392 



BELL SYSTEM TECHNICAL JOURNAL 



(3) The resistance of the fault on one wire must be higher than the 



resistance of the fault on the other wire. 



(4) The conductor resistances of the faulty wires must be equal. 



In the practical application of the method, care must be exercised 

 in selecting the wires to be used for measurements. The resistance, 

 M, of the fault on the wire which is connected to the ratio arm of the 

 bridge when measuring V should be appreciably higher (at least 25 

 per cent higher) than the resistance, F, of the fault on the wire con- 

 nected to the rheostat arm of the bridge. This can be understood by 

 considering that as M and F approach each other in value the correc- 

 tion factor becomes larger and the Varley balancing resistance, V, 

 approaches zero, i.e., the apparent location of the trouble approaches 

 the distant end of the cable. Errors in measurement become in- 

 creasingly important as V and Fo become smaller. 



Accurate results will not be secured if the resistances of the faults 

 vary while a set of measurements to determine V and the correction 

 factor is being made. It is advisable, therefore, to make a number of 

 separate sets of measurements, and to base the location on those sets 

 which appear to be consistent. 



Straight Resistance Method 

 In many cases of complete cable failure the faults on all of the wires 

 are of practically equal resistance, and the Corrected Varley method 

 cannot be used successfully. The Straight Resistance method de- 

 scribed below has the advantage that the wires used for measurement 

 need not be unequal in fault resistance. 



Schematic circuit — straight resistance method. 



The Straight Resistance method is based on the assumptions that 

 the wires on which the tests are made are of equal conductor re- 



