -fuz 4T7-4 1 . 2. 3-For e*f< 6 see reference lK page 25. Reference 10 page 199 Reference 10 page IKS B. Reflections 1. Theory, For a f i n i t e line closed at x « o by an impedance^ and at by an impedance we have \ 1 T U J, >-In general At'• J e = A + 3 / i f l e t r Rearranging 0 Rearranging let /*-! '- ' " 4 Then from®, 3 = ~ < ^ -^4 r -/4 1. See reference 115 page 1°9 ^2-Hence from£g?<£© 3 = (/-/t^/ £~a-AJL)^^/) ^ ^ Then fvom®^© A " (r£~ X^ X('*0 ^ ^ Expand denominator as (/~ = /+ -f ••• Similarly For an i n f i n i t e line ~& -= £ ^ y / /3 -Jx ^~ £* ~~ 6 j (('+0 " 4 - r ^ JZ3-The voltage and current In a f i n i t e line are sums of disturbances which have travelled successively longer distances by being reflected at the discontinuities at the ends of the line. While running on the smooth line, however, the dis-turbances act as If each were alone on an Infinite line. This statement i s apparent when i t i s noted that the equation for the increment in _g or i due to the second term i s of the same form as that of the in f i n i t e line except that x i s replaced by (2 - A - x). As the solutions of the operational equations for the Infinite line are zero before t = ^  , then the solutions of the f i r s t echo equations w i l l be zero before t = 2^-x . v Consequently, each echo i 3 delayed by a time which varies directly with the total length of i t s route. The problem then resolves Itself into producing the i n i t i a l pulse and accurately timing the arrival of the echoes. I -J-2 L IF If the line continues on past the shunt impedance, then replace zfr^ i n the above formulas by _^  The input voltage to .if*^. w i l l be the same as across , viz: 5 - 7~~Z 3 £ where and e then the input current to j£ i s As the denominator only adds in the multiple echoes from the sending end and — — ^ — i s the voltage incident on the discontinuity, then /CyU.^ i s the voltage reflection operator and i s the refraction or transmission operator. Tables of these operators for various circuit arrangements may be found i n reference j/*^ 2. Terminating Impedances The type of echo and the attenuation produced by reflection and refraction operators can be estimated i f the operator impedances are replaced by surge impedances. The voltage-reflection operator: If the cable i s open-circuited, = / If the cable i s short-circuited, S T ^ - & If the cable i s terminated in i t s surge impedance, This approximation holds quite closely providing the incident pulses are of short duration."'" Hence for cable terminations less than the cable surge Impedance, e.g., shunt faults or taps, the polarity of the echo w i l l be the reverse of the incident pulse; whereas for cable terminations greater than the surge Impedance, the echo w i l l be of the same polarity. The transmitted or refracted pulse w i l l resemble the incident pulse. For maximum efficiency of reflection, the cable should be short- or open-circuited and not terminated in i t s surge impedance. Now consider a uniform cable containing a discontinuity somewhere along its. length. As — JEF :^ (same type of cable) then the voltage reflection operator f-7 f'^ ^ /-f-z 1 See reference £ and the voltage refraction operator / If g 3 i s large, then very l i t t l e energy w i l l be reflected. Hence a relatively high resistance fault may go unnoticed unless i t s echo i s highly amplified. If — i s small, very l i t t l e energy w i l l be refracted past the low-impedance fault. Echoes from discontinuities beyond the fault w i l l be doubly attenuated as they must pass the discontinuity twice i n order to return to the detecting apparatus. As the echo reflected from an open circuit i s of the same polarity as the Incident pulse, the voltage at the open circuit at the instant of arrival of the pulse w i l l be double the voltage of the arriving pulse. Consequently, the most sensitive detector should present an open circuit to the cable. As the thyratron i s a unilateral impedance which varies with time, the value of • .JLLf w i l l depend on the polarity of the incoming pulse, the voltage on the grid of the thyratron, and the deioniz-atlon time of the thyratron. However, unless the shunt fault i s less than several hundred feet from the sending end of the cable, the thyratron acts like a high impedance to incoming pulses. Distribution boxes w i l l appear as low impedances unless the total admittance of a l l the outgoing cables equals the admittance of the single incoming cable. As transformers are generally connected to the tap joint by a flve-^to ten-foot cable, the termination f i r s t appears to be a drop i n impedance. •The large inductive loop between the fuses and the transformer presents a high series impedance which w i l l give an echo even though the stray capacity of the transformer winding effectively shorts the far end of the inductive loop. Grounding reactors between insulated sections of the sheath may have enough stray capacity to appear to short the sheath discontinuity. However, this possiblity has not yet been experimentally checked. A distant good joint i s too small a discontinuity to be detected without highly sensitive apparatus. However, nearby Joints or joints having a series resistance of several ohms can be detected as rises i n impedance. Crushed faults should give a minute echo showing a slight decrease i n impedance. However, two discontinuities close together, e.g., the junctions to good cable at both, ends of a wiped Joint, produce echoes of opposite polarity which partly cancel one another to leave short pips which are very quickly attenuated. Multiple echoes from complicated distribution grids should be mapped while the cable i s unfaulted. If this pre-liminary work has not been done, the positions of the known echoes can be calculated and compared with the oscilloscope trace to determine any discrepancies which indicate faults. Lattice diagrams proposed by L.V. Bewley1 simplify the problem of Identifying multiple echoes on lines or cables whose velocity, impedance, and attenuation vary from section to section. 