PUC-Rio, Department of Civil Engineering

ENG1211 – Mecânica dos Solos – Prof. Celso Romanel

2. THEORY OF PRIMARY DENSIFICATION
2.1 Introduction
Densification is the gradual reduction in volume of a saturated soil of low.
permeability due to the drainage of water from its voids, with the process
hydraulic-mechanical continuing until the excess pore pressure generated by
a total voltage increase has been completely dissipated. The
densification involves the following interconnected phenomena: increase in
effective stress, decrease in pore pressure, decrease in moisture content
decrease in soil volume and rearrangement of solid particles.
If the excess pore pressure is negative, with the soil increasing in volume in
tempo, o processo é chamado de expansão.
Mathematical theories that describe the phenomenon of excess dissipation.
poropressure, with the consequent gradual increase in effective stress in the skeleton
solid and the occurrence of soil deformations over time are known
as theories of primary densification. These theories can be classified into
two major categories: in the first, associated with the names of Terzaghi (1923) and
Rendulic (1936) assumes that the total stresses generated by loading
external remain constant during consolidation. The theory proposed by
Terzaghi was developed for the state of 1D deformation (vertical deformations.
only) and generalized by Rendulic for the 3D state of deformations. One of
difficulties of this type of approach is the absence of coupling between the
deformation of the solid skeleton and the dissipation of excess pore pressure
water present in the pores of saturated soil.
In the second category of primary consolidation theories, the soil is modeled as
a more realistic form as a saturated porous medium and linearly elastic under
state of 3D deformations. This approach, known as theory of
poroelasticity was developed by Biot (1935, 1941) and reformulated by Biot
(1955, 1956), Verruijt (1969) and Rice and Cleary (1976). The hydro-coupling
mechanics is described by a set of partial differential equations,
involving mechanical variables (stresses, pore pressure) and kinematics
(displacements, velocities) relative to the solid and fluid phases of saturated soil.
Most of this chapter is dedicated to the study of Terzaghi's 1D theory.
including experimental methods for evaluating geotechnical parameters,
approximate solutions (algebraic and graphic), formulations for calculating the
consolidation recalculation, adaptations of Terzaghi's theory for consideration
of 3D effects, techniques for accelerating the dissipation of excesses
pore pressure (vertical drains), imposition of pre-loading for acceleration
of reminders, among other topics.
2.2 GOVERNING DIFFERENTIAL EQUATION OF FLOW IN POROUS MEDIA
Figure 2.1 shows the movement of water along curved paths, called
of flow lines, in the foundation soil of a large concrete dam
C. Romanel
58
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
perpendicular length to the plane of the paper, with a waterproof curtain to
amount, characterizing a 2D flow problem.
Consider the flow conditions at point P, represented at an enlarged scale.
through the prism of infinitesimal edges dx, dy, dz in the same figure. The speed in the
point P is tangent to the line of flow and can be represented by its
horizontal componentsxthe verticalyremembering thatz= 0 in this 2D problem.
(a)
vy
vy
dy
y
vx
x
vx
vx 
dx
vy
(b)
Figure 2.1‐(a) 2D flow in the foundation soil of an impermeable concrete dam; (b)
components of velocity in the infinitesimal element that represents point P.
Assuming that water is incompressible, the infinitesimal flow in the element,
defined as the difference between the input flow dQentand the output flow
ai
dQin theit can be written as,
dQ dQent dQsai
(2.1)
dQent vxdydz vydxdz
(2.2)
(2.3)
vy


vx
x


dQsai v
d d d v 
dydx dz




x
y
z
y

x




y

C. Romanel
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
(2.4)
vy


vx
x




dQ 

dxdydz
y


The flow in the volume element dV = dxdydzcan also be expressed in
function of the variation over time of the volume of water dVwpresent in the voids dVw
= SdVvwhere the saturation degree and dVvthe infinitesimal volume of voids.
(2.5)
edV
1 e
t




S

d d d


SdVv




t

If
x y z

dQ

dVw




t
t
1 e
observing whatxdydzdV = (1+e)s it is the constant value of the volume of solids,
regardless of the time.
As Eqs. (2.4) and (2.5) are equivalent,
2.6
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Eq. 2.10 governs the 2D consolidation where the soil remains saturated but the
the volume of voids decreases (compaction) or increases (expansion) over time.
Hypothesis 4: in equation (2.9) the void index remains constantandt 0e
saturation level remains constantSt 0.
2h
x2
2h
y2
2h2h
k
k
0


0

2h 0
(2.11)
x2y2
Eq. 2.11 governs the permanent 2D flow where the soil remains with volume and
degree of saturation (S =1) constants. This linear partial differential equation is
known as Laplace's equation.
2.3 PRIMARY CONSOLIDATION THEORY OF TERZAGHI
The theory of Terzaghi's consolidation studies the variation of the field of
poropressures over time, with the deformation of the solid skeleton being determined
independently based on the assumed equivalence between the percentage
average dissipation of excess overpressure Uvand the percentage of
reprint U, at time t. Terzaghi's theory is widely used in practice.
Geotechnical engineering despite some limitations introduced by its
hipóteses básicas: a) solo isotrópico, completamente saturado; b) grãos minerais
and water are incompressible; c) validity of Darcy's law; d) the coefficient of
dense cvremains constant during densification; e) there is a single
linear relationship between void ratio and effective vertical stress that remains
constant during compaction; f) the deformations that occur in the soil are
infinitesimals.
The greatest limitations1they relate to hypotheses d), e). Evidence
experiments in real soils show that the consolidation coefficient
decreases rapidly when the effective vertical stress reaches the pre-pressure
the densification and the stress x strain behavior is neither linear nor
elastic. It should be emphasized at this point that the non-linearity of the stress x curve
deformation can be incorporated in obtaining the final settlement of densification
primary but not in the solution of the consolidation equation established by
Terzaghi.
Consider the governing differential equation for 2D consolidation (Eq. 2.10)
2h
x2
2h
y2
1e
1 et
(2.10)
k
k

For the 1D situation, assuming z as the vertical axis
1
Considering that Terzaghi's theory produces acceptable results in many field cases, some
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
2h
z2
1e
1 et
(2.12)
kz

where
2h2
1
(2.13)





he 

uss ue

z2z2


w


How the lifting loadandvaries linearly with the vertical coordinate z,
2
2
h
z
0
e
it results in Eq. (2.13) that
  As in the permanent condition the
2
2
u
z
0
poropressão uss varies linearly with depth3, then
  and a
2
ss
Equation 2.13 can be rewritten as:
kz2ue
1e
1 andt
(2.14)

 wz2
Considering a linear stress-strain relationship for the solid skeleton,
4
it is defined as the coefficient of compressibility av o f d 'v(Fig. 2.2). Thus,
by the chain rule of differential calculus and in conjunction with the principle of
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
2ue ue
(2.17)
cv

z2
t
The partial differential equation Eq. 2.17 is linear, that is, the excess pore pressure ue
and its derivatives appear only in the first power and there is also no
products between the variable ue and its derivatives. The superposition of solutions is
valid and the linearity of the governing equation makes the use of
normalized parameters with the aim of expressing Eq. 2.17 in a way
dimensionless and thus obtain a general mathematical solution that facilitates the
presentation of results in the form of graphs and tables.
Then define the dimensionless parameters,
(2.18)
z
c vt
H2
Z
e
Tv
H
where z represents the depth of a point from the top of the soil layer
saturated, Z the dimensionless depth, Tvthe time factor, H the thickness of
drainage of the layer, equal to its actual thickness divided by the number of faces
of drainage (2 for drainage at the top and base; 1 for drainage at the top or in
base).
cv2ue
1ue
H2Tv
cv
2ueue



(2.19)
H2Z2
Z2Tv
The solution of the partial differential equation (Eq. 2.19) will naturally depend
still of the specific boundary conditions of the problem and of a condition
initial, according to some typical solutions that will be addressed in the section
next.
Figure 2.2 – Assumed linear relationship between the variation of the voids index (e) and the variation of

effective vertical stress ('vCraig (2004).

C. Romanel
63
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
2.4 SOLUTIONS OF THE DIFFUSION EQUATION
2.4.1 – Case 1
a) Initial condition - at the initial moment t0excess of pressure u0uniform to
length of thickness b of the layer for 0≤z≤b.
b) Boundary conditions - for t > 0, null excess pore pressure uand= 0 nos
contours at z = 0 (Z = 0) and at z = b (Z = 2), assuming two faces of
drainage (b = 2H).
Before proceeding, one wonders what kind of load would produce a
initial excess of poropressure u0uniform with depth and deformations
null lateral? Consider the perfectly flexible circular foundation of radius R,
uniformly chargedq), on the surface of a linear half-space
elastic, homogeneous and isotropic. As mentioned in chapter 1, the distribution
of the increases in principal stresses caused by the loading of the foundation
it can be determined analytically (Poulos and Davis, 1974) or graphically, as
Fig. 2.3.
In this figure, note that the horizontal deformations are null at x/R = 0 due to
to the symmetry of the problem. Assuming that the radius of the circular foundation increases
infinitely such that, for any point in the soil mass P(x,z), it results in the
normalized valuesx / R 0ez/ R 0then the increase in vertical tension
tends to become constant with depth and equal to the value of the load
appliedq. Note also that the horizontal deformations would also be
null because any vertical axis for a foundation with infinite radius can be
admitted as an axis of symmetry (x / R 0The designation 1D for the densification
it means, therefore, unidimensional in terms of deformations, i.e. they occur
only vertical deformations. In the field, the conditions for creation of
1D deformations therefore imply the application of a load (fill,
large dimension radier) and, in the laboratory, in the lateral confinement of the
tested soil samples.
Another question that could be asked is why the initial excess of pore pressure
u0is accepted as an increase in loadingq?
Consider a saturated soil element of volume V subjected to an increase in
total stress in the undrained condition, in the 1-D state of deformation (without
horizontal deformations). A volume variation dV must occur. As the
the compressibility of mineral grains is much lower than the compressibility
of water, which can be ignored, then at the moment of loading and before that
drainage begins, the variation in the volume of the soil element corresponds to
a variation in the volume of water present in its voids.
C. Romanel
64
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.3 - Distribution of the increases in principal stresses in a linear half-space
elastic, isotropic, and homogeneous generated by the uniform loading ∆q on a circular foundation
perfectly flexible with radius R. Distribution of increments3paraν = 0.45 (Lambe and Whitman,
1969).
The compressibility of water Cw is defined as
dVwVw
Cw 
(2.20)
you
where dVwit is the variation of the volume of water resulting from the application of the increment
of the pressure in the volume Vw
e
HowVv

Vw 
V
so
1 e
e



dVw CwVwyou C
In you
w


(2.21)
1 e
The variation in the volume of the soil element represents not only a
compression of water as well as a compression of the solid skeleton, whose
the compressibility coefficient was previously defined as


dVv
of
d v'
1
dV 1 e




av 


(2.22)
Vsd v'
In d v'


The variation in volume of the solid skeleton is therefore
C. Romanel
65
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
av
dV 
Vd v'
(2.23)
1 e
The variations in the volumes of water and the solid skeleton must be equal. Therefore,
av
e
C
Vdu 
Vd v'
w 1 e
1 e
Cwe
av
d v' 
you
(2.24)
According to Terzaghi's effective stress principle for saturated soils,
Cwe
d v of you
you
av
you
1
Cwe
av

2.25
d v
1
The compressibility coefficient5from water is Cw= 0.458 × 10-9Pa
approximately. For typical values of void indices and coefficients of
compressibility of the solid skeleton, Eq. (2.25) results approximately
you d v 0.999Logo, under the assumed loading conditions,
essentially all the total vertical tension increase

 v  q is broadcast

for the water in the form of initial excess pore pressure u0.
The analytical solution of the differential equation 2.19 is given by the Fourier series.
(Taylor, 1948)

2
2u0
M
m0

2
u

senMZ

PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
vertical, for any values of Z and Tvis proportional to the value of the gradient
hydraulic that generates the transient vertical flow of water.
Between points A and B of Fig. 2.5, the transient, vertical, and ascending flow at the factor
time TvIt occurs due to the following difference in hydraulic pressure
A
B
uss ueA
uss ueB

 

A
B
A
B

 
 he


 h  h h he
(2.28)
(2.29)

 
 
w
w


uA
uB
ss 

ueA ueB
u0





 

A
e
B
 h h
 he
 



hsAs hsBs



PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.4 - Percentage of excess pore pressure dissipation Uzzdue to the
normalized depth Z and the time factor Tvfor uniform excess of initial pore pressure u0
(Lambe and Whitman, 1969).
Figure 2.5 - Hydraulic head differenceh between points A and B in time factor Tv.
C. Romanel
68
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
c) for TvThe 0.3 isochrones are almost perfect sine curves, i.e.
only the first term of the series from Eq. 2.26 (m = 0 with M =
important for the definition of the isochrone.
becomes
2
4

Tv
4u0

2





ue
You are Z
for
Tv  0.3
2.32


In terms of geotechnical engineering, it is more convenient to know, for certain
value of the time factor Tvthe average percentage of dissipation of excesses
U poropressurev throughout the entire thickness of the saturated soil layer and not only
at isolated points in the normalized depth Z.
u0 ue
ue
u0
Uv
 1
(2.33)
u0
wheree it is the average excess pore pressure in the saturated soil layer for
determined value of Tv.
The value wasecan be calculated based on the equivalence of the areas in Fig. 2.6,
a comprehended between the generic isochrone Tvand the vertical axis Tv  (when
Uz= 100%) and another area of rectangular shape, with the same normalized thickness
(Z=2), and the width equal to the average value of the excess pore pressure uein the same
time factor Tv.
2
2

M2Tv

 2u
1

2

ue
e dZ with M 2m 1
0
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel

2
2u0
u
eM T
v
M
e
2
(2.35)
m0

2
ue
2

2
Uv(Tv )  1
 1
eM T with M

2m 1

v

m0M
2

2.36
u0
Equation (2.36) can also be approximated, with good precision, by the following
algebraic expressions proposed by Fox (1948):

4
Tv  Uv2for U 0.60
(2.37)
Tv 0 .933log(1U)0.085 for U 0.60
or, alternatively, be estimated through graphical solution represented in the
Fig. 2.7 by the curve (1). A comparison of the values obtained by the equations
(2.36) and (2.37) are shown in Table 2.2.
t 
Theoretically, primary consolidation extends infinitely over time (
)
but, for engineering purposes, a value of the time factor T could be chosenv
what indicated, in practice, the end of primary consolidation? Based on the
average percentage values of excess pore pressure dissipation Uvyes
Table 2.1 is generally considered for engineering purposes the end of
primary densification when Tv= 1 (Lambe and Whitman, 1969).
Table 2.1 - Average percentage of dissipation of excess pore pressure for some values
of time factor Tv.
Tv
Uv(%)
Tv
Uv(%)
Tv
Uv(%)
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
simple drainage (case 2) will achieve the same average percentage of dissipation
of the excesses of poropression Uv at time t2= 4t1.
(2.38)
cvt1
cvt2
b2
Uv

Tv1

Uv

Tv2


Tv1 Tv2


 t2 4t1
2
b
 
 
2
 
Table 2.2 - Comparison of average percentage values of excess dissipation
poropression Uv.
Tv
Uv (%)
Eq. 2.36
03:57
11.28
15.96
19.54
22.57
25.23
27.64
29.85
31.92
33.85
35.68
50.41
61.32
69.79
76.40
81.56
85.59
88.74
91.20
93.13
Uv(%)
Eq. 2.37
03:57
11.28
15.96
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
(Fig. 2.8d) where the distribution of pore pressure increments along the axis
can be approximated by a trapezoidal distribution (compare with the
distribution of the vertical stress increments obtained from the Boussinesq solution.
curve 1
curve 1
curve 1
curve 1
curve 2
curve 3
Figure 2.7 – Relationship between average percentage of excess pore pressure dissipation Uv e
time factor Tv(Craig, 2004).
C. Romanel
72
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Figure 2.8a - A clay layer initially subjected to the permanent 1D flow generated in
lower sand layer (artesian aquifer) whose pressure load is instantly reduced
for the value of the hydrostatic pressure load through pumping.
Figure 2.8b - The hydraulic load of the clay layer is in equilibrium with the level of the water table.
phreatic 1, which is suddenly elevated to level 2 due to the filling of a reservoir of a
dam located nearby. It is assumed that the hydraulic load of the lower sand layer
do not be modified. Note the combined occurrence of transient flow (upward and downward)
the permanent flow (downward, from the upper sand layer to the lower sand layer).
Figure 2.8c – Triangular distribution of the initial excess pore pressure caused by its own weight
do material (Jumikis, 1962).
C. Romanel
73
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.8d - Approximate trapezoidal distribution of initial excess pore pressure caused by
superficial foundation loading (Jumikis, 1962)
a) Initial condition - at the initial moment t0triangular distribution of excess
poropression u0along the thickness b of the layer for 0≤z≤b.
b) Boundary conditions - for t > 0 null excess pore pressure uthe= 0 nos
contours at z = 0 (Z = 0) and at z = b (Z = 2), assuming two surfaces of
drainage (b = 2H).
The analytical solution of the differential equation Eq. 2.19 in this case is given by
2 2
n1
n

T
v
2u0b
n Z
2






(2.40)
4
ue 

1

you
e
com
n1,2 ...

