Geoscience Reference
In-Depth Information
uniform. The DSS test was developed to improve on the DS test, but also it suffers
from non-uniformity by the lack of shear stress on the vertical faces. Since
frictional soil can fail in two modes (double sliding), the failure state in a DSS test
is not necessarily unique. It depends on the actual stress state (horizontal
stress/vertical stress), see Chapter 9.
interlocking
friction
friction
c
c
cv
cohesion
(a) Terzaghi's concept (b) Taylor's concept
Figure 3.4 Mohr-Coulomb model
The Mohr-Coulomb model is widely in use in geotechnical engineering.
Terzaghi suggested for clay a concept based on cohesion, independent of normal
stress, and friction (Fig 3.4a). Taylor suggested, based on shear box tests on sand, a
different (peak strength) model. His concept does not include cohesion but
interlocking (related to dilation or asperity theory), which decreases with normal
stress (Fig 3.4b), and friction following
cv , the angle of repose of drained granular
soil. These two concepts led to a serious debate on true friction and true cohesion,
and whether or not cohesion exists, lasting to the present day. In stiff clay,
interlocking develops just as in dense sand, but the increased shear resistance is
mainly a result of the work required to suck in water. Localised weakened shear
surfaces develop and strength is subsequently reduced. According to Schofield,
Coulomb (1773) supported by Rankine (1857) stated that disturbed soils do not
possess cohesion. In design, it is safe not to count on cohesion. Normally
consolidated clay, the consolidation stress of which is above or equal to the yield
stress, is known not to possess cohesion, unless significant cementation of particles
and/or particle aggregates is involved.
Triaxial test
The triaxial test apparatus (see Fig 3.5) allows measuring axial and horizontal
deformations of a cylindrical soil sample wrapped in an impervious membrane
under varying vertical loading (called deviator stress, see later) and a supporting
ambient cell fluid pressure. Also the pore pressure in the sample (usually at the
bottom) can be measured. From triaxial test data various constitutive parameters
can be determined, such as elasticity E 50 , cohesion c , friction angle
, and dilation
angle
. The various stages in triaxial testing (impose cell pressure, impose a
vertical loading step or a certain deformation, etc.) and the choice to allow the
sample to drain (consolidate) or not during a stage provide for different
approaches. One distinguishes CU (consolidated under uniform cell pressure load
and undrained during a loading step), CD (consolidated-drained) and UU
 
Search WWH ::




Custom Search