1 See reference 9B page ^"^5 J28 3. Effect of Change of Velocity As the length of the trace on the oscilloscope screen i s a function of time or ~ , care must be taken to interpret Intervals between echoes in terms of the velocity of propagation of the section of cable bounded,by the echoes. If a piece of flexible ^Q-oim polyethylene cable connects the apparatus to a 50-ohm rubber ot varnished-cambric power cable, the reflection from the junction would be almost undetectable; yet the different velocities of propagation wi l l compress the time-distance scale of the polyethylene cable to 75$ o f the scale of the power cable, A similar phenomenon occurs when the cable sheath i s 1 not connected to the grounded sheath terminal of the Echo Ranger. The sheath then acts only as a capacitive voltage-divider between the core and ground. Consequently,. most of the energy in the applied pulse travels at the velocity of light i n free space until the pulse reaches the f i r s t place where the sheath i s grounded. An echo from the junction w i l l be produced because the pulse w i l l be entering a cable from an aerial line. This change of parameters may also appear at the insulated sheath-joints sometimes used to control the induced currents which heat the sheath. Only the sending end of the sheath must be grounded, for the whole of the i n i t i a l pulse i s then impressed directly between the core and sheath without an intervening inductive loop between line and ground. 1 See reference page 2-3 When the 'sheath's; and cores are continuous, j u n c t i o n s between,cables having d i f f e r e n t c h a r a c t e r i s t i c impedances w i l l be d e t e c t a b l e from the p o l a r i t y o f the echo. A drop i n impedance at the echo w i l l be i n d i c a t e d when the pul s e i s f e d i n from one end of the ca b l e , whereas a r i s e i n impedance a t the same j u n c t i o n w i l l appear when the pul s e i s f e d i n from the 1 o t h e r end o f the ca b l e . 1 See r e f e r e n c e 1 page, ^ ko1 3o C, Apparatus, 1. Description Basically, the apparatus consists of a low-voltage impulse generator combined with a timing device for measuring the interval between the transmission of the i n i t i a l pulse and the reception of echoes from discontinuities in the cable. The input pulse approximates a 0 by 0.0} wave which reaches a peak of about $00 volts when feeding a 5<~,-°nm line. The impulse generator consists of a high-power hydrogen thyratron which discharges a condenser directly into the cable. A compact 2-kv power supply recharges the condenser while the echoes are being received. (See block diagram) The timer or marker pips and the echo pattern are compared on a Cossor double-beam oscilloscope adjusted to pro-duce a single sweep each time a pulse i s sent into the cable. As the marker pips, common sweep, and i n i t i a l pulse are synchronized within 0.002yu-sec. of one another, small parts of the trace can be expanded without affecting the relationship between the marker pips and the echoes. Hence the starting of the high-speed sweep can be delayed whenever echoes from distant parts of the cable are to be examined in detail. The apparatus, exclusive of the oscilloscope, i s 23-g- inches high, 10^ - Inches wide, and 7-| inches deep. The weight i s 36^ lb. The unit consumes 175 watts from a 115-volt 60-cycle line. As the input impedance of a cable may be as low as 50 ohms, the coupling device must provide a high burst of current 31 i n order to impress a pulse of a reasonably high voltage. Cathode followers are limited by low transconductance and low peak plate-current unless enormous tubes are used. Ordinary Marx-type impulse generators are too bulky for easy portability, although their hlgh-poiirer steep-wavefront pulses could probably be used to advantage on long cables. Mercury or argon thyratrons are too slow in ionizing to provide extremely steep-fronted pulses. Also, the deionization time i s so long that the thyratrons act like a short circuit to Incoming inverted echoes. The sluggishness i n ionizing and deionizing has been overcome in small hydrogen thyratrons, such as the Sylvanla 5C22, which can deliver into a 50-ohm load 2-microsecond pulses of 5-megawatt peak power $00 times a second.^ In this apparatus, the 5^ 22 i s used as a switch to discharge a small condenser into the cable. As the pulse can reach a maximum of only SO kw with 2 kv driving i t , the pulse repetition frequency was stepped up to about three kilocycles. However, as the pulses are very short - jx-aec. - the average current through the thyratron i s only 1.7 nia. Consequently only a small high-voltage "power-supply i s needed; 2. Accuracy On RG-SU Polyethylene Coaxial Cable 1023 feet long, a single 100-ohm shunt fault 71.6 feet from the sending end appeared to be 73*6 feet away - an error of 0.2$. On a 1044-foot piece of RG^ U coaxial cable with tx*o 100-ohm shunt faults 71-6 feet and 92.4 feet respectively from the sending end, the faults appeared to be distant feet 1. See reference /2- page 32-and 93 .5 feet respectively . - (Errors of 0 .63$ and 0,06# respectively). Note that the faults were only 20.8" feet apart. On a three-conductor o i l - f i l l e d lead-sheathed power cable 1389 feet long, a transformer tap K&K feet from the sending end was located within 25 feet before the non-linearity of the start of the trace was realized. In a later test with improved apparatus the error was reduced to l 6 feet (1 ,15$ error). A Joint, 320 feet from the sending end gave a small echo 11 feet short of i t s true position; When the oscilloscope was connected to the untapped core while the pulse generator fed the tapped core of the above power cable, the transformer appeared to be 21 feet beyond i t s true position. It has not yet been determined whether the frequency variation of the marker-pip oscillator or the distortion of the echo i s responsible for these errors. 