206 



Prof. Ayrton and Dr. Sumpner on Alternate 



was small, that is when the current passing through ab 

 was small, that it would be most difficult, on account of lag, 

 to measure with ordinary methods the power given to ab. 

 But that is exactly the case when it is most easy to use our 

 one-voltmeter and two-ammeter method, since when the 

 dynamo has to send little current through ab there is little 

 objection to requiring it to send a current through cd in 

 parallel with ab. 



If the voltmeter (fig. 3) be a hot wire instrument, then, 

 since an appreciable current will pass through this voltmeter, 

 r must be taken as the parallel resistance of cd and of the 

 voltmeter. It is important to observe, however, that there 

 is no necessity to know either of these resistances separately, 

 since the value of r can be determined when the three instru- 

 ments A 1? A 3 , and V are relatively calibrated thus : — First, 

 break the circuits of cd and of the voltmeter, and compare 

 the deflexion of A 3 with A l ; this calibrates ammeter A 3 

 relatively to ammeter A 1? the calibration of which we will 

 assume to be correct ; second, close the circuits of cd and of 

 the voltmeter, but break the transformer circuit ab, A 3 is now 

 in series with the parallel circuit containing cd and the volt- 

 meter. The value of r is therefore at once known, since 

 r must equal the volts as read by V divided by the amperes 

 as read by A 3 . 



II. 



As an illustration of the general proposition to which we 

 have referred, it may interest the Members to see what are 

 the other analogies that we have traced out between alternate 

 P.Ds. in series and alternate currents in parallel in connexion 

 with the measurement of power. 



Mr. Blakesley's method, communicated to this Society in 

 February of this year *, is the current analogue of our electro- 

 meter method of 1881. For with the electrometer method 

 (fig. 4) we make two measurements, one giving us the mean 

 * Phil. Mag. April 1891, p. 346. 



