Dielectric measurements at centimetre wavelengths

Abstract

Some u.h.f. measurement techniques based on the use of coaxial line resonators are described.The methods cover the measurement of permittivity and power factor of disc and tubular samples, and in the latter case it is a simple matter to measure the effect of temperature.In both cases a fixed oscillator is employed, resonance being obtained by movement of the inner conductor of the line with a micrometer drive.For disc samples, machining or grinding is necessary in order that they may be a close but sliding fit in the line and, though not essential, time in computation is reduced if a standard thickness is employed, The experimental procedure involves so placing the disc in the line that the line length of resonance is a minimum and then measuring the resonant length and current. These quantities together with the disc dimensions and frequency are sufficient to determine the permitivity and power factor.In the case of thin discs the formulae used are very simple, but with thick discs, which are often necessary to obtain reasonable accuracy in power-factor measurements, more complicated formulae are involved.The accuracy obtainable is about 1% for permittivity and 5% for power factor.Temperature dependence of the dielectric constants cannot be measured with this apparatus as differential thermal expansion between he metal and the dielectric disc prevents free movement of the disc.In the case of tubular samples the only machining necessary is j utting the material to a given length. This is a great advantage when ceramic materials or glass are involved. To carry out the measurement the tube is cut to the required length and placed symmetrically in the measuring line over the inner conductor, the inner and outer liametcrs of the tube being immaterial so long as this can be done. The change of length of the inner conductor to re-establish resonance hen allows the permittivity to be calculated, and the reduction in esonant current enables the power factor to be found. As no close ncchanical clearances are involved this method enables measurements o be carried out at other than room temperature without difficulty. fhe accuracy obtained depends upon the uniformity of the dielectric tube: with a reasonable uniformity the error is in the range ±10%.