5-B] TEST BY LOSSES. 167 



i 



32. Maximum efficiency* occurs at such a load that the cop- 

 per loss is equal to the core loss ; in Fig. 8, this is at 0.9 full load. 



Note the similarity between the curves for a transformer, shown 

 in Fig. 8, and the corresponding curves for a shunt motor, Fig. 3, 

 Exp. 2-B. 



33. All-day efficiency is computed on some assumption, as 

 5 hrs. full load and 19 hrs. no load. Other assumptions can be 

 made to suit specific service conditions. Except under special con- 

 ditions, the term " all-day efficiency " has no useful significance. 



34. Regulation. The regulation of a constant potential 

 transformer is the per cent, increase in secondary voltage in 

 going from full load to no load. See Appendix L, Exp. 5~C. 

 There are various graphical methods for determining regulation, 

 which are necessarily unsatisfactory on account of the small 

 values of some of the quantities and the consequent difficulty 

 in making an accurate drawing to scale. There are also various 

 analytical methods, many of which are equally unsatisfactory 

 on account of their involved character and the unnecessary labor 

 required in using them. 



The regulation of a transformer can be determined for all 

 power factors current lagging or leading by the same method 

 as is used in determining the regulation of an alternator by the 

 electromotive force method ( 16-22, Exp. 3~B), either graph- 

 ically or analytically. 



A modified method, however, is easier to apply to a trans- 

 former on account of the fact that the resistance and reactance 

 drops in a transformer are comparatively small. 



35. What the writer believes to be the simplest and most 



practicable methodf for determining the regulation of a trans- 



r- 



* See 28, Exp. 2-B. 



t (35a). From a paper "Transformer Regulation," by F. Bedell, Elec. 

 World, Oct. 8, 1898; the term with I'M is now dropped on account of 

 difference of definition (see Appendix L, Exp. 5-C). 



