DHARM286 GEOTECHNICAL ENGINEERING
and roughness and particle size distribution. Its value increases with density index, with the
angularity and roughness of particles and also with better gradation. This is influenced to
some extent by the normal pressure on the plane of shear and also the rate of application of
shear.
The ‘angle of repose’ is the angle to the horizontal at which a heap of dry sand, poured
freely from a small height, will stand without support. It is approximately the same as the
angle of friction in the loose state.
Some clean sands exhibit slight cohesion under certain conditions of moisture content,
owing to capillary tension in the water contained in the voids. Since this is small and may
disappear with change in water content, it should not be relied upon for shear strength. On the
other hand, even small percentages of silt and clay in a sand give it cohesive properties which
may be sufficiently large so as to merit consideration.
Unless drainage is deliberately prevented, a shear test on a sand will be a drained one
as the high value of permeability makes consolidation and drainage virtually instantaneous. A
sand can be tested either in the dry or in the saturated condition. If it is dry, there will be no
pore water pressures and if it is saturated, the pore water pressure will be zero due to quick
drainage. In either case, the intergranular pressure will be equal to the applied stress. How-
ever, there may be certain situations in which significant pore pressures are developed, at
least temporarily, in sands. For example, during earth-quakes, heavy blasting and operation
of vibratory equipment instantaneous pore pressures are likely to develop due to large shocks
or dynamic loads. These may lead to the phenomenon of ‘liquefaction’ or sudden and total loss
of shearing strength, which is a grave situation of lack of stability.
Further discussion of shear characteristics of sands is presented in the following sub-
sections.
8.11.1 Stress-strain Behaviour of Sands
The stress-strain behaviour of sands is dependent to a large extent on the initial density of
packing, as characterised by the density index. This is represented in Fig. 8.28.
It can be observed from Fig. 8.28 (a), the shear stress (in the case of direct shear tests) or
deviator stress (in the case of triaxial compression tests) builds up gradually for an initially
loose sand, while for an initially dense sand, it reaches a peak value and decreases at greater
values of shear/axial strain to an ultimate value comparable to that for an initially loose speci-
men. The behaviour of a medium-dense sand is intermediate to that of a loose sand and a
dense sand. Intuitively, it should be expected that the denser a sand is, the stronger it is. The
hatched portion represents the additional strength due to the phenomenon of interlocking in
the case of dense sands.
The volume change characteristics of sands is another interesting feature, as depicted
in Fig. 8.28 (b). An initially dense specimen tends to increase in volume and become loose with
increasing values of strain, while an initially loose specimen tends to decrease in volume and
become dense. This is explained in terms of the rearrangement of particles during shear.
The changes in pore water pressure during undrained shear, which is rather not very
common owing to high permeability of sands, are depicted in Fig. 8.28 (c). Positive pore pres-
sures develop in the case of an initially loose specimen and negative pore pressures develop in
the case of an initially dense specimen.