Retaining Structures and Stability of Soilmass

Project number
415660-21 v04
Date
December 2007
Preface
Geotechnical properties of soil generally differ from those of rock. Soil properties are deter-
mined by texture, i.e. its appearance, which depends on the grain size, the grain shape, the void
ratio and the water content.
The science which deals with the engineering behaviour of soil is called soil mechanics (see
lecture note Soil mechanics). A sufficient knowledge of soil mechanics is essential for the de-
sign of foundations, retaining walls, sheet pile walls, embankments, etc. One should realise that
the behaviour of soil under the influence of loads is generally more complicated than that of
other building materials. Stress-strain relations of soil are usually non-linear. Furthermore,
irreversible deformations and time effects play an important role.
The lecture series Geo Engineering 1 deals with: Earth retaining structures; gravity wall, analy-
sis of sliding and overturning and allowable soil pressures; sheet pile wall, analytical and
(Winkler) spring models, screwed anchors, grout anchors, anchor walls, struts, and anchor
piles. Slope stability, according to Fellenius and Bishops theorys including the effect of an
earthquake load and groundwater flow. Shallow foundations, calculations of bearing capacity
under vertical and inclined loads according to Prandtl-Buismans and Meyerhofs theory, set-
tlement calculations, allowable deformations, mutual influencing of foundations. Deep founda-
tions, overview of piling systems, determination of end bearing capacity and of positive and
negative friction.
This syllabus with lecture notes deals with Earth retaining structures and the Stability of a soil
masses. These lecture notes contain parts of books and syllabi that describe certain geotechni-
cal aspects. Chapters 2, 3 and 4 are basically a copy of TUDelft lecture notes for the lecture
notes series CT5331 Sheet pile walls which in itself is a translation of the Dutch code of prac-
tice CUR166. The lecture notes are provided by H.J. Everts. Further use has been made of the
lecture notes of C. van Veen, a former member of the UNESO-IHE staff.
The full contents of Chapter 3 to 5 in all its details go beyond the goals of this lecture series.
During the lectures the parts that are of special interest will be notified. The remainder of con-
tent of these chapters can be used as reference material.
The topics Shallow foundations and Deep foundations are dealt within two separate lecture
notes.
Project number
415660-21 v04
Date
December 2007
Contents
1 Introduction in earth retaining structures 1
1.1 Types of earth retaining structures 1
1.2 Horizontal earth pressures 1
1.3 Retaining walls 5
1.3.1 General 5
1.3.2 Vertical force equilibrium: 7
1.3.3 Horizontal force equilibrium: 7
1.3.4 Moment equilibrium of the wall 7
1.3.5 Moment equilibrium of the wall and the soil as a unit 7
1.4 Sheet pile walls 8
1.4.1 Types of sheet pile walls 8
1.4.2 Stability of a sheet pile wall 8
1.4.3 Anchorages and struts 11
1.5 Reinforced earth 14
2 Sheet pile walls Design, Safety and Quality 16
2.1 Comments 16
2.2 Design process 16
2.2.1 Introduction 16
2.2.2 Description of the design process 17
2.2.3 Programme of demands 23
2.2.4 Norms, guidelines and recommendations 25
2.2.5 Construction time 26
2.2.6 Costs of the project 28
2.3 Description of alternative constructions 30
2.3.2 The sheet pile wall 34
2.3.3 The alternative soil retaining constructions 37
2.4 Safety considerations 47
2.4.2 Testing the maximum soil resistance, the maximum moment in
the sheet pile and the maximum anchor force 51
3 Parameters for sheetpile desgin 61
3.1 Design parameters 61
3.2 Loads 63
3.2.2 Loads due to soil 63
3.2.3 Surcharge load 75
3.2.4 Horizontal loads due to water level differences in free water
and groundwater 76
3.2.5 Anchor forces and strut forces 86
3.2.6 Ice loads 86
3.2.7 Ship impact 88
3.2.8 Boulder forces 90
3.2.9 Loads due to earthquakes and vibrations 91
3.2.10 Loads due to compaction 92
3.2.11 Loads due to swelling soil 93
3.2.12 Loads due to temperature changes 93
3.3 Parameters of the sheet pile wall 94
3.3.2 Oblique bending 94
3.4 Parameters of the soil 107
3.4.2 Determination of parameters 109
Project number
415660-21 v04
Date
December 2007
3.4.3 Determination of the representative value of soil parameters,
based on test results 128
3.4.4 Partial material factors for soil 129
3.4.5 Type and size of soil investigation 129
3.4.6 Report of the soil investigation 132
4 Calculation models for sheetpile wall desgin 134
4.1 Classical calculation models 134
4.1.2 Determining the maximum earth pressure and soil friction 135
4.1.3 Angle between the slip surface and the horizontal axis 135
4.1.4 The influence of a not horizontal ground level in Coulombs
method 136
4.1.5 The influence of a piled structure 138
4.1.6 The influence of surcharge loads 139
4.1.7 The influence of an embankment 142
4.1.8 Reduction of the moments and the increase of the anchor
force 143
4.1.9 Undrained soil behavior 145
4.1.10 Water pressures and groundwater pressures 145
4.1.11 Other loads 145
4.1.12 Influence of compaction of the soil behind a sheet pile wall
construction on the earth pressure 145
4.1.13 Design method according to Blum 145
4.1.14 Computer code for the analytical calculation according to
Blum 155
4.2 Calculation model for a beam on elastic foundation 160
4.2.2 Basic equation 160
4.2.3 Pile wall as elastic supported beam 163
4.2.4 Basicsl of the calculation model 166
4.2.5 Limitations 177
4.2.6 Input parameters 179
4.2.7 General view on parameters 184
4.2.8 Input check 186
4.2.9 Execution of the calculation 187
4.2.10 Speci fic aspects 189
4.3 Calculating using the finite element method 192
4.3.2 General description of the method 193
4.3.3 Software for the finite element method 195
4.3.4 Application of the finite element method 196
4.3.5 Aspects when modeling with the finite element method 197
4.4 Stability check 206
4.4.2 Failure due to shearing along a deep straight slide plane
(Kranzs method) 207
4.4.3 Loss of total stability due to shearing along a circular slide
plane 211
4.4.4 Piping 215
4.4.5 Heave 219
4.5 Designing the anchor construction 223
4.5.2 Applicability tests and verification experiments 223
4.5.3 Elaboration of the experiments 226
4.5.4 Free anchor length 228
4.5.5 Anchor struts and walls 229
4.5.6 Screw anchors 232
4.5.7 Grout anchors 234
Project number
415660-21 v04
Date
December 2007
4.5.8 Screw-injection anchors 238
4.5.9 Anchor piles 239
4.5.10 Group functioning of anchors and piles 243
4.5.11 Variable loads 243
4.5.12 Specific execution aspects 244
4.5.13 Designing the wales 244
4.6 Sheet pile wall constructions loaded in the plane 245
4.6.2 Failure mechanisms 246
4.7 Coffer dams 246
4.7.2 Function, application options, construction and parts of coffer
dams 246
4.7.3 Loads and testing procedures 248
4.7.4 Force envelope in coffer dams 249
4.7.5 Possible failure mechanisms 250
4.7.6 Rules of thumb dimensioning 251
4.7.7 Boundary conditions, starting points and calculation scheme 252
5 Stability of a soil mass 254
5.1 Introduction 254
5.2 Slope stability calculations 254
5.3 Stability of an infinite slope 255
5.4 Circular slip surfaces 258
5.4.1 The method of slices, Fellenius 260
5.4.2 The method of slices, Bishop 263
5.4.3 The method of slices, Bishop with Earthquake effect 265
5.5 Special cases 266
5.5.1 Stability of clay slopes 266
5.5.2 Stability of slopes in fine sand 268
5.5.3 Composite slip surface 269
5.6 Influence of groundwater on the stability of a soil mass 269
5.6.1 General 269
5.6.2 Groundwater flow and stability of dams 270
5.6.3 Effect of rapid draw down of reservoirs on dam stability 271
5.6.4 The influence of a horizontal filter on the pore pressure and
stability 273
5.6.5 The influence of a filter on the stability in homogeneous soil
with anisotropic permeability 275
5.6.6 Rapid drawdown: pore pressure and stability 276
5.7 Magnitude of the required factor of safety 277
6 List of references 279
Project number
415660-21 v04
Date
December 2007
Page
1 of 279
1 Introduction in earth retaining structures
1.1 Types of earth retaining structures
Earth retaining structures are used if there is not enough space available to construct a stable
slope, to create mooring facilities for ships (quay walls) or to enable an excavation in a con-
fined space (sheet pile walls). The retaining structure can be temporary or permanent and is
often also used as a foundation element to transfer vertical loads to a soil layer with sufficient
bearing capacity.
(1) (2) (3)
In these lectures a number of types of structures will be described. Sheet pile structures are
dealt within more detail. The following types of structures can be distinguished (see Figure
1.1):
- a gravity wall which is prevented from overturning by its weight
- a sheet pile wall which is embedded in the subsoil and obtains its stability from passive
earth pressure and often also from anchors or struts
- a structure of reinforced earth which obtains its strength from metal strips or geotextiles in
the backfill.
These are also structures which consist of a combination of the above mentioned types of struc-
tures.
1.2 Horizontal earth pressures
Before earth retaining structures are dealt with, at first some remarks are made concerning
terms such as active, neutral and passive earth pressure, which play a role in the design of these
structures. The definition of the active, neutral and passive earth pressures are described in the
lecture notes for Soil Mechanics
The coefficients of active, neutral and passive earth pressure (K
a
, K
o
and K
p
) denote the ratio
between the horizontal effective stress and the vertical effective stress. Their value depends on:
- the direction and magnitude of soil deformation
- the angle of internal friction |'
- the angle of wall friction o' (for detailed analysis).
Figure 1.1 Earth retaining structures: gravity wall (1), sheet pile wall (2) and wall of reinforced earth (3)
Project number
415660-21 v04
Date
December 2007
Page
2 of 279
o
'
v
Kp
K
0
K
a
deformation on active side deformation on passive side
o
'
h
o
Figure 1.2 shows that the value of the earth pressure is dependent on the magnitude and the
direction of the horizontal soil deformation. As described in the lecture notes Soil Mechanics
from the coefficients of active and passive earth pressure (K
a
, and K
p
) are defined by:
|
.
|
\
|
' = |
2
1
45 tan
2
a
K
(1.1)
|
.
|
\
|
' + = |
2
1
45 tan
2
p
K
(1.2)
The active (minimum) and the passive (maximum) horizontal effective pressure can be derived
from the following equations:
a v a a
K c K ' ' = 2 o o (1.3)
p v p p
K c K ' + ' = 2 o o
(1.4)
In the neutral situation without deformation, the horizontal stresses in the soil can be estimated
using equation (1.6).
The coefficient for neutral earth pressure (K
o
) is a parameter that is very difficult to determine.
In practise a correlation of Jacky (1.5) is used for normally consolidated soils, which gives a
reasonable estimation.
|' = sin 1
o
K
(1.5)
So for normally consolidated soil the horizontal stress at rest is:
v o nc
K o o ' = '
(1.6)
In situation there the soil is overconsolidated the horizontal pressure at rest (o'
oc
) can be sig-
nificantly higher then predicted by (1.6) but never can exceed the passive value (o'
p
).
There is also a different way to illustrate how its work with there active and passive stress
states. Consider a retaining wall ABDE, subjected to active earth pressure, caused by cohe-
sionless soil (unit weight ). The smallest earth pressure (active earth pressure) is obtained
Figure 1.2 Relation between earth pressure coefficient and the deformation of soil
a
o
p
o
nc
o '
Project number
415660-21 v04
Date
December 2007
Page
3 of 279
when the retaining wall moves over a sufficient distance away from the soil it supports. The
height of the wall is H (see Figure 1.4).
If the ground level of the backfill is horizontal and the angle of wall friction o' = 0, the slip
surface BC is plane and makes an angle with the vertical back side of the wall (see lecture
notes Soil Mechanics) for the situation for the minimum (active) horizontal pressure:
|
.
|
\
|
' = | u
2
1
45
The soil wedge BCD is subjected to the following forces:
- the weight G of the soil wedge
- the (horizontal) reaction force P of the wall
- the reaction force R, which acts under an angle |' with the normal on the slip surface (the
direction of R is such that the shear force component acts in an opposite direction to the
movement along the slip surface of soil wedge BCD).
u
E D C
A
H P
G
B
R
G R
P
o
The weight G of the soil wedge BCD for a thickness of 1 metre is:
|
.
|
\
|
' = B = |
2
1
45 tan
2
1
CD area
2
G
The reaction force P, which is equal and opposite to the active earth pressure on the wall, is
determined from the force polygon in Figure 1.3:
| tan =
G
P
where:
|
.
|
\
|
' = |
.
|
\
|
' ' = | | | |
2
1
45
2
1
45 90
which gives:
|
.
|
\
|
' = = |
2
1
45 tan
2
1
2 2
P P
a
(1.7)
Figure 1.3 Forces acting on soil wedge behind retaining wall
|
Project number
415660-21 v04
Date
December 2007
Page
4 of 279
90
u
o

u
R
(a)
(p)
R
u
(R = R
a
for active earth pressure, see Figure 1.4).
In a similar manner the passive earth pressure can be determined. As the active earth pressure
is developed when the retaining wall is moved slightly away from the soil it supports, the pas-
sive earth pressure is developed when the retaining wall is moved in the direction of the soil it
supports. The soil wedge BCD moves then upward along the slip surface BC. The direction of
the shear force component is now directed downwards along the slip surface, which in turn
determines the direction of the reaction force R (= R
p
, see Figure 1.4).
The formula for the passive earth pressure becomes:
|
.
|
\
|
' + = = |
2
1
45 tan 2 / 1
2 2
P P
p
.
(1.8)
If the soil behind the retaining wall has also cohesion c', (1.1) can be extended as follows:
|
.
|
\
|
' ' |
.
|
\
|
' = | |
2
1
45 tan 2
2
1
45 tan 2 / 1
2 2 2
H c P
a
where:
P
a
= active earth pressure on the wall, force per linear metre (kN/m)
= unit weight of soil (kN/m
3
)
H = height of wall (m)
|' = angle of internal friction (degrees)
c' = cohesion (kN/m
2
).
This formula can be rewritten as:
a a a
K c K P ' = 2
2
1
2
(1.9)
where:
|
.
|
\
|
' = |
2
1
45 tan
2
a
K = coefficient of active earth pressure.
Figure 1.4 Determination of magnitude of angle o
|
Project number
415660-21 v04
Date
December 2007
Page
5 of 279
In a similar way formula 4.2 can be extended in the case of cohesion:
p p p
K c K P ' 2
2
1
2
+ = (1.10)
where:
P
p
= passive pressure on the wall, force per linear metre (kN/m)
|
.
|
\
|
' + = |
2
1
45 tan
2
p
K = coefficient of passive earth pressure.
If it is assumed that horizontal effective stresses (thus exclusive water pressures) increase
linearly with depth for cohesionless soil, the effective stress diagrams can be depicted as in
Figure 1.5.
P = hatched area
a
H
1
o
h
= H k
a
o
h
= H k
0
H K
a
2
=
H
0
2
= K H
p
2
= K p
P
o
v
= H o
v
= H o
v
= H o
h
= H k
p
The above given solutions for the determination of the horizontal effective pressures on a wall
1.3 Retaining walls
1.3.1 General
The stability of a retaining wall is obtained from its large own weight, sometimes in combina-
tion with passive earth pressure. Retaining walls are usually monolithic structures and their
weight is determined by the unit weight of masonry or concrete, see Figure 1.6. Nowadays
often cantilever walls are constructed, which obtain most of the required weight to meet stabil-
ity from backfill see Figure 1.7. Such cantilever walls are generally made of reinforced con-
crete, so that a relatively low quantity of concrete is necessary.
Figure 1.5 Active, neutral and passive effective stresses on a vertical wall for cohesionless soil
Figure 1.6 Gravity walls
Project number
415660-21 v04
Date
December 2007
Page
6 of 279
Gravity walls and cantilever walls (sometimes called L-walls) are sensitive to rotation and
translation. To construct/design a stable wall, stability conditions for 4 hypothetical mecha-
nisms have to be satisfied. For the stability 4 important conditions have to be fulfilled:
- Vertical force equilibrium:
- Horizontal force equilibrium:
- Moment equilibrium of the wall
- Moment equilibrium of the wall and the surrounding.
Figure 1.8 shows these failure mechanisms.
Beside these four soil mechanical conditions there should not be structural failure of the wall
element itself. Further it should be checked if the resulting deformations dont exceed the limi-
tations for serviceability of the construction. These excessive deformations can easily occur
when one is constructing a wall on soft peat and clay.
a b
c d
Figure 1.7 Cantilever or L-wall
Figure 1.8 Ffailure mechanisms of a retaining wall
Project number
415660-21 v04
Date
December 2007
Page
7 of 279
In case of large deformations it should checked if geometrical nonlinear effects (the so called
second order effects) become of significant importance. For instance settlement/compression of
the soil under a wall can lead to a situation that a wall leaning forward. The leaning of the wall
has a negative effect on the stability: the rotational moment increase by the shift in the centre
of gravity. The maximum pressures under the foundation will also increase, so the bearing ca-
pacity of the soil has to be checked again.
1.3.2 Vertical force equilibrium:
The bearing capacity of the soil below the wall (see Figure 1.8c) should be higher than the
weight of the wall (Prandtl / Brinch Hansen). For this mechanism a safety factor of F
s
= 2.0
generally used.
For this analysis, first all the resultant and its line of action of all the loads on the retaining wall
have to be determined. This resultant force should be compared with bearing capacity accord-
ing to the calculation method for shallow foundations (see lecture notes shallow foundations)
for the determination of the vertical bearing capacity of the soil underneath the retaining wall.
1.3.3 Horizontal force equilibrium:
The friction between the base of the wall and the subsoil plus possible passive earth pressure
(also called passive resistance) should be sufficiently higher than the horizontal force due to
active earth pressure (passive resistance is often partly omitted as a relatively large displace-
ment is required to mobilise this resistance fully). For this mechanism (see Figure 1.8a) a
safety factor of F
s
= 1.5 is generally used.
1.3.4 Moment equilibrium of the wall
The sum of the resisting moments about the toe must exceed the sum of the overturning mo-
ments. (mobilisation of passive forces of the soil in front of the wall is usually ignored in the
calculation because considerable rotation is required to obtain resisting forces). For this mecha-
nism (see Figure 1.8d) a safety factor F
s
= 2.0 is generally used.
1.3.5 Moment equilibrium of the wall and the soil as a unit
In addition to the assessment of the moment equilibrium for stability of the retaining wall de-
scribed above, the stability of the surrounding soil mass has to be assessed. The stability of this
soil mass including the wall should be assessed by considering the several hypothetical circular
slip surfaces underneath the retaining wall. (see Figure 1.8b). This takes the form of a slope
stability problem.
For this mechanism a safety factor of F
s
= 1.3 to 1.5 is generally used.
Project number
415660-21 v04
Date
December 2007
Page
8 of 279
1.4 Sheet pile walls
1.4.1 Types of sheet pile walls
A retaining wall is a stiff structure, which does not deform. A wall of sheet piles is relatively
flexible, which may lead to horizontal deformations. A sheet pile wall consists of a series of
sheet piles, which are driven or vibrated into the soil. Sheet piles can be made of wood, steel or
concrete. The length of the sheet piles is always larger than the retaining height (see figure
4.10). The part of the sheet pile wall that is embedded in the soil can mobilise passive earth
pressure (also called passive resistance) which is necessary for stability purposes.
Sheet pile retaining walls are classified as cantilever, anchored or strutted types (see figure
4.10a, 4.10b and 4.10c, respectively). Cantilever sheet pile walls can be used when the retain-
ing height is relatively small, struts are often used to support sheet piles in trenches.
(a)
anchor
strut
strut
(b) (c)
1.4.2 Stability of a sheet pile wall
For the stability of a sheet pile wall three conditions have to be satisfied:
- the sum of the forces in the horizontal direction must be zero
- the sum of the forces in the vertical direction must be zero
- the sum of the moments about any point must be zero.
Figure 1.9 shows an anchored sheet pile wall and the forces that act on it:
- active earth pressure force P
a
- anchor force T
- passive earth pressure P
p
.
There should be equilibrium of moments. So about point B the moment equilibrium requires
that 0 = +
p BD a BC
P l P l which results in the required sheet pile length. From the horizontal
force equilibrium the anchor force can be derived as P
a
= P
p
+T. Since, in this case, no addi-
tional loads in vertical direction are present; it can be assumed that the condition of vertical
equilibrium is stratified. This situation with a minimal embedded length is called free earth
support where the tow is allowed to translate and rotate.
Figure 1.9 Cantilever wall (a), anchored wall (b) and wall with struts (c)
Project number
415660-21 v04
Date
December 2007
Page
9 of 279
Figure 1.10 shows a sheet pile wall with an embedded part, which is required to obtain a stable
construction. The length of the embedded part should be sufficient to satisfy the demands for a
stable condition. Under the influence of the active pressure the embedded portion moves to the
left and the passive earth pressure is mobilised over the entire embedded length. For an ulti-
mate limit state analysis generally a factor of safety is included by reducing the passive earth
pressure to an allowable value of P
p;allowable
:
5 . 1
allowable p;
p
P
P =
If the penetration depth of the sheet piles is sufficiently increased, the bottom part of the em-
bedded portion will practically not deflect to the left or will even move slightly to the right.
The bottom part is then able to mobilise passive earth pressures on the right hand side of the
wall. This situation is indicated as fixed earth support.
K
p
xo
,
v,l
- K
a
xo
,
v,r
M
max
a
c
F
a
anchor rod
moment diagram
C
B
A
K
a
xo
,
v, r
D
E
M
min
E
E
Example of a cantilever sheet pile wall
The penetration depth of a cantilever sheet pile wall has to be determined for a retaining height
h = 4 m. The subsequent soil parameters can be used:
- angle of internal friction |' = 25
- angle of wall friction o '= 0
- coefficient of active earth pressure K
a
= 0.41
- coefficient of passive earth pressure K
p
= 2.5
- unit weight of soil = 16 kN/m
3
.
Figure 1.10 Resultant earth pressure diagram for free earth support
T
Pa
Pp
Project number
415660-21 v04
Date
December 2007
Page
10 of 279
passive earth
pressure
P
p
resultant
earth pressure
h
u
x
a
P
P
c
active earth
pressure
A
B
C
h
t
o
t
o
In Figure 1.11 the penetration depth is t
o
. For point C the sum of the moments must be zero:
, ) 0
3
1
3
1
= +
o a o p
t h P t P
in which
2
2
1
o p p
t K P =
, )
2
2
1
o a a
t h K P + = .
For h = 4 m we obtain:
, ) , )
, )
m. 82 . 4
0 50 . 12 38 . 9 34 . 2
0 4 09 . 1 67 . 6
0 4
3
1
4 16 41 . 0
2
1
3
1
16 5 . 2
2
1
2 3
3
0
3
2 2
=
=
= +
= + +
o
o o o
o
o o o o
t
t t t
t t
t t t t
For t
o
= 4.82 m the passive and the active force can be calculated:
kN/m 6 . 464 82 . 4 16 5 . 2
2
1
2
= =
p
P
, ) kN/m 2 . 255 82 . 4 4 41 . 0
2
1
2
= + =
a
P
As P
p
is not equal to P
a
, it can be concluded that another horizontal force is required to enable
equilibrium of horizontal forces. That force is P
c
(see figure 4.12), which is equal to:
kN/m 4 . 209 =
a p
P P .
The active and passive earth pressures in Figure 1.11 are a schematisation of the horizontal
loading shown in Figure 1.12. When the sheet pile wall is subjected to active pressures, the
Figure 1.11 Design of cantilever wall
Project number
415660-21 v04
Date
December 2007
Page
11 of 279
wall rotates about point F. On the left hand side passive earth pressures are mobilised, which is
also the case on the right hand side below point F. To be able to calculate the installation (or
penetration) depth of the sheet pile wall fairly easily, the bottom part of the earth pressure dia-
gram of figure 4.13 is replaced by the line load P
c
in figure 4.12. However, P
c
is not a line load
but the resultant of the passive earth pressures that act on the bottom area of the wall. Thus the
calculated penetration depth t
o
has to be increased by an additional depth A t
o
. This additional
depth can be determined by:
, ) , )
m 60 . 0
5 . 2 82 . 4 4 16
4 . 209
=
+
=
+
= A
p o
c
o
K t h
P
t
The additional length A t

