Civil Engineering Reference
In-Depth Information
retained on each sieve recorded. The smallest practical sieve has an aperture size of
about 0.07 mm, corresponding roughly to the division between silt and sand. Silt-sized
particles can be separated by sedimentation making use of Stoke's law, which relates
the settling velocity of a sphere to its diameter.
A rapid estimate of grading can be made by sedimentation in a jam jar or milk bottle.
Take a sample about three quarters of the height of the container, fill the container
with water and shake it up. Quickly stand the jar or bottle upright and leave it for
several hours. You can see and estimate the grading of gravel, sand and silt; clay will
remain in suspension for a long time and any material floating on the surface is likely
to be organic (i.e. peat). This is a test frequently used by gardeners.
(b) Measurement of water content and unit weight
The water content of a soil is defined as
W
w
W
s
w
=
(7.1)
and the unit weight
γ
is defined as
W
V
γ
=
(7.2)
where
W
w
is the weight of water evaporated by heating soil to 105
◦
C until the weight
is constant,
W
s
is the weight of dry soil and
W
=
+
W
s
is the weight of a sample
with volume
V
. These weights can be measured by simple weighing and the volume of
a cylindrical or cubic sample determined by direct measurement.
W
w
(c) Measurement of Atterberg limits
For coarse-grained soils the engineering properties are governed largely by the grading
and, to a lesser extent, by the shape, texture and mineralogy of the grains, but the
properties of fine-grained clay soils depend largely on the type of clay. The basic
behaviour of clay soils can be assessed from the Atterberg limits (i.e. liquid limit,
plastic limit and plasticity index) described in Sec. 5.6. The liquid limit determines
the water content at which the soil has weakened so much that it starts to flow like a
liquid. The plastic limit determines the water content at which the soil has become so
brittle that it crumbles.
Liquid limit tests
The two alternative liquid limit tests are illustrated in Fig. 7.1. In the Casagrande test
in Fig. 7.1(a) a small slope is failed by bumping a dish on to a rubber block. In the
fall cone test a small cone-shaped foundation penetrates the soil. The precise details
of the geometries, weights and so on are arranged so that the soil has a strength of
approximately 1.7 kPa when it is at its liquid limit. In each case the sample has a
high water content and is soft enough to be moulded into the container using a knife
