464 



ELECTRICAL EQUIPMENT 



The above discussion shows that a reactance somewhat above 

 that required for current limiting protection does not materially 

 affect the regulation or the power-factor, and in many cases it 

 may, therefore, be advantageous to use a somewhat higher reac- 

 tance than that which would be required, and thereby gain the 

 advantage of reduction in cost which can be obtained by using 

 standard ratings. 



Losses. The losses in reactors are not a serious matter but 

 should, of course, be taken into consideration in laying out the 

 system. They are due to the PR and eddy-current losses in the 

 conductors and possibly average 5 per cent of the rating of the 

 reactor. In some cases, however, the losses may be somewhat 

 higher and in others considerably less. 



Assume, for example, a 4 per cent feeder reactor on a 3000-Kv.A. 

 feeder, the three coils would have a combined capacity of 120 



Kv.A. or 40 Kv.A. per coil. The 

 losses at 5 per cent would equal 

 about 2000 watts per coil or 6 kilo- 

 watts on the 3000 Kv.A. feeder; 

 that is to say, one-fifth of one per 

 cent at the maximum load of the 

 feeder which may last only for a 

 comparatively short period during 

 the day. Since the losses are 

 nearly all copper losses which go 

 down as the square of the current, 

 at one-half load, the losses would 

 only be one-fourth of the above. 



Bus reactors, on the other 



hand, carry normally very little, if any, current and the losses 

 under normal operations are, therefore, negligible. 



Inductance. The inductance of current limiting reactors 

 may be calculated with sufficient accuracy by the following for- 

 mula by Prof. Morgan Brooks: 



h-H 



FIG. 288. Reactance Coil. 



in which (see Fig. 288), 



r = mean radius of coil in centimeters; 

 6 = axial length of coil in centimeters; 

 t = thickness of winding in centimeters. 