1n


n
b
u
where Z is measured from the vertex of the triangular distribution and 0represents the
initial excess of pressure at the base of this distribution.
The percentage of dissipation of the excess pore pressure Uzin depth
normalized Z in the time factor Tvis obtained by Eq. 2.41 and illustrated in Fig. 2.9.
u0b ue
ue
u0b
Uz
1
u0b
(2.41)
2 2
n T
v

PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.9
Percentage of excess pore pressure dissipation Uz in depth
normalized Z, in the time factor Tvfor initial triangular distribution of u0.

2
Figure 2.10 - Equivalence of distribution areas of excess pore pressure
for a certain time factor Tv.
2 2
n T
v
ub2
2u0b
n Z
2
2
 u dZ
ub 2

1n1

you
e
dZ n1,2,...






4




e
2
n1n


0
0
2 2
n

T
v



2u0b
2


ub 
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
regarding the initial excessive porosityu0bon the basis of the triangular distribution,
is determined by
n2 2Tv
4
ub
8
n2 2

Uv (Tv)  1
 1
e
with n 1,3,5,...,
2.42

u0b
n1

2

2
Comparing Eqs. 2.36 and 2.42, withM (2m1) n, it is confirmed that the
average percentage of dissipation of excess pore pressure Uv= Uv(Tv) for
the triangular distribution with double drainage face (Eq. 2.42) is exactly
similar to case 1 (Eq. 2.36). The approximate algebraic solutions (Eqs. 2.37),
as well as the graphical solution of curve 1 of Fig. 2.7, can then be applied.
Values of Uv= Uv(Tv) remain the same also for a distribution
trapezoidal of the initial excess pore pressure, according to Table 2.2.
2.4.4 – Case 4
a) Initial condition - at the initial moment t0triangular distribution of excess
poropressure u0along the thickness b of the layer for 0≤z≤b.
b) Boundary conditions - for t > 0 null pore pressure excess ue= 0 no
contour at z = 0 (Z = 0) and zero flow velocityi h z ue z 0) no
waterproof contour in z = b (Z = 1), allowing for a single drainage face (b =
H). Alternatively, with boundary conditions exchanged between the
contours at Z = 0 and Z = 1.
The distribution of excess pore pressure with depth for t > 0 must
to be obtained through an approximate numerical solution of Eq. 2.27, since in the
During the densification process, there is a reversal of the transient flow direction (Fig.
Until a certain value of the time factor Tvdrainage occurs on both
senses, delimited by the isochrone point with vertical tangent. From this
value of Tvthe drainage only occurs from the impermeable contour to
the permeable contour. Regarding the average percentage of dissipation of
excesses of poropressure Uv= Uv(Tvthe graphic solutions are shown in
Fig. 2.7 through curves 2 and 3, depending on the position of the impermeable face
in relation to the initial triangular distribution of u0Table 2.3 shows the relationship
between Tve Uvfor both cases of initial distribution of excesses

poropression when the base of the triangular distribution is next to the

waterproof surface when the vertex of the triangular distribution is
next to the waterproof surface. Note that
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.11 - Isoclines for initial triangular distribution of excess pore pressure with single
drainage face, with reversal of flow direction during the densification (Craig, 2004).
Table 2.3 - Average percentage of excess pore pressure dissipation Uvfor distribution
initial triangular of excess pore pressure and single drainage face (Jumikis, 1962).
Tv
Uv(%)
Curve 2 of Fig. 2.7
0.85
Uv(%)
Curve 3 of Fig. 2.7
15.05
0.004
0.008
0.012
0.020
0.028
0.036
0.048
0.060
0.072
0.083
0.100
0.125
0.150
0.167
0.175
0.200
0.250
0.300
0.350
0.400
0.500
0.600
0.700
0.800
0.900
1
1.62
2.41
4.00
5.60
7.20
9.50
19.14
22.55
27.96
32.18
35.62
39.78
43.30
46.20
48.20
51.48
55.36
58.54
60.46
61.30
63.78
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
2.4.5 - Case 5
Consider that at an unknown time a granular soil landfill was placed
on the surface of a saturated clay deposit (Fig. 2.12). Currently,
it is intended to add a final layer of compacted fill for a
construction at this location. Piezometers were installed in the clay layer.
saturated and it was found that there are excess residual poropressure, due to the
previous loading, not uniform with depth but that can be
approximated by a sinusoidal curve (note that in cases 1 and 2 for Tv 0.3
the isochrones are almost sinusoidal curves). For the new landfill, the excesses of
Poropressure at a certain time t is calculated in two stages: a) layer
with double drainage, initial excess of uniform pore pressure with the
depth; b) layer with double drainage, initial excess of pore pressure
described by a sine curve, and the partial results summed.
Excesses of poropressure due to an initial sinusoidal distribution of u0, with
maximum value us in the center of the layer, can be obtained for 0≤Z≤2 through
from Eq. 2.45, while the corresponding values of the average percentage of
dissipation of excessive pore pressure Uv= Uv(Tvcan be calculated by
Eq. 2.46.
1
4
2
 T
Z
2
v






u you are
e
e
s
(2.45)
(2.46)
1
4
2
v
U 1 e  T
The average percentage of dissipation of excess pore pressure due to the
composite loading can be obtained by the weighted average (Eq. 2.47)
2
U A

j
j
j1
U
(2.47)
2
Aj

j1
where Ujit is the average percentage of dissipation of pore pressure of the loading j
PUC-Rio, Department of Civil Engineering
ENG1211 – Mecânica dos Solos – Prof. Celso Romanel
Figure 2.12 - Initial distribution of pore pressure excess composed of a distribution
uniform with depth (new landfill) and a sinusoidal distribution due to excesses
residuals of pore pressure (previous landfill) - (Olson, 1989)
Table 2.4 - Average percentage of dissipation of excess pore pressure Uvfor distribution
sine initial of excess pore pressure with double drainage face (Jumikis, 1962).
Tv
0.004
0.008
0.012
0.020
0.028
0.036
0.048
0.060
0.072
0.083
0.100
0.125
0.150
0.167
0.175
0.200
0.250
0.300
0.350
0.400
0.500
0.600
0.700
0.800
0.900
1
6.49
8.62
0.98
1.95
2.92
4.81
6.67
10.49
13.67
16.38
18.76
21.96
24.81
27.43
29.67
32.88
37.19
41.12
43.61
44.73
48.09
54.17
59.50
64.21
68.36
75.28
80.69
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
2.5 TIME-DEPENDENT LOADING
The solutions presented in the previous item were obtained with the assumption of
instant loading, without variation of the total voltage over time. In terms
practical, this equates to considering the construction time very small in
relation to the time necessary for the dissipation of excess pore pressure
generated by the loading (i.e. time required to reach Tv= 1).
There are situations, however, where the construction time cannot be ignored in
calculations such as, for example, in the construction of landfills that can take time
weeks or months to be completed. In such cases, the usual practice is
subdivide the load versus time diagram (q x t) into a number
adequate for partial loadings and consider them applied instantaneously
in the center of the different time intervals as shown in Fig. 2.13. The settlement curve
the time of each partial loading is obtained and all are summed to
produce an approximate final curve of settlement x time considering the
time-dependent loading.
q
q
qc
q2
q1
qc
q2
q1
t1
t2tc
t
t
Figure 2.13 - Subdivision of time-dependent loading into instantaneous loads
partials.
Another procedure was suggested by Olson (1977) for loadings that
they grow linearly over time, reaching a constant valuecat the final time of
constructionc(Fig. 2.14).
For the case of a layer with dual drainage, the infinitesimal excess of
initial poropression0constant with the depth, generated by the application of
infinitesimal increment of instantaneous loadiis expressed by
qc
you0 dq dtI
(2.48)
tc
In time at time t– tiAt the normalized depth Z, the infinitesimal excess
the oppression comes from Eq. 2.26,
2








qc
M cv(t  ti)
2

of you
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Figure 2.14 - Loading function versus time, linear for t≤tcit is constant for t > tc
(Olson, 1989).
Equation 2.49 must be integrated considering the loading periods.
increasing (t < tc) and of constant loading (t > tc).
For t < tcresult,
ue
t
2








q
M cv(t ti)
2
2
cyou


MZ

exp
dtI

d
ue 

(2.50)
(2.51)
H2
 
0M t
m
c
0
0

2
qc
u

you

MZ


1 exp

 M2Tv



e
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel

1
2
M4
U 1

exp

M2Tc

1

exp

 M2Tv

(2.55)
Tc
m
0
The graphs in Fig. 2.15 show the evolution of the average percentage of
dissipation of excess poropressure Uvfor several values of the time factor of
T constructionc.
Figure 2.15 – Average percentage of excess pore pressure dissipation Uv= UvTv) for
various values of the construction time factor Tcconsidering linear loads
crescent (Olson, 1977).
2.6 PRIMARY DENSIFICATION OF ADJACENT LAYERS
Solutions for a system formed by two contiguous layers with double or
the only drainage face was presented by Gray (1945), solving the
Terzaghi's differential equation in each layer separately and imposing the
continuity conditions (equal values of pore pressure excess ueand of
flow velocity) at the interface.
Such equations are not widely used in practice because they are laborious to
will be evaluated for specific cases and also because of the number of variables.
involved is large for the presentation of a general solution in the form of
graphs or tables. For systems composed of multiple layers, a solution
general was presented by Schiffman and Stein (1970), but is also difficult
determination for engineering applications.
An approximate analysis method was suggested by the U.S. Navy (1962) for the case
of 2 contiguous layers, with double drainage (base and top of the system), when
one of the layers has a permeability coefficient (or coefficient of
density) at least twenty times higher than the coefficient of the adjacent layer.
C. Romanel
82
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
In this case, one can consider the densification process subdivided into two
independent stages: a) first, the most permeable layer densifies with
simple drainage until reaching time factor Tv= 1; b) from this time on, it starts-
if the densification of the other layer (less permeable) with double drainage.
2.7 DETERMINATION OF SOIL PARAMETERS IN THE LABORATORY
The experimental determination of the geotechnical parameters required for
estimation of the consolidation speed (consolidation coefficient) well
as for the calculation of primary consolidation settlement (parameters of
compressibility) is usually done in a device developed by
Terzaghi (1923), refined by Casagrande (1936), called an oedometer
It is basically made up of an application system
vertical loading (compaction press), a compaction cell that
contains the soil sample, a confinement ring, porous stones, and a
rigid plate for load transmission.
The lateral deformation of the soil sample is prevented by the ring, occurring therefore
only vertical deformations during the test. Drainage is allowed through
the porous stones placed at the top and bottom of the sample, so that the
water expelled from the pores during densification should flow in a vertical direction.
When a loading increment is applied to the soil sample, there is a
instantaneous increase in pore pressure equal to the value of the stress increase
vertical total. As drainage occurs, the excess pore pressures
dissipate and the effective tensions grow over time, until the loading
be fully transferred to the solid skeleton in the form of effective stresses, the
which gradually deforms, causing a vertical displacement (settlement)
measured by a strain gauge during the test.
The testing procedure consists of applying incremental loads to the
compaction cell (20kPa, 40kPa, 80kPa, 160kPa, 320kPa, 640kPa)
keeping them constant for up to 24 hours (or more) depending on the type of soil.
For each of the charging stages, the variation over time of the
the thickness of the sample is recorded through readings taken on the extensometer in the
sequence of times 1/8, 1/4, 1/2, 1, 2, 4, 8, 15, 30 minutes and 1, 2, 4, 8, 24 hours
or more. If it is necessary to understand the sample expansion behavior
from alone, the loads are removed in the reverse order, with the recordings being made again
extensometer readings over time during the unloading phase.
For each loading stage, the readings are used for determination
of the consolidation coefficient cv, normally done through two methods of
interpretation: a) Taylor method ort; b) Casagrande method oulog t.
C. Romanel
83
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.16 – Compaction cell, porous stones, and filter paper (above), schematic of the cell
compaction with soil sample and execution of the consolidation test at the PUC-Rio compaction press
(below).
2.7.1 Consolidation coefficient cv
a) Taylor method (1948)
a.1) Theoretical justification
Considering the determination of the average percentage of dissipation of
excesses of poropressure for Uv≤0.60 is approximated by Eq. 2.37, then in
a U graphvversusTVthat equation is represented by a straight line with

inclination
in relation to the vertical axis (Fig. 2.17). A new line drawn
2
by origin (Uv= 0), with a 15% higher inclination, intersects the theoretical curve at
Uv = 90%, corresponding to the time factor 0.848.

C. Romanel
84
 PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Figure 2.17 - Theoretical and experimental curves of the Taylor method for determination of
coefficient of consolidation.
In the laboratory, the readings of the extensometer are plotted against the square root
of the measurement times and an experimental curve is then plotted through these
points. How the densification recalculation of the soil sample and the drainage
they occur in the vertical direction (1D), then there is a direct proportionality
between the evolution of the extensometer readings (experimental curve) and the
average percentage of excess pore pressure dissipation Uv(curve
theoretical). Similarly, as the consolidation coefficient is considered
constant during the charging stage, there is also proportionality
entre o fator tempo Tvand the time
t
.
a.2) Determination of the consolidation coefficient cv
i) For each load increment, construct the consolidation curve
marking on the y-axis the readings of the extensometer (variation of
thickness of the soil sample) and on the x-axis the corresponding
values of the square root of the time since the beginning of the trial;
ii) determine the corrected initial reading point of the extensometer dsprolonging the
line defined by the initial points of the test;
iii) from point dsdraw a new line with inclination relative to the axis
vertical 1.15 times higher than the slope of the initial segment,
extending it until it intercepts the laboratory curve at the point with coordinates (
t90, d90);
2
.0
c

iv) the consolidation coefficient is calculated asv
t90
C. Romanel
85
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
in the Taylor method, the end of densification can also be estimated
10
calculating the value of the readingd100 ds

ds  d90

.
9
b) Casagrande method (1940)
b.1) Theoretical justification
Uv(Tvis plotted on a logarithmic scale for the values of
The theoretical U curvev=
time factor TvThe curve is composed of a parabolic section for Uv≤0.60
followed by an intermediate linear section and a final curve for which the axis of
abscissas is a horizontal asymptote (Fig. 2.18). The corresponding value to Uv= 0
is obtained by selecting two time factor values in the initial parabolic section
(Tv1e Tv2) such that Tv2/Tv1= 4. Logo,

4

Uv22
2.56
Tv2
Tv1
4
1
Uv2
Uv1



2
Uv21
4
The position of the starting point can be marked as indicated in Fig. 2.18.
The densification theoretically ends (Uv= 100%) for a time t 
.
However, a finite approximation can be made for considering the

intersection of the asymptote (x-axis) with the extension of the segment

intermediate linear of the curve. The Casagrande method is based on the value of
time factor corresponding to 50% average dissipation of excesses

poropressure, calculated as /2. For the theoretical value

the correspondent of the time factor is 0.197.