3 . Rang;e Although the 5 0~* nicrosecond brightened trace limits the maximum range to two miles, the size of amplified multiple echoes indicates that faults three miles away on a five - mile cable should be readily visible. If both ends of the cable are available, the effective range could be doubled, depending on the position of the fault. Ultimately, the weak echoes from distant faults w i l l be masked by multiple echoes from inter-mediate discontinuities. • ' 33 Circuit Mechanism a. Multivibrator. The master timer i s a double-pentode multivibrator\rhich produces nearly square waves under light load without appreciably altering the period of the square waves. The screen grids of the 6AG7 rs (V^ and V 2 ) act as the anodes of a triode multivibrator which are shielded from the power-.. handling plates by the grounded suppressor grids. The plate current i s then almost independent of the plate voltage. The total width of the asymmetrical square waves of 52 and 22>5^i~sec. respectively corresponds to a repetition frequency of 297° P e r second. The amplitude (peak to peak) of the output i s 267 volts. b. Trigger Network. The pulse-shaping network for triggering the thyratron i s similar to the f i r s t shock-excited stage of the cascade trigger ci r c u i t used in the Marine 2 Type-26g Radar Set. A negative-going square wave from Vg i s fed onto the grid of which immediately cuts off the steady plate current through the inductance L^t The high Inductance causes the plate voltage of V, to rise to 3 try to keep the current constant through the c o i l . As the Inductance has appreciable stray capacity, i t begins to oscillate to produce a da'mped sine wave of See reference l 6 page 534 See reference 21 Diagrams 40, 45, 47. 34-about 3° kc» However, the 3° inductance in the grid circuit of V received a shock when the negative-going square wave was impressed on i t . (Note the resulting pip in the multivibrator output). As the natural frequency of i s about 5° kc, the voltage across i t and the grid of V w i l l complete the f i r s t 3 half cycle several microseconds before the voltage across L^, Consequently the grid of w i l l be going positive when the plate voltage nears a maximum. The sudden flow of pla,te current damps out the remaining oscillations of the plate inductance. Also, as the grid i s driven positive, i t draws enough grid current to damp out the succeeding oscillations of the 3° ^ inductance. As the thyratron (V^) requires at least^ 20-ma grid current to f i r e i t at the low plate voltage of 2 kv., a. cathode follower (V^) i s inserted between the plate of and the thyratron grid. As the cathode follower i s suddenly overloaded when the grid,space ionizes in the thyratron, the grid of draws current to flatten the peak of the damped sine voltage appearing on the plate of V . The negative charge, stored on the coupling capacitors while the grids were drawing current, w i l l hold and V". well below cutoff when the square wave from finally.goes positive, c. Thyratron. When the thyratron (V^) fires, a 0 . 0 0 C 4-uf con-denser i s suddenly discharged into the cable. As the 3& cable appears as a resistance of only 50 ohms, the 0.5-^ih discharge circuit w i l l be much under-damped i f a smaller condenser i s used. C » for c r i t i c a l damping. R* * Also, the thyratron i s slowly discharging the con-denser while the grid plasma i s ionizing. (The thyratron acts like a vacuum triode.) As a larger condenser i s more slowly charged, a smaller charging resistance i s needed. Hence, the small current drawn by the thyratron does not appreciably lower the voltage applied to the condenser immediately before f i r i n g . A s t i l l larger condenser takes too long to discharge and requires too large a power supply. The amplitude of the pulse fed to the cable can be decreased by the insertion of larger f i l t e r i n g : resistors i n the high-voltage power supply. Voltage Divider and F i l t e r . In order to provide a condenser-charging path Independent of the cable, a tapped 2000-ohm metallized resistor i s shunted across the cable terminals. This resistor also serves as a voltage divider to feed ^ >Q% of the i n i t i a l pulse and echoes from -the cable to the plates of the oscilloscope. As the Input capacitance of the oscilloscope Is across the lower half of the resistor, the ultrahigh-frequency components of the pulses are shunted to ground so that they cannot. Increase the distortion of the trace on the screen. 36 For weak echoes, the two-megacycle video amplifier in the Cossor oscilloscope can be used providing i t i s not overloaded, (e) Beam Intensifying. In order to black out the return trace, a 65-volt peak-to-peak (i.e. 5^ »5 volts above zero and 8.5 volts below) square wave from i s fed onto the grid (lower tag) of the Gossor cathode-ray tube through a 0.01-uf 1200-volt coupling capacitor. The grid resistor i s 0.1 megohm. As this square wave i s going positive when the thyratron i s fi r i n g , the trace showing the i n i t i a l pulse and subsequent echoes i s intensified. During the flyback, the square wave i s going negative so that the beam i s blacked out. As the wave i s not truly flat-topped, the focus and brilliance vary along the sweep, (f) Triggered Sweep. When the self-repeating trigger voltage Is short-circuited, the Puckle time base in the Cossor Model 339 oscilloscope can be started and stopped by an external trigger voltage applied to the "synch", terminal. If the "synch", terminal i s driven negative, the sweep condenser i s discharged and the beam f l i e s back to the l e f t side of the screen where i t remains u n t i l the "synch", terminal i s driven positive and the normal sweep begins. 