Search

Analysis is performed of the geometry of a Hertel exophthalmometer to determine the intrinsic error of reading that results from using the instrument in the manner that the manufacturer intended. Recommendations for improvement of the design and/or use of the instrument are made based on the findings.

A recently commercially available probe design employs remote diode shorting and differential waveform detection to create a segmented time domain reflectometry (TDR) probe using two parallel rods. The objective of this study was to investigate the accuracy of the multilevel probe in a laboratory soil column and in a gravely soil in the field. We outlined two procedures for calibrating the TDR probe. One technique uses physically based parameters, whereas the second type of calibration applies a curve-fitting process relating volumetric water content to measured travel times in the bulk soil. We show that instrument-related water content errors of the TDR are less than 0.010 cm3cm−3. If the physically based calibration approach is used, probe and segment-specific calibration yields an accuracy of about 0.015 cm3cm−3or less. However, we also show that excellent calibration results are obtained by simple calibration using segment-specific corrected bulk soil travel times, in which case a standard error of estimate of volumetric water content of about 0.015 cm3cm−3is obtained for both laboratory and field measurements. If TDR probes are combined and each depth interval is calibrated separately, as is usually done for neutron probe calibration, the standard error of estimate is 0.020 cm3cm−3or less, with R2values ranging from 0.63 to 0.92.