Figure 2.18 - Theoretical Curve Uv= Uv(Tv) plotted with time factor on a logarithmic scale.
C. Romanel
86
 PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
b.2) Determination of the consolidation coefficient cv
i) For each loading increment, construct the settlement curve.
marking on the y-axis the readings of the extensometer (variation of
thickness of the soil sample) and on the x-axis, on a logarithmic scale,
the corresponding time values;
ii) determine the point corresponding to the corrected initial reading of the extensometer ds
selecting, in the initial parabolic section of the essay, two separate points
for the reason t2/t1Measure the difference between the readings of the extensometer in these
time and transport it above the reading of t1to thus establish the value of ds;
iii) determine point d100corresponding to Uv= 100%, at the intersection between the
reta do trecho intermediário da curva de adensamento e a tangente ao seu trecho
final (Fig. 2.19). Due to the possibility of secondary compression occurrence
In the soil sample, this tangent may not be horizontal, as expected by
primary consolidation theory of Terzaghi;
iv) considering the average reading of the extensometer (ds+ d100)/2 determine the time
t50corresponding to 50% density;
2
.0
c

v) the consolidation coefficient is calculated asv
t50
Figure 2.19 - Theoretical and experimental curves of the Casagrande method for determination of
consolidation coefficient.
One of the difficulties in estimating the consolidation coefficient, both by
Taylor's method as well as Casagrande's is that the value of the thickness of
drainage H, squared, strongly influences the final result. It can
there is a big difference between values of cvcalculated with the initial thickness of the
sample and with some other reduced thickness value, such as the
correspondents for the readings of the extensometer d100you d50, for example.
Duncan (1993) recommends using the value of the initial drainage thickness of the
C. Romanel
87
PUC-Rio, Departamento de Engenharia Civil
ENG1211 - Soil Mechanics - Prof. Celso Romanel
sample but usually in the interpretations of laboratory tests considers
if the value of the average drainage thickness H = (d100- ds) / 2.
Both methods (Taylor and Casagrande) use the same equation (Eq. 2.37)
to establish the value of the corrected initial reading ds but differ in relation to
method by which they estimate reading of100corresponding to the end of the densification
primary. Generally speaking, the Taylor method is more affected by compression
initial of the sample, whose effect is to increase the value of cv, while the method of
Casagrande is more affected by the occurrence of secondary compression.
producing values of cvusually lower. Both methods are
theoretically valid and should produce the same value for the coefficient of
thickening if the theoretical curve Uvversus Tvwas faithfully reproduced by the
laboratory results. Duncan (1993), based on the finding that in
in most cases, the pumping speeds estimated by conventional theory
the density measurements are lower than the field measurements, it suggests the use of
Taylor method which, by using larger values of cvwould tend to decrease the
difference between predicted and observed pumping speeds.
The literature records several other methods, much less known than the two.
classic methods presented here, which seek to establish the value of
coefficient of consolidation through the consideration of other points and
characteristics of the densification curve, such as the speed method (Parkin
1978, 1981); the rectangular hyperbola method (Sridharan and Rao, 1981; Sridharan
et al., 1987); the log method (H2(t) – U (Sridharan et al., 1995), the log method –
log t (Sridharan and Prakash, 1997), the methods proposed by Scott (1961), Cour
(1971), Sivaram and Swamee (1977), Robinson and Allam (1996), Robinson (1999)
Mesri et al. (1999), Feng and Lee (2001), Singh (2007), Al-Zoubi (2008), among others
others.
2.7.2 Compressibility parameters
Typical graphs of void ratio values (e) after compaction versus
effective vertical stress

 v'
are shown in Fig. 2.20, displaying a compression

initial, followed by a cycle of expansion and recompression. The shape of these
curves depend on the soil's history of stresses, being approximately linear
for normally consolidated clays (NC) and having a curved initial section for
pre-compressed clays (PA).
In the unloading, it is also observed that the soil does not follow the same
tension trajectory of the charging phase, indicating that deformations
irrecoverable plastics occurred.
The compressibility of the soil, based on the behavior of the graphs in Fig.
2.20 can be represented by the following parameters:
C. Romanel
88
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Figure 2.20 - Graphs of void ratio versus effective vertical stress (Craig, 2004).
a) Compressibility index Cc the recompression index Cr
The results of the odometer test are normally presented in a
semi-log graph (Fig. 2.20, on the right) where on the y-axis, on the scale
arithmetic, the values of the void index are plotted at the end of the stage of
loading and on the x-axis, in logarithmic scale, the
corresponding values of effective vertical stress.
For low values of effective voltage, the rate of decrease of the void index is
normally small, highlighting that it is about the recompression of a soil
pre-densified. In fact, when taking an undeformed sample from the interior of
soil mass has a relief of tensions and the sample will behave as pre-
compressed while the maximum voltage value it has previously been subjected to
field (called pre-consolidation pressure) v'mis not surpassed by the
effective vertical tension induced by the loading of the test.
The recompression index Crcan be defined in the pre-densification section or,
similarly, in an expansion and recompression cycle. In general, it is recommended
the determination of Crin the expansion and recompression cycle because in the section
the initial sample may still display effects of disturbance experienced during the stage
of sample preparation. Leonards (1976) suggests that this cycle be executed
close to the value of pre-consolidation pressure and that the same slope
average to be adopted to identify the recompression and expansion trajectories.
e
log v'
Cr 
(2.57)
For effective vertical stress greater than the preconsolidation pressure, it is defined in
linear segment (called virgin line) the compressibility index Cc
e
log v'
CC 
(2.58)
C. Romanel
89
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Typical values of Cc in saturated clays vary from 0.1 to 0.5, increasing with the
plasticity index of the soil. In peats and organic soils, the index of
compressibility can reach up to 3, although in some cases (clay from the city
from Mexico, for example, can be as high as 10 (Mesri et al., 1975). Ladd
(1971) suggested the range 0.1 Cr/Cc 0.2 while for Leonards (1976)
typical values range from 0.015 Cc 0.035, decrescendo with the decrease
the plasticity index (PI) of the soil.
The pre-consolidation pressure  v'm, defined as the maximum vertical tension
effective that the soil sample has already been subjected in its history, can be
determined by various methods proposed in the literature, among which the method
from Schmertmann (1955), the Tavenas method (1979), the Butterfield method
(1979), the Casagrande method (1936), the Pacheco Silva method (ABNT, 1990),
among others. Only the last two methods are described here, as the
Pacheco Silva method (ABNT, 1990) was standardized in Brazil by MB-3336 of
Brazilian Association of Technical Standards (ABNT).
Knowing the pre-consolidation pressure, it is then possible to determine the ratio
of soil pre-densification OCR 7defined as the'ratio between the vertical tension
'
v
maximum effective vmand the current effective vertical tension
.
a.1) Preconsolidation pressure: Casagrande method (1936)
Regarding Fig. 2.21, the process of determining the pre-pressure
densification follows these steps:
i) mark point A of maximum curvature (or minimum radius) on the curve;
ii) from this point, draw a horizontal line and a tangent to the curve, forming between
they or the angle;
iii) the intersection of the angle bisector with the prolongation of the virgin line
then determine the pre-consolidation pressure v'm .
a.2) Preconsolidation pressure: Pacheco Silva method (1990)
Regarding Fig. 2.22, the process of determining the pre-pressure
thickening follows the following steps:
i) By the initial void index and0draw a horizontal line and determine its point of
intersection with the prolongation of the virgin line;
ii) from this point, draw a vertical line until it crosses with the test curve and, in
next, from this intersection point draw a horizontal line until it meets the
extension of the virgin line, thus determining the pre-pressure
thickening v'm .
7
overconsolidation ratio
C. Romanel
90
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
(log scale)
Figu

ra 2.21 - House Method

great for determining the pre-consolidation pressure v'm

'
 v(log scale)
Figure 2.22 - Pacheco Silva method for determining pre-consolidation pressure v'm
C. Romanel
91
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
b) Coefficient of compressibility av
e
 v'
Previously defined in Fig. 2.2 as
a relationship between tension and deformation because
In reality, it is about
av 
 Vv
Vs
 v'
V
V0
A L

1 e0


1 e0

(2.59)
e
AL0
 v
av 





1 e0

 v'
 v'
 v'
 v'
where A is the cross-sectional area of the sample, L = L - L0the variation of
thickness of the sample at the loading stage being L0and the thicknesses
initial and final, respectively, and0the initial void index. As L < L0o
the final result of Eq. 2.59 is positive.
c) Coefficient of volumetric variation mv
By definition,
V
V0
 v'
AL
AL0
2.60
2.61
 v
 v'
mv 

PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
immediate or elastic) when loaded. Also note that as the
densification occurs, the values of the Poisson's coefficient vu= 0.5 and of the module
of undrained elasticity Eugradually changes to the values of
elastic parameters of the solid skeleton ' eE' (drained condition). Assuming the
behavior of the soil as linear elastic, among the parameters under the conditions
there is the following relationship for drained and undrained
E'
1 '
3
Eu
1 u
Eu
E'
G



(2.65)
1 '

2


2


2

where G is the shear modulus. Lambe and Whitman (1969) report that in
only the reason EuIt is higher than the theoretical elastic value, which is not uncommon.
to find values from 3 to 4 in the case of clays NA.
Note 2 - Table 2.5 lists the existing relationships between the various parameters of
compressibility used for calculating settlements in one-dimensional theory.
Table 2.5 - Relationship between the compressibility parameters for confined compression
(Lambe and Whitman, 1969)
M
mv
av
1 e0
Cc
1 e0
0.435CC


 v'
 v'
M
1
M
M
M
M
mv
av
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
2.8 UNDRRAINED RECOIL AND PRIMARY CONSOLIDATION
As already mentioned, the total recalculation of a layer of soil can be
decomposed into three installments, calculated using different formulas: a) the settlement
e
non-drained
or elastic settlement or immediate settlement, assessed by
formulation of the theory of linear elasticity, which occurs immediately after the
{"loading application":"loading application","primary consolidation settlement":"primary consolidation settlement"}cwhose
evolution over time, in the 1D deformation hypothesis, can be predicted based on
in the average dissipation of excess pore pressure Uvcalculated by the theory of
Terzaghi's densification; c) secondary compression settlement what can
to be significant in turf and highly organic soils.
2.8.1 Undrained Recalculation

e

Settlement parcel
what happens immediately after loading, without
any change in the volume of saturated soil. Usually calculated by
formulation of the theory of linear elasticity considering the parameters Eu
(undrained Young's modulus) and u0.5 of an elastic, isotropic half
homogeneous.
For the 1D state of deformations, the undrained settlement is null because, as already
0
z
mentionedM  and  .
2.8.2 Final settlement recalculation - direct method
Consider a layer of homogeneous saturated soil with initial thickness b.0
subject to an infinite intensity loadWhat causes a repressed feeling?

t 

end of densification
This displacement, considered positive in
soil mechanics when indicating shortening of the thickness of the layer of
only (emphasis), is defined as
  b b
(2.67)
c
0
1
where1represents the value of the final thickness.
Defining now the increase in vertical compressive strain as
b b
(2.68)
(2.69)
0
1
 
v
b
0
obtained from Eq. 2.67
  b
c
0
v
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Considering that in the 1D case of deformations the volume change (contraction)
V0V
V0
V
V0
is expressed as

V
e
1 e
(2.70)
  b  b
b
c
0
v
0
0
V
0
0
with e e0 andthe difference between the initial and final void index values,
respectively.
In terms of the parametermv  v v' the final settlement of primary consolidation
it can be expressed by Eq. 2.69 as
'
2.71
(2.72)
  b  b m
0 v
c
0
v
v
In terms of the parameter M = 1/mvit directly results from Eq. 2.71 that
'
v

b
c
0
M
and in terms of the parameterav mv

1 e0what

a
(2.73)
'
v
v
b

c
0
1 e
 PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
evaluate, for each voltage range, its contribution to the final settlement of
denseness. In other words, in the successive applications of Eq. 2.70, for
for each voltage interval the values of the compressibility parameters are
updated.
Note 2 - Equations 2.70 to 2.76 refer to the calculation of the final settlement of

t 

primary densification
In engineering practice, it is also necessary
to understand the evolution of settlement over time. As in Terzaghi's theory, the
deformations and drainage associated with the dissipation of excess pore pressure
both occur in the vertical direction, so the primary consolidation settlement in
time t is directly proportional to the average percentage of dissipation of
excesses of poropressure Uvat the same time.
 ct Uv

t c
(2.77)
Note 3 - One of the basic premises for calculating the settlement by the formulation
1D means that the deformations of the solid skeleton are infinitesimal. The errors of this
Hypotheses can be significant in the case of landfills over highly...
compressible, as reported by Weber (1969) where the thickness of the
the compressible layer decreased by approximately 80% compared to the thickness
after the densification. Olson and Ladd (1979), based on the results of
finite difference method analyses of layer densification
highly compressible under instantaneous loading, concluded that the
average percentage of dissipation of excess pore pressure (Eqs. 2.36 or
2.37) can still be used in the case of large deformations occurring
since the original drainage thickness H is corrected by:
c
n
H* H

2.78
 
where H* is the corrected drainage thickness and the number of faces of
drainage of the layer (1 or 2). According to Olson and Ladd (1979) this
adaptation results in errors below 2% forc(2H*) < 0.25, less than 5%
forc(2H*) < 0.46 and less than 10% forc(2H*) < 0.65.
2.8.3 Primary consolidation recalculation - indirect method
The final settlement of primary consolidation can be calculated by the method
indirectly subdividing the soil layer into an appropriate number of sub-
layers, as explained in chapter 1, using the parameter values of
1D deformation state: M, av, mv, CCe CR.
2.8.4 Primary consolidation settlement - non-instantaneous loading
An approximate solution (Taylor, 1948) can be applied for forecasting of
recalculation of densification in the case of linearly increasing load with the
time until t = tc(construction time) and then constant (Δq = qc). O
consolidation settlement at time t for a non-instantaneous load
C. Romanel
96
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
you
c
t

can be determined based on the densification retraction for
instant loading Iby
t
c
a) in the construction period (0 < t

tc)
recalque loading = recalque loading x fraction of the load
non-instantaneous in t instantâneo em 0.5t applied in t
q
 qc
 cyou

t

  ci0.5t

(2.79)
(2.80)
b) after the construction period ( t > tc)
recalque carregamento = recalque carregamento
non-instantaneous in t
instantaneous at (t - 0.5tc)
you
c
i
t
t
.5tc
c

 



The instantaneous settlement values determined by Eqs. 2.79 and 2.80 also
they can be estimated graphically, as shown in the construction of Fig. 2.23.
Figure 2.23 - Graphical method for obtaining the settlement curve x time considering
linearly increasing load during the construction period.
The similarity between triangles OAB, OCD, OPt, OQtcallows to write
CD
Pt
Pt
q
 qc

 CD AB
  tyou  0i.5t
2.81
AB Qtc
Qtc
C. Romanel
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
A common fact in engineering practice is the occurrence of loads.
discontinuous, that is, a work (for example, landfill) is built during a certain
time, then interrupted and finally, after some time,
resumption with the continuation of the application of the assumed loading as
linearly increasing until it reaches the final value qcat time tcIn this situation, the
solutions for each charging stage can be obtained in a way
independent, applying the equations and/or graphical method for each
loading separately, and the results summed to obtain the curve
final recalque x time.
2.8.5) Skempton and Bjerrum's Method (1957)
The formulation of the 1D theory for calculating the final settlement of consolidation
primary assumes that the loading area is of large extent for
cause only (or mainly) vertical deformations in the soil layer
saturated and that the excess initial poropressure is equivalent to the value of
applied loading. As most of the loadings in the practice of
engineering has smaller dimensions, often questioning what for
cases, if and when, a 1D formulation can be used. This issue was
studied by several authors, among which are Skempton and Bjerrum (1957) who
they proposed a method that still allows for the repression to occur in the state
1D deformation, on the other hand, incorporates 3D effects into the estimate of excess.
the initial pressure determined based on the known equation of
Skempton
(2.82)
u0 u B3 A

