Correction of Geometric Influence in Permittivity Determination

Xiangdong Xu,Stanislaw Gubanski,Jörgen Blennow,Tord Bengtsson

doi:10.5324/nordis.v0i23.2459

Abstract

Determination of relative permittivity of dielectric materials is generally done with a rather low accuracy, in the order of several percent. This is in a sharp contrast to the accuracy of measurements of the dissipation factor, both usually being determined in the same measurement. A common understanding for the inferior accuracy in permittivity measurements is the effects of electrode edges. However, further studies indicate that geometric effects, arising from electrode shielding box, guard ring, electrode supporting materials, etc., also influence the accuracy significantly if the responding voltage present at the measuring electrode is non-negligible. With help of the Finite Element Method (FEM), geometric correction factors are estimated from an electrode model to increase the accuracy. This study is specially focused on the application of contact-free electrode arrangement using the air reference method. In this paper, a few examples of how geometric influences affect results are presented as well as a comparison of experimental results. From these insights, we discuss how to minimize and compensate the geometric effects.

Highlights

Determination of relative permittivity of dielectric materials is generally done with a rather low accuracy, in the order of several percent
It is often determined together with loss factor by applying dielectric response measurements in the frequency domain, known as frequency domain spectroscopy (FDS) [1] which is an important technique for both dielectric characterization and assessment of insulation system condition
The accuracy of permittivity determination in dielectric material specimens is generally low, in the order of several percent. This is a sharp contrast to the accuracy of the loss factor, both of which are often determined in the same measurement

Summary

Introduction

Determination of relative permittivity of dielectric materials is generally done with a rather low accuracy, in the order of several percent. With voltages present at the electrode, the measured capacitances will be larger than the capacitance between the two electrodes as the surrounding objects, such as shielding box, electrode supporting materials, connection cables etc., will have capacitive couplings with the measuring electrode and result in geometric influences. The finite element method (FEM) is employed to estimate a correction factor from an electrode setup model to improve the accuracy of permittivity determination.

Results

Conclusion