ALTERNATORS. 



249 



switch. In this case the current will vary, and should be 

 recorded. Judgment must be exercised in the experiment 

 as regards overloading the alternator ; machines vary 

 with regard to the magnitude of their current on short circuit. 

 It is usually three or four times the normal ' full-load current. 

 Probably it will be necessary to employ a lower value of 

 the excitation in this case in order to prevent the current 

 becoming excessive at full speed. 



The results should be entered as below. Curves for the 

 readings (1) and (2) should be plotted on the same sheet of 

 squared paper, speed horizontally and voltage vertically. 

 The curve for readings (3) should be plotted separately, 

 if the excitation employed varies from that adopted for the 

 first two sets of readings, and in this case the vertical scale 

 is made to read amperes. 



SPEED CURVES. 



Alternator No Type 



Normal output . . . .amps volts, at ... .revs, per rnin. 



Normal excitation frequency 



The curve (1) comparing speed and voltage at no load 

 should be a straight line, as already explained. 



If the loss of voltage in the armature were the same for a 

 given current independently of speed, the curve (2), in which 

 the armature carries a constant current, would be a straight 

 line making the same angle with the horizontal axis and lying 

 at a constant distance above curve (1). 



In the case of the curves given in Fig. 119, this is seen not 

 to be the case. The variations in the armature drop caused 

 by a change of speed are mainly due to two causes : (a) change 

 in armature reactance ( = 2 * n L) at different speeds ; (6) 

 eddy currents induced in the pole faces, and to a less extent 

 in the armature core. These eddy currents themselves set 

 up lines of force opposing the field producing them. 