1 3


(2.83)
u0 A1


1 A3


with B = 1 for saturated soil
where A8e B are the Skempton pore pressure parameters for state of
axial tension2 3(Tab. 2.6 and Fig. 2.24).
Table 2.6 – Typical values of parameter Affor some types of soils (Craig, 2004)
Alone
A
sensitive clay
clay NA
slightly pre-consolidated clay
OCR < 4
1
0.5 to 1.0
0 to 0.5
strongly pre-consolidated clay
OCR > 4
- 0.5 to 0
8
The value of parameter A varies with the stress levels and is usually defined at rupture.
C. Romanel
98
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.24 – Typical relationship between OCR and pore pressure parameter at failure (Af) – Craig, 2004.
Considering a loading of a perfectly flexible, circular foundation,
the final consolidation settlement by the Skempton and Bjerrum methodSCB
)
é
calculated by
b0
(2.84)
 CSB
 v' dzas time t → ∞
 m
v
0
or, equivalently at time t = 0,
b0
b0
 CSB
mvu0dz mv

A 1


1

A

 3dz



(2.85)
0
0
observing that Eq. (2.85) introduces soil type influences through
parameter A, of the increases in principal stresses  generated  by
e
loading but the calculation of settlement is still done with the parameter mv(or M,
av, CCthe CRof the 1D deformation state.
1D
C
The final settlement of 1D consolidation by conventional theory
2.71

)
comes from Eq.
b0
 C1D
 v' dzas time t→ ∞
 m
v
(2.86)
0
or, similarly, at time t = 0
b0
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
b0
mv

A 1

1 A

 3dz

(2.89)

 circular

0
b0
mv 1dz

0
If the compressibility parameter mvand the pressure parameter A is
constants with depth (if they are not, the layer can be subdivided
in sublayers with assumed constant average values in each of them
so
circular
A
1 A
(2.90)

   circul ar


where the value ofcircularit is a function only of the dimensions of the loading and the
thickness of the soil layer.
b0
 dz
(2.91)
3
 circular

0
b0
 dz
1
0
Note from the transition from Eq. (2.89) to Eq. (2.90) that the increments of
tension are equal, i.e. in Eq. (2.86) the increase in tension is that caused
by the circular foundation and not by an infinite load.
For A = 1/3 it only behaves like a linear elastic medium under loading
axissymmetric.
For loading in strip (continuous foundation), flat state situation of
deformation with2 0.5
predicted by (Scott, 1963)

1 3

, the initial excess of poropressure can be
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Observe from Fig. 2.25 that when the loaded area of width L is small in
relation to the thickness b0from the soil layer and, especially in the case of clays
pre-consolidated (low values of the Skempton parameter A), the correction of
calculated recalque by the conventional formulation is important to incorporate the
effects of foundation size.
Finally, one should not ignore the contribution of undrained 3D consolidation in
estimation of the total settlement of the foundation,
3D
and
3D
nd
 3D



 CSB



 C1D
(2.95)
Table 2.7 - Values of in the method of Skempton and Bjerrum (1957).
b0/L
Circular Foundation
Running foundation
(see also Fig. 2.25)
uniformly charged
uniformly charged
0
0.25
0.50
1
2
4
1.00
0.67
0.50
0.38
0.30
0.28
0.26
0.25
1.00
0.74
0.54
0.37
0.26
0.20
0.14
0.00
10

Figure 2.25 - Correction factor in the method of Skempton and Bjerrum (adapted from Scott, 1963).
2.8.6) Davis and Poulos Method (1968)
Davis and Poulos (1968) employed the theory of 3D linear elasticity for calculation.
final settlement caused by shallow foundations on clay layers
saturated
C. Romanel
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
 3D  3D  c3D
(2.96)
nd
3D
nd
where 3D, 
and c3Drepresent the total, undrained and of
primary densification of the shallow foundation, respectively.
For a homogeneous soil layer, the final settlement values (drained) and
initial (undrained) can be calculated directly by the elasticity theory
linear based on the formulation described in chapter 1,
2.97
B q
 3D I
E'
(2.98)
B q
Eu
3D
and
I
or indirectly for a layer of initial thicknessb0by
n
2.99
1
E'j
 3D b

 z v'j

 x  y



0j

j1
n
(2.100)
1
Euj
 3D b

 z 0.5

 x  y



and
0j

j1
where B is an appropriate dimension of the foundation (width or diameter), I a factor
of influence (depending on the Poisson's coefficient, geometry, and rigidity of
foundation, the depth of embedding, the thickness of the soil layer
E' e Euthe values of the modulus of elasticity under drained and undrained conditions
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
vertical deformations and the dissipation of excess pore pressure due to
flow in the vertical and horizontal directions.
From the perspective of geotechnical engineering, an interesting procedure is
so directly adapt the results of Terzaghi's 1D theory for prediction
of the pumping speed in 3D problems. Poulos (1999) based on the
results of analyses of the pseudo 3D consolidation of shallow foundations
(Davis and Poulos, 1972) proposes that the consolidation coefficient cvfrom the theory of
Terzaghi should be multiplied by a correction factor Rfthus establishing a
c
equivalent 3D consolidation coefficient andto be used in the formulation
conventional 1D for calculating the percentage of settlement at time t.
(2.102)
cand

Rf cv
Values of Rfbased on the thickness of the soil layer, normalized in
Regarding the size of the foundation (circular or running), values were obtained for three
types of boundary conditions (Fig. 2.26): a) PTPB - permeable top, base
permeable; b) PTIB - permeable top, impermeable base; c) IFPB - top
impermeable (foundation), permeable base.
As the thickness of the layer increases relative to the size of the
foundation, the correction factor Rfalso increases, highlighting an acceleration
of the recirculation speed due to the contribution of the dissipation of excesses
horizontal drainage pore pressure. Poulos (1999) concludes that for
3D loadings the introduction of the equivalent consolidation coefficient in
Terzaghi's formulation leads to satisfactory results for prediction of the
recalculation speed, and this adaptation must be considered by the
engineer when a truly 3D solution (Biot solution) is not available
available or difficult to obtain it.
Figure 2.26 - Equivalent consolidation coefficient for correcting the settlement velocity
predicted by Terzaghi's 1D theory. Cases of perfectly flexible circular foundation (left)
the race (right) - Poulos, 1999.
C. Romanel
103
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
2.8.7) Comparison between 1D and pseudo 3D settlement predictions
From the comparison between settlement predictions obtained by 1D and pseudo theories
3D, based on theoretical studies (Davis and Poulos, 1968) and experimental
(Burland et. al, 1977), the following procedure was recommended by Poulos
(1999, 2000) for an approximate estimate of the 3D settlement made from the
results obtained by the formulation 1D:
i) For pre-consolidated stiff clays (PA)
 t3D  C1D C3D  t3D  3D  C1D  3D
2.103
e
and
nd
ii) For normally consolidated clays (NC)
 C3D  C1D t3D  3D  C3D  3D  C1D
(2.104)
e
nd
and
2.9) SECONDARY COMPRESSION RECALCULATION
Secondary compression is generally interpreted as that which occurs after
the end of primary consolidation (Fig. 2.27). This definition is quite
simplified because it suggests that secondary compression only begins when everything
the soil layer has completed the compaction process. If that were the case,
soil elements along the drainage contours should not experience
no secondary compression effect until the end occurred of
compaction in the soil elements located in the central region of the layer.
Secondary compression is likely the result of the action of different
mechanisms, still not completely understood, among which:
a) only has voids of different sizes, and water can drain from the voids
larger according to the theory of primary densification and be expelled more
slowly from the smaller voids, producing secondary compression effects.
b) in organic soils containing plant remains, water can drain from the voids
of soil according to the primary consolidation theory and being expelled
slowly from the vegetables through the cell walls, producing effects of
secondary compression;
c) in some organic soils, hydraulic conductivity decreases more than
an order of magnitude during the densification. The primary densification is
naturally faster in the initial phase of the process, gradually decreasing
with the reduction of the hydraulic conductivity of the soil and producing apparent effects
of secondary compression;
d) the shear resistance between particles can be of viscous nature,
dependent on the rate of deformation. The greater the movement of the
particles (in the initial phase of primary consolidation) the greater the resistance to
variation of soil volume and the smaller the movement of the particles (in the phase
of secondary compression) lower resistance to volume variation;
C. Romanel
104
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
f) highly non-linear stress-strain curves can produce a
load behavior x settlement similar to that observed in the curves
experiments involving primary consolidation and secondary compression.
Figure 2.27 - Occurrence of secondary compression in the Casagrande method for determination
of the consolidation coefficient.
It is unknown whether, and to what degree of importance, any of the (or others)
the mechanisms above control the effects of secondary compression and that can
vary for different types of soil.
Under such circumstances, although there is a mathematical formulation in the literature for
modeling of secondary compression (the model presented by Gibson and Lo
(1961) is the most well-known) its application in analysis is rare.
recalculation problems. Most practical methods are still based on the
Buisman's proposition (1936) which found in field trials a relationship
linear between the secondary compression and the logarithm of time, with slope
represented through the secondary compression coefficientC
e
log
C  
(2.105)
The characterization of secondary compression through linear curves is
criticized by several authors who claim that the settlement versus time curve
(log scale) tends to gradually become horizontal and that a representation
the secondary compression by line segment is justified only because the
time period for conducting the field and/or laboratory test was not
long enough. On the other hand, long-term laboratory trials
they are also difficult to execute and interpret, as questions arise
C. Romanel
105
PUC-Rio, Department of Civil Engineering
ENG1211 – Mecânica dos Solos – Prof. Celso Romanel
about the influence of organic matter growth due to the presence of
oxygen, the oxidation of iron minerals present in the sample, the influence of
small tremors in the laboratory environment, etc.
Mesri et al. (1994) suggest thatC it relates to the index of
compressibility Ccaccording to Tab. 2.8.
Table 2.8 - Values ofC CCfor solos10 (Mesri et al., 1994).
Alone
C CC
Clay and inorganic silt
Clay and organic silt
Peat
0.04 ± 0.01
0.05 ± 0.01
0.06 ± 0.01
From a practical point of view, it would still be necessary to know from what time t* the
Recalculation due to secondary compression could be added to the value of
primary consolidation settlement. The most logical, simple, and used way
(although arbitrary) seems to be when the slope of the settlement curve of
settlement versus time (log scale) is equal to the slope of the settlement curve
secondary compression versus time (log scale).
Considering Eq. 2.70, then the secondary compression settlementscan
so to be obtained as
C
0 1 e*
e
t
b
b

og

s
01 e*

t*
(2;106)
where is* it
is the voids index at the moment accepted as the beginning of compression
secondary and it is the time when one wishes to assess the settlement portions.
In general, it can be said that among all the aspects related to the
estimate of settlements in geotechnical engineering, the theme of compression
secondary is the least developed, both in the sense of understanding its
mechanisms and the influence and importance of each of its possible agents,
how in the systematic of incorporating research results into practices of
engineering. In the absence of a fully accepted approach, the method of
Buisman (1936) is still the most used to provide estimates of the share of
settlement caused by secondary compression of saturated low soils
permeability.
2.10) VERTICAL DRAINS
Most clay deposits are anisotropic, formed by deposition of
horizontal layers (or nearly horizontal) with permeability coefficient in
horizontal direction khsuperior to the value of the permeability coefficient in
10
In the case of PA soils, the values in Tab. 2.8 refer to the relationshipC Cr, indicating that the
secondary compression in pre-consolidated soils is less significant than for NA soils.
C. Romanel
106
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
vertical direction kvLogo, if drainage with flow in the horizontal direction occurs,
so the average percentage of dissipation of excess pore pressure, the
evolution over time of primary consolidation settlement, the increase of
effective stresses and the increase in soil shear resistance
would occur at speeds higher than the predicted values in
1D densification.
A technique to accelerate the dissipation of excess pore pressure consists of
installation of vertical drains, with the execution of holes in the clay layer and
backfill using well-graded sand that allows for flow
efficient in water (drain function) but also prevents the migration of
fine soil particles through their voids (filtering function). Holes are
executed at depths of up to 30m, with typical diameters ranging between
200 to 400mm.
Currently, prefabricated drains (Figs 2.28 and 2.29) are used, inserted
in previously opened holes or inside driven casing tubes
through the soil mass up to a depth of 65m. For the determination of the radius
equivalent rdin the case of prefabricated drains in rectangular shape axb, several
suggestions exist in the literature, but for practical applications the following
the expression is recommended (Atkinson and Eldred, 1981; Jansen and den Hoedt, 1983;
Rixner et al. 1986) :

(2.107)

Drenos são normalmente instalados em configurações quadrada ou triangular

Figure 2.30 and the definition of the spacing between them, usually between 1 to 4m, is
the main objective of the project. The spacing should naturally be smaller than
that the thickness of the layer, not being advantageous, therefore, the use of
drains in relatively thin layers.
The drainage radius R is determined by the equivalence of areas between a circle.
it is a square (square configuration) or between a circle and a hexagon
(triangular configuration).
For square configuration,
 R2 S2
(2.108)

R S  0.564S
where S is the distance between the vertical drains.
For the triangular configuration
S2
3
4
 R2 6

R
S 0.525S
(2.109)
4 3
2
The ratio between the consolidation coefficients in the horizontal direction chand in
vertical direction cvis usually located in the interval1 ch cv 2, being the drain
C. Romanel
107
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
more efficient the greater this relationship is. Both coefficients of
thickening in the area of soil immediately adjacent to the perimeter of the hole can
be significantly reduced due to the flattening caused by the
installation of the drain, a phenomenon known as smear effects. In the formulation of
radial densification, the effect can be incorporated by decreasing the value of
horizontal consolidation coefficient chor reducing the value of the radius of the
drain. Another complication that could occur is the tendency for sand drains.
they will act as columns, reducing the increase in vertical stress in the layer of
clay, which can result in low pore pressure values not completely
dissipated and primary consolidation repressions not fully occurred.
This effect is minimal in the case of prefabricated drains due to the large
flexibility of geosynthetics (Fig. 2.31).
Figure 2.28 - Vertical drains reduce the drainage length and accelerate the dissipation of
excesses of poropressure generated by loading.
Figure 2.29 - Prefabricated vertical drains made of polypropylene, with channels for drainage.
water on both sides, surrounded by a high permeability geosynthetic filter,
flexibility and resilience.
C. Romanel
108
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.30 - Configuration patterns (square and triangular) and influence radius R of the drains
separated by a distance S.
Figure 2.31 - Deformation of prefabricated drains during consolidation.
Observations:
Vertical drains can be inefficient in pre-consolidated clays if the
an increase in effective vertical stress is not sufficient to exceed the value of
preconsolidation pressure. According to Bjerrum (1972), the use of
vertical drains in clays PA are recommended when

0.6
(2.110)

ii) The amalgamation of pre-densified clays due to the installation process

two drains can increase the values of settlement by consolidation.
C. Romanel
109
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
iii) Vertical drains cannot control the occurrence of compression settlement.
secondary clay layer.
iv) Part or all of the water collected by the drainage system will flow into the carpet of
drainage on the surface (Fig. 2.28). As the permeability of sand is
considerably higher than that of clay, it is usually considered in
project that there will be no development of hydraulic resistance in the carpet,
specifying a sufficiently thick layer (more than 0.5m) of
clean sand (percentage of fines less than 5%).
2.10.1) Primary densification with radial drainage
The derivation of the governing differential equation of 1D consolidation in the direction
radial was obtained by Rendulic (1935), considering the infinitesimal element of
only from Fig. 2.32 with radial permeability coefficient krsubject to radial flow
in the positive sense of r.
The infinitesimal inflow rate, normal to the cross section. given
for
dQent  krbird dz
(2.111)
and the normal output flow to the cross section (r+dr)d dz
(2.112)
i
r



dQknow k i doctor

r dr d dz


r


The infinitesimal flow dQ in the element, defined as the difference between the flows
of entranceentand of saidsai, disregarding the second infinitesimals
order, results in
i
r
I
r






dQ dQent dQhe/she/it is k ird dz k i dr

r  dr

d dz k i r drd dz



(2.113)
r
r


r
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel



i


the rd

drdz


h
2

h
and








k i

r drd
 dz


kr

r



(2.115)
r

r2
r
t
1
e
r
t
1
e


Figure 2.32 - Representation of infinitesimal soil element in cylindrical coordinates.
However


he

ue

PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Barron (1948) obtained the solution to Eq. 2.120 for both cases, but here it will be
considering only the situation for equal vertical deformations (equal strain
case), commonly applied in engineering projects. The difference between both
the solutions will probably be less than the committed in the determination
experimental the coefficient of radial consolidation which, in a way, justifies
the use of a less complex mathematical solution developed
for the case of equal vertical deformations. Barron's solution ([Link].)
consider that the influence zone of the drain is circular and axisymmetric, that the
the permeability of the drain is infinite compared to that of the soil and that the law
of Darcy is valid.
The excess of poropression uandat time t in the radial distancerd r Rit is expressed
by the following equation, considering n = R/rdthe normalized drainage radius
regarding the drain radiusd.
2







