AN ABSOLUTE DETERMINATION OF RESISTANCE

Abstract

An absolute determination of resistance has been completed in the laboratories of the National Research Council, using a bridge which was originally derived as a modification of Campbell's network. It can be shown that this network can also be treated as a special case of the bridge used by Hartshorn for the comparison of equal mutual inductors. This latter aspect is a very valuable feature, for it means that the same basic network can be used for nearly all the a-c. operations, i.e. calibration of inductors in terms of the computed standard, measurement of self-capacitances and intercapacitances, and the inductance bridge. The change from the comparison bridge to inductance bridge is accomplished simply by replacing one of the inductors by one having twice its value and by shunting its primary by an appropriate resistor. By doing so, the very low value of the "potentiometer" in the comparison bridge increases so as to be equal to R1 = 1 Ω, the latter becoming the resistor whose value is to be determined in terms of inductance (length) and frequency (time). The computed value of the standard inductor being equal to 10 mh. nominal, the value 20 mh. is obtained by the addition method on a specially designed inductor.A complete analysis of the circuit, including all small terms (self-capacitances, intercapacitances, residual inductances), is made for all operations: comparisons of inductors, addition of inductances, and finally the main bridge. The methods for the evaluation of these small terms are the most important and are therefore given with some detail. This includes also the evaluation of a ratio of resistances that figures in the equation for R1. This is made by means of a resistance comparator and its auxiliary equipment. A complete measurement is quoted from our laboratory book.A total of over 100 such measurements of R1 in absolute units were made. Their average shows that the unit presently used in the N.R.C. laboratory (deduced from calibrations made on N.R.C. standards by other national laboratories) is about 8 parts in a million smaller than the unit given by our method. This absolute unit agrees very closely with units resulting from similar work done in various other national laboratories.