1 In order to delay the start of the high-speed sweep until the echoes return from the distant parts of. the cable, the square-wave trigger voltage i s fed through a tapped 1 See reference 3 page /£? 3/ four-microsecond delay l i n e . Owing to the l o a d of the delay l i n e on the m u l t i v i b r a t o r , the d i s t o r t i o n along the delay l i n e , and the m u l t i p l e echoes i n the delay l i n e , the f r o n t of the t r i g g e r pulse i s toppled over so that the amplitude now v a r i e s d i r e c t l y w i t h time. As the v o l t a g e to cancel the negative charge on the g r i d of the b u f f e r tube i n the Puckle time base i s p i c k e d o f f the "synch", potentiometer, the "synch." i s used as a v e r n i e r to vary the time at which the g r i d of the b u f f e r tube i s d r i v e n p o s i t i v e . As a t o t a l delay of 10 microseconds i s obtained f o r the f a s t e s t sweep of 2.5 cm p e r y j - s e c , the e f f e c t i v e t r a c e l e n g t h i s expanded to 2>5 c m» a* 2.5. cm per yu-sec. The whole l i n e can be u n r o l l e d across the screen by r o t a t i n g t h e s y n c h . c o n t r o l so that measuring of time i n t e r v a l s by counting the marker p i p s i s f a c i l i t a t e d . E i t h e r another delay l i n e can be i n s t a l l e d or the sweep speed can be decreased to take i n f a u l t s . between \ mile and 2 1/& m i l e s d i s t a n t . Hence, the device can be used to l o c a t e f a u l t s which are between f i v e f e e t and two miles from the end of the cable, (g) Marker P i p s . Marker p i p s are used as a means of measuring equal time i n t e r v a l s on the h o r i z o n t a l t r a c e . With the common X sweep and double beam on the Cossor, the marker p i p s appear on the 1^ t r a c e simultaneously w i t h the echo p i p s on the tr a c e . As the marker p i p s are l o c k e d i n step w i t h the i n i t i a l p u lse f e d to the cable, any p o r t i o n of the sweep can be expanded and the p o s i t i o n of the 38' expanded echo can be more accurately interpolated between adjacent marker pips. For simplicity, only one frequency of marker pip i s used. The oscillator frequency of k,k5 M.C. provides pips 0.225 -u-sec. apart. Although these pips are dis-crete at the low speed at which the right end of the 50 - -u-sec. brightened trace just appears on the screen, they are only 19 mm apart on the fastest sweep where 1 mm equals 0.012 ji-aec. A modified Hartley oscillator overdamped during the flyback time and shock excited at the beginning of the brightened trace provides the basic marker frequency. •The oscillations are damped out by a shunting triode (V£ A) which conducts during flyback while forcing the oscillator triode (Vgg) down near cutoff by means of the common cathode resistor (R 2^). The small cathode bypass condenser (C^) increases the amplitude of the f i r s t few cycles by aiding the grid c o i l i n delaying conduction of(Vgg)during the f i r s t half cycle and by reducing degeneration during the succeeding cycles. The amplitude of the steady oscillations i s controlled by the de-coupling condenser and resistor in the common plate-supply lead. This isolating f i l t e r prevents plate-supply variations from disturbing the o u t p u t frequency. The sine wave i s then clipped by the asymmetrical cathode follox^er ( V y A ) . A second cathode-follower buffer triode (V ) drives a differentiating circuit 39 which produces sharp peaks at the beginning of each d i s c o n t i n u i t y i n the r e c t i f i e d sine wave f e d .from V . 7A This peaked wave impressed on a small inductance (L^) produces 25-volt rounded p i p s O.O3 jx-sec. wide and 0.225 j sec. apart. Each p e r i o d i s f u r t h e r subdivided by a p a r a s i t i c o s c i l l a t i o n of the X sweep which,produces b r i g h t spots 0.03 jx-sec. apart on the t r a c e s . By i n c r e a s i n g the r e s i s t a n c e of the condenser-charging c i r c u i t , the t h y r a t r o n can be made to f i r e at only each second t r i g g e r pulse. I n t h i s way, the degree of i n t e r - m o d u l a t i o n of the X.^  and Y g voltages by the echo pulses from the Y 1 p l a t e s can be determined from a comparison of the non-coinciding t r a c e s of a l t e r n a t e sweeps both of which remain on the screen. I t i s found t h a t , both before and a f t e r the cable pulse i s r e c e i v e d , the timing pulses are undisturbed. However, whil e the cable pulse i s going negative, the t i m i n g p i p s are d i s p l a c e d about O.O3 u-sec to the r i g h t . However, t h i s displacement i s recovered while the pulse i s dropping back to zero. As t h i s e f f e c t i s not n o t i c e a b l e u n t i l the cable pulse has reached an amplitude of 4 mm (12 v o l t s ) ^ t h e toe o f the pulse i s not d i s t u r b e d . Hence the f r a c t i o n of a t i m i n g i n t e r v a l immediately before the cable pulse should be compared w i t h the preceding and not the f o l l o w i n g i n t e r v a l . 