Determination of Energy dissipated in Condensers. .229 



J, the mechanical equivalent of heat, to be 41,972,000 ergs. 

 This is the value derived from Rowland's and Griffiths's work, 

 assuming the specific heat of water at an average temperature 

 of 20° C. to be unity. Column (j) gives the frequency. 

 Sometimes this was estimated from the average frequency of 

 the dynamo supplying the lines of the Middle town lighting- 

 circuits at the time ; and in other cases it was determined by 

 measuring the speed of a small synchronous motor. The 

 electromotive force (k) was measured with an electrometer, the 

 current (I) with a Siemens dynamometer. Column (??) gives 

 the values of cos cf> of the expression power = EI cos <£. Care 

 was taken in every instance to avoid the presence of upper 

 harmonics, in some cases using a resonance-coil to quench the 

 harmonics as well as increase the voltage on the condenser. 

 Column (o) gives the per cent, loss, 1007T cot </>, and column 

 (p) the net efficiency. This relative loss, tt cot <$>, has been 

 proved* to be the ratio of w to W, where w is the energy 

 dissipated per half-period, and W is the energy stored in the 

 condenser at each charge. 1— 7rcot<£, the net efficiency, is 

 therefore 



Energy stored — Energy lost 

 Energy stored 



5. Beeswax and Rosin Condensers. 



The first experiment, with condensers Nos. 3, 4, and 7, 

 showed a net efficiency of 93*39 per cent., or a loss of 6*61 

 per cent. The temperature of the condensers was not deter- 

 mined ; but from the fact that the dielectric was softened and 

 the quantity of heat generated was more than in any suc- 

 ceeding experiment, we feel sure that it was considerably 

 above 40° 0. Six condensers were then placed in the calori- 

 meter, joined in three pairs. Nos. 1 and 5 gave no sound 

 when joined to an alternating E.M.F. of 1000 or more volts, 

 and we called it the " best pair/' In Nos. 3 and 4 vibrations 

 were distinctly felt when the fingers were placed in contact 

 with them, while the condensers gave a clear musical note 

 and on the higher voltages a hissing sound ; this we called 

 the " poorest pair." Nos. 8 and 9 were intermediate. 



These six condensers were first of all joined in parallel and 

 connected to a low-frequency circuit of 1520 volts and 26 

 periods per second. Care was taken to exclude upper har- 

 monics. The experiment continued over three hours after 

 the temperature of calorimeter and condensers had become 



* See our paper, Phil. Mag. Jan. 1893. 

 Phil. Mag. S. 5. Vol. 47. No. 285. Feb. 1899. R 



