Civil Engineering Reference
In-Depth Information
φ
and
s
u
are the appropriate residual or critical state strengths. The values
chosen for the factors of safety depend on many things, including uncertainties in
determination of strength and the consequences of failure.
Notice that in Eqs. (18.3) and (18.4) the strength depends not only on the friction
angle but also on the total stress (which depends on the unit weight
where
and on the
external loads from foundations) and on the pore pressure. Some engineers apply
partial factors to each of these to reflect the different levels of uncertainty in their
determination.
γ
18.6 Serviceability limit state: peak strength with a
load factor
A further fundamental requirement of engineering design is to examine the service-
ability limit state and to demonstrate that the movements will not exceed some limit
determined by the design team. In geotechnical engineering there are two principal
methods for examining serviceability limit states. One is to apply a load factor to a
collapse analysis and, again, the question is whether the collapse analysis should be
done with the peak or with the critical state strength.
The critical state strength is clearly not appropriate for serviceability limit state
design because this would mean that you would design the same structure for dense
and loose sand or for normally consolidated and overconsolidated clay and this is
clearly illogical. A dense sand is stiffer than a loose sand and, for the same movement,
it can have a larger load and the same for overconsolidated and normally consolidated
clays.
Figure 18.5(a) shows the behaviour of two samples of the same soil in a triaxial
compression test: up to the peak the behaviour would be similar for drained and
undrained tests. (Note that in a triaxial test
q
σ
a
−
σ
r
) and the same symbol
q
is used
for loading.) The behaviour shown in Fig. 18.5(a) is similar to that shown in Fig. 10.15.
Sample 2 is further from the critical state line, it is more heavily overconsolidated and
its state parameter is larger than that of sample 1. Both samples reach a peak strength
=
(
Figure 18.5
Behaviour of soils in triaxial tests and behaviour of a structure.
