5-B] TEST BY LOSSES. 163 



readings) at any convenient value of current is sufficient for all 

 results. Slight changes in wave form are immaterial and a 

 series resistance or any other means may be used for adjusting 

 the current (45). 



22. These readings vary with temperature, being dependent 

 upon resistance, and should be taken at some definite tempera- 

 ture or under some definite temperature conditions to be specified 

 in the report, as hot after a heat run of a certain duration, or 

 cold before the transformer is heated up. In this latter case 

 readings must be taken quickly to avoid rise in temperature due 

 to the testing current. Commercial tests should be under speci- 

 fied service conditions, commonly after a three hour heat run at 

 full load or the equivalent, the room temperature being 25 C. ; in 

 this the A. I. E. E. Standardization Rules should be consulted 

 and followed. 



23. At a second frequency repeat the readings. (In a com- 

 mercial test, readings would be taken at rated frequency only.) 

 It will be found that the copper loss and apparent resistance 

 vary but slightly with frequency and that the leakage reactance 

 is proportional to frequency. Known for one frequency, it can 

 be computed for any other. 



24. Results. The impedance voltage, EZ, and the copper 

 losses, We, can be plotted directly from the voltmeter and watt- 

 meter readings, with primary current as abscissae, as in Fig. 7 

 and Fig. 8. It is better, however, to proceed as follows: 



25. From the readings of the ammeter, voltmeter and watt- 

 meter in the short-circuit test, compute, for each observation, the 

 values of Z, R and X, as given below, and determine an average 

 value for all the observations. 



Impedance : Z = Ez-^-I 1 . 

 Resistance : R = W C ^- A 2 . 

 Reactance : X =Z 2 R 2 . 



