510 AGNEW: TESTING INSTRUMENT TRANSFORMERS 



phases of a three-phase source. If the meters are working at the 

 power factor cos d, current lagging. 



tan a2 = tan ai + 



tan 6 



«i — 02 61 — 62 , R2 — Rl 



2ai 2bi Ri 



where the a's are the phase angles of the transformers, counted 

 as positive for the reversed secondary voltage leading the primary- 

 voltage. 



For current transformers the method of connecting the coils 

 is inverted compared with the case of the voltage transformer. 

 An auxiliary voltage is applied to the voltage coils of the meters 

 in parallel, the current coils being connected alternately to the 

 two transformers. Care should be taken not to open the second- 

 ary circuit of a current transformer while cm-rent is passing in 

 the primary. The formula for the difference in ratios is the same 

 as for the voltage transfonner. The formula for phase angle is 

 correct as given both for the voltage transformer with lagging 

 cmi'ent and for the current transformer with leading current. 

 If the conditions are vice versa the + sign before the bracketed 

 term should be changed to — . 



It is, however, not necessar}^ to depend upon this equation 

 for determining which transformer has the greater phase angle. 

 The following facts can be used as criteria for experimentally 

 determining the question. 



1. Adding a noninductive load to a voltage transformer al- 

 ways tends to lag the secondary voltage. 



2. Increasing the resistance in the secondary of a current 

 transformer tends to advance the phase of the secondary current. 



The experimental results show that by taking runs of approxi- 

 mately 100 turns each the method is capable of determining 

 ratio to 0.02 per cent or 0.03 per cent and phase angle to one, 

 or two minutes. In commercial work shorter runs would suffice. 

 Results as good as this were also obtained with the speed of the 

 meter doubled by shunting the magnets. Separate direct 

 experiments showed that the modern induction meter, even with 

 the drag magnets shunted so as to give double speed for normal 

 torque will repeat in consecutive runs under constant conditions 

 to a precision of about 0.01 per cent at full load. 



