524 



SCIENCE 



[N. S. Vol. XXV. No. 640 



The Compensated Two-circuit Electro- 

 dynamometer for Alternating Current 

 Measurements of Precision: Edward B. 

 Rosa, Bureau of Standards. 

 If an electric current pass in series 

 through a shunt S and the fixed coil F of 

 an electrodynamometer, and a small de- 

 rived current, taken from the potential 

 points ah of the shunt flow in series 

 through the resistance B and the moving 

 coil, a deflection will occur proportional to 

 the square of the current and inversely as 

 the resistance, R. The current may be 

 either direct or alternating. Hence if a 

 known direct current, measured by means 

 of a potenliometer, standard cell and 

 standard resistance, be used first to get the 

 constant of the dynamometer, the value of 

 an alternating current can then be ob- 

 tained. It is shown in the paper mathe- 

 matically that any self-inductance in the 

 shunt S or the fixed coil F does not pro- 

 duce an error in the alternating current 

 measurement, but that the self-inductance 

 of the moving coil and any Foucault cur- 

 rents in the fixed coil or in any neighbor- 

 ing conductor or metallic parts do cause 

 an error. It is shown how to detect such 

 an error and to compensate for it, so that 

 the deflected readings of the instrument 

 are the same as though these sources of 

 error were absent. The mathematical 

 theory is verified by experiment, and alter- 

 nating currents up to 500 amperes are thus 

 measured with precision. The compensa- 

 tion holds good at different frequencies, 

 and the error of a Kelvin balance, due to 

 frequency, is measured. 



The Power Factor and Temperature Coef- 

 ficient of Mica Condensers: E. B. Rosa 

 and F. W. Grovee, Bureau of Standards. 

 Absorption and leakage in a condenser 

 are a cause of expenditure of energy when 

 the condenser is placed on an alternating- 

 current circuit. This energy is, of course, 



smaller for good condensers than for poor 

 ones, and for very good condensers is ex- 

 tremely small, so small that the heat result- 

 ing is inappreciable unless the voltage is 

 higher than can safely be applied to the 

 condenser. In the comparison of con- 

 densers by means of an alternating cur- 

 rent bridge, it is shown in the paper how 

 the phase angle of the currents through the 

 two condensers can be compared, the differ- 

 ence indicating which is the better con- 

 denser. In an air condenser the current 

 is 90° in phase ahead of the electromotive 

 force acting on the condenser, and if a 

 mica condenser is compared with an air 

 condenser and shows a difference in phase 

 of 10', the angle for the mica condenser is 

 89° 50'. The power factor is then cos 

 89° 50' = .0029. If the power factors of 

 standard mica condensers are determined, 

 other condensers may be compared with 

 them and their differences (+ or — ) being 

 directly measured, giving the power factors 

 of the condensers under test. This is the 

 best single test of the quality of a con- 

 denser, and capacities measured in this 

 way are not subject to the error due to the 

 leakage to v/hich direct-current measure- 

 ments are subject. 



The temperature coefficient of a mica 

 condenser will be affected by the paraffine 

 in which it is embedded. Some grades of 

 paraffine melt at a relatively low tempera- 

 ture and soften and expand rapidly at still 

 lower temperatures. The temperature co- 

 efficient of the condenser will be small and 

 nearly constant up to a certain temper- 

 ature and then rather suddenly increase 

 greatly. The temperature at which this 

 sudden increase occurs is sometimes low 

 enough to occur in summer weather without 

 any additional heating. A good mica con- 

 denser should be embedded in high-grade 

 paraffine, so that its temperature coef- 

 ficient may remain small through all ordi- 

 nary ranges of laboratory temperatures. 