o
(without safety factor) is sometimes approximated by 0.2 x (see
Figure 1.11). In practice most of the time the length of the sheet pile wall is taken L = h+1.2t
0
to include some safety. The practical application of cantilever walls is for economical reasons
limited to retaining heights of approximately 4 m as relatively long and strong sheet piles are
required.
A
B
C
F
E
D
h
d
resultant
earth
pressure
1.4.3 Anchorages and struts
Sheet pile walls designed as cantilever wall usually undergo relatively large lateral deflections.
As the lateral support of the cantilever wall is supplied by the passive earth pressure exerted on
the embedded portion, installation (or penetration) depths can become quite high. The penetra-
tion depth of a sheet pile wall can be reduced by the installation of a support or anchorage sys-
tem.
The following support or anchorage systems can be distinguished (see Figure 1.13):
- anchors with concrete slabs or continuous anchor walls
- raking pile anchorages
- ground anchors
- struts.
Figure 1.12 Resultant earth pressure diagram for cantilever wall
Project number
415660-21 v04
Date
December 2007
Page
12 of 279
[a] [b] [c] [d]
Anchor slabs
The location of an anchor slab should be selected in such a way that a maximum anchor force
can be mobilised. The minimum required length can be determined according Figure 1.14. In
the first place the anchor slab should be placed outside the zone that causes active earth pres-
sures on the sheet pile wall. Furthermore, the passive zone in front of the anchor slab should
not intersect the active zone behind the sheet pile wall. The minimum required length becomes
therefore equal to l
AB
+ l
BC
in Figure 1.14. Surcharges behind a sheet pile wall and behind an
anchor have a negative effect on the wall if they are present on top of the so called active
wedges Figure 1.14. In Figure 1.14 there is also an area given (on top of the passive wedge of
the anchor) where one should not take a surcharge in account as it has a positive effect on the
strength of the anchor After the dimensions of the sheet pile wall and the anchorage system
have been established, the overall stability of the whole sheet pile structure including the an-
chorage system has to be checked. This is not further dealt with here.
(45 - )
2
(45 + )
2
h
w
ha
key - nut
A B C
surcharge no surcharge surcharge
Ground anchors
A flexible anchorage system can be constructed with ground anchors. For the installation lim-
ited space is required and therefore they are often used for sheet pile structures in built-up ar-
eas. The principle of a ground anchor is fairly simple (see Figure 1.15). After the installation of
a sheet pile wall, excavation in front of the wall is carried out to the level where the anchor
rods have to be connected to the wall. At that level, holes are cut in the sheet pile wall. With a
Figure 1.13 Anchor slab (a), raking pile anchorage (b), ground anchor (c) and a strut (d)
Figure 1.14 Anchored sheet pile wall
Project number
415660-21 v04
Date
December 2007
Page
13 of 279
drilling rig a casing is installed through a hole in the wall to the required depth (a). Upon reach-
ing that depth a steel anchor tendon (consisting of one steel bar or one or more strands) is in-
stalled (b). Finally the casing is connected to a grout pump, where after the casing is filled with
grout using a pressure of 1 to 2 N/mm
2
(c). In this way an anchor is formed with a length of 3
to 5 m (fixed anchor length). The remaining part of the space that becomes available after re-
tracting the casing is filled with grout under a low overpressure (free anchor length (d)). The
anchor is supposed to transfer its load to the subsoil over the fixed length of the anchor.
drilling shoe
casing
casing
rejected
free length
free length
fixed
anchor
length
avoid
harcracks
(a) (b)
(c) (d)
Factors which determine the location and the length of a ground anchor are:
- the soil conditions; the soil should be able to provide sufficient pull-out resistance
- groundwater level; the location where an anchor is fixed to the sheet pile wall should pref-
erably be located above the groundwater level
- the shape of a sheet pile structure
- the presence of foundations, ducts, sewers, etc.
- the magnitude of the required anchor force and the spacing of the individual anchors which
can provide this force
- the location of property boundaries.
The pull-out resistance of the fixed length of a ground anchor depends on:
- soil type and density (in the Netherlands the fixed anchor length is usually located in a
sand layer)
- the effective stress level
- the surface area of the fixed length of a ground anchor.
After the pull-out force for each ground anchor has been determined, it is still necessary to
analyse the overall stability of the sheet pile wall with its anchorage.
Figure 1.15 The installation of a ground anchor
Project number
415660-21 v04
Date
December 2007
Page
14 of 279
1.5 Reinforced earth
A retaining wall consisting of reinforced earth is shown in Figure 1.16. The vertical wall con-
sists of reinforced concrete facing panels. Each facing panel is attached to a metal anchor strip.
The construction of such a wall starts with placing a row of facing panels. Then a metal strip is
attached to each panel, with each strip extending backwards on the horizontal surface of the
backfill. A metal strip has to be long enough to be able to counterbalance the active earth pres-
sure which acts on a panel (see Figure 1.17). Subsequently, a new layer of backfill is placed,
where after the next row of facing panels is installed.
o
active zone passive zone
lean concrete base
[a] cross section
= (45- )
1
2
1.50
1
.
5
0
[b] front view
dowel pvc tube
Instead of concrete panels and metal strips also geotextiles are used for the vertical face and the
anchorage.
The soil mass which is intersected by the anchor strips acts as a gravity wall, but the reinforced
earth structure is more flexible and generally cheaper than a gravity wall of reinforced con-
crete.
Figure 1.16 Principle of reinforced earth
Project number
415660-21 v04
Date
December 2007
Page
15 of 279
panel side
active zone
resisting zone
M N
strip
t
t
max.
t
O
t
O
< 0.75 max
t
Apart from the internal stability (the stability of each separate element of the wall), also the
external stability must be satisfied. The external stability is analyzed in the same way as is
done for a gravity wall in the following way:
- check the bearing capacity of the subsoil
- check whether there is enough resistance against horizontal sliding of the elements
- check the global stability, e.g. by means of a circular slip circle that goes completely un-
derneath the wall.
Figure 1.17 Shear stresses along a metal anchor strip
Project number
415660-21 v04
Date
December 2007
Page
16 of 279
2 Sheet pile walls Design, Safety and Quality
2.1 Comments
As for nearly all geotechnical constructions, the realisation of a sheet pile wall for temporary as
well as for permanent facilities requires special attention for safety and quality aspects. In this
chapter both parts, as well as the design process will be discussed. It must be remarked that the
design method, proposed in this handbook is rather green and therefore must be applied criti-
cally.
This is especially the case for the safety testing procedure and possibly also for the serviceabil-
ity of sheet pile walls, that are heavily loaded or loaded in a special manner. In such cases it is
recommended to, if possible, carry out also a probabilistic or semi-probabilistic analysis, based
on the reliability-index from 2.4.1.
If a sheet pile wall retains exclusively, or almost exclusively water or another liquid, according
to NEN 6702; 1991 TGB 1990 Belastingen en vervormingen [NNI, 1991], it does not be-
long in the category of geotechnical constructions, when it comes to the determination of the
loads. For the analysis of the ultimate limit state 1, the load factor to be applied on liquid pres-
sure is:
f;g
= 1.2. Suppose that in the design of a sheet pile, which is almost exclusively loaded
by water pressures, a load factor
f;g
= 1.0 is used, which is suitable for geotechnical construc-
tions, in stead of
f;g
= 1.2. In nominal sense there can hardly be a margin between the occur-
ring moment in the sheet pile wall and the maximum moment of resistance, when a plastic
hinge is formed in the sheet pile wall.
The methods mentioned in this handbook can also be used for the design of water retaining
sheet pile walls without difficulties or restrictions.
2.2 Design process
2.2.1 Introduction
Designing, in the broadest sense of the word, means the making of and the weighing of plans
for possible solutions for technical problems. Based on this, a proper design for the construc-
tion can be chosen. Not all of the possible solutions are relevant here, only the ones of which
has been established in an intelligent manner that they are technically and economically attain-
able. The solutions are of course never equal in all respects, but must all meet the performance-
demands made by the investor.
In order to come to a well-weighed decision, it is desirable that the different solutions are
tested for the most relevant criteria such as safety and usefulness, environment, costs, feasibil-
ity and construction time, by objective assessment-criteria. Such a multi-criteria evaluation
generally requires a rather profound and thorough research of all of the possible solutions.
After such a thorough weighing, it may even appear that not solving the problem is the best
solution, for example because the damage to the environment is unacceptable for all solutions.
Project number
415660-21 v04
Date
December 2007
Page
17 of 279
Designing in the sense mentioned above almost always concerns total projects that serve an
economical goal in one way or another. In the past few years however, solving environmental
problems has played a more and more important part.
In total projects, earth-retaining structures are merely components; but these components can
be important.
Besides, there is a design process for earth-retaining structures where practically the same cri-
teria should be tested as for the total project, but on a smaller scale.
The design process is a dynamic process. This means that from the moment that initiatives are
taken up to the completion of the work, there is a continuous interaction between what is
documented in the contract documents and what is really happening. Even a structuring of the
design process using computerised multi-criteria analysis can only partially control this dy-
namic process and must therefore stay open for the continuous interaction mentioned before.
In this handbook, the sheet pile wall takes the centre stage. This means the definition of the
design process is fenced in even further. On the level of the total project, many often com-
pletely different solutions are possible. Descending to the components of the project, the only
things that are left for the designer are the economical and the technical optimisation of the
component concerned.
If designers want to do their job well, they must make high quality-demands for the means and
the methods that are used for testing the behaviour of the sheet pile walls. This generally means
that research, calculation, design and details and also the execution must be tested for the ac-
tual state of technology and possibly even for the state of science in the area concerned. Natu-
rally this includes the magnitudes of the risks the product of chance and damage.
If the risk is small, then it is not necessary to use very advanced design methods. Usually it is
more obvious to choose safe starting points and to make the design based on these as well as on
the experience that has been acquired while building sheet pile walls under similar circum-
stances.
The state of technology concerning soil investigation, the determination of the soil parameters
and the calculations, is usually fixed in standards, guidelines, handbooks, etc. The designer of
the sheet pile wall should be certain of the standards and rules in the field concerned.
For exceedingly complex sheet pile walls, where high demands are made for the deformations,
traditional methods will almost always be deficient. More advanced methods will have to be
called upon.
2.2.2 Description of the design process
The goal of the design process is to come to a technically, economically and spatially optimal
design of a sheet pile wall that meets the demands made in the programme of demands, which
is to be discussed later on, by a structured process. In this process a number of phases can be
discerned. At the end of each phase, there is a test for the established programme of demands,
on the basis of which a decision is made concerning the following steps.
For larger projects, this decision will be made in consultation with the investor and based on
the information and advice of the designer.
If the result of a design phase does not yet meet the demands made or has a too large safety
margin, this phase will be repeated with another sheet pile wall profile, another sheet pile wall
length or another anchor design.
Project number
415660-21 v04
Date
December 2007
Page
18 of 279
It can also appear that the programme of demands must be adjusted. Principally, a design proc-
ess is a cyclic process which optimises the output of the process.
In 2.5 the quality assurance aspects concerning the design process are treated. An important
aspect is the formalisation of the structured design procedure in which the following must be
clearly fixed: Who does What, and When is it done; Who, What and When checks this and
which means will be used to solve the technical problems?
Initiation phase
After the client or the investor has determined that a sheet pile wall is required, this person
shall take the initiative to conduct a study from which the attainability of the project in the
given situation will become apparent. If the project is attainable, then a global programme of
demands shall be established, based on the results of the study. Based on this programme of
demands, the client or the investor will decide if the project should be continued or if it must
stop. If the decision is made to continue, the definition phase is started.
Definition phase
In the definition phase, the project is analysed in great detail. The project is defined based on
this analysis and the demands and wishes formulated by the investor. The designer and the
investor set up the definite, functional as well as operational, programme of demands.
Based on the information that has been collected, the starting points for the next phase are
fixed.
Preliminary design phase
In the preliminary design phase a number of possible solutions are developed into global pre-
liminary designs of the earth retaining structure.
The designs should be worked out in such a manner that a well-founded choice between the
generated solutions is possible. For this, all solutions should be worked out to the same level.
Data of the soil should be available for the global design calculations. If there is no information
available, then a soil investigation must be carried out. For the global design, all aspects that
can determine the choice will have to be taken into consideration. Strength, stability and de-
formations must be tested according to the same design criteria as the definite design. Also,
attention must be paid to the manner of execution, construction time and construction costs.
A choice is made between the different solutions based on the testing criteria for the total con-
struction, such as technical practicability, construction costs, construction time and durability.
The chosen solution is worked out even further in the design phase.
Design phase
In the design phase the definite design of the sheet pile wall construction is made. It will gener-
ally be necessary to conduct an extra soil investigation. An estimate of the costs and a provi-
sional programme for the execution are also made.
In connection with quality guarantee, the design phase is a very important phase, where the
designer must be able to show that he is working or has worked according to the state of tech-
nology at all times. Also a number of fixed internal as well as external control procedures must
be carried out.
Project number
415660-21 v04
Date
December 2007
Page
19 of 279
Preparation phase
In the preparation phase the tender-specifications and the construction drawings are made.
These serve as contract documents of the agreement between the building contractor and the
investor. Before the tendering takes place, the required permits are applied for.
The designer can of course play an important role in the award of the contract work, especially
in relation with the presentation of possible alternatives.
Execution phase
After the work has been awarded and the contracts have been signed, the work is started. In
many cases the designer also directs the execution of the work. On behalf of the investor he
supervises the quality of the work and guards over the costs and the construction time up to the
definite completion of the work.
If the work is executed with a guarantee of quality, the supervision of the designer will be lim-
ited to carrying out the checks, set in the quality plan of the building contractor. These checks
are made for points which are specified in the execution plan in advance of the execution.
These checks can also be made by audits, executed by independent external experts. In the
case of quality guarantee the designer is not the supervisor. The building contractor is com-
pletely responsible for the testing. Generally, after completion of the sheet pile wall, an as-
built document must be present, which contains all the realised details of the wall. This in-
cludes directions and possible limitations concerning the use and the maintenance.
User phase
Usually in the beginning of the user phase, a number of tests are conducted for the behaviour of
the construction. In the case of quality guarantee these must be explicitly fixed in the quality
plan of the work. Because the starting points of the design are concerned, the designer will play
an important role in the testing, especially if the designer is also the supervisor. This so-called
monitoring can cover a period which goes on beyond the definite completion of the work.
Figure 2.1 shows the schematised design process for a sheet pile construction.
During the total design process, a number of different information streams are required in order
to inform the concerned parties adequately and in time. Besides the design process, the figure
shows which parties are involved in the design process and when transfer of information
should take place.
The following parties are usually involved in the design and the execution:
the investor;
the designer: this can be a consultancy, engineering firm, building contractor and
sometimes the supplier of the sheet pile wall;
the specialists, such as the soil mechanical adviser for example;
the permit distributors, authorities and possibly the insurers;
the supervisors, if these are not the designer or the building contractor;
the building contractor and the subcontractors.
Project number
415660-21 v04
Date
December 2007
Page
20 of 279
Figure 2.1a Schematisation of a design process and schematisation of the relations for the exchange
of information and controlling.
Project number
415660-21 v04
Date
December 2007
Page
21 of 279
Figure 2.1b Schematisation of a design process and schematisation of the relations for the exchange
of information and controlling (continued).
Project number
415660-21 v04
Date
December 2007
Page
22 of 279
Figure 2.1c Schematisation of a design process and schematisation of the relations for the exchange
of information and controlling (continued).
Project number
415660-21 v04
Date
December 2007
Page
23 of 279
Although for different contract forms the same parties are involved in the design process, the
construction process and the management and maintenance, the mutual relations can be rather
different, and therefore also the information exchange.
If the work is done under guarantee of quality, the relation scheme is especially about the for-
malisation of the relations between the officials of the parties.
2.2.3 Programme of demands
Introduction
The programme of demands is a document that serves as reference for the design and the exe-
cution of the sheet pile construction. It is established in close consultation between the inves-
tor, the designer and possibly also the building contractor. It must contain precisely specified
demands concerning use, risks for the surroundings and the execution. This also includes the
demands made by the authorities with respect to working conditions, environment, health and
safety.
The programme of demands must also include directions about specific loads, load situations,
design limitations and the tests that must be executed.
The programme of demands for a sheet pile construction is based on the demands that are made
in relation with the function(s) provided and the use of it. Most importantly, the sheet pile wall
must sufficiently provide in these functions, and must keep doing so throughout the entire life-
span. The demands formulated from this requirement are called functional demands.
The functional demands for sheet pile constructions are derived from one or more of the fol-
lowing functions:
the earth retaining function, where the sheet pile wall provides the possibility of a sudden
difference in ground level;
the water retaining function, where the sheet pile construction provides the possibility of a
sudden difference in water level;
the bearing function, usually combined with one or more of the other functions, where the
sheet pile construction resists the external horizontal and vertical forces;
the protective function, often also combined with other functions, where the sheet pile wall
is used for the following purposes;
prevention of erosion due to water movements;
guiding water movements;
guiding ships;
prevention of deformations and settlements of areas nearby;
limitation of groundwater level lowering and its results on existing foundation
constructions due to negative shaft friction;
isolation of an area in order to prevent soil pollution from spreading;
water tightness to prevent the leakage of drinking water from a drinking water area to
a deep excavation.
Project number
415660-21 v04
Date
December 2007
Page
24 of 279
In the programme of demands the functions mentioned above will have to be translated into
formal data concerning:
the exact location of the sheet pile construction;
the retaining height or excavation depth;
the designed water level difference that must be retained;
the maximum leakage in total, or per unit of sheet pile wall area;
special loads that are not prescribed in standards or guidelines;
maximum deformations of the sheet pile wall and the neighbouring area;
the hydraulic roughness in connection with the water flow.
Apart from the functional demands, there are also operational demands, which, as was men-
tioned earlier, concern the realisation and the use of the project in which the sheet pile wall
fulfils certain functions.
These operational demands intend to realize optimal security with regard to the functioning of
the project with the sheet pile wall, to the investors satisfaction. The chance of an extreme load
situation occurring during the execution phase, which hardly ever takes longer than a year, is
much smaller than the chance of such a situation occurring in the user phase, which can take
over 50 years. Nevertheless, it is advised to let the load situations weigh as heavily during the
execution as they do during the lifespan of the structure when designing. This means that the
partial factors given in table 2.5 in 2.4 are valid for the representative values of the loads and
the representative material properties.
Operational demands are coupled with rules for safety, the serviceability and the effect on the
environment and the surroundings during the construction and the use over the full lifespan of
the sheet pile construction.
Durability, management and maintenance are important in connection with the lifespan. This
means that, besides regular inspections of the situation, measurements must be made during the
entire lifespan in order to check if the construction still meets the demands made. The follow-
ing activities are carried out during this so-called monitoring:
measuring the displacements of the sheet pile wall, of the neighbouring area and of
possibly neighbouring constructions;
measuring ground water levels in the surroundings of the sheet pile wall;
determining the extent in which chemical pollution moves out from the isolated area into
the surrounding soil.
The rules concerning the safety, the serviceability, the environment and the working situations
are fixed in ministerial resolutions and in laws or in local regulations, which are part of the
ARBO-wet. The existing national and European standards play an important part in this.
Naturally the investor can make additional demands concerning the aspects mentioned previ-
ously.
In the programme of demands for the sheet pile construction, these operational demands must
also be converted into formal concrete directions.
Design limitations can result from the project itself, but also from circumstances during the
execution.
Project number
415660-21 v04
Date
December 2007
Page
25 of 279
Some examples of design limitations are:
the construction costs;
the construction time;
demands made concerning the water discharge and shipping;
the choice of the anchoring system due to constructions present;
the soil condition due to which, for example, special facilities are required in order to
install the sheets till the required depth.
There are also limitations and conditions which are related to the surroundings of the sheet pile
wall. Generally these concern the following aspects:
the area available;
fitting into the landscape and surroundings aesthetically;
the noise and vibration hinder during execution.
The demands mentioned are usually fixed in the building permits and therefore do not have to
be mentioned specifically in the programme of demands. The investor will usually take care of
all the required permits.
2.2.4 Norms, guidelines and recommendations
Norms, Guidelines and Recommendations are important documents for designers and building
contractors. They are linked to the Bouwbesluit.
Norms, Guidelines and Recommendations go beyond the Bouwbesluit, which is only aimed
at the granting of the permit for the construction of the project. The Bouwbesluit only organ-
ises business of public law with respect to the realisation of the building project. Standards also
show the state of the technology on the area concerned and are important under private law for
realizing the highest possible level of quality within the programme of demands.
Next an inventory is made of the most relevant standards and guidelines from the Netherlands,
Europe, Germany and Great Britain that are related to sheet pile constructions.
The Netherlands
National Standards:
NEN 6701; 1991 TGB 1990 Namen en symbolen voor grootheden;
NEN 6702; 1991 TGB 1990 Belastingen en vervormingen;
NEN 6720; 1991 TGB 1990 Voorschriften Beton Constructieve eisen en
rekenmethoden (VBC 1990);
NEN 6740; 1991 TGB 1990 Geotechniek, Basiseisen en belastingen;
NEN 6760; 1991 TGB 1990 Houtconstructies Basiseisen Eisen en
bepalingsmethoden;
NEN 6770; 1991 TGB 1990 Staalconstructies Basiseisen en basisrekenregels
voor overwegend statisch belaste constructies.
Project number
415660-21 v04
Date
December 2007
Page
26 of 279
Guidelines for sheet pile walls:
Rijkswaterstaat, Richtlijnen Directie Bruggen voor het ontwerpen van betonnen
kunstwerken, versie II, rapport nr. 14;
Richtlijnen voor funderingen van gebouwen - RFG 1985 (Deel II);
Houten damwanden en meerpalen (CHR-brochure 78-1).
Europe
CEN-ENV: Eurocode 7.
The European guideline, concerning minimum regulations with regard to safety and
health for temporary and mobile construction sites, is implemented by changing the
ARBO-wet. The regulation allows member states to make choices concerning policy
with regard to giving responsibility to the investor and the to be assigned coordinators
concerning safety and health. It is particularly important that the regulation gives the in-
vestor responsibility concerning safety and health.
Germany
Empfehlungen des Arbeitsausschusses Ufereinfassungen (EAU);
Empfehlungen des Arbeitskreises Baugruben (EAB);
DIN 4085 Baugrund Berechnung des Erddrucks fr starre Sttzwnde und
Widerlager.
Great Britain
British Standard Code of Practice for Maritime structures, BS 6349, part 1 General
criteria;
British Standard Code of Practice for Maritime structures, BS 6349, part 2 Design of
quay walls, jetties and dolphins.
2.2.5 Construction time
The construction time can play an important role in the choice of the construction type and the
method of execution.
According to the UAV, the date on which the work is started is taken to be the fifth working
day after the day on which the building contractor has acquired the job. The construction time
lasts up to the moment of completion of the work, unless another deal has been made.
The date of completion is mostly fixed in the tender-specifications, according to the Standard
RAW determinations (see paragraph 8 of the UAV)):
a. using a number of workable days;
b. using a number of calendar days, calendar weeks or calendar months;
c. by noting the date on which the work must be completed.
The following aspects influence the total construction time of a project:
Project number
415660-21 v04
Date
December 2007
Page
27 of 279
Mobilisation
Mobilisation concerns the furnishing of the working area, the supply of material and the prepa-
ration of its transport. On average, the mobilisation may take a number of weeks. If special
equipment is required, the mobilisation time can last up to some months.
Work preparations
These are the necessary preparatory activities such as making a planning, making deals with
the suppliers, making detailed calculations and drawings, etc. This usually starts at the same
time, or possibly even sooner than the mobilisation, and continues parallel to the execution.
Supply time of the materials
The supply time of the different materials depends, among other things, on the size of the lot, if
conservations are applied in the factory or not, the situation of the market, standard or non-
standard products, special treatments, etc. For standard products the manufacturers have a spe-
cial order of fabrication, so the waiting time can be longer than usual. If the investor supplies
the sheets, the supply time will generally not be a problem for the construction time. If neces-
sary, the supply time can be shortened by using another appropriate sheet profile that can be
supplied in a shorter period of time. It may be necessary to adjust the design of the sheet pile
construction so the available profile will fit.
The execution of the work
The choice of the material and the possibilities to start the work in more than one place at the
same time, have great influence on the construction time and must therefore be included in the
design process. They can indirectly influence the choice of the sheet type.
The construction time can also be decreased by using temporary facilities during the execution,
such as a driving frame, a mobile bridge between sheet pile walls (in dutch: traverse) and a
building pit.
In order to install a sheet pile wall in wet conditions, a temporary fill can be made with sand, so
the work can be executed dryly. This can also decrease the construction time considerably. Due
to this method special circumstances and loads may occur, which have to be taken into consid-
eration in the calculation of the design. They can influence the spring constants of the soil in
the different construction phases and in the user phase.
If there are periods in which the work has to be stopped, this can have great influence on the
construction time. These temporary pauses can be caused by environmental demands, factors in
the surroundings, traffic, and specifically by demands made by the user or the person who
grants the permits.
If there is good consultation between the designer, the contractor, the investor and the authori-
ties that grant the permits, the design can be altered or temporary facilities can be applied in
time, so the negative influence on the construction time can be partially or even completely
cancelled.
For a sheet pile construction, the presence of unexpected obstacles in the soil can have great
influence on the both the quality of the work and the construction time. Therefore it is neces-
sary to conduct a very detailed soil investigation in the design phase, so that the chance of en-
countering an obstacle in the soil is minimal.
Project number
415660-21 v04
Date
December 2007
Page
28 of 279
Obstacles can be the following:
remains of old foundations and pipes;
erratic boulders such as boulders in stiff clay, boulder clay and moraines;
remains of trees;
explosives.
2.2.6 Costs of the project
In the design process the construction costs are nearly always an important boundary condition.
Naturally the designer must find a balance between construction costs and quality. This is an
optimisation process, in which the founding costs, the maintenance and reparation costs and the
insurance costs are involved. The minimal costs over the entire life span must be found. For a
decreasing chance of construction failure, the founding costs increase but the costs for mainte-
nance, reparation and insurance decrease. The minimum of the total curve resulting from the
optimisation test shows the optimal point, where the total costs are minimal.
In figure 2.2 this is shown schematically for the founding costs and the maintenance costs. In
chapter 6 of the original CUR166 edition, Supervision and maintenance of sheet pile construc-
tions, this subject is treated in more detail.
Figure 2.2 Optimisation of total costs of permanent sheet pile wall.
But this philosophy is hardly ever applied for one hundred percent. Usually other factors also
play a part in the choice for material, equipment and method of execution.
Optimisation of costs does not automatically mean that the construction is optimised in techni-
cal sense. A good designer will always have to investigate how sensitive the design is regarding
costs
optimized
costs
development costs
maintenance costs
total costs
section modulus sheet pile wall
Project number
415660-21 v04
Date
December 2007
Page
29 of 279
the safety margin in the area around the optimisation point. If the chance of failure decreases
strongly for a slight strengthening of the construction against low costs, the designer will most
likely choose to make this slight adjustment.
For sheet pile constructions the chance of failure will not be very sensitive to the choice of the
weight and strength of the sheet pile profile.
The construction costs are determined for a great deal by the following types of costs:
wages;
buying or renting the materials and the means of conservation;
costs for renting the materiel, the supply and removal of equipment and materials;
costs of the construction site: accessibility, supply roads, supporting constructions,
excavations, temporary fills and special measures concerning hinder for the surroundings;
approximate estimates, costs concerning permits, insurance costs;
design costs and management costs;
costs of the soil investigation and possible checks carried out during and after the
construction;
costs for transportation and storage of the excavated soil, which can be extremely high for
polluted soil;
profit and risk cover.
The costs are estimated using unit prices from the following sources of information:
internally made documentation of calculation prices based on post calculation of similar
projects that have already been executed.
directives and documentation, for example:
Misset Bouw: GWW-costs foundation techniques;
Bouwmarkt: actual construction costs information;
COBOUW: companyinformation and -documentation.
suppliers of sheet piles and other materials.
Paragraph 8.1.14 of the EAU 1990 gives a global indication of the cost distribution of a sheet
pile construction over the cost entries supply sheet piles, supply anchors and girders and
execution. This division only concerns a steel sheet pile construction. The costs for dewater-
ing, filling out, excavating, dredging and paving are not considered here and must be estimated
separately.
The division of costs for an anchored steel sheet pile wall is the following:
supply of the sheet pile wall: approximately 45 %
supply of the anchorage, anchor wall and girders: approximately 25 %
execution costs: approximately 30 %
For an unanchored steel sheet pile wall:
supply of the sheet pile wall: approximately 60 %
execution costs: approximately 40 %
Project number
415660-21 v04
Date
December 2007
Page
30 of 279
The supply costs are an important part of the total costs for the realisation of the sheet pile
wall. This means that it is worth the trouble to make strict designs in order to come to a most
economical design, of course taking into account the demands for performance made in the
programme of demands.
Investments made in soil investigations and in the constructive design are easily profitable and
give a good impression of the sensitivities and the insecurities of the design.
2.3 Description of alternative constructions
2.3.1 Introduction
In this paragraph alternative earth retaining constructions as well as sheet pile walls will be
treated. Principally, the alternative constructions have the same functions as a sheet pile wall.
Embankments will not be included in this discussion.
The description of the sheet pile wall is given in paragraph 2.3.2; the alternatives such as for
example the diaphragm wall or the pile wall will be described in paragraph 2.3.3.
For every alternative construction some general information will be given. Also the advantages
as well as the disadvantages compared to a sheet pile wall will be discussed.
In this handbook the alternative constructions will not be treated outside this paragraph. Natu-
rally a number of design methods or parts of them are also applicable in the case of the calcula-
tions for the alternative solutions.
With the exception of the application for reinforced soil, the anchors will not be treated in this
paragraph. Anchor constructions are treated in paragraph 5.5.
The different main functions that the sheet pile wall or alternative construction can have are
mentioned in 2.2. An overview of the different main functions and sub-functions of the sheet
pile wall and the alternatives is given in table 2.1. This table also shows if the function is tem-
porary or permanent. Table 2.2 shows the applicability of the sheet pile wall and the alterna-
tives for different structure types. An overview of the different main functions and sub-
functions of the earth retaining constructions for the constructions mentioned in table 2.2 are
given in table 2.3.
Table 2.4 shows which anchoring methods are suitable for the different earth-retaining struc-
tures.
For each alternative the following points will be discussed:
the special characteristics;
the material used and an indication of the size of the main parts;
the normal construction method and possible alternative execution methods;
the specific properties;
the advantages and disadvantages.
Project number
415660-21 v04
Date
December 2007
Page
31 of 279
Table 2.1 Main functions and sub-functions of sheet pile walls and alternatives.
Project number
415660-21 v04
Date
December 2007
Page
32 of 279
Table 2.2 Applicability of sheet pile walls and different alternatives for certain constructions.
Project number
415660-21 v04
Date
December 2007
Page
33 of 279
Table 2.3 Main functions and sub-functions of soil retaining structures in certain constructions.
Project number
415660-21 v04
Date
December 2007
Page
34 of 279
Table 2.4 Anchoring methods for different types of soil retaining structures.
anchoring
construction alternatives
soil anchor
+ anchor-
construct.
screw-
anchor
grout-
anchor
screw-
anchor
tension
piles
strutting
sheet pile walls:
steel +++ + ++ ++ + +++
wood ++ + +
concrete ++ + ++ ++ + +
pile wall ++ + +
diaphragm wall ++ + ++
composed walls:
Berlin wall ++ + ++
combined wall ++ ++ +++ ++
PSP-wall/HZ-wall ++ ++ ++ ++
cofferdam ++
cellular wall +++
terre arme
foils + soil body
gravity walls:
small quay walls
L-walls
different alternatives:
betonite-cement wall
caisson
+++ : frequently
++ : regularly
+ : rarely
2.3.2 The sheet pile wall
Sheet pile walls are vertical elements that are linked by a interlock construction or a groove and
tongue joint and which stand alone. The soil pressures and the water pressures are transferred
to the subsoil and possibly to the anchors due to the stiffness of the sheet pile wall profiles. The
sheet pile wall behaves as an overhanging elastic beam that is clamped in the soil, or as a
beam on two supports; the soil at the bottom and an anchor at the top. The support at the bot-
tom in the soil can be free or it can be either completely or partially clamped. At the bottom the
supporting pressure that is necessary for the balance is mobilised by the passive soil pressure.
The anchorage at the top is made up of a tie rod with an anchor plate at the end and grout an-
chors or tension piles, which transfer the anchor force to the soil behind the sheet pile wall.
The length of the rod is chosen long enough to transfer the load to a soil mass that is not or
only slightly influenced by the sheet pile wall.
For building pits the anchorage is often obtained by strutting on the opposite sheet pile wall.
An anchor or strut is installed in order to limit the displacements and also to obtain an eco-
nomical sheet pile construction.
Project number
415660-21 v04
Date
December 2007
Page
35 of 279
The vertical sheets are mutually attached by a girder (or a wale) or a top beam in order to di-
vide the anchor force uniformly and to bridge possible local variations in the soil condition and
to limit oblique bending.
Sheet piles are supplied in the following materials:
steel: warmly rolled and cold-formed Z-, U-, and H-shaped profiles.
The steel thickness varies between 6 mm and approximately 31 mm. Lengths of up to
31 m are normally available.
Figure 2.3 shows a photograph of the construction of a steel sheet pile wall.
Steel sheet pile walls must meet the demands made in NEN 6770;1991 [NNI, 1991].
concrete: prefabricated sheets with thickness varying between 120 mm to 300 mm.
Concrete sheet piles should satisfy NEN 6720;1991 [NNI, 1991]. But specific demands
regarding the use of concrete for sheet pile walls are not mentioned in this document.
In 5.2.8 and 5.4 this is treated in more detail. Figure 2.4 shows the construction of a
concrete sheet pile wall.
wood: hardwood, such as azob, bangkirai, karra, jarrah and European oak or softwood, such
as European types of pinewood, Douglas and lariks. The latter are naturally less dura-
ble. If they are used for sheet pile walls with an economic life span of 25 years or
more, they must be preserved.
The brochure Houten damwanden of the Centrum Hout [Centrum Hout, 1987] gives
directions for the preservation of different types of softwood, for tolerances and for de-
sign and execution. NEN 6760; 1991 [NNI, 1991] gives demands concerning the
strength for different classes. For this the climatologic circumstances have also been
taken into consideration.
Figure 2.5 shows a wooden sheet pile wall with a girder and an anchorage in the
execution phase.
Figure 2.3 Installation of a steel sheet pile wall.
Project number
415660-21 v04
Date
December 2007
Page
36 of 279
Figure 2.4 Installation of a concrete sheet pile wall.
The sheet piles are inserted from ground level, either separately or with a few at the same time.
They can be installed by vibrating, driving or pressing, possibly in combination with jetting,
fludation or preboring.
Table 2.2 gives an overview of the range of application of steel, wooden and concrete sheet
pile walls. It appears that almost all materials are universally applicable except for environ-
mental applications where wood can hardly ever be used.
For temporary sheet pile constructions wood and steel are most frequently used. Sheets made
of these materials are usually easy to pull from the soil after use, so they can be reused.
If high vertical loads need to be transferred by the sheet pile wall to the soil, relatively thick
concrete sheets can be a good solution. For very high retaining heights, for example 15 m or
more, usually only steel is an option.
Advantages and disadvantages compared to the alternatives
Sheet pile constructions are easy to construct because standardised construction elements
are used.
The construction time is relatively short.
Sheets can be reused. But concrete sheets are hardly ever used for temporary projects.
It is not complicated to install anchors at different levels.
It is not necessary to use a separate trench in order to install the wall. But for L-walls and
caissons this is necessary.
For steel sheet pile walls the maximum vertical bearing capacity of the tip is very small.
But the bearing capacity can be improved in a very simple manner, for example by
welding an enlargement at the bottom of the sheets.
The installation of the sheets usually gives vibration and noise hinder.
Deep firm layers can only be penetrated by combining driving and vibrating with jetting,
fludation and preboring.
At the location of the interlock the sheet pile wall is usually not completely watertight.
There are means available to make the interlock watertight, but these are not always full
proof.
Project number
415660-21 v04
Date
December 2007
Page
37 of 279
Figure 2.5 Installation of a wooden sheet pile wall.
2.3.3 The alternative soil retaining constructions
In the following description of the different alternative systems, the different constructive and
execution aspects will only be treated superficially. More details can be found in literature
[Van Weele, 1987].
COMBINED WALLS
Combined walls can be considered as a special type of a sheet pile wall. They are made up of
steel pipe piles, the so-called primary elements, in between which the sheet pile profiles are
installed. Sheet pile interlocks have been welded to the pipe piles. The piles are available in
every desired length and diameter. In order to obtain economical constructions, the primary
Project number
415660-21 v04
Date
December 2007
Page
38 of 279
elements can be assembled from tubes that are welded onto each other with a wall thickness
that is adjusted to the moment distribution.
In 5.4 more details are given about the installation of a combined wall in the soil. Generally
combined walls are applied in earth retaining constructions where the retaining height is very
large and where substantial loads can act. This is the case for quay walls of sea harbours. The
steel pipe piles have a large resistance against torsion, so they encounter few problems during
heavy driving work. Figure 2.6 shows a combined wall.
Advantages and disadvantages
For an equal capacity for bearing moments, combined walls use less steel than walls that
are made up of many of the same sheet pile profiles.
The construction behaves more stiffly than a sheet pile wall, so fewer levels of anchors
and struts are required.
Combined walls have a great vertical bearing capacity.
It is technically possible to adjust the thickness of the wall to the moment distribution,
over the entire height of the pipe piles.
Penetrating firm sand layers is possible, so high retaining heights can be realised.
Most of the time a girder is not necessary. By installing a concrete prop in the pipe pile it
is possible to fasten the anchor bar to the pipe piles.
A slight batter of the pipe piles increases the risk of declutching during installation.
The sheets in between of the pile pipes can cause problems during heavy driving work.
The thickness of the sheets must therefore also be dimensioned for the loads caused by
driving and vibrating.
Figure 2.6 Installation of a combined wall.
Project number
415660-21 v04
Date
December 2007
Page
39 of 279
PSP-WALLS AND HZ-WALLS
PSP-walls and HZ-walls can also be considered as a special type of a sheet pile wall. They are
closely related to the combined wall mentioned previously. The PSP-wall and the HZ-wall are
made up of steel I-profiles that can be linked together and which are installed either closely
fitting or with spaces between them. The sheet pile profiles are then fitted in these spaces.
Principally the execution of the PSP-walls and the HZ-walls is the same as the execution of a
combined wall. But the single I-profiles have only limited resistance against deformations by
torsion. During driving torsion can occur very easily. By using double profiles this shortcoming
can be remedied.
PSP-walls and HZ-walls have the same application area as combined walls. They also have the
same advantages and disadvantages compared to the sheet pile wall, but the distinction be-
tween retaining function and bearing function is not present.
PILE WALL
A pile wall is made up of a row of non-soil displacing piles, which are formed in the soil and
are made of concrete, cement grout, bentonite cement or injected soil. These piles are placed
next to each other in slightly overlapping positions. The horizontal loads on the pile wall are
taken up by bending moments in the piles. These bending moments are in turn transferred to
the soil via the elastic restraint of the pile in the soil, and possibly also by the anchoring con-
struction. For the piles to be able to take up the bending moments they must be reinforced.
Steel profiles can be used for this reinforcement, or the piles can be pretensioned.
For this type of wall especially screw auger piles or auger piles and screw pipe piles are used.
Drilled piles are mostly used for large diameter piles.
Figure 2.7 shows a pile wall that functions as a support for new buildings. It has been executed
with screw piles.
Pile walls are executed in concrete. But if only dry soil needs to be retained, a row of steel pipe
piles placed next to each other will suffice.
Figure 2.7 Construction of a pile wall.
Project number
415660-21 v04
Date
December 2007
Page
40 of 279
The diameter of the piles can reach up to 2.50 m for drilled piles. When screw piles and similar
systems are used, the usual diameters vary from 300 mm to 800 mm.
For non-soil displacing piles the soil is removed by screwing or boring, after which the hole is
filled with concrete mortar, cement grout or bentonite cement.
If prefabricated concrete piles or steel piles are used, these are practically always inserted by
driving.
The advantage of pile walls over sheet pile walls is that besides their retaining function, they
can also bear high vertical loads.
But generally, when the use of anchors is not possible, a stiff wall formed in the soil is chosen.
This can be the case if, for example, the wall must retain soil right next to an existing construc-
tion. In this case the deformations must obviously remain very small.
If an anchor is used, it can be attached to a reinforced concrete heading beam on the pile heads
or to a steel girder. In this case the heading beam can also function as a foundation beam.
A pile wall can take up both vertical and horizontal loads.
A pile wall can be installed without vibrating and noiselessly.
Execution is possible up to great depths, even in very firm layers.
A pile wall has a high bending stiffness, due to which the horizontal displacements are
small.
Fixed joints with other construction parts are difficult to realise.
Anchoring on a level below the top can produce problems, except in the case of struts.
DIAPHRAGM WALLS
A diaphragm wall is a concrete wall that has been constructed in the soil by excavating. It can
be partially prefabricated. In principle the diaphragm wall has many similarities with the pile
wall mentioned previously. Generally the diaphragm wall offers more possibilities as construc-
tional element in an underground structure.
Diaphragm walls are practically always made of reinforced concrete. The thickness of the wall
varies between 0.4 m and 1.2 m. In certain cases, for example if only ground water needs to be
retained, a hardening cement-bentonite mixture can be used instead of concrete.
During the installation, first of all guiding beams are installed at ground level, between which a
shallow trench is excavated over the length of the diaphragm wall that is to be installed. The
guiding beams also prevent the crumbling away of the side of the trench during the excavation
of the deep trench sections. Between the guiding beams excavation takes place one section at a
time, by excavating with a grab, drilling or fraising. During the excavation the side of the
trench is supported by the so-called bentonite suspension. The stability of the trench that is to
be excavated is taken care of by a number of things. These are the hydrostatic pressure of the
bentonite-slurry in combination with the arching in the soil and the forming of a hard imperme-
able bentonite layer on the side of the trench at the location of the contact surface between the
soil and the supporting fluid.
In relation with the maintenance of the stability during the excavation and the pouring of the
concrete in the sections, the potential of the bentonite fluid must be at least 1 m higher than the
potential of the ground water in the layers that the diaphragm wall cut through.
Normally a self-guiding wire grab or the Kelly-bar are used for the excavation [Van Weele,
1987]. These are used discontinuously. The fraising cutter is used in a more or less continuous
Project number
415660-21 v04
Date
December 2007
Page
41 of 279
process. The mixture of the soil that is cut loose and the bentonite are removed through pipes
in the machine.
When the section is ready the joints and the cages are placed. Next the concrete is poured in
the trench from the bottom upwards in a controlled manner, using the contractor method. At the
same time the bentonite fluid is pushed out.
The diaphragm wall sections are usually not excavated closely-fitting because the concrete in
the recently filled trench has not yet hardened and must be supported.
After a hardening time of at least one day the intermediate or neighbouring sections can be
excavated, of course after the joint moulds (in dutch: voegmallen) have been removed. These
joint moulds can present problems concerning the watertightness. By inserting so-called water
stops this drawback can be removed.
Sometimes prefabricated elements made of reinforced concrete are placed in the trench sec-
tions that are filled with bentonite. In some cases this can offer constructive advantages, but it
is mostly of importance if a remaining clean wall must be obtained in the underground space.
Figure 2.8 shows a photograph of a wire grab for the construction of a diaphragm wall. In fig-
ure 2.9 the method for constructing diaphragm walls is shown schematically.
Besides having an earth retaining and a water retaining function, diaphragm walls can also be
used as permanent elements of the foundation.
They are used for constructions that are partially or completely located below ground level,
such as parking garages, tunnels and deep cellars. In these situations the diaphragm wall is part
of the construction and the foundation of the building.
Figure 2.8 Wire grab making trench-sections for diaphragm wall.
Project number
415660-21 v04
Date
December 2007
Page
42 of 279
Figure 2.9 Construction method of a diaphragm wall.
If an anchor is used, a girder is not necessary.
A diaphragm wall has a high bearing capacity for vertical loads.
A diaphragm wall is a watertight screen, if the execution is carried out carefully.
The execution is practically vibration-free and the noise pollution is slight.
Diaphragm walls can reach down to a great depth, even if firm soil must be penetrated.
Depths of up to 120 m have been reached.
A diaphragm wall can be integrated in the construction as a permanent wall.
supporting fluid
pipes
bayonet catch
locked
(suspension reinforcement)
supporting pipe
c.
filling funnel chute
supporting pipe
reinforcement
poured concrete
supporting fluid
a. b.
concrete
Project number
415660-21 v04
Date
December 2007
Page
43 of 279
A diaphragm wall has a large bending stiffness, so the deformations of the soil behind it
remain small.
A diaphragm wall is relatively expensive and is therefore only used in special
constructions.
A diaphragm wall cannot be reused.
A diaphragm wall is indeed watertight, but the concrete is slightly porous, so the wall can
sweat. This can be a drawback if the diaphragm wall is used as a permanent wall and it
remains in sight.
The reinforcement percentage must be limited to approximately 0.75 % in connection with
the execution, especially during the pouring of the concrete.
Densification of the concrete filling is practically impossible.
BERLIN WALL
A Berlin wall is made of steel I-profiles, the so-called primary elements or soldiers, that are
deeply inserted in the soil at distances between 1 and 3 metres, between which horizontal
wooden planks are placed synchronous to the excavation. The bottom plank will be located
only slightly deeper than the bottom of the final excavation. The I-profiles are inserted in the
soil by driving or by vibrating. Figure 2.10 shows a sketch of a cross-section over a Berlin wall.
The figure also shows how the planks are placed behind the flanges of the I-profiles.
The construction is mostly applied for temporary and not too deep building pits. A condition is
that the ground water level is below the excavation depth.
The deformations of the soil body that is supported by the wall can become quite large. In con-
nection with this the distance to the present constructions must be larger than for other earth
retaining structures.
The execution is relatively simple.
The retaining height is limited.
The deformations in the soil behind the wall can become quite large due to the relatively
large relaxation of the soil.
Execution is only possible if the ground water level is below the excavation depth.
RETAINING WALLS
Retaining walls obtain their earth retaining function from the weight of the soil and the shear
resistance in the soil below the bottom of the wall. In areas with poor soil conditions the retain-
ing wall is often placed on a pile foundation.
Retaining walls in the shape of quay walls are applied in combination with steel sheet piles in
sea harbours.
Retaining walls used to be built with bricks, now concrete is generally used.
Retaining walls are mostly built in L-shapes. The weight is obtained through the mass of the
soil on the floor plates of the L-shaped elements. In small versions they are used on a large
scale as platform walls in train stations. Prefabricated concrete caissons are also used as retain-
ing walls for large harbour structures. These are floated to the location and are sunk there.
They are filled with soil to obtain in sufficient mass. Caissons can be regarded as a special type
of retaining walls.
Project number
415660-21 v04
Date
December 2007
Page
44 of 279
Figure 2.10 Berlin wall.
CEMENT-BENTONITE WALLS
Cement-bentonite walls are vertical screens with no, or only limited strength, that are inserted
in the soil. They are mainly used if there are only requirements concerning watertightness.
If they are applied in combination with an embankment, in the case of sealing walls they can
function as a replacement for sheet pile walls.
The suspension that is applied is a mixture of cement and bentonite, which will harden after
some time.
I-profiles
c.t.c. 1,500 mm
wooden planks
50 mm thick
g.w.l.
Project number
415660-21 v04
Date
December 2007
Page
45 of 279
The watertightness can be improved by fitting a synthetic foil in the cement-bentonite mixture
before it has hardened.
The wall can be constructed in the following manners:
According to a variation of the diaphragm wall procedure:
The trench is made using diaphragm wall equipment, while the trench is filled with the
cement-bentonite suspension, or with bentonite fluid, which can be replaced by cement-
bentonite in a later stage;
By inserting a relatively heavy steel I-profile into the soil, onto which supply pipes have
been attached. The profile is pulled, while cement-bentonite is pressed in at the same time,
forming a thin water retaining screen.
The I-profiles are inserted with an overlap. The flange of the I-profile that is to be inserted
follows the trench, which has been made in the soil during the insertion of previous pro-
file. In this manner a sufficiently watertight screen can be constructed down to depths of
approximately 15 m.
Walls made of cement-bentonite are used as practically watertight screens around deep excava-
tions, building pits, rubbish dumps, and in order to limit the seepage through dikes.
COFFERDAMS
A coffer dam is a construction made of two sheet pile walls that are coupled by anchor rods, in
between which there is soil. The cofferdam obtains its earth-retaining function from the shear
resistance and the weight of the soil in the cofferdam. A cofferdam is sort of a combination of a
gravity wall and a sheet pile wall.
Usually steel sheet pile profiles are used for cofferdams. These are driven or vibrated into the
soil. After this both walls are coupled by rods that are connected to the girder. The space be-
tween the sheet pile walls is then filled with soil.
Cofferdams are used as training walls, for dike reinforcement, for the construction of elevated
roads, lock walls of sluices and as dam for the construction of a building pit in water.
In the last case they often have a temporary function.
A cofferdam is a stable and robust earth and water retaining construction that stands alone. It
can resist a large retaining height. The execution usually does not present many problems, be-
cause the sheet piles do not have to be driven deeply.
The strength has to be obtained from mobilisation of the shear resistance in the filling material,
so rather large deformations can take place.
CELLULAR COFFERDAMS
A cellular cofferdam is a cofferdam that is made up of round, or partially round cells. The cells
are made of special, flat, steel profiles.
Because of the shape of the cellular cofferdam, anchoring is not necessary. The sheet pile pro-
files are loaded in tension in their plane, such as for a tube under pressure, so they are fitted
with strong interlocks.
The photograph in figure 2.11 shows a cellular cofferdam.
In principle the execution of a cellular cofferdam is not different than that of a cofferdam. But
the planks for a cellular cofferdam have only a slight bending stiffness, so they must be prop-
Project number
415660-21 v04
Date
December 2007
Page
46 of 279
erly guided when they are installed. Usually special frames are used for this. The frames have
the same shape as the cell and the profiles are placed one next to another in this frame in ad-
vance. They are then inserted for a short distance at once, one after another, by driving or vi-
brating. Cellular cofferdams are mostly used as training walls and for quay constructions in sea
harbours. They are flexible constructions, which can generally stand the soil movements during
earthquakes very well. Because the sheet pile profiles are flexible, a lot of energy is lost during
driving. This is definitely the case if there is insufficient guidance and if the sheets have to be
installed deeply into dense soil. Like coffer dams, cellular cofferdams can also resist a large
retaining height. For a cellular cofferdam also quite some deformation is required for the mobi-
lisation of the shear resistance in the soil.
Figure 2.11 Photograph of a cellular cofferdam under construction.
REINFORCED SOIL
For reinforced soil, of which the most important application is the so-called terre arme, the
reinforcing effects of tensional elements are used, which have been inserted in the soil massive.
These tensional elements can be steel strips, steel pins, geotextiles and sometimes also fibres.
Terre arme uses steel strips that are connected to lining panels (in dutch: bekledingspan-
elen). During the raising in layers, the lining panels and the steel strips are placed layer per
layer until the desired height has been reached. So the placement of the panels and the strips
occurs simultaneously with the raising per layer. A similar method is used for the reinforce-
ment using geotextiles, but in this method panels are usually not used. The lining at the front is
obtained by folding over the piece of geotextile.
The soil of a terre arme wall is usually made of sand of another coarse-grained cohesionless
material. Prefabricated concrete elements are usually used for the lining. Figure 2.12 shows a
photograph of a terre arme wall. Terre arme constructions are usually constructed dryly.
Sometimes they are submerged under water afterwards. Detailed descriptions of terre arme
can be found in literature [Jansen, 1982 and Jansen et al., 1983].
Terre arme is mostly applied for abutments of bridges and viaducts and for the boundaries of
soil embankments.
Project number
415660-21 v04
Date
December 2007
Page
47 of 279
Figure 2.12 Wall constructed in terre arme.
2.4 Safety considerations
2.4.1 Introduction
Failure of the sheet pile construction
In principle a design method is chosen based on a probabilistic concept, meaning: calculating
with failure chances. Within this scope, making a design for a sheet pile construction boils
down to determining such dimensions of the construction and construction-parts that the failure
chance is sufficiently small.
When the desired safety level of a construction or the maximum accepted chance of failure of
the construction is determined, a comparative assessment is made between the damage due to
failure of the construction and the costs to realise a certain safety level.
A class-system has been made for the safety levels of a generally current sheet pile construc-
tion, which is expressed in the reliability index |.
The failure of a construction is regarded as no longer fulfilling at least one of the following
main functions:
retaining soil or water;
bearing loads;
protection against erosion, ground water flow or the guiding of flows.
According to NEN 6700; 1991 [NNI, 1991] and NEN 6740; 1991 [NNI, 1991] the so-called
ultimate limit state and the serviceability limit state are distinguished.
When the ultimate limit state 1A is reached, the stability of the entire sheet pile construction is
only just guaranteed.
Project number
415660-21 v04
Date
December 2007
Page
48 of 279
But in an earlier stage sufficiently large deformations of the sheet pile construction and the
neighboring terrain can occur, so the stability of the constructions that are founded on these is
endangered. This is the ultimate limit state 1B, which requires the testing of the occurring de-
formations for the demands made with respect to those deformations on before hand..
The ultimate limit state 2 also concerns deformations, but here the usefulness of the sheet pile
wall itself is tested. Or in other words, the deformations of the sheet pile construction and those
of the neighbouring terrain become so large that the usefulness is seriously damaged. For ex-
ample if traffic is hindered due to unequal settlements of the terrain located behind the sheet
pile wall.
Practical execution
In practice the execution comes down to the testing of stability and deformations:
stability of the main wall due to the exceeding of the soil resistance, plastic deformation of
the sheet or anchor fracturing;
failure of the anchor system, among which the so-called Kranz-instability;
heave;
total loss of stability;
deformations of the sheet pile wall and terrains located behind the wall.
In this method, for uncertain parameters in the design, partial safety factors or safety margins
are applied to the so-called representative values. These parameters are divided in strength
parameters and load parameters. Essentially these partial safety factors depend on the spread of
the parameter and the influence of the parameter on the failure of the construction.
During the development of the practical execution, the conservative values of the spread were
used. The influence of a parameter, for example the cohesion, on the ultimate limit state differs
per mechanism. Checking the ultimate limit state with regard to the stability of the main wall is
at the centre point, namely:
the exceeding of the maximum passive resistance;
the origination of the moment of plastic deformation in the sheets;
the exceeding of the bearing capacity of the anchors.
The partial safety factors for these mechanisms have been determined in a probabilistic man-
ner.
It is also possible to derive separate sets of partial safety factors for the other mechanisms, such
as the Kranz-stability, but these are not determined.
A method has been chosen in which the partial material factors that have been derived for test-
ing the ultimate limit state of an earth retaining sheet pile wall (where the maximum passive
soil resistance is exceeded), are also used for the analysis of the other failure mechanisms. This
means that for example for testing the moment of plastic deformation using the calculation
model beam on elastic foundation, in order to determine the design value of this moment a
certain adjustment is required of the maximum moment in the sheet pile wall which is calcu-
lated using this method. This is also true for the anchor force.
Project number
415660-21 v04
Date
December 2007
Page
49 of 279
The following must be taken into account when determining these adjustment factors:
the use of the design values of the parameters for the testing of the maximum soil
resistance for the purpose of determining the minimal required driving depth of the sheet
piles;
the limited part of the total failure chance space that is available for the different
mechanisms.
Design values of a parameter are obtained after applying the partial safety factor or safety mar-
gin that is valid for the parameter concerned, to the representative value of the parameter. The
representative value of a parameter is often put level with the so-called characteristic value.
The characteristic value of a parameter is the value with a certain probability of non-
exceedance or probability of exceedance. Principally the choice of the probability of non-
exceedance for the characteristic value is not set. Conform to the existing standards and regula-
tions a 5 % probability of exceedance or non-exceedance is used.
An extensive soil investigation due to which the spread of the average value of the parameter is
reduced, results in a more favourable design value. For the calculation of the partial factors a
conservative value for the spread was used. If the real spread is smaller, then it is possible to
reduce the partial factors. Possibly the reduced values for the partial factors can be derived
using the calculation rules given in the research reports [CUR, 1991].
Backgrounds of the determination of the safety levels
The failure of a sheet pile construction is regarded as no longer fulfilling the main functions
mentioned previously.
The failure of a sheet pile wall can be caused by the mechanisms that have also been men-
tioned previously. These are:
loss of stability of the main wall due to the exceeding of the maximal passive soil
resistance, plastic deformation of the sheet and anchor failure;
failure of the anchor system, for example Kranz-instability;
heave;
loss of total stability;
exceeding the deformation criterion.
The failure mechanisms and their mutual connection are fixed in the failure tree in figure 2.13.
Starting from the failure chances for the different mechanisms, the chance of the main event, or
in other words, failure of the entire construction, can be calculated using the failure tree. The
failure tree is calculated from bottom to top.
When determining the acceptable failure chance for each of the mechanisms, the tree is used
the other way around.
Starting from the acceptable chance of the main event occurring, the failure space, moving
downwards in the tree, a matching division of this space over the different mechanisms is
searched for. The failure space for the main event is determined using the by calibration to the
present practice [RWS, 1990].
Project number
415660-21 v04
Date
December 2007
Page
50 of 279
Figure 2.13 Failure tree for failure of a sheet pile wall.
Project number
415660-21 v04
Date
December 2007
Page
51 of 279
Based on this calibration the following classification with matching levels of the reliability
index | has been made for the mechanisms considered:
Class I: |
frac
~ |
yield,pl
~ |
anchor
~ 2.75 (simple wall protection);
Class II: |
frac
~ |
yield,pl
~ |
anchor
~ 3.75 (building pit);
Class III: |
frac
~ |
yield,pl
~ |
anchor
~ 4.50 (quay wall).
The classification is conform the classification of NEN 6702; 1991 [NNI, 1991]:
Class I: relatively simple constructions, no personal safety risks due to failure,
relatively slight damage due to failure;
Class II: substantial damage for failure; slight personal safety risks;
Class III: great damage due to failure and/or substantial personal safety risks.
The |-values that have been found have been used as starting point for the filling in of the
complete failure tree.
For the mechanisms in the failure tree, such as exceeding of the soil resistance, loss of total
stability, etcetera, values of 0.2 p and 0.6 p of the total failure chance space p have been used.
These are based on a rational weighing. In this manner the failure tree has been entirely filled
in.
After this, the failure tree is gone through from bottom to top for every class and the failure
chance of the top is determined. In this approach, the correlation of mechanisms is taken into
account. The failure chances of the entire construction that have been calculated in this manner
are 2.4 p to 3.6 p, where p is the failure chance of the three main mechanisms.
These failure chances lead to the following class-division:
Class I: |
construction
~ 2.5;
Class II: |
construction
~ 3.4;
Class III: |
construction
~ 4.2.
The highest safety class connects with the demanded safety level for primary dams, such as
fixed in the guidelines for the design of dams [TAW, 1985] and [TAW, 1989]. It appears that
the lowest safety class roughly connects with the safety level where no human lives or large
economical damage are at issue [CUR, 1991].
2.4.2 Testing the maximum soil resistance, the maximum moment in the sheet pile
and the maximum anchor force
General
The reliability index | is translated in partial load factors and partial material factors for the
mechanism, where the maximum passive soil resistance is exceeded.
For determining the design values of the moment and the anchor force respectively, the same
partial factors are used as those that are used for the mechanisms that lead to failure of the
sheet pile wall due to exceeding of the so-called moment of plastic deformation or failure of
Project number
415660-21 v04
Date
December 2007
Page
52 of 279
the anchor. This is why an additional safety factor is applied to the maximum value of the an-
chor force.
Failure of the sheet pile construction
Practical execution in steps
Step 1 Determine the representative cross-section, all relevant external loads and water
pressures and the safety class, possibly taking the length effects into
consideration.
Step 2 Determining the parameters
Determine the expected values of the parameters, their variation coefficients V and
the representative values X
rep
, if necessary. Referring to NEN 6740; 1991 [NNI, 1991]
the representative value can be a nominal value or the characteristic 5 % upper
boundary value or 5 %lower boundary value X
rep
.
Relevant parameters are:
strength parameters of the soil: cohesion, angle of internal friction and wall
friction angle;
unit weight of the soil;
stiffness parameters: spring constant, stiffness of the anchors, bending stiffness of
the wall;
surcharge load;
geometric parameters; retaining height, water levels.
How to determine these parameters is discussed in chapter 3.
Step 3 Calculate the minimal length of the sheet using partial factors
3.1 Determine the design values using table 2.5 for the representative values of the pa-
rameters. The design value of the unit weight is equal to the expected value; the design
value of the wall friction angle o is determined according to article 12.2.3 of NEN 6740
shown in table 3.1, in which the design value of the angle of internal friction is taken
for . For cohesion and angle of internal friction, the design value is found with a
(partial safety) factor via division. For geometric quantities, retaining height and
ground water levels, of which the uncertainty depends on the spread and not on the av-
erage, the safety margin is expressed in an additional safety margin (with respect to the
average). The margin, which has a minimum (A), depends on the standard deviation o
of the parameters concerned and is found by multiplying by a partial factor, so margin
= o.
For the surcharge load a design value must be chosen in agreement with table 3.3.
If high loads occur behind the sheet pile wall, for example due to the storage of
ore or the piling up of containers, the design value must be determined based on
the maximum value for the sheet pile construction given in the programme of
demands, and taking into consideration the indicated or derived load division in
the direction of the sheet pile construction and also perpendicular to it. The
maximum value of the surcharge load obtained in this manner may directly be
inserted in the test as design value.
Project number
415660-21 v04
Date
December 2007
Page
53 of 279
3.2 Calculate the minimal length of the sheet, freely supported, using the Blum method or
using a similar method.
Table 2.5 Partial factors and A concerning X
rep
.
and A concerning X
rep
class I class II class III
parameter
A A A
design value
cohesion c 1.00 1.00 1.10 X
rep
/
internal friction 1.05 1.15 1.20 X
rep
/
retaining height [m]
1)
1.60 0.20 2.20 0.30 2.60 0.35 max(+o; +A)
g.w.l. low side [NAP] 1.30 0.15 1.70 0.20 2.10 0.25 min(o; A)
g.w.l. high side [NAP] 0.66 0.05 0.87 0.05 1.50 0.05 max(+o; +A)
1)
after determining the design value of the retaining height, it has to been taken into ac-
count by lowering the ground level at the passive side and by keeping the ground level
constant at the active side.
Note: for the design value of the groundwater level at the high side sometimes just the
ground level can be used, because a higher value is physically not possible.
Step 4 Optimisation
Calculate the optimal length of the sheet piles based on the moment distribution, using
the model of the elastically supported beam. Use the design values for the bending
stiffness of the sheet pile, the stiffness of the anchor and the stiffness of the soil. This
length should at least be equal to the length determined in step 3.2.
The design values for the bending stiffness (EI)
d
and for the spring stiffness of the an-
chor (K
a
)
d
, must be derived from the respective representative values by using a partial
material factor
m
= 1.0.
The design value for the modulus of subgrade reaction (in dutch: beddingsconstante)
must be obtained by dividing the representative value of the low average from table
3.15 by
m;E
= 1.3.
The low design value for the modulus of subgrade reaction can also be determined
without the use of table 3.15, by dividing the expected value (i.e. the best estimate of
the average) by a factor 2.0.
Step 5 Checking the moment
5.1 Calculate the moment distribution for the length of the sheet as determined in step 4
with the high design value of the modulus of subgrade reaction of the soil layers. The
high design value for that modulus is obtained by dividing the representative value of
the high average from table 3.15 by
m;E
= 1.0.
The high design value of the modulus of subgrade reaction can also be determined
without using table 3.15. The expected value must be multiplied by a factor 1.5.
Possible uncertainties concerning the bending stiffness are also included in these sensi-
tivity calculations.
5.2 Determine the largest field moment and fixing moment from the steps 4 and 5.1. The
greater of these two values is the design value M
s;d
.
Project number
415660-21 v04
Date
December 2007
Page
54 of 279
5.3 The design value of the moment M
s;d
must be checked with the design value of the
strength of the sheet pile according to the standards concerning materials. Correspond-
ing with the standards, the following must be satisfied:
; ; s d r d
M M s
;
;
r rep
r d
m
M
M
=
in which:
M
r;rep
is the representative value of the moment of the sheet pile at the instant the ex-
terior fibre starts to yield.
m
is the material factor, which is equal to:
steel :
m;st
= 1.0
concrete:
m;c
= 1.1
wood:
m;w
= 1.2
Step 6 Checking the shear force and the normal force
The design values of the of the normal forces and the shear forces that come
forward as being decisive obtained from steps 3 up to 5, must be compared with
the design value of the strength of the sheet pile according to the material
standards concerned.
Step 7 Checking the anchor force
Sensitivity calculations regarding the anchor stiffness are only relevant if the anchor
stiffness is uncertain.
7.1 Determine in which step step 4 or step 5.1 the largest value of the anchor force was
calculated. For step 7.2 use the same parameters.
7.2 Calculate the determining anchor force F
a;max
using the parameters from step 7.1 and
the 5 %- upper boundary value of the anchor stiffness. The decisive moment that has
already been calculated does not have to be adjusted.
7.3 Extra factors must be applied to the value of F
a;max
obtained in step 7.2 in order to de-
termine the design values F
a;max;gr;d
and F
a;max;st;d
.
For testing the strength of the anchor plate or the grout mass of a grout anchor and
such:
F
a;max;gr;d
= 1.1 F
a;max
For the design of the anchor rod:
F
a;max;st;d
= 1.25 F
a;max
Project number
415660-21 v04
Date
December 2007
Page
55 of 279
Possibly effects of factors that have not been included in the calculation of the
anchor force must be taken into account, for example ground level settlements
and corrosion.
Comment: For the strength of the anchor rod extra safety is demanded, because for
yield and failure of the anchor rod, the functioning of the anchor completely fails. Be-
sides, the costs for overdimensioning the anchor rod are reasonably low.
7.4 The design values of the anchor force must be compared with the design values for the
strength of anchor rod and the strength of the anchor plate or the anchor body.
Anchor rod:
The design value of the strength of the anchor rod is equal to the representative value
of the force at which the anchor rod yields, divided by the material factor
m;st
.
Anchor plate:
The design value for the strength of the anchor plate is equal to the calculated bearing
capacity based on the design values for soil parameters and geometry parameters,
where the material factors from table 2.5 in step 3 are used.
Grout anchors:
Ultimate limit state:
In general the design of grout anchors is based on the results of suitability tests. For the
ultimate limit state, due to the exceeding of the maximum resistance of the grout body
in the soil, the value of F
a;max;gr;d
found in step 7.3 is used. For testing the anchor rod or
the anchor cable, it must be remembered that this must be based on F
a;max;st;d
.
The following conditions hold for the design:
F
a;max;gr;d
s
the lowest value of F
k;d
and 0.85 F
f;d
and
; ;
;max; ;
1.4
br st d
a st d
F
F s
in which:
F
k;d
is the design value of the force on the anchor, where the demand for the creep
coefficient is exceeded
F
f;d
is the design value of the force on the anchor, where the anchor undergoes a
continuous displacement
F
br;st;d
is the design value of the failure force of the anchor rod
In the last case the yield point is not used, because pre-tensioned steel does not have a
yield point.
F
k;d
and F
f;d
must satisfy:
Project number
415660-21 v04
Date
December 2007
Page
56 of 279
;
;
;
k repr
k d
m k
F
F
=
and
;
;
;
f repr
f d
m f
F
F
=
in which:
F
k;repr
and F
f;repr
are the values from the suitability tests carried out on both the
force where the demand concerning the maximum value of the
measure of creep k is exceeded, and the force, where the an-
chor undergoes a continuous displacement, respectively;
m;k
and
m;f
are the partial material factors, which are set to 1.25 in this
case.
A condition for using this method is that so-called control tests must be carried out on
all grout anchors in the construction. During those tests the tension force on the an-
chors is raised up to at least F
a;max;gr;d
.
Serviceability limit state:
If demands are made for the displacement of the main wall and pre-tensioning is not
used, the serviceability limit state can be decisive. This situation is tested using a
method similar to the method mentioned above. Except:
F
a;max;gr;d
s
min(F
k;d
; F
u;d
)
in which:
F
u;d
is the tension force for which the criterion for the displacement u
limit
is ex-
ceeded.
If pre-tensioning is used, F
u;d
is the tensional force for which the sum of the displace-
ment criterion and the displacement at the pre-tensioning force is exceeded. The partial
factor for determining the design value of the maximal tension force is taken to be 1.0.
Comment:
Generally, in practice no suitability tests will be available during the design phase. The
design values of the at least required strength F
a;max;gr;d
and F
a;max;st;d
and the u
limit
are in
fact tendering demands.
Suitability tests are conducted in order to test the grout anchor, which is designed by
the contractor, for the tendering demands. Based on these results, the decision is made
to either change the design of the anchor or to adjust the anchor distance, based on the
procedure mentioned above.
Step 8 Checking deformations
8.1 Determine the u
limit
, the maximum allowable displacement caused by displacement of
the sheet.
Then show if this is limit state 1B or 2, such as described in NEN 6740; 1991 [NNI,
1991].
Project number
415660-21 v04
Date
December 2007
Page
57 of 279
The u
limit
can have a much higher value for limit state 1B than for limit state 2. This
value is a kind of w
req
, for the critical settlement of a foundation for limit state 2, such
as prescribed in NEN 6740; 1991 [NNI, 1991].
If the maximum displacement u
max
from step 4 is lower than the u
limit
for the service-
ability limit state, no further calculations are necessary. If u
max
is not lower than u
limit
,
and step 5 has produced a different sheet pile profile than the profile that was origi-
nally used in the calculations, a new deformation calculation can be made using the pa-
rameters from step 4, with the deformation demand:
u
max
s
u
limit
Step 9 Checking other mechanisms
9.1 Kranz-stability
Determine the design value of the anchor force F
a;max;gr;d
from step 7.3.
Use the Kranz-model to determine the minimum required anchor force F
kr;d
with the
design value of the parameters according to step 3.1.
The following counts for all safety classes:
1.3 F
a;max;gr;d
s
F
kr;d
As an alternative, the testing for the Kranz-stability can also be carried out using the
representative values of the soil properties. Then the following demand must be met:
1.5 F
a;max
s
F
kr;rep
in which:
F
a;max
is the value of the anchor force from step 7.2
F
kr;rep
is the minimum required anchor force according to Kranz, which has been
determined using the representative values of the soil properties.
9.2 Heave
Determine the deterministic safety factor
frac
using the representative values of the pa-
rameters. This safety factor
frac
must satisfy the values in table 2.6.
In safety class III heave will hardly ever be decisive.
Table 2.6 Safety factor
frac
for heave.
safety class minimal value of
frac
I 1.3
II 1.5
III 1.7
9.3 Total stability
Calculate the total stability using the partial material factors in NEN 6740; 1991 [NNI,
1991]. This standard concerns constructions in safety class II. In order to express the
difference between the different safety classes, the driving moment in the base equa-
Project number
415660-21 v04
Date
December 2007
Page
58 of 279
tion in art. 13.4.3 of NEN 6740; 1991 must be multiplied by the class factors in table
2.7.
For sheet pile constructions in primary water retaining structures the Leidraden ten
behoeve van het ontwerpen van rivierdijken can be used [TAW, 1985 and 1989].
Table 2.7 Class factors.
safety class class factor
I 0.9
II 1.0
III 1.1
9.4 Wales design
Determine the decisive anchor force F
a;max
from step 7.2. The force distribution in the
girder should be calculated using the following design values for the anchor force F:
steel sheet pile wall: F

