Introduction
In the traditional triaxial systems, where saturated samples are tested, the volume change measurement is a simple monitoring of the water entering or leaving the sample by a volume change transducer.
On the contrary, in the unsaturated systems volume change measurements are complicated by the compressibility of air.
If an increase of confining pressure is applied to an unsaturated sample, a movement of water out of the sample will occur but at the same time the size will change due to the compression of the air in the voids.
A correct measurement requires the volume of water leaving the sample and the total volume change of the sample as shown in the sketch.
With these two measurements, by difference the volume change due to water being squeezed out of the sample and the volume change due to the compressibility of air can be determined.
A double walled triaxial cell can be the solution: the same pressure inside and outside of the inner cell wall will produce zero expansion of the inner cell and allow to measure the total volume change from a volume change transducer inserted in the cell pressure line.
The inner wall of the cell above is made from glass: this eliminates the problem of water absorption.
The cell pressure is applied equally to the inside and outside of the glass wall: this eliminates the problem of expansion.
The total volume change of the sample can then be measured using a standard volume change transducer.
Cell pressure line supplying pressure to the outer cell and the inner cell via the volume change transducer.

Axis translation method with High Air Entry Stone (HAES). Operating principle.
One of the problems when a sample with high suction is to be tested, is to prevent the sample from sucking the water from the porous stone on the base pedestal and cause cavitation in the triaxial cell pore water measuring system.
To prevent this happening the porous disc has been replaced with a high air entry stone, cemented into the base pedestal. The high air entry stone will allow water to pass through but not air, at various values. For example, a 5 bar stone will not allow air under 5 bar pressure to pass through the stone. The stone is cemented into the base pedestal to prevent water passing around the outside of the stone. The saturated stone will then allow the passage of water but not air. This will make it very difficult for the water to be sucked out of the stone; it will stop air entering the stone but it will not stop cavitation under the stone. This requires another modification to our triaxial system. To stop cavitation happening and to enable the suction to be measured, an air pressure is applied to the pore space in the sample. Suction is caused by the surface tension forces providing a difference in pressure between the air and the water pressure. If the air pressure is zero (atmospheric) then the water pressures will be negative. If we increase the air pressure in the pore space, the water pressure will also increase, keeping the difference between the air and water pressure the same. The air pressure is increased until the water pressure becomes positive. The suction is still maintained because the water pressure is still lower than the air pressure. The air pressure is applied via the top cap (in the same way as a water back pressure in a saturated test) at about 200 kPa below the air entry value of the porous stone. This will raise the pressure inside the sample to a positive value and in turn will apply a positive pressure to the porous stone and the pore water transducer.

Description
Double wall triaxial cell, inner wall made from glass, complete with acces ring for transducer cables. The cell has to be completed with the base pedestal with High Entry Stone.