 
 


c ht
R2
q
n2Fn
r
1
r
2

2
(2.121)
 
 
 




uand(r,t)
n

1 exp


 

rd
2r
Fn


PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
only hollowed out, saturated, with external radius rsthe permeability coefficient ks,
followed by another cylindrical shell of intact homogeneous soil with external radius R
the radial permeability coefficient kr.
For the case of the theory of equal vertical deformations, the following solution was
presented by Moran et al. (1958) for the average percentage of dissipation of
excesses of poropressure Ur


2
Ur(Tr)  1 exp Tr
(2.126)
(2.127)





2Tr  n1Ur

0
where
n2
n
s2
n2 s2
kr
ks
 

n
 0.75

n

s

withs  rs rd
 
(2.128)
(2.129)
n2 s2
4n2
n2
s
 
or, approximately (Hansbo, 1981),
n
kr
 
 n
 0.75 ns

withs rsrd
 
s
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
suggest usingkskr kv / krthat, in the case of Brazilian soft clays, corresponds to
values between 0.5 and 0.66 (Almeida, 1996).
Several researchers (Hansbo, 1997; Jamiolkowski et al., 1983; Bergardo et al.,
1993; Holtz and Holm, 1973; Sathananthan and Indraratna, 2006 indicate that the radius
The area of the amolgado soil varies from 2 to 3 times the equivalent radius of the mandrel.
used for the installation of the drain.
2.10.3) Time-dependent loading
Vertical drains are often used in cases where it is important
ensure that a significant portion of the densification occurs during or
immediately after the construction period. So, the construction time
represents an important variable of the project, which may lead to the hypothesis of
instant loading can lead to significant forecasting errors in speed
of repression.
Following the same methodology used in section 2.5 for vertical drainage,
Olson (1977) obtained the following equations for the average percentage of
dissipation of excess pore pressure in the radial direction, considering the
application of a linearly increasing load (Fig. 2.13), without effect
smear
a) for t≤tc
2.130
1
1




Ur(Tr) 
T

1 exp

 ATr




r
Trc
A
b) for t > tc
1
Ur(Tr )  1

exp

ATrc

1

exp

 ATr

(2.131)
ATrc
2F
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
(2.133)

1U



1Uv1Ur

where U is the average percentage of excess pore pressure dissipation
considering flow in vertical and radial directions, Uvthe average percentage of
dissipation of excess pore pressure due to flow in the vertical direction (Eqs.
2.36, 2.37) and Urthe average percentage of dissipation of excesses
poropressure due to the flow in the radial direction.
Figure 2.34 - Variation of Urvs. Tr for linearly increasing loading, vertical drains with
s = 1, for n = 5, 10 (Olson, 1977).
2.10.4) Primary 1D consolidation with vertical and radial drainage
Considering the combined drainage by simultaneous flow in the vertical directions
In radial, the governing differential equation of primary consolidation can be
expressed by
2ue
z2
1ue2u
rr




the


ue
t
 cv
c

(2.132)
h
r2

Carrillo (1942) obtained the solution of Eq. 2.132 as
C. Romanel
115
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
(2.133)

1U



1Uv1Ur

where U is the average percentage of dissipation of excess pore pressure
considering flow in the vertical and radial directions, Uvthe average percentage of
dissipation of excess pore pressure due to flow in the vertical direction (Eqs.
2.36, 2.37) and Urthe average percentage of dissipation of the excesses of
poropressure due to the flow in the radial direction.
2.11) PRELOADING WITH FILL
Preloading consists of applying a temporary overloadqs in
a layer of soil before the construction of the permanent structure. The intensity
The preloading should be sufficient to eliminate a large part of the settlement.
end of primary consolidation. In pre-consolidated soils, for the technique to be
efficient, it is necessary to exceed the value of the pre-consolidation pressure.
Removed the temporary loading, with the soil in the PA condition, it is possible to
estimate that any potential secondary compression settlements should be
lower than those that would occur with the soil layer in the NA condition.
The main advantages of preloading can be cited:
a) for its construction, only conventional movement equipment of
land is needed;
there is a long history of success stories with the use of this technique;
c) can present a very favorable cost-benefit relationship in the case of
works that generate small loads (roadways, flooring of establishments)
commercial establishments such as supermarkets, etc.) as long as the fill material is
easily available.
The disadvantages mentioned are:
a) landfills must extend horizontally beyond the planned perimeter of
construction, which may not be possible in confined spaces;
b) may require the transportation of large amounts of soil;
c) the duration of the overload may delay the schedule of the
construction. Pre-loadings are generally designed to last from 3 to 9
months and, in the case of deposits of great thickness, this time can
be reduced with the simultaneous installation of vertical drains.
The principle of the preloading technique is illustrated in Fig. 2.35, where the
temporary loading(qs that exceeds the construction load
permanent
) is applied for a certain time [Link] double
q f
drainage, it is observed that when the temporary load is removed, the
the central region of the clay layer will have compacted less than the regions
close to the contours. If the average percentage of dissipation of the excesses of
poropression Uvused as a criterion for removal ofqs so the region
the central layer will continue to densify after the removal of the load
temporary while the edge regions will expand. Although these
C. Romanel
116
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
effects are likely to offset each other, the retraction will exceed the expansion given that the
compressibility index Ccis higher than the recompression index CrJohnson
(1970) recommends using the percentage of excess dissipation
poropressure in the center of the U layerz=1as a removal criterion of
temporary loading and not the average value Uvfor the entire thickness of
layer.
Admitting the existence of a NA at the center of a layer of clay.
unit thickness sublayer (b0= 1), with coincident medium plans in Z
= 1.
a) Final settlement recalculation of the sublayer with unit thickness due to
to the temporary and permanent loadsqs+
q f
 v' 0  qs   qf
1
 cs f  b0 v 1
CClog
(2.134)
1 e0
b) Final settlement recalculation of the sublayer of unit thickness due to
 v' 0
permanent loading
q f
 v' 0   qf
1
 cf  b0 v 1
CClog
(2.135)
1 e0
 v' 0
'
v0
where it is the initial effective vertical stress at the center of the layer (Z = 1).
At time tsthe recalibration of the loading due to theqs +
equal or repress
q f
final due to loading
(Fig. 2.35), it is possible to define the percentage
q f
of the dissipation of excess pore pressure in the center of the thickness layer
unitary by
 cs f

t ts

  cf

Us f  cs f  cf
(2.136)
Z1

 qf
 qf

PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
value of Tv corresponding to the center of the layer with the help of the isochrones of
Fig. 2.4 and calculate the time tsconsidering the usual definition of the time factor.
the value of time tsit is known and you are asked to calculate the value of the surchargeqs.
In this case, calculate the value of the time factor [Link] then the percentage of
dissipation of excess pore pressure at Z = 1, through Eq. 2.27 or
isocronous of Fig. 2.4. Next, solve Eq. 2.137 for
(therefore
qs qf
qs) considering v' 0the effective initial vertical stress at the mid-plane of the layer.
q
qs
qf
ts

Figure 2.35 - Preloading scheme to accelerate consolidation settlement.
2.12 ADMISSIBLE WITHHOLDINGS
The loading of soil masses with a structure always produces
displacements. If the movements are significant, the structure stops
function properly, reaching a limit state that can provoke
from damage to its appearance (cracks, tilting walls, ...), going through the
loss of its usefulness (stuck doors, misalignments of equipment,
pipe rupture due to differential settlements, ...) which can even reach
the structural collapse (foundations tilting leading to pillar ruptures,
ground foundation rupture, loss of support due to landslides, ...
The calculation of structural displacements is a difficult and complicated task, and the
Experience shows that forecasts are generally unsatisfactory. Some
the reasons for the difficulty of this task are as follows: most of the structures
it is too complicated to be modeled in detail, the properties of
the materials involved are known only approximately, analyses are
made as if the structures were built exactly as designed,
long-term effects are generally ignored, etc.
Building movements can be classified as: (a) body settlement
rigid; (b) rotation of a rigid body; (c) distortion.
C. Romanel
118
PUC-Rio, Department of Civil Engineering
ENG1211 – Mecânica dos Solos – Prof. Celso Romanel
Rigid body backlash is not usually a problem, although it can be
to cause damage to piping and external infrastructure connected to the building
walkways, for example). Rigid block rotation can be a problem for
tall buildings, which becomes noticeable when the slope exceeds 1/250.
The distortion implies the deformation of a structure and this fact causes the
occurrence of cracks and other damages to buildings. Bjerrum (1963) presented
criteria for maximum angular distortion limitsdepending on the type of damage,
as shown in Table 2.9. In this investigation some structures
they apparently experienced greater distortions without any damage occurring,
but none of them were damaged under values lower thanindicatedIt must be
highlighted that these data apply to traditional structures (prior to
1960s) and that for modern, more flexible structures, the values
The listings in Tab. 2.9 can be considered conservative.
In practice, the limit values of distortion are not used by the engineer.
geotechnical because the projects usually predict zero distortion, but
Distortions can occur due to a lack of uniformity in soil conditions.
of foundation.
Correlations were then obtained between the distortion and the maximum total repression max,
expressed in inches, according to the following equations:
For isolated foundations,
1
in clays
 max