a;d
= 1.1 F
a;max
concrete sheet pile wall: F

a;d
= 1.1 F
a;max
wooden sheet pile wall: F

a;d
= 1.1 F
a;max
The design value of the moment in the girder must be determined and tested according
to 4.5.13 using these design values.
9.5 Thermal loads
If thermal loads (fire) can play a part near the sheet pile wall, this must be taken into
account in a deterministic manner.
9.6 Piping and hydraulic soil fracturing
Assuming that the water supply on the active side is sufficient, and that there is a free
water level in front of the sheet pile wall, the critical outflow gradient must satisfy the
following:
i
(1 + 1.64 V
i
) < 0.5
in which:
i
is the average value of the calculated outflow gradient
V
i
is the variation coefficient of the calculated outflow gradient
For safety class III the hydraulic gradient i must be calculated in such a precise manner
that V
i
< 0.3.
For the other classes less precision is required.
Moreover, if there is not enough data available, piping must be tested using the Lane
method according to 4.4.4.
L
1
+ L
2
>
p
C
L
AH
in which:
Project number
415660-21 v04
Date
December 2007
Page
59 of 279
L
1
is the length of the sheet pile wall in m
L
2
is the distance from the lower groundlevel till the tip of the sheet
AH is the difference in waterlevel over the construction in m
p
is the partial factor for piping from table 2.8
C
L
is Lanes constant (from table 4.3)
L
1
, L
2
, AH are design values
Table 2.8 Partial factors for piping.
class I II III
P
1.0 1.5 2.0
9.7 Vertical bearing capacity
Limit state 1A must be tested using NEN 6743; 1991 [NNI, 1991] taking into consid-
eration the specific characteristics of the sheet pile wall.
For the calculation of the vertical load component of the anchor, the anchor force
F
a;max;gr;d
calculated in step 7.3 must be used.
Step 10 Check the executional aspects
In the design enough attention must be paid to the possible load situations and failure
mechanisms in the execution phase, such as excavations that are temporarily deeper
than was planned.
Step 11 Aspects of the material choice
For the choice of the material of the sheet, different aspects play a part, such as the
cost-price in relation with the type of construction, the intended safety level and aes-
thetic demands. More specific properties of the material that can influence the choice
are:
the fluid tightness or the soil tightness of the final wall in relation with the
function of the construction;
strength and deformation behaviour;
influence of the wall friction and the differences per material;
execution;
method of installation;
sustainability.
2.4.3 Length-effect of the sheet pile construction
For a certain class of constructions the total length in the longitudinal direction, must be taken
into account. For these constructions failure of a cross-section leads to failure of the entire
construction. An example is a sheet pile wall that is a part of a dike ring. Failure of the sheet
pile wall leads to failure of the dike ring.
For bank protections (in dutch: overbeschermingen) that have no function as a primary or sec-
ondary water retaining structure, the length effect does not play a part. In this situation failure
only leads to local damage.
Project number
415660-21 v04
Date
December 2007
Page
60 of 279
The length effect leads to an increase in the maintained reliability index |. Using figure 2.14 it
must be checked if, due to the length effect, the construction belongs in a higher class than the
class that was chosen.
Figure 2.14 Length-effect for sheet pile walls on the required value of |.
|

(
r
e
q
u
i
r
e
d
)
Class III
Class II
Class I
Class III
C;ass II
Class I
Longitude (m)
Project number
415660-21 v04
Date
December 2007
Page
61 of 279
3 Parameters for sheetpile desgin
3.1 Design parameters
In the design calculation for the testing of the different limit states of a sheet pile construction,
the following parameters are of importance:
the loads and possibly prescribed deformation demands;
the properties of the soil, of the construction materials and of the profiles;
the geometry of the construction.
Fig 3.0 In the execution phase sometimes temporal stability measures have to be taken.
An important part of the loads is made up of forces that are exerted on the sheet pile wall by
the soil, the ground water, the free water, or by ice- and anchor forces.
Of course, other forces can also be active, such as vertical forces on the top of the sheet pile
wall, shipping forces on boulders, and traffic loads. As a rule the size of the design values for
these forces does not have to be determined by the designer. The design values can usually be
obtained from directions and/or recommendations, such as the EAU [EAU; 1990]. The same
counts for the possibly constraining deformation demands in case the deformations can cause
danger to the neighbouring constructions or they influence the serviceability of the terrain be-
hind the sheet pile wall.
Generally the designer of the sheet pile construction will have to determine by himself the de-
sign values for the forces that are exerted by the soil and the water. Instructions for this can be
found in 3.2.
Project number
415660-21 v04
Date
December 2007
Page
62 of 279
In order to determine the size of the design values of the forces that are exerted on the sheet
pile wall by or through the soil, it is obvious that the designer must be familiar with the proper-
ties of soil. These should be fixed according to the rules in NEN 6740; 1991 TGB 1990
Geotechniek; Basiseisen en belastingen [NNI; 1991]. Paragraph 3.4 describes the soil investi-
gation and the determination of the design values of the soil properties for the benefit of the
design of a sheet pile construction.
Of course, the soil is part of the earth retaining construction as a whole and therefore the prop-
erties of the soil are also necessary data for the determination of the strength and the deforma-
tion of the sheet pile construction. From this aspect the determination of the soil properties is
also a very important part of the soil investigation.
The properties of the construction materials can be divided into:
The strength properties and deformation parameters of the materials that have been used in
the different parts of the construction. Usually these are steel, concrete or wood. The
design values of the tensional strength, the compressive strength, the bending strength and
the Young's modulus are important. These properties will have to be obtained from the
material standards for steel (NEN 6770; 1991), concrete (NEN 6720; 1991) and wood (NEN
6760: 1991) [NNI; 1991].
The cross-section properties such as on the surface of the cross-section, the moment of
inertia I and the moment of resistance W.
In 3.3 the effect of oblique bending is discussed, and directives are given for the reduction
factors that are used for the moment of inertia and the moment of resistance. Wooden and
concrete sheet pile profiles do not have the problem of oblique bending.
For a sheet pile construction the geometry of the construction means:
the design value of the height and the slope of the terrain at both sides of the wall, the
length of the wall in horizontal direction, the locations of the points of application and the
direction of the anchor forces and extreme loads due to ship impact, boulder forces and ice
forces;
if relevant, the design values for the pressure head of the groundwater level in front of and
behind the wall, the design values of the free water level at the front of the wall and the
design value of the wave height.
A number of these geometric parameters is discussed in 3.4.
Generally, unfavourable situations are assumed when determining the design values of geomet-
ric parameters, where the chance that a number of the special circumstances mentioned coin-
cide is large. This is a judgement of situations that are not normal, such as: excavating holes
at the front of the wall, a strong raise of the ground water level due to extreme rainfall or inun-
dation of the area behind the wall and possible groundwater level raises up to ground level in
case there is no reliable and controlled drainage behind the sheet pile wall.
Project number
415660-21 v04
Date
December 2007
Page
63 of 279
3.2 Loads
3.2.1 Introduction
NEN 6740; 1991 [NNI; 1991] gives the loads, external as well as internal, from the soil and the
ground water, that must be taken into account in the design of a sheet pile wall.
The design value of the external loads must be determined according to NEN 6702; 1991 [NNI;
1991] by multiplication of the representative value of the load in question with the partial fac-
tor belonging to it. This factor depends on the safety class in which the sheet pile construction
is placed.
The design value of the maximum value of the active, the passive and the neutral soil loads
must be determined using the design values in of the soil parameters in the calculation models.
The loads following from this do not require an extra safety factor.
3.2.2 Loads due to soil
General
In the calculation for sheet pile walls where the sheet pile wall can move along with the pres-
sure of the soil at the backside, the minimum value of the soil pressure can be viewed as a load.
If movement is almost excluded, at least the neutral soil pressure must be taken into account as
load.
In the case that horizontal wall displacements are of a number of centimetres are permitted in
opposite direction of the soil resistance, the maximum value of the soil resistance may be seen
as a load.
In this situation, soil pressure means the load that the soil exerts on the wall on the active side
and soil resistance means the load on the passive side.
For the calculation of size, direction and location of the resulting soil pressure the following
aspects must be taken into account:
the surcharge load on the soil next to the earth-retaining structure;
the ground water level;
the unit weight of the soil;
the shear strength of the soil;
the friction between the wall and the soil;
the slope of the backside of the wall;
the level and the slope of ground level at both sides of the wall;
the levels of the free water surface and the ground water level;
the movement and deformation of the wall with respect to the soil;
if relevant, swelling, for example due to freezing of the soil that is retained.
The minimum respectively maximum value of the active and passive soil pressure occurs as
soon as the shear resistance of the soil is completely mobilised. They are determined by the
state of plastic balance in the soil.
Project number
415660-21 v04
Date
December 2007
Page
64 of 279
Maximum soil pressure and maximum soil resistance in the drained situation
The horizontal component of the maximal effective soil pressure at a depth z below ground
level for the 'active' and the 'passive' case is (see figure 3.1):
, )
; ; ; ; ; ; ea h d d sur d d z d d d c d
p p K z u K c K
= + +
(3.1)
in which:
p
ea;h;d
is the horizontal component of the design value of the soil pressure
(ea: 'earth', h: 'horizontal', d: 'design value');
p
d
is the design value for the surcharge load over a width behind the sheet
pile
wall of at least the length of the sheet pile wall;
K
sur;d
is the design value of the soil pressure factor due to the surcharge load
(sur: 'surcharge');
d
is the design value of unit weight of the soil;
z is the depth;
u
z;d
is the design value of the water pressure at depth z;
K
;d
is the design value of the soil pressure factor due to the weight of the
soil;
c
d
is the design value of the effective cohesion;
K
c;d
is the design value of the soil pressure factor for the part of the cohe-
sion;
d is the depth of the wall;
The horizontal component of the design value of the effective force follows from:
; ; ; ; ; ; ; ; ; ; ; ; ;
0
z d
ea h d ea h d ea h sur d ea h d ea h c d
F p dz F F F
=
= = + +
} (3.2)
The first term indicates contribution of the surcharge load. In case of homogeneous soil this
pressure p
d
K
sur;d
is constant over the depth z. The representative value of the contribution in the
resulting force from this pressure for the active case at depth z is:
ea;h;sur;a;d d sur;a;d
F = zp K
(3.3)
The point of application is located at a depth of 0.5 z.
For the passive case the resulting force is:
ea;h;sur; p;d d sur; p;d
F = zp K
(3.4)
where the point of application is also at 0.5 z.
Project number
415660-21 v04
Date
December 2007
Page
65 of 279
The second term indicates the contribution of the effective dead weight of the soil. This pres-
sure:
, )
d z;d ;d
z u K (3.5)
increases linearly with depth. The representative value of the resulting force for the active case
follows from the equation:
, )
; ; ; ; ; ; ;
0
z d
ea h a d d z d a d
F z u K dz

=
=
} (3.6)
The location of the point of application can be calculated using the theory of mechanics. For
the passive case the resulting force follows from:
, )
; ; ; ; ; ; ;
0
z d
ea h p d d z d p d
F z u K dz

=
=
} (3.7)
where the point of application can be determined using the theory of mechanics.
The third term describes the part of the effective cohesion c'
d
. This pressure c'
d
K
c;d
is constant
over the depth in a soil layer i with thickness d
i
. For the active case the resulting force in layer i
is:
ea;h;c;a;d i d c;a;d
F = d c K (3.8)
The point of application is located in the middle of layer i. For the passive case the resulting
force in layer i is:
ea;h;c; p;d i d c; p;d
F = d c K
(3.9)
The point of application is located in the middle of layer i.
Soil pressure factors for active and passive soil pressure for straight slip surfaces
The design values of the soil pressure factors K
, K
sur
, K
c
for the active and passive values of
the horizontal component of the soil pressure due to the self weight of the soil and possibly a
uniformly distributed load p are:
, )
, ) , )
, ) , )
2
; ; 2
; 2
;
cos
sin sin
cos 1
cos cos
d
a d
d a d d a
a d a
K
| o
| o | |
o
o o o |
+
=
(
+
(
+
+ (

(3.10)
and
Project number
415660-21 v04
Date
December 2007
Page
66 of 279
, )
, ) , )
, ) , )
2
; ; 2
; 2
;
cos
sin sin
cos 1
cos cos
d
p d
d p d d p
p d p
K
| o
| o | |
o
o o o |
=
(
+
(
( +

(3.11)
, )
; ; ; ;
cos cos
cos
a
sur a d a d
a
K K
o |
o |
=
+
(3.12)
and
, )
; ; ; ;
cos cos
cos
p
sur p d p d
p
K K
o |
o |
=
+
(3.13)
, ) , )
, )
;
; ;
;
2cos cos 1 tan tan cos
1 sin
d a a a d
c a d
d a d a
K
| | o | o o
| o o |

=
+ +
(3.14)
and
, ) , )
, )
;
; ;
;
2cos cos 1 tan tan cos
1 sin
d p p p d
c p d
d p d p
K
| | o | o o
| o o |

=
+ +
(3.15)
in which:
|'
d
is the design value of the effective angle of internal friction of
the soil;
o, |, o
a
and o
p
are the angles according to figure 3.1.
The values for o can be taken from table 3.1
Table 3.1 Wall friction angles o
a
and o
p
.
design values o
a
and o
p
wall surface roughness
straight slip surface curved slip surface
toothed > 10 d
50
0.67 |'
d
s |'
d
rough 0.5 10 d
50
0.67 |'
d
s |'
d
2.5 and s 27.5
half rough 0.1 0.5 d
50
0.33 |'
d
0.5 |'
d
smooth < 0.1 d
50
0 0
Remark: Generally, sheet pile walls will belong to the category half rough or rough.
Project number
415660-21 v04
Date
December 2007
Page
67 of 279
Figure 3.1 Relevant angles for the calculation of active and passive earth pressure.
The equations for the soil pressure factors for the active situation are valid for wall friction
angles o
a
> 0 and a slope of the wall o that meets one of the following requirements:
20 > o > + 10 for 0 s |
a
s |'
d
(3.16)
10 > o > o
min
for |'
d
s |
a
s |'
d
(3.17)
and for a wall friction coefficient o
a
s 0 for a slope of the wall o that meets the following re-
quirement:
+ 20 > o > o
min
for |'
d
s |
a
s 2/3 |'
d
(3.18)
, )
, )
min
cos
tan
sin
sin
sin
d
d a
d
d a
|
o
| |
|
| |
=
+
+
(3.19)
The equations for the soil pressure factors for the passive situation may only be used when:
the wall friction angles o
p
s 0 and |'
d
s 30 for rough steel or similar walls, respectively
|'
d
s 35 for a rough concrete wall surface; o s 0 and |
a
> 0;
wall friction angles o
p
> 0 for every value of |'
d
; o s 0 and |
a
> 0.
o is the slope of the wall at the backside
| is the slope of the groundlevel next to the wall
F
ea;h;p
F
ea;v;p
F
ea;v;a
F
ea;h;a
F
ea;a
p
p
F
ea;a
F
ea;p
z
z
o +o
o
+|a
|a
+oa
+op
|p
+|p
+o
F
ea;p
Project number
415660-21 v04
Date
December 2007
Page
68 of 279
In all other cases the calculation must be carried out using curved or non-linear slip surfaces.
Further it must be checked if the vertical force on the sheet pile construction exerted by the wall
friction does not exceed the vertical bearing capacity.
In order to take the wall friction angle o into account a relative (vertical) displacement between
the sheet pile wall and the adjacent soil massive is necessary. If this relative displacement is
cancelled or reversed because of the settlement of the sheet pile wall (due to a too high vertical
load), o = 0 or o with an opposite sign must be used for the calculation of the horizontal soil
pressures.
The wall friction angles following from the previous consideration and table 3.1, are design
values if a design value has also been used for |'
d
.
Soil pressure factors for active and passive soil pressures for curved sliding surfaces
In reality the sliding surfaces will be curved due to the influence of the wall friction o. Theo-
retically straight slip surfaces only occur for o = 0.
For active soil pressures the differences in resulting soil pressures between straight and curved
sliding surfaces are slight. For passive soil pressures the differences can be quite considerable,
especially for high values of |'
rep
. Straight slip surfaces can give a much too favourable impres-
sion. This is the reason that clear limitations have been made concerning the use of the formula
for the soil pressure factors given previously.
A large number of authors have published soil pressure factors for curved sliding surfaces
[Krey; 1963], [Sokolowski; 1965] and [Caquot et al.; 1973].
Figure 3.2 compares a number of results from literature. Also table 3.2 gives a collection of
values that are frequently used. The values recommended by [Caquot et al.; 1973] can be de-
termined from the diagrams in figure 3.3a.
Project number
415660-21 v04
Date
December 2007
Page
69 of 279
Coulomb
Streck et al.
Ohde
Caquot et al.
|' = 30
o

c
p
K
;p;d
10
5
0
0
o
-10
o
-20
o
-30
o

Figure 3.2 Example of values for K
;p;d
in case of curved slip surfaces according to different au-
thors.
Table 3.2 Soil pressure factors K
;p;d
for passive soil pressures according to Sokolowski (o = 0)
[Sokolowski; 1965].
| 10
o
20
o
30
o
40
o

|
o 0
o
5
o
10
o
0
o
10
o
20
o
0
o
15
o
30
o
0
o
20
o
40
o

-30
o
1.04 1.10 1.14 1.26 1.47 1.62 1.49 2.01 2.42 1.86 2.98 4.15
-20
o
1.18 1.28 1.33 1.51 1.80 2.00 1.90 2.69 3.29 2.50 4.42 6.30
-10
o
1.31 1.42 1.50 1.77 2.16 2.41 2.39 3.50 4.35 3.37 6.36 9.42
0
o
1.42 1.55 1.63 2.04 2.51 2.86 3.00 4.46 5.67 4.60 9.10 13.9
+10
o
1.49 1.64 1.73 2.30 2.88 3.32 3.65 5.62 7.29 6.16 13.06 20.4
+20
o
1.53 1.69 1.80 2.53 3.26 3.79 4.42 7.13 9.27 8.34 18.3 29.9
+30
o
1.52 1.70 1.82 2.76 3.61 4.24 5.28 8.76 11.69 11.3 26.7 43.4
Project number
415660-21 v04
Date
December 2007
Page
70 of 279
-40
o
-30
o
-20
o
-10
o
0
o
10
o
20
o
30
o
40
o
K
;p;d
15
10
5
|' = 40
|' = 30
|' = 30
|' = 10
|' = 0
c
p
o
p
20 30 40 20 30 40
o = |'
|'
d
K
;p;d
|'
d
0.8
0.6
0.4
0.2
-10
0
10
20
30
2.5
3
4
5
7.5
10
2
k
;p;d
=
Figure 3.3 Passive soil resistance factors according to [Caquot et al.; 1973].
Graphical method according to Culmann
This method is only suitable for non-cohesive soils, so the effective cohesion c' = 0. Besides it
must be realistic to regard the soil as homogeneous.
The advantage of this graphical method is that irregular shapes of the ground level (slopes,
etc.) and possible surcharge loads can be included relatively easily in the determination of the
earth pressure.
Project number
415660-21 v04
Date
December 2007
Page
71 of 279
a. b. c.
Q
G
G
Q
F
ea;d
F
ea;d
F
ea;d
Q
G
o
o
|'
|'
o
Per unit depth a number of straight slip surfaces is assumed, where the resulting earth pressure
is calculated graphically per slip surface.
Next the slip surfaces can be determined where the resulting earth pressure is maximal for the
active side and minimal for the passive side. These slip surfaces are determining for the calcu-
lation.
Per slip surface the force polygon of the soil wedge concerned is determined using the angles
|'
d
and o, such as shown in figures 3.4a and 3.4b.
By placing this polygon in the tip of the sheet pile wall, and tilting it 90 |'
d
for the active
side or 90 + |'
d
for the passive side, the force Q coincides with the slip surface, such as figure
3.4c shows for the active side.
Figure 3.4 Graphical method according to Culmann. Force polygon per slip surface (only active
side is shown).
Now the force G, the force F
ea;d
and the direction of F
ea;d
with respect to G can be determined
from the intersection of F
ea;d
and the shear band. This is repeated for a number of shear bands,
so F
ea;a;max
and F
ea;p;max
are obtained. Figure 3.5 shows the method.
The influences of a non-horizontal ground level and a surcharge load can be included in the
force G.
In this manner only the size of the earth pressure force can be determined. The distribution of
the earth pressure along the wall is not known yet. By repeating this procedure for a number of
apparent foots which are located higher, the distribution of the earth pressure over the depth
is obtained as shown in figure 3.6. The location of the resulting earth pressure force can be
determined from this figure.
Project number
415660-21 v04
Date
December 2007
Page
72 of 279
F
ea
until depth h
n
= F
ea;n
F
ea
until depth h
n+1
= F
ea;n+1
P
ea;n-n+1
=
F
ea;n+1
F
ea;n
h
n+1
h
n
h
n+1
h
n
e
n.n+1
Figure 3.5 Graphical method according to Culmann. Determination of E
a;max
and E
p;min
.
Figure 3.6 Graphical method according to Culmann. Method for the envelop of the earth pressure
over the depth.
slip surface for max F
ea;a
slip surface for
F
ea;p;min
F
ea;a;max
F
ea;p;min
u , )
|
o , +)
= 90
o
o
Project number
415660-21 v04
Date
December 2007
Page
73 of 279
Earth pressures in an undrained situation
This situation can be calculated in two manners:
a. by making the calculation using the present effective earth pressures and the formulae and
the earth pressure factors from 3.2.2. The part of the water pressure that has not been
consolidated must be added to the effective horizontal pressure that is obtained from the
calculation;
b. by making the calculation using the undrained shear strength f
undr;d
.
For o = 0 and | = 0, per definition |' = 0 gives o = 0.
Then K
q
= 1, K
sur
= 1, K
fundr;a
= -2 and K
fundr;p
= 2.
The following formulae are valid:
For the upper load:
F
ea;h;sur;a;d
= z p
d
(3.20)
F
ea;h;sur;p;d
= z p
d
(3.21)
where the point of application is located at 0.5 z.
Due to the self weight:
; ; ; ;
0
z d
ea h a d d
F zdz

=
=
}
(3.22)
f
undr
is the undrained shear strength.
; ; ; ;
0
z d
ea h p d d
F zdz

=
=
}
(3.23)
N.B.: In the method for the undrained situation the calculation is made in the domain
of the total stresses, so the unit weight of the soil or the saturated weight
sat
must be used.
The location of the point of application can be determined using the theory of mechanics.
Project number
415660-21 v04
Date
December 2007
Page
74 of 279
The part of the cohesion for an adhesion = 0:
F
ea;h;fundr;a;d
= 2 d
i
f
undr;d
(3.24)
F
ea;h;fundr;p;d
= 2 d
i
f
undr;d
(3.25)
The point of application is located at 0.5 d
i
.
For normally consolidated clay in the undrained situation, the calculation may also be made
using the adhesion between soil and wall, where:
0.5
adh undr
f f =
(3.26)
For the part of the cohesion plus adhesion the following is found:
F
ea;h;fundr;adh;a;d
= 2.4 d
i
f
undr;d
(3.27)
F
ea;h;fundr;adh;p;d
= 2.4 d
i
f
undr;d
(3.28)
The point of application is located at 0.5 d
i
.
Neutral earth pressure
In case of a wall that does not displace and where only the self weight of the fill acts, the
horizontal component of the force from the so-called neutral earth pressure becomes:
, )
; ; ; ; ;
0
z d
ea h o d d z d o d
F z u K dz
=
=
}
(3.29)
in which:
F
ea;h;o;d
is the design value of the horizontal force from the neutral earth pressure;
d
is the design value of the unit weight of the soil;
d is the driving depth;
z is the depth;
u
z;d
is the design value of the water pressure at depth z;
K
o;d
is the design value of the neutral earth pressure factor.
Project number
415660-21 v04
Date
December 2007
Page
75 of 279
For a horizontal soil surface and normally consolidated incoherent soil, the neutral earth
pressure factor K
o;d
must have at least the value obtained from the following formula:
K
o;d
= 1 sin |'
d
(3.30)
in which:
|'
d
is the design value of the effective angle of internal friction.
The formula given above may also be used for cohesive soils. Possible cohesion may not be
taken into account.
3.2.3 Surcharge load
Surface load
This includes surface loads and loads due to foundations and such. The design value of the
surcharge load must be held as principle. If there is uncertainty about the surcharge load on the
active side, in any case a uniformly distributed load of at least 10 kPa must be held. Table 3.3
gives the design values of loads for the different classes of the sheet pile constructions.
Table 3.3 Loads for different safety classes.
class type of construction design value of the load in [kPa]
I shore protection 5 10
II quay, building pit 10 20
III quay for sea-harbours, dike improvements > 20
If the terrain is meant for storage the design value of the actual load must be taken into
account. Directions for this can be found in the EAU [EAU; 1990]. For one storage the load
outside the traffic road can be 300 kPa.
Traffic load
For the design of sheet pile walls close to traffic roads or railways the traffic load is taken into
account according to the Wegenverkeersregelement or the VOSB respectively.
If the point of application of the traffic load is not located directly next to the sheet pile
construction, the wheel loads may be modelled by an equivalent uniformly distributed load.
Project number
415660-21 v04
Date
December 2007
Page
76 of 279
The loads can be schematised to a uniform vertical load of 50 kN/m
2
over a rectangular area
with length 2 m and width 1 m, placed next to the edge of the road- or terrain paving.
For the design of sheet pile walls in the geotechnical categories 1 and 2 (GC1 and GC2) it is
allowed to regard the traffic load directly next to the sheet pile wall as a uniform static load in
the manner as described previously.
For crane foundations that are supported by a sheet pile wall the maximum static load must be
increased by an impact factor.
The influence of impact loads is usually evaluated by considering the energy that is absorbed
by the construction. In this case the same impact factor could be used that was applied for the
design of the legs of the crane. For sideways impacts against the sheet pile wall it is necessary
to consider the higher stiffness of the retained soil due to the dynamic loads.
The calculation of the load on the sheet pile wall due to this type of surcharge load is treated in
4.1.6.
3.2.4 Horizontal loads due to water level differences in free water and groundwater
A sheet pile wall that only retains free water
If the sheet pile wall retains only water, according to 2.1 it is not a geotechnical construction.
The design value of the water pressure must be determined according to NEN 6702; 1991 [NNI
1991]. This means that the representative value of the water pressure, determined using the
representative values of the water height and the unit weight of the water, must be multiplied
by the partial load factor
f;g
= 1.2.
Quasi-stationary water movement
When the design values for ground water pressures are chosen, long-term observations of
ground water levels in the direct surroundings of the construction must be taken into
consideration, even if a drainage system is placed.
For sheet pile constructions that retain soil with a low permeability such as silt and clay, it
must also be assumed that the ground water behind the wall can reach up to ground level,
unless a reliable drainage system is placed at a lower level.
If the assumption has been made that a drainage system has been placed that permanently keeps
the ground water pressures low, provisions must be made for maintenance and effective
functioning of the drainage during the life span of the construction. The consequences of
possible blocking or failure of the drainage system must be taken into account during the
designing. The average water levels may be used for these calculations.
For stationary conditions of the open water and the ground water the loads according to figure
3.7 must be held.
Project number
415660-21 v04
Date
December 2007
Page
77 of 279
model water pressures resulting
water pressure
impermeable
layer
h
w;1o
w;o

h
w;2o
w;o

h
w;2id
h
w;1id
5 kPa
Ah
w:d
> 0.5 m
For slight water level differences outside without tidal effects and if drainage is possible
through the sheet piles wall, the design value of the pressure difference is taken to be
minimally 5 kPa.
Slight water level differences outside mean: average water level outside lowest water
level s 1 m.
Figure 3.7 Resulting water pressures and groundwater pressures for stationary conditions.
If there are no tidal effects and the difference between the average and the lowest water level
outside is bigger than 1 m, the following is valid for the difference in water level Ah
w;d
:
Ah
w;d
is the design value of the difference in water level between the
groundwater level and the lowest water level outside that can occur within 24
hours.
If this situation can occur for different groundwater levels, the design groundwater level should
be taken to be the most unfavourable level.
In case of water outside with tidal effects and no drainage the load is shown in figure 3.8.
Project number
415660-21 v04
Date
December 2007
Page
78 of 279
H.W.
L.W.
L.L.W.S.
0.3 m
AH
W

Ah
w
w
a
a
design g.w.l.
Figure 3.8 Resulting (ground)water pressures for big differences in water level outside, with tidal
influence, undrained situation.
In this figure:
L.L.W.S. (low- low water- spring): the average of the lowest low water spring of every
month calculated over a number of years;
L.W. (low water): average of the low water calculated over a number of years;
H.W. (high water): average of the high water calculated over a number of years.
If drainage can take place through the valve, Ah
w
is determined according to figure 3.9.
Figure 3.9 Resulting (ground)water pressures for big differences in water levels outside, with tidal
influence, drainage feature with valve.
design g.w.l.
L.W.
L.L.W.S.
0.3 m

Ah
w
w
valve
Project number
415660-21 v04
Date
December 2007
Page
79 of 279
hydrostatic pressure
hydrostatic pressure
+ A
a
A
a
= i
a

w
A
p
A
p
= i
p

w
i
p
d
1

w
i
a
d
2

w
d
2
d
1
AH
w;d
Loads due to ground water flow
When a sheet pile wall is placed in a permeable layer and there is ground water flow under the
sheet pile wall, this will lead to vertical pressures on the grains. These pressures will influence
the effective stresses. At the side of the retaining wall with highest pore pressures these
stresses will increase and on the other side of the wall the effective stresses will decrease. On
the other hand there is a decrease in the water pressure on the side of the sheet with the highest
pore pressures and an increase on the other side.
In case of homogeneous soil the following approach shown in figure 3.10 will suffice.
;
2 1 2
0.7
w d
a
H
i
d d d
A
=
+
(3.31)
;
1 1 2
0.7
w d
p
H
i
d d d
A
=
+
(3.32)
Figure 3.10 Simplified method to approach the effects due to groundwater flow.
Project number
415660-21 v04
Date
December 2007
Page
80 of 279
The current pressure causes the effective unit weight of the soil (') at the back side of the wall
to increase by i
a
w
and decrease by i
p
w;d
at the front of the wall.
At the same time the resulting water pressure due to i
a
and i
p
must be adjusted in the same
manner as is shown in figure 3.10.
For non-homogeneous soils d
1
and d
2
respectively are the summated layer thickness of the less
permeable layers. These are determining for the ground water flow.
A more precise determination of the water pressures can be obtained using the theory of the
potential flow. This procedure is shown in figure 3.11. For this method it is assumed that there
is a badly permeable layer located at a depth of approximately 50 % of the retaining height
below the tip of the wall. One line along the sheet pile walls edge and one streamline along the
upper edge of the badly permeable layer are chosen to be the ultimate streamlines. Between
these lines the so-called square net is constructed or calculated.
The pressure acting on the active side of the sheet pile wall can be calculated by subtracting the
course of the pressure over the streamline bordering on the sheet pile wall from the hydrostatic
pressure. For the pressure on the passive side the pressure course should be added to the
hydrostatic pressure.
For building pits on which large excess pressures act, these pressures can have great influence
on the dimensions of the sheet pile construction.
Project number
415660-21 v04
Date
December 2007
Page
81 of 279
a.
b.
impermeable layer
significant
active failure surface
equipotential lines
piezometer
sand
Ah = h
wij
/ n
i
= 4.50 / 15 = 0.30 m
water pressure excess water pressure
pressure difference
pressure difference
hydrostratic water
pressure
streamline
h
wij
13 Ah
2 Ah
h
w13
h
w2
2 Ah
w
ij2
= w
2
= w
Aw
2
= h
w2

w
= 2 Ah
w
=
= 2 . 0 . 3 . 10 =
= 6 kN/m
2
w
13
Aw
13
w
ij
Aw
g
w
g
w'
g
Aw w
g
= w
g
Aw
g
w
g
= (15+7) . 10
w
g
= 220 kN/m
2
g
w
2
2w'
2
decisive passive
failure surface
Figure 3.11 a. and b. Earth pressures and water pressures for groundwater flow beneath a sheet pile
wall.
Project number
415660-21 v04
Date
December 2007
Page
82 of 279
Figure 3.11 c. and d. (continued).
c.
d.
number of squares n
piezometer
passive active
[kN/m] [kN/m]
h
w
h
w2
= 13 Ah
h
v13
= 13 Ah
h
ij
h
w13
= 2 Ah
h
w2
= 2 Ah
4
3
2
1
0
(+7.0) 0 1 2 3 4 5 6 7 13 14 15
number of squares n
h
w

[
m
]
0 20 40 0 10 20
h
ea
t = h
ep
p
ea;p;hw
= w
g
. k
;g
p
ea;a;h0
= . h
ea
. k
;a
p
ea;a;hw
= Aw
g
. k
;g
p
ea;p;h0
= . t . k
;p
1
2
3
4
5
14
13
9
Project number
415660-21 v04
Date
December 2007
Page
83 of 279
The change in the earth pressures on the active and passive side of the sheet pile wall can be
calculated in the following manner:
On the active side there is an increase in earth pressure of:
; a h a
K w o = A
(3.33)
On the passive side there is a decrease in earth pressure of:
; p h p
K w o = A
(3.34)
Wave loads
Differences in water pressures due to waves that are caused by wind and that are higher than
0.5 m working on an earth retaining structure, located at an open water area, must be taken into
account [Sainflou; 1928].
For the determination of the wave pressures on the vertical wall the following load cases must
be investigated:
the waves are entirely or partially reflected;
the construction is loaded by waves that break a small distance from the wall.
Which situation is applicable depends on the following factors:
the water depth in front of the wall;
the wave movement;
the morphological and topographical circumstances of the bottom in front of the wall.
In connection with this the probable frequency of known movements of the water as well as
those that have been calculated based on values from an analysis of the wind and the waves
predicted from this, are assessed.
In case of a high vertical wall and such a water depth that the waves do not break, the incoming
waves are superposed, so a standing wave pattern originates. If the waves come in
perpendicular to a vertical wall and no relevant energy losses are expected, (reflection
coefficient o = 1.0), the wave height of this standing wave is two times the height of the
incoming wave. There is also a raise of the average water level outside with respect to the
water level outside in rest. In order to determine the load, the dynamic wave pressure may be
simplified to a quasi-static linear water pressure for wave tip and wave trough, that is added or
respectively subtracted from the hydrostatic pressure of the water level outside in rest
according to figure 3.12.
Project number
415660-21 v04
Date
December 2007
Page
84 of 279
Figure 3.12 Loading due to standing wave pattern for complete reflection.
In the figure:
H
d
= 1.5 H
s
; H
s
is the significant height of the incoming waves;
L
s
is the length of the incoming significant wave (L
s
= 1.56 T
2
); T is the period;
h is the raise of the average water level outside due to the wave load; this
corresponds with the difference between the water level outside in rest and the
average water level of the reflected wave at the sheet pile wall.
This raise can be approached using the following formula:
2
2
coth
d
s s
H d
h
L L
t t
=
[m] (3.35)
d is the water depth for the water level outside in rest;
w
is the unit weight of the water;
p
1
is the increase in pressure at the tip of the wave or the decrease in pressure at the
trough of the wave at the foot of the construction due to the wave load, where the
following is valid:
1
2
cosh
d
w
s
H
p
d
L
t
=
[kPa] (3.36)
The resulting water pressures and earth pressures are calculated in the manner shown in figure
3.13
water level outside, in rest water level outside, in rest
average water level outside, as a result of a
standing wave pattern
d
p
1
p
1
h
H
d
H
d
H
d
H
d
Project number
415660-21 v04
Date
December 2007
Page
85 of 279
Figure 3.13 Distribution of the wave pressure on a vertical sheet pile wall for complete refraction, as
well as the difference in water pressure for the top of the wave and for the minimum of
the wave according to [Sainflou; 1928].
In which:
Ah is the difference between the water level outside in rest and the
groundwater level;
p
0
is the maximal difference in water pressure at the level of the
groundwater table, for which the following counts:
, )
1 o
H h h
p p d
H h d
+ A
= +
+ +
[kPa] (3.37)
p
z
is the difference in water pressure at the minimum level of the wave
according to:
, )
z
p H h h = + A
[kPa] (3.38)
Wave loading due to shipping
Shipping can cause wave loads by bow waves and stern waves, and also secondary shipping
waves. Also in a canal a decrease of water level will occur due to the return stream. The sum of
both effects of the decrease of water level and the stern waves is called the (in dutch:)
haalgolf.
For ships suitable for canals (excluding the double pushed load carriers (in dutch:
duwbakeenheden)) the calculation must be carried out with a maximal decrease of water level
of 0.8 m and a maximal wave height of 0.6 m above the water level in rest for bow waves, stern
incoming wave
water level outside
in rest
water level outside
sheet pile construction
standing
wave
area of reflection
undisturbed area
Ah
d
s
p
x
g.w.l.
d
H h
L
H
H
L
p
0
Project number
415660-21 v04
Date
December 2007
Page
86 of 279
waves and secondary waves. For sea ships and double pushed load carriers these values must
be raised to 1.5 m and 1.1 m respectively. A hydrostatic course of the water pressure can be
maintained from the decrease of water level or the wave tip. The decrease of the water level is
almost always the most critical [CUR-report 89-9; 1989].
3.2.5 Anchor forces and strut forces
Anchor forces and strut forces only count as loads for:
cofferdams; because the walls function as retaining walls, or as anchor walls, the anchor
force from one wall can be considered as a load on the other wall;
pre-tensioned anchors; the pre-tension force can be considered as a load;
inclined anchors; the vertical component of the anchor force must be considered as a load.
When using pre-tensioned anchors the anchor forces must be treated as independent loads.
For an anchor plate or an anchor wall and anchor piles the anchor forces depend on the
movement of the retaining construction and follow from the calculation.
3.2.6 Ice loads
Generally ice loads will not be determining for the design. But it must be taken care of that the
ice loads cannot damage essential parts of the retaining construction such as anchor
connections.
For retaining walls located near open water the loads due to ice, ship impacts and waves do not
have to be taken into account simultaneously. The size of the ice load or the sum of wave loads
and ship impacts is determining.
In case they are relevant, the following situations must be considered:
drifting ice;
floating ice.
The analysis may be omitted if measures are taken that prevent ice loads.
Generally, for determining the horizontal ice loads on sheet pile walls, a layer of ice of 0.5 m
thick and a compression strength of 1.5 MPa for salt water or 2.5 MPa for fresh water may be
assumed. Because this load will vary strongly over the contact surface between 0 and the
compression strength, a flattening coefficient of 0.33 may be applied to the calculated line
load. These values may be considered as design values.
Project number
415660-21 v04
Date
December 2007
Page
87 of 279
The calculation must be made with at least:
an average horizontal line load of 250 kN/m (salt water) or 400 kN/m (fresh water),
assumed at the location of the least favourable water level that has been taken into
consideration;
a local load of 1.5 MN for the strength calculation of construction details in case they can
be loaded by ice.
Figure 3.14 Ice load as a result of temperature increase.
Remark:
In tidal areas, where ice floes originate as a result of differences in water levels, a line load of
100 kN/m' may be used in the calculations.
These values only count for long-drawn sheet pile constructions. For short walls, such as for
bridge pillars, the ice load must be determined based on data from literature [EAU; 1990].
In tidal areas possibly also vertical ice loads must be taken into account.
When an ice floe collides with a construction, the impact force can be approached if the
thickness and the speed of the ice floe are known. The impact force can be found by regarding
the loss of kinetic energy E
kin
= mv
2
due to the increasing counteracting force that is
necessary for the ice floe to fail.
Ice forces can also occur due to the expansion of a layer of ice that is trapped between the sheet
pile wall and a fixed construction when the temperature rises.

