Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

The Use of Pore-Pressure Coefficients in Practice

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5
At a glance
Powered by AI
The article discusses the use of pore-pressure coefficients to estimate pore pressure changes in soils due to changes in stress, and gives examples of their use in earth dams during construction and rapid draw-down.

Pore-pressure coefficients relate changes in pore water pressure to changes in stress. They help estimate pore pressures and stability in problems involving soil masses under changing stresses, such as during earth dam construction and slope failures.

Pore-pressure coefficients are used to estimate excess pore pressures that would develop in an earth dam's impervious fill during construction, assuming no pore pressure dissipation. Changes in stress due to overlying fill layers are estimated, and the appropriate pore-pressure coefficients are applied.

THE USE OF PORE-PRESSURE COEFFICIENTS IN PRACTICE

by
A. W. BISHOP, M.A., PH.D., A.M.I.C.E.

SYNOPSIS
The pore-pressure coefficients defined by Les coefficients de pression interstitielle definies
Skempton (1954) have been applied to the problems par Skempton (1954) ont ete appliques aux pro­
blemes du calcul des tensions efficaces dans une bar­
of determining the effective stresses in an earth dam
rage en terre, en cours de construction ainsi que
during construction and during rapid draw-down.
durant un affaissement rapide. Il est fait mention
Reference is made also to several other practical egalement de diverses autres applications de ces
applications of these coefficients. coefficients.

INTRODUCTION

In many practical problems involving the stability or deformation of soil masses, it is


necessary to estimate the magnitude of the changes in pore-pressure resulting from changes
in the state of stress. Where such an estimate is also associated with subsequent field
measurements of pore pressure, it provides a means whereby the margin of uncertainty in
construction work can be greatly reduced.
Typical cases are :
(i) The stressing of natural strata forming the foundation of a structure or of an
earth dam, in which a knowledge of the initial pore pressure is necessary to
estimate the stability during and after construction.
(ii) The stressing of the compacted impervious fill of an earth dam during construction,
due to the weight of the superimposed layers. In this case, the magnitude of
the initial pore pressure controls the stability and the amount of settlement
which can be expected after completion.
(iii) The removal of the water load on the impervious fill of an earth dam due to rapid
draw-down.
(iv) The formation of slopes and cuttings in natural strata in which pore-pressure
changes result from the removal of the weight of the overlying soil.

In each of these cases, the changes in normal stress are accompanied by changes in shear
stress. It is for this reason that the pore-pressure coefficients have proved increasingly
useful during the past few years-firstly, in providing a qualitative picture of the type of
pore-pressure changes which can be expected in any particular problem; and, secondly,
in cases where the stress changes are known with reasonable accuracy, in providing the
basis for quantitative design methods.
The overall accuracy of an estimate of stability or deformation will, of course, be influ­
enced by additional factors, such as the rate of dissipation of the initial pore pressure, and, with
natural slopes in particular, by changes in the values of the shear parameters due to changes
in the state of stress. However, the first step in solving these problems is to determine the
pore pressures which would be set up under undrained conditions.
The use of the pore-pressure coefficients in estimating the stability of an earth dam
founded on a soft clay layer is described by Bishop (1952) arid by Skempton and Bishop
(1955). A siinilar problem-that of estimating the residual pore pressure in "drained"
shear tests carried out in the laboratory under various rates of loading-has been treated on
this basis by Gibson and Henkel (1954).
The pore-pressure coefficients have also been used in a discussion by Henkel and Skempton
(1954) of the decrease in stability of cuts made in over-consolidated clays, and in a discussion
by Bishop and Henkel (1953) of the discrepancy between in-situ shear strength and that
obtained from consolidated undrained tests.
148
THE USE OF PORE-PRESSURE COEFFICIENTS 149
In this article, the use of the coefficients will be briefly illustrated jn te�s of the pore­
pressure changes which occur in the impervious fill of an earth dam during construction and
during rapid draw-down.

PORE PRESSURES SET UP IN THE IMPERVIOUS FILL OF AN EARTH DAM DURING


CONSTRUCTION

An element of soil in the impervious fill of an earth dam (Fig. 1) will be subjected, during
construction, to additional stresses due to the increasing weight of fill placed above .this
point. If, in the simplest case, zero dissipation of pore pressure is assumed to take place,
the excess pore pressure is a function only
of the applied increases in stress. The
problem then consists, firstly, of estimat­
ing the magnitude of the changes in
principal stress, and, secondly, of deter­
mining the appropriate pore-pressure co­ f.tlldenllt)'
= Y
efficients.
Although the directions of the principal Fig.i. Change• in principal Btre•• OD aD
ele.ment during con•tructiOD
stresses will vary considerably along a (.:la1 �y. h)
potential slip surface, the magnitude of the
major principal stress can be taken, as a first approximation, as being equal to the vertical
head of soil above the point considered (Bishop, 1952); The accurate determination of the
other principal stresses would require a detailed analysis in each particular case, and it is
therefore convenient to express the pore pressure as a function of the major principal stress,
and to examine the extent to which this relationship is influenced by variations in the principal
stress ratio.
The pore-pressure ratio is then given by the expression (Skempton, 1954):