 max
(2.138)
(2.139)
1200
1
in sands
 max

 max
600
As an example, considering max 1,300 in both equations result in the
values max 4"  100mmfor clays and max 2"  50mmfor sands.
C. Romanel
119
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Table 2.9 - Maximum angular distortion for various types of building damage12(Bjerrum,
1963).
12
H/L in Table 2.9 refers to the height/length ratio of a masonry wall.
C. Romanel
120
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
BIBLIOGRAPHIC REFERENCES
ABNT - Brazilian Association of Technical Standards (1990) - Consolidation test
unidimensional, MB 3336, Rio de Janeiro, 15p.
Almeida, M.S. (1996) - Landfills on soft soils: from design to performance evaluation,
UFRJ publisher, 215 pages.
Atkinson, M.S. and Eldred, P.J.L. (1981) – Consolidation of soils using vertical drains, Géotechnique,
v. 31, n. 1, pp. 33-43.
Barron, R. A. (1944) - The Influence of Drain Wells on the Consolidation of Fine-Grained Soils,
U.S. Army Corps of Engineers, District of Providence.
Barron, R. A. (1948) - Consolidation of Fine-Grained Soils by Drain Wells,Trans., ASCE, vol. 113,
pp. 718-754.
Bergado, D.T., Mukherjee, K; Alfaro, M.C. and Balasubramaniam, A.S. (1993) – Prediction of vertical-
band-drain performance by the finite element method, Geotextiles and Geomembranes, v.12,
n.6, pp. 567-586.
Biot, M.A. (1935) – The problem of the consolidation of clay materials under a load.
Annals of the Scientific Society of Brussels, Series B, vol. 55, pp. 110-113.
Biot, M.A. (1941) – General theory of three-dimensional consolidation, Journal of Applied Physics,
vol. 12, pp. 155-164.
Biot, M.A. (1955) – Theory of elasticity and consolidation for a porous anisotropic solid, Journal of
Applied Physics, vol. 26, pp. 182-185.
Biot, M.A. (1956a) – General solutions of the equations of elasticity and consolidation for a porous
material, Transactions, Journal of Applied Mechanics, ASME, vol. 78, pp. 91-96.
Bjerrum, L. (1963) - Allowable settlement of structures. Proc. European Conf. on Soil Mech. and
Found. Eng., Wiesbaden, Germany, v. 3, 135-137.
Bjerrum, L. (1972) – Embankments on soft ground. Specialty Conference on Performance of Earth
and Earth Supported Structures, v. 2, pp. 1-54.
Booker, J.R. (1974) - The Consolidation of a Finite Layer Subject to Surface Loading. lnt. J. Solids
and Structures., vol. n10, pp. 1053-1065.
Bolton, M.D. (1979) - A Guide to Soil Mechanics,Palgrave Macmillan, 456 p.
Burland, J.B., Broms, B.B. and de Mello, V.F.B. (1977), Behaviour of Foundations and Structures
Proc. 9 ICSMFE, Tokyo, vol. 2, pp. 495-546.
Buisman A.S. (1936) - Results of long duration settlement tests, Proc. 1st ICSMFE, Cambridge
Mass., vol.1, pp. 103-106.
Butterfield, R. (1979) – A Natural Compression Law for Soils (an advance over e – ln p ’),
Geotechnique 29 (4), pp. 469-480.
Carrillo, N. (1942) - Simple Two- and Three-Dimensional Cases in the Theory of Consolidation of
Soils, Journal of Mathematics and Physics, vol. 21, n. 1, pp. 1-5
Casagrande, A. (1936) - The determination of preconsolidation load and its practical significance,
Proc. 1st Int. Conf. Soil Mech. and Found. Eng., Harvard, vol. 3, pp. 60—64.
C. Romanel
121
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Casagrande, A. e Fadum, R.E. (1940) - Notes on soil testing for engineering purposes, Harvard
University Graduate School Publications, 268, 74 pp.
Cour, F.F. (1971) - Inflection Point Method for Computing cv, Journal of the Soil Mechanics and
Foundation Engineering Division, ASCE, 97 (5), pp. 827-831.
Cray, R.F. (2004) – Craig’s Soil Mechanics, seventh edition, SPON Press, 447 p.
Davis, E.H. and Poulos, H.G. (1968) - The use of elastic theory for settlement prediction under three-
dimensional conditions, Geotechnique 18(1), pp. 67-91.
Davis, E.H. and Poulos, H.G. (1972) - Rate of Settlement Under Three Dimensional Conditions,
Geotechnique, 22(1), pp. 95-114.
Duncan, J.M. (1993) – Limitations of conventional analysis of consolidation settlement, Journal of
Geotechnical Engineering, ASCE, vol. 119, no. 9, pp. 1333-1359.
Feng, T.W.e Lee, Y.J. (2001) - Coefficient of Consolidation from the Linear Segment of the t1/2
Curve, Canadian Geotechnical Journal, 38, pp. 901-909.
Fox, E.N. (1948) - The mathematical solution for the early stages of consolidation. Proc. 2nd Int.
Conf. Soil Mech. Found. Engng., Rotterdam, vol.1, pp. 41-42.
Gibson, R.E. and Lumb, P. (1953) - Numerical solution of some problems in the consolidation of clay,
Proc. Institution of Civil Engineers, 2, part 1, 182-198.
Gibson, R.E. and Lo, K. Y. (1961) - A theory of consolidation for soils exhibiting secondary
consolidation, Acta Polytech. Scand., 296 p.
Gray, H. (1945) - Simultaneous Consolidation of Contiguous Layers of Unlike Compressible Soils,
TransactionsASCE, vol. 110, pp. 1327-1344.
Hansbo, S. (1981) - Consolidation of fine grained soils by prefabricated drains, 10th International
Conference on Soil Mechanics and Foundation Engineering, Stockholm, vol.3, pp. 677-682.
Hansbo, S. (1986) – Preconsolidation of soft compressible subsoil by the use of prefabricated
vertical drains, Annales des Travaux Publics de Belgique, v.6, pp. 553-563.
Hansbo, S. (1987) – Design aspects of vertical drains and lime column installations, 9thSoutheast
Asian Geotechnology Conference, Bangkok, pp. 1-12.
Hansbo, S. (1997) – Aspects of vertical drain design: Darcian or non-Darcian flow, Géotechnique,
v.47, n.5, pp. 983-992.
Hird, C.C. and Moseley, V.J. (2000) – Model study of seepage in smear zones around vertical drains
in layered soil. Geotechnique 50, no. 1, pp. 89-97.
Holtz, R.D. and Holm, B.D. (1973) – Excavation and sampling around some sand drains at Ska-
Eddeby, Sweden, Sixth Scandinavian Geotechnical Meeting, Trondhein, Norwegian Technical
Institute, pp. 75-79.
Jamiolkowski, M., Lancelotta, R. and Wolski, W. (1983) - Precompression and speeding up
consolidation, 8th European Conference on Soil Mechanics and Foundation Engineering, v. 3
pp. 1201 – 1226.
Jansen, H.L. and den Hoedt, G. (1983) – Vertical drains: in situ and laboratory performance and
design consideration in fine soils, 8th European Conference on Soil Mechanics and Foundation
Engineering, v.2, pp. 647-651.
C. Romanel
122
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Johnson, S.J. (1970) - Precompression for improving foundation soils, J. Soil Mech. and Found.
Eng., Amer. Soc. of Civ. Engrs. vol. 96, SM1, pp. 110-144.
Jumikis, A.R. (1962) – Soil Mechanics, Van Nostrand Company, 791 p.
Ladd, C.C. (1971) – Settlement analysis of cohesive soils, Soil Publication 272, Massachussetts
Institute of Technology, Department of Civil Engineering, Cambridge, Mass., 92 pp.
Lambe, T.W. and Whitman, R.V. (1969) – Soil Mechanics, John Wiley & Sons, 553 p.
Leonards, G. A. (1976) - Estimating consolidation settlements of shallow foundations on
overconsolidated clays, Special Report 163, Transportation Research Board, pp. 13-16.
Mesri, G. and Rokhsar, A. (1974) - Theory of consolidation for clays. Journal of Geotechnical Division
ASCE, vol. 100, no. GT8, 889-904.
Mesri. G; Rokhsar, A. and Bohor, B.F. (1975) - Composition and compressibility of typical samples of
Mexico City clay, Géotechnique 25, pp. 527-554.
Mesri, G., Lo, D.O.K. and Feng, T-W. (1994), Settlement of embankments on soft clays. Geot. Spec.
Pub. 40, ASCE. 1, pp. 8-56.
Mesri, G., Feng, T.W. e Shahien, M. (1999) - Coefficient of Consolidation by the Inflection Point
Method, Journal of Geotechnical and Geoenvironmental Engineering, ASCE, 125 (3), pp. 716-
718.
Moran, Proctor, Meuser and Rutledge (1958) - Study of Deep Soil Stabilization by Vertical Sand
Drains, Publ. PB151692, U.S. Dept. of Commerce.
Newmark, N.M. (1942) – Influence charts for computation of stresses in elastic foundations,
University of Illinois Bulletin No. 338.
Olson, R. E. (1977) - Consolidation under time dependent loading, Journal of Geotechnical
Engineering, ASCE, vol. 103, no. 1, pp. 55-60.
Olson, R.E.; Ladd, C.C. (1979) – One-Dimensional Consolidation Problems, Journal of the
Geotechnical Engineering Division, ASCE, vol. 105, GT1, pp. 11-30.
Olson, R.E. (1989) - Advanced Soil Mechanics (class notes), Chaoyang University of Technology,
Department of Construction Engineering.
Parkin, A.K. (1978) - Coefficient of Consolidation by the Velocity Method, Géotechnique, 28 (4), pp.
472-474.
Parkin, A.K. (1981) - Consolidation analysis by the velocity method, 10th International Conference
on Soil Mechanics and Foundation Engineering, Stockholm, vol. 1, pp. 723–726.
Poulos, H.G. and Davis, E.H. (1974) – Elastic Solutions for Soil and Rock Mechanics, John Wiley &
Sons, 411 p.
Poulos, H.G. (1999) – Common procedures for foundation settlement analysis – are they
adequate?, 8thAustralia New Zealand Conference on Geomechanics, pp. 1-25.
Poulos, H.G. (2000) - Foundation Settlement Analysis - Practice versus research, The Eighth
Spencer Buchanan Lecture, College Station Hilton, Texas.
Rendulic, L. (1935) - Der hydrodynamische spannungsausgleich in zentral entwasserten
cylinders, water management, 2, pp. 250-253.
C. Romanel
123
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Rendulic, L. (1936) - Porenziffer und Porenwasserdruck in Tonen, Der Bauingenieur,vol.17, n º
51/53, pp 559-564.
Rice, J.R. and Cleary, M.P. (1976) - Some basic stress-diffusion solutions for fluid saturated elastic
porous media with compressible constituents, Rev. Geophys. Space Phys., 14, pp. 227-241.
Rixner, J.J., Kraemer, S.R. and Smith A.D. (1986) - Prefabricated vertical drains, Federal Highway
Administration, Washington DC, Report noHWA-RD-86/169.
Robinson, R.G. and Allam, M.M. (1996) - Determination of Coefficient of Consolidation from Early
Stage of Log t Plot. ASTM Geotechnical Testing Journal, 19 (3), pp. 316-320.
Robinson, R.G. (1999) - Consolidation analysis with pore pressure measurements, Géotechnique
49(1), pp. 127-132.
Sathananthan, I. and Indraratna, B. (2006) - Laboratory evaluation of smear zone and correlation
between permeability and moisture content, Journal of Geotechnical and Geoenvironmental
Engineering, ASCE, vol. 132, no. 7, pp. 942-945.
Schmertmann, J.H. (1955) – The undisturbed consolidation of clay, Transactions, ASCE, v. 120,
pp. 1201 – 1233.
Schiffman, R. L. and Stein J. R. (1970) - One-Dimensional Consolidation of Layered Systems,
Journal of Soil Mechanics and Foundations Division, ASCE, vol. 96, n SM4, pp. 1499-1504.
Schmertmann, J.H. (1955) – The Undisturbed Consolidation of Clay, Transactions ASCE, vol. 120,
pp. 1201 – 1233.
Scott, R.F. (1961) - New Method of Consolidation-Coefficient Evaluation, Journal of the Soil
Mechanics and Foundation Engineering Division, ASCE, 87, SM1, pp. 29-39.
Scott, R.F. (1963) – Principles of Soil Mechanics, Addison-Wesley Publishing Company, 550 p.
Singh, S.K. (2007) - Diagnostic Curve Methods for Consolidation Coefficient, International Journal
of Geomechanics, ASCE, 7 (1), pp. 75-79.
Skempton, A.W. and Bjerrum, L. (1957), A contribution to the Settlement Analysis of Foundations on
Clay, Geotechnics, 7(3), pp. 168-178.
Sridharan, A. and Rao, A.S. (1981) - Rectangular Hyperbola Method for One-Dimensional
Consolidation, ASTM Geotechnical Testing Journal, 4 (4), pp. 161-168.
Sridharan, A., Murthy, N.S. e Prakash, K. (1987) - Rectangular Hyperbola Method of Consolidation
Analysis, Geotechnical, 37 (3), pp. 355-368.
Sridharan, A., Prakash, K., e Asha, S.R. (1995) - Consolidation Behaviour of Soils. Geotechnical
Testing Journal, 18 (1): 56–68.
Sridharan, A. and Prakash, K. (1997) - The logδ– logt Method for the Determination of the
Coefficient of Consolidation. Proceedings of the Institution of Civil Engineers, Geotechnical
Engineering 125, pp. 27–32.
Tavenas, F.; Des Rosiers, J.P.; Leroueil, S; La Rochelle, P; Roy, M. (1979) – The use of Strain
Energy as a Yield and Creep Criterion for Lightly Overconsolidated Clays, Géotechnique 29
(3), pp. 285-303.
Terzaghi, K. (1923) - On the Calculation of the Permeability Coefficient of Clay from the Flow of
Hydrodynamic
tension phenomena,
Session reports,
Academy
the
Sciences, Vienna, part I, 132 (3/4), pp. 125—138.
C. Romanel
124
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Terzaghi, K. (1925) – Erdbaumechanik auf Bodenphysikalischer Grundlage, Leipzig, Alemanha.
Taylor, D.W. (1948) – Fundamentals of Soil Mechanics, John Wiley & Sons, New York, 700p.
Terzaghi, K. and Fröhlich, O.K. (1936) - Theory of the settlement of clay layers, Vienna, Austria.
University of Sydney (1997) - CIVL 3401 Soil Mechanics A, Senior Year Examination
U.S. Navy (1962) - Design Manual - Soil Mechanics, Foundations and Earth Structures,
NAVDOCKS DM-7.
Verruijt, A. (1969) - Elastic storage of aquifers, in Flow Through Porous Media, Academic Press,
New York.
Weber, W.G. Jr (1969) – Performance of Embankments Constructed Over Peat, Journal of the Soil
Mechanics and Foundation Division, ASCE, vol. 75, pp. 53-76.
C. Romanel
125
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
SOLVED EXERCISES
1) A densification test with double drainage was performed with a sample of
clay with an initial thickness of 2 cm. The values listed below refer to the
results measured in 2 charging stages. It is requested to determine the
values of the excess of poropressure at the center of the sample and of the thickness of the
sample when the extensometer reading is 0.696 cm.
Loading
Reading of the extensometer
Void Index
(kPa)
(cm)
e
0
---
---
200
400
0.769
0.665
0.815
0.710
Average percentage of excess pore pressure dissipation Uvwhen
the reading of the extensometer is equal to 0.696cm
 ct 0.769 0.696
 c0.769 0.665
Uv

 0.7019
Calculation of the time factor Tvfor
Uv0.60
Tv 0.933 og(1Uv)  0.085 0.933 og(1 0.7019 0.085 0.4055
In the center of the clay sample (Z = 1) for Tv= 0.4055 is determined
approximately of the isochrones of Fig. 2.4 the value Uz0.53
ue
Uz 1

ue

1Uz

u0 

10.53 200 94kPa

u0
The initial void index is0and the initial reading of the extensometerℓ0they can be
estimated from Eq. 2.70 of the final settlement calculation.
the loading stage ∆q = 200 kPa
e00.815
(A1)
e
1 e0
 c  b0
 00.769 2
1 e0
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
A granular fill of 6m width and great length was built
about a layer of clay 3m thick overlying a rock
waterproof. The water level is located 0.5m below the ground surface. If
the landfill applies a load of 75 kPa on the surface of the ground, and the clay
is saturated above the water level, it is requested to calculate, considering the
specific weight of waterw= 10kN/m3:
a) the final settlement of primary consolidation of the clay layer by the 1D theory
considering the following properties:
Void index in the center of the layer
Grain density Gs
0.8
2.6
Preconsolidation Ratio OCR
Compressibility index Cc
Compression index Cr
4
0.21
0.035
Saturated specific weight of clay
Gs If
1 e
2.6 10.8
 sat
w 
10 18.9 kN / m3
1 0.8
Initial effective vertical stress at the center of the clay layer
 v' 0  sat1.5  w1.0 18.35 kPa
Pre-consolidation pressure
 v' m
OCR
 4  v' m 418.35 73.4 kPa
 v' 0
Final effective vertical tension after loading
Considering Eq. 1.34 or graphs from Fig. 1.9 comes

with = 2.214
(1.34)

∆ 71.96
 v'   v' 0  v'  18.3571.96 90.31 kPa
As'v > 'vmso the recalculation should be calculated as the sum of the
following installments:
C. Romanel
127
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
'
'










b0

v


vm





 c
c 
C log
 CClog
r
'
'

1 e0
 v0
 vm




3
73.4
90.31










0.035 log
 0.21 log
 0.067m


PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
It is intended to build a building in a location where an underground duct passes.
containing highly sensitive cables. The project provides for a rectangular foundation
(Fig. 2.36) that transmits a uniform load of 220 kPa on the surface of the
alone. A floor plan of the building and the position of the cable can be seen in the figure.
The soil consists of 2m of sand over a clay deposit 5m thick.
underlying the sound rock. The duct is buried in the middle of the clay layer,
as shown in the vertical cut of Fig. 2.37. The water level coincides with the top
from the clay layer. The sand is dry withd= 15 kN/m3and the clay hassat=
18 kN/m3. For both types of soil Gs 2.6 and the pre-consolidation ratio
OCR = 3 across the entire clay layer. Sand can be assumed to be incompressible.
and for the clay consider Cr= 0.1, Cc= 0.5.
It is requested to determine, considering the specific weight of waterw= 10kN/m3:
a) the final settlement of primary consolidation at points A, B, and C of the duct through
from the 1D theory.
15m
Pipeline line
buried
4m
4m
C
B
A
4m
7.5m
Figure 2.36 - Plan of the rectangular shallow foundation
220 kPa
Sand
2m
NA
Clay
Duto
2.5m
2.5m
Clay
Rock
Figure 2.37 - Soil profile cross-section
C. Romanel
129
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Initial voids ratio in the clay layer
Gs If
1 e
2.6 1 e
 sat
 w 18
10

e 1
1 e
Note that the duct's subsidence will be produced by vertical deformations of the
clay sublayer between depths 4.5m < z < 7m
Initial effective vertical stress at the average depth z = 5.75m of this sublayer,
 v' 0 15 2 8 3.75 60 kPa
 v' m  v' 0 OCR 60 3180 kPa
i) Increase of vertical tension at point A (Fig. 2.38)
Pipeline line
buried
4m
4m
A
7.5m
Figure 2.38 - Location of point A in the corner of 4 rectangular subareas (7.5m x 4m).
4
7.5
(m e n are interchangeable)
mz 4m
 0.7
nz 7.5n
 1.3
5.75
5.75
From the Fadum graph (Fig. 2.39)13In Table 2.10, f(m,n) = 0.16
 v'  4 q f(m,n) 4 220 0.16 140.8kPa
 v'   v' 0  v'  60140.8 200.8kPa
How'v> 'vmthen the recalculation must be calculated as the sum of the
following payments:
'
'










b0

 v0
v
2.5
180
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Figure 2.39 – Graphs of the influence factor f(m,n) for calculating the increase in vertical tension in
depth z below the corner of a uniformly loaded rectangular area on material
elastic, homogeneous and isotropic. (b) At point A it is obtainedv f (m,n)q(Newmark, 1942)
C. Romanel
131
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Table 2.10 – Influence factor f(m,n) for calculating the increase in vertical stress in

depth z below the corner of a uniformly loaded rectangular area on material

elastic, homogeneous and isotropic.
C. Romanel
132
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
ii) Increase of vertical stress at point B (Fig. 2.40)
15m
Pipeline line
buried
4m
4m
B
Figure 2.40 - Location of point B in the corner of 2 rectangular subareas (15m x 4m).
4
15
mz 4
m
 0.7
nz 15
n
 2.6
5.75
5.75
From Fadum's graph (Fig. 2.39) or Tab. 2.10 comes f(m,n) = 0.17
 v'  2  q f(m,n)  2 220 0.17 74.8 kPa
 v'   v' 0   v'  6074.8 134.8 kPa
How'v< 'vmso the recalculation should be calculated as
'


b0
v
2.5
134.8
60










c 
Crlog

0.1log
 0.04 m
'
1 e0
 v0
1 1


iii) Increase of vertical stress at point C (Fig. 2.41)
Considering the rectangular area extended to point C
4
19
mz 4
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
19m
Pipeline line
buried
4m
4m
A
C
4m
15m
Figure 2.41 – Location of point C in the corner of 2 rectangular subareas (19m x 4m).
4m
Pipeline line
buried
4m
B
C
Figure 2.42 - Location of point C in the corner of 2 rectangular subareas (4m x 4m).
 v'  2  q

f1(m,n)  f2(m,n)   2 220 0.04 17.6 kPa
 v'   v' 0  v'  60 17.6 77.6 kPa
How'v< 'vmso the recalculation should be done by
'


b0
v
2.5
77.6
60










c 
Crlog

0.1log
 0.01 m
'
1 e0
 v0
1 1


b) These pressures could cause some damage to the buried duct assuming
that the maximum angular distortion of the duct is 0.005?
Angular distortion between points A and B
0.09 0.04
  tan 
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
4 - In the soil profile of Fig. 2.43, at time t = 0, a pumping is performed.
in the sand layer 3 causing an instantaneous reduction of the pressure load
hp = 2m at all its points. At time t = 30 days, a superficial layer
of sand (γt20 kN/m3), of large extent and thickness of 2m, is removed from the
surface of the land. Considering the specific weight of water w 10 kN/m3can
se determinar o diagrama de poropressões em: (a) t < 0; (b) t = 0; (c) t = 30 dias;
(d) t = 90 days; (e)t 
Also fill in the poropressions table (Tab. 2.11).
piezometers
Areia 1
0
20 kN/m3
NA
4m
6m
cv= 4 x 10-3cm2/s
k = 2 x 10-7cm/s
clay 1
8m
9m
Areia 2
clay 2
cv= 2 x 10-3cm2/s
k = 1 x 10-8cm/s
11m
12m
Areia 3
clay 3
k = 1 x 10-8cm/s
cv= 2 x 10-3cm2/s
14m
Areia 4
Figure 2.43 - Soil profile with alternating layers of clay and sand.
Time for compaction of clay layers 2 and 3
c vt
H2
2103t
Tv
 1
 t  5106s  57.9 days
2