4-0 (h) Power Supplies. As i t was originally intended to bias one of the tubes 100 volts negativetyyya 4 0 0 -volt condenser^ input power-supply was built. As this tube was later discarded, the extra voltage was dropped through a series resistor. This method'provided an additional f i l t e r section, the condenser of which was installed in the timer chassis close to the multivibrator. The plate supply delivers 90 ma at 300 volts to the timer. As the thyratron needed only two ma at two fcv, a 1500-volt RMS oscilloscope transformer was used. The filament winding for the r e c t i f i e r tube was connected Internally to the plate winding^because the positive side of the output In an oscilloscope i s grounded. However, i n order to ground the thyratron heater to produce a negative i n i t i a l pulse, the negative side of the power' supply must also be grounded. An Insulated 2 .5-volt filament supply was available in the spare centre-tapped 5-volt filament winding on the thyratron heater transformer. A red pilot lamp lights when the high-voltage transformer i s on, whereas a green pilot lamp lights when only the heaters and 300-volt supply are energized. Operation (a) I n i t i a l Adjustment. After connecting the oscilloscope and the pother input to the pulse generator, turn on the oscilloscope and only the top heater switch of the pulse generator. Only the green light should show. If the light i s red, turn off the 2000-volt plate supply (bottom switch), otherwise the thyratron cathode w i l l develop hot spots and soon f a i l . While waiting 300 seconds for the thyratron cathode to heat uniformly, adjust the oscilloscope and pulse-generator controls. 1. On the generator, turn the Sweep Delay to zero and the Pulse Amplitude to the centre. 2. On the oscilloscope^* turn the Condenser or coarse velocity to number 9. 3 . Turn the Velocity and Amplitude to the maximum clockwise position. Turn the Trigger to the maximum counterclockwise position. 5. The Sync, control should be turned almost f u l l y clockwise. 6. Adjust the Focus and Brilliancy controls so that the timing wave i s clearly seen on the screen. Focusing i s simplified i f the timing wave on Y^ i s above the zero line of Y,. See reference 9 page H.O • ^ 2 . 7» If only one trace i s visible, rotate either Y j - or Y^- shift until the other trace floats into view. S. Now adjust the Sync. and X-shift to get the i n i t i a l part of the trace expanded at the l e f t side of the screen. 9. I n i t i a l l y the pulses from the cable should be connected to Y^ directly and the selector switch should be set at Plates A.C. Later on, i f amplification i s desired, the pulses should be fed Into and the selector sx^itch should be turned to 2HFY1. 10. Connect the cable to the Sheath-Core terminals by short twisted or coaxial leads containing as small an inductive loop as possible. 11. When the 5-minu.te heating time has elapsed, turn on the 2000-volt supply to the thyratron. The pilot light should change from green to red. Do not touch the cable leads while the 2000-volt supply i s energized. A downward-going pulse should appear near the l e f t side of the screen. 12. Now turn the Velocity control counterclockwise to compress the sweep until the total number of timing pips indicates that the total cable length appears on the screen. (Allow approximately 6o circuit-feet of cable between pips). If no irregularities appear, Increase the pulse 4-3-amplitude until the open end of the cable (down-ward pip) i s visible. For cables longer than half a mile, use the amplifiers controlled by A^ and A r , G-ain. Do not overload the amplifiers, for the pulses will.be distorted and uninterpretable. •The maximum positive amplitude with the bu i l t - i n video amplifiers i s about half an inch. The Y^ marker pips also slightly modulate the Y^ cable trace to produce regularly spaced bumps. As the magnitude of these bumps i s independent of the.amplified signals, no ambiguity should arise. However, to identify very weak signals, the marker-pip lead should be temporarily disconnected from Y^. If the fault i s very close to the sending end of a long cable, the multiple echoes can be damped out by a 5°- t 0 100-ohm carbon resistor connected across the cable terminals. A 100-ohm carbon resistor in series with the core of the cable w i l l give longer pulses which are more easily interpreted when the fault i s less than 50 feet away. A small echo from a known discontinuity can be more positively identified i f the cable i s alternately shorted and opened at this discontinuity. Trace Expansion. After the fault has been approximately located on the compressed sweep, the trace can be expanded for a detailed study of the echoes. <44 After allowing 250 circuit-feet per microsecond, set the sweep-delay switch to the desired range. Set the condenser switch to position ten. Readjust the focus and brilliancy i f necessary. By turning the velocity and "sync." controls slowly clockwise, one can then expand the sweep to cover up to half a mile. For more distant faults, the X-shlft and velocity controls must be turned slowly counterclockwise. Measurements. The distances between echoes on the trace can be measured in terms of the 0.225-microsecond intervals between the timing pips appearing simultaneously on the screen. These larger intervals are further subdivided by a. succession of bright and dim spo'ts produced by a parasitic oscillation of the X sweep. A l l measurements should be made to the begin-ning of the pulse, not the peak. If two adjacent pulses partly cancel one another, the second pulse begins where the f i r s t pulse changes slope. As the high-amplitude negative-going pulses temporarily expand the sweep whereas positive-going pulses compress the sweep, an error of about ten feet w i l l creep in. Although the marker pips are also distorted, they maintain the same time interval between their peaks because the oscillator i s isolated. However, Just count f u l l intervals and then compare the additional fractional Interval :• to the immediately preceding interval. To increase the accuracy, i t i s advisable to work from both ends of the cahle and take the weighted mean of the two fault positions so found. 