[ 19 ] 



II. A Resonance Method of Measuring En^gy dissipated 

 in Condensers. By Edwakd B. Rosa and Akthur W. 

 Smith *. 



THAT the dielectric of a condenser becomes warmed when 

 an alternating electromotive force is applied to the 

 terminals of the condenser has long been known, and the 

 study of this heating effect has been undertaken by numerous 

 observers. Kleiner f used a thermo-electric couple imbedded 

 in the dielectric to measure the rise of temperature, and noted 

 a considerable heating effect in ebonite, gutta percha, glass, 

 and mica, but none at all in paraffin and kolophonium. He 

 reports that in spite of all attempts by variations of the con- 

 ditions of the experiment, no heating could be detected 

 in these two last-named substances. On the other hand, 

 Boucherot % has made paraffin-paper condensers for use on the 

 3200-volt commercial circuits of Paris, some of which became 

 so hot in use that they were obliged to be cut out. Boucherot 

 says of the heating effect that if a condenser rises as much as 

 30° C, it should be rejected. That this is good advice is 

 evident from the fact that paraffin melts at 54° C. ; and hence 

 when 30° C. above the temperature of a summer's day (say, 

 25° C. or 77° F.), the paper would be floating in melted 

 paraffin. 



Bedell, Ballantyne, and Williamson § report experiments 

 upon a paraffin-paper condenser of 1*5 microfarad capacity, 

 the efficiency of which was found to be 95*6 °/ , or 4*4 °/ 

 lost in heat. The loss was determined by a three-voltmeter 

 method, similarly to measurements on a transformer. It was 

 put upon a 500-volt circuit at a frequency of 160, and the 

 current was therefore about 07 ampere, the apparent watts 

 about 350, and the heating effect 15*4 watts. The tempe- 

 rature rose one degree per hour. 



Threlfall |] reports a test on a paraffin-paper condenser of 

 his own construction which had a capacity of 0'123 micro- 

 farad, and on a circuit of 3000 volts and a frequency of 60 

 the rise of temperature was less than one-fifth of a degree per 

 hour. The apparent watts would be about the same as in the 

 experiment of Bedell, Ballantyne, and Williamson, although 

 the capacity was only one-twelfth as much. Threlfall concludes, 

 apparently, that since his condenser had only one-twelfth the 



* Communicated by the Authors. 



t Wied. Ann. vol. 1. p. 138. 



% LEclairage Electrique, Feb. 12, 1898. 



§ ' Physical Review,' October 1893, vol. i. p. 81 



|| ' Physical Review/ vol. iv. p. 458 (1897). 



C2 