100

Time for densification of clay layer 1
c vt
H2
4103t
Tv
 1
 t  2.5106s 29 days
2

100

PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
a) t < 0 (Fig. 2.44)
areia 1
0
NA
4m
6m
clay 1
8m
9m
sand 2
clay 2
11m
12m
sand 3
clay 3
14m
sand 4
Figure 2.44 – Distribution profile of pore pressures with depth at t<0.
b) t = 0 (Fig. 2.45)
0
areia 1
NA
4m
6m
clay 1
8m
9m
areia 2
clay 2
11m
12m
areia 3
clay 3
us
14m
areia 4
Figure 2.45 - Profile of pore pressure distribution with depth at t=0.
C. Romanel
136
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
c) t = 30 days (Fig. 2.46)
2m
areia 1
clay 1
NA
4m
6m
8m
9m
sand 2
clay 2
11m
12m
sand 3
clay 3
14m
sand 4
Figure 2.46 – Pressure distribution profile with depth at t=30 days.
d) t = 90 days andt  (Fig. 2.47)
2m
areia 1
clay 1
NA
4m
6m
8m
9m
areia 2
clay 2
11m
12m
areia 3
clay 3
14m
areia 4
Figure 2.47 - Profile of pore pressure distribution with depth at t=90 days andt  
.
C. Romanel
137
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Table 2.11 - Pore pressures (kPa) in the soil profile at different times.
t 
0
20
depth
t<0
0
20
t=0
0
20
t = 30 days
0
20 / -20
0 / 40
t = 90 days
4m
6m
8m
0
20
40
40
40
40
9m
50
50
50 / 10
50 / 90
100 / 60
60 / 100
50
90
100
100
50
90
100
100
11m
12m
14m
110
120
100
110 / 90
100 / 120
100
5 - A large road embankment will be built over a layer of clay.
6m thickness overlying a deep sand layer. To ensure
that the construction be completed in 2 years, and to minimize the occurrence of
road surface imprints, a pre-loading is applied over the
clay layer for a period of 1.5 years. The preloading is
built linearly (Figure 2.48) taking 1 year to reach the maximum value
∆q (kPa). At t = 1.5 years, part of the preloading is removed, along with the fill.
remaining applying a constant vertical tension of ∆qf = 60 kPa, and the construction
The pavement work is carried out immediately. The fill material is highly
permeable and the properties of clay are cv= 4.5 m2/year and mv0.001 m2/kN.
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
 c b0mv v'   c  6 0.001 60 0,36m
b) The final settlement under instantaneous loading ciif
related to the final settlement rebound under loading not
linearly increasing instantaneous cyoufor
 cyou

t

  cIt  0.5tc

  cyou1.5   cI

1

for construction time tc= 1 year
Considering an instantaneous preloading of intensity Δq for the period
of 1 year, then the final settlement recalculation of a central sublayer of
unit thickness results in
1mv60 60
1mv  q  q
cvt
4.51
s f
Z1
Us f


U
com
Tv
 Tv
 0.5
H2
32
s f
U

From 1

In Fig. 2.4, considering Z = 1 and Tv0.5 is determined

= [Link],

60
0.63
  q 95.2 kPa
q
6 - A landfill will be built over a layer of clay 10m thick
overlying the sound rock. The construction will increase the total vertical stress on the clay
at 65 kPa. The project specifies the average percentage of dissipation of the
excesses of poropressure U = 0.85 after 6 months of instantaneous loading.
Determine the necessary spacing of drains in a square mesh.
verticals (400mm diameter) that meets the project conditions.
Consider for the claycv 1.5107m2/sech  2.5107m2/s.
a) Determination of Uvafter 6 months
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
2.5107 630 24 60 60 97.2
c ht
R2
Tr


n2 0.22
n2
a) Rigorous numerical solution
From Eqs. 2.123 and 2.124 it follows
n2
n2 1
3n2 1


2 97.2

nn


n1 0.82  0


n2
4n2


that the bisection method results in n = 8.8
b) Approximate numerical solution
From Eqs. 2.123 and 2.125 it follows
2 97.2

 nn 0.75 n1 0.82  0
n2
by the bisection method results in n = 8.8
c) Graphical solution (Fig. 2.49)
n
Fig. 2.33
5
Tr
Fig. 2.33
0.8
n* = 97.2Tr
11.0
7
1.1
1.5
9.4
8.0
10
The sought value is obtained by the intersection of the interpolated curve with the 3 points.
from the third column and the rectangle* = designed by the origin of the axes, with
inclination of 45as shown in Fig. 2.49, determining the value
approximate n = 8.5. The drainage radius is calculated as R = nrd= 8.8(0.2) =
1.76m and finally, for a square mesh of vertical drains the
spacing is determined as
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Figure 2.49 – Approximate graphical solution for obtaining n = R/rd.

7 - A fill with natural specific weight 20kN/m3was built on the profile of

only indicated in Fig. 2.50. On May 2, 2010, the water level quota at
the interior of the piezometer was +5.5m, in relation to the ground level (GL), and on the day
May 2, 2011 was at +4.0m. Calculate the date of the landfill construction.
admitting instantaneous loading and specific weight of waterw= 10 kN/m3.
Normalized coordinate of the midplane of the clay layer
z
3
Z
  0.5
H 6
Excessive poropressure on May 2, 2010
ue1  w

h1p hpss

 10

10.5 4.0 65 kPa

Excessive poropressure on May 2, 2011
hp2 hpss
ue2  w


 10

9.0 4.0 50 kPa

Percentage of excess pore pressure dissipation at Z = 0.5 on the 2nd day of
May 2010
ue1
u0
65
c vt
62
UZ1 1
 1
 0.1875 Fig.2.4 Tv1 0.075
(A11)
(A12)
141
4 20
Density percentage at Z = 0.5 on May 2, 2011
ue2
u0
50
cv

t1

PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
cv 0.075

0.15


 t  1 year Início da construção em 02/05/2009
62
t
t 1
Figure 2.50 – Measurement of the pressure load in the mid-plane of the clay layer.
In the soil profile of Fig. 2.51, the groundwater level has been lowered.
snapshot from NA1 to NA2( w  10 kN/m3Assuming absence of
capillarity in sand and considering the effects of lowering on the variation of
specific weight of the upper sand layer, graphs of distribution are requested
of pore pressure with depth for the moments: a) before the lowering
a) at t < 0; b) at the moment of lowering at t = 0; c) at  t  Fill in the
Table 2.12 with the pore pressure values at the requested points.
a) t < 0 (Fig.2.52)
Considering the continuity of vertical flow in the upper layers of clay
between the depths of 5m to 10m, comes from Darcy's law
hD hC
hC hB
hB hA
k1
A k2
A k1
A
with A 1m2
2
2
1
hD heD hpD 511 16m
hA heA hpA 10 4 14m
Logo
16 hC
hC hB
hB hA
310 
210 
310 
9
9
9
2
2
1
hC 15.33m hpC hC heC 15.33 7 8.33m uC 83.33 kPa
hB 14.33m hpB hB heB 14.33 9 5.33m uB 53.33 kPa
C. Romanel
142
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
-1m
0
sand
sand
1m
NA1
NA2
A
B
5m
6m
clay 1
clay 2
C
D
8m
clay 1
10m
piezometers
sand
confined aquifer
E
F
12m
clay 2
sand
15m NR
Clay 1
Clay 2
k = 3 x 10-7cm/s
cv 6 x 10-3cm2/s
sat15 kN/m3
k = 2 x 10-7cm/s
cv= 4 x 10-3cm2/s
sat= 15 kN/m3
Figure 2.51 - Soil layer profile and properties.
In the other depths,
uA 40kPauD 110kPa
b) t = 0 (moment of lowering – Fig.2.53)
uE 130kPa
uF 140kPa
Decrease in pore pressure in all clay layers due to variation of
specific weight of sand between 1m ≤ z ≤ 5m
   4 3 4 12kPa
u0 
uA  0
uB 53.3312 41.33kPa
uC 83.3312 71.33kP
uA  4012 28kPa
uD 11012 98kPa
uD  110kPa
uE 130kPa
uE 13012118kPa
uF 14012128kPa
uF 140kPa
C. Romanel
143
PUC-Rio, Department of Civil Engineering
ENG1211 – Mecânica dos Solos – Prof. Celso Romanel
c) t  (Fig. 2.54)
c.1) Due to the variation in the specific weight of the sand (disregarding
capillarity
c.2) Due to the variation of the groundwater level from NA1 to NA2 (Fig 2.55)
Due to the variation of hydraulic load at a depth of z = 5m (point A) the
Hydraulic loads as t → ∞ in the upper layers of clay must be
again calculated based on the permanent flow condition
hD hC
hC hB
hB hA
k1
A k2
A k1
A
with A 1m2
2
2
1
hD heD hpD 511 16m
hA heA hpA 10 0 10m
Logo
3109
16 hC
hC hB
hB hA
2109 
3109
2
2
1
hC 14m
hCp hC heC14 7 7m
 uC 70kPa
hB11m

hpB hB heB11 9 2m
 uB 20kPa
In the other depths,
uA 0uD 110kPa
uE130kPa
uF140kPa
0
sand
sand
1m NA1
A
B
5m
6m
NA2
clay 1
clay 2
t<0
C
D
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
0
sand
sand
1m
NA1
NA2
12 kPa
A
B
5m
6m
clay 1
clay 2
C
D
8m
t<0
clay 1
t=0
10m
12m
sand
confined aquifer
E
F
clay 2
sand
15m NR
12 kPa
us
Figura 2.53 – Perfil das poropressões no instante do rebaixamento (desenho sem escala).
0
sand
1m
NA1
NA2
12 kPa
sand
A
B
5m
6m
clay 1
clay 2
C
D
8m
t→∞
clay 1
t=0
10m
12m
sand
confined aquifer
E
F
t approaches infinity
clay 2
sand
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
0
sand
sand
1m NA1
A
B
5m
6m
NA2
clay 1
clay 2
t→∞
t=0
C
D
8m
clay 1
10m
12m
sand
confined aquifer
E
F
clay 2
sand
15m NR
Figure 2.55 - Profile of pore pressures as t → ∞ due to the variation of the water table from NA1 to
NA2. Note that the graph also represents the final distribution of pore pressures considering the
superposition of the 2 simultaneous events (lowering and unloading due to the variation of
specific weight of sand). Drawing not to scale.
Table 2.12 - Pore pressure values at t < 0, t = 0 and t 
t 
Depth
t<0
t=0
m
1
0
0
0
5
40
0 / 28
0
6
8
10
12
15
53.33
83.33
110
130
140
41.33
71.33
98 / 110
130 / 118
128 / 140
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Negative imprint represented by the hatched area of Fig. 2.54 multiplied
for the corresponding values of mv1or mev2(in the case mv1= mv2).
 cunloading 5 104 12 8 0.048m
d.3) Recalque due to the lowering of the water table from NA1 to NA2
Recalque represented by the shaded area of Fig. 2.55 multiplied by the
corresponding values of mv1you mv2(in the case of mv1= mv2).
40

53.33 20



53.33 20


 (83.33 70)
2
(83.33 70) 2


 crelegation 5104
1
 2




2
2
40

53.33 20

53.33 20
 (83.33 70)
2
(83.33 70)

 crelegation 5104
1
 2


2
2
 cdowngrade 0.048m
 ctotal  cunloading  clower mento 0 .048 0.048 0
The construction schedule provides that 90% of the final settlement surcharge
caused by road fillingt= 20 kN/m3h = 3m) in Fig. 2.56 should
to occur within 200 days after the start of its construction. The construction that
it took 90 days to be completed, with the loading applied linearly
in time.
Please check:
a) whether the project condition will be met;
b) in a negative case, one of the options for the engineer is to use a preloading
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Final settlement due to road fill
 c  b0mv  v'  4103 203 0.24m
landfill road
sand
3m
0
NA
3m
5m
sand
clay
9m
sand
Figure 2.56 – Road embankment over a soil mass containing a clay layer.
Design criterion: non-instantaneous settlement at t = 200 days must be at least
equal to 90% of the final settlement recalculation of the landfill.
 cyou

t  200days0.240.90 0.216m

Consolidation settlement at t = 200 days for linearly applied load
increasing over time is equivalent to the instantaneous settlement at t' = t - tc/2 = 200
- 90/2 = 155 days
cvt 6108155246060
Tv

 0.20
H2
22
4Tv
40.20
Uv

 0.50 0.60


 cyou

t 200days

  ci

t155days  cUv 0.240.5 0.12m 0.216m

Therefore, the design criterion is not met considering the dissipation of
excesses of poropressure only by vertical drainage.
b) Alternative 1: preloading with a maximum fill height of 2.5m
Additional compaction recalque
 c b0mv v'  4103 20 2.5 0.20m
C. Romanel
148
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Additional consolidation recalculation at t = 200 days for applied load
linearly with time.
For Z = 1, Tv= 0.20 from the isochrone graph in Fig. 2.4 resultsUZ1 0.23
 cyou

t 200days

 ci

t155days  cUFrom1 0.200.23 0.046m

Verification of the design criterion
t 200days t 200days 0.12 0.046  0.166 0.216m
 cyou  cyou




Therefore, the design criterion is not met considering the application of the pre-
loading.
c) Alternative 2: pre-loading with a maximum height of 2.5m together with
vertical drains
Consolidation thickness to be achieved by the use of vertical drains,
excluding the amount due for pre-loading
0.17
 ncit 200days 0.216 0.046 0.17m

U
 0.787
0.216
Average percentage of dissipation of excess pore pressure due to
radial drainage Ur

1U



1Uv1Ur



1 0.787



1 0.501 Ur

Ur  0.574
Considering R = rdn = 0.2n and t = 200 days,
c ht 15108 200 24 6060 64.8
Tr