-=-
Llu
Ll�
[
B B 1 -(1 - A) 1 -
=

.
( -)]Llu8
Ll�
(1)

where B is the overall pore-pressure coefficient and A and B are defined by the equation :
Llu
= B[Liu3 + A(Liu1 - Llus)].
where Llu denotes the increase in pore pressure and
Llu1 and Llu3 denote the increases in major and minor total principal stresses respectively.
Since, in compacted earth fill materials, A is less than unity and, .in many cases, clo� to
zero, it follows immediately from equation ( 1) that the shear stresses set up during c,gnstruction,
which are represented by values ofLiu3fLiu1 of less than unity, lead to values of Blower than
the pore-pressure ratio B obtained by applying an all-round pressure to the sample.
To determine the extent of this reduction in practice, it is necessary to consider the
limiting states of stress between which the actual case must lie. If the bank were composed
of horiwntal layers of infinite extent, there would be no lateral deformation and the ratio
of the minor and major principal effective stresses would be equal to the coefficient of earth
pressure at rest :
i.e., (2)

If, on the other hand, the slope were sufficiently steep to be in a condition of limiting
equilibrium, the ratio of the principal effective stresses would have its minimum value,
corresponding to failure in the triaxial test. This limit is represented by the value K1.
Actual construction carried out with a factor of safety greater than unity will lie between
these values, and may be represented by the intermediate value K.
150 BISHOP

Equation (1) may be expressed in tenns of the effective stress ratio K by making the
substitution :
Aoa' Ao8- Au Ao1- B.Ao1
K-
_ _
- _
Ao1' Ao1- Au- Ao1- j.Ao1
• (3)

This leads to the expression :


1-(1-A)(I - K)
J =B.1-B.(1-A)(l-K) (4)

Representative values are given in Table 1 to indicate the range within which jj should
lie for two typical soils-a sandy clay and a clay-gravel-placed on the wet side of the
optimum moisture content.

Tahle 1
I


Soil Ko KJ A B
ForK=Ko ForK=KJ

Sandy clay .. .. 0·5 0·3 0·5 0·8 0·75 0·72


Clay-gravel .. .. 0·4 ()-25 0 0·8 0·62 0·50

In practice, the accuracy of such an esti­


-

mate is limited by the fact that both A and B
.,
are influenced by the magnitude of the
"? stresses, and that A is not independent of
. .,; . ,.
., the principal stress ratio. For a more exact
""'
... estimate of J, tests have also been carried
"'
;:)
� •j•K•. fl(
... out in which the natural sequence of stressing
f is followed more closely, by increasing both
...
"'
fli•K. .-.· principal stresses simultaneously so as to
2,. If ofS.•I-51
� maintain the condition either of zero lateral
yield or of a constant factor of safety against
3 .; • K.f.a.'
failure. A typical set of results (Fig. 2) indi­
(f of5 •1·0)
"'
u
� cates that the same type of result is obtained
0 so 100 l$1 as is predicted in Table 1.
INCREASE IN TOTAL I1AIOR "-'NOI'AL 5TU5S
.:ia\ Ub/oq.MI
PORE-PRESSURE CHANGES OCCURRING
Ftc· 2. The iDfl1MoDce ol priDcipal•tz-M8 :ratio DURING RAPID DRAW-DOWN
·
OD the pore-pr- coefllcleDt .1.
(Boulder clay: w = opt. + 1 per cent)
In a similar manner, the initial pore
pressure on rapid draw-down may be esti­
mated in terms of the reduction in the principal stresses due to the removal of the water
load.
The initial pore pressure in an element of soil beneath the upstream slope of an earth
dam in which steady seepage is established (Fig. 3) is flo, where :

Uo =,•. [he+ hr + hw- It'] (5)


After rapid draw-down this will change to u, where

u=flo +Au (6)


and Au= Jl.Ao1 (7)
Thus, u =flo+ B.Ao 1 (8)
THE USE OF PORE-PRESSURE COEFFICIENTS 151
�==---- - _______ ,
1 _, ,YV_.Je.., ...., draw­

�.

