Shear Strength of Unsaturated Soils - Unsaturated Soil Mechanics in Engineering Practice - page 559

angle φ can be drawn tangent to the Mohr circle at failure
(e.g., at stress point C ). The failure envelope intersects the
shear strength versus matric suction plane at a cohesion
intercept c . The cohesion intercepts obtained at various
matric suctions can be joined to give the φ b angle.

11.6.3 Consolidated Undrained Test with Pore
Pressure Measurements
CU tests are performed with the soil specimen first consol-
idated and then sheared under undrained conditions. The
pore-air and pore-water phases are undrained during the
shearing process, as shown in Fig. 11.51. The consolida-
tion process brings the soil specimen to the desired initial
stress state (i.e., σ 3 −

u w ). The axis translation
technique is used to establish matric suctions greater than
1 atm. Once equilibrium conditions are achieved, the soil
specimen is sheared by increasing the axial load σ 1 −

u a and u a −

(a)

σ 3

until failure is reached.
The drainage valves for the pore-air phase and the pore-
water phase are closed (i.e., undrained conditions) during
shear. Excess pore-air and pore-water pressures are devel-
oped during undrained loading. The excess pore pressures
are related to the deviator stress through use of the D pore
pressure parameter (see Chapter 15). The pore-air and pore-
water pressures should be measured during the shear pro-
cess. The net confining pressure σ 3 − u a and matric suction

(b)

u a − u w , are altered throughout the test due to changing

pore-air and pore-water pressures. The magnitudes of the
net major and minor principal stresses as well as the matric
suction are a function of the pore pressures at failure.
A typical stress path for a consolidated undrained test
is illustrated in Fig. 11.52. The stress state at the end of
consolidation is represented by stress point A where the
net confining pressure is σ 3 − u a and the matric suction is

u a − u w . Shear causes the st res s state to move from point A

to point B along stress path AB . The stress state at failure is
represented by stress point B , corresponding to a different
net confining pressure and matric suction from those
associated with stress point A . In the example shown, the
pore-air pressure is assumed to increase continuously during
shear. This causes the net confining pressure to decrease

(c)

Figure 11.49 Constant-water-content triaxial tests on Dhanauri
clay: (a) stress versus strain curve; (b) matric suction change versus
strain; (c) soil specimen volume change versus strain (after Satija,
1978).

[i.e., σ 3 − u a f <σ 3 −

u a ]. Matric suction is also assumed

to decrease continuously [i.e., u a − u w f <u a − u w ]. The

failure envelope is tangent to the Mohr circle at failure (e.g.,
at stress point C ) and inclined at an angle φ with respect
to the σ

stress point is as sum ed to move from point A to point B
along stress path AB as the soil is compressed during shear.
Stress point B represents the stress state at failure.
The net confining pre ssure remains constant at σ 3 −

u a axis. The failure envelope intersects the
shear strength versus u a −

−

u w plane at a cohesion intercept
c . The intersection line joining the cohesion intercepts
produced by tests at different matric suctions gives
the angle φ b .
It should be noted that it may be difficult to maintain a
fully undrained condition for the pore-air phase since air can
diffuse through the pore-water, the rubber membrane, and
other parts of the triaxial apparatus.

u a
along stress path AB since the pore-air pressure is
maintained at the pressure used during consolidation. The
pore-water pressure is assumed to increase continuously
during shear. The result is a reduction in matric suction

[i.e., u a − u w f <u a −

u w ]. The failure envelope with an