32 Calibration of Galvanometers by a Constant Current. 



(k being a constant), we have to plot 4 — ^r and tan (i. e. 



sin 6 sin 



■ 7i). Thus sin 6 is common to both variables ; and hence, 



cos 6 / ' 



if we plot sin D and cos 0, we should get a straight line if the 



galvanometer be really a tangent one. 



About the law of sine-galvanometers there can be no 

 question if the controlling field be uniform ; and herein lies 

 one great advantage of sine-galvanometers. 



In " proportional " galvanometers, where the deflection is 

 supposed to be proportional to the current, the curve obtained 



by plotting 6 and — — ^ should be a straight line ; or 



sin 6 

 -— — — = a constant, ' 

 6 sin I) 7 



if calculation be preferred. 



It may be pointed out that, by choosing our unit current 

 such that the maximum value of D is nearly 90°, the method 

 admits of great sensibility. 



The quantity of apparatus required to perform the calibra- 

 tion is reduced to a minimum, nothing but the galvanometer 

 and a suitable current-generator being needed. 



As described above, the angle D is measured by breaking 

 the circuit and allowing the needle to come to rest in the 

 magnetic meridian. This requires some time, if the needle is 

 not dead-beat, and the observations could be greatly facilitated 

 by the use of a fixed divided circle or pointer, by which D 

 could be read off simultaneously with 6 without breaking the 

 circuit. If so arranged, the whole operation could be com- 

 pleted in a few minutes ; but as the method is chiefly 

 characterized by the fewness and simplicity of the apparatus 

 required, it was thought undesirable to mar that simplicity by 

 the addition of something not essential. 



In conclusion, I desire to thank Professor Ayrton, F.K.S., 

 for his kindness in suggesting the simple way of describing 

 the method adopted in this note, and also for the statement of 

 a problem which led to the results herein stated. 