t
h
i
c
k
n
e
s
s

[
m
]
temperature increase
per hour
8
6 4
1

0

0 150 300
1

ice load [kN/m]
Project number
415660-21 v04
Date
December 2007
Page
88 of 279
These forces depend on the original temperature of the ice, of the rate of the temperature
increase and of the thickness of the ice layer.In this case the forces on the sheet pile wall can be
obtained from figure 3.14.
3.2.7 Ship impact
Generally so-called fenders are placed in front of the sheet pile wall in order to cut off the
impact load peak. If these fenders are connected to the sheet pile wall, the flattened impact load
can be considered to be a load of the sheet pile wall itself.
A fender together with the sheet pile wall form an elastic system in which the working-power
of movement is transformed into potential energy.
Paragraph 13.2 of the EAU [EAU; 1990] contains detailed considerations about the
determination of the working-power of the ship that must be taken by the combination of the
fender and the sheet pile wall. A consideration is also dedicated to determine the spring
constant of dolphins and fenders.
Data about the speed of ships when they berth can also be found in the EAU in paragraph 5.3.
The speed perpendicular to the quay vary between 0.1 and 0.75 m/s.
The spring constant c is determined by the elastic stiffness of the fender and the resistance
against deflection of the sheet pile wall. Figure 3.15 shows this principle.
The spring constant of the fender follows from the used springs in the fender. The stiffness of
the sheet pile wall can be derived using a computer programme for sheet pile calculations,
where the sheet piles wall is considered as an elastic supported beam. The modulus of subgrade
reaction of the soil must match the short loading time; for example the high values from table
3.15 in 3.4.
The magnitude of the deflection of the sheet pile wall due to horizontal loading at the impact
point is considered.
In this manner the deformation behaviour of the fender sheet pile wall system is determined.
Generally the behaviour will not be completely elastic, although the goal must be to have as
little plastic deformation a possible, because this leads to a remaining displacement. Figure
3.16 shows the general course of the deformation. If a demand has been made concerning the
outward bending of the sheet pile wall, a given kinetic energy E
kin
can be used to calculate the
spring stiffness of the fender sheet pile wall system for which the demand is met.
Project number
415660-21 v04
Date
December 2007
Page
89 of 279
Figure 3.15 Principle of the spring-system fender sheet pile wall.
Figure 3.16 Displacement-force-diagram of a system fender sheet pile wall.
From the deformation diagram of figure 3.16 it follows that for a deformation demand u s u
max
the potential energy that can be taken up E
pot
is:
2
1 1
1
2
pot pl
E cu cu u = +
(3.39)
in which:
c = F
r;h;max
/ u ;
F
r;h;max
is the maximum horizontal force that can be resisted by the fender
sheet pile wall system;
u
1
is the displacement at the beginning of the plastic deformation;
u
pl
is the plastic deformation.
soil spring
fender spring
v [m/s]
C = tan o = F
r;h;max
/ u
u
F
rih
F
r;hmax
u
i
u
max
u
pl
E
p.t
o
Project number
415660-21 v04
Date
December 2007
Page
90 of 279
And therefore:
2
1 1
1 max
1
2
kin pot
kin pl
pl
E E
E cu cu u
u u u
=
= +
=
(3.40)
From which follows that:
min
2 2
max
2
kin
pl
E
c
u u
>
(3.41)
If there is only elastic deformation, this formula becomes:
min
2
max
2
kin
E
c
u
>
(3.42)
In this case the soil behind the sheet pile wall does not fail at any location.
If the demand is made that the soil behind the sheet pile wall may not fail during the ships im-
pact, the previous formulae lead to the following:
2
; ;max;
min
2
r h d
kin
F
c
E
>
(3.43)
in which:
F
r;h;max;d
is the design value of the maximum horizontal force that can be resisted by the
fender sheet pile wall system.
In order to determine this maximum horizontal force the high design values for the
shear resistance parameters of the soil must be filled in. The ship may also fail locally
due to the impact. If the maximum impact force is S
max
, the maximum stiffness c
max
of
the fender sheet pile wall system can be calculated from:
kin
E
S
c
2
2
max
max
s (3.44)
3.2.8 Boulder forces
The forces of ships act on the sheet pile wall via boulders, either directly or indirectly. Gener-
ally these forces have a static character, and they are determined by the size of the ship, the
water currents and the wind load.
The boulder forces are especially important for the design of the anchoring construction. Table
3.4 gives the boulder forces for sea ships. These have been taken from paragraph 5.12 of the
EAU [EAU;1990].
Project number
415660-21 v04
Date
December 2007
Page
91 of 279
Table 3.4 Design values of the boulder forces for sea ships.
water displacement [kN] boulder force [kN]
< 20,000 100
< 100,000 300
< 200,000 600
< 500,000 800
< 1,000,000 1,000
< 2,000,000 1,500
> 2,000,000 2,000
For the design this boulder force is inserted in the calculation model as design value in the
direction perpendicular to and away from the sheet pile wall.
For ship berths with strong water currents, the values from table 3.4, higher than 500,000 kN,
must be increased by 25 %.
For main bollards at the ends of berths for large sea ships with strong water currents, the values
from the table must be increased with 2,500 kN for ships with a water displacement of maxi-
mally 1,000,000 kN. For ships with a water displacement that is greater than 1,000,000 kN they
must be doubled. In the Netherlands the inland navigation has been divided into the classes I to
VI and recreational navigation. Table 3.5 shows the bollard forces for these classes [CVB;
1989]. The second column contains the design values that should be used in the new testing
method.
Table 3.5 Design values of the boulder forces for inland ships [CVB; 1989].
ship class boulder force [kN]
(design value)
recreational ships 55
professional shipping:
class I & II
class III & IV
class V & VI
140
210
280
3.2.9 Loads due to earthquakes and vibrations
In earthquake areas the influence of earthquakes is taken into account by:
using a reduced shear resistance of the sand embankment and possible sand layers in the
subsoil. This reduced shear resistance is the result of excess pore pressures, which originate
due to horizontal soil movements. Eventually, these can lead to complete saturation or
liquefaction of the soil layers concerned. This phenomenon can only take place in soil
layers which are not densely packed and are saturated with water.
By carrying out cyclic triaxial tests on representative soil samples the reduction in
question can be determined.
Project number
415660-21 v04
Date
December 2007
Page
92 of 279
applying a horizontal acceleration force on the soil mass in the active and passive failure
zone near the sheet pile wall and the anchor wall according to the pseudo static method,
which is shown in figure 3.17.
Figure 3.17 Determination of loading due to earthquakes.
Then the extra horizontal force for the sliding soil masses is:
;
h
h earthq g
a
F F
g
A =
(3.45)
in which:
a
h
is the peak value of the horizontal acceleration during the design-earthquake;
g is the strength of the gravity field;
F
g
is the wet unit weight of the sliding soil wedge.
The reduction of the shear resistance develops gradually during the earthquake and therefore it
often reaches its maximum at the end of the earthquake. This is why the effects of a and b will
not occur simultaneously.
3.2.10 Loads due to compaction
If the soil behind the wall is placed and compacted in layers, an increased earth pressure must
be used. Figure 3.18 shows how the representative value of this increased earth pressure should
be determined.
F
h;earthq
F
h;earthq
F
h;earthq
F
h;earthq
F
g
F
g
F
g
F
g
Project number
415660-21 v04
Date
December 2007
Page
93 of 279
Figure 3.18 Determination of the representative value of the earth pressure after compaction of the
fill.
3.2.11 Loads due to swelling soil
If relevant, forces due to swelling of the soil must be taken into account. For example forces
due to freezing of the soil that is being retained or due to excavation have to be considered.
3.2.12 Loads due to temperature changes
In principle the influence of temperature changes must be taken into consideration for sheet
pile walls or parts of them that have been exposed to climate influences such as direct sun ra-
diation during their economical lifetime. As a rule this influence is not big, except for struts.
For increasing temperature the strut forces can increase strongly and there is danger of outward
buckling.
z
infl
depth of influence of the compaction machine
b width of the compaction plate of the roller
strut or anchor
sheet pile wall
d
z K
;p;rep
2 z
infl
z
infl
0.4 b
d
z K
;a;rep
Project number
415660-21 v04
Date
December 2007
Page
94 of 279
Because of temperature influences the sheet pile wall will deform. For relatively stiff and
strong sheet pile walls the soil pressures on the active side can increase if the sheet pile wall
wants to displace in that direction.
Generally this will not have to be taken into consideration during the judgement of the ultimate
limit state, because the starting point is that one or more plastic hinges have formed in the sheet
pile wall.
If, due to the temperature change, the wall wants to deform in the direction of the passive zone,
this will be counteracted by the anchor. This will cause the force in the anchor to increase.
Usually the anchor will be able to resist this increase in load without deforming very much. But
if there are little reserves in the anchors, the anchors will not be able to prevent the temperature
deformations in the direction of the passive zone.
A great number of repetitions of this situation can lead to large remaining deformations.
3.3 Parameters of the sheet pile wall
3.3.1 Introduction
The parameters of the sheet pile wall, such as the moment of inertia and the moment of resis-
tance can generally be given by the suppliers. For steel sheet pile walls these parameters have
been collected in profile books. The values are valid for the beginning of yielding in the ulti-
mate fibre. The fully plastic moment of resistance is not used.
If the design allows some corrosion or damage, the values from the table should be adjusted to
the decreased thickness of the material. Concerning the bending tensional stress, compression
strength and the Young's modulus, NEN 6720; 1991 must be used for concrete, NEN 6760;
1991 must be used for wood and NEN 6770; 1991 must be used for steel [NNI; 1991]. A spe-
cific problem related to steel sheet pile walls is the phenomenon oblique bending. A different
behaviour occurs in that situation.
In case strict deformation demands are made for a sheet pile wall, the use of single sheet piles
is not recommended. For slight slipping in the interlocks, the calculations must be made with a
considerable reduction of the strength and stiffness, compared to the values given in the suppli-
ers tables, which are based on displacement/slipping in the interlocks.
3.3.2 Oblique bending
Steel sheet pile walls made up of so-called U- or Z- shaped profiles have been used for earth-
and water-retaining structures for about three quarters of a century. In the U-shaped profiles the
interlocks are located in the centre line of the sheet pile wall; in the Z-shaped profiles they are
located on the outside of the sheet pile wall (at the utmost fibre). Both profiles have been de-
signed in such a manner that through a favourable use of materials good strength and stiffness
properties can be reached. The interlocks have been shaped in such a manner that the profiles
can simply and reliably be inserted in the soil and for the most part the constructive connection
remains assured.
In an early stage experiences of slipping in the interlocks have been published in cases where
U-shaped profiles were used in weak soil and water. This phenomenon is called oblique bend-
ing.
Project number
415660-21 v04
Date
December 2007
Page
95 of 279
In the Netherlands, the long existing problem has become topical again due to the experiences
of the Nederlandse Spoorwegen (NS) during the construction of the Schipholtunnel. During
construction, deformations of the sheet piles were recorded, that exceeded the calculated de-
formations, which were determined with the assumption that all sheets cooperate completely.
These large deformations could only be explained by not cooperating sheets, which implied
that slipping occurred in the interlocks. Since then, for sheet pile walls that are made up of
single U- shaped sheets or double U-shaped sheets, that are made up of two single sheets that
are connected to each other by means of punching, the NS applies a reduction to the moment of
inertia and the moment of resistance given by the supplier.
Gemeentewerken Rotterdam and the Rijkswaterstaat Directie Sluizen en Stuwen in Utrecht have
also carried out measurements at projects that were executed under their supervision, which
gave rather varying results. The measurements in Rotterdam were carried out by the sheet pile
walls for the trench for the construction of the Metrotunnel in the Schiedamseweg. The sheet
pile wall reached down to approximately 23 m depth below ground level and retained mainly
clay and peat layers. The measurements of Rijkswaterstaat were carried out in the water retain-
ing structure between the building pits for a road in Amelisweerd.
In the subsoil of Rotterdam it appeared that there were no, or hardly any displacements in the
interlocks.
The test in Amelisweerd however showed almost complete slipping. In the building pit in Rot-
terdam two strut layers were applied and in Amelisweerd none. Of course this has affected the
results.
The occurrence of oblique bending is naturally counteracted by the soil. But also constructional
factors can play a part, such as the number of strut layers or anchor levels. This means that
oblique bending does deserve attention, but does not always have to be taken into account.
The following consideration is based on the report Oblique Bending of the Commissie Dam-
wanden of the KIvI [Commissie Damwanden; 1987].
Behaviour of profiles with interlocks in the centre line
Due to loads caused by soil and water the elements will deform like isolated elements with
their own moment of resistance and moment of inertia. Initially there will be very small mutual
shifting in the interlocks. Depending on the stresses in these interlocks, local shear stresses are
activated. These shear stresses originate because of for example the initial deformations of the
sheet pile wall in the soil and the presence of soil particles in the interlocks. The sheet pile
walls will behave as a monolith, up to the moment that the maximum value of the shear forces
are reached. In this phase the moment of resistance and the moment of inertia are the same as
those for completely co-operating sheet pile profiles and the differences with the values given
by the manufacturer will be practically negligible. The remark must be made that activating the
required shear forces in the interlocks involves very slight displacements in those interlocks
according to mechanism I in figure 3.19. From the moment that the maximum shear forces are
reached, the bending behaviour of the sheet pile wall will be identical to the behaviour of the
separate sheet pile sheets.
This means a reduction of the moment of resistance and the moment of inertia:
the sheet pile wall elements that have been inserted as single sheets will show a
displacement of the bending axis from the centre of the sheet pile wall to the centre of the
elements (see figure 3.19, mechanism II);
the elements that have been inserted as double sheets with shear-resistant intermediate
interlock (due to welding or punching) show a distortion of the bending axis from the centre
of the combined profile. The sheet pile elements bend in the direction perpendicular to the
Project number
415660-21 v04
Date
December 2007
Page
96 of 279
I
II
III
IV
axis of the sheet pile wall as well as in the plane of the sheet pile wall. This is why the term
oblique bending is used (see figure 3.19, mechanism III);
when elements are used that are made of three sheets with shear-resistant connections in the
interlocks, it appears that the axis shifts slightly. This occurs together with a decrease in the
moment of resistance and the moment of inertia that can be neglected (see figure 3.19,
mechanism IV);
when elements are used that are made of four or more sheets it appears that the
displacement and/or the distortion of the axis of bending can be neglected.
Figure 3.19 Bending axes for a theoretical sheet pile wall with completely fixed interlocks (I), a wall
consisting of single sheets (II), double sheets (III) and a wall consisting of elements of 3
sheets (IV).
Factors that influence the oblique bending for U-shaped profiles
Oblique bending does not only depend on the interlocks between the U-shaped sheet pile pro-
files. The freedom that is required for the displacement in the interlocks, can be limited by a
number of factors, for example:
the soil type in which the sheets have been inserted:
due to the penetration of sand and or silt particles in the interlocks the shear resistance in
those interlocks will increase. Also, for stiff soils the shear resistance in the interlocks will
be high due to the slight deformations;
application of lubricants:
in order to decrease the friction in the interlocks between the elements, lubricants or fillers
Project number
415660-21 v04
Date
December 2007
Page
97 of 279
are sometimes used. When these are used the interlock friction will be low for a longer
period of time;
the number of supports or anchors:
the shear force in the interlocks is proportional to the shear forces (in dutch: dwarskracht)
in the wall. By using one or more supports like struts or anchors, the shear force will de-
crease. This means that the interlock is capable to resist higher shear at those supporting
points.
applying shear-resistant connections in the interlocks:
by locally applying shear-resistant connections in the interlocks between the elements,
oblique bending is limited. The interlock can be fastened near the top of the sheet pile wall,
for example using a concrete top beam or by welding. Constructive measures to limit the
displacements near struts and anchors, are also possible.
When one shear-resistant connection is applied, for example only at the top of the sheet pile
wall, the displacement in the interlock is not prevented. In the calculation one should take
the oblique bending into account.
In the user phase the shear-resistant connections will only be effective, if they have been in-
serted before the soil pressure has developed;
the manner in which the sheet pile elements are installed (driving or vibrating): as a result
of heavy driving high stresses will have originated in the interlocks, causing deformations
that can hinder slipping. If the sheets are installed by vibrating the deformations in the
interlocks will be smaller and the resistance against displacement in the interlocks will also
be smaller.
Reduction factors R
1
and R
w
The effect of oblique bending for U-shaped steel sheet pile wall profiles is taken into account
by using so-called reduction factors, to be applied on the moment of resistance as well as on
the moment of inertia, as indicated in the tables of the suppliers. As was mentioned earlier, the
magnitude of the reduction factors depends on the soil condition, method of strutting and the
connection between the sheet piles or sheet pile elements.
Figure 3.20 shows the theoretically derived relation between the reduced stiffness and strength
of diversely combined sheet pile profiles on one hand and the percentage of the available fric-
tion with respect to the required friction for an optimal connection on the other hand. This rela-
tion is in agreement with the guidelines for the reduction factors that are to be used for deter-
mining the moment of inertia and the moment of resistance given in table 3.6. These reduction
factors also depend on the soil condition and the construction circumstances such as the num-
ber of struts and anchor layers and the shear-resistant measures in the interlocks.
The reduction factors presented in table 3.6 have been derived from different tests and publica-
tions (see figure 3.20). In fact they have relations with the expected behaviour of the sheet pile
wall under the working load.
For multiple intermediate supporting points, the ability to redistribute loads is taken into ac-
count in the size of the reduction factors.
The reference moments of resistance and of inertia are respectively the moments of
resistance and inertia that are present if all pairs of sheets co-operate completely. This
means that the occurring shear stresses in the punched interlocks as well as in the open
interlocks can be resisted without the sheets displacing mutually.
When sheets are used with h/b < 0.2 (see figure 3.21) the reduction factors for the moment
of resistance (R
W
) and the moment of inertia (R
I
) given in the table must be lowered. In case
Project number
415660-21 v04
Date
December 2007
Page
98 of 279
single sheet
double sheet
triple sheet
wall
not partly completely
co-operation
E
D
DR
100
50
DR (I)
DR (W)
D (I)
E (W)
E (I)
100 % resp.
pres
l
l

100 %
pres
W
W

0 0.5 1.0
[%]
0.43
92.8
78.5
57.0
35.5
52.7
51.5
there is sand both behind the wall and at the toe, the factors must be lowered by 0.1. For all
other cases they must be lowered by 0.05.
In case there is sand above the ground water level, R
W
and R
I
may be increased with 0.1 up
to a maximum value of 1.0.
Figure 3.20 Acting moment of inertia and moment of resistance as a function of the shear force in
the not wellded or punched interlocks for Larssen 23 (Hoesch Estel publication 26-09-
1980).
In case shear-resistant connections are applied in the driven interlocks, R
W
and R
I
may be
taken to be 0.85 for the section between these connections.
A double sheet is a sheet pile that is made up of two single sheets, where displacement of
the two sheets with respect to one another is largely prevented by the interlock through
which they are connected. Usually this is realised by applying punching in a factory by he
supplier or by wellding.
Driven interlocks are the loose interlocks that connect the sheet pile elements to one
another.
The required number of punching points or welds per [m
1
] of the interlock should be given
to the supplier in an early stage. This number can be found using the calculation method
according to 3.3.3.
Project number
415660-21 v04
Date
December 2007
Page
99 of 279
nutral axis wall
b
h
Table 3.6 Reduction factors with respect to the reference moment of resistance and moment of
inertia
reduction factor
sheet pile
element
(number of
single sheets)
number of
supporting
points
driving interlock
fixed = 1
loose = 0
soil
*
s/s: sand/sand
n/s: no sand/sand
n/n: no sand/no sand R
w
R
1
1 all 0.6 0.35
2 0 0 s/s 0.8 0.7
n/s 0.7 0.6
n/n 0.7 0.6
1 s/s 0.9 0.8
n/s 0.8 0.7
n/n 0.7 0.6
1 0 s/s 0.9 0.8
n/s 0.8 0.7
n/n 0.8 0.7
1 s/s 1.0 0.9
n/s 0.9 0.8
n/n 0.8 0.7
2 0 s/s 1.0 0.9
n/s 0.9 0.8
n/n 0.9 0.8
1 s/s 1.0 1.0
n/s 1.0 1.0
n/n 0.9 0.8
3 0 and 1 all 1.0 1.0
4 0 and 1 all 1.0 1.0
*
see figure 3.22
Figure 3.21 Dimensions h and b for U-shaped sheet pile profiles.
Project number
415660-21 v04
Date
December 2007
Page
100 of 279
In table 3.6 the distinction has been made between two types of soil material: sand and
not sand.
Sand indicates the soil types sand and silt. Not sand includes clay, peat and
gravel, as well as (free) water. Figure 3.22 shows this schematically.
The values for R
W
and R
I
given in the table are valid for a soil mass that is saturated with
water. In the case of unsaturated soil the values for R
W
and R
I
may be increased to the s/s-
values from table 3.6.
For a sand fill behind the wall, after the wall has been placed, the reduction factors must be
determined based on the soil type not sand.
A favourable effect of a shear-resistant connection between the sheets may only be expected
if the toe of the wall is inserted in a sand layer. As an indication for the part of the sheets
that must be in sand, one can think of a section from the tip of the sheets to the first point of
zero moment located above it.
For the cases in which there is no sand near the toe of the wall, the reduction factors may
not be increased, even in the case of shear-resistant connections in the driven interlocks.
The reason for that is that such connections can not be made in the part that is entirely
surrounded by soil.
Supporting points are supports that change the shear force in the sheets considerably. A
restraint in the soil mass may not be considered to be a supporting point.
Shear-resistant connections must be applied at locations where the expected displacements
in the interlocks are large. Generally this will be the case at places where the moments
change sharpest (dM/dz is large).
A favourable effect of shear-resistant connections may only be expected if the connections
are applied before the load acts on the wall.
Figure 3.22 Supporting conditions for the sheet pile wall to use in table 3.6.
''sand'' or
''no sand''
''sand'' or
''no sand''
toe
of the
wall
Project number
415660-21 v04
Date
December 2007
Page
101 of 279
clay
peat
clay
sand
anchor
envelope of moments
Examples for the determination of the reduction factors R
W
and R
I
for a wall that is made of
sheet pile elements with two (double) U-shaped profiles
Example 1
Data (see figure 3.23):
a single anchored wall;
a moment course such as sketched in the figure below;
no shear-resistant connections have been made between the double sheets;
h/b according to figure 3.21 > 0.20.
Factors:
R
W
= 0.8 and R
I
= 0.7.
Figure 3.23 Schematic view of example 1.
Project number
415660-21 v04
Date
December 2007
Page
102 of 279
sand
clay
sand
anchor
envelope of moments
Example 2
after the excavation a concrete floor is poured, which also functions as a strut;
before the floor is poured, the moment course in the wall must be like the figure below;
no shear-resistant connections have been made between the pairs of sheets;
according to figure 3.21 h/b > 0.20.
Factors:
Behind the wall there is mostly sand.
For the phases up to and including the excavation:
R
W
= 0.9 and R
I
= 0.8.
As soon as the concrete floor has hardened:
According to situation 2: R
W
= 1 and R
I
= 0.9.
Project number
415660-21 v04
Date
December 2007
Page
103 of 279
deksloof
clay
sand
envelope of moments
Example 3
an cantilever wall;
a moment course as is shown in the figure below;
the sheets are fitted with a concrete top beam, in such a manner that displacement of the
pairs of sheets with respect to one another is opposed;
according to figure 3.21 h/b > 0.20.
Factors:
At the location of the toe there is sand and the pairs of sheets are connected by a top beam.
From the moment course it appears that the sand does not reach to the first point of zero mo-
ment located above the toe.
The concrete top beam therefore has no influence on the strength and stiffness of the wall.
R
W
= 0.70 and R
I
= 0.6 are found.
Project number
415660-21 v04
Date
December 2007
Page
104 of 279
envelope of moments
sand
clay
anchor
Example 4
the part of the wall from the toe of the wall up to the first point of zero moment located
above it, is located in a sand deposit;
at the location of the point of zero moment shown in the figure, shear-resistant
connections are made;
according to figure 3.21 h/b s 0.20.
Factors:
R
W
= 0.9 0.05 = 0.85 and
R
I
= 0.8 0.05 = 0.75.
For the trajectory between the shear-resistant connections a reduction factor of 0.85 may be
used for R
W
and R
I
.
Project number
415660-21 v04
Date
December 2007
Page
105 of 279
Y
Y
X
X
y
2
y
1
n
e'
T T
n
e
3.3.3 Shear forces in the interlocks
No displacement in the interlocks
If mutual displacement in the interlocks of the sheet is not possible, the shear force T in every
interlock can be determined from (see figure 3.27):
' '
2
n
M Fe
T
I
=
(3.46)
in which:
M' bending moment in the cross-section of the single sheet;
T integrated shear force in the interlock from the cross-section to the end of the
sheet;
e' distance from the median of the single sheet to the centre of the sheet pile wall;
y local distance from the considered point w.r.t. the median of the single sheet;
I
x
moment of inertia of the single sheet w.r.t. the median of the single sheet;
F cross sectional area of the single sheet;
I
n
= I
x
+ F(e')
2
, it represents the moment of inertia of the single sheet with respect
to the centre of the sheet pile wall.
Figure 3.27 Calculation parameters for U-shaped sheet pile walls.
The shear force in the interlock per unit length along this interlock is:
2
n
DS
t
I
=
(3.47)
in which:
D is the shear force in the single sheet;
S is the static moment of the single sheet with respect to the centre line of the
sheet pile wall;
t is the shear force per unit length.
Project number
415660-21 v04
Date
December 2007
Page
106 of 279
0 0.5 1.0
100
50
pres
T
t
| =
j %
pres
l
l
Larssen 23
Displacement in the interlocks
If there is displacement in the non-punched or non-welded interlocks, as a result the shear force
in the punched interlock will increase. In order to obtain an impression of the maximum shear
force that is to be resisted, figure 3.28 can be used, in which the relation is given between the
reduced moment of inertia of the double sheet and the shear force in the interlocks that is to be
resisted.
If a maximum shear force in the interlocks is equal to |T, the shear force in the punched inter-
locks can rise to:
T (punched interlock) = |T + 1.15 (1 |) T (3.48)
in which:
| is the ratio between the present shear force in the interlock and the shear force
in the driven interlock, if displacements in the interlocks do not take place (see
figure 3.28);
T is the shear force in the interlock if no displacement takes place.
From different test loads it has appeared that per punching, a shear force of at least 75 kN can
be resisted. This value must be considered as a boundary, unless it can be proven that the
punching can resist a higher shear force.
Figure 3.28 Maximum resisting shear force.
Project number
415660-21 v04
Date
December 2007
Page
107 of 279
3.4 Parameters of the soil
3.4.1 Introduction
In order to use calculation models for the sheet pile wall analysis soil parameters are needed.
These have been collected in table 3.7 for different models.
Remarks:
|' and c' still relate to the drained situation and are shear strength parameters in relation with
effective or grain stresses.
f
undr
is the undrained shear strength that exclusively has relations with cohesive soils in the
undrained situation.
The parameters |', c', f
undr
, k and E depend on the stress and the deformation.
When the choice is made for the value of the mentioned parameter for the limit state that is to
be analysed, this must be taken into consideration.
For the ultimate limit state 1A and 1B according to NEN 6740; 1991 [NNI; 1991] the shear
strength parameters |', c', f
undr
can be taken from the top in the oc diagram, obtained from
triaxial tests or celtests. It is also possible to use the values that belong to an axial strain of 5
%. Using the peak values can lead to the danger of progressive failure. The possible differ-
ence in deformation levels for which the peak value is reached in different soil layers is not
taken into account. For layered soils, for example layers of sand, clay and peat, these differ-
ences can be large. This is why for large deformations, one layer after another can fail. There-
fore, for layered soil with large differences between the properties it is recommended to start
from a value of the shear force for an axial strain of 5 %.
Table 3.7 Required soil parameters in the different calculation models
calculation model parameters
all models configuration of soil layers
level of layer boundaries
groundwater pressure
classical model , |' and c' or f
undr
, o
supporting soil springs , |' and c' or f
undr
, o, k
h
finite element method , |' and c' or f
undr
, o, E, v and
groundwater flow models homogeneous soil: k
layered soil: k
v
and k
h
critical hydraulic gradient, according to Terzaghi and
Peck or EAU E 115

anchors , |' and c' or f
undr
, o
failure surfaces, Kranz , |' and c'
If the serviceability limit state 2 must be tested, values for a strain of 2 % may be used. This
limit state is determined by the deformation demands in the user situation and therefore actu-
ally only the deformation parameters of the soil layers are of importance. These are the
Young's modulus E and the horizontal modulus of subgrade reaction k
h
that are discussed in
figure 3.4.2. The values of the 5 % and 2 % axial strain are not strict, but in practice they ap-
pear to be satisfying for the calculation of sheet pile walls as elastically supported beams. For
computer programs where deformation depending parameters can be filled in, the real envelope
Project number
415660-21 v04
Date
December 2007
Page
108 of 279
|', c'
c
a
2 % 5 %
c
a
2 % 5 %
o
1
o
3
Ao
c'
a
|'
c'
a
E
o
c
A
=
'
up to or even past failure can be used. The envelope is usually divided in one, two or three
linear parts in the manner shown in figure 3.29.
Finite element programs sometimes use the shear modulus G instead of the Young's modulus E.
The relation is as follows:
, ) 2 1
E
G
v
=
+
(3.49)
Figure 3.29 Results of triaxial tests on soil samples.
Project number
415660-21 v04
Date
December 2007
Page
109 of 279
3.4.2 Determination of parameters
Introduction
The values for the soil properties must generally be determined according to chapter 8 NEN
6740; 1991 [NNI; 1991].
As a rule this will mean determining the so-called representative values of the concerned aver-
age soil properties for every layer.
In special cases the representative values from the collection of the single observations can be
decisive. This can be the case for example for the determination of the maximum bearing ca-
pacity of an anchor if the girder is not stiff or no present and therefore no proper force transfer
for a redivision of the forces can take place. If this is the case, then the methods described in
chapter 5 of NEN 6743; 1991 can be used for determining the representative value.
Next the manner is shown in which the different parameters for the different aspects concern-
ing the composition and the structure of the soil and for the soil properties can be found. This
can be done either in an indirect manner for example using probing results or in a direct
manner by doing tests on representative samples of the different soil layers.
INDIRECT PARAMETER DETERMINATION
Soil types and layer boundaries
The composition of the different soil layers in the soil and the location of the layer boundaries
can be derived using table 3.8. This table is a version of table 1 from NEN 6740; 1991. In this
table the soil types are named in the manner prescribed in NEN 5104; 1989 Geotechniek
Classificatie van onverharde grondmonsters [NNI; 1989]. The cone resistance qc is used as
starting point for the table.
It is possible for the soil types to be derived from the graph of figure 3.30 that is based on a
relation between the cone resistance and the friction ratio [Searle; 1979]. From the course of
the cone resistance a clear picture of the thickness and the location of the different soil layers
can be obtained.
Groundwater pressures
Normally groundwater pressures can be derived from information about the water level of
neighbouring open water or the groundwater, where the assumption has been made that the
course of the pressure is hydrostatic.
Soil properties
The representative values of the following soil properties can be obtained from table 3.8, for
the low values as well as for the high values:
the unit weight ;
the angle of internal friction |';
the cohesion for the drained situation c';
the undrained shear strength f
undr
;
the Young's modulus E.
Project number
415660-21 v04
Date
December 2007
Page
110 of 279
Figure 3.30 Relation between cone resistance, friction ratio, soil type with an indication of the soil
properties.
friction number [%]
c
o
n
e

r
e
s
i
s
t
a
n
c
e
q
c

[
M
P
a
]
u
n
d
r
a
i
n
e
d

s
h
e
a
r

s
t
r
e
n
g
t
h
f
u
n
d
r
angle of internal friction |
relative density R
e
Project number
415660-21 v04
Date
December 2007
Page
111 of 279
Table 3.8 Relation between soil type and soil parameters (table 1 from NEN 6740; 1991).
type of soil representative mean value of the soil parameters
name admixture consistency
1)
2)
[kN/m
3
]
sat
[kN/m
3
]
qc
3) 6)
[MPa]
C'p C's Cc
little silty loose
medium
dense
17
18
19 or 20
19
20
21 or 22
15
25
30
500
1,000
1,200 or 1,400
0.008
0.004
0.003 or 0.002
gravel
very silty loose
medium
dense
18
19
20 or 21
20
21
22 or 22.5
10
15
25
400
600
1,000 or 1,500
0.009
0.006
0.003 or 0.002
clean loose
medium
dense
17
18
19 or 20
19
20
21 or 22
5
15
25
200
600
1,000 or 1,500
0.021
0.006
0.003 or 0.002
sand
little silty/clayey
very silty/clayey
18 or 19
18 or 19
20 or 21
20 or 21
12
8
450 or 650
200 or 400
0.008 or 0.005
0.019 or 0.009
little sandy soft
medium
stiff
19
20
21 or 22
19
20
21 or 22
1
2
3
25
45
70 or 100
650
1,300
1,900 or 2,500
0.168
0.084
0.049 or 0.030
silt
4)
very sandy 19 or 20 19 or 20 2 45 or 70 1,300 or 2,000 0.092 or 0.055
clean soft
medium
stiff
14
17
19 or 20
14
17
19 or 20
0.5
1.0
2.0
7
15
25 or 30
80
160
320 or 500
1.357
0.362
0.168 or 0.126
little sandy soft
medium
stiff
15
18
20 or 21
15
18
20 or 21
0.7
1.5
2.5
10
20
30 or 50
110
240
400 or 600
0.759
0.237
0.126 or 0.069
very sandy 18 or 20 18 or 20 1.0 25 or 140 320 or 1,680 0.190 or 0.027
clay
organic soft
medium
13
15 or 16
13
15 or 16
0.2
0.5
7.5
10 or 15
30
40 or 60
1.690
0.760 or 0.420
not preloaded soft 10 or 12 10 or 12 0.1 5 or 7.5 20 or 30 7.590 or 1.810 peat
medium preloaded medium 12 or 13 12 or 13 0.2 7.5 or 10 30 or 40 1.810 or 0.900
variation coefficient 0.05 0.25
The table gives the low representative value of means of the concerned soil type. Within an area, defined by the row of the admix-
ture and the column of the parameter (a box), it applies:
- for , sat, C'p, C's, E', |', c' and fundr: if raising the value leads to an unfavorable situation (larger/bigger dimensions of the fou-
dation), then the right-value in the same row has to be used, or, if there is no right-value mentioned, the value of the row below;
- for Cc, Cu and Csw: if lowering the value leads to an unfavorable situation, then the right-value of the same row has to be used, or,
if there is no right-value mentioned, the value of the row below.
1) loose: 0< Rn < 0.33
medium: 0.33 s Rn s 0.67
dense: 0.67 s Rn s 1.00
2) for a natural moisture content
3) the here mentioned values of qc (cone resistance) have to be used as entry to the table and are not allowed be used in calculations
4) saturated silt/loam
5) values of Ca are valid for an increase of stresses of 100 % as a maximum
6) qc and E are standardized on an effective vertical stress of 100 kPa. To get a right entry into the table via qc, the measured values
of qc have to be converted to a level of effective vertical stresses of 100 kPa.
The factor of conversion CN has to be determined with the graph in figure 2A (NEN 6740) see also next page. Doing so, one
finds the E-modulus for a vertical effective stress level of 100 kPa. For other stress levels the E has to be converted to that
specific stress level. E has to be interpreted as E'50;rep
Project number
415660-21 v04
Date
December 2007
Page
112 of 278
Table 3.8 (continued).
type of soil representative mean value of the soil parameters
name admixture consistency
1)
Ca
5)
Csw E'50
6)
[MPa]
|'
[]
c'
[kPa]
fundr
[kPa]
little silty loose
medium
dense
0
0
0
0.003
0.002
0.001 or 0
75
125
150 or 200
32.5
35
37.5 or 40
gravel
very silty loose
medium
dense
0
0
0
0.003
0.002
0.001 or 0
50
75
125 or 150
30
32.5
35 or 40

clean loose
medium
dense
0
0
0
0.007
0.003
0.001 or 0
25
75
125 or 150
30
32.5
35 or 40
sand
little silty/clayey
very silty/clayey
0
0
0.003 or 0.001
0.006 or 0.001
25 or 35
20 or 30
27 or 32.5
25 or 30

little sandy soft
medium
stiff
0.004
0.002
0.001
0.056
0.028
0.017 or 0.005
2
5
10 or 20
27.5 or 30
27.5 or
32.5
27.5 or 35
0
2
5 or 7.5
50
100
200 or 300
silt
4)
very sandy 0.002 0.031 or 0.005 5 or 10 27.5 or 35 0 or 2 50 or 100
clean soft
medium
stiff
0.013
0.006
0.004
0.452
0.121
0.056 or 0.042
1
2
4 or 10
17.5
17.5
17.5 or 25
0
10
25 or 30
25
50
100 or 200
little sandy soft
medium
stiff
0.009
0.005
0.003
0.253
0.079
0.042 or 0.014
1.5
3
5 or 10
22.5
22.5
22.5 or
27.5
0
10
25 or 30
40
80
120 or 170
very sandy 0.004 0.063 or 0.025 2 or 5 27.5 or
32.5
0 or 2 0 or 10
clay
organic soft
medium
0.015
0.012
0.550
0.250 or 0.140
0.5
1.0 or 2.0
15
15
0 or 2
0 or 2
10
25 or 30
not preloaded soft 0.023 2.530 or 0.600 0.2 or 0.5 15 2 or 5 10 or 20 peat
medium preloaded medium 0.016 0.600 or 0.300 0.5 or 1.0 15 5 or 10 20 or 30
variation coefficient 0.25 0.10 0.20
The factor of conversion CN (figure 2A NEN 6740)
Project number
415660-21 v04
Date
December 2007
Page
113 of 278
The matching Poisson's ratio v follows from table 3.9.
Table 3.9 Values of Poisson's ratio v
soil type Poisson's ratio v
sand 0.2 0.3
clay and peat 0.3 0.45
clay and peat in undrained conditions 0.5
1)
1)
the value 0.5 implies a volume resistant material. Some computer programs
are not able to use this. It those cases a value of 0.49 is recommended.
The angle of dilation can be derived using table 3.10.
Table 3.10 Values of angle of dilation.s
soil type angle of dilation
sand: |' > 30
o
|' s 30
o
|' 30
o
0
clay and peat 0
Remark:
Essentially these values are only valid for quartz sands and normally consolidated or lightly over consoli-
dated clay.
The angle of wall friction o
From |' and the roughness of the wall surface, the angle of wall friction o for clay, silt and
gravel can be determined from table 4, NEN 6740; 1991. This table, that has also been included
as table 3.1 in 3.2.2, gives values that are valid for straight sliding surfaces.
Horizontal modulus of subgrade reaction k
h
Values of the horizontal modulus of subgrade reaction of the soil that is being compressed such
as the soil at the passive side of the sheet pile wall, can be determined in the following man-
ners:
Terzaghis method [Terzaghi; 1955], which has been worked out in detail below;
Mnards method [Amar; 1991], which is also described in more detail below.
N.B.: Mnards method is actually a direct method but for the clarity it has been included
in the indirect methods.
The values of k
h
that have been obtained using these methods do not take into account the oc-
curring deformation of the soil. The following methods do so:
k
h
values for sand for the passive side that depend on deformations can be obtained
according to DIN 4085 [DIN; 1982] in the manner described below. For clay and peat there
is not yet a method available that can be used to determine values of k
h
that depend on
deformations;
Project number
415660-21 v04
Date
December 2007
Page
114 of 278
values of k
h
have also been derived from the methods mentioned in combination with data
from practice in deep building pits in the West of the Netherlands. These will be discussed
later, in the paragraph Dutch data from experience.
Terzaghis method
For sand Terzaghi gives the following methods for the horizontal modulus of subgrade reaction
in relation to the relative density:
for u, the horizontal displacement of the wall, larger than 0.0002 D, in which D is the depth of
the toe beneath ground level:
h h
z
k L
D
=
(3.50)
and for u s 0.0002 D:
'
h h
z
k L
D
=
(3.51)
in which:
k
h
is the horizontal modulus of subgrade reaction in [kN/m
3
];
L
h
, L
h
' is a constant of which the magnitude depends on the relative density
and
the effective unit weight of the sand according to table 3.11;
z is the depth of the regarded point beneath the momentary ground level
at the regarded side of the wall;
D is the depth of the toe of the wall beneath the momentary ground level
at the regarded side of the wall;
u is the horizontal displacement of the sheet pile wall in [m];
R
e
is the relative void ratio:
max
max min
e
e e
R
e e
(3.52)
q
c
is the cone resistance in [MPa] for an effective vertical stress of 100
kPa;
e is the void ratio. This is the volume of the void divided by the volume
of the solids.
Project number
415660-21 v04
Date
December 2007
Page
115 of 278
Table 3.11 Modulus of subgrade reaction for z = D.
characteristics values of L
h
in [kN/m
3
], q
c
in [MPa]
quality low average high very high
relative void ratio R
e
0.15 0.35 0.35 0.65 0.65 0.85 0.85 1.0
cone resistance q
c
2.0 4.5 8.5 12
mod. of subgrade reaction:
dry or humid sand: L
h
L
h
'
sand under water: L
h
L
h
'
800
40,000
500
24,000
2,600
64,000
1,600
38,500
6,400
96,000
4,200
57,600
14,000
130,000
9,000
80,000
Remark:
The values for R
e
= 100 % have been determined by extrapolation of the values given by Terzaghi.
Terzaghi considers the values of L
h
he has given to be slightly too low.
The method is essentially only valid for normally consolidated sand. In case an excavation
takes place, the horizontal stress below the level of the excavation will not decrease propor-
tionally with the decrease of the vertical stress.
In this case the horizontal modulus of subgrade reaction can be found using:
, ) 0.5 ' 2 '
v hi h
h
L
k
D
o o
+
=
(3.53)
in which:
o'
v
is the effective actual vertical effective stress;
o'
hi
is the horizontal effective stress present before the excavation
(o'
hi
~ 0.5 o'
vi
; o'
vi
is the present vertical effective stress before the excavation.)
For clay the horizontal modulus of subgrade reaction follows from the undrained shear strength
f
undr
according to:
for a freely supported sheet pile wall, according to figure 3.31a:
1
14.75
sl
h
k
k
D
=
(3.54)
for a clamped sheet pile wall, according to figure 3.31b:
1
14.75 '
sl
h
k
k
D
=
or
3
14.75 ''
sl
h
k
k
D
=
(3.55)
The values of k
sl
have been given as a function of f
undr
in table 3.12.
Project number
415660-21 v04
Date
December 2007
Page
116 of 278
a. freely supported sheet pile wall b. clamped sheet pile wall
D
D'
D''
Figure 3.31 Different types of support according to Terzaghi.
Table 3.12 values of k
sl
as a function of f
undr
and stiffness.
clay soft moderate stiff stiff very stiff hard
f
undr
[kN/m
2
] < 20 20 50 50 100 100 200 > 200
k
sl
[kN/m
3
] s 8,000 8,000 16,000 16,000 32,000 32,000 64,000 > 64,000
Remark:
The values of k
sl
for weak and moderately stiff have been determined by extrapolation of the values
given by Terzaghi, and they must therefore be treated with some reserve.
Mnards Method
From the Young's modulus E
m
, which has been determined using Mnards pressiometer test,
k
h
can be determined in the following manner:
, )
1
4
13 9 10
2
h M
a
k E a
o
o

