A METHOD OF AND APPARATUS FOR DETERMINING THE DISTANCE FROM A FAULT IN A LINE

1275374 Detecting faults in lines VSESOJUZNY NAUCHNO-ISSLEDOVATELSKY INSTITUT ELEKTROENERGETIKI 30 Sept 1970 46413/70 Heading H4D In a method of detecting a fault in a line &c.- of the type in which a radio pulse is transmitted down the line and the fault causes a reflection of the pulse-the influence of noise in the line on the reflected signals is reduced by taking a sample of the reflected signals in a time position when no fault echo signal can be received (such that the sample comprises noise only), integrating the sample over a plurality of consecutive transmit/ receive cycles and subtracting the resulting integral from the integral of the sampled fault echo signal pulse noise. Thus the noise content cancels out, leaving the fault echo signal. A clock pulse generator 1, Fig. 1, produces clock pulses (a) Fig. 2. These are frequency halved at 3 to give the transmitter trigger pulses (b) and from transmitter 4, the transmitted pulse (d). The signals (f) received back from the line and including noise U 3 as well as a fault echo U 2 , are fed to a gated receiver 6. The clock pulses are also fed to a variable delay 2 where they are delayed by time ty to give gating trigger pulses (c). The delay is hand set and varied to sweep the gate through the received signal. Gating pulses (e) produced from pulses (c) gate the fault echo (plus noise) U 2 and also a sample of the noise. An amplifier 7 amplifies the gated signals with a rising swept gain to counteract the attenuation of the line. The fault (h) and noise (i) signals are diverted to respective integrators g 1 and g 2 by a switch 8 operated by the clock pulses. The integrator outputs (j) and (k) are subtracted in a pen type recorder 10 having its record sheet moved in synchronism with the variation of the delay Tg. A typical recording is shown at the bottom of Fig. 2.