Environmental Engineering Reference
In-Depth Information
11.3
SELECTION OF SHEAR STRENGTH FOR DESIGN
11.3.1
Drained, effective stress parameters
11.3.1.1
Peak, residual or fully softened strength in clay soils?
Whether peak, softened or residual strengths are used in the analysis of slope stability
depends on the presence or absence of existing slide “planes” (actually surfaces which
may not be truly planar) and fissuring. The following guidelines are given:
-
Where there is an existing slide plane, e.g. in the foundation of a dam or where a dam
has failed by sliding, the field residual strength should be used for the slide plane. This
applies regardless of how long it has been since sliding last occurred. The exception
could be where there is definite evidence of recementing, e.g. in some slickensided
joints in weathered rocks.
-
Bedding surface shears formed by folding of rock strata have strengths approaching
residual and, unless extensive tests on the surfaces show otherwise, residual strengths
should be used.
-
Compacted soils and soils which have no fissuring or cracks should be assigned peak
strength parameters c
determined from saturated remoulded samples com-
pacted to the same density ratio and moisture content the soil would be in the dam. The
majority of dam embankment design is therefore based on peak strength parameters.
- Cuts in soils are often designed using fully softened strength to allow for the softening
due to the unloading of the slope and some progressive failure mechanisms which may
be present.
- Fissured soils have a strength between peak and residual strength depending on the
nature of fissuring, orientation, continuity and spacing of the fissures. Such soils do
occasionally occur in dam foundations, e.g. Ross River Dam (McConnel, 1987) and
Prospect Dam (Landon-Jones, 1997). This is discussed in more detail in Section 6.1.8.
Soils with relict joints, such as extremely weathered basalt, behave in a similar manner.
-Triaxial and direct shear tests do not properly simulate the actual plane strain stress
conditions which exist in most slope stability problems. However, the uncertainty aris-
ing from this is not significant in practical slope stability problems and, provided the
tests are carried out and interpreted correctly, the results from both tests can be
adopted for design.
and
It should be noted that, as pointed out by Lade (1986) and Mitchell (1993), apart from
cemented soils the effective cohesion (c
adopted
is particularly critical in the analysis of the stability of smaller dams, the upper parts of larger
dams and in smaller landslides, because it has a major effect on calculated factors of safety
when failure surfaces are shallow. It is recommended that unless there is definite evidence
of higher values, the effective cohesion adopted for design should be between 0 and 10 kPa
for peak strength and should be zero or say 1 kPa for residual and softened strengths, at least
at low normal stresses. In some situations where the strength envelope is markedly curved a bi-
linear envelope such as that shown in Figure 6.43 may be used.
) should be zero or very small. The value of c
11.3.1.2
Selection of design parameters in clay soils
When several triaxial tests have been carried out on the one soil, it is recommended that
the design shear strength parameters are obtained from a p
-q plot of the test results,
rather than by, say, averaging the individual c
values from each test, or plotting all
Mohr's circles on one diagram. It is also important to use results from the effective stress
range applicable to the field problem. In many cases this is at low stress, e.g. less than
50 kPa to 100 kPa for slide surfaces at 5 m depth.
,
