Shear Strength of Soils - Elements of Soil Mechanics - page 121

If the results of an undrained test are to be quantified in terms of effective stress, the nature of the test
must be considered. In the standard compression undrained triaxial test, the soil sample is placed in the
triaxial cell, the drainage connection is removed, the cell pressure is applied and the sample is immediately
sheared by increasing the axial stress. Any pore water pressures generated throughout the test are not
allowed to dissipate.
If, for a particular undrained shear test carried out at a cell pressure p c , the pore water pressure gener-
ated at failure is u, then the effective stresses at failure are:

′ = − u ; ′ = − = − u p u

σ σ σ σ 1 1 3 3 c

Remembering that, in a saturated soil, the pore pressure parameter B = 1.0, it is seen that if the test
is repeated using a cell pressure of p c + Δ p c, the value of the undrained strength of the soil will be exactly
as that obtained from the first test, because the increase in the cell pressure, Δ p c , will induce an increase
in pore water pressure, Δ u, of the same magnitude ( Δ u = Δ p c ). The effective stress circle at failure will
therefore be the same as for the first test (Fig. 4.30) , the soil acting as if it were purely cohesive. It is
therefore seen that there can only be one effective stress circle at failure, independent of the cell pressure
value, in an undrained shear test on a saturated soil.

4.12.2 Drained and consolidated undrained shear

The triaxial forms of these shear tests have already been described. It is generally accepted that, for all
practical purposes, the values obtained for the drained parameters, c ′ and φ ′ , from either test are virtually
the same.
The c ′ value for normally consolidated clays is negligible and can be taken as zero in virtually every situ-
ation. A normally consolidated clay therefore, has an effective stress strength envelope similar to that
shown in Fig. 4.31 and, under drained conditions, will behave as if it were a frictional material.
The effective stress envelope for an overconsolidated clay is shown in Fig. 4.32. Unless unusually high
cell pressures are used in the triaxial test, the soil will be sheared with a cell pressure less than its precon-
solidation pressure value. The resulting strength envelope is slightly curved with a cohesive intercept c ′ .
As the curvature is very slight it is approximated to a straight line inclined at φ ′ to the normal stress axis.
In Fig. 4.32, the point A represents the value of cell pressure that is equal to the preconsolidation pres-
sure. At cell pressures higher than this, the strength envelope is the same as for a normally consolidated
clay, the value of φ ′ being increased slightly. If this line is projected backwards it will pass through the
origin.
Owing to the removal of stresses during sampling, even normally consolidated clays will have a slight
degree of overconsolidation and may give a small c ′ value, usually so small that it is difficult to measure
and has little importance.
The shearing characteristics of silts are similar to those of normally consolidated clays.
The behaviour of saturated normally consolidated and overconsolidated clays in undrained shear is
illustrated in Fig. 4.33 which illustrates the variations of both deviator stress and pore water pressure
during shear.

Fig. 4.31 Strength envelope for a normally consolidated clay subjected to a drained shear test.