R2
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
the result by the bisection method n = 9
Logo
R r n 0.2 9 1.8m
d
In a square configuration, the spacing S between the drains is given by
S R  1.8  3.2m
10 - (Olson, 1977) Consider a layer of clay (cv0.0045 m2/day) with 3m
of thickness and double-sided drainage. A large area landfill is
built between times t = 0 and 30 days, with linear application of the loading,
until reaching a thickness of 3m, maintained constant until t = 50 days. A new landfill
is built upon the previous one, with linear application of the load, between t = 50
and 100 days, until reaching 6m in thickness, kept constant until t = 200 days,
when 3m of fill is removed instantly. The factor values
construction time Tcfor each stage are presented in Tab. 2.13.
It is requested to determine the settlement due to consolidation at times t = 10, 30, 50, 100,
200 to 400 days, considering the following values of final settlements caused.
for each of the stages: a)  landfill1 25mm; b) landfill2 50mm; c)
 unloading 25mm .
Table 2.13 - Construction Time Factor Tcfor each stage
Slow Landfill1
Slow Landfill 2
Instant Removal
0
0.06
0.10
Considering the curves in Fig. 2.15, with interpolation for the case Tc= 0.06
it results in the partial settlement values in Table 2.14.
Table 2.14 - Average compaction percentage Uvthe primary consolidation recalculation t
Landfill 1
Uv
Landfill2
Uv
Instant Removal
Uv
t
t
t
t
mm
1
4.25
6.75
11.75
17.00
22.25
t
mm
-
-
0
10.50
28.50
42.00
t
days
Tv
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
Table 2.15 - Evolution of consolidation settlement over time, considering two stages of
slow loading linearly increasing.
Time
(days)
10
30
50
100
200
400
Trace
(mm)
1
4.25
6.75
22.25
45.50
46.75
11 - A perfectly flexible circular foundation with a radius R = 20m is uniformly
loadedq = 100 kPa) at the surface of the soil mass shown in Fig. 2.57.
It is requested to calculate the final settlement of primary consolidation by the following
methods:
a) state of deformation 1D considering infinite loading (theory
conventional)
b) 1D deformation state considering only stress increments
verticals generated by the loading of the circular foundation;
c) Skempton-Bjerrum method considering foundation loading
circular and only with pore pressure parameters B = 1 e A = 0.85
(disregard the influence of the superficial sand layer);
d) theory of linear elasticity 3D
R = 20m
q = 100 kPa
0m
2m
r
Sandt = 15 kN/m3
Clay NAsat16 kN/m3
E' = 3 MPa
v’ = 0.25
z
B = 1A = 0.85
42 m
Sandt18 kN/m3
Figure 2.57 - Circular foundation with uniform loadingq = 100 kPa
Note: For the purposes of indirect calculation of the settlement, consider the clay layer divided into 5.
sub-layers of 8m thickness. The increases in tension along the z-axis that passes through the
Center of the circular foundation are:
C. Romanel
151
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
z/R
0.3
0.7
1.1
1.5
1.9
⁄
0.98
0.44
0.81
0.15
0.59
0.04
0.42
------
0.31
------
∆ ∆
⁄
∆ ∆
a) state of 1D deformation considering infinite loading (theory
conventional
E

1

3000(1 0.25
(1 0.50 )(1 0.25)
M

 3600 kPa

1 2 1

 v'
M
100
 C1D b0
 40
 1.11m
3600
b) state of 1D deformation considering only increases in vertical stress
generated by the loading of the circular foundation.
Considering the increases in vertical tension at the center of each sublayer of 8
meters, along the vertical axis z
z (m)
∆ (kPa)
6
98
14
81
22
59
30
42
38
31
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
z (m)
6
14
22
30
38
98
44
0.025
81
15
0.024
59
04
0.019
42
------
0.014
31
------
0.010
∆ (kPa)
∆ (kPa)
∆
 3D b  8 (0.025 0.024 0.019 0.014 0.010 0.736m

0
zz
Considerando que o módulo de cisalhamento G é o mesmo para as condições
drained and undrained (vu= 0.5) results
′

0 3000 3600

21 ′
21 0.5
2.5
In the undrained condition (Eu = 3600 kPa, vu= 0.5) vem
1
∆
∆ ∆ ∆

z (m)
6
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
c) Skempton-Bjerrum method considering the foundation loading
circular and only with pore pressure parameters B = 1, A = 0.85
 u B

 3 A

 1  3

with 3  rr e 1  zz on the z-axis
d) theory of 3D linear elasticity
 zz  2 rr
 you
or admit B = 1 and A = 1/3 in Skempton's equation
3
PROPOSED EXERCISES
1 - Consider a clay layer 6m thick, with double face of
drainage and the following properties: vertical consolidation coefficient cv=
0.009 m2/day, horizontal consolidation coefficient ch0.045 m2/day, coefficient
of vertical permeability kv6 times 10-8cm/s and permeability coefficient
radial kr= 30 x 10-8cm/s. For a period of 600 days, plot the graph of
average percentage of dissipation of excess pore pressure U versus t in
following situations:
1.a) only vertical drainage;
1.b) vertical drain with radius rd= 0.20m, drainage radius R = 3m, s = 1 (without effect
smear);
1.c) vertical drain of radius rd= 0.20m, drainage radius R = 1.5m, s = 1 (without
smear effect
1.d) vertical drain of radius rd= 0.20m, drainage radius R = 3m, s = 1.5 (with
effects smear) and relationshipkrks 4.
Answer: Fig. 2.58

Figure 2.58 - Variation over time of the average percentage of excess dissipation
propression for the conditions of the proposed exercise 1.
C. Romanel
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
2 - Consider a clay layer 6m thick, with double face of
drainage and the following properties: cv9 times 10-3m2/day, ch= 45 x 10-3m2/day,
mvlinearly varying from 12 x 10-4m2/kN (layer top) up to 6 x 10-4m2/kN
(base of the layer). It is requested to calculate:
a) the consolidation settlement at time t = 8 months considering the application
instantaneous of an infinite load with intensity ∆q = 40 kPa;
b) the engineering project specifies that only 3cm of the total settling are
admissible after t = 8 months. Check how to meet this condition, using
vertical drains (rd= 20cm) in a square mesh;
c) another possible solution, without considering the use of vertical drains, for
meeting the design conditions is to increase the loading intensity with
the objective of reaching the specified value of subsidence at time t = 8 months.
Determine the value of the overload ∆qs necessary, considering that the
total charge (40 + ∆qskPa is applied instantly.
 ct 0.119m
Respostas: a)
b) S≈5 m c)∆qs= 74.81 kPa
3 - An engineering project predicts that the existing structure will present a
final settlement of 10cm due to circular loading (R
= 300 cm) placed nearby, as shown in Fig. 2.59. Determine the
settlement value at the center of the existing structure at time t =
1.5 years.
Answer: . 7.56
4 - A cylindrical reservoir (R = 6m, h = 3m) for oil storage

 oil 8 kN/m3
was built on a layer of saturated clay of 24m

thickness, according to Fig. 2.60. Consideringw 10 kN/m3it is requested to determine
the final settlement recalculation in the center of the reservoir, accepted as
perfectly flexible foundation, by the following methods:
a) 1D state of deformation considering only the distribution of increments
of vertical tensions induced by the loading of the reservoir;
b) formulation of Skempton-Bjerrum;
c) 3D theory of linear elasticity, with the aid of equations A.9 and A.10 or
graphs of Fig. 2.3;
d) in your opinion, what is the best estimate of the final settlement subsidence?
among the values calculated above? Why?
Answers: Dividing the clay layer into 4 sublayers of 6m each.
thickness, comes: a) 6.3 b) 3.3 (c) 2.2 recalque
final of primary densification determined by the theory of elasticity, due to
3D nature of the formulation and the fact that the soil, in this example, is assumed
linearly elastic material (A = 1/3).
C. Romanel
155
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
Vertical cut
Horizontal projection
structure
existing
structure
existing
q
600cm
sand
300cm
300cm
clay
cv= 80 cm2/day
rock
Figure 2.59 - Vertical section and horizontal projection of existing structure and circular loading
in the vicinity.
3m
NA
6m
clay
24m
sand
Figure 2.60 - Cylindrical reservoir for oil storage over a clay layer.
5 - The construction schedule foresees that the final settlement caused by
by the road landfill

 t 20 kN/m3,h 2m
must occur within 30 days after

its construction, considered instantaneous (Fig.2.61). Disregarding effects of
capillarity, it is requested:
C. Romanel
156
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
a) check if the instantaneous lowering of the water table from NA1 to NA2 (w
10 kN/m3it could be a method to achieve this design condition,
including the effects of the variation of the specific weight of the sand layer;
b) in the negative case, in addition to the lowering of the water table, determine the height
additional fill (preloading) to achieve the design condition,
considering snapshots of both events (the lowering of the water table and
launch of the overload.
Respostas: a) F 6.1 PQR 8 PQR b)h = 1,72m
h
landfill
0
sand
NA1
3m
sand
NA2
5m
clay
9m
 sat 20kN/m3
sand
Figure 2.61 - Densification caused by road embankment and lowering of the water table level
phreatic.
6 - In the soil profile of Fig. 2.62, the level of the groundwater table has lowered.
snapshot from NA1 to NA2. Assuming absence of capillarity and
 w 10 kN/m3it is requested:
a) the poropressure distribution graphs with depth before the
downgrade (t<0), at the moment of downgrade (t=0) and in the permanent condition

t

;
b) the values and meanings of the flow velocities at the above moments, in the plane
average of the 3 layers of clay;
c) considering clay layers NA (A = 0.5, B = 1), you would expect that the
final settlement due to primary consolidation calculated using the Skempton method and
Bjerrum would be equal to, greater than, or less than the value of the 1D settlement obtained by the theory.
Conventional 1D? Explain.
C. Romanel
157
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
d) what is the estimate of the settlement at t = 0? Why?
Answers: a) to observe that the downgrade only causes a change in
pore pressure in the upper clay layer; b) 3m clay layer z 5m

⁄
( 0 e → Layer of clay 6m z 9m

⁄
( → 1.43 10 ↑); c) for infinite loading the
pore pressures values by the conventional 1D theory and by Skempton's method
Bjerrum are equal; d) undrained recoil is null for state 1D of
deformation.
piezometer
0
NA1
sand
1m
sand
NA2
3m
clay I
5m
sand
6m
clay I
8m
clay II
9m
sand
Figure 2.62 – Stratified soil profile with alternating layers of sand and clay.
Answers: a) observe that the downgrade only causes a change of
pore pressure in the upper clay layer; b) clay layer 3m z 5m

⁄
( 0 → Layer of clay 6m z 9m

⁄
( → 1.43 10 ↑); c) for infinite loading the
recalculation values by conventional 1D theory and by Skempton's method
Bjerrum are equal; d) undrained surcharge is null for 1D state of
deformation.
7 - Regarding the soil profile of Fig. 2.63 and Tab. 2.16, assumed as material
linear elastic, the following events occurred:
a) Instantaneous lowering of the water table from NA1 to NA2 at time t0
= 0;
b) Instant removal of the top layer of dry sand of 2m
thickness at the end of primary settlement caused by lowering of the
water table at time t1;
C. Romanel
158
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
c) Instant construction on the new surface of the excavated land of a
circular foundation, perfectly flexible, with radius R = 6m with load
uniformly distributed ∆q = 100 kPa at the moment corresponding to the end
of primary thickening2caused by the removal of the surface layer of
sand
d) Instantaneous reburial of the original sand layer at time t3
corresponding to the end of primary consolidation caused by
circular foundation loading;
e) Instantaneous rise of the groundwater table from NA2 to NA1 at time t4
corresponding to the end of the primary consolidation caused by the backfill of
superficial layer of sand.
It is requested to determine, considering the specific weight of waterw= 10 kN/m3e
disregarding the effects of capillarity and the variation of total vertical tension due to
the variation of the specific weight of the sand in the region 2m < z < 4m:
i)
The distribution of pore pressures with depth at time t0do
instantaneous lowering of the groundwater table from NA1 to NA2.
ii)
iii)
iv)
v)
The distribution of pore pressures with depth at time t1yes
removal of the top layer of sand. What is the value of this time?
The distribution of pore pressures with depth at time t2do
loading of the circular foundation. What is the value of this time?
The distribution of pore pressures with depth at time t3do
Greater the removal of the surface layer of sand. What is the value of this time?
The distribution of pore pressures with depth at time t4yes
elevation of the groundwater table from NA2 to NA1. What is the value of this?
time?
vi)
What is your best estimate of the final settlement due to primary consolidation?
from the center of the foundation circular perfectly flexible at time t→ ∞?
vii)
Repeat the previous question considering the loading of the foundation
took 40 days to be completed.
Note: in addition to the distribution graphs of pore pressures with depth
fill in Tab. 2.17.
C. Romanel
159
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
0
Dry sand
Areia
2m
4m
NA1
NA2
clay
7m
8m
areia
clay

areia
12m
Figure 2.63 - Soil Profile
Table 2.16 - Soil Properties
weight
alone
specific
cv
mv
coefficient of
Poisson v'
(kN/m3)
m2/ s
m2/ kN
sand
dry
sand
18
-----
-------
---------
saturated
clay
20
18
-------
-------
----------
0.3
4 x 10-7
3 times 10-4
Table 2.17 – Pore pressure values in the soil profile (kPa)
prof. (m)
t0
0
20/
0
t1
0
0/
t2
0
0/
100
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
105.42
90/
111.41
100/
118.13
116
90/
126
100/
136
11
12
90
54
64
90
90
100
100
100
Answers: ii) t1= 65.1 days; iii) t2= 115.7 days; iv) t3= 231.4 days; v) t4347.1 days;
vi) observe that due to the soil elasticity hypothesis, the displacements
verticals caused by the lowering and raising of the NA, as well as by removal
the restoration of the surface layer, if mutually annual; vii) the final value of
primary consolidation settlement of the circular foundation, calculated by the theory of
3D linear elasticity, does it remain the same considering instantaneous loading or
slow loading.
The foundation of a large building consists of two flexible square footings,
with 20m on each side, and a separation of 10m between them. A horizontal projection of the
foundation is shown in Fig. 2.64. The building applies a uniform load of
200 kPa on the foundation. The footings are placed on the surface of a reservoir.
of soil formed only by a upper layer of dense sand with a thickness of 9m,
followed by a layer of clay 2m thick and, finally, by a
thick layer of gravel. The water table is 4m below the
the surface and the sand situated above is dry. The properties of the sand ared= 15
kN/m3, sat19 kN/m3And the properties of clay are listed in Tab. 2.18.
Table 2.18 - Clay Properties
Saturated specific weightsat
Grain density Gs
16.5 kN/m3
2.6
Poisson's coefficient v'
0.3
Reloading index CR
Compressibility index Cc
Pre-consolidation pressure´vmax
Consolidation coefficient cv
0.06
0.35
175 kPa
0.5 m2year
Ignoring the settlements of the layers of sand and gravel and assuming
that the construction of the foundations took a year, with a loading rate
constant, it is requested:
a) according to the 1D theory, the primary consolidation settlement at the center of one of the
shoes int 
b) by the theory of 1D consolidation, the primary consolidation settlement in
PUC-Rio, Department of Civil Engineering
ENG1211 – Soil Mechanics – Prof. Celso Romanel
20 m
10 m
20 m
Figure 2.64 - Horizontal projection of two footings of a large building
Answers: a) c  0.055m
b)  cyou(2) 0.048m
9 - A perfectly flexible circular foundation with a radius of R = 10m is uniformly
loadedq = 100 kPa) at the surface of the soil mass formed by a
surface sand layer of 5m thickness and a saturated clay deposit
NA (Skempton parameter A = 0.85) with a thickness of 20m, as shown in the
Fig. 2.65. Table 2.19 lists the increments of vertical stress and deviation along
from the foundation axis to some points of the clay layer, considering the case
of 2 different soils with contrasting modules 10.
It is requested to calculate the final settlement of primary consolidation by the method of
Skempton-Bjerrum, considering the influence of the surface sand layer.
In your opinion, which method would provide a better estimate of the settlement of
primary density in this case?
Respostas: a)  CSB 6.3cmb) Observe in the development of the solution that the
influence of(1 A) 3' is small in the Skempton-Bjerrum method and there is no
significant difference in the backpressure value compared to formulation 1D, where 3'
is ignored. Note also that in both methods the settlement is calculated with a
compressibility parameter of the 1D deformation state, valid hypothesis
only for large extent shipments A better estimate of
recalque is obtained by the 3D theory of linear elasticity, including the
immediate recalculation parcel.
C. Romanel
162
PUC-Rio, Department of Civil Engineering
ENG1211 - Soil Mechanics - Prof. Celso Romanel
R = 10m
q = 100 kPa
0m
5m
r
Medium sandt15 kN/m3
E = 30 MPa
Soft clay NAsat16 kN/m3
E' = 3 MPa
v’ = 0.25
z
B = 1A = 0.85
25 m
Sandt = 18 kN/m3
Figure 2.65 - Perfectly flexible circular foundation on stratified soil deposit.
Table 2.19 – Increases in vertical stress and deviator stress as a percentage of
loading for v = 0.5
Depth (m)
∆
29.2
16.8
10.5
7.0
∆ ∆
18.8
13.5
9.2
5
10
15
20
25
6.4
4.7
5.0
C. Romanel
163