Fig. 3. Calc11lat.ion of pore pre•IIUI'e after rapid


draw-doWD

u = y.,. [he+ /ar. (I- E.n) + IJw. (I-ii) -II']

If, as before, the major principal stress is taken as being equal to the vertical head of
material (soil and water) above the element, then,
before draw-down (a1)0 =y.,.h.,+Yr·hr + y.,.h,. (9)
and after draw-down a1 =y.,.h., + Yrll·hr • (10)
where Yc denotes the saturated density of the clay fill ,
Yr denotes the saturated density of the rock fill,
Yftl denotes the drained density of the rock fill,
and y., denotes the density of ·water.
Thus, .1al = (ai)-(al)o =-[ (Yr-Yltl) .h,.+y.,.h,.] (11)
Now Yrtl =Yr-n.y., (12)
where n denotes the specific porosity of the free-draining rock fill.
Hence, .da1 =-y.,.[ n.hr+hw] . (13)
From equations (5) and (8) it then follows that :
u=y.,.[hc+hr. (1-B.n)+hw.(1-B) -h'] (14)
It can be seen from equation (14 ) that, on draw-down, the lower the value of B the greater
will be the residual pore pressure and, th!:_refore, the lower will be the factor of safety. It is
thus of interest to examine the range of B values which may be expected in practice.
For fully saturated soils, B will be almost equal to unity, and the upper limit of A, in
practice, is also unity. The changes in total stress on draw-down are represented by a
decrease in major principal stress and, as the shear stress is increased, by an even larger drop
in minor principal stress. It follows immediately from equation (1) that, with A =1,
B = B=1, and that, for lower values of A, 1i > B, and hence B > 1. As a safe working
rule, therefore, B may be taken as unity (Bishop, 1952).
The expression for u then simplifies to the form :

u=y..,. [he +hr. (1-n) -h'] (15)


This has been found to give satisfactory agreement with field measurements in the case
of the Alcova Dam, in which values of both h' and u were obtained (Glover, Gibbs, and
Daehn, 1948).
Full saturation will not necessarily always occur in practice. In this case, B is less than
unity but, as its magnitude depends on the sign of the change in stress, the values of A and
B measured in the conventional undrained test with increasing principal stresses are not
applicable. The test procedure which has therefore been used consists, firstly, of allowing
the sample to come to equilibrium under values of total major and minor principal stress
and pore pressure corresponding to a typical element under the steady seepage condition­
a flow of water being maintained through the sample to reproduce a representative degree of
152 BISHOP: PORE-PRESSURE COEFFICIENTS

Before draw-down After draw-down


o1 = 44 lbjsq. in. o1 = 30 lbjsq. in.
aa = 36 , , a1 15
= , ,
a0 = 28 , ,. u = 12 .. .. (measured in the triaxial test)
- Au 12-28
Hence B = 401 = = 1·14
30-44
Fig. 4. Laboratory teat to determine pore-preBBlU'e
coefllcient li on draw-down
(Moraine: w = opt. + I per cent)

saturation. The principal stresses are then reduced, under undrained conditions, to re­
produce the decrease in major principal stress and increase in shear stress consequent
on draw-down, and the corresponding changes in pore pressure are measured. The shear
stresses at each stage are deduced from a typical slip circle passing through the element under
consideration.
Tests carried out on a moraine and on a clay-gravel (boulder clay) give values of B
between 1-1 and 1·4. As the degree of saturation is best indicated by the value of B, the
reduction in stress is carried out in two stages-an equal reduction in both ot and u3 to give
the value of B, and a further reduction in u3 to give the additional shear stress and the value
of il
In the test illustrated in Fig. 4, the value of B was 1·14. The corresponding value of B
for a reduction in all-round stress was 0·57. This difference indicates the marked influence
on the pore pressure of the change in shear stress, which is associated, in this case, with an
A value of - 1·0.
Values of B less than unity are, of course, possible for lower values of the increment of
shear. stress. The probable range of values has not yet been investigated in detail, but it
is possible that the lower B values may, in practice, be associated with shear stresses whose
magnitude is insufficient to lead to draw-down failure.

ACKNOWLEDGEMENTS

The laboratory tests referred to in this article were carried out in the Civil Engineering
Department of Imperial College, University of London, and the Author is indebted to
Dr A. R. H. El Ramli and Mr A. J. Watt, who performed most of the tests.

REFERENCES
BisHOP, A. W., 1952. Ph.D. Thesis, London.
BISHOP, A. W., and Henkel, D. ]., 1953. Pore Pressure Changes during Shear in Two Undisturbed Clays.
Proc. Third Int. Conf. Soil Mech., I : 94.
GIBSON, R. E., and HEN KEL, n. J., 1954. Influence of Duration of Tests at Constant Rate of Strain on
Measured " Drained " Strength. Geotechnique, 4 : 1 : 6.
GLOVER, R. E., GIBBS, H. J., and DAEHN, W. W., 1948. Deformability of Earth Materials and its Effect
on the Stability of Earth Dams following a Rapid Drawdown. Proc. Second Int. Conf. Soil Mech. 5:77.
HENKEL, D. ]., and SKEMPTON, A. W., 1954. A Landslide at Jackfield. Proc. European Conf. on Stability
of Earth Slopes, Stockholm, I : 90.
SKEMPTON, A. W., 1954. The Pore-Pressure Coefficients A and B. Geotechnique, 4 :4: 143.
SKEMPTON, A. W., and BISHOP, A. W., 1955. The Gain in Stability due to Pore Pressure Dissipation in a
Soft Clay Foundation. Proc. Fifth Cong. Large Dams, Paris.

You might also like