Measurement and Estimation of State Variables - Unsaturated Soil Mechanics in Engineering Practice - page 130

electrical resistance sensor. The gypsum blocks are based on
the principle that changes in water content produce changes
in electrical resistance. Unfortunately, electrical resistance is
even more sensitive to the presence of salts. With time salts
can move into the gypsum and result in a substantial change
in the calibration of the sensor. Gypsum block sensors have
generally been found to be unsuitable for geotechnical engi-
neering applications.
Amore recent variation of electrical-resistance-type suction
sensor is the Watermark soil suction sensor. The Watermark
sensors consist of a granular matrix with embedded electrodes
for the measurement of electrical resistance. The granular
matrix is next to a gypsum wafer and forms a partial buffer
from salts in the soil. The entire sensor is placed within a stain-
less steel enclosure. Themeasurement range of theWatermark
sensor is 0-200 kPa. The Watermark sensor provides an indi-
cation of the wetness of the soil but does not provide a reliable
indication of the matric suction in the soil. The Watermark
sensor was patented in 1985 and is manufactured by Irrom-
eter Company. The Watermark suction sensors are relatively
inexpensive and can be connected to a data acquisition system
where they are used to trigger irrigation systems.

Phene et al. (1971) developed a thermal conductivity sen-
sor using a germanium p - n diode as a temperature sensor.
The sensor was wrapped with 40-gauge Teflon-coated cop-
per wire that served as the heating coil. The sensing unit
was embedded in a porous block. The optimum dimensions
of the porous block were calculated based on a theoreti-
cal analysis. The block must be large enough to contain
the heat pulse without interference from the surrounding
soil. The distribution of the pore sizes in the sensor was
important.
Gypsum, ceramics, and mixtures of ceramics and castone
were examined as potential porous block materials by Phene
et al. (1971). Ceramic blocks appeared to exhibit a linear
response and provided a stable solid matrix.
In the mid-1970s, Moisture Control System (MCS) of
Findlay, Ohio, manufactured the MCS 6000 thermal con-
ductivity sensor. The sensor was built using the design and
construction procedures used by Phene et al. (1971). The
manufactured sensors were subjected to a two-point calibra-
tion. The suggested calibration curves were assumed to be
linear from zero suction to a suction of 300 kPa. Above
300 kPa, the calibration curves were empirically extrapo-
lated. In the region above 300 kPa, the calibration curves
became highly nonlinear and less accurate.
The MCS 6000 sensors were used for making some matric
suction measurements in the laboratory and in the field
(Picornell et al., 1983; Lee and Fredlund, 1984). The sensors
appeared to be quite suitable for field usage, being insensitive
to temperature and salinity changes. Relatively accurate
measurements of matric suction were obtained in the range
from 0 to 300 kPa. Curtis and Johnston (1987) used the MCS
6000 sensors in a groundwater recharge study. The sensors
were found to be quite responsive and sensitive. However,
Moisture Control System discontinued production in early
1980, and the MCS 6000 sensor is no longer commercially
available.
In 1981, Agwatronics in Merced, California, commenced
production of the AGWA thermal conductivity sensors. The
design of the sensor was changed from previous designs but
was based on the research by Phene et al., (1971). There
were several difficulties associated with the AGWA sensor
that resulted in a new design called the AGWA-II sensor in
1984.
A detailed calibration study on the AGWA-II sensors was
undertaken at the University of Saskatchewan, Canada (Wong
et al., 1989; Fredlund and Wong, 1989). Several difficulties
were reported with the use of the AGWA-II sensors. These
include the deterioration of the electronics and the porous
block with time. The AGWA-II sensors have been used for
laboratory and field measurements of matric suctions on sev-
eral research studies (van der Raadt et al., 1987; Sattler and
Fredlund, 1989; Rahardjo et al., 1989).
In 2000, the GCTS company (Tempe, Arizona) under-
took the production and marketing of the heat dissipation
suction sensors that had been researched and developed at
the University of Saskatchewan, Canada. The production of

4.2.9 Thermal Conductivity Suction Sensors
The thermal properties of a soil have been found to be
indicative of the water content of a soil. Water is a bet-
ter thermal conductor than air. The thermal conductivity of
a soil increases with increasing water content. This is partic-
ularly true where the change in water content is associated
with a change in the degree of saturation of the soil.
Shaw and Baver (1939) developed a device consisting
of a temperature sensor and heater which could be installed
directly into the soil to measure thermal conductivity. It was
found that the presence of salts did not significantly affect
the thermal conductivity measurements. However, different
soils required different calibrations in order to relate ther-
mal conductivity measurements to the water contents of the
soil. Johnston (1942) suggested that the thermal conductiv-
ity sensor could be enclosed in a porous medium that would
have a standard calibration curve. The porous medium could
then be brought into equilibrium with the (negative) water
pressure in the soil. Johnston (1942) used plaster of Paris as
the porous medium to encase a heating element.
In 1955, Richards patented an electrothermal element for
measuring moisture in porous media. The element consisted
of a resistance thermometer which was wrapped with a small
heating coil. The electrothermal element was then mounted
in a porous cup and sealed with ceramic cement. Richards
proposed the use of a sandy silt material for the porous
block. It was suggested that the porous cup should have an
air-entry value less than 10 kPa.
Bloodworth and Page (1957) studied three materials for use
as a porous cup for the thermal conductivity sensors. Plaster of
Paris, fired clay, or ceramic and castone (i.e., a commercially
available dental stone powder) were used in the study. The
castone was found to be the best material for the porous cup.