CABLE TESTING 673 



to the impairment of the insulation between two wires so that 

 the current may pass between them. 



The location of grounds and crosses is effected by the same 

 methods. In a multiple-conductor cable, the first step is to pick 

 out the conductors which are faulty, for, in general, some con- 

 ductors remain in perfect condition. 



For this purpose, both ends of the cable are disconnected from 

 the service apparatus, and the insulation resistances between 

 the various conductors and ground and between the con- 

 ductors themselves are measured; the voltmeter method may be 

 used. If the faults are of sufficiently low resistance, bridge 

 measurements with reversed currents may be made. Continuity 

 tests should also be made to determine whether any of the con- 

 ductors have been burned off or broken. The above tests enable 

 one to decide on the subsequent procedure. 



If the resistance per unit length of the line is uniform, three 

 things must be known in order that a ground or a cross may be 

 located: 



1. The total length of the faulty line. 



2. The total conductor resistance of the faulty line at the time 

 of test. 



3. The resistance of the faulty line from the testing station to 

 the fault. 



The length of the line is given by the office records. When 

 there are only two wires connecting the stations at the ends of 

 the lino, it is not possibe to measure the line resistance after the 

 fault has occurred. The best approximation possible must 

 then be made by taking the stated resistance per unit length 

 and correcting it for temperature; in this correction there may 

 be considerable uncertainty, for the temperature coefficient of 

 the copper is large (0.4 per cent, per degree C.) and it is often 

 difficult to form a just estimate of the temperature of the con- 

 ductor, especially if it is in a duct near heavily loaded cables. 



Blavier Test. In case the faulty wire is the only one connect- 

 ing the stations, Blavier's method furnishes the only means of 

 ing the fault by measurements made from one end of the 

 line. In order that the test may be carried out, it is necessary 

 that the observer be able to send his instructions over the line 

 to the attendant at the other end. 



43 



