BRIDGE METHODS FOR LOCATING RESISTANCE FAULTS 407 

 and Formula (12) can then be written: 



D = T 



Rqi + -Ro 



In cases where the fault resistance appears to be affected appreciably 

 by the testing current this scheme of measuring may be found desirable. 



It has been found that, when a battery and manually operated 

 battery reversing switch are used and the balance point of the bridge 

 is determined by observing the galvanometer kicks as the battery is 

 reversed, the action of the galvanometer is somewhat as follows: For 

 settings appreciably below the balance point the galvanometer kicks 

 are definitely in one direction while for settings which are too high the 

 kicks are definitely in the opposite direction (assuming, of course, that 

 the polarity of the battery is taken into account). When the rheostat 

 setting is very close to the point of balance but slightly too low, the 

 galvanometer gives a quick double kick, i.e., the needle moves away 

 from galvanometer zero, then returns toward zero a short distance 

 and again moves away from zero. When the rheostat setting is 

 slightly too high, the galvanometer gives a single kick and then coasts 

 toward the end of the scale. The balance point of the bridge is where 

 the transition from double to single kick occurs. 



When the value of Rq is low a rheostat variable in steps of 0.1 ohm 

 may be necessary if the transition point is to be accurately obtained. 



Seasoned judgment is an essential adjunct to a knowledge of theory 

 in the practical application of fault locating methods. This is espe- 

 cially true in the case of methods such as those discussed here, with 

 which accurate results cannot be secured unless the fault resistances 

 remain constant in value while a set of measurements to determine 

 location is being made. Experience has indicated that cable faults of 

 the types discussed are apt to be inconstant in resistance. Great 

 care must be exercised, therefore, in interpreting the results of meas- 

 urements. It is very important to make a sufficient number of separ- 

 ate sets of measurements to insure that consistent data are being 

 obtained. 



