Abstract :
The paper deals with the problems which arose in the adaptation ofthe Schering bridge network for service at radio frequencies, anddescribes the finalform taken by the arrangement of components, screened high-frequencysource, and screened detector-amplifier, which has given a satisfactory performance at frequencies as high as 1 million cycles per sec. The steps which had to be taken to ensure a simultaneity of the main and the auxiliary bridge balances and to expedite the convergence of the succession of balances upon the final simultaneousbalance are explained. An account is given of the modifications which have been introduced into the well-known Schering bridge network, and of the provisions which had to be made in the disposition and the linking-up of the component parts to render the bridge capable of precision measurements. The arrangement described is essentially an equal-arm bridge composed of two capacitance arms, two resistance ratio-arms, and two auxiliary resistance ratio-arms, the latter forming part of a Wagner earthing system. The resistance arms are provided with condensers in parallel for phase-angle adjustments. Tappings are brought out in the main arms which permit of a choice of resistance for the whole arm and for the fraction across which the condensers are shunted. The artifice of connecting the power-factormeasuring condenser across only a fraction of the main resistance was adopted in order to supply a means of maintaining the changes of capacitance which are required when dealing with a large variety of power-factor differences over a large range of frequencies, always of the same convenient order of magnitude. The procedure followed in the measurements is strictly one of substitution of a variable air condenser for an actual, or an effective, capacitance of approximately the same value. Apparatus such as coils are tested in series or in parallel with such a value of the capacitance in the arm as brings the effective capacitance of the comp- osite arm to the value required for a bridge balance. Observations are taken of the change in reading of the standard air condenserand of the difference in setting of one of the power-factor-measuring condensers required by the inclusion and the exclusion of the impedance under test. From these two differences the two components of the impedance may be computed. In the case of condensers, the condenser is replaced by the variable air condenser. The capacitance is obtained directly from the settings of the latter. The power factor is derived from the formula ?? = 1/?2R?C?KT/KS where R is the resistance of the ratio-arm, and 1/? is the fraction across which the difference ?C is observed. KT represents the total capacitance forming the arm, and KS the substituted capacitance under test. The corrections to the simpler formul?, when the ratio-arms and the arm under measurement possess large phase angles, are discussed and evaluated. In order to reduce the phase angles of the resistance arms to a minimum, the latter have been provided with a shield, independently earthed, in preference to a separate screen connected to their common point. The successful performance of the arrangement is to be attributed in the first place to the attention which has been given to the design of the high-frequency screened source and of the detector-amplifier, and secondly to the arrangement of the component parts of the network and of the system of wiring which has been developed. .The adoption of toroidal forms for the tuning coils of the source and of the high-frequency stage of the amplifier, together with the screening arrangements employed, has practically eliminated all stray inductive and any capacitative coupling between the source and the bridge arms or the amplifier, respectively. Reversal of the leads from the source at 1 million cycles per sec. affects the power-factor balance by only 0.000002. In approaching the final balanced state of the bridge, the usual procedure is follo
