Measurement and Estimation of State Variables - Unsaturated Soil Mechanics in Engineering Practice - page 126

Phase III: After the suction response curve has come to
equilibrium (i.e., curve is essentially horizontal), there
is a slow increase in the observed suction measurement
for both saturated and unsaturated soils. The slow drift
in suction appears to be mainly attributable to mois-
ture loss due to evaporation from both the soil and the
high-suction tensiometer. Measurements showed that
the water content generally decreased by up to 0.1%
during the measurement of matric suction.
Test results to-date show that it is possible to sustain ten-
sile stresses in high-suction tensiometers for a considerable
time. It is important that the tension in the water phase not
exceed the air-entry value of the ceramic disk. The slow
drift in measured negative pressures should be further stud-
ied. The high-suction tensiometers are extremely sensitive
to minute losses of water during the measurement process.
The measurement of matric suction should be made in a
constant-temperature, high-humidity environment. The soil
specimens should also be at the same temperature as the
high-suction tensiometer at the start of the test (Fredlund
et al., 1997b).
High-suction tensiometers appear to have potential for rou-
tinely measuring soil suction in the laboratory. Direct mea-
surements can be made with suctions in excess of 1000 kPa.
It is not likely that the high-suction probes will find extensive
application for field usage in its present form because of the
need to occasionally recharge the high-suction tensiometer.

as the probe was inserted into the specimen. The tendency
of the water in the measuring system to become negative
was countered by increasing the air pressure in the chamber.
Eventually, an equilibrium condition was achieved with the
mercury plug (i.e., the null indicator) remaining stationary.
The difference between the air pressure in the chamber and
the measured water pressure at equilibrium was taken to be

the matric suction of the soil u a − u w .
The matric suction value was numerically equal to the
negative pore-water pressure when the air pressure was atmo-
spheric (i.e., u a = 0). The axis translation technique simply

translates the origin of reference for the pore-water pressure
from standard atmospheric condition to the final air pressure
in the chamber (i.e., axis translation). The water pressure in
the measuring system does not become highly negative and
as a result the problem of cavitation is prevented.
Hilf (1956) demonstrated that the measured pore-water
pressure increased by an amount equal to the increase in the
ambient chamber air pressure. In other words, the matric

suction in the soil u a − u w remained constant when mea-
sured at various ambient air pressures. The condition of no
water flow during the measurement of matric suction justi-
fies the use of the axis translation technique.
The axis translation technique has also been used by other
researchers (Olson and Langfelder, 1965; Fredlund, 1989c).
The procedure used is as follows. A soil specimen is placed
on top of a saturated high-air-entry disk inside an air pres-
sure chamber. The air-entry value of the disk must be higher
than the matric suction in the soil. A 1-kg mass can be placed
on top of the specimen to ensure good contact between the
soil specimen and the high-air-entry disk. The placement of
the specimen onto the ceramic disk and the assemblage of
the cell are performed as rapidly as possible (i.e., within
approximately 30 s).
The water pressure in the compartment below the high-air-
entry disk is maintained as close as possible to zero pressure
by increasing the air pressure in the chamber. The pressure
transducer connected to the water compartment is used as
a null indicator. It is important for the apparatus to have a
system to flush air bubbles from below the high-air-entry
disk in order to keep the compartment above the transducer
saturated with water.

4.2.7 Axis Translation Technique for the Laboratory
Laboratory measurements of negative pore-water pressure
can be made using the axis translation technique. The mea-
surement is performed on either undisturbed or compacted
specimens. This technique was originally proposed by
Hilf (1956) and is illustrated in Fig. 4.26. Hilf placed an
unsaturated soil specimen in a closed pressure chamber. A
pore-water pressure measuring probe consisting of a needle
with a saturated high-air-entry ceramic tip was inserted into
the soil. The probe was connected to a null-type pressure
measuring system through a tube filled with deaired water
and a mercury plug in the middle of the line.
The water in the tube tended to go into tension and the
Bourdon gauge began registering a negative pressure as soon

Figure 4.26 Null-type, axis translation device used to measure negative pore-water pressures