| Nine time-domain
reflectometry (TDR) probes, 2 to 10 cm long, were evaluated by comparing their
measurement accuracy of TDR-pulse travel time in a sand and sandy loam soil, and
electrical conductivity in NaCl solutions. TDR probes <2.5 cm in length
generated trough-haped TDR waveforms with rounded corners at the points of the
pulse reflection from the probe ends. The sharpness of the pulse reflection on
the waveforms increased with both the increasing probe length and soil-water
content. The transition time for the propagation of TDR pulse at the probe
entrance increased as the soil dried up. The increased transition time caused a
rightward movement of the first peak of the waveform at the probe entrance.
Because of such peak movement, TDR-support software algorithm determined travel
path of TDR pulse through the probe that was smaller than the actual travel
path. TDR-measured pulse travel time tTDR varied erratically
with the predicted pulse travel time tg (from volumetric
soil-water content) for the probes <2.5 cm in length. But, for all probes
³2.5 cm in length, tTDR varied linearly with tg
and followed the 1:1 line. TDR could not measure tTDR <300
ps accurately. A minimum probe length Lmin and the lowest
allowable soil-water content qmin that the probe can
accurately measure govern this lowest pulse travel time tmin.
The mean absolute deviation between tTDR and tg
was 77 ps for the 2.3 cm long probe and 1.39 ps for all probes ≥2.5 cm
in length. All probes ≥2.5 cm in length measured electrical
conductivity of salt solutions sTDR that compared well with the
electrical conductivity measured by a conductivity meter sm. The
length of the probes did not exert any noticeable influence on the accuracy of
electrical conductivity measurement.
Keywords: TDR probe, pulse travel time, dielectric
constant, electrical conductivity |