DIELECTRIC REFERENCE MATERIALS FOR MATHEMATICALLY MODELING AND STANDARDIZING GRAIN MOISTURE METERS

Dielectric-type instruments are the most widely used technology for grain moisture measurement. Measuring fundamental dielectric parameters (dielectric constant and loss factor) rather than instrument-specific parameters is advantageous in terms of understanding the physical basis of the measurement and in transferring calibrations; however, few grain moisture meter models report such fundamental parameters. One reason for this is the difficulty in calibrating instruments to measure dielectric parameters. High molecular weight alcohols are commonly used as dielectric reference materials for calibrating research-type instruments. However, grain moisture meters are designed to handle grain rather than liquids, so performing meaningful tests on alcohols is very difficult. Suitable granular dielectric reference materials are needed. This article describes research assessing the suitability of three alcohols (1-pentanol, 1-hexanol, and 1-decanol) and low-loss dielectrics (glass beads, plastic beads, plastic pellets, and air) as dielectric reference materials for grain moisture meters. Complex reflection coefficient measurements were performed from 1 to 501 MHz with an HP-4291A material/impedance analyzer and a transmission-line dielectric test cell. Mathematical models based on ABCD matrix methods were created to predict complex permittivity from the measurements of complex reflection coefficient. The low-loss materials were used successfully for calibrating the system to measure permittivity in spite of the fact that their permittivity values were unknown. Two characteristics were assumed: constant dielectric constant over the frequency range, and constant (near zero) loss factor over the frequency range. These two constraints (along with electrical test cell parameters defined by finite element analysis and the test cell physical dimensions) permitted unambiguous determination of all parameters for the test cell mathematical model. The resulting test cell model accurately predicted dielectric characteristics of alcohols.