6. Sample Calculation Refer to the photographs of the oscilloscope traces showing the echoes from a power cable. By comparing the expanded traces with the compressed trace one can see that pips 2 and 3 are badly distorted. From figure 1, echoes return almost at pips 6, 9, 11, 17, and.23. From the remaining figures (which are much more distinct on the screen than in the photos) the echoes are more closely logged. Echo logged at Intervals from beginning Calculate distance (by i'ratlo) L Expected Echo known distance Error % Error 0.93 0 0 feet beginning of cable 0 f t . 5.91 K3& Joint 325 -11 ft, 0.75$ g.Ol 505 trans-former tap 16 1.15 n.05(?) 10.12 63s [joint (re-echo 6^ 7 653 -q -17 • 1.2 i6.sk- 16.01 1009 nultlple echo 97S" 32 2.2 23.05 22.12 139^ (assumed) Par end of cable 139*1-x: 1391*- as 22.12 Hence x » 31^ Propagation velocity - 2 (length) . (intervals)(time of interval) = 2 (1394 ( 2 2 . 1 2 ) 1 0 . 2 2 5 ) = 5 6 0 f t . per micro sec. Velocity relative to light - 5 6 0 = 5 6 . 9 $ +7 IV DISCUSSION A. Advantages Although the Echo Ranger has i t s limitations, i t possesses many desirable features: 1. Multiple steady, variable, or Intermittent dis-continuities can be accurately located, 2. The general nature and severity of the fault Is Indicated. 3 . Only one end of the cable Is required. 4. Only ratio and proportion are used in the simple calculations. 5. The whole line i s visi b l e for rough checking of arithmetic. 6. Known discontinuities can be used as reference points. 7. Few controls in addition to those on a moderately priced commercial oscilloscope simplify operation. 25. The measuring current w i l l not destroy the fault. 9. As the. device requires l i t t l e power from a light socket, i t can be run from batteries and an inverter. 1 0 . Stable operation f a c i l i t a t e s reading, measuring, and checking. 1 1 . The apparatus i s compact and easily ca.rried. 1 2 . Temperature variations should not significantly affect the accuracy. 1 3 . The propagation velocity of surges can be easily determined. B, Limitations 1, High-resistance shunt faults, i.e. greater than 2000 ohms, cannot be detected without additional amplifiers or flashover equipment. 4-e 2. A combined series and shunt fault equalling the surge impedance of the cable appears as an i n f i n i t e line and w i l l not produce echoes. 3. Multiple faults may be missed i f they are .just beyond a major fault. However, the accuracy of. location of the f i r s t fault i s not impaired. 4. As the lower limit for series faults i s about five ohms, a poor joint may not appear. 5 . A "roadmap" may be required to identify echoes from complicated networks such as distribution systems. 6 . The distance between any two discontinuities, such as the ends of the cable, must be accurately known. On overhead lines, the variation In velocity i s small because the dielectric i s air. C. Suggested Improvements. Several additional problems appeared while the Echo Ranger was being developed. Further refinements of the apparatus are also necessary. A method of separating arc surges from the echo-ranging pulses when self-healing faults are being flashed over could take advantage of the longer time constant of the fault-generated surges. Overloading of the amplifiers could be prevented by a variable-gain stage which i s allowed to operate at maximum sensitivity only during a controllable period of the trace. \The i n i t i a l pulse could be cancelled by a clipper tube or a bridge. The range and the sweep delay should be increased to handle cables up to 10 miles long and overhead lines up to ~$00 miles. The oscillator frequency should be stabilized by a crystal. Transient disturbances of the frequency at the beginning of the trace should be investigated. If every f i f t h or tenth ^9 marker pip were increased in amplitude, highly compressed sweeps would he more easily measured. A circuit similar to the present timing oscillator would provide a circular sweep locked to the i n i t i a l pulse. A delayed spiral could then be used to determine the number of complete revolutions between the arrivals of echoes. Remove the i n i t i a l triode current from the cable while the thyratron i s ionizing. Methods of reducing intermodulation should be Investigated. A cathode follower to isolate the sweep delay-line w i l l prevent defocusing of the trace. The hydrogen thyratron could be used instead of gaps in an accurately controlled high-voltage surge generator. D, Preliminary Note on Flashover. If the faults are self-healing or have a high resistance, high voltage can be applied to flash over the fault while the Echo Ranger i s monitoring the cable. When this high steady d-c voltage Is applied so that the echo-ranging pulse voltages add to i t , the arc starts more readily. The cable w i l l flash over at a variable time after the pulse arrives at the weak spot because of the time taken to completely ionize the flash-over path. Owing to the intermittent nature of the discharge, the ruptured o i l has time to escape and be pa.rtly replaced by fresh o i l . Bubble formation and release w i l l increase the agitation. When the water vapour bolls out of a wet fault, the breakdown volta.ge w i l l rise un t i l the arc i s extinguished. When the fault flashes over, the arc begins to discharge the cable on both aides of i t . Once the discharge wave reaches the sending end of the cable, the supply w i l l begin to feed current into the cable. With a poorly-regulated power supply, the 3 t o r e d energy i s dissipated faster than i t can be replenished and the arc goes out long before the next timed pulse arrives. If a large enough power supply i s not available, high-voltage echo-ranging pulses applied while the cable i s normally energized but isolated from the a-c supply by a current-limiting reactor would probably be enough