TRANSFORMERS 383 



Rating. A transformer should be rated by its kilovolt-ampere 

 (Kv.A.) output. It is simply equal to the product of the voltage 

 and current, and is, therefore, the same whether the different coils 

 are connected in series or parallel. If the load is of unity power- 

 factor, the kilowatt output is the same as the kilovolt-ampere 

 output, but if the power-factor is less, the kilowatt output will be 

 correspondingly less. For example, a 100 Kv.A. transformer will 

 have a full-load rating of 100 Kw. at 100 per cent power-factor, 

 90 Kw. at 90 per cent power-factor, etc. 



The A. I. E. E. Standardization Rules identify self- and water- 

 cooled oil immersed transformers as to Kv.A. rating by their 

 maximum continuous capacity at 55 rise. With an ambient 

 room temperature of 40 C. air for the former and 25 C. incoming 

 water for the latter, the observable temperatures would be 95 

 and 80 C. respectively. The rules further specify that the tem- 

 perature of the windings of transformers is always to be ascer- 

 tained by Method II, i.e., the resistance method. (See page 310, 

 " Rating of Generators ") This method allows for -a correction 

 factor of 10 C., so that for self-cooled transformers the hottest- 

 spot temperature is limited to 105 C. and for water-cooled to 

 90 C. The oil shall in no case have a temperature, observable 

 by thermometer, in excess of 90 C. 



For air-blast transformers the rules specify that a correction 

 sliall be applied to the observed temperature rise of the windings, 

 and it is to be noted that air-blast transformers constitute the 

 only instance wherein it is required that a correction shall be ap- 

 plied to take into account the precise ambient temperature at 

 time of the test. This is due to the difference in resistance, when 

 the temperature of the ingoing cooling air differs from that of the 

 standard reference. This correction shall be the ratio of the 

 inferred absolute ambient temperature of reference to the inferred 

 absolute temperature of the ingoing cooling air, i.e., the ratio 



274.5 , . ., . . 



(234 54-fl ' wn ere t is the ingoing cooling-air temperature. 



Thus, a cooling-air temperature of 30 C. would correspond 

 to an inferred absolute temperature of 264.5 on the scale of copper 

 resistivity, and the correction to 40 C. (274.5 inferred absolute 



274 5 

 temperature) would be ' =1.04, making the correction factor 



1.04; so that an observed temperature rise of say 50 C. at the 



