This type of yield gives conditions that approximate to the wedge theory, the centre of pressure moving

up to between 0.45 and 0.55 h above the wall base, and is referred to as the arching–active case.
The differences between the various pressure diagrams can be seen in Fig. 7.30d where the three diagrams
have been superimposed. It has been found that if the top of a wall moves 0.1% of its height, i.e.
a movement of 10 mm in a 10 m high wall, an arching–active case is attained. This applies whether the
wall rotates or slides. In order to achieve the totally active case the top of the wall must move about 0.5%,
or 50 mm in a 10 m wall.
It can therefore be seen that if a retaining wall with a cohesionless backfill is held so rigidly that little
yield is possible (e.g. if it is joined to an adjacent structure) it must be designed to withstand earth pressure
values much larger than active pressure values.
If such a wall is completely restrained it must be designed to take earth pressure at rest values, although
this condition does not often occur; if a wall is so restrained that only a small amount of yielding can take
place, arching–active conditions may be achieved, as in the strutting of trench timbers. In this case the
assumption of triangular pressure distribution is incorrect, the actual pressure distribution being indeterminate
but roughly parabolic.
If the wall yields 0.5 per cent of its height then the totally active case is attained and the assumption
of triangular pressure distribution is satisfactory. Almost all retaining walls, unless propped at the top, can
yield a considerable amount with no detrimental effects and attain this totally active state.
In the case of a wall with a cohesive backfill, the totally active case is reached as soon as the wall yields
but due to plastic flow within the clay there is a slow build-up of pressure on the back of the wall, which
will eventually yield again to re-acquire the totally active pressure conditions. This process is repetitive
and over a number of years the resulting movement of the wall may be large. For such soils, one can
either design for higher pressure or, if the wall is relatively unimportant, design for the totally active case
bearing in mind that the useful life of the wall may be short.
7.12 Design parameters for different soil types
7.12.1 Active earth conditions
Owing to various self-compensating factors, the operative values of the strength parameters that determine
the value of the active earth pressure are close to the peak values obtained from the triaxial test,
even although a retaining wall operates in a state of plane strain. As has been discussed in Chapter 4,
Fig. 7.30 Influence of wall yield on pressure distribution.
Lateral Earth Pressure 217
the values of these strength parameters vary with both the soil type and the drainage conditions. For
earth pressure calculations, attention should be paid to the following.
Sands and gravels
For all stages of construction and for the period after construction the appropriate strength parameter is
φ′. It is appropriate to take c′ as being equal to zero.
Clays
The manner in which a clay soil behaves during its transition from an undrained to a drained state depends
upon the previous stress history of the soil and has been described in Chapter 4.
Soft or normally consolidated clay
During and immediately after construction of a wall supporting this type of soil the vertical effective stress
is small, the strength of the soil is at a minimum and the value of the active earth pressure exerted on to
the back of the wall is at a maximum. After construction and after sufficient time has elapsed, the soil will
achieve a drained condition. The effective vertical stress will then be equal to the total vertical stress and
the soil will have achieved its greatest strength. At this stage therefore the back of the wall will be subjected
to the smallest possible values of active earth pressure (if other factors do not alter).
Obviously it is possible to use effective stress analyses to estimate the value of pressure on the back of
the wall for any stage of the wall’s life. A designer is interested mainly in the maximum pressure values,
which occur during and immediately after construction. As it is not easy to predict accurate values of pore
water pressures for this stage, an effective stress analysis can be difficult and it is simplest to use the
undrained strength parameters in any earth pressure calculations, i.e. assume that φ = 0° and that the
undrained strength of the clay is cu.
As mentioned in Chapter 4, the sensitivity of a normally consolidated clay can vary from 5 to 10. If it is
considered that the soil will be severely disturbed during construction then the c u value used in the design
calculations should be the undrained strength of the clay remoulded to the same density and at the same
water content as the in situ values.
If required, the final pressure values on the back of the wall, which apply when the clay is fully drained,
can be evaluated in terms of effective stresses using the effective stress parameters φ′ (c′ = 0 for a normally
consolidated clay). Soft clays usually have to be supported by an embedded wall (see Chapter 8)
and water pressures acting on the wall must be considered in the design.
Overconsolidated clay
In the undrained state negative pore water pressures are generated during shear. This simply means that
this type of clay is at its strongest and the pressure on the wall is at its minimum value during and immediately
after construction. The maximum value of active earth pressure will occur when the clay has reached
a fully drained condition and the retaining wall should be designed to withstand this value, obtained from
the effective stress parameters φ′ and c′.
With an overconsolidated clay, c′ has a finite value (Fig. 4.32) but, for retaining wall design, this value
cannot be regarded as dependable as it could well decrease. It is therefore safest to assume that c′ = 0
and to work with φ′ only in any earth pressure calculations involving overconsolidated clay. The assumption
also helps to allow for any possible increase in lateral pressure due to swelling in an expansive clay as its
pore water pressures change from negative (in the undrained state) to zero (when fully drained).
Silts
In many cases a silt can be assumed to be either purely granular, with the characteristics of a fine sand,
or purely cohesive, with the characteristics of a soft clay. When such a classification is not possible then
218 Smith’s Elements of Soil Mechanics
the silt must be regarded as a c–φ soil. The undrained strength parameter cu should be used for the evaluation
of active earth pressures which will be applicable to the period of during and immediately after
construction. The final active earth pressure to which the wall will be subjected can be determined from
an effective stress analysis using the parameters φ′ and c′.
Rain water in tension cracks
If tension cracks develop within a retained soil and if the surface of the soil is not rendered impervious,
then rain water can penetrate into them. If the cracks become full of water we can consider that we have
a hydrostatic, triangular distribution of water pressure acting on the back of the wall over the depth of
the cracks, z0. The value of this pressure will vary from zero at the top of the wall to 9.81 × z0 kPa at the
base of the cracks. This water pressure should be allowed for in design calculations, see Section 7.5.1 and
Example 7.4.
The ingress of water, if prolonged, can lead eventually to softening and swelling of the soil. Swelling
could partially close the cracks but would then cause swelling pressures that could act on the back of the
wall. The prediction of values of lateral pressure due to soil swelling is quite difficult.
Shrinkage cracks may also occur and, in the UK, can extend downwards to depths of about 1.5 m below
the surface of the soil. If water can penetrate these shrinkage cracks then the resulting water pressures
should be allowed for as for tension cracks.
7.12.2 Passive earth conditions
Granular soils
It is generally agreed that, for passive pressures in a granular soil, the operative value of φ is lower than
φt, the peak triaxial angle obtained from drained tests, particularly for high values of φt.
With a granular soil φt is most often estimated from the results of some in situ test such as the standard
penetration test (see Chapter 6). It is suggested therefore that values for φ, to be used in the determination
of passive pressure values, can be obtained from Fig. 7.31 (which is a modified form of Fig. 4.34).
The corrected N value can be used in place of the direct blow count N.
Normally consolidated clays
As with the active state, this type of clay is at its weakest when in its undrained state, i.e. during and
immediately after construction. For a normally consolidated clay the operative strength parameter is c u.
Fig.