to flash over a weak spot and hold the arc long enough for i t s position to be read off the oscilloscope screen. Complete burn-down i s unnecessary as the arc should have a low enough Impedance over part of each a-c cycle to give a. • recognizable echo. The amplitude, of the intermittent arc-discharge Wave i s so high that the relatively small echoes superimposed on i t produce vertical lines which are too faint to be interpreted. As the arc surges are not locked to the sweep, they d r i f t faintly across the screen leaving bright and steady the synchronized echo-ranging pulses which were sent and received while the arc was temporarily extinguished. When the arc path i s approaching complete breakdoxtrn, a bright pip begins to grow on the screen indicating where the fault l i e s . Multiple echoes from only the arc surges could be used to locate the faults if.the resulting pattern on the oscilloscope screen could be interpreted. Further research 'is needed on this aspect of fault location. £2-V. CONCLUSIONS A. From Theory. The beginning of the disturbance, the toe of the pulse, does not reach a. point a distance x awayuntil has elapsed. Current and voltage waves propagate at the same velocity. At the wave front, — — . 2± After the arrival of the wave front, the shape of the wave depends on the total distance the wave travelled, the type of discontinuity from which i t was reflected, and the severity of the discontinuities i t passed. Each echo travels as i f i t were alone on an Infinite line. The velocity of propagation varies inversely as the square root of the dielectric constant of the insulation and i s sensibly Independent of temperature. The polarity of the echo from a total impedance lower than the surge impedance of the cable w i l l be the reverse of the Incident pulse. A negligible echo w i l l be returned from a terminating impedance equalling the surge Impedance. An impedance higher than the cable surge impedance w i l l return an echo of the same polarity as the incident pulse. For maximum efficiency of reflection, the cable should be short- or open-circuited. Skin effect causes a distant echo to start apparently later than i t should. However, amplification w i l l reduce this error. 53 10. The attenuation i s much larger in cables than in open-wire lines because the inductance i s decreased without a corresponding decrease in resistance. 11. After the i n i t i a l rounding of the pulse, further distortion takes place less rapidly. B. From Experiment. 1. The surge impedance of cables i s of the order of f i f t y ohms. 2. The propagation velocity in o i l - f i l l e d cable i s approximately $60 feet per microsecond. •3. The accuracy of location of an internal discontinuity i s within 0.6% on RG8U polyethylene cable and i s within 1.2$ on o i l - f i l l e d power cable. k. Multiple shunt faults are discernable i f they are more than 30 feet apart, 5. Amplified echoes must be shown i n their entirety or compared with unamplified echoes, i f a particular wavefront i s to be positively identified. 6. The pulse from the Echo Ranger i s high enough to aid the applied steady voltage in i n i t i a t i n g a flash-over at an Incipient fault. 7. The conclusions from theory have been t r i f l e d . 5-4 VI. LITERATURE CITED 1. Abraham, L.G., Lebert, A.W.,. Maggio, J.B.,. Schofct, J.T., "Pulse-echo Measurements on Telephone and Television F a c i l i t i e s " , Technical Paper 47-26, Transactions of the American Institute of El e c t r i c a l Engineers, New York, 1947, vol. 66, pp. 54-1-6. 2. Beck, E., "Sheath Grounds Affect Travelling Waves In Cables", Elect r i c a l Engineering, New York, vol. 5 2 pp. 232-9, April, 1933. 3 . Benson, F. S., H i l l , (J.L.,. Machen, C.R., "Sonic Detector", Electrical World, Albany, N.Y.,. vol. 125, no. 25, pp. 22-5 , June 22, 1946. 4 . Bewley, L.V., "Travelling Waves on Transmission Systems", Transactions of the American Institute of Electrical Engineers, New York, 1931, vol. 50, part 2, pp. 532-50 5. Blankmeyer, W.H., "Power Line Fault Locator^ Electronics, New York, vol. 17, p. l 6 6 , January, 1944. 6. The Br i t i s h E l e c t r i c a l and All i e d Industries Research Association, "Surge Phenomena - Seven Years11 Research for the Central E l e c t r i c i t y Board", Reference S/T 35 , London, 1 9 4 l . 7. Brown, G.H., "Impedance Determinations of Eccentric Lines" in Markus and Zeluff, ed., Electronics for Engineers, f i r s t ed., McGraw-Hill Book Co., Inc., New York, 1945. 55 g . Bush, V., "Operational Circuit Analysis" New York, John Wiley and Sons, 1929. 9. Cossor, A.C., Ltd,, "Cossor Double Beam Oscillograph Model 339 Instruction Manual" CB.55E, sixth ed., A.C. Cossor, Ltd., London, February, 19^ -6. 10. Coulthard,. W.B. ,• "Transients in,Electric Circuits", London, Sir Isaac Pitman and Sons, Ltd., 19k6. 11. Danner, G.L.,. "R.F. Oscillator Aids Locating of Cable Faults", Ele c t r i c a l World, Albany, N.Y.,.. vol. 123, p. I32, January 20, 19^5. 12. Heins, H., "Hydrogen Thyratrons", Electronics, vol. 19, No. 7, p. 96, July, 1946. 13. Hpyle, W.G., "Transmission Line Fault Locator", National Research Council of Canada, Report PRA-I35, September, 19I+6. \\. Jahnke, E., and Emde, P., "Tables of Functions with Formulae and Curves," fourth ed., New York, Dover Publications, 19^5. 15. Karman, T.von.,. and Blot, M*A., "Mathematical Methods in Engineering", f i r s t ed., New York, McGraw-Hill Book Co., Inc., 19^0. 16. Kiebert, M.V.jr., and Inglis, A.F., "Multivibrator Circuits", Proceedings of the Institute of Radio Engineers, Mew York, vol.. 33, No. S, p..53ir, August, 19^5. 17. Leslie, J.R., and Kidd, K.H.,. "The Linascope", The Hydro-Electric Power Commission of Ontario, Research and Testing Department, December, 19^ 7 • (see also A.I.E.E. Technical Paper lJ-8-207) IS, Margenau, H., and Murphy, G.M.,. "The Mathematics o f Physics and Chemistry", New York, D. Van Nostrand Co.,. Inc., 1 9 4 7 . 