ALTEBNATORS. 203 



The lower curve shown in Fig. 94 is a load magnetisation 

 curve taken on the same alternator as the no-load curve 

 shown in Fig. 92. In order to enable an easy comparison 

 to be made between the two curves, the curve of Fig. 92 

 has been re-drawn in Fig. 94. The vertical distance 

 between the two curves shows the amount of " armature 

 drop " produced by the load, when measured on the vertical 

 scale of volts. Thus with an excitation of 1-8 amps., the 

 drop of volts is seen to be 6. Similarly, the horizontal 

 distance between the curves, when measured on the horizontal 

 scale of amperes, shows the amount by which the excitation 

 would have to be increased in order to bring the voltage at 

 full -load to its value at no-load. The direct experimental 

 determination of this quantity is the subject of a later 

 experiment. 



The causes producing the armature voltage drop have been 

 already enumerated. The most important of these are the 

 armature resistance and the armature self-induction. The 

 armature resistance forms the chief cause of loss of 

 efficiency in the armature. It is readily obtained by 

 sending a measured direct current through the armature 

 and noting the voltage drop on a milli-voltmeter, or by 

 one of the more sensitive galvanometer methods employed 

 for measuring the resistance of direct-current armatures. 

 The power lost in the armature at any current output is 



W = C 2 R 

 where R is the armature resistance. 



The self-induction does not produce an actual loss of power, 

 although it does produce a larger drop of voltage, and con- 

 sequently greater diminution in the output of the machine 

 than the armature resistance. 



The impedance of the armature is made up of the resistance 

 and self-induction, and is given by the usual formula 



Impedance = ^/R- + (2 T n L)~ 

 where R = armature resistance. 



L = armature self-induction. 

 n = periodicity. 



The method of determining the armature impedance has 

 already been given in Experiment VII., page 52. Another 

 method, which is of considerable practical utility, is' the 

 following : 



