Determination of Energy dissipated in Condensers. 233 



6. Paraffined- Paper Condensers. 



The second lot of condensers used were commercial paraffined- 

 paper condensers made by the Stanley Electric Co. A finished 

 condenser is a solid, slab about 25 x 30 cm. and 2 cm. thick, 

 thus having a volume of 1500 c. c, and is enclosed in a tight 

 tin case, the lead-wires coming out through ebonite bushings. 

 Nos. 1 to 4 of our condensers have a capacity of about 

 1*7 microfarad each ; Nos. 5 to 10, which were purchased 

 about a year later, have a capacity of about 3*2 microfarads 

 each. The condensers of the second lot are made of paper 

 about -0038 cm. thick, two sheets being placed together in 

 each stratum. This we learned by dissecting some which we 

 had broken down. It ought to be stated, however, that 

 while the condensers are guaranteed by the makers to stand 

 500 volts alternating electromotive force, we have repeatedly 

 subjected them to 1000 to 2000 volts, and in some cases for 

 several hours at a time. Nos. 9 and 10 were upon one 

 occasion maintained at 2250 (effective) volts, at a frequency 

 of 130, for over an hour, and showed no signs of being over- 

 taxed. We have, however, broken several at voltages between 

 1000 and 2000. The paper of the first lot of condensers is 

 thicker, but as we have never broken one of this lot we 

 cannot state its thickness. From the fact that the capacity 

 of each of these is about 60 per cent, as great as that of the 

 others, while their volumes are substantially the same, we 

 conclude that the thickness of the paper is about "0048 cm., 

 supposing there are, as in the other, two sheets together in 

 every stratum of the condensers. 



In Table III. are given the results of seven separate expe- 

 riments with Stanley condensers, which were made at intervals 

 during the past three months. The frequency in every case 

 except experiment 4 was estimated from the average frequency 

 of the two dynamos of the Middletown lighting circuits. 

 Experiments 1 and 3 were made when the faster dynamo was 

 supplying the lines, the others were with the slower dynamo. 

 All but No. 4, however, were with a relatively high frequency. 

 No. 4 was made using a two-pole rotary transformer, sup- 

 plying it with direct current, and running it at a speed of 

 1600 per minute. The percentage losses (o) vary more among 

 the different condensers at the same frequency than one would 

 expect. 



The percentage loss at the frequency 28 (Experiment 4) is 

 '78 per cent., whereas at a frequency five times as great it is 

 (Experiment 3) 1*00 per cent. At 120 it is, as would be ex- 

 pected, nearly as great as at 140; that is, it is *96 per cent. 