1 9 . Margoulies, S., and Fourmarier, P., "The L o c a l i z a t i o n o f F a u l t s on Overhead L i n e s by Means of Impulse Waves", paper 3 ° 7 o f Conference'Internationale des Grands Re^seaux E l e c t r i q u e s k Haute Tension, P a r i s , France, G a u t h i e r - V i l l a r s , June 2 4 to J u l y 3 , 1942. 2 0 . Nalos, E.J., "High Frequency.Method of L o c a t i n g Power Cable F a u l t s " , Master's Thesis i n the Department of E l e c t r i c a l Engineering, U n i v e r s i t y of B r i t i s h Columbia, August, 1 9 4 7 . 2 1 . N a t i o n a l Research Council of Canada, E l e c t r i c a l Engineering and Radio Branch, "Technical Manual f o r Merchant Marine Type 2 6 2 Radar w i t h C i r c u i t Diagrams", (Draft Copy) 1 9 4 7 ( ? ) 2 2 . Noakes, F. " High Frequency Method f o r L o c a t i o n of F a u l t s I n Poxirer Cables. " E l e c t r i c a l News, v o l . 5 3 * n o»' ^3* PP» 4 6 - 7 J u l y 1 , 1944.. 2 3 . Puckle, O.S.,. "Time Bases", New York, John Wiley and Sons, Inc., 1 9 4 3 . 24. Race, H.H., " V a r i a t i o n s xirith Temperature and Frequency of D i e l e c t r i c Loss i n a Viscous M i n e r a l I n s u l a t i n g O i l " P h y s i c a l Review, Minneapolis, Minn., v o l . 3 7 , pp. 4 3 0 - 4 4 6 , February 1 5 , 1 9 3 1 . 2 5 . Ramo, S.,. and Whinnery, J.R., " F i e l d s and Waves i n Modern Radio? New York, John Wiley and Sons, Inc., 1 9 4 6 . 2 6 . Roberts, F.F., "New Methods f o r L o c a t i n g Cable F a u l t s Particularly on High Frequency Gables", Journal of the Institution of Electrical Engineers, London, vol. 93, no. 26, part 3, pp. 325-95, November, 1946. 27. Savage, J.H. "Localization of Faults in Low Voltage Cables, with Special References to Factory Techniques", Journal of the Institution of Elec t r i c a l Engineers, London, vol. 92, part 2, no. 3°, pp. 5^0-93, December, 1945. 23. Simmons, D.M.,. "Calculation of the Electrical Problems of Underground Cables", The Electric Journal, vol. 29, May to November, 1932. 29. Stevens, R.F., and Stringfield, T.W., "Transmission Line Fault Locator using Fault-Generated Surges," A.I.E.E. Technical Paper 42-202, presented at the A.I.E.E. Pacific General Meeting at Spokane, Washington on August 26, 1942. 30. Ware, L.A., and Reed, H.R., "Communication Circuits",. second ed.,. New York, John Wiley and Sons, Inc., 1944. V l l . ACKNOWLEDGMENTS Vancouver, B.C., September 10, 19^2. Dr. H.J. MacLeod, Head, Department of Mechanical and Electrical Engineering, University of British Columbia, Vancouver, British Columbia. Dear Sir: In presenting this Thesis, I wish especially to thank Dr. Frank Noakes for his encouragement and guidance throughout the investigation. The British Columbia Electric Railway Company Limited Research Scholarship which made the work possible was gratefully accepted. Respectfully yours, Thomas K. Naylor. V l l l DIAGRAMS A. Apparatus THE ECHO RANGER (one n i n t h f u l l s i z e ) Oscilloscope and Pulse Generator with coaxial cable attached. PULSE GENERATOR MARKER-PIP GENERATOR Trigger Network f U Mdstir Timer Multivibrator Shock Osc///af6r Hydrofen Thyrafron T CaJ>/e Vo/fctfe Divider de filler A. Sweep De/a-y Syrr-JL 2HFYI A* / Connection ST /Veo |— Vernier Siny/e Sneef> 3ri///'anee 4 ^ BRIGHTENING PULSE Vie CaSS or #339 Osc/V/osm/re Pulse - 'modu/dfeoi Osciflafhr —<MMH/lr Rectifier- 8* P ifferen-fitCfor s Marker Pea. £ my Co// THE ECHO-RANGER P L A T E 11 B L O C K D I A G R A M THE ECHO-RANGER PLATE 111 CIRCUIT DIAGRAM - U f S . C . "JN \ Parts List for Echo-Ranger 63-(See wiring diagram) RESISTORS CONDENSERS Resistor # Ohms Watts Type Condenser micro- Volt- Type # farads age 1 2 I 9 10 i i 12 11 15 lS H IB 19 20 21 22 II 11 27 2g 29 30 2500 7500 20 K 20 K 220 K ^70 K 10 K 1500 11 K 6g K 100 K 50 K 20 K lj-00 1000 1000 500 K 3 meg 2 meg 2 meg 0.1 meg 0.1 meg 15 K goo 0.1 meg 1000 39 K 1000 100 1350 2 1. I 10 i 10 2 £ 1 2 2 2 2 2 1 2 2 1 2 10 carbon 1 1 1 1 n WW C WW c c c c c c c c c c c c c c c c c c c c WW TRANSFORMERS T L Hammond II36X 6 . 3 V - 5 a ; 6 . 3 v - 5 1 2 I I 9 10 11-12 II II ig 19 20 21 v c 3 V,-5 - 3 CT 215 1 5 0 0 v Cathode Ray Trans. 275 4-00-0-li-OO; 5 V - 3 a -6 . 3 v - 5 a T^ #2g D.C.C. on 1 5/g" form, space-wound; overlap top two coils 0.05 0.05 0.001 0.0001 0.01 0.001 0.015 1.0 0.00025 0. OOOM-0.00k 0.5 0.5 0.00025 0.1 o.oooif-5 0.01 0.001 g g g 600 600 500 500 600 500 600 600 500 2500 1200 3000 3000 500 600 500 500 500 % 0 600 600 Paper 1 Mica 1 P M P Oil M M M Oil Oil M P M P M Elec-tro I— lytic 1 TUBES 6AG7 6AG7 6AG-7 6L6 Sylvania 5C22 or ij-035 Vg B 6SN7-GT V ? A 6SN7-GT Vg 2 X 2 V q 5Y3&T/G INDUCTANCES L^ 30 mh iron-core R.F.C. L„ Two g5 mh iron-core R.F.C. in series L^ Hammond 10H - 150 ma liu 31 turns #1^S.C.E. -pdiam. form.-coll 3^-" long. Delay* 1| u-sec. T-ll*4-Line ) J (L-ll+7-152 incl.) War Assets -(C-172-176 incl.) Dubin Elec-* tronics. £4. B. Echoes from a Power Cable Length Type Sheath Core Outside Diameter Voltage Bed Manholes Grounding Terminations 1 3 2 9 feet O i l - f i l l e d paper Lead three #3 copper I . 0 3 inches 23OO volts (At least) Conduit in concrete A l l dry No reactors Open-circuited pot-heads 1. Panoramic view (compressed sweep) 0.225-microsecond marker intervals. 2 . I n i t i a l Pulse 0.225-microsecond inte»als. Echo from a Joint 0.225-microsecond intenals. 4. Echo from the transformer tap. 0.225-microseoond i n t e r v a l s . 5. Jransformer-tap re-echo or an echo from a second, j o i n t . 0.223-microsecond i n t e r v a l s . Multiple echo reflected from sending end, 0.225-microsecond intesals. Echo from open-circuited far end of cable 0 . 225-micro second internals. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items