| |
= +
|
\ .
(3.56)
in which:
k
h
is the horizontal modulus of subgrade reaction in [MN/m
3
];
E
M
is the Young's modulus according to Mnard, for the pseudo-elastic area
shown in figure 3.32 in [MN/m
3
];
o is the rheological coefficient following from table 3.13;
a is a length in [m], that satisfies:
a = 2/3 D for a clamped wall;
a = D for a freely supported wall, in which D is the length of the sheet pile
sheet below the ground surface at the low side of the sheet pile wall.
Table 3.13 Rheological coefficient o.
peat clay silt sand sand and gravel
over-consolidated 1 1 2/3 1/2 1/3
normally consolidated 1 2/3 1/2 1/3 1/4
not consolidated 1/2 1/2 1/3 1/4
Project number
415660-21 v04
Date
December 2007
Page
117 of 278
pl
pressiometer pressure p
volumetric
strain
adaption pseudo-elastic area plastic area
Ap
AV
V
m
V
0
V
0
AV / V
0
= 1
In the pseudo-elastic area of the curve in figure 3.32 the following formula is valid:
, ) 1
M m
p
E V
V
v
A
= +
A
(3.57)
in which:
AV is the change in volume of the cylindrical drilling hole in the pseudo-elastic
area;
p is the pressure exerted on the wall of the drilled hole;
p
1
is the limit pressure;
V
m
is the volume of the pressure cell during the measurement of Ap and AV;
v is the contraction coefficient, which is usually set to 0.33;
v
Figure 3.32 Result of a pressiometer test according to Mnard.
Project number
415660-21 v04
Date
December 2007
Page
118 of 278
A B C
H
% of passive
soil resistance
p
ea;h;p;rep
100
67
50
33
0.09 S
B`
wall displacement
S
B`
(c
0
)
; ; ;
;33 50%
0.50 0.33
0.09 0.04
ea h p rep
h
B
p
k
S
0.04 S
B`
0.2 S
B
Method according to DIN 4085
Essentially this method is only valid for dry sand. The values for the horizontal beddingcon-
stant are related to the deformations.
The manner in which this can be done for the different possibilities of deformation and rotation
is shown in figures 3.33 and 3.34. Table 3.14 gives values for S
B
for the different displacement
types in the case of loose and stiff sand. S
B
is the displacement required for the development of
the complete soil resistance. S
B
is expressed as a percentage of the retaining height.
Table 3.14 Values for S
B
in percent for different displacement types for loose and stiff sand accord-
ing to figure 3.33.
way of displacement loose sand stiff sand
A 30 10
B 10 5
C 15 5
Figure 3.33 Way of displacement and the meaning of A, B and C.
Figure 3.34 Construction of a displacement dependent bedding constant.
Project number
415660-21 v04
Date
December 2007
Page
119 of 278
Dutch experience
In table 3.15 data has been collected from measurements made in the Dutch practice. In combi-
nation with the graph in figure 3.35 the characteristic of the spring constant can be determined.
Table 3.15 Representative values for the low- and high- average of the horizontal modulus of sub-
grade reaction for stress increase, belonging to figure 3.35.
k
h;1
[kN/m
3
] k
h;2
[kN/m
3
] k
h;3
[kN/m
3
]
1)
p
0
< p
h
< 0.5 p
ea;h;p;rep
0.5 p
ea;h;p;rep
< p
h
< 0.8 0.8 p
ea;h;p;rep
< p
h
< 1.0
1
2)
2
3)
1
2)
2
3)
1
2)
2
3)
zand:
loose
moderate
dense
q
c
[MPa]
5
15
25
12,000
20,000
40,000
27,000
45,000
90,000
6,000
10,000
20,000
13,500
22,500
45,000
3,000
5,000
10,000
6,750
11,250
22,500
clay:
soft
moderate
stiff
f
undr
[kPa]
25
50
200
2,000
4,000
6,000
4,500
9,000
13,5000
800
2,000
4,000
1,800
4,500
9,000
500
800
2,000
1,125
1,800
4,500
peat:
soft
moderate
f
undr
[kPa]
10
30
1,000
2,000
2,250
4,500
500
800
1,125
1,800
250
500
560
1,125
1)
the values in this column can be used for active soil pressure P
ea;h;a;rep
2)
representative value of the low average if a low bedding constant is unfavourable
3)
representative value of the high average if a high bedding constant is unfavourable
p
0
is the neutral soil pressure in [kN/m
2
]
p
ea;h;p;rep
is the representative value of the maximum horizontal soil pressure for (passive) stress increase in
[kN/m
2
]
p
h
is the maximum horizontal pressure in [kN/m
2
] for the course of the bedding constant
Permeability coefficient k
Design values for the permeability coefficient for the different soil types can be determined
using table 3.16. Generally it is sufficient to take the middle of the course.
Table 3.16 Permeability coefficient k
soil k-value [m/s]
gravel > 1.0*10
-2
coarse sand with fine gravel 1.0*10
-2
1.0*10
-3
coarse sand 1.0*10
-3
1.0*10
-4
fine sand 1.0*10
-4
1.0*10
-5
very fine sand 1.0*10
-5
1.0*10
-6
clay and strong loamy sand 1.0*10
-6
1.0*10
-8
sandy clay 1.0*10
-8
1.0*10
-10
clay < 1.0*10
-10
For clean sand the permeability can very well be derived from the particle size distribution:
2 2
10
10 k d
=
(3.58)
Project number
415660-21 v04
Date
December 2007
Page
120 of 278
where:
d
10
is the particle diameter at which 10 % of the soil weight is finer [mm];
k is the permeability in [m/s].
'DIRECT' PARAMETER DETERMINATION
Consistency of soil layers and levels
Determining the consistency of the soil can be done be means of drilling or sampling according
to NEN 5119; 1991, followed by a classification by visual inspection. This classification can be
improved by the following lab-tests:
unit weight, only for cohesive soil;
particle size distribution;
degree of saturation of the sample;
plastic limit, only for cohesive soil;
liquid limit, only for cohesive soil;
percentage of organic matter.
Pore water pressure
The pore water pressure in different layers of soil can be determined by the following tests:
piezometer measurements, according to NEN 5120; 1991 [NNI; 1991];
cone penetration test with a piezo-conus;
measurements of water pressure; this method can also be used for determining excess pore
water pressures in clay or peat layers.
Unit weight of soil
The unit weight of the soil can be determined by using the unit weight. The following formula
applies: = .g.
For cohesive soil the unit weight can be determined by using the cylinder, which can be pushed
into the soil in the field as well as filled with soil in the laboratory. The weight of the soil in the
cylinder, of which the volume is known, can be measured. The unit weight of the soil is deter-
mined by dividing the measured weight of the sample by the known volume of the cylinder.
For the lab-determination a soil sample of class 1 or 2, according to NEN 5119; 1991 [NNI;
1991] must be available.
To determine the unit weight of humid non-cohesive soil above the ground water table, a
method called replacing by gravel can be used. According to this method, a pit of approxi-
mately 1 liter must be dug. The excavated soil must be weighed after it has dried; this for ap-
proximating the water content. The volume of the pit can be measured indirectly by filling it
with uniform gravel and measuring the weight of the gravel necessary to fill the pit. The dry
unit weight can now be determined using the dry weight of the excavated soil and the measured
volume of the gravel.
In the field the unit weight can also be determined by using nuclear equipment, which measures
the mass of the soil by using -rays.
Project number
415660-21 v04
Date
December 2007
Page
121 of 278
failure envelope
t
o'
c'
|'
2u
o'
3
o'
f
o'
1
t
max
t
f
t
f
u
o'
3 o'
3
o'
1
o'
1
o'
f
Angle of internal friction and cohesion for drained circumstances
For a reliable determination of effective shear friction parameters |' and c' by means of triaxial
tests according to paragraph 8.8.5.2 of NEN 6740; 1991 a soil sample of class 1, according to
NEN 5119 is necessary. For non-cohesive soil, however, obtaining a class 1 sample is usually
not possible. In this case a class 3 sample must be prepared, of which the unit weight resembles
the unit weight of the soil in the field. The triaxial test is executed according to CU-procedures
with measurement of the pore water pressure or according to the CD-procedure.
The shear strength t
f
can be defined as follows:
t
f
= c' + o
n
' tan |' (3.59)
where:
c' is the cohesion;
o
n
' is the effective normal stress;
tan |' is the friction coefficient (|' is the effective angle of internal friction).
For the triaxial test this can be expressed as:
t
f
= c' + o'
f
tan |' (3.60)
and
t
max
= (o
1
o
3
) (3.61)
This is clarified in figure 3.36. In order to indicate the stress path, the results are rendered in a
pq diagram, as shown in figure 3.37.
In order to determine the failure line of Mohr at least three failure circles are necessary. Using
the single stage procedure, three different samples are required. On the other hand, when using
the multi stage procedure, one sample is sufficient.
In figure 3.38 the axial strain is indicated on the stress paths. o' and a' can be determined in
relation with the axial strain. This can be rendered separately for |' and c' in a so-called mobi-
lization curve (see figure 3.39).
Figure 3.36 Graphical representation of the results of a triaxial test.
Project number
415660-21 v04
Date
December 2007
Page
122 of 278
adapted failure envelope
stress point
(o'
1
o'
3
)
o'
, o'
3
) (o'
1
)
(o'
1
+ o'
3
)
a'
45
o
45
o
(o'
1
+ o'
3
)
(o'
1
o'
3
)
1
2
3
c
a
= 5 %
2 %
1 %
|', c'
|'
c'
2 % 5 %
c
a
Figure 3.37 Stress path method of representation of the results of a triaxial test.
Figure 3.38 Stress paths in triaxial tests.
Figure 3.39 The effective angle of internal friction |' and the effective cohesion c' as a function of
the axial strain.
Project number
415660-21 v04
Date
December 2007
Page
123 of 278
Undrained shear strength
This parameter, which only relates to the strength of cohesive soil, is determined in the field by
means of a Vane-test according to NEN 5106; 1991 [NNI; 1991], and can (but not necessarily)
be done in a borehole. In order to petrform a Vane-test, a vane, consisting of four blades, must
be placed in undisturbed soil. This vane must then be rotated to determine the required torque
for shearing the enclosed cylindrical soil element.
This required torque is then converted to a shear stress along the perimeter of the cylindrical
soil element. This shear stress is equal to the shear strength f
undr
of the soil. A similar test, using
a smaller vane, can be performed in the laboratory on an undistrubed soil sample.
The undrained shear strength can also be determined with triaxial test (type UU or CU) on a
class 1 soil sample. First the compressive strength o
1
of the soil must be determined by a com-
pression test on a cylinder of soil, with a cell pressure o
3
in the triaxial cell.
Than follows:
f
undr
= (o
1
o
3
) (3.62)
where:
o
1
is the total vertical stress on the sample on the moment of collapse;
o
3
is the cell pressure in the triaxial cell.
Young's modulus
From a triaxial test according to the CU or CD procedure, the value of the Young's modulus
can be derived as follows:
E
o
c
A
=
A
(3.63)
In the triaxial test Ao is expressed in the deviatoric stress o
1
o
3
and Ac in the axial strain c
1
,
as shown in figure 3.40. The stress-strain diagram of soil is not linear. Because of this the
Young's modulus will decrease when the deformation increases. Normally in calculations the
so-called 50 % secant modulus of the maximum deviatoric stress is used.
The Young's modulus also depends on the stress level of the soil element. In the triaxial test
this stress level is represented by the cell pressure o
3
. A higher stress level implies a higher
value of the Young's modulus. Deep layers also show a stiffer behaviour than shallow layers.
The cell pressure in the triaxial test should represent to the actual stress level in the field.
Poisson's ratio's.
By means of a drained triaxial test or an undrained triaxial test with measurement of the water
pressure, Poisson's ratio v can be determined by:
3
1
c
v
c
=
(3.64)
Project number
415660-21 v04
Date
December 2007
Page
124 of 278
(o
1
o
3
)
75 %
c
a
50 %
25 %
100% (o
1
o
3
)
max
c
1.0
c
.0.5 0
0
E
0
1
1
, )
1 3
50
0.5
0.5
E
o o
c

=
(o
1
o
3
)
c
1
(o
1
o
3
)
c
1
c
c
c
va
c
1
E
1
1
1 2v
Figure 3.40 Determination of the Young's modulus E using the results of a triaxial test.
The following equations apply to the drained triaxial test:
) 2 (
E
1
) 2 1 (
E
) 2 (
V
3 1
3 1
3 2 1
o
o v o c
v
o o
c
c c c c
A A =
+
=
+ + =
A
=
v
v
v
V
(3.65)
Figure 3.41 Determination of Poisson's ratio v.
Project number
415660-21 v04
Date
December 2007
Page
125 of 278
By measuring c
v
and c
1
, v can be determined according to figure 3.41. In a triaxial test Ao
3
= 0
applies, this leads to:
1
(1 2 )
v
c
v
c
=
(3.66)
Angle of dilation
The angle of dilation can also be determined by a drained triaxial test by means of the follow-
ing relation:
1
sin
2
v
v
c
c c
+ =
(3.67)
or
1
2sin
1 sin
v
c
c
+
=
+
(3.68)
How to determine this angle is shown in figure 3.42.
Figure 3.42 Determination of the angle of dilatation.
Wall friction angle
The wall friction angle o can be determined by means of a direct shear test, where a representa-
tive soil sample is being sheared along a wall sample. Such tests are often difficult to execute
and not very reliable. Therefore they are not used regularly. In general table 3.1 is used, or data
from literature [Potyondy; 1961].
(o
1
o
3
)
c
1
(o
1
o
3
)
c
1
c
v
c
1
c
v
c
1
1
1
2sin
1 sin

E
Project number
415660-21 v04
Date
December 2007
Page
126 of 278
Horizontal modulus of subgrade reaction
The horizontal modulus of subgrade reaction can be derived from the Young's modulus E of
the soil. To do this finite element calculations are necessary. This is in itself redundant, be-
cause then a sheet pile calculation with the finite element method is necessary to determine the
horizontal modulus of subgrade reaction. In that case an analysis of the sheet pile has been
performed in the same time.
The pressiometer test is also a direct method for determining k
h
as is described in the para-
graph about indirect methods. There are different types of pressiometers and pressiometer
tests, of which the results cannot always be compared. [Amar et al.; 1991].
Permeability coefficient
The permeability of soil can be determined by means of different direct methods. Deciding
which method to use depends on the soil type - permeable or impermeable and the nature of
the problem. In the laboratory this is done by:
the constant head test, for permeable soil;
the falling head test, for impermeable soil;
consolidation coefficient determined by a triaxial test;
consolidation coefficient determined by a compression test.
In situ this is done by:
pump tests;
determining the velocity of the dropping of the water level in a borehole or piezometer.
The scheme in figure 3.43 shows which tests are suitable for which magnitude of permeability,
mostly based on experiences with different types of soil. In general, pumping tests are only
used for designing dewatering systems.
Figure 3.43 Suitability of different methods to determine the permeability.
permeability coefficient k in [m/s]
From grain size distribution
Reliable if clean sand
From 'constant head' lab. test
Quite reliable; experience not required
From 'falling head' lab test
Reliable
Experience not required
From 'falling head' lab. test
Only indicatory
Experience required
From 'falling head' lab. test
Quite reliable, experience req.
From 'in situ' tests
Reliable; experience required
From 'oedometer' test
Reliable, experience req.
10
-5
10
-10
Project number
415660-21 v04
Date
December 2007
Page
127 of 278
3.4.3 Determination of the representative value of soil parameters, based on test
results
The representative value of the mean of the geotechnical soil parameters X
mean;rep
, must, with
exception of the geometrical parameters and the consistency of the soil layers, be determined
while taking into consideration the spread in the test results. This can be done according to the
method, described in chapter 8 of NEN 6740; 1991 [NNI; 1991].
The representative value of the mean soil parameter can be found by:
; ;
i
mean rep n v mean
X
X R X
n
= =
(3.69)
where:
X
mean;rep
is the representative value of the soil parameter of the concerning soil
layer;
X
mean
is the mean value of the soil parameter determined from the test re-
sults;
X
i
is the value of the soil parameter of test i (the test result of test i);
n is the number of test results;
v is the variation coefficient, e.g. from table 3.8 or table 1 of NEN 6740;
1991;
R
n;v
is a multiplication factor according to table 2 of NEN 6740; 1991 or
table 3.17.
Table 3.17 Values of R
n;v
for different numbers of test results n and different variation coefficients
v.
R
n;v
n v = 0.05 v = 0.10 v = 0.15 v = 0.20 v = 0.25
1
2
in this area the representative values, according to table 3.8,
apply for the concerning soil parameters.
3 0.92 0.83 0.75 0.66 0.58
4 0.94 0.88 0.82 0.76 0.71
5 0.95 0.90 0.86 0.81 0.76
6 0.96 0.92 0.88 0.84 0.84
8 0.97 0.93 0.90 0.87 0.87
10 0.97 0.94 0.91 0.88 0.88
Project number
415660-21 v04
Date
December 2007
Page
128 of 278
3.4.4 Partial material factors for soil
The partial material factors for general applications in geotechnics are given in table 3 of NEN
6740; 1991. In 2.4 partial material factors are given, which can be applied when designing a
sheet pile wall, specifically in the case when the sheet pile wall is considered beam, supported
by soil springs. Application of these factors leads to a design, which is tuned to the probability
of failure, acceptable for sheet pile wall constructions in different safety classes.
3.4.5 Type and size of soil investigation
The type and extent of the necessary soil investigation depends on de geotechnical category
under which the project classifies and the limit state to be analysed. Also the calculation
method and the necessary parameters for this method determine the type of investigation.
According to NEN 6740; 1991, there are three geotechnical categories: GC1, GC2 and GC3.
Geotechnical Category 1 (GC1)
This category mainly consists of relatively simple sheet pile constructions, e.g. wooden or con-
crete sheets of small size or very small-scale or temporary projects. The following conditions
must be met:
the soil (and groundwater) conditions are known, if relevant;
no life-threatening situations in case of collapse;
economical loss in case of collapse is clearly defined and relatively small;
adjacent ground surface and water levels must be horizontal;
at grade no surcharge load is present;
maximum retaining height of the sheet pile wall _ 2 m, with exception of temporary
projects such as sewerage trenches;
experience with comparable sheet pile walls in comparable terrain and soil conditions is
present and pile driving is no problem in the area;
the surrounding area is not classified as vulnerable, and is not influenced negatively;
no anchors or struts, with the exception of singular screw anchors, prefabricated anchor
blocks or a singular strut in temporary projects such as sewerage trenches;
groundwater flow has no significant role, the normal water level differences over the sheet
pile wall are < 0.5 m.
All sheet pile wall constructions that do no classify as GC1 or GC3 classify as GC2
This category covers very large-scale and complicated sheet pile wall constructions, and sheet
pile walls that are part of complex projects. Sheet pile wall constructions in relatively difficult
and possibly dangerous terrain conditions, soil consistency or groundwater situations also clas-
sify as GC3.
Project number
415660-21 v04
Date
December 2007
Page
129 of 278
If one of the following circumstances applies, then the sheet pile wall construction must be
classified as GC3:
imminent life-threatening situations in case of collapse;
extremely large economical loss in case of collapse;
no experience with comparable sheet pile walls in comparable terrain, soil and groundwater
conditions;
the maximum retaining height of the sheet pile wall is > 6 m;
the sheet pile wall is multiple strutted or anchored;
the sheet pile wall is subjected to exceptional load;
the direct vicinity is considered vulnerable; e.g. buildings of historical value or pipes
containing hazardous materials;
the soil is strongly polluted and possibly corrosive;
the sheet pile wall is constructed in open (commercial) water;
the sheet pile wall is constructed in open water with relatively big water level differences;
the sheet pile wall is part of:
tunnels, big bridges and sluices;
nuclear installations;
installations for treatment or storage of hazardous materials;
offshore constructions;
storm surge protection constructions;
constructions subjected to high dynamical load.
Classification in one of the geotechnical categories must be done according to the following
aspects:
character of the sheet pile wall itself (type and vicinity);
the terrain, soil and groundwater conditions;
the vulnerability of the surrounding area;
influences from abroad, e.g. exceptional loads and presence of hazardous materials.
In general the classification is done before a possible soil investigation is executed, but must be
reassessed after this investigation and during construction. In order to obtain an economical
design, or if a qualified and experienced designer deems it necessary, the sheet pile wall con-
struction may be placed under a higher geotechnical category.
Soil investigation for GC1
If experience shows that the primary demands are met, then a sheet pile wall in GC1 can be
constructed without further soil investigation. If experience and knowledge of the soil in the
direct vicinity are compatible, then a visual inspection of the construction site before and dur-
ing construction is sufficient. This inspection must at least result in knowledge of the consis-
tency of the upper layers and the groundwater level. Complementary soil investigation by
means of c.p.t.'s, approximately 25 meters centre to centre, and till a depth of at least 2 m be-
low the assumed toe level are usually advisable, like simple borings.
It is advisable with regard to the ultimate limit state 1A, to check with traditional calculation
models, the collapse of sheet pile wall elements, the pulling out of anchors and insufficient
passive soil resistance. Necessary representative parameters can be found in table 3.8.
Project number
415660-21 v04
Date
December 2007
Page
130 of 278
The inspections and soil investigation for GC1, the before mentioned testing of the design
situation in comparison with the ultimate limit state worst-case scenario, must be seen as
minimal for a sheet pile wall construction in GC2.
NEN 6740; 1991 however, demands that the construction site must be examined by means of
field tests and lab tests. The following GC3 instructions can be seen as a maximum for sheet
pile wall constructions in GC2
The following instructions must be seen as minimal for constructions in GC3. Soil investiga-
tion for GC3 must always take place in different phases, with intermediate evaluations. These
evaluations are necessary to determine the final extent and type of the soil investigation. The
investigation contains minimally:
cone penetration tests with measurement of the local friction resistance, approximately 25
meters centre to centre, and to a depth of at least 5 meters below the intended toe level of
the sheets, or if need be, even deeper till at least 2 meters in a soil layer of which is proven
that it is not a part of any limit state or design situation;
if the maximum retaining height H _ 10 meters, then c.p.t.'s are necessary at distances of
0.75*H and 1.5*H, approximately 25 meters centre to centre;
borings at 50 meter intervals maximum.
For all these borings and c.p.t.'s applies that if the results differ significantly, then extra borings
and c.p.t.'s are necessary in between the original test locations.
During performance of a boring, the complete profile of the cohesive layers must be deter-
mined by means of a series of undisturbed soil samples. (Un)disturbed samples of the non-
cohesive layers must be taken continuously. The consistency of the soil layers and levels of the
layer boundaries will be determined using the results of the boring and c.p.ts.
Data about ground and surface water levels must be gathered through observations in the field
over a period of time, in which the highest as well as the lowest water level is expected. The
envelope of the water pressure in the soil with depth must also be determined by means of ob-
servations. If, due to the presence of impermeable layers, differences of the standard hydro-
static situation are expected, which can be relevant to the limit state to be analysed for one of
the design situations, then the water pressures ought to be determined specifically by measur-
ing.
From hereon the expression significant layer indicates a soil layer that is predominantly pre-
sent over the depth, or that has properties which have a significant influence on the limit state
to be analysed. To confirm the classification of the soil, it is necessary to perform additional
laboratory tests. For each boring and each significant cohesive layer or at least once per 2 m
boring, the Atterberg Limits and a laboratory vane test should be performed. This is also ap-
plicable for the determination of the organic matter, if a visual inspection indicates that there is
indeed organic matter present in the layer. Also a grain size distribution has to be determined at
least three times per significant non-cohesive layer, or once per 5 meter boring. Per boring the
unit weight and the degree of saturation of all the cohesive layers must be determined twice
every meter. In case of non-cohesive soil, the unit weight can be determined by using table 3.8.
Per boring, a series of triaxial tests must be carried out three times per significant cohesive or
non-cohesive layer, or once per 5 meter boring. A series of triaxial tests consists of at least
three determinations of the critical stress circle according to Mohr. From these tests the shear
Project number
415660-21 v04
Date
December 2007
Page
131 of 278
strength parameters c' and |' can be determined, and if needed the deformation parameter E as
well.
Note:
Taking undisturbed sand samples from layers is virtually impossible. The values of c' and |'
can in this case be determined using table 3.8.
For projects that require calculation of the undrained situation, in certain limit states and de-
sign situations, per boring at least three times per significant cohesive layer, or once every 5
meter boring, an in-situ vane test must be performed to determine the in-situ shear strength f
undr
.
Values for o, v and will be determined by experience.
If the ultimate limit state 1B is analysed by means of a calculation method, based on a beam on
soil springs, per 100 m' sheet pile wall, at least three times per significant cohesive layer a pres-
siometer test according to Mnard has to be carried out to determine the horizontal modulus of
subgrade reaction k
h
. If no demands are made concerning the displacements, then the values
from table 3.15 can be used.
If for a project, classified as GC3, limit states for design situations must be analysed, where
groundwater flow is significant, then, per 100 m' sheet pile wall, the permeability coefficient
must be determined by one of the indicated methods; preferably by means of a pump test.
3.4.6 Report of the soil investigation
For every project the results of the soil investigation must be recorded in a factual report. The
used methods and test procedures must also be indicated.
The report usually consists of the following parts:
A. General part;
B. Results of the in-situ investigation, followed by: (if needed);
C. Results of the laboratory tests;
D. Geotechnical length-profile.
Part A. General part.
Consists of, at least, the following:
table of contents;
name or identification of the project;
location of the project;
name of the company that carried out the geotechnical investigation;
date;
situation drawing;
a record of the used system for determination of the location and depth of the soil data;
an explanation of the used symbols and codes.
Project number
415660-21 v04
Date
December 2007
Page
132 of 278
Part B. Results of the in-situ soil investigation
This part contains all the results of the soil investigation in graphs, diagrams, statements and
tables. Every result must be provided with:
project name or identification;
date;
identification number, code or coordinates;
indication of depth, or level relative to N.A.P.
Some investigation methods and field tests are standardized. The methods of reporting are in-
dicated in these codes, and naturally have to be copied in the report. The following investiga-
tion methods and field tests are implied:
cone penetration tests according to NEN 3680; 1982;
boring and sample taking according to NEN 5119; 1991;
piezometers according to NEN 5120; 1991;
in-situ vane-test according to NEN 5106; 1991.
For the description and reproduction of soil consistencies NEN 5104; 1989 has to be used.
Part C. Results laboratory testing
This part contains all the results of the laboratory tests in graphs, diagrams, statements and
tables. Every result must be provided with:
project name or identification;
date;
identification number, code or coordinates;
indication of depth, or level relative to N.A.P.
Of the laboratory tests, the triaxial test is standardized in NEN 5117; 1991 [NNI; 1991]. This
code also indicates the method of reporting, which, naturally, must be copied in the report.
Part D. Geotechnical length-profile
Strictly considered, the geotechnical longitudinal soil profile is not a part of the reporting of
soil data, but a part of the interpretation of the soil data on behalf of the project. This interpre-
tation can also be left to the designer. In this case the soil profile becomes a part of the design
report of the sheet pile wall.
In general, however, a factual report will often surpass the sole report of test results, and can
perfectly well be completed with geotechnical profiles (longitudinal and cross sectional pro-
files).
Projects classified as GC2 and GC3 require at least a geotechnical profile of the projected loca-
tion of the sheet pile wall. The horizontal and vertical scale of the profile must be indicated.
Depths and levels have to be indicated relative to N.A.P. All c.p.t's, borings and piezometers in
geotechnical profiles and the results of field en laboratory tests must be provided with identifi-
cation numbers or codes. The results of c.p.t.'s and borings must be drawn at the exact location
in the completely explored depth.
Project number
415660-21 v04
Date
December 2007
Page
133 of 278
4 Calculation models for sheetpile wall desgin
4.1 Classical calculation models
4.1.1 Introduction
The over the years developed classic calculation models for sheet pile structures are either em-
pirical or analytical or a combination of both. In the Netherlands the method of Blum is used in
general. This method is based on an analytical concept, which applies the beam-theory. In Dutch
designing the following classical methods are not often used:
Danish codes (Brinch-Hansen method);
Tschebotarioff;
Ohde.
More about these methods can be found in literature [Smoltczyk; 1982].
Blum schematizes the statically undefined system, which is formed by a sheet pile and the sur-
rounding soil, in such a way that it becomes a statically defined calculation model. The earth
pressure is determined by Coulombs theorem, but other methods can be used as well, e.g. the
method using arched slip surfaces. In the first phase the earth pressure load on the sheet pile
wall is calculated. This is done using a schematization, which only takes the strength of the soil
into consideration. The deformation behavior of the soil and stiffness of the sheet piles have no
influence on the result of the calculation. Assumed is that the displacement of the wall results
in respectively active and passive failure of the soil on both sides of the sheet pile wall. Based
on this determined load distribution, the depth of penetration can be determined. The deflection
line of the sheet pile can be deduced from the moment distribution. However, the displace-
ments calculated with this method are not very realistic.
In case of a flexible sheet pile in combination with non-cohesive soil on the active side, arching
may be taken into consideration. This arching leads to higher anchor forces and lower field
moments.
Blums method is not reliable for taking into account the following influences and will result in
rough approximations:
deformation of soil as well as deflection of the sheet pile;
the pre-tensioning of anchors;
the different building stages;
over consolidation as a result of former surface loads (e.g. ice) or former heightening.
Blums method however, is a fast and simple way to make a rough estimation of the minimally
required depth of penetration and sheet pile properties A further optimization can be made
using the more sophisticated calculation methods. For this optimization the following models
can be used:
a spring model;
a finite element model.
Project number
415660-21 v04
Date
December 2007
Page
134 of 278
4.1.2 Determining the maximum earth pressure and soil friction
For determining the maximum earth pressure and soil resistance Coulombs or Culmanns
method is mostly used. Principally every method that results in reliable values of this maxi-
mum earth pressure and soil resistance can be used, e.g. the method using arched slip surfaces.
In 3.2 these methods are described. In this description the following aspects are considered:
the direction in which the wall displaces;
the necessary displacement for total development of the soil resistance in the passive part of
the soil;
angle of wall friction o;
the influence of cohesion c';
the maximum earth pressure and soil resistance for the undrained situation (parameter f
undr
);
inclining or declining terrain on either side of the sheet pile.
Coulombs method is based on straight slip surfaces in both the active and passive zone. This is
an unsafe approximation because in case of a wall friction angle o > 2/3 | and arched slip sur-
faces, the soil resistance in the passive zone is only 60 % to 70 % of the value according to
Coulomb. In case of arched slip surfaces the value of the earth pressure in the active zone is
also higher then the value according to Coulomb.
Culmanns method is based on Coulombs theory of straight slip surfaces. Due to its graphical
solution method this method however has the option to take into account the influence of a
sloping terrain, locally applied surcharge loads and other direct influences. However, the same
restrictions as for Coulombs method apply, so Culmanns method is also unsafe in case of a
wall friction angle o > 2/3 |.
4.1.3 Angle between the slip surface and the horizontal axis
Sometimes it can be of importance to know the location of the slip surface, for example in con-
nection with the surrounding structures. Culmanns method determines the slip surface graphi-
cally. For homogeneous non-cohesive soil and a vertical wall, the following formulas apply:
2
2
(1 tan )(tan tan )
tan tan
tan tan
(1 tan )( tan tan )
tan tan
tan tan
a
a
p
p
| | |
u |
| o
| | |
u |
| o
+
= +
+
+
= +
+
(4.1)
in which:
u
a
is the slip surface angle in case of active earth pressure;
u
p
is the slip surface angle in case of passive earth pressure;
| is the angle of internal friction;
o
a
is the angle of wall friction in case of active earth pressure;
o
p
is the angle of wall friction in case of passive earth pressure;
| is the angle between the surface and the horizontal axis.
Project number
415660-21 v04
Date
December 2007
Page
135 of 278
4.1.4 The influence of a not horizontal ground level in Coulombs method
This paragraph describes some practical methods to implement the influence of a not horizon-
tal ground level and a local surcharge load. The calculation method determines an extra load, to
be added on the regular earth pressures in case of an horizontal, unloaded ground level at the
top level of the sheet pile wall.
Influence of a non-linear surface
According to Jenne [Jenne; 1960] the not horizontal ground level situation sketched in figure
4.1 can be calculated as follows:
At a considerable depth, the vertical earth pressure increase Ag equals:
Ag = Ah (4.2)
Figure 4.1 Addition to the effective vertical stress in case of a not horizontal ground level.
At top level of the wall the stress is reduced as follows:
For | _ | applies:
;0
; ;0
;0
; ;0
a
a a
a
a a
K
x a ka
K K
K
y h k h
K K
|
|
= =
= A = A
(4.3)
in which:
K
a;0
is the horizontal earth pressure factor for a horizontal surface; according to 3.2
this equals K
;a
(o = 0; | = 0);
K
a;|
is the horizontal earth pressure factor for an inclining ground level with an an-
gle |; according to 3.2 this equals K
;a
(o = 0; | = 0);
x, y see figure 4.1 and 4.2.
x = k . a = 1.3 . 1.0 = 1.3 m
y = k . Ah = 1.3 . 3.0 = 3.9 m
Ag = . Ah = 16 . 3.0 = 48 kN/m
2
g = . h = 19 . 7.5 = 142.5 kN/m
2
6.0 2.0
Ah = 3.0
h = 7.5
a = 1.0
x =1.3
y = 3.9
k = 1.3
|
| = 26.6
o
g Ag
= 16 kN/m
3
| = 30
o
o = 2/3 |
= 19 kN/m
3
Project number
415660-21 v04
Date
December 2007
Page
136 of 278
For | > | the following applies:
tan
'
tan
y y
|
|
=
(4.4)
In figure 4.2 the relation between |, | and k is given in a diagram. Another simple method to
take into account a not horizontal ground level is to assume a virtual second ground level, and
for this virtual ground level the horizontal earth pressure graph can be calculated. Using the
angle of internal friction | and the slip surface angle u
a
, the transformation line between the
virtual and the actual earth pressure graph can be constructed according to figure 4.3. A third
option is to use Colmanns graphical method.
Figure 4.2 Relation between |, | and k.
15
o
20
o
25
o
30
o
35
o
|
k
5
4
3
2
1
0
Ah
a
x
y
Ah
a
x
y
|
|
|
. Ah . Ah
slope angle | = |
x = k . a
y = k . Ah
slope angle | > |
x = k . a
y =( tan | / tan | ) k .
Ah
suppose
o
a
= + 2/3 |
Project number
415660-21 v04
Date
December 2007
Page
137 of 278
Figure 4.3 Approximation method for the horizontal earth pressure in case of a non horizontal
ground level.
4.1.5 The influence of a piled structure
Analogous to the above described methods for non horizontal ground levels, | and u
a
can also
be used to take into account the influence of a piled structure. This according to figure 4.4.
Figure 4.4 Horizontal earth pressure due to a piled structure.
Ah
h
|
|
u
a
e
ah
(h)
e
ah
(h+Ah)
Ag = . Ah
Ah
h
|
u
a
g = . h
Project number
415660-21 v04
Date
December 2007
Page
138 of 278
4.1.6 The influence of surcharge loads
In 4.2, where the method for determining an elastic supported beam has been described, is also
described how to determine the influence of the surcharge load, using Culmanns method, in
case of an active stress situation. An uniform surcharge load can be replaced by a virtual sur-
face located at a height h' = p / as shown in figure 4.5.
Figure 4.5 Horizontal earth pressure in case of an uniform surcharge load.
The determined earth pressure graph consists of a rectangular earth pressure envelope with a
pressure value e
ahp
= K
sur;a
p and a triangular envelope with a bottom level earth pressure of
e
aho
= hK
;a
. For a vertical wall and a horizontal ground level K
sur;a
= K
;a
. This implies that
the influence of the load p can also be taken into account with the earth pressure factor K
sur;a
directly, as is shown in 3.2.
Ohde has already given general calculation methods for one sided or two sided local surcharge
loads [Ohde; 1938, 1952 and 1956]. Determining the exact angle and location of the slip sur-
face is very meticulous and difficult. The approximation method, sketched in figure 4.6, for
one sided and two sided local surcharge loads, however, leads to acceptable results. The slip
surface angle u
a
can be determined while neglecting the influence of the surcharge load, ac-
cording to 4.1.3.
Analogous to the determination of the earth pressure in the Rankine-case, a formula can be
derived for the total extra horizontal load on a wall, in case of a two sided local surcharge load.
This method is sketched in figure 4.7.
ahp a
E psk =
(4.5)
virtual ground level
p [kN/m
2
]
h
e
ah0
e
ah
e
ahp
E
a
E
ah
u
a
h' =
p

e
ahp
= h' . k
ah
= p . K
ah
Project number
415660-21 v04
Date
December 2007
Page
139 of 278
Figure 4.6 Horizontal earth pressure in case of a one sided local surcharge load.
The equation for u
a
is:
sin( )
cos
cos( )
a
a a
a a
u |
k o
u | o
=

(4.6)
where:
E
ahp
is the horizontal extra load;
p is the surcharge load per [m'];
s is the width where the load acts;
k
a
is an auxiliary value.
In table 4.1 some values for u
a
are given for a vertical wall and a horizontal ground level.
p [kN/m
2
]
p [kN/m
2
]
u
a
u
a
|
b
b/2
h
b
a
a.
b.
e
ah0
e
ah0
e
ahp
e
ahp
= . h . K
a
= p . K
a
= . h . K
a
= p . K
a
Project number
415660-21 v04
Date
December 2007
Page
140 of 278
h
c
b
a
u
a
u
a
d s
p [kN/m']
e
ah0
e
ahp
= . h . K
a 2 E
ahp
C a
=
E
ah0 E
ahp
= p .
k
L + 2d
d s d s
= p . [kN/m']
L
L + 2d
theoretically loaded area
actually loaded area
p [kN/m']
L
L + 2d
45
o
45
o
Table 4.1 Values for factor u
a
.
k
a
| ()
o
a
= 0 o
a
= 2/34|
15 0.77 0.56
17.5 0.73 0.53
20 0.70 0.51
22.5 0.67 0.48
25 0.64 0.46
27.5 0.61 0.44
30 0.58 0.41
32.5 0.55 0.39
35 0.52 0.37
Figure 4.7 Horizontal earth pressure in case of a two sided local surcharge load.
Figure 4.8 Load distribution in case of a foundation behind the sheet pile wall.
h
c
b
a
u
a
u
a
e
ah0
e
ahp
= . h . K
a
2 . p . s . k
C a
=
S
p [kN/m
2
]
Project number
415660-21 v04
Date
December 2007
Page
141 of 278
Figure 4.9 Soil pressure due to a horizontal load on a foundation.
4.1.7 The influence of an embankment
The influence of an embankment can be determined accurately using Culmanns method. This
results in an envelope of the earth pressure, which is not linear. The approximation method
sketched in figure 4.10 shows the value of extra resistance given by an embankment. The trans-
formation of the earth pressure envelope in case of a declined surface with an angle equal to
the slope of the embankment to an earth pressure envelope of a horizontal surface is assumed.
c
h
b
a
a
b
|
u
a
H
ahH
=
2 . E
ahv
C a
e
ahv
H
H
E
ahv
H
ahH
= H
e
ahH
s
|
u
a
u
a
P
1
P
2
= . s . k
P
1
+ P
2
2
=
2 . H
C a
h
2 . H
b a
e
ahH
=
Project number
415660-21 v04
Date
December 2007
Page
142 of 278
Figure 4.10 Passive soil resistance in case of an embankment.
The distance a beneath the horizontal level where both pressure lines cross can be calculated
with:
;
;0 ;
'
p
p p
K
a h
K K
|
|
=
(4.7)
where:
K
p;|
is the horizontal earth pressure factor in case of an inclined ground level with
an angle |; according to 3.2 this equals K
;p
(o = 0, | = 0);
K
p;0
is the horizontal earth pressure factor for a horizontal ground level; according
to 3.2 this equals K
;p
(o = 0; | = 0).
For the determination of K
p;|
the wall friction is not taken into account.
4.1.8 Reduction of the moments and the increase of the anchor force
If a calculation is made using the classical earth pressure theory, corrections can be made if the
following demands are met:
the earth pressure calculation is made using the classical earth pressure theory according to
Coulomb;
the anchor of the sheet pile wall must be rigid compared to the rigidity of the sheet pile
wall;
the soil behind the sheet pile wall must be present and undisturbed to the height of the
anchor, since the installation of the sheet pile wall; there is no reduction in case of a soil
fill; the reduction is only valid if after the installation of the sheet pile wall, soil is removed
in front of the wall;
the sheet pile wall must behave relatively flexible compared to the soil rigidity;
in soil type is non-cohesive.
h
|
a
. K
p0
. K
p;|
e
ph
= h . K
p;|
(o = 0)
Project number
415660-21 v04
Date
December 2007
Page
143 of 278
The actual earth pressure will diverge significantly from the pressure calculated using Cou-
lomb. The fixation of the rigid anchor will cause the flexible wall to make relatively big deflec-
tions. Due to the fact that the soil is already present behind the sheet pile wall before excava-
tion, in the soil mass shear stresses will originate, if the soil follows the deformation of the
sheet pile wall. The deflections will be biggest where the field moment is maximal. This will
cause the shear stresses to be maximal at these locations. This will cause arching in the soil.
Due to the developed shear stresses the soil will start transferring forces to the area surround-
ing the anchor and to the zone where the sheet pile wall is imbedded. Movement in these zones
is relatively small and the deformation behavior is relatively rigid.
This leads to the pressure envelope sketched in 4.11c. An increased pressure is present around
the anchor point and in the area where the free or fixed support acts, while simultaneously the
earth pressure decreases in the area of the maximal field moment.
The figure shows clearly that the present field moment is smaller than the moment to be calcu-
lated using the earth pressure envelope according to Coulomb. The anchor force however, is
greater. That is why the following corrections are allowed:
the field moment due to the horizontal load of the effective soil stresses is reduced by 30 %;
the anchor force is increased by 30 %.
The reduction of he moment does not apply for the water pressure component of the field mo-
ment.
Figure 4.11 Influence of the wall movement on the soil pressure and the slip surface pattern.
In figure 4.11 is:
type of deformation type of failure
a. top deformation zone failure
*
b. foot deformation wedge failure
**
c. deflection combined failure
d. wall displacement zone failure
*
the occurrence of a pattern of slip surfaces in the failed zone;
**
the shearing wedge of soil translates as a block.
a. b. c. d.
anchor
arching
pressure
shear band
pattern
Project number
415660-21 v04
Date
December 2007
Page
144 of 278
4.1.9 Undrained soil behavior
Due to the changes in earth pressure in cohesive layers, a so called undrained situation can
occur.
Due to the undrained behavior, changes in soil pressures will not cause an immediate increase
or decrease of the effective stresses. This means that the cohesive soil layers initially maintain
their initial shear strength f
undr
.
4.1.10 Water pressures and groundwater pressures
Paragraph 3.2.4 discusses methods to determine the water pressures on the sheet pile wall in
case of a hydrostatic water pressure envelope, which is the case when the sheet pile wall is
situated (fully or partially) in a poorly permeable soil layer. It also discusses methods to deter-
mine the water pressures in case of flow under the sheet pile wall.
4.1.11 Other loads
Indications to determine the effects of the following loads:
anchor forces and strut forces 3.2.5;
ice load 3.2.6;
ship loading 3.2.7 and 3.2.8;
earthquake load 3.2.9;
loads due to swelling soil 3.2.11;
loads due to fluctuations in temperature 3.2.12.
4.1.12 Influence of compaction of the soil behind a sheet pile wall construction on
the earth pressure
If the soil behind the sheet pile wall is compacted in layers, then the earth pressure can increase
locally to values above the maximal active earth pressure. Indications to determine the increase
are discussed in 3.2.10.
4.1.13 Design method according to Blum
Blum developed an analytical graphical method to design anchored or not anchored sheet pile
walls. He distinguished sheet piles with free earth support and sheet piles with fixed earth sup-
port. These methods can be applied in most soil types. Because of its simplicity and the possi-
bilities for a wide application, this method has been most popular for a long time. Even in
Project number
415660-21 v04
Date
December 2007
Page
145 of 278
actual
ground level
cross-section
p = 20 kN/m
2
layer 1
layer 2
layer 3
resulting load figure
(values in [kN/m
2
])
| = 25
o
= 18 kN/m
3
| = 25
o
= 18 kN/m
3
c = 11 kN/m
2
| = 25
o
= 18 kN/m
3
g.w.l.
7
13
20
26
33
39
47
50
54
57
41
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.55
5 . 0.8 = 4.00
x
2
= 1.85
x
1
= 3.55
x
44
-152
-231
-310
-389
-467
-546
a = 0.45
0
-2
-6
-10
-16
-9
-5
modern design cases this method is often applied to achieve a first indication of the dimensions
of a sheet pile wall. After that the dimensions are optimized using a more advanced calculation
model, that can take into account the different construction stages.
Blum assumes complete plastic earth pressure and soil resistance. Initially the sheet pile wall is
assumed to be completely rigid.
Blums graphical method
This method was published by Blum in 1931 [Blum; 1931]. The maximum earth pressures on
the active and passive side are determined. From these load distributions an equivalent, result-
ing earth pressure envelope is derived. This figure is split up into horizontal parts, and the
resulting force per part is then calculated. See figure 4.12.
Figure 4.12 Example of Blums graphical method;
u = 44 / ( (K
:p:d
K
:a:d
)) = 44 / (11(9.15 0.22)) = 0.45 m.
The forces are then combined in a pole figure with an height H
p
which also functions as the
scale for the calculation (see figure 4.13 and 4.14). In this polar figure a link polygon is
sketched through the lines of action of the forces. By means of the drawing of a closing line for
the different support/fixation possibilities through the constructed link polygon,, different solu-
Project number
415660-21 v04
Date
December 2007
Page
146 of 278
tions can be judged in the same link polygon, on value of the field moment, the length of the
sheet pile wall and the overall deformation.
Figure 4.13 Example of Blum's graphical method;
M
max
= 2.35 . 150 = 353 kNm/m;
B
w
= (4.45 . 10 (3.50 + 4.45/2)) / 7.95 = 32 kN/m;
M
w
= B
w
2
/ 2 W
u
= 32
2
/ 20 = 51 kNm/m;
M
red
= 2/3 (M
max
M
w
) + 51 = 252 kNm/m.
The support/fixation conditions are:
unanchored, completely fixed;
singularly anchored, free supported;
singularly anchored, completely fixed;
singularly anchored, partially fixed.
A double anchored solution can also be calculated by trail and error. From the constructed
moment distribution figure, a new link polygon can be constructed using a new pole figure.
This new link polygon will show the deformation of the sheet pile wall, after the closing lines
have been added. Figure 4.13 shows two types of foot supports:
a free support solution;
a completely elastic fixed solution.
In case of a free support, as shown in figure 4.15, the toe of the sheet pile wall can rotate
freely.
excess water moment line deflection
pressure loads loads line
[kN/m
2
] [kN/m] [m] [m
2
/m] [m]
w
u
= 10kN/m
2
closure line, fixed sup-
porting at the toe
B
w
A
w
closure line, free sup-
porting
Project number
415660-21 v04
Date
December 2007
Page
147 of 278
Figure 4.14 Example of Blum's graphical method.
In that case the construction is statically determined and comparable to a beam on two sup-
ports. The closing line encloses one area: the moment area.
According to Blum, the theoretical depth of penetration found with this method is also a mini-
mal demand. Achieving equilibrium is not possible anymore in case of a lesser depth of pene-
tration, and the demands relating to ultimate limit state 1A are not met anymore.
As the sheet pile wall is installed deeper, the rotation rigidity at the foot increases until a com-
pletely elastic fixation can be assumed, as shown in figure 4.13. Increasing the depth does not
result in an increased fixation, and a further decrease of the field moment is not possible. Find-
ing this theoretical depth of penetration requires some trail and error calculation. By means of
using the elastic line of the sheet pile wall, the angular rotation of the toe should be checked to
remain zero.
This demand is met when the static moment of the moment areas related to the anchor point
equals zero.
To make a fast estimation, the following method can be used. See figure 4.16.
P-pole-figure
50 [kN/m
2
]
12 13 14
1 2 3 4 5 6 7 8 9 11
H
p
= 150
A
1
= 175
A
2
= 206
H = 50
P-pole-figure 2 [m
2
]
9 8 7 6 5 4 3 2
13 12 11 10
Project number
415660-21 v04
Date
December 2007
Page
148 of 278
Figure 4.15 Rotation stiffness at different fixations of the toe of the wall;
a. free supported;
b. (partially) fixed.
Figure 4.16 Determination of the theoretical penetration depth.
The link polygon can be assumed to consist of two parabolas. The areas of the moment enve-
lope are respectively:
I = 2/3 ab
II = 2/3 cd (4.8)
a. b.
A
A
o
o
B B
A
anchor
e
2
b
a
e
1
d
c
II
Project number
415660-21 v04
Date
December 2007
Page
149 of 278
The distances e
1
and e
2
are the distances to the centre of gravity of the moment areas related to
the anchor point. Both the static moments can be described as:
S
1
= 2/3 abe
1
S
2
= 2/3 cde
2
(4.9)
The condition (S
1
+ S
2
) = 0, then leads to the condition:
2/3 (abe
1
cde
2
) = 0 (4.10)
The closing line can now be rotated until this condition is met. It is unnecessary to construct a
new elastic line.
In figures 4.13 and 4.14 this method is elaborated. These figures also show that the foot rota-
tion of the sheet pile wall equals zero indeed. The anchor is assumed to remain in place.
At the foot of the wall an extra reaction force develops next to the fixation moment, the so
called C-force. This force must be a result of the earth pressure on a part of the wall underneath
the theoretical fixation point.
Blum states, on empirical reasons, that the actual depth of penetration must meet the following
demand:
t = 1.2 x + u (4.11)
in which:
u is the distance from the excavation level to the point of zero load on the wall in
the resulting earth pressure diagram;
x + u is the theoretical depth of penetration.
It is recommended to determine the depth of penetration t with the use the so called C-force,
according to the method described in figure 14.4.
If the actual depth of penetration is situated between the calculated depth for complete fixation
and free support, then one can speak of a partially fixed solution. In that case the foot of the
sheet pile wall can rotate and the fixation moment is significantly smaller then the field mo-
ment. Equal to the case of a fixed-end sheet pile wall, the same addition to the theoretical depth
of penetration applies.
Free-end sheet pile walls have the biggest anchor forces, the biggest field moments, but the
smallest sheet pile wall lengths. Due to the statically determined nature there is no possibility
for another envelope of the loads.
Free-end sheet pile walls are often used for temporary sheet pile walls. The load situation has
to be rather well known.
The calculation model with fixed earth support is statically undetermined. These sheet pile
walls behave more rigid when compared to other sheet pile walls. If overloading acts, the extra
load will cause some redistribution of forces between the anchor, the wall and the fixation in
the soil. Some sort of extra safety is apparently available. To limit the field moment as much as
possible, it is recommended to apply the anchors on a not too high level.
Generally is it more economical to limit the depth of penetration. The influence of the level of
penetration on the field moment is smaller then the influence on the fixation moment. In real-
ity, a partially fixed solution often leads to the most economical design. Displacement of the
anchors leads to increased fixation moments and decreased field moments.
Project number
415660-21 v04
Date
December 2007
Page
150 of 278
Blums analytical method manually executed
In 1950 Blum developed an analytical method which applies the Hedde method for determina-
tion of the theoretical penetration depth.
Figure 4.17 shows the load distribution. Because the earth pressure above the point of zero
resulting load can have any form or shape, it is recommendable to divide the figure in a few
horizontal parts, as is done in the graphical method. Of these parts the areas are represented by
forces P
n
with distances to the load reference point a
n
. If the sheet pile wall is fixed in a point x
meters below the point of zero resulting load, and loaded by the anchor force A, the forces P'
n
and P
n
, and the passive earth pressure, then the sheet pile wall calculation can be executed us-
ing the nomograms displayed in figures 4.18 and 4.19.
A limitation with regard to the graphical method is that below the point of zero resulting load
the resulting earth pressure must be assumed linear.
The case of a singularly anchored sheet pile wall can be solved if the earth pressures below the
point of zero resulting load are schematized as different layers. By checking if the sum of the
moments at the point of zero resulting load equals zero, the theoretical depth of penetration can
be determined.
To design a not anchored fixed-end wall, an estimation can be used as well. In this case the point
of zero shear forces is assumed at the location of the fixation moment.
Figure 4.17 Analytical method according to Blum.
a
n
P
n
-a
n
P
n
P
n+1
Aa
C
u
x
l
l
0
A
load reference point
Project number
415660-21 v04
Date
December 2007
Page
151 of 278
The level of penetration is determined according to the Blum-Jenne method for not anchored
walls, because the sum of the moments at this point equals zero. The following situations can
be calculated with this method:
singularly anchored completely fixed;
not anchored with soil load, local load or water pressure;
singularly anchored, free supported;
singularly anchored with top fixation;
singularly anchored completely fixed with top fixation.
Figure 4.18 shows the method for the singularly anchored and completely fixed-end wall. In fig-
ure 4.19 the same is shown, but for a not anchored wall.
Figure 4.18 Nomogram for the analytical approximation method according to Blum (anchored wall).
The nomogram in figure 4.18 is used as follows:
determine the force P;
calculate:
0
3
3 5
0
6 6
l l
l R R
m Pa en n Pa
K l K l
+
= =

; (4.12)
take and calculate: x = l; (4.13)

M
max
A
a
x t
l
0
l
Aa
u
P
n m
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
0.7
0.6
0.5
0.4
0.3
Project number
415660-21 v04
Date
December 2007
Page
152 of 278
calculate t = u + 1.2 x; (4.14)
calculate:
1
1
0
max
Q
Q
M Q a
= A
; (4.15)
The field moment is maximal where the shear force Q = 0.

Figure 4.19 Nomograms for the analytical approximation method according to Blum (unanchored
wall).
The nomogram in figure 4.19 is used as follows:
determine the force P;
calculate: 2 3
0 0
6 6
l l
R R
m P en n Pa
K l K l
= =

; (4.16)
take and calculate: x = l; (4.17)
calculate t = u + 1.3 x; (4.18)

M
max
M
a
x t
l
u
P
n m
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.5
1.4
1.3
1.2
0.3
1.1
1.0
0.9
0.8
0.7
0.6
0.5
O
Ax
y
K
R
Project number
415660-21 v04
Date
December 2007
Page
153 of 278
calculate:
1
1
0
max
Q
Q
M Q a
= A
; (4.19)
the field moment is maximal where the shear force Q = 0;

the location of the maximal moment:
2
n
R
Q
y
K
=
; (4.20)
where Q
n
is the shear force at depth l.
Ersatzbalkenverfahren, according to Blum
The so called Ersatzbalkenverfahren is an approximation method published by Blum in 1930
[Blum; 1930]. It is still widely used to make orientating calculations, with quick results.
In this method the sheet pile wall is cut, at the depth where the moment equals zero, which is
assumed at depth where the resulting soil pressure equals zero. At that location the shear force
B
0
is applied as a reaction force. This is shown in figure 4.20.
Figure 4.20 Ersatzbalkenverfahren according to Blum-Jenne.
Due to the presence of the anchor and the reaction force, a statically determined system is de-
veloped, which is comparable to a beam on two supports, of which the anchor force and the
moment are easy to determine.
The depth of penetration can then be calculated with the following formula:
t = u + 1.2 x (4.21)
in which:
0
6
R
B
x
K
=
(4.22)
t
h
C
A
l
u
x
l
0
u
0
B
0
M
max
A
Project number
415660-21 v04
Date
December 2007
Page
154 of 278
K
R
= K
;p
K
;a
(4.23)
where:
is the unit weight of the soil in the passive zone;
B
0
is the shear force force at the level of zero moment;
K
;p
, K
;a
are the soil pressure factors according to 3.2.
The objection to this method is that the value of u
0
, the exact level of zero moment, can not be
determined. Somtetimes the following is assumed:
u = 0.1 h (4.24)
If level of zero moment does not coincide with the assumed level, it may have a significant
influence on the design. Blum acknowledged this complaint and published an analytical calcu-
lation method in 1950, in which this complaint has been annihilated. The imperfections of the
Ersatzbalkenverfahren method were stimulation for Jenne, to develop a quick calculation
method, that did not have these same imperfections. [Bolderman et al.; 1968].
4.1.14 Computer code for the analytical calculation according to Blum
Here follows a description of a simple computer program which can execute calculations using
Blums method. The definitions for the calculations are displayed in figure 4.21.
In this computer program [Centrum Hout; 1987] the resulting earth pressure can have a multi-
ple linear course. By means of an interactive calculation the program determines the depth of
penetration, depending on the entered boundary conditions, fixed or free supported, and the
number and depth of the anchors.
Starting points are:
equilibrium of moments;
horizontal equilibrium;
a completely fixed-end sheet pile wall with a fictive fixation point at the foot of the wall, at
the location of the fixation force F;
a fixed anchor;
homogeneous soil below excavation level.
Project number
415660-21 v04
Date
December 2007
Page
155 of 278
Figure 4.21 Definitions for the computer program.
The following values can be entered:
the angle of the sheet pile wall with the vertical o [];
the value of the surcharge load q [kN/m
2
];
if an anchor is used and, if so, the height of the top of the sheet pile wall above the anchor
m [m] (if no, then m = 0);
for every soil layer:
the location of the top of the layer, for both the front and the back of the wall z [m];
the unit weight of water
w
[kN/m
3
];
the unit weight of wet soil
n
[kN/m
3
];
the angle of internal friction | [];
the wall friction angle o [];
Dry soil can be simulated by entering for
n
the dry unit weight of the soil, and
w
= 0. If the
only load on the wall is water pressure, then
n
= 0 and
w
= 10 kN/m
3
. If there is no load on the
wall then
n
and
w
both equal zero.
First of all the coefficients for the horizontal effective stresses are determined, assuming non-
cohesive soil (c = 0). Then the resulting uniform horizontal load f is calculated.
The fixation in the bottom layer is characterized by the parameter K
R;
:
K
R;
= (
sat

w
) (K
p
K
a
) (4.25)
Here the characteristics of the bottom layers are entered.
In Blums method the horizontal load f above the level of zero resulting load at z = l character-
ized by a force coefficient C
0
, a moment coefficient C
1
and a displacement coefficient C
3
.
layer
z(1)
z(2)
z(3)
z(4)
z(5)
z(6)
1
2
3
4
5
6
q
F
a
m
+ z
z = 0
+ o
left right
homogeneous
soil
Project number
415660-21 v04
Date
December 2007
Page
156 of 278
0 2
;
1 3
;
3
3 5
;
6
6
6
l
R m
l
R m
l
R m
C fdz
K l
C fzdz
K l
C fz dz
K l
=
=
=
}
}
}
(4.26)
These factors are determined numerically by dividing the load figure in intervals with a con-
stant load. After this the dimensionless coefficient , can be determined by solving the follow-
ing equation:
without anchor: ,
3
= C
0
(1 + ,) C
1
(4.27)
with anchor:
3 2 2
1 3
(0.8 2.5 2) (1 ) C C , , , , + + = +
(4.28)
The resulting passively loaded sheet pile wall length x follows from:
x l , =
(4.29)
The anchor force F
a
(horizontal component) per [m'] sheet pile wall equals:
without anchor: F
a
= 0 (4.30)
with anchor:
3
;
1
6( )
l l
R
a
m m
K x
F fdz fzdz
l x l x

=
+ +
} } (4.31)
The fixation force F
c
(horizontal component) per [m'] sheet pile wall is calculated from the
horizontal equilibrium:
2
;
2
l
c R a
m
x
F K F fdz
= +
}
(4.32)
after which the extra necessary penetration depth Ax according to the EAU [EAU; 1990] fol-
lows from:
'
2
c
c v
F
x
K o
A =
(4.33)
where:
o'
v
is the vertical effective stress at the fixation force on the right side of the sheet
pile wall;
Project number
415660-21 v04
Date
December 2007
Page
157 of 278
K
c
is the horizontal soil pressure factor on the soil retaining side at the location of
the fixation.
The total required sheet pile wall length l
d
is then:
o cos
x x l m
l
d
A + + +
= (4.34)
Finally the maximum moment has to be determined. In the no anchor situation a (negative)
fixation moment M
i
is present. In the anchored situation a distinction must be made between a
negative moment at the anchor M
a
, a positive field moment M
v
over the part of the wall be-
tween the anchor and the fixation in the soil, and a negative fixation moment in the soil M
i
.
The maximum moments are found at a distance z, for which applies that the shear force at that
location equals zero. The program contains a calculation method for this.
Appendix A displays a listing of the program. This program is not applicable for every situa-
tion.
Determining the depth of penetration
Theoretically the depth of penetration t
0
that is determined by Blums method, is the minimal
depth of penetration in case of a sheet pile wall with free earth support. In case of a fixed earth
support however, the undetermined force C must be delivered, which, working from the active
side of the wall, delivers the fixation moment. Lackner describes the AX, necessary for devel-
oping half of the C-force as:
' ' '
2 cos
p p
C
X
h K o
A =
(4.35)
Together with figure 4.22 this delivers the wanted penetration depth:
t = u + X + AX (4.36)
where:
u is the distance between the excavation level and the level of zero resulting
load;
X is the distance between the level of zero resulting load and the theoretical level
of penetration of the toe of the sheet pile wall;
' is the effective unit weight at the toe of the sheet pile wall;
h' is the theoretical comparable soil height above the foot;
K
p
is the passive earth pressure factor;
o
p
' is the wall friction angle..
Project number
415660-21 v04
Date
December 2007
Page
158 of 278
Figure 4.22 Schematization of the so called C-force according to Blum.
The C-force grows from zero (in case of a free earth support) to a maximum value in case of a
complete fixed earth support. With the following formula:
t = o (u + X) = o t
0
(4.37)
the necessary penetration depth can be determined using the values of o stated in table 4.2 and
the Blum formula given in 4.1.13.
Table 4.2 Factor o for the determination of the depth of penetration.
wall type
small resulting water
pressure
high resulting water
pressure
very high resulting water
pressure
unanchored 1.20 1.30 1.40 1.60
anchored with fixation 1.10 1.15 1.20 1.30
anchored free supprted 1.05 1.10 1.15 1.20
4.1.15 Sheet pile wall anchors
Anchoring a sheet pile wall is necessary for:
limiting deformations;
limiting the length of the sheet pile wall;
creating an as light as possible sheet pile wall profile;
spread of the effects of external concentrated loads over separate sheet pile wall sections, by
means of a wale construction.
p
h
t
D = point of rotation
e
ph
e
ph
e
ah
N
t
0
u
x
F
Ax
C 2Ax
h
F = theoretical point of the toe
o
p(+)
K
ph
K
ph
K
ah
(K
p
K
a
)
t = u + x +Ax
or
t = o . t
0
cross-section load figure approximated load figure
Project number
415660-21 v04
Date
December 2007
Page
159 of 278
The height on which the anchor is applied influences the field moment, the depth of penetra-
tion and the anchor force. The method to determine the bearing capacity of anchors by means
of tests and calculations, will be elaborated in 4.5.
The distance between sheet pile wall and anchor wall is determined when the stability of the
soil body between the sheet pile wall and the anchor wall is checked. The stability analysis is
described in 4.4.
4.2 Calculation model for a beam on elastic foundation
4.2.1 Introduction
Using a calculation method based on the theory of a beam on an elastic foundation, it is possi-
ble to implement complex boundary conditions in the sheet pile calculations, e.g. multiple an-
choring and execution stages. Even this calculation method requires adapted modelling of the
soil behavior.
The soil body is strongly schematized to an elasto-plastic medium without creep effects. The
chosen model has two significant consequences. First, taking into account the complexity and
the magnitude of the calculation, a computer is necessary to come up with a solution. Secondly,
it is necessary to have knowledge of the soil characteristics, specific for this calculation
method. The calculation model, based on the theory of the beam on elastic foundation, is cur-
rently accepted in the Netherlands. The plastic component of the model is taken into account.
4.2.2 Basic equation
The theory of the beam on elastic foundation was originally developed by Winkler and Zim-
merman [Hetenyi; 1946] for the calculation of sleepers and rails for the railway development.
The theory is applicable in other civil engineering problems, and is the basis for the following
calculation method for sheet pile walls and other retaining constructions.
A starting point is the well known elementary beam method, which states that:
flat cross sections remain flat (Bernoulli);
Hookes law applies;
normal forces are small, so they do not contribute to the deformation;
the angular displacements are small.
In this assumption the differential equation is:
2
2
d w M
dx EI
=
(4.38)
The balance relations for shear force and moment for a beam with a load q(x) result in:
2
2
( )
d M
q x
dx
=
(4.39)
These equations form the general equation for a beam with a constant EI:
Project number
415660-21 v04
Date
December 2007
Page
160 of 278
4
4
( )
d w
q x EI
dx
=
(4.40)
See also figure 4.23.
Figure 4.23 Calculation principle for a beam with an uniform distributed load.
For a beam on elastic foundation the external line load q(x) consists of the effective load f(x)
and the reaction of the elastic bed. This reaction is noted with g(x), so the equation transforms
to:
4
4
( ) ( )
d w
EI f x g x
dx
=
(4.41)
See also figure 4.24.
q(x)
Q
M
dx
dx
ds
dy
d|
d|
|
M
y
x
dM
M dx
dx
dQ
Q dx
dx
+
+
Project number
415660-21 v04
Date
December 2007
Page
161 of 278
Figure 4.24 Principle of a beam on elastic foundation according to Winkler.
The reaction of the soil g(x) depends on the soil characteristics, the geometry of the beam, and
the displacement w. For simplicity reasons, the relation between g and w is assumed linear:
cw g = (4.42)
In this case, the differential equation becomes:
4
4
( )
d w
EI cw f x
dx
+ =
(4.43)
The reaction of the soil g(x) and the load f(x) are both expressed in a force per length dimen-
sion. The counter pressure of the soil is expressed per area unit is:
g
p
b
=
(4.44)
where:
p is the soil pressure per area unit;
b is the width of the beam.
From this follows:
c
p w kw
b
= =
(4.45)
g(x)
f(x)
Q
M
dM
M dx
dx
dQ
Q dx
dx
+
+
dx
Project number
415660-21 v04
Date
December 2007
Page
162 of 278
Here k is the coefficient of subgrade reaction of the soil, expressed in [kN/m
3
]. This coefficient
of subgrade reaction is assumed independent of the width of the beam. For a two dimensional
schematization of the pile wall b equals the length unit and value of k, which is expressed as k
h
,
the horizontal coefficient of subgrade reaction, which is equal to that of c.
The differential equation, with these notations, becomes:
4
4
( )
d w
EI kbw f x
dx
+ = (4.46)
The analytical solutions for this differential equation, for a homogeneous bed, with a constant
coefficient of subgrade reaction, are given by Hetenyi [Hetenyi; 1946]. Generally the pile wall
bed is non homogeneous, and consists of several different layers.
Nevertheless, this analytical solution can be used for a first estimate. An assumption can be
that the pile wall is elastically fixed below excavation level. Above that level one may assume
that the active soil pressure and anchor forces act as loads, which implies that infinitely stiff
soil springs are assumed.
4.2.3 Pile wall as elastic supported beam
Soil characteristic
For the elasto-plastic method a schematization of the horizontal soil pressure as a function of
the sheet pile wall displacement is necessary. The relation between the soil pressure and the
sheet pile wall displacement, used in Blums method, is displayed in figure 4.25.
Blum only recognizes plastic areas with this schematization. The soil yields immediately pas-
sive or active, depending on the previously assumed direction of displacement of the wall;
against or with the soil pressure. The passive and active soil pressures are based on the balance
of a shearing body of soil in the failure situation. The disadvantage of this schematization is the
abrupt transition from active to passive yielding. For the elasto-plastic calculations, used in
most computer programs, a transition area is implemented between passive and active yielding,
which implies a linear relation between stress and displacement. For the relation between soil
pressure and pile wall displacement in elasto-plastic calculations, figure 4.26 is generally used.
Project number
415660-21 v04
Date
December 2007
Page
163 of 278
Figure 4.25 Soil pressure displacement diagram according to Blum.
Figure 4.26 Soil pressure displacement diagram for a spring supported beam.
displacement w
effective horizontal
stress o
p
passive failure surface
against the soil wedge
with the soil wedge
displacement w
against the soil wedge
with the soil wedge
elastic deformation
passive failure surface
effective horizontal
stress o
p
Ao
Aw
o
a
o
n
w
o
k
A
=
A
Project number
415660-21 v04
Date
December 2007
Page
164 of 278
Horizontal coefficient of subgrade reaction
For an elasto-plastic calculation it is assumed that every soil particle, for the entire height of
the pile wall in the model, reacts as a spring with a specific spring characteristic. These springs
are assumed to be present on both sides of the wall and not coupled. The rigidity of the soil is
determined by the horizontal coefficient of subgrade reaction k
h
, which is defined by the fol-
lowing relation:
h
k
w
o A
=
A
(4.47)
The horizontal coefficient of subgrade reaction is among others dependant on:
the soil type;
the water level;
the direction of displacement of the wall;
the stress history;
the effective pressure;
the length of the sheet pile wall;
the displacement;
the speed of loading, drained or undrained.
Horizontal effective pressure
The counter pressure of the soil against the pile wall is directly proportional to the displace-
ment of the pile wall, until it reaches a limit value. Principally at every section of the wall
where soil is present, minimally the active earth pressure and maximally the passive earth pres-
sure acts, while, depending on the displacement, the earth pressure has a value between these
two extremes.
Without displacement of the wall, this means, without deformation of the soil, a neutral effec-
tive pressure o
n
is present. In case of displacement against the earth body, the pressure will
increase linearly:
o
h
= o
n
+ k
h
w (4.48)
where:
o'
h
is the present effective pressure;
o'
n
is the present neutral effective pressure;
k
h
is the horizontal coefficient of subgrade reaction;
w is the horizontal displacement.
The effective pressure increases until passive failure occurs; this means the maximum (passive)
effective pressure o'
h;p;max
is reached. The pressure cannot increase any further. In case of a
displacement in the opposite direction, away from the earth body, the effective pressure will
decrease linearly:
o
h
= o
n
k
h
w (4.49)
Project number
415660-21 v04
Date
December 2007
Page
165 of 278
The effective pressure decreases until the minimum (active) effective pressure o'
h;a;min
is
reached. This will cause the earth body to shear.
o'
h;a;min
, o'
n
and o'
h;p;max
are calculated by the program for every location on the wall with values
that have been entered by the user.
Passive and active effective pressures
To determine the magnitude of the maximum values of both the active and the passive pressure
a reference is made to paragraph 3.2. The method using curved failure surfaces is recom-
mended.
If also the cohesion of the soil in the drained situation is added, then the interval for which the
soil behaves elastically, is longer, provided that the coefficient of subgrade reaction remains
constant, as is shown in figure 4.27.
Figure 4.27 Influence of the cohesion on the soil pressure displacement diagram.
4.2.4 Basicsl of the calculation model
Elaboration of the basic equation
In a one-dimensional finite element program, based on the displacement method, the differen-
tial equation for a beam on an elastic foundation can be solved. The springs are assumed not to
be linked, and to have bi-linear characteristics, with an elastic and a plastic component. Even
multi-component spring characteristics can be solved with such programs.
The calculation process is iterative. Displacements, rotations, moments, shear forces, effective
pressures, and water pressures are calculated.
The basic equation for a beam with a unit width is:
4
4
( , ) ( )
d w
EI k x w w f x
dx
+ =
(4.50)
influence of cohesion
increase of elastic deformation
displacement w
horizontal effecti-
ve stress o
o
n
Project number
415660-21 v04
Date
December 2007
Page
166 of 278
This equation is composed of the following three parts:
part 1: the rigidity of the beam;
part 2: the elastic foundation, the characteristic of soil reaction;
part 3: external loads and the loads in the initial phase of the elastic foundation.
The elastic foundation consists of:
earth springs;
elastic anchors.
The external loads are:
external (line) loads on the sheet pile wall;
neutral horizontal effective pressure in the initial situation;
water pressures;
initial anchor forces (including pre-tension forces).
The spring characteristic of the soil on both sides of the sheet pile wall is generally schema-
tized to a bi-linear spring, of which the second part is horizontal in case of respectively the
minimum active effective pressure or the maximum passive effective pressure. The spring
characteristic of the earth springs is of the form shown in figure 4.26.
The integration constants for the solution of the basic equation can be determined by entering
the prescribed displacements and forces.
The stability of the solution in case the displacement w is limited, can be proven. For this it is
necessary that, for a sheet pile wall without rigid supports, in at least two locations of the sheet
pile wall the soil springs are inside the elastic displacement interval.
The course of the calculation
The calculation procedure starts with an initially not deformed, sheet pile wall without internal
stresses, which is subjected to neutral effective pressures and water pressures.
On before hand the deflection of the sheet pile can only be predicted very roughly. Therefore
also the magnitude of the anchor forces and counteracting forces of the soil springs i only
known roughly, so the calculation process is executed iteratively. After each iteration, it is
checked if the assumed loads and stresses are coherent to the calculated displacements, and
reverse, it is checked if those displacements can cause the stresses and loads. Finally, equilib-
rium shall be reached when the dimensioning process is accurate.
In case of a combi-wall, piled wall or slurry wall, at the location of the toe, there is an extra
spring in the earth, due toe resistance of the foot in the soil. Generally, this can be taken into
account by increasing the rigidity of the lower soil springs.
Numerical approximation
For an accurate numerical approximation it is important to flawlessly describe the balance of
an element of the beam. This can be done by using the principle of equilibrium of an element
of random length, see figure 4.28.
Project number
415660-21 v04
Date
December 2007
Page
167 of 278
Figure 4.28 Schematization of the calculation.
The element is loaded by a) a uniform load of which the resulting force has a magnitude F
i+1
,
and is considered to be applied in the middle of the element, and b) a concentrated force P
i
,
which is applied just to the right of point x
i
. The reaction of the soil is a uniform load, which
can be assumed proportionate to the average displacement and the length of the element, and
the coefficient of subgrade reaction. Also it is assumed that the beam can still be supported by
a discrete spring with a rigidity f
i
at location x
i
. Balance perpendicular to the axis requires:
1 1 1 1
1
( )
2
i i i i i i i i i
Q Q F P R w w f w
+ + + +
= + + +
(4.51)
where R
i+1
= k
i+1
d
i+1
and d
i+1
= x
i+1
x
i
. If i is replaced by i 1 then the equation transforms
into:
1 1 1 1 1
1
( )
2
i i i i i i i i i
Q Q F P R w w f w

= + + +
(4.52)
Adding these two equations leads to:
1 1 1 1 1 1
1 1 1 1
1 1
( )
2 2
1
2
i i i i i i i i i i i
i i i i i i
Q Q F F P P R w R R w
R w f w f w
+ + +
+ +
= + + +
+ + +
(4.53)
The equilibrium of moments requires:
1 1 1 1 1
1 1 1
( )
2 2 2
i i i i i i i i i i
M M d Q Q Pd f wd
+ + + + +
= + +
(4.54)
If i is replaced by i 1 then the equation transforms into:
1 1 1 1 1
1 1 1
( )
2 2 2
i i i i i i i i i i
M M d Q Q P d f w d

= + +
(4.55)
Eliminating Q
i
from this equation leads to the following equation:
{
1 1
1 1
1 1 1 1 1
1 1 1 1
1
2
i i i
i i i i
i i i i i i i
i
M M M
d d d d
Q Q P f w P f w
+
+ +
+
| |
+ + =
|
\ .
+ +
(4.56)
M
i+1
Q
i+1
F
i+1
P
i
Q
i
M
i
d
i+1
Project number
415660-21 v04
Date
December 2007
Page
168 of 278
or, by using the previously given equation:
1 1
1 1
1 1 1 1
1
1 1 1 1
1 1 1
( )
4 4 4
1 1
2 2
i i i
i i i i
i i i i i i i
i i i i i
M M M
d d d d
R w R R w R w
F F P f w
+
+ +
+ + +
+
| |
+ +
|
\ .
+ =
+
(4.57)
This is the first basic equation (B1) for the numerical solution.
The second basic equation (B2) is derived as follows:
Observe the two elements to the left and right of point x
i
in figure 4.28. In the left element,
element i, the following applies:
x < x
i
:
2
1 2
1
( )
2
i i
d w
EI M M
dx

= +
(4.58)
for which is assumed that the bending moment in the element is equal to the average of the two
bending moments at both ends of the element.
On the right side of the element, element i + 1, the following applies:
x > x
i
:
2
1 2
1
( )
2
i i
d w
EI M M
dx
+
= +
(4.59)
Integrating these two equations leads to (provided that the terms on the right hand side are con-
stants):
x < x
i
:
2
1
1
( )( ) ( )
4
i i i i i
EIw M M x x A x x EIw
= + + +
(4.60)
and for
x > x
i
:
2
1
1
( )( ) ( )
4
i i i i i
EIw M M x x A x x EIw
+
= + + +
(4.61)
where the integration constants are chosen in such a manner that for x = x
i
the displacement
always equals w
i
and the slope remains constant A/EI.
Substitution of x = x
i-1
and x = x
i+1
in respectively (4.60) and (4.61) leads to two different equa-
tions for A. After eliminating A these equations lead to:
1 1
1 1 1
1 1 1 1
1 1 1
( ) 0
4 4 4
i i i
i i i
i i i i i i i
EI EI EI EI
w w w
d d d d
d M d d M d M
+
+ +
+ + +
| |
+ + +
|
\ .
+ + + =
(4.62)
This is the second basic equation (B2).
Project number
415660-21 v04
Date
December 2007
Page
169 of 278
Accurate values of the coefficients can only be obtained by means of a careful analysis. The
boundary conditions have to be approximated numerically. This also requires a careful consid-
eration. It is preferable that the boundary conditions on both sides of the beam are expressed in
the displacement w and the bending moment M. In case of a hinge-joint both w and M are obvi-
ously zero. In case of other boundary conditions, e.g. a fixed or free support, then it might be
necessary to manipulate the boundary conditions in order to express them in the two variables
w and M. This to avoid that there are two boundary conditions for one variable and none for the
other. That might lead to numerical problems when solving the system of equations.
The numerically discretized system of equations (B1) and (B2) can be described in matrix no-
tation as follows:
j j
t e
S V U f ( + =

(4.63)
j S U f =
(4.64)
in which:
[S
t
] is the contribution to [S] as a result of the beam elements;
[V
e
] is the contribution of the system matrix [S] as a result of the spring elements;
U is the vector with the unknown displacemens and forces;
f is the vector with loads.
At first in the vector f only terms that result from external line loads, water pressures, neutral
effective pressures and pre-tension forces of the anchors are taken into account. All springs are
assumed to react elastically. The solution is then found from:
j j
1
t e
U S V f
( = +

(4.65)
The assumption that all springs react elastically is checked by means of this solution. Then the
following might be the case:
1. the springs react as was assumed:
The final solution has been reached. The program continues with the next task.
2. not all the springs react as was assumed:
a. there are less then two springs which are elastic:
The sheet pile wall is instable, the calculation is terminated
b. there are at least two springs which are elastic;
In this case the active and passive earth pressures on the sheet pile wall are categorized
as loads in the vector with loads, as long as the force in a spring exceeds the passive
pressure, or is smaller than the active pressure. The next iteration step might show that
a spring reacts elastically, while it was assumed plastically. In this case the concerned
load will be extracted from the load vector.
The total matrix [[S
t
] + [V
e
]] is then decomposed and the new iteration for the dis-
placement vector is determined.
The calculation is halted when it is shown that all the springs react as assumed, or when the
sheet pile wall turns out to be instable. The calculation procedure is described schematically in
figure 4.29.
Project number
415660-21 v04
Date
December 2007
Page
170 of 278
Figure 4.29 Calculation procedure.
Transition to another construction stage in the calculation
The calculation model can calculate different building stages in line, where the stress history of
the sheet pile wall is taken from one stage to the next. It is also possible to change the soil
characteristics in the new stage, in case of excavation or fill of soil. In the new stage new
spring characteristics are applied.
The processing is as follows:
In case of multiple building stages the total load on the wall, including those which result from
elastically reacting springs, is calculated at the end of each stage.
At the transition from one stage to the next, the characteristics and earth pressures in a soil
layer can vary significantly, in case of excavation or fill of soil. Therefore new spring charac-
teristics are applied in the new stage.
To determine the new spring characteristic in a point along the wall, the following principles
are valid:
yes
no
end of calculation
process for this stage
calculate new approximation of the dis-
placement
determine effective horizontal pressures
that belong to these displacements
are these effective horizontal pressures
different from these from the previous
iteration step?
determine effective horizontal pressure
following from the deflection moment line
of the previous stage or from the 0-stand
start calculation process for a new phase
Project number
415660-21 v04
Date
December 2007
Page
171 of 278
Figure 4.30 Alteration spring characteristic in case of excavation or fill of soil.
Assume the spring characteristic (1), shown in figure 4.30, is valid for the old situation. For the
considered point on the wall the displacement w
0
and horizontal effective pressure o
h
are valid.
The wall is then assumed fixed during the excavation or fill of soil. Also is assumed that the
increase of the horizontal effective pressure equals the difference between the neutral effective
pressure before and after the stress alteration:
*
h h
o o o A =
(4.66)
On the old spring characteristic the point (w
0
, o
h
) is located. From this point the initial stress
situation can be determined:
Point A:
* *
0
( , )
h h h
w with o o o o = + A
(4.67)
This point is located on the new spring characteristic (2), shown in figure 4.30, with its own
specific parameters:
k
*
is the slope of the elastic interval, the coefficient of subgrade reaction;
* *
,
a p
o o
are the effective stress limits.
It is checked that the new stress point is located inside these two limits. If this is not the case,
the stress point is assumed equal to the limit stress.
The wall is then released and the displacement as a result of the extra stress in the new stage is
calculated according to the previously discussed calculation procedure, which will use the new
spring characteristics.
displacement w
horizontal
soil pressure
o
*
p
o
p
(2)
(1)
o
*
h
o
h
o
*
a
o
a
w
0
A
arctan k
*
arctan k
Ao = o
*
h
o
h
Project number
415660-21 v04
Date
December 2007
Page
172 of 278
Dealing with surcharge loads
Dealing with the effects of surcharge loads is done by means of determining the effective hori-
zontal earth pressure on the wall as a result of this surcharge load. In this case a homogeneous
linear elastic halfspace, or active or passive slide planes, as occur when failure slides act, may
be assumed. The equations apply for line loads and strip loads and satisfy the principles of
superposition.
The determined horizontal effective pressures are entered in the load vector. The method for
entering surcharge loads in sheet pile wall programs is analysed [Van den Berg; 1990]. These
studies show that a distinction must be made between neutral pressure situation and active
pressure situation.
It is proposed to enter surcharge loads by means of the following models.
For a neutral pressure situation it is recommended to determine the horizontal effective pres-
sure by means of Boussinesqs method, corrected with a factor f(x,l). This factor takes into
account the distance between the load and the wall, and the length of the wall.
2
; 4
2 2
2
( , )
( , ) 2
( , ) 1
( )
xx z
px z
f x l
r
x
f x l for x l
l
f x l for x l
r x z
o
t
=
= s
= >
= +
(4.68)
where:
o
xx;z
is the horizontal stress against the wall as a function of the depth z;
x is the distance between the line load and the wall;
p is the line load;
l is the length of the wall.
In case of an active stress situation in the active zone behind the wall, Culmanns method is
recommended, with the following additions:
the cohesion of the soil must be taken into account by means of assuming a force T
c
along
the slip surface, as a result of the effective cohesion of the soil:
'
cos
c
l
T c
u
=
(4.69)
where:
c' is the effective cohesion of the soil;
l is the length of the sheet piles;
u is the angle between the slip surface and the wall;
Project number
415660-21 v04
Date
December 2007
Page
173 of 278
the effect of multiple soil layers must be taken into account by means of the following
method:
The slip surface in each layer must have the opportunity to determine the optimal direction.
The decisive failure plane intersects n layers and therefore consists of n straight parts of the
failure surface. The total horizontal force Q
h
on the wall is then a summation of n different
contributions.
In case of active earth pressure the optimal direction is the direction of the part of the
failure surface of which the contribution to the horizontal earth pressure is maximal. In case
of passive earth pressure the optimal direction is the direction of the part of the slip sur-
face of which the contribution is minimal.
For each layer, to determine the contribution of each layer to the earth pressure on the wall,
the balance of the triangular wedge, which is cut by the failure plane, is considered. Here a
surcharge load is applied to the top of the wedge, which can be determined by means of the
local effective vertical pressure (See W
3
for the determination of Q
3
in figure 4.31);
Figure 4.31 Failure plane with multiple layers behind the sheet pile wall;
(Q
h
)
tot
= (Q
h
)
1
+ (Q
h
)
2
+ (Q
h
)
3
.
pressure envelope over the depth:
To get an impression of the envelope of the pressure against the wall, the horizontal pres-
sure can be determined on a number of levels. The wall is the divided in a number of ele-
ments. Then the decisive failure surfaces are determined, which intersect the bottom of each
element, as is sketched in figure 4.32. For each decisive failure surface its force on the wall
Q
h;j
is calculated. The horizontal pressure on element j is then calculated by:
1
( ) ( )
( )
h j h j
h j
e
Q Q
l
o
+

=
(4.70)
where:
l
e
is the length of the element j;
Q
1
Q
2
Q
3
o
1
o
2
o
3
N
3
T
3
N
2
T
2
N
1
T
1
W
1
W
2
W
3
O
1
O
2
O
3
layer 1
layer 2
layer 3
4 3 2 1
Project number
415660-21 v04
Date
December 2007
Page
174 of 278
1
2
(Q
h
)
j
(Q
h
)
j+1
(o
h
)
j
element j with
length l
e
1 = controlling shear band j
2 = controlling shear band j+1
Determining the pressure envelope against the wall takes a lot of calculation time,
especially with the presence of a random surcharge load or a layered soil mass. This
however, can be done with a computer program.
Figure 4.32 Construction of the course of soil pressure, over the depth of the sheet pile wall.
Application options
The available sheet pile wall programs, based on the beam on elastic springs theory, offer the
possibility to design complex sheet pile wall constructions, which are only approximated
roughly by Blums and other methods.
These programs offer the following application options and advantages, which have also been
schematized in figure 4.33:
calculating multiple anchored sheet pile wall constructions;
anchors and struts can be entered as elastic or rigid supports, by means of a bi-linear spring
characteristic;
horizontal forces and moments can be applied to the sheet pile wall; this creates the option
to pre-stress anchors, with even the option to change the magnitude of this pre-stress force
per construction stage;
a vertical load can be applied to the head of the sheet pile wall;
the influence of the construction stages on the dimensioning can be analyzed. Construction
stages can be successively analysed. In these calculations the stress envelope and
deformation, calculated in the last stage, are taken into account in the next stage. That
includes the elastic deformation of the wall and supports and the elastic and plastic
deformations of the soil;
because the earth pressure envelope in the elastic phase of the spring characteristic is a
derivative of the deformations and the deformation direction of the sheet pile wall, a more
realistic earth pressure envelope is calculated. Its effect is more prominent in case of sheet
pile wall displacement in the elastic interval of the spring characteristic. The effect of
increase of soil stress at the top of the wall, which is a result of the deformation of the sheet
Project number
415660-21 v04
Date
December 2007
Page
175 of 278
above a relatively rigid anchor, is taken into account in the calculation. The effect is
sketched in figure 4.34;
by means of tuning the coefficient of subgrade reaction and the maximum passive and the
minimum active earth pressure, the sheet pile wall programs can be used to design
horizontally loaded piles, Berlin-walls, dolphins and others.
Figure 4.33 Possibilities of the beam on elastic foundation method.
surcharge load
normal force
excess water pressure
different soil types
and layers
variable wall stiff-
ness
loads
or mo-
ments
fixed or
spring
support
anchors with/without pre-tensioning
Project number
415660-21 v04
Date
December 2007
Page
176 of 278
Figure 4.34 Increase of soil stresses above anchor level.
4.2.5 Limitations
When applying the sheet pile wall programs, based on the theory beam on elastic foundation,
the relation between the earth pressure on the sheet pile wall and the displacement of the sheet
pile wall in the elastic interval is determined by the horizontal coefficient of subgrade reaction.
This relation is assumed linear in the elastic interval for most of the sheet pile wall programs.
The springs, which represent the behavior of the soil, are assumed unlinked in both the hori-
zontal and vertical direction. This is shown in figure 4.35. This means a rather significant sim-
plification of the real behavior of the soil. The horizontal coefficient of subgrade reaction de-
pends on of the magnitude of the loaded area, and therefore also implicitly on the flexural ri-
gidity of the sheet pile wall and on the level of stress.
I reality the stress-displacement behavior of the soil is not bi-linear and would be better repre-
sented by a multi-linear spring characteristic. Some computer programs can process such char-
acteristics.
Next to the rough schematization of the soil by means of the coefficient of subgrade reaction,
the sheet pile wall programs, based on the beam on elastic foundation theory, have the fol-
lowing limitations:
the wall is in principle vertical. By means of alteration of the active and passive earth
pressures the influence of a slope of a sheet pile wall can be taken into account within a
certain interval;
only the component of the anchor force and the strut force perpendicular to the wall can be
taken into account;
in the wall plastic joints cannot develop. This implies that in dimensioning, one cannot use
the fully plastic moment, but must use the moment with which the yielding stress in the
outer particles of the pile section has been reached;
active
passive
increase of soil
stress
Project number
415660-21 v04
Date
December 2007
Page
177 of 278
the calculation is valid for a plane strain situation. The sheet pile wall construction must be
of significant length (horizontally). If this is not the case, e.g. in case of a single pile, then
the 3-dimensional effect has to be implemented in the input values. Neglecting this,
however, results in a safe approximation;
only the layer system over the height of the sheet pile wall influences the calculation results.
If weaker layers are present below the toe of the wall, then total stability, of the wall and the
soil mass, must be checked by means of a slip failure calculation;
most sheet pile wall programs are most suitable for checking a design rather than making a
design. So at the start of the calculation the flexural rigidity and the length of the wall need
to be known. During the calculation the stability, strength and deflection can be checked;
the ground levels on both sides of the wall are in principle horizontal. By means of
alterations of the active and passive earth pressure, the influence of slopes in the ground
level, to certain extends, can be taken into account;
because the springs in the soil are assumed unlinked, the effects of arching, as described in
EAU [EAU; 1990], are not automatically taken into account in the computer calculations.
Figure 4.35 Spring models, coupled and uncoupled;
a. homogeneous wall; the geometry fulfils almost the requirements of a plane strain
situation. The fact that the springs are coupled in lateral direction has no influence
on the mechanical behavior;
b. non-homogeneous wall; here the coupling between the soil springs is significant.
uncoupled springs
transfer of loads only by
bending stiffness of the
sheet pile wall
a.
cross-section vertically uncoupled
top view-section horizontally uncou-
pled
uncoupled coupled
pile
Project number
415660-21 v04
Date
December 2007
Page
178 of 278
4.2.6 Input parameters
Of the following parameters the design values for the considered limit state must be entered in
the computer program:
parameters that define the geometry;
parameters that represent the characteristics of the sheet pile wall;
soil information, ground water levels and soil characteristics;
stages during construction, including installation of anchors and struts;
externally applied loads on the sheet pile wall and surcharge loads on both the active and
passive side of the wall.
Paragraph 2.4 shows how to determine these design values.
Geometry and sheet pile wall characteristics
Sheet pile wall
The discussed sheet pile wall or anchor wall consists of by means sheet pile elements, con-
nected by interlocks. Every separate sheet pile element has a moment of inertia, which depends
on the thickness, shape, material and width of the section.
The input parameters usually relate to a sheet pile wall width of one meter, but are easily
changed by means of a conversion factor, so even walls, consisting of sheets with alternating
longer and shorter elements (in dutch: getaffelde wanden), Berliner walls, piles and dolphins
can be analysed.
Rigidity of the sheet pile wall
The rigidity of the sheet pile wall is determined by the EI-value of the applied section, where
the I-value must be entered after reduction because of oblique bending (see paragraph 3.3).
Due to the elastic soil mass, the earth pressure on the wall is deformation dependent and the
earth pressure envelope is different compared to Blums method.
In case of a increasing flexural rigidity of the sheet pile wall, and a similar coefficient of sub-
grade reaction, the field moment will increase.
The load is transferred by the sheet pile wall to the anchors and the supports or the fixation at
the toe of the wall. An increasing rigidity of the wall, compared to the rigidity of the soil
springs, causes an increase of the span and thereby an increase of the moment in the wall.
A sheet pile wall with sheets of high flexural rigidity will, in combination with a rigid anchor,
develops above the anchor level a larger earth pressure than the active earth pressure, due to
the rotation of the wall into the soil mass above the anchor level. This results in a moment at
the level of the anchor and the anchor force, which increases when the flexural rigidity of the
sheet pile wall increases.
Length and penetration depth of the sheet pile wall
As shown in paragraph 2.4, the minimal depth of penetration, or the length of the wall, can
initially be determined by means of a calculation using Blums method. After executing a cal-
culation by means of a computer program based on the beam on elastic foundation- theory, a
decision to increase the penetration depth can always be made. This may optimize the section.
Project number
415660-21 v04
Date
December 2007
Page
179 of 278
Top displacement and top fixation
In some computer programs, based on the beam on elastic foundation- theory it is possible to
prescribe a top displacement, meaning a horizontal displacement at the top of the sheet pile
wall. It can also be possible to enter an angular displacement of zero into the program, in case
of a sheet pile wall that is fixed at the top.
Geometry soil characteristics and ground water levels
Surface levels in front of and behind the sheet pile wall
In the Netherlands, the soil levels on both sides of the wall are generally horizontal. For every
calculation the altitude of the ground level on both sides of the wall has to be entered.
Future heightening of the ground level has to be entered as surcharge loads. Heightening,
which is a part of the construction stage can be included by means of an alteration of the
ground level in a next stage (see figure 4.36).
Figure 4.36 Ground level heightening and construction stages with fill.
phase 1 phase 2
phase 1 phase 2
surcharge load
excavation
fill
Project number
415660-21 v04
Date
December 2007
Page
180 of 278
Soil layers
The following must be taken into account specifically during the input of soil layers:
the type of soil, thickness of layers and characteristics must be determined according to
paragraph 3.4. Design values for these parameters must be used in the calculation.
Directives for the determination of these design values can be found in paragraph 2.4.
In this case the following is of importance:
the earth pressure factors for the active and passive failure situation, and for the neutral
situation: K
;a;d
; K
;p;d
; K
sur;a;d
; K
sur;p;d
; K
c;a;d
; K
c;p;d
; K
o;d
;
the wall friction o
d
;
the angle of internal friction u
d
';
the cohesion c
d
';
the dry and wet unit weight
d
and
sat;d
;
the permeability k
d
;
the horizontal coefficient of subgrade reaction k
h;d
;
the unit weight of water
w
;
the water pressure;
the location of the top of the soil layers.
In most programs there is an option to choose to calculate the earth pressure factors by the pro-
gram itself, or to enter them directly from tables or calculations.
Water level and ground water level
The design values of the phreatic water level and the groundwater level on both sides of the
wall must be re-entered every calculation stage.
If no excess water pressure is entered, then the calculation is made using hydrostatic water
pressure.
Excess water pressures
Most programs have a feature to enter excess pore pressures in different soil layers. The magni-
tude of the excess pore pressure depends on, for example, changing phreatic water levels, flow
of groundwater, permeability of the soil. The presence of drains behind the wall plays a signifi-
cant role.
Excess water pressure can occur, for example, in layers that either have a low permeability or
are enclosed by non-permeable soil layers. Excess water pressure can occur during the consoli-
dation phase due to the presence of a surcharge load, or during the installation of piles. Be-
cause the ground level is unable to flow immediately, the increased earth pressure is com-
pletely or partially, depending on the permeability, converted to excess water pressure.
After some time the grain skeleton of the soil will adapt to the load and the excess water pres-
sure will decrease and disappear. The effective vertical stress will have increased proportion-
ally. Generally it is necessary to calculate the sheet pile wall for both the situation without
consolidation and with complete consolidation. The load on the wall will usually be unfavor-
able in case of no consolidation.
Calculations with 100 % excess pore water pressure are labeled as undrained situations,
where in the concerning layer calculations must be made using undrained behavior of the soil.
This for both the coefficient of subgrade reaction and the shear strength f
undr;d
.
Project number
415660-21 v04
Date
December 2007
Page
181 of 278
Negative excess pore pressures can occur due to excavations or dewatering systems. Entering
excess pore water pressures into a computer program is possible because, for each layer, the
difference with the hydrostatic pressure course can be entered. Entering excess pore water
pressure is generally done linearly between layers. If the excess pore water pressure course is
somewhat strange or irregular, then by means of choosing more layers (as shown in figure
4.37), it can be approximated more accurately.
Figure 4.37 Course of water pressure in case of irregularities.
Phasing with anchors and struts
Construction stages and calculation phases
Sheet pile wall programs based on the beam on elastic foundation theory can calculate con-
secutive construction stages. A construction stage causes a change in the load on the wall.
These changes can be caused by:
excavations;
fills;
surcharge loads (e.g. traffic loads);
changes of water level;
placing or removing supports (e.g. anchors, struts);
increasing, decreasing or removing loads directly applied to the wall (e.g. pre-tension
forces);
placing or removing extra vertical loads on the wall.
The system calculates the displacement relating to the previous construction stage and thereby
follows the stress and displacement history of the wall. These characteristics have to be entered
prior to every stage.
surcharge load
excess water pressure
Project number
415660-21 v04
Date
December 2007
Page
182 of 278
Elastic and rigid supports by means of anchors and struts
A sheet pile wall can be supported at every level by means of translation springs and rotational
springs. Rigid supports, where translation and rotation equal zero, can also be simulated. This
can differ in every phase. The wall can be supported by multiple anchors (pre-stressed or not).
These anchors have to be entered in and removed from the program conform the actual situa-
tion, this means per stage. The traditional anchor (not pre-stressed) is assumed to be an elastic
support of which the stiffness is calculated from the modulus of elasticity, the cross sectional
area of the anchor beam and the fictive length of the anchor. In this fictional length not only the
influence of the total anchor has to be considered, but also its slope. The failure anchor force
(in case of total failure), which has to be entered, counts as the ultimate anchor force. The
minimal anchor force is zero, so there can be no compressive force in the anchor. After poten-
tial failure, hysteresis is taken into consideration in the course of the force-displacement-
diagram.
Besides anchors, which can only take tensile forces, the wall can be supported by one or more
struts, which can handle significant compressive forces (and sometimes even tensile forces).
The fact that the maximum force is limited by the possible buckling of the strut has to be taken
into account.
Pre-stressing of the anchor
The earth pressure envelope on the sheet pile wall and the anchor displacement are closely
related and determine, in case of a beam on an elastic foundation, largely the dimensions of the
wall. See figure 4.38.
When applying pre-stressed anchors the pre-stress can be entered into the program either as a
pre-stress force in the anchor or as a horizontal force on the wall. The pre-stress force has to be
entered in every phase for every anchor. The load on the wall, caused by this pre-stress force, is
applied as a line load. The processing of these pre-stress forces differs from program to pro-
gram. For every program it has to be checked thoroughly how pre-stress forces have to be en-
tered.
Figure 4.38 Influence of pre-stressing the anchor.
w
anchor
stiff anchor
flexible anchor
Project number
415660-21 v04
Date
December 2007
Page
183 of 278
Externally applied forces on the sheet pile wall
Directly applying forces
The programs offer the opportunity to enter loads due to horizontal forces and moments, which
apply directly on the wall. These loads remain present in the each next stage until they are re-
moved. In most programs however, it is impossible to enter axial forces in a manner, which
takes into account the second order effects of these forces. Some programs do possess that
ability. In that case it is then also possible to take into account the frictional forces along the
wall (see figure 4.39). The influence of these second order effects on the deformation is taken
into account as well.
Figure 4.39 Frictional forces and vertical loads on the wall (second order effects).
Surcharge loads
Surcharge loads by means of local or uniformly distributed loads, for both sides of the wall,
can be entered in each stage.
4.2.7 General view on parameters
Although the coefficient of subgrade reaction describes the relation between horizontal effec-
tive pressure and the displacement of the soil, it is usually the case that the calculated load
envelope over the wall (and the moments) are more realistic then the calculated displacement.
A less accurately estimated value of the coefficient of subgrade reaction has a relatively small
influence on the calculated moments. This, on the one hand, is caused by the fact that the mo-
ment is proportional to the root of 4
th
-order of the coefficient of subgrade reaction. On the other
hand, most of the springs are located in the active or the passive branch of the spring character-
istic during the failure situation. This causes the load on the wall, and thereby the occurring
N
N
w
N
2nd order
= N . w
Project number
415660-21 v04
Date
December 2007
Page
184 of 278
moment on the wall, to be partly independent of the magnitude of the coefficient of subgrade
reaction.
The present sectional forces and displacements are, in case of constant load, depth of penetra-
tion and such, not only determined by the coefficient of subgrade reaction, but also by the ri-
gidity of the wall, the coefficient of subgrade reaction and the stiffness of the anchors in pro-
portion to each other.
In case of a not anchored sheet pile wall, the coefficient of subgrade reaction has a minimal
influence on the maximum moment, but a big influence on the deformation. The influence of
the flexural rigidity of the wall is insignificant for both, which is shown in figure 4.40.
Figure 4.40 Influence of the flexural rigidity on the (unanchored) sheet pile wall. The occurring
moment is the same in both cases.
In case of an anchored sheet pile wall the relation between the sheet pile wall parameters, the
coefficient of subgrade reaction, the moments and the deformations is not so self-evident and
less simple. In figure 4.41 the influence of variations of the above stated factors is schemati-
cally sketched.
In case of anchored, relatively flexible walls, the deformation pattern of the wall and the an-
chor forces is strongly influenced by the stiffness of the anchor. Once appeared plastic defor-
mations are usually permanent. To undo these deformations the wall must be moved in the
direction of the soil mass, which will lead to relatively high passive pressures during pre-
tensioning of the anchors. These forces can usually not be delivered by the anchors.
high
bending stiffness
low
bending stiffness
high modulus of sub-
grade reaction
k = big
low modulus of sub-
grade reaction
k = small
Project number
415660-21 v04
Date
December 2007
Page
185 of 278
Figure 4.41 Influence of rigidity of the sheet pile wall and the soil stiffness bed in case of an an-
chored wall.
4.2.8 Input check
The input can usually be checked by means of an automatic sketch of the geometry of the sheet
pile wall and the soil. The program usually checks both for syntax and logical errors.
The non fatal errors are in this case called warnings. This implies that the program will, after
diagnosing such an error, correct the error itself and complete the run with the given input val-
ues.
In case of fatal errors an error message will appear in which case the program will not continue
the calculation. The program stops after diagnosing such an error.
low modulus high modulus of
of subgrade reaction subgrade reaction
low modulus high modulus of
of subgrade reaction subgrade reaction
flexible anchor
stiff anchor
low bending stiffness
high bending stiffness
Project number
415660-21 v04
Date
December 2007
Page
186 of 278
4.2.9 Execution of the calculation
Output data and interpretation
The output data is logically arranged for each phase. All relevant input data per phase is
shown, combined with the calculated data from that particular phase.
The output can for example consist of the following:
E. display of general input values;
F. for each construction stage or calculation phase:
G. a complete display of the relevant input values;
H. the calculated data per level of terrain envelope;
I. level in relation to [N.A.P.];
J. horizontal displacement in [m];
K. rotation in [rad];
L. shear force force in [kN];
M. moment in [kNm];
N. water pressure on left side in [kN/m
2
];
O. effective pressure on left side in [kN/m
2
];
P. water pressure on right side in [kN/m
2
];
Q. effective pressure on right side in [kN/m
2
];
R. proportion of the occurring effective pressure to the maximum effective pressure
in [%].
The forces and pressures are per [m'] sheet pile wall.
The percentage of the last point in the above list displays to which extent the effective stress is
mobilized at that given location. 100 % indicates that the maximum or the minimum force of
the given soil spring, respectively passive or active, has been reached. Not all programs calcu-
late this percentage.
Besides, the following are executed as well:
the possible support reactions;
indications referring to the stability and possibly still present reserves. Some programs
calculate the proportion of the mobilized support pressure to the maximum passive earth
pressure on the passive side.
Review of the output
Initially the following output data has to be checked:
do the surface levels match with the calculated effective stresses;
does the input concerning the water tables, combined with the excess pore pressures, match
with the calculated pressures against the wall;
is the influence of the entered loads processed correctly.
By using the calculated moments in the wall, possibly combined with the shear forces, the
strength of the chosen sheet pile section can be checked.
The calculated anchor forces can be used as input values when checking the anchor system.
Project number
415660-21 v04
Date
December 2007
Page
187 of 278
The influence of the calculated deformations on the following has to be checked:
construction details of the sheet pile wall construction, like corner-design and the
connections of other construction elements;
adjacent foundations, in connection with possible vertical and horizontal displacement of
the soil behind the wall.
It also has to be checked if the deformation criteria and demands have been met.
At last the present stability reserve in the soil has to be judged. The proportion of the total of
the mobilized passive earth pressure to the total of the completely passive earth pressure can be
used as a judgment criterion. If design values for loads, geometry and soil characteristics have
been entered, then this proportion value may equal to 1.
When entering values for a sheet pile wall calculation, the values of sheet pile wall length,
rigidity, coefficient of subgrade reaction and stiffness of the anchors must be chosen realisti-
cally. A too high sheet pile wall rigidity, in combination with a too long sheet pile wall and a
too rigid anchoring, produce the results shown in figure 4.42. These results are technically
correct, but do not lead to a correct sheet pile wall design.
Figure 4.42 Input data and calculation results for the calculation example with varying parameters
for the moment of inertia I;
I = 10
-5
m
4
/m
1
; * I = 10
-4
m
4
/m
1
;
o I = 10
-3
m
4
/m
1
; A I = 10
-2
m
4
/m
1
.
0 2 4 6 8 10 12 14
depth [m]
0 2 4 6 8 10 12 14
depth [m]
0 2 4 6 8 10 12 14
depth [m]
r
e
s
u
l
t
i
n
g

e
a
r
t
h

p
r
e
s
s
u
r
e

[
k
N
/
m
2
]
m
o
m
e
n
t
s

[
k
N
/
m
]
d
i
s
p
l
a
c
e
m
e
n
t
s

[
m
m
]
150
130
110
90
70
50
30
10
-10
-30
60
40
20
0
-20
-40
-60
60
40
20
0
-20
5
10
15
k = 10,000 kN/m
3
strut
construction stage:
excavation
0
-2
-4
-6
-8
-10
-12
-14
-16
Project number
415660-21 v04
Date
December 2007
Page
188 of 278
4.2.10 Specific aspects
Groundwater flow
Differences in water levels can be taken into account by means of the method shown in para-
graph 3.2.4. The unit weights, adapted to the water flow, can be entered in the programs.
Over-consolidation
The naturally present neutral earth pressure is determined by the load history of that particular
layer. Due to high surcharge loads in the past (e.g. due to the weight of glaciers during the gla-
cial periods) the horizontal pressures can remain significantly larger then the reduced vertical
pressures. This is called over-consolidation.
Soil investigation can determine the extent of over-consolidation, by means of e.g. a pressiome-
ter test. Usually the value for this over-consolidation has to be estimated. The equations below
[Brookes et al.; 1965] give an estimate of the neutral earth pressure factor K
0
:
K
0
= K
0;nc
(OCR)
x
(4.71)
where
K
0;nc
= K
0
for normally consolidated soil;
previous effective stress
current vertical effective stress
OCR = ;
x = an exponent that depends on the soil type;
for sand: x = 0.45;
for clay: x = 0.5.
Figure 4.43 shows how this high neutral earth pressure factor due to over-consolidation is
processed in the calculation. The spring characteristics are determined by it during the first
iteration.
Figure 4.43 Influence of overconsolidation.
w
o
o
p

o
a

o
n
= k
0;nc
. o
v

o
n;0
= k
0;nc
(OCR)
x
. o
v

influence of overconso-
lidation on o
n
Project number
415660-21 v04
Date
December 2007
Page
189 of 278
Inclined ground level
One of the starting points of the calculation model is a horizontal ground level on both the ac-
tive and the passive side of the wall.
A sloping ground level on either the passive or the active side of the wall can be taken into
account by means of the method presented in paragraph 3.2.2. It is customary to neglect the
effect of a sloping ground level on the coefficient of subgrade reaction.
When installing a sheet pile wall in a sloping area, horizontal effective pressures are already
present which differ from the regular effective pressures in case of a horizontal ground level.
One has to be careful when using the regular starting points for a horizontal or sloping ground
level. It is advisable to perform the sheet pile wall calculation by means of a program that uses
the finite element method. Such a program calculates the stresses in the initial phase before
installing the sheet pile wall; after that the loads (excavations) were put into the calculation.
Increased pressures at the top and moment reduction
By connecting the earth pressure envelope to the deformations of the sheet pile wall, the influ-
ence of the increased soil stresses near the top and the moment reduction on the field moment
will be processed by the program itself. The arching at the active side of the wall which occurs
in non-cohesive soil, as sketched in 4.44, however, is not simulated by the calculation model,
because the springs are entered as unlinked in vertical sense. In some cases, under specific
circumstances, a reduction of the field moment is allowed, as is shown in the EAU for Blums
method [EAU; 1990] It is, however, unadvisable to take this mitigating influence into account.
Figure 4.44 Arching of the soil behind the sheet pile wall.
arching in
non-cohesive soils
Project number
415660-21 v04
Date
December 2007
Page
190 of 278
Figure 4.45 Determination of the influence of cracks in a slurry wall on the course of the moments
by iteration.
First calculation
Last calculation after adaption of stiffnesses
EI
0
EI
g
EI
0
EI
0
crack
criterion
uncracked
uncracked
cracked
uncracked
uncracked
cracked
Project number
415660-21 v04
Date
December 2007
Page
191 of 278
Influence of densification of fill
Due to influence of densification of thin layers behind the sheet pile wall, it might occur, in
case of a multipe anchored wall, that the lowest anchor gets overloaded due to the increased
earth pressures.
Because of the uncertainty concerning the anchor force and the wall displacement, it is advised
to pre-stress the anchor only partially, or not at all, before applying a fill.
Diaphragm walls
When applying diaphragm walls, due to cracking, a relation exists between the occurring bend-
ing moment in the wall and the flexural rigidity.
When exceeding the maximum strength the rigidity of the wall will decrease. The parts of the
wall where the flexural rigidity diminishes can be determined iteratively and processed accord-
ing to the method sketched in figure 4.45. It is, however, advised to use programs, specially
designed for diaphragm walls [Genesys; 1980]. In those programs it is possible to enter the M-
k-diagram of the concrete section of the diaphragm wall.
Piles, dolphins and Berlin-walls
In some cases the retaining structure consists of a combination of I-profiles, where, to a certain
depth, wooden planks are placed in between the flanges of the I-profiles. This retaining struc-
ture is called a Berliner wall. It also occurs that on regular intervals a sheet pile wall section
is installed to a greater depth (in dutch: gestaffelde wand). In these cases the soil resistance
will be primarily be caused by the soil underneath the shallower sheet pile wall sections, which
implies that the piles or longer sheet pile wall sections, must provide the stability.
The plane strain deformation situation, like that of a regular sheet pile wall, does not apply in
this case. The three-dimensional effect is taken into account by means of a so-called shell
factor. This shell factor indicates the proportion of the passive earth pressure factor of the pile
to the passive pressure factor of the wall. Weissenbach and Brinch Hansen published methods
which can be used to determine the earth pressure factors in case of such a shell factor. [Weis-
senbach; 1975] and [Brinch Hansen; 1960].
Also isolated horizontally loaded piles can be calculated by sheet pile wall programs, if the
shell factor is properly entered [Everts; 1984]. If the piles are situated close to each other, then
the shell shaped passive areas of the individual piles may overlap. In that case a plane strain
situation will be more representative.
4.3 Calculating using the finite element method
4.3.1 Introduction
This manual, when applying the finite element method, assumes a computer program which is
capable of calculating stresses and deformations in both a soil mass and the construction ele-
ments in that mass, such as a sheet pile wall, anchor bars and anchor walls.
Such models have been developed for the last twenty years.
Equilibrium of stresses and deformation of the soil and bending behavior of construction ele-
ments is described by a linked system consisting of partial and ordinary differential equations.
This system can usually only be solved numerically and a general method for this is the finite
element method. The finite element method can be applied for solving two and three dimen-
sional mechanical problems.
Project number
415660-21 v04
Date
December 2007
Page
192 of 278
Other applications of the finite element method as a mathematical numerical method are:
b. solving equations for groundwater flow;
c. solving heat transportation problems.
The design of sheet pile walls by means of a beam on elasto-plastic supports is also based on
the finite element method. It has become almost a custom, in geotechnics, to identify a finite
element method with a computer model for numeric soil mechanics.
The following subjects will be discussed in this paragraph:
c. general description of the method;
d. computer programs;
e. when to use them;
f. aspects when modeling sheet pile wall constructions by means of the finite element method.
4.3.2 General description of the method
The description is concentrated on a sheet pile wall construction. It concerns a sheet pile wall
test of which the cross-section has been displayed in figure 4.46. The problem is assumed as
two dimensional.
Equilibrium and deformation of the soil and the construction elements in the soil, is described
by the following mathematical equations:
in the soil:
partial differential equation for equilibrium of forces and the relations between stresses and
strains, and strains and deformations;
in the sheet pile wall and the anchor:
the equation for the bending behavior of a flexural rigid beam;
in the anchor bar:
a relation between axial tensile force and axial deformation;
on the transition planes between the soil and the sheet pile wall, and the soil and the anchor:
equations which describe the slide behavior; transfer of normal forces occurs in the direc-
tion perpendicular to these planes;
external forces:
e.g. horizontal forces on the wall, surcharge loads on ground level and the self weight of the
soil, which acts uniformly on the soil mass.
Due to the entered load history, where equilibrium is always possible, the system of equations
leads to only one possible displacement field. This implies that for each point of the soil mass
and the construction elements it is determined which displacement is necessary to lead to an
internal stress envelope between the soil and the construction elements, which results in equi-
librium with the external forces. This displacement field can be calculated numerically by
methods such as the finite element method.
In the calculation method, the soil mass is divided into small sections, e.g. triangles or rectan-
gles, the so-called finite elements. These are then marked by nodes, the so-called nodes of
the elements. These are usually the corners of the elements, but that does not necessarily have
to be the case. The internal stress situation in an element is determined as a function of the
displacement of the nodes of the element in respect to each other. This is based on a chosen
Project number
415660-21 v04
Date
December 2007
Page
193 of 278
interpolation for the displacement in the area of an element and the relation between the
stresses
Figure 4.46 Cross-section of the sheet pile wall construction. The bending stiffness of the left sheet
pile wall is ten times the stiffness of the right wall.
and the strains of the material of which the element is composed. Adjacent elements have joint
nodes. These nodes provide the equilibrium of forces between the different elements.
The stiff sheet pile wall is divided into linear elements, the so-called beam-elements. These
too are bordered by nodes, which, if necessary, are connected with nodes of the soil mass by
means of interface elements. In a beam-element the bending equation applies. This equation
is elaborated in such a manner that the bending and the axial deformation of the element is
expressed as a function of the displacement of the nodes of the element. The sliding behavior
and the transformation of forces between the beam-elements and the soil elements are de-
scribed by the interface elements. These are elements with nodes that coincide on one side with
nodes of the beam-elements, and on the other side with nodes of the soil elements.
1 m 4 m
4.00 m ~5.00 m
sheet pile wall
(Arbed PU 8)
as retainment
(fully fixed)
1st prediction after
excavation up to 4 m
spindle
strut
HEB 240
HEB 140
test wall
(Krupp KD VI)
W = 242 cm
3
framework
water basin
loading 10 kN/m
2
+1.00 m
-1.25 m
-4.00 m
-5.00 m
-5.50 m
-6.00 m
0.00 m
-8.00 m
Project number
415660-21 v04
Date
December 2007
Page
194 of 278
These elements combined lead to the element mesh, shown in figure 4.47. The border of the
mesh at the bottom and sides are placed at such a distance that the influence of the borders on
the behavior of the construction can be neglected. By connecting the equations for all the soil
elements, beam-elements and interface elements, a system of algebraic equations originates, of
which the displacement of the nodes is unknown. Solving this system results in the displace-
ment field, from which the stress envelope in the soil elements, moments and shear forces in
the sheet pile wall and the tension in the anchor bar can be derived.
Figure 4.47 Element mesh for the construction displayed in figure 4.46, with supports at the borders
and detail of the sheet pile wall, interface elements and soil elements.
4.3.3 Software for the finite element method
Many programs for solving finite element problems are available nowadays. The most popular
Dutch programs, specified for geotechnical application, at the publication time of this manual,
are:
PLAXIS, by Delft University of Technology and Rijkswaterstaat, which is meant for general
use and supported by Plaxis BV;
PLUTO, by Rijkswaterstaat and GeoDelft, which is primarily used for research purposes;
FEGRO, owned by FUGRO;
mesh with prescribed displacements
Project number
415660-21 v04
Date
December 2007
Page
195 of 278
there are some programs that are not specifically for geotechnical purposes but which can
be used to solve geotechnical problems, e.g. DIANA by TNO. This is also meant for general
use.
Most of these programs have versions available for both DOS personal computers and UNIX
workstations or mainframe computers. Working with a personal computer imposes limits in the
schematization, for example in the number of elements.
Paragraph 4.3.5 discusses the finite element method calculation for sheet pile wall problems.
4.3.4 Application of the finite element method
The finite element method primarily calculates displacement of the soil; not only at the loca-
tion of the sheet pile wall or the anchor, but also of the soil in front of or behind the wall. So
the finite element method is necessary if there are objects in the direct vicinity of the wall,
which are sensitive to soil deformations that can occur due to the execution near the sheet pile
wall construction. Sensitive objects might be shallow or piled foundations of buildings or
cranes.
The calculation provides an indication of the horizontal and vertical displacements of the soil
at the location of the foundation, with which can be checked if they are admissible for the
building. Demands for this are included in NEN 6740; 1991 [NNI; 1991]. Soil displacements
can cause horizontal pressures on piles and pipes and might cause serious damage.
When compared to other calculation methods for dimensioning a sheet pile wall (for example
Blums method or the beam on elastic bed method) the finite element method implicitly in-
cludes the mutual shear stress transformation between the soil layers. This is important in the
case of, for example, an embankment, which has a layer of peat underneath. This embankment
will not be able to provide much support if the peat layer is displaced horizontally and takes the
embankment with it. Blums method cannot directly include this interaction in its calculation.
Of course indirectly, by means of adequate assumptions, this effect can be taken into account.
Also arching on the active side of an anchored sheet pile wall is a result of internal transforma-
tion of shear stresses in the soil mass.
For all sheet pile wall designs the total stability must be checked. Usually this is done by means
of a slip circle analysis or Kranzs method, according to paragraph 4.4.
The finite element method has as an advantage that the checking of the mechanism of total
stability loss is expressed implicitly in the method. The analysis finds for the deciding shape of
the system and displays it directly in the graphical representation of the displacement field.
In the following cases the finite element method is the most logical method to use:
if the realization of the building pit might have any effects on other constructions in the
vicinity;
if important or expensive sheet pile walls are concerned, of which it is important to be able
to accurately estimate realistic processes like arching and interaction. This leads to a higher
quality of the design and, if the processes are to its advantage, optimization;
if situations are concerned, where it is a priori not evident if the schematizations and
simplifications, necessary for the classic calculation method, are sufficiently conform
reality.
In any case, for a useful application it is necessary to have a good insight in the soil type and
characteristics of the soil in the area of the sheet pile wall.
Project number
415660-21 v04
Date
December 2007
Page
196 of 278
In cases where three dimensional effects have to be taken into account to come to an accurate
and reliable solution, the finite element method offers good perspectives. With this method
reality can be simulated accurately.
However, the finite element programs mentioned before offer limited options in this aspect,
and programs that do offer these options are generally hard to come by.
The development of these programs, however, is relatively fast and it can be expected that
three dimensional problems can be solved efficiently in the near future.
4.3.5 Aspects when modeling with the finite element method
Material model and soil parameters
The reliability of the results of the finite element method depends on the entered constitutive
equations for the relations between the stresses and deformations. These are far more compli-
cated for the soil elements then for the construction elements. For the construction elements
elastic models are used, which are implied in the bending equation for the sheet pile wall and
the force-strain relation for the anchor bars. For soil elements elastic relations are completely
inadequate.
In the last twenty years lots of research has been done concerning development of elasto-
plastic material models for soil. This resulted in a significant number of models, varying in
level of detail, complexity, number of necessary parameters, numerical robustness, and overall
practical usability. This is not surprising, considering the fact that soil is a very complex mate-
rial which, depending on composition, density and load conditions, shows very different behav-
ior.
The elastically-ideal plastic model, which is relatively simple, proves in practice to be rather
useful. It has been accepted by many. In this model the relation between stress and strain is
assumed linearly elastic, as long as the shear stress remains smaller then the shear strength of
the material. If the shear stress reaches the shear strength of the material, plastic flow will oc-
cur. When using this model, the designer must be aware of unrealistic solutions. When using
the rather stiff initial modulus of elasticity, the occurring strains will be bigger then the calcu-
lated stresses. This is caused by the fact that a part of the plastic strain, in case of failure, is
neglected, and can cause the effect of arching to be over estimated. This is why in practice
often E
50
and G
50
(see figure 3.40) are used, which causes this effect to decrease. Only in areas
where the strains are small, generally between 0 % and 0.5 %, the rigidity is under estimated.
This might cause the magnitude of the development of the tip fixation of the sheet pile wall to
be under estimated.
The continuing development of calculation programs, especially the inclusion of the effect of
hardening plasticity, will remove this complaint.
The shear strength is determined by the classic soil parameters. For undrained analysis this is
the undrained shear strength f
undr
and for drained analysis this is the effective cohesion c' and
the angle of internal friction |. The shear strength in drained situations is determined by the
Mohr-Coulomb criterion:
t
max
= c' cos |' + p sin |' (4.72)
where:
t
max
is the maximum shear resistance;
c' is the effective cohesion;
Project number
415660-21 v04
Date
December 2007
Page
197 of 278
|' is the effective angle of internal friction;
p is the average effective stress according to figure 4.48.
In case of an undrained situation, where total stresses are used in the analysis, then, for nor-
mally consolidated soil, the undrained shear strength can be determined according to the above
mentioned relation by using for p the stresses, acting at the moment just before the load
changes happen. In addition to these parameters two more parameters are necessary for the
characterization of the elastic deformation behavior, namely, the modulus of shear G or the
modulus of elasticity E in combination with Poisson's ratio v.
Figure 4.48 Explanation of the Mohr-Coulomb failure criterion.
Estimation of the elastic parameters is based on the results of in situ soil tests and laboratory
testing of undisturbed samples, such as the pressiometer test and the triaxial compression test.
The modulus of elasticity can also be derived from an empirical relation with the cone resistance.
The value of the shear modulus G depends on the stress level. Therefore it is necessary to per-
form the triaxial compression test under the same stress conditions as in the field.
For the elastic parameters a distinction must also be made for drained and undrained behavior.
In the undrained analysis the material must have a fixed volume. In the analysis of total stresses
this can be achieved by assuming the Poisson's ratio to be somewhat smaller than 0.5; for ex-
ample v = 0.49. The shear modulus is not influenced by the type (drained or undrained) of re-
action of the soil. If the modulus of elasticity has to be entered in the program, instead of the
shear modulus, then a fictive E-value will be used for an analysis of total stresses, which is
chosen so that the relating shear modulus is correct. The equation that can be used for this is:
3
2(1 )
fict
E E
v
=
+
(4.73)
where:
E and v are the parameters for drained behavior, which are determined by the
grain skeleton.

t
t
|'
c'
c' cot |'
o'
t
max
= (c' cot |' + p) sin |' = c' cos |' + p sin |
Project number
415660-21 v04
Date
December 2007
Page
198 of 278
Some programs are capable to execute undrained analyses. In the calculation they take effec-
tive stresses and water stresses explicitly into account. In these programs the soil parameters
for the drained situation must be entered.
In the calculations the choice of parameter values demands specific knowledge related to the
choice of the so-called constitutive model. This model must describe the soil behavior as
precisely as possible, concerning the effects that may determine the design.
Calculating the drained and undrained situation
Soil is a two-phased medium, consisting of grains and water. The shear strength is delivered by
the grains and depends on the effective isotropic stresses between the grains. The change of the
total stresses in the soil due to in- or decreasing loads, is primarily absorbed by water stresses.
Depending on the permeability of the soil, a situation establishes fast or slowly, where the
change in stresses is absorbed by the grains. In case of clay this consolidation process is slow;
in sand it is almost instantaneously. In the first case this is called undrained behavior; in the
second case it is called drained behavior. It also depends on the speed of changing loads.
When analyzing the process of the execution of a sheet pile wall construction, undrained be-
havior of the soil must often be taken into consideration.
Primarily, there are the two following options to calculate undrained situations:
an analysis where effective soil stresses and water pressures, are taken into account
explicitly;
an analysis where only the total soil stress is taken into account.
Not al finite element method programs offer the possibility to calculate soil and water sepa-
rately.
In the first case the calculations can be made using the soil parameters for drained behavior.
In the second case calculations must be made with the undrained shear strength and the elastic
parameters of the soil for undrained behavior.
Apart from that, undrained calculations do not necessarily implicate a safe estimate. With ex-
cavations, which are usually the case with sheet pile walls, the undrained shear strength will be
bigger then the drained shear strength, which, in cohesive soil, adjusts itself only after some
time.
For each case it must be determined which type of analysis is to be used to obtain a reliable
result.
Generally a situation with clay layers in the soil, or sand layers enclosed by clay layers, in case
of an increase in stresses, must be analysed with undrained behavior. In case of a decrease in
stresses it is allowed to use undrained behavior in case of a temporary situation, after which (in
the following stage) the strength of the sheet pile wall construction is increased by, for exam-
ple, applying struts or anchors.
The long term situation will always have to be checked based on drained behavior.
In the user phase, when e.g. non-permanent or mobile surcharge loads must be taken into ac-
count, undrained behavior might be determining.
The considerations concerning drained or undrained behavior also apply when dimensioning
the sheet pile wall construction by means of other types of calculation models.
An option to take into account automatically the drained, undrained and all the situations in
between is offered by a program which takes consolidation into the calculation implicitly.
When calculating construction stages, where also the time path must be specified, the occurring
water pressure is calculated at each time step. For this analysis the permeability of all the soil
Project number
415660-21 v04
Date
December 2007
Page
199 of 278
layers, in addition to the already mentioned soil parameters, must be specified and entered into
the program.
Due to their time dependant character, these types of calculations demand large amounts of
computer time, and it is therefore obvious that they are only executed when the time effect due
to consolidation is of great importance for the design, the effects in the vicinity, or in case of
dangerous execution situations.
Construction stages
Calculating consecutive building stages, is a condition for an accurate analysis of the effects
during the execution and use of the sheet pile wall construction. The last building stage may
concern an analysis of the long term effects. When selecting a finite element method program,
a check has to be made if the program has options for analysing consecutive stages and options
for adding or excavating soil. Changes of strength properties and deformation parameters, in-
cluding the transition between drained and undrained calculations, must also be simulated by
the program if relevant. If the possibility exists to change the elastic parameters in consecutive
stages, this option can be used to imitate the difference in behavior between relieving and re-
loading. Programs where this can be taken into account automatically, as a part of the stress-
strain module, are currently being developed.
Mesh generation and accuracy of the finite element method
The finite element method starts, as any other geotechnical calculation, with the definition and
schematization of the problem. Here the following is determined:
the geometry, by means of partitioning the soil mass in layers, defining dimensions of the
construction parts and the location of the boundaries of the problem area;
the schematized partition in construction stages;
the soil parameters;
the loads.
Next up is the mesh-generation. This implies the partition of the problem area in elements and
applying the proper characteristics to those elements.
When choosing the finite element method program it must be checked that the program has to
its disposal the following elements, which are necessary for sheet pile analysis:
soil elements, where the proper stress-strain relations are implemented for the modeling of
the behavior of the soil;
interface elements, which can accurately simulate slip between the sheet pile wall parts and
the adjacent soil;
beam-elements, for modeling flexural rigid and axial stiff sheet elements;
line elements with axial stiffness for modeling anchor bars or struts.
Most programs have a mesh generator to establish the element network. In this network it is
generally possible to indicate the schematized soil layers and their characteristics, the locations
of the construction elements, interface elements and boundaries of the problem areas. In addi-
tion the fineness of the partition of the soil layers in elements can be indicated.
The user-friendliness of the mesh generator differs per program.
Project number
415660-21 v04
Date
December 2007
Page
200 of 278
When generating the element network the following aspects, which determine the accuracy of
the results, play a role:
the fineness of the element network;
the height-width proportion of the soil elements;
the chosen boundaries of the problem area.
It is usually the case that the finer the element network, the more accurate the results.
Increasing the fineness implies a longer calculation time. The optimal situation is to determine
a level of fineness so that the calculation results are obtained with a level of accuracy adjusted
to the problem. This does not only depend on the dimensions of the elements, but also on e.g.
the gradients of the stresses, thus on the calculation result itself. It is therefore advised to apply
finer elements at locations where great gradients are expected. It is impossible to prescribe a
general recipe for the choice of fineness of the element network.
By means of calculations with increasingly finer element mesh the gain in accuracy can be
evaluated. In case of a sheet pile wall, for example, the calculated moment line is an indicator
to check if the dimensions of the beam-elements are chosen sufficiently small.
A local increase in fineness of soil elements can, for example, be necessary if big differences in
rigidity are present between two adjacent soil layers.
When generating soil elements, very small or large height-width proportions must be avoided,
because they have a negative influence on the numerical procedure. A general indication for
the proportion is somewhere in between 6:1 and 1:6.
The boundaries of the element network must be chosen at such a distance that the influence of
the sheet pile wall construction at the location of the boundaries is negligible. A general indica-
tion can be that the vertical boundaries of the element network must be located at the passive
side at a horizontal distance of three times the penetration depth of the sheet pile. At the active
side this distance is the horizontal anchor length plus about one or two times the distance be-
tween the surface area and tip of the anchor.
The boundary conditions are:
no horizontal displacement;
free vertical displacement.
The horizontal bottom boundary of the network will often coincide with the top of a firm layer,
which is not effected by displacements, but must be at a depth of at least 5 meters under the
bottom of the sheet pile wall.
The boundary conditions here are:
no vertical displacement;
depending on the situation a free or not-free horizontal displacement. For example, in case
of high water pressures the horizontal displacement remains free.
Research of the sensitivity of the choice of boundaries is almost always necessary.
In addition, an accurate calculation of a defined problem with the finite element method does
not necessarily mean an accurate solution of the practical problem. This, of course, depends on
how realistic the schematization of the soil layers was, and on the quality of the determination
of the sheet pile wall parameters and soil characteristics, the loads, water pressures and other
circumstances. If these factors contain large uncertainties, then a sensitivity analysis has to be
executed.
Project number
415660-21 v04
Date
December 2007
Page
201 of 278
Presenting the results of the finite element method
Practically speaking it is only possible to judge the results of the finite element method if, in
addition to the numerical output, there is an option to obtain a graphical representation of the
calculated displacement field, the stresses and the plastic zones. Figures 4.49 through 4.54
show an example of a graphical output.
PC programs are usually equipped with such options. Mainframe computers or workstations
often use separate graphical software, which are called post-processors.
Figure 4.49 Deformed mesh, after excavation.
Figure 4.50 Vector rendering of the displacement in the last iteration.
deformed mesh, scaled up
extreme displacement: 1.31E-02 units
incremental displacement field
extreme displacement increment: 2.77E-03 units
Project number
415660-21 v04
Date
December 2007
Page
202 of 278
Figure 4.51 Contour lines of the displacement field.
Figure 4.52 Soil stresses (main stresses and directions).
contours of total displacements
minimum value: 0.00E+02 units; maximum value: 1.31E-02 units
total principal stresses
extreme stress: -1.72E+02 units
Project number
415660-21 v04
Date
December 2007
Page
203 of 278
Figure 4.53 Plastic points in the soil mass.
Figure 4.54 Moments in right sheet pile wall.
moments in beam chain 2
extreme moment: 7.19E+00 units
plane strain
Project number
415660-21 v04
Date
December 2007
Page
204 of 278
Safety aspects
The finite element method primarily determines displacements of the soil and sheet pile wall,
and the forces in the sheet pile wall and anchors. The displacements must meet demands con-
cerning the serviceability limit state and possibly the ultimate limit state. Also, the determined
anchor forces may not exceed the strength of the anchors.
This check must be executed with the design values of the parameters that have been entered in
the finite element method, according to paragraph 2.4.
If only the ultimate limit state is checked, then this check must indicate that not a single point
of the sheet pile wall will undergo an infinitely large displacement. Such a check has almost no
practical value. It is therefore advised to determine, in advance, demands concerning horizontal
displacement of the sheet pile wall and vertical displacement of the surface level behind the
sheet pile wall. These demands can be linked to deformation demands, which apply for some
constructions behind the wall, related to the ultimate limit state and the serviceability limit
state of those constructions.
Design values for parameters, entered in the finite element method, are determined by dividing
the representative values of those parameters by their partial factors. A construction is consid-
ered sufficiently safe if with these design values no displacements are calculated that exceed
the determined demands.
The partial factors in paragraph 2.4 are derived for the ultimate limit state which is determined
by failure of the soil wedge on the passive side of the sheet pile wall, as determined by Blums
method.
When checking the design with a finite element method, this concept is taken as a directive.
It is recommended to start with the execution of a calculation with the most unfavorable design
values, which are recommended in paragraph 2.4 for respectively the failure of the passive
zone and the total stability loss. If this does not lead to problems, then the demanded safeties
for the two mechanisms (failure of the passive wedge and total stability loss) are in any case
met.
If this does lead to problems, then the following can be done:
in case the passive zone in front of the wall fails, when design values are used which match
the values of the total stability loss mechanism, according to NEN 6740; 1991, then a new
finite element method calculation is executed using the design values of paragraph 2.4. The
results of this calculation determine if the design has to be revised;
in case the total stability is insufficiently secure when using the design values which match
the values of the failure of the passive wedge mechanism, the same rules apply.
In other cases after the first analysis it is apparent that the design must be adjusted during the
different construction stages.
When applying the finite element method it is a fact that there is insufficient practical experi-
ence available to judge the safety. The before mentioned recommendations must be considered
as such. Older methods have the advantage that the applicability has been proven. They have a
sufficient margin of safety when considering the ultimate limit state, even when applying the
design values of the loads and soil characteristics recommended in paragraph 2.4. A reliable
check of the serviceability limit state, which is determined by the occurring deformations, is
almost impossible when using the older methods.
Considering this it is therefore obvious to use both older and newer dimensioning methods in
cases where the risks, which come with the execution and usage of big and heavy sheet pile
wall constructions, are relatively big.
Project number
415660-21 v04
Date
December 2007
Page
205 of 278
This method supplies the designers with a good insight in the expected behavior of the sheet pile
wall construction in the different stages of execution and usage, and also in the possible weak
links in the construction. They can then possibly anticipate these problems and interfere with
effective constructive measures.
4.4 Stability check
4.4.1 Introduction
Anchored sheet pile wall constructions can, under specific circumstances, suffer a number of
different sorts of failure, where the total construction literally collapses. When checking the
design for the ultimate limit state, this total loss of stability must also be analysed.
By exceeding the shear stress along a deep slide plane between the sheet pile wall and the an-
chor, it is possible that the entire mass of soil, including wall and anchor, displaces, and the
sheet pile wall collapses forward, as shown in figure 4.55.
Figure 4.55 Shearing along a deep failure surface.
Figure 4.56 Shearing along a circular failure surface.
deep failure surface
circular failure surfa-
ce
R
Project number
415660-21 v04
Date
December 2007
Page
206 of 278
Another possibility, e.g. in case of a slope, is the occurrence of a circular failure surface that
continues underneath the sheet pile wall and the anchor wall, if the shear resistance of the soil
is not sufficient. This situation is sketched in figure 4.56.
Also, research must be done for an anchored or strutted sheet pile wall to check if a collapse of
the bottom of the pit, due to upward pressure of groundwater flow or piping, can cause a total
loss of stability of the wall. These types of failure are usually not determining for unanchored
walls, because the ratio of the retaining height and the penetration depth is too small.
Strutted walls with a relatively small penetration depth, which are used e.g. when placing pipes
in weak soil, there is a possibility that around the foot of the wall a sort of instability occurs,
where the soil mass near the bottom is pushed upwards due to the difference in vertical effec-
tive pressure at the foot. This sort of failure mechanism is somewhat similar to Prandtls
wedge, and is usually called heave.
In this paragraph the following types of total stability loss are discussed:
shearing along a straight slide plane between sheet pile and anchor;
shearing along a deep circular slide plane;
piping and local collapse of the bottom of the building pit due to flowing groundwater;
heave on the passive side of the wall.
4.4.2 Failure due to shearing along a deep straight slide plane (Kranzs method)
This calculation is to determine the required distance between the anchor and the sheet pile
wall. If the distance is to short a slide plane will occur from approximately the point of zero
shear forces in the wall to the tip of the anchor wall or the center of the grout element of a
grout anchor or tension pile. Model research proves that the shape of the slide plane follows a
logarithmic spiral. Kranzs method [Kranz; 1953], however, assumes a straight slide plane. By
means of comparing calculations, Kranz proved that this assumption is acceptable. In case of
free supported sheet pile walls and anchor walls, the slide planes occur from the tip of the sheet
pile walls to the tip of the anchor walls or anchor struts.
For fixed sheet pile walls or anchor walls the deepest point of zero shear forces is used.
To guarantee sufficient stability it must be proven that the design value for the anchor force,
obtained from the sheet pile wall calculation, can be supported by the body of soil in between
the sheet pile wall and the anchor strut, taking into account the representative values for the
shear resistance characteristics. In paragraph 2.4 an exception is made for Kranzs method, by
means of design values, and the stability factor is assumed to be 1.5. With this in mind a bal-
ance check is performed for the body of soil, which is limited by:
ground level;
vertical planes through the sheet pile wall and the anchor wall;
the assumed straight slide plane, from the tip or the point of zero shear forces of the sheet
pile wall and the tip of the anchor wall.
This body of soil is the area ABCD, shown in figure 4.57. In this case it is a free supported
sheet pile wall and anchor wall and a homogeneous, non-cohesive soil layer.
Project number
415660-21 v04
Date
December 2007
Page
207 of 278
On this body of soil the following forces act, which must finally be in equilibrium:
G is the mass of the body ABCD;
E
a
is the force on the shearing active soil wedge AB, during the failure phase. This
force can be derived from the active soil pressure on the sheet pile wall E
a
and
the weight G
A
of the body of soil between the wall and slide plane AB;
E
1
is the soil pressure on the vertical through the anchor wall. The influence of
the surcharge load behind the anchor strut must be included in this soil pres-
sure;
F
A
is the anchor force on the soil mass due to the anchor wall;
Q
1
is the force in shear plane BC, which, in the failure phase, has an angle |' with
the normal in this plane.
For the determination of these forces, representative average values of the soil parameters must
be used. Paragraph 2.4 discusses further directives for the safety analysis.
If no equilibrium is reached and ABCD shears along BC, then the direction of Q
1
is determined
and the force diagram can be sketched.
If cohesion is present it acts parallel to the slide plane.
For the chosen anchor length the representative value of the resistance of the anchor can now
be determined analytically or graphically. This force must satisfy 2.4.
Figure 4.57 shows the graphical determination of the strength of the anchor for homogeneous
non-cohesive soil by means of a force diagram.
Figure 4.57 Calculation scheme for Kranzs method.
A
A
D
C
A
G
G
1
G
A
A
B
Q
1
Q
1
E
1
E
a
G
1
+ G
A
= G
|
E
a
E
1
strength of anchor
F
A;opn
, sup
;max
1.5
cr rep
a
F A
F A
= >
Project number
415660-21 v04
Date
December 2007
Page
208 of 278
Figure 4.58 shows how to implement the influence of cohesion. In case of an undrained situa-
tion, for example in peat or clay layers, equilibrium of the soil mass can be determined in a
similar way. For these layers only the undrained shear strength in the slide plane, f
undr
, must be
taken into account. The angle of internal friction |' equals zero in this situation.
Clearly, the anchor force must also be determined for this undrained situation.
It must be taken into consideration that in case of a horizontal groundwater level, this ground-
water only influences the weight of the soil mass. There is no resulting horizontal groundwater
pressure. Water pressure, due to a difference in water levels along the sheet pile wall, is taken
into account in the analysis of the total stability by means of the in the sheet pile wall calcula-
tion determined anchor force, which is partially determined by the water level.
Figure 4.58 Kranzs method for cohesive soil.
In case of a non horizontal groundwater level behind the sheet pile wall, the influence of the
difference in water level must be included in the analysis of the stability of the soil mass, be-
tween the wall and the anchor screen.
In this case, in addition to the forces described above, a resulting horizontal force is also pre-
sent due to the difference in water pressure. This force, of which the magnitude is shown in
figure 4.59, must also be included in the analysis of the equilibrium.
The total stability of a layered soil mass can be checked in a similar way. The soil mass is par-
titioned into different layers that are limited by the intersection of the soil layers and the slide
plane, as shown in figure 4.60.
If the stability check shows that the demands for the ultimate limit state are not met, then either
the penetration depth of the sheet pile wall or the distance between the wall and the anchor
must be increased.
A
D
C G
B
Q
1
Q
1
E
1
E
A
|
E
A
E
1
G
strength of anchor
force F
Asup
c
F
A
F
A
E
A
= only as a result of
effective stresses
'
c
F c BC =
Project number
415660-21 v04
Date
December 2007
Page
209 of 278
Figure 4.59 Influence of difference in water pressure over the anchor length.
Figure 4.60 Kranzs method for layered soil mass.
If grout anchors or tension piles are applied then the total stability is checked by means of an
analysis of the straight slide plane between the point of zero shear forces of the wall and the
center of the anchor body. With grout anchors, the slide plane will intersect the grout anchor at
a distance from the head of the anchor that equals the sum of the free anchor length and half
the length of the grout body, as shown in figure 4.61; for tension piles the method shown in
figure 4.62 must be used.
In the analysis of equilibrium the force E
a
, caused by the effective pressure on the wall, should
equal the summed effective pressure on the wall, with which the load on the anchor is deter-
mined.
excess pore pressure to be
taken into account
Q
1
Q
1
E
1
E
A
E
A
E
1
strength of anchor
F
Asup
F
A
F
A
G
1
G
2
Q
2
Q
2
G
1
G
2
layer 1
layer 2
Project number
415660-21 v04
Date
December 2007
Page
210 of 278
Figure 4.61 Stability analysis in case of grout anchors.
Figure 4.62 Stability analysis in case of tension piles.
For pre-stressed anchors the applied pre-stress can exceed the anchor force, determined by the
sheet pile wall analysis. If this is the case, then only the calculated anchor force will be taken
into account when determining the total stability. The excessive force due to the pre-stressing
will be neglected.
Instead of using the graphical approximation shown in figure 4.57, the total stability can also
be checked by means of an analytical calculation.
4.4.3 Loss of total stability due to shearing along a circular slide plane
In addition to Kranzs method for stability consideration described above, it is necessary, in
specific cases, e.g. relatively small penetration depths and short anchors, to check the stability
along a circular failure plane, which is located underneath the pile wall and the anchor wall.
There are some calculation methods available to determine the stability of slopes. These meth-
ods are also usable to determine the stability of a body of soil, where the difference in height of
the terrain is obtained by a sheet pile wall. These calculations start with a balance against rota-
A
B
C
L
eff
L
free
D
grout body made
under pressure
active
B
D A
L
R
L
A
C
L
K
L > 5 m
D = 0
L
A
> L
K
+L
W
Project number
415660-21 v04
Date
December 2007
Page
211 of 278
tion of a mass of soil along a circular slide plane, as shown in figure 4.56. The design value of
the driving moment M
a
must be equal or be smaller then the design value of the stabilizing
moment M
r
. For the determination of these moments the assumed slide circle, the mass of soil
which is assumed to rotate, is divided into vertical layers. For each of these layers an equilib-
rium equation is considered.
The sum of these layer equilibrium equations results in M
a
and M
r
. The difference between the
calculation methods is primarily determined by the level of simplification of the equilibrium
conditions for the horizontal direction, where the forces in between the vertical layers are taken
into account.
Figure 4.63 Calculation method for circular slide planes according to Bishops method.
Note:
The test that is discussed here differs from the slide plane calculation according to Bishop in
the CUR-manual Construeren met grond [CUR; 1992] where the calculations are made with
a stability factor instead of design values.
Bishop assumes that the horizontal forces E
1
and E
2
are equal and that their direction coincide.
The vertical layer forces T
1
and T
2
are neglected.
The driving moment in the center of the slide circle O per layer equals:
AM
a
= (G + Q) R sin o (4.74)
The total driving moment then equals:
M
a
= R Al E(q + h) sin o (4.75)
The maximum shear stress t is calculated from the conditions of vertical equilibrium between
the sum of Q and G and the sum of the vertical components of T, N and U.
0
R
q
o
Q
G
E
1
T
1
E
2
T
2
T
o
N
u
cos
l
o
A
Project number
415660-21 v04
Date
December 2007
Page
212 of 278
This leads to the maximum resisting moment per lamel:
r
M TR A =
(4.76)
where:
cos
l
T t
o
A
=
(4.77)
'
( ) tan
(1 tan tan )
c q h u
t
o
+ + u
=
+ u
(4.78)
And therefore:
cos
r
l
M R t
o
A
A =
(4.79)
From which follows:
'
( ) tan
cos (1 tan tan )
r
c q h u
M R l

o o
+ + u
A = A
+ u
(4.80)
where (with displaying the normal dimensions):
A is the layer width in [m] (width of all layers is equal);
q is the surcharge load in [kPa];
is the unit weight of the sliding soil in [kN/m
3
];
h is the layer height in [m];
c' is the effective cohesion (drained situation) in [kPa];
|' is the effective angle of internal friction in [];
u is the water pressure in [kPa].
Now it can be checked if M
r
_ M
a
.
This calculation is executed for a number of slide circles, which are chosen just so that from
them the most critical circle; the circle with the lowest ratio F = M
r
/ M
a
, can be determined.
For the safety check the neglects and simplifications are usually acceptable.
More accurate calculations, where the mutual layer forces are taken into account, can be found
in literature.
If it is concluded that there is a circular slide surface present, where M
r
_ M
a
, then the demand
regarding the ultimate limit state has not been met, and that the design has to be altered.
The most critical slide circle will generally pass a the tip of the sheet pile wall. The center
point will often lie anywhere along the vertical extension of the wall.
The available computer programs automatically check a number of center points and radii and
supply, for every center point, the lowest value of the previously mentioned ratio and, for every
circle, its radius. From this the most critical slide circle can be determined.
The flexural rigidity and strength of the wall is usually so high that no slide circle will occur
through the sheet pile wall. This is not the case for an anchored wall with anchor bars that have
a relatively low flexural rigidity and a small cross section, so the surrounding soil can move
Project number
415660-21 v04
Date
December 2007
Page
213 of 278
through these bars with relative ease. Anchor piles will generally have such a rigidity and
strength that they will prevent the development of a slide circle through the piles. The anchor
piles will resist rotating with the surrounding soil mass. The critical slide circle will then be
located outside the area of the anchor piles. However, in weak soils, situations can occur where
the critical slide circle passes the anchor piles. See figure 4.64a and b for more information.
If the anchor body is located outside the slide circle, the influence of the anchor force may be
taken into account in the manner shown in figure 4.65. The driving moment is then decreased
by aF
a
where for F
a
the unfavorable (low) design value must be used.
A low point of application of the anchor under the top of the sheet pile wall has a favorable
influence.
If the demands regarding the ultimate limit state are not met, the penetration depth of the sheet
pile wall has to be increased.
Figure 4.64 Slide plane for an anchor bar (a) and anchor pile (b).
Project number
415660-21 v04
Date
December 2007
Page
214 of 278
1 2 3
4 5
Figure 4.65 Influence of the anchor force on the circular shearing.
4.4.4 Piping
General
If the difference in water level between both sides of a sheet pile wall is relatively large and the
soil consists of fine granular, permeable soil and there is free water on both sides of the sheet
pile wall, then on the lower side water will flow upwards. This causes a decrease in effective
stress.
If the flow velocity reaches a certain critical level, then on the passive side a situation can oc-
cur where fine soil particles (grains) are transported by the water flow. This can cause erosion
in specific cases, followed by originating wells and a process of erosion starts. This can cause
the formation of channels, as is shown in figure 4.66.
Figure 4.66 Development of piping.
A
R
a
M
Project number
415660-21 v04
Date
December 2007
Page
215 of 278
These channels can have very irregular shapes. The locations of these channels depend on the
coincidental presence of weak spots in the soil. In this case there exists a great danger that the
phenomenon will expand, due to the development of an open connection with the free water
behind the wall.
This phenomenon is called piping, and can occur if the exit drop exceeds a critical value.
An analytical approximation of piping is possible by considering the vertical equilibrium of a
slice of soil against the front side of the wall.
This is the so called macro-stability approximation according to Terzaghi [Terzaghi; 1954]
and the determination of the exit velocity and critical drop at the location of the soil at the pas-
sive side of the wall.
Critical drop according to Terzaghi
The critical drop i
kc
is, according to Terzaghi, determined by considering the equilibrium of a
soil slice under influence of the groundwater pressure applied to that slice. Based on model
tests Terzaghi concluded that piping is manifested in a zone in front of the sheet pile wall of
approximately half the length of the penetration depth of the wall. This distance also deter-
mines the width of the soil body of which the equilibrium is considered. The height of the zone
is determined by a random horizontal plane at a distance D bottom level. The horizontal pres-
sures applied to the body ABCD can be neglected in the stage of loss of equilibrium. De up-
ward flow pressure on the downside CD of the soil mass can, for example, be determined by
mens of a so called flow net, formed by flow lines and equipotential lines.
The principle of the method is described in figure 4.67.
1
( )
sat w w
D m H > A (4.81)
where:
sat
is the saturated unit weight of the soil;
m is a factor (_ 1), with can be derived from the flow net by means of the theory
of potential flow.
The critical value for the difference in hydraulic head AH
crit
for the horizontal reference level
D
1
is:
1
( )
sat w
krit
w
D
H
m

=
(4.82)
By executing the calculation shown above for horizontal planes on different levels, the deter-
mining value for H
crit
can be obtained. Generally the location of this plane located at the bot-
tom of the wall is determining.
De margin with regard to this form of piping is determined by the factor F
piping
:
krit
piping
H
F
H
A
=
A
(4.83)
Project number
415660-21 v04
Date
December 2007
Page
216 of 278
When checking the ultimate limit state, this factor should at least equal 1 if design values for
the different parameters are used. More directions are given in paragraph 2.4.
The critical difference in pressure, relation to piping, which is determined by the method
shown above, is confirmed by model tests, executed with clean sand.
Figure 4.67 Calculating piping according to Terzaghi.
Determining critical exit hydraulic gradient from effective pressure and seepage forces
The critical exit hydraulic gradient is obtained by comparing the groundwater pressure on a
soil element at the bottom of the building pit, to the weight of that soil element. A critical situa-
tion occurs when the groundwater pressure on the soil element and the vertical effective pres-
sure of the weight of the soil element are almost equal. The critical gradient can be calculated
from figure 4.68 as follows:
Ah
crit

w
= Az
sat
Az
w
(4.84)
or
crit sat w
krit
w
h
i
z

A
= =
A
(4.85)
AH
D
D
D
1
G
B
D
A
C
U
m Ah
w
course of excess groundwater
pressure at C D with respect to
groundwater level in front of the
wall
Project number
415660-21 v04
Date
December 2007
Page
217 of 278
For sand this leads to:
20 10
1
10
krit
i

= =
(4.86)
Figure 4.68 Determining the critical gradient.
Checking piping by means of Lanes method
With Lanes method the required length of the flow line can be determined, which is necessary
to prevent piping [Lane; 1935].
The available length of the flow line is determined by the sum of the length of vertical flow
lines and a third of the length of flow lines, which are at an angle of less then 45 with the hori-
zontal.
Sheet pile walls only have the vertical flow lines L
1
and L
2
shown in figure 4.69. The flow line
factor C
L
for different soil types is given in table 4.3.
Figure 4.69 Determining the length of the flow line for sheet pile walls.
o
g
Ah
hydraulic head at depth Az
Az
Az .
w
Au = Ah .
w
Au
o
g
= Az .
sat
u
z
L
2
AH
L
1
Project number
415660-21 v04
Date
December 2007
Page
218 of 278
Table 4.3 Flow line factor C
L
.
soil sort recommended factor for C
L
very fine sand or silt 8.5
fine sand 7.0
middle course sand 6.0
course sand 5.0
fine gravel 4.0
middle course gravel 3.5
course gravel 3.0
rocks 2.5
A condition is that:
1 2 L
L L C H + > A (4.87)
in which:
AH is the difference in hydraulic head over the sheet pile wall.
This method has not been tested specifically for piping for sheet pile walls, so it must be used
with significant reservation.
In any case, piping does not occur if there is no free water surface present at the high side of
the wall. In addition, the groundwater level at the lower side must at least equal the terrain
level.
4.4.5 Heave
General
Sheet pile walls with a limited penetration depth can suffer a form of instability where the sur-
face at the lower side of the wall is pushed upwards, due to the influence of the earth pressure
at the backside of the wall. Here the equilibrium is interrupted. This form of instability occurs
at multiple anchored or strutted walls and cellar walls, where the depth of the wall below
ground level at the lower side of the wall is relatively small or even zero. In addition, the shear
strength of the soil has to be small. This phenomenon will therefore occur in soft clay or peat
layers.
The problem can be approximated by considering the vertical equilibrium of the mass ABCD in
figure 4.70, which is limited by the tip of the sheet pile wall and the vertical planes formed by
sheet pile wall AB and a vertical plane CD located at a distance B.
The equilibrium of the mass is maintained by the vertical friction forces in CD and the bearing
capacity underneath BD. The maximum value of this bearing capacity can be determined with
the calculation method for shallow foundations according to NEN 6744; 1991 [NNI; 1991].
For non cohesive soil (c = 0) and cohesive soil (| = 0) the equilibrium will be analysed after-
wards. The assumption is made that the tip of the sheet pile wall equals the terrain level at the
lower side of the wall.
Project number
415660-21 v04
Date
December 2007
Page
219 of 278
Figure 4.70 Heave in non cohesive soil.
Non-cohesive soil
The situation shown in figure 4.70 applies. The load on the horizontal plane BD is determined
by the weight of the body ABCD, a possible surcharge load on AC, and the upwards friction
forces in DC. It is clear that there are no vertical friction forces in AB, because in the failure
situation the sheet pile wall, with struts, collapses with body ABCD.
The friction force approximately equals the resulting force of the active earth pressure multi-
plied by the tangent of the angle of internal friction |. The vertical load on BD equals, if there
is no top load:
F
v
= G W (4.88)
This load is opposed by the maximum support force of the plane at the bottom of BD. This
equals the support force of a strip foundation with a width B and a depth of z = 0 in case of a
vertical load. The horizontal load is supported by the struts or anchors. The maximum support
force is determination according to NEN 6744; 1991 [NNI; 1991].
It appears that heave in non-cohesive soil cannot occur without the presence of groundwater
flow.
Because of the assumption that the penetration depth equals the terrain level at the lower side
of the wall, there can be no difference in potentials present over the sheet pile wall. If there is a
difference in groundwater level, then a drainage system with a good filter composition is nec-
essary at the foot of the wall.
Cohesive soil
The friction force in DC of figure 4.71 is determined by the length of DC and the undrained
shear strength f
undr
. The vertical load F
v
on BD hereby, without top load, equals:
F
v
= G W (4.89)
F
v
= BH Hf
undr
(4.90)
B
A C
G
W
F
v
H
N
B D
strut or
anchor
strut or
anchor
N = H
2
K
0
o
v
H
W= N tan |
Project number
415660-21 v04
Date
December 2007
Page
220 of 278
Figure 4.71 Heave in cohesive soil.
The maximum support force of the soil for a vertical force underneath BD, equals, according to
NEN 6744; 1991:
Q
v
= 5.14 Bf
undr
(4.91)
The condition is now, when using the design values of the loads and soil characteristics:
F
v
_ Q
v
(4.92)
so:
BH Hf
undr
_ 5.14 Bf
undr
(4.93)
From which follows for the situation where a balance is just reached:
undr
undr
5.14
Hf
B
H f
=
(4.94)
This leads to a realistic solution if:
5.14 f
undr
< H (4.95)
or:
undr
5.14
H
f

<
(4.96)
B
A C
G
W
F
v
H
B D
strut or
anchor
strut or
anchor
45
o
45
o
Project number
415660-21 v04
Date
December 2007
Page
221 of 278
If, for example, H = 5 m and = 13 kN/m
3
then:
f
undr
< 12.6 kPa (4.97)
This means that a ground breach can only occur in soft clay or peat.
If a qualitatively good, hard soil layer is present at a relatively short distance underneath the
foot of the wall, then the width of the body ABCD is determined by the distance between the
foot of the wall and this layer, as is shown in figure 4.72.
In that case a ground breach can hardly occur anymore, so it does not have to be taken into
consideration.
In cases where the wall is penetrated further then the level at the lower side of the wall, similar
calculations can be executed, where the loaded surface, according to which the equilibrium is
considered, goes through the foot of the sheet pile wall.
The influence of the groundwater level and differences in groundwater levels must be taken
into account. Groundwater flow to the bottom of an excavation has a negative influence.
Figure 4.72 Influence of a shallow stiff layer heave.
D
H
soft clay
D
Project number
415660-21 v04
Date
December 2007
Page
222 of 278
4.5 Designing the anchor construction
4.5.1 Introduction
This chapter discusses the calculation methods for sheet pile wall anchor construction. Here a
distinction is made between:
anchor struts and walls;
screw anchorss;
screw injection anchors;
grout anchors;
anchor piles.
The calculation includes the determination of the support capacity of an anchor element, the
mutual influences of the anchors, verifying equilibrium between anchors and anchor walls for
which indications have been made in paragraph 4.4, the necessary strength of the anchor bar or
strand and the calculation of the wales.
Good determination methods by calculations do not exist for every anchor type. This goes for
screw anchors, screw injection anchors and grout anchors. For these types an approximation of
the support capacity is made by means of an empirical relation between the extraction force and
e.g. the measured c.p.t.-values or based on experience. This approximation is usually verified by
means of applicability tests and verification experiments.
4.5.2 Applicability tests and verification experiments
General
The purpose of the applicability tests is to determine the maximum support force and creep
behavior of screw anchors, screw injection anchors and grout anchors. The verification ex-
periments can be seen as a form of final check, to verify if the anchors meet the meet the re-
quired demands. These experiments are usually performed on all anchors.
The procedures for the applicability tests and verification experiments are bound to pre-
determined rules.
Determining the type and dimensions of the anchor depends in many cases on experience ob-
tained in that specific area. If this experience is missing, then applicability tests must be per-
formed during the design phase to determine the extraction force and the deformation behavior
of the soil. If there is experience obtained from previous implementations of the same anchor
type in that area, performing an applicability test at the beginning of the construction phase is
sufficient. If necessary, alterations to the design can then still be made.
During the execution, in most cases, a verification experiment is performed on all screw an-
chors, screw injection anchors and grout anchors. When applying MV piles or other piles, veri-
fication experiments are usually performed on a limited number of piles.
Project number
415660-21 v04
Date
December 2007
Page
223 of 278
time [min]
a
n
c
h
o
r

f
o
r
c
e

[
%
F
a
;
m
a
x
;
g
r
;
d
]
120
100
80
60
40
20
t
2
= 30 min
t
3
= 60 min
t
2
t
3
t
3
eventually continued
destructive
Execution of the tests
Applicability test
A design value for the load is assumed, F
a;max;gr;d
.
During the test the anchors are, in a number of steps, burdened to at least 1.2 F
a;max;gr;d
.
The load scheme is displayed in figure 4.73.
At the beginning an initial force F
i
is applied of approximately 10 % of F
a;max;gr;d
with a maxi-
mum of 50 kN. This eliminates the unwanted backlash in the measuring device.
Figure 4.73 Procedure of the applicability test.
The loads are increased in steps of 0.2 F
a;max;gr;d
and then kept constant for 30 minutes. After
every load step the load is decreased to F
i
. From 0.8 F
a;max;gr;d
the load is kept constant for 60
minutes.The increase and decrease of the load must be performed with a maximum of 20 kN
per second.The measuring device for the applicability test is shown in figure 4.74.
Figure 4.74 Measuring device for the applicability test.
1
2
3
4 5
6
7
8
9
10
1
2
3
4
5
6
7
8
9
10
Project number
415660-21 v04
Date
December 2007
Page
224 of 278
time [min]
a
n
c
h
o
r

f
o
r
c
e

[
%
F
a
;
m
a
x
;
g
r
;
d
]100
80
60
40
20
0
t
pr stress
at desired force
The displacement of the anchor is measured with respect to a measuring frame which is com-
pletely independent of the sheet pile wall.
By means of a theodolite the following observations are made:
the horizontal displacement of the top of the sheet pile wall;
the horizontal displacement of the measuring frame;
a fixed referential point on the surface, which lies outside the influence of the test.
The theodolite measurements are meant to correct possible influence of the measuring frame
on the forces applied to the anchor and sheet pile wall.
The theodolite measurements are performed several times per load step.
The displacement of the top of the wall with respect to the measuring frame is measured by
means of a theodolite of an electric displacement recorder with 0.01 mm. accuracy. These
measurements, possibly corrected for the displacement of the measuring frame, are used for
interpreting the test.
The readings occur at timestep 0, 1, 2, 3, 4, 5, 10, 15, 20, 25 and 30 minutes after the beginning
of each load step. After the 0.8 F
a;max;gr;d
load step readings also occur at 45 and 60 minutes.
After decreasing the load to F
i
the displacement is also measured. The test can also be contin-
ued to the moment the anchor is extracted from the soil. In case of pre-stressed grout anchors it
must be taken into consideration that the load may not exceed 90 % of the collapse strength of
the pre stress elements.
The applied force is measured by means of a pressure box with an accuracy of 0.25 kN.
To prevent local deformations of the sheet pile wall, it must be locally stiffened by means of a
wale.
Verification experiment
During the verification experiment the load is increased in steps of 0.2 F
a;max;gr;d
from the initial
force F
i
of 50 kN, to F
a;max;gr;d
. The load speed has a maximum of 20 kN per minute. After
reaching the maximum load it is kept constant for 5 minutes and then decreased to F
i
. The load
scheme is shown in figure 4.75.
Figure 4.75 Load scheme verification experiment.
Project number
415660-21 v04
Date
December 2007
Page
225 of 278
The measurements can be limited the top of the anchor and it displacement.
Directly after reaching F
a;max;gr;d
and every minute after that the displacement must be meas-
ured. The reading interval is extended if the anchor behavior does not meet certain conditions.
4.5.3 Elaboration of the experiments
The representative value for the maximum tensile force, or the strength of the anchor, has to be
determined by the interpretation of the applicability test.
The representative value of the maximum tensile force is determined by:
the collapse of the anchor rod of anchor strand F
br;st;rep
;
reaching the maximum resistance of the anchor element in the soil F
a;max
;
reaching the demands for creep F
k
.
The force F
a;max
is the force with which a continuing displacement of the anchor occurs due to
the collapse of the connection between the grout and the surrounding soil. If this force is not
reached in the applicability test, it has to be approximated by extrapolation of the load-
displacement diagram.
The representative value of this force F
a;max;rep
must satisfy:
a;max;rep a;max;gem
a;max;rep a;max;min
3
3
F F for N
F F for N
= >
= <
(4.98)
where depends on the following factors:
if the tests are performed on all anchors = 1;
if the tests are preformed on a limited number of anchors, depends on the number of
applicability test N, which is assumed representative for the considered area and of the
amount of anchors M that is assumed to cooperate as a result of the transmission of forces
due to the wales.
and where:
F
a;max;gem
is the average F
a;max
from N tests;
F
a;max;min
is the lowest F
a;max
from N tests.
The value for can be obtained from table 4.4. This table is copied form NEN 6743; 1991
[NNI; 1991].
Table 4.4 Values for .
N
M
1 2 3 4 5
1 0.75 0.78 0.79 0.81 0.82
2 0.78 0.81 0.83 0.84 0.86
3 0.81 0.84 0.86 0.87 0.89
Project number
415660-21 v04
Date
December 2007
Page
226 of 278
In addition, the tensile force of the anchor is determined by the characteristics of the tensile bar
or strand. According to paragraph 2.4, the following should apply:
br;st;d
a;max;st;d
1.4
F
F s
(4.99)
where:
F
br;st;d
is the design value of the collapse force of the bar or strand is reached.
The creep behavior of the anchor is expressed in the degree of creep k. This is defined as:
2 1
2 1
mm
log log
u u
k
t t
(4.100)
where:
u
1
is the displacement in mm at timestep t
1
in minutes;
u
2
is the displacement in mm at timestep t
2
in minutes.
If the settlement behavior in time is plotted on half logarithmic scale, then the degree of creep
can be obtained from the graph directly. This is shown in figure 4.76. When elaborating the
applicability test the degree of creep is calculated form the course between timesteps t
1
= 15
minutes and t
2
= 30 minutes, respectively 60 minutes.
The demand with respect to the degree of creep is determined by the performance demands
required of the sheet pile wall construction. It is normal for permanent anchors to use a maxi-
mum value of k
max
_ 2.0 mm.
Figure 4.76 Graphical determination of degree of creep k.
time [min]
d
i
s
p
l
a
c
e
m
e
n
t

[
m
m
]
0
1
2
3
4
5
6
7
8
log t
1
log t
2
1 10
1
10
2
10
3
u
1
u
2
u
2
u
1
1.2 F
a;max;gr;d
1.0
0.8
0.4
0.6
log (t
2
/t
1
)
k = 0.4 mm
1.6 mm
0.9 mm
0.6 mm
0.6 mm
Project number
415660-21 v04
Date
December 2007
Page
227 of 278
The accessory tensile force is F
k
. The representative value of this force is determined by:
k;rep k;gem k;rep k;min
or F F F F = =
(4.101)
For the same applies as stated above with respect to the maximum resistance of the anchor
element in the soil F
a;max
.
For the verification experiment the degree of creep may reach a maximum of 1.0 mm. in case
of a load of 100 % of F
a;max;sr;d
. If this cannot be proved within a time interval of 5 minutes,
under constant load, the time interval must be increased to 15 minutes.
If the degree of creep is between 1 and 2 mm, then the result is compared to the results of the
applicability tests. Based on this comparison the anchor is approved or disapproved. If the de-
gree of creep exceeds 2 mm the anchor is definitely disapproved and a new anchor has to be
constructed, or the design of the sheet pile wall has to be altered.
4.5.4 Free anchor length
The part of the anchor bar or strand that in the schematization can deform freely, independent
from the surrounding soil, is considered as the free anchor length. The underside of the free
anchor length is the fictional anchor point. So this is the point of application of the resulting
force of the shear stresses between the grout body and the soil. The location of the anchor point
determines if the shear stress transmission occurs at the desired location.
The extension Al of the anchor bar is determined in the applicability test by the measured dis-
placements of the top of the anchor between the loads of 1.2 F
a;max;gr;d
and F
i
. The theoretical
free anchor length L
vt
is then derived from:
vt
a;max;gr;d i
1.2
lEA
L
F F
A
=
(4.102)
The explanation of L
v
and L
a
are shown in figure 4.77.
For anchors where the tensile force is transmitted from the top of the anchor body, as shown in
figure 4.77a, for the so called mono anchor, the following must be met:
0.8
2
a
v vt v
L
L L L < < +
(4.103)
The fictional anchor point can be located somewhat in front of the anchor body, because part of
the force is already transmitted in the free anchor part.
For anchors where the force transmission occurs at the foot of the anchor body, e.g. the duplex
grout anchor, shown in figure 4.77b, the following must be met:
( )
2
a
v vt v a
L
L L L L + < < +
(4.104)
where:
L
v
is the length of the free anchor part to the point of application of the force;
L
a
is the length of the anchor body;
L
vt
is the theoretical free length.
Project number
415660-21 v04
Date
December 2007
Page
228 of 278
If the fictional anchor point is located behind the foot of the anchor body it indicates that the
anchor as a whole has moved in the soil body. This, of course, is unacceptable.
Figure 4.77 Fictional anchor points for mono- and duplex-anchors.
4.5.5 Anchor struts and walls
The calculation of the anchor strut is generally based on Coulombs theory for straight slide
planes. The active slide plane behind the sheet pile wall and both the active and passive slide
planes in front of and behind the anchor strut are shown in figure 4.78. Generally, the distance
between the anchor and the wall are chosen in such a manner that the active slide plane of the
sheet pile wall, according to Blums sheet pile wall calculation, and the passive slide plane of
the anchor strut do not intersect.
The anchor strut normally does not extend to the surface. In the earth pressure and resistance
calculation it can be assumed that the anchor strut extends to the surface if the following condi-
tion is met:
1
2
1.5
h
h
s
(4.105)
where:
h
1
is the height of the strut;
h
2
is the distance between the surface and the underside of the strut.
a. mono-anchor
L
v
+ L
a
/ 2
b. duplex-anchor
0.8 L
v
L
v
L
a
F
F
head of
spanvijzel
Project number
415660-21 v04
Date
December 2007
Page
229 of 278
Figure 4.78 Slip lines.
Anchor constructions are often composed of separated independent struts.
Based on Buchholzs research [Buchholz; 1930] a directive is developed for square anchor
struts in sand, for the maximum distance a
max
and the influence of the depth of the anchor on
the passive soil resistance F
ea;h;p
of a row of anchor struts. The wall may be calculated as con-
tinuing if the following condition is met:
max
max 1
a a
a h |
s
=
(4.106)
In addition for separate anchor plates with h
1
= b the following applies:
F
ea;h;p
= q 0.5 b h
2
2
(4.107)
For the continuing wall:
F
ea;h;p
= 0.5 h
2
2
(4.108)
The values for |, q and are shown in figure 4.79 as a function of h
2
/h
1
. Extrapolation of h
2
/h
1
above the displayed maximum value of 5.5 is not allowed.
This graph generally applies for sand with an average effective angle if internal friction of
|' = 32.5. The factor actually equals the earth pressure factor K
;p
. For other |'-values the
value for q can be derived quite accurately. In case of rectangular struts with a height h
1
the
above stated equation can be used in the following adapted form:
a
max
_ h
1
(| + o 1) (4.109)
where:
o is the proportion of the width of the strut to its height h
1
.
h
1
h
2
45
o
|/2
45
o
+ |/2
u + x
acc. to
Blum
Project number
415660-21 v04
Date
December 2007
Page
230 of 278
Figure 4.79 |, and q-values.
Figure 4.80 Graphical determination of F
a;max;d
.
h
2
h
1
b
b
a
0 1 2 3 4 5 6
h
2
/h
1
12
10
8
6
4
2
0
q

|
|, q,
p
E
2
Q
2
G
2 G
1
Q
1
F
a;d
E
1
Q
1
G
1
G
2
P
Q
2
E
2
E
1
F
a;d
F
u
= 216 kN/m
| = 30
o
o
a
= 20
o
c = 0
= 18 kN/m
3
p = 10 kN/m
2
o
|
|
45
o
|/2
45
o
+ |/2
Project number
415660-21 v04
Date
December 2007
Page
231 of 278
The design value for the maximum support force F
a;max;d
can be determined graphically in the
manner shown in figure 4.80.
The forces on the passive wedge in front of the anchor and the active wedge at the back are in
equilibrium for an anchor force the size of the closure forc of the polygon of forces.
Possible top load behind the sheet pile wall has a desirable effect on the anchor force and can
be neglected on the passive side of the anchor strut. The following equation can also be used:
a;max;d 1 2
1 sin ' 1 sin ' 1 sin '
0.5( 1) ( 1)
1 sin ' 1 sin ' 1 sin '
d d d
d d y z
d d d
F h h p h h
| | |
| o o |
| | |
| | +
= + +
|
+ +
\ .
(4.110)
where:
| is Buchholzs factor shown in figure 4.79;
o is de proportion of the width and the height of the anchor strut;
h
1
is the height of the strut;
h
2
is the distance from the surface to the underside of the strut;
p
d
is the design value of the top load at the backside of the anchor strut;
d
is the design value for the volumetric weight of the soil, which can be obtained
in table 1 of NEN 6740;1991 [NNI;1991], where, in this case, the situation
must be considered as unfavorable;
|
d
is the design value of the angle of internal friction of the soil, to be determined
according to NEN 6740; 1991.
4.5.6 Screw anchors
A screw anchors consists of a steel anchor bar with a diameter of approximately 30 mm, which,
at the end, has a welded steel screwblade with a diameter of approximately 300 to 500 mm.
The anchors are installed by means of an adjustable vibrator drill engine.
Support capacity
In case of failure of the anchor a failure pattern forms in the soil surrounding the screw blade
of which the shape depends on the depth of the blade. For a shallow screw blade the proportion
of the depth to the blade diameter is: H/D _ 5 and the slide lines of the failure pattern reach the
surface. For a deep anchor blade is H/D > 5 and a failure pattern as shown in figure 4.81 oc-
curs.
For the calculation of the support capacity of this type of anchor no theoretical model exists.
From tests performed on deep screw anchors in sand [Hergarden; 1983], the following empiri-
cal relation between cone resistance q
c
and the collapse force F
a;max;rep
was derived:
a ; m a x
FAq =
(4.111)
where:
F
a;max
is the maximum support force;
A is the surface area of the screw blade;
q
c
is the average cone resistance in the influence area.
Project number
415660-21 v04
Date
December 2007
Page
232 of 278
Figure 4.81 Failure pattern around a deep anchor blade.
The design value F
a;max;d
is then determined from the representative value F
a;max;rep
according
to:
a;max;rep
a;max;d
m
F
F
=
(4.112)
where F
a;max;rep
is obtained in a manner similar to the one shown in paragraph 4.5.3:
F
a;max;rep
= F
a;max;gem
or
F
a;max;rep
= F
a;max;min
;
m
is the partial material factor which, according to NEN 6740; 1991, equals 1.4.
In this case N equals the number of c.p.t.'s at the construction site.
For cohesive soil with a design value for the average undrained shear strength f
undr;d
F
a;max;d
becomes:
F
a;max
= 10 f
undr;d
A (4.113)
From the failure pattern shown in figure 4.81 it appears that the influence area is limited to a
distance of 2D from the centre of the anchor. For determination of the support force a zone of
3D above and below the centre is assumed, with respect to possible deflections in the depth of
the anchor, which occur during the drilling.
Both the exercised rotating moment and the vibrationlevel of the vibrator have an influence on
the extraction force of the anchor. The density of the sand increases due to the vibration. To
obtain a good effect the adjustments of the vibrator must meet:
amplitude > 1.0 mm;
frequency > 15 Hz.
When the distance between two anchors is smaller then 8D it is advised to reduce the calcu-
lated maximum strain force. The reduction factor can be obtained through linear interpolation
between 1.0 and 0.5 for distances of respectively 8D or 1D.
H
D
F
a;d
Project number
415660-21 v04
Date
December 2007
Page
233 of 278
The occurring displacement of the head of the screw anchor is relatively small if the anchor
force F
a
is smaller then, or equal to 0.4 times the strain force F
a;max
.
This head displacement s can be estimated with the following equaltion:
;max
0.006
a
a
F
s H
F
=
(4.114)
where:
H is the depth of the hart of the screw blade in relation to the ground surface.
Applicability test and verification experiments
The strain force, determined by means of the empirical relation, must be checked according to
an applicability test if it has not been tested in the regarded area yet.
The test must be executed according to the description in paragraph 4.5.2, where, however, a
lower loading speed must be taken into account. This must be done to prevent the positive in-
fluence of the decrease of initial horizontal earth pressure behind the anchor. If the test is pre-
formed rapidly an extra reduction factor of 1.5 must be introduced in addition to the already
applied -factor, from table 4.4, for the determination of the representative value of the support
force from the test results.
Execution aspects
The previously mentioned test results [Hergarden; 1983] show that there is a relation between
the strain force and the turning moment M
i
. The turning moment is registered during installa-
tion with the application of the anchor by means of a vibrator drilling engine If the turning
moment does not meet the following relation, a verification experiment, according to paragraph
4.5.2, must be executed:
M
i
_ 2.5 F
a;max
(4.115)
where:
M
i
is the turning moment in [kNm];
F
a;max
is the necessary strain force in [kN].
4.5.7 Grout anchors
Support strength
A grout anchor consists of an in the soil constructed cylindrical body of cement grout, which
usually has a diameter of approximately 100 mm. The anchor force is transmitted from the
sheet pile wall to the grout body by means of a steel pre-stressed bar or strand.
The strain force is derived from the shear resistance between the grout body, which is formed
under high pressure, and the surrounding soil. The high pressure applied to the grout results in an
increase in shear resistance of the sand compared to the initial value.
Project number
415660-21 v04
Date
December 2007
Page
234 of 278
gravel, sandy
mean/coarse sand
fine sand
2,000
1,600
1,200
800
400
0
0 2 4 6 8 10
anchor length [m]
F
a
;
m
a
x
;
g
r

[
k
N
]
very dense
dense
dense
dense
very dense
mean dense
mean dense
loose
mean dense
The anchor force can be transmitted to the grout body in two different manners:
through the anchor bar, at the front of the grout body: mono anchor;
through the backside of the anchor body: duplex anchor.
Both mechanisms are sketched in figure 4.82.
The advantage of duplex anchors is that the force is applied to the anchor body as a compres-
sive force. This will prevent the formation of cracks in the grout body, and results in a better
protection of the anchor bar or strand against corrosion.
A theoretically based method to calculate grout anchors is not available so the model must be
largely based on field experience. The stretch force and creep force determined by means of
this experience must therefore, in case no experience with grout anchors is available in the
regarded area, be tested prior to the design phase by means of applicability tests.
Figure 4.82 Mono anchor and duplex anchor.
Figure 4.83 Relation between stretch force, grout characteristics and anchor length.
course of the
shear stresses
mono-anchor
duplex-anchor
F
a
F
a
Project number
415660-21 v04
Date
December 2007
Page
235 of 278
Furthermore, all anchors must subject to a verification experiment.
Field tests and model tests [Jelinek et al.; 1976] lead to the following qualitative conclusions:
the support strength increases with higher density of the sand;
with equal density the extraction force increases when the U-number, which is a measure
for the grain surface, increases;
the extraction force increases less then proportionately with the anchor length.
Approximately 5.0 meters must be used as a minimal length;
an increase of the diameter of the grout body has almost no influence on the extraction
force.
This is displayed in the form of a graph in figure 4.83, for anchor diameters of 100 150 mm.
and a depth of H _ 4m.
Execution aspects
Influence of the method of application
Grout anchors can be installed by means of driving or drilling. The installing method influ-
ences the extraction force of the anchor.
Generally driven anchors have a higher extraction force then drilled anchors. When driving the
anchor, the surrounding soil is pushed away and the pressure between the surrounding grains
increases. When drilling the cavity for the anchor is formed by means of removing the soil,
which will cause the pressure between the grains to decrease. When applying the grout anchor
under a pressure of 10 to 20 bars, the contact pressure increases significantly. When applying
the anchor under even higher pressures the formation of cracks in the soil becomes more risky.
Placing and distances in between
To develop the necessary extraction force the grout body must be fully enclosed by a sand
layer. As a rule of thumb it can be assumed that the head of the grout anchor must be located at
least 1 meter under the underside of a possibly present weak layer, and that a soil body of at
least 5 meters must be present above the head of the grout body. This is sketched in figure 4.84.
Figure 4.84 Demands regarding the depth of the grout anchor.
H _ 5.0 m
_ 1.0 m
L
a
_ 5.0 m
soft layer
sand
Project number
415660-21 v04
Date
December 2007
Page
236 of 278
The length of the grout body must be at least 5 meters. An acceptable approximation of the
extraction force is obtained by regarding the equilibrium of a slice of soil above and possibly in
between the anchors, as is shown in figure 4.85.
During the collapse of the soil slice a cone shaped rupture surface will originate. Its length
depends on the distance between the anchors t. According to van Weele [Van Weele; 1987] the
following equation applies:
4tan '
d
t
d
|
=
(4.116)
where:
|
d
is the design value of the angle of internal friction of the regarded soil layer.
From the extraction force F
a;max
calculated according to the above mentioned method, the rep-
resentative value F
a;max;rep
is determined by means of:
F
a;max;rep
_ F
a;max
(4.117)
Where _ 0.7 must be used.
The minimal distance in between anchors must be 1.5 meters to prevent mutual influence.
Figure 4.85 Determining the extraction force from the equilibrium of the soil mass on top of the
grout anchor.
soft layers
F
F
Q
Q
E
a
E
a
d
W
P
P + W
|
sand
Project number
415660-21 v04
Date
December 2007
Page
237 of 278
When the distance is small a row of anchors will function as a sheet pile. The design value for
the maximum shear stress in the plane of this sheet then depends on the effective pressure ac-
cording to:
t
max;d
= o'
o
tan |'
d
(4.118)
where:
o'
o
is the effective pressure on the anchor plane, which makes an angle of o with
the horizontal plane;
|'
d
is the design value of the angle of internal friction of the soil.
In addition to the above mentioned tests, which concern the maximum extraction force, the
equilibrium demands regarding the stability of the sheet pile wall including the anchors must
also be met. Indications for these tests are given in paragraph 4.4.
Prestressing of grout anchors
After the execution of the verification experiment the grout anchors are generally pre-stressed
to a force of approximately 0.75 F
a;max;gr;d
. This prevents significant displacements before the
necessary anchor force has been developed.
If only a minimal displacement of the wall is allowed the anchors are pre-stressed to 0.9
F
a;max;gr;d
.
F
a;max;gr;d
can be determined according to step 7 of paragraph 2.4.2.
4.5.8 Screw-injection anchors
Figure 4.86 Screw-injection anchors.
tube ca. 48.3 x 12.5
coupling
M48 x 100
grout body
ca. 5 m
anchor blade
nut
M48
Project number
415660-21 v04
Date
December 2007
Page
238 of 278
General
A screw-injection anchor as sketched in figure 4.86 consists of a spiral drill which is welded to
a thick walled steel tube. Anchors are applied where the extreme support strength ranges from
400 to 1,750 kN.
The demanded maximum tensile force of the anchor determines the length of the grout body
which lies in between 4 and 10 meters. The standard version has a spiral drill length of 1 meter.
During the screwing a grout mixture is pumped to the bottom of the tube under a pressure 5 to
10 bars and is intensely mixed with the sand where the anchor obtains its support strength. The
displacement of the anchor must be at least half of the pitch of the spiral. This equals a scrape
factor of 2.
Support capacity
The executed applicability test leads to a safe approximation of the maximum extraction force:
F
a;max
= 0.015 O L
a
q
c;av
(4.119)
where:
F
a;max
is the calculated value of the maximum extraction force in [MN];
L
a
is the length of the anchor part in the sand layer in [m];
O is the circumference of the spiral in [m];
q
c;av
is the average measured cone resistance in the sand layer over the anchor part
in [MPa].
It is very important to understand that this approximation may only be used if verification ex-
periments have been performed on all anchors of this type, and if the scrape factor is _ 2.
The determination of the design value of the calculated maximum tensile force is done accord-
ing to paragraph 4.5.6.
By using a thick walled tube of ordinary steel, this anchor behaves relatively rigid.
Applicability tests and verification experiments
Applicability tests for screw injection anchors must be executed if no prior tests have been
performed in similar soil characteristics, for this type of anchor.
All similar anchors must be checked by means of verification experiments.
Applicability tests and verification experiments must be executed according to paragraph 4.5.2.
If these tests lead to a design value for the maximum extraction force, according to 5.4.3,
which is lower then the design value of the maximum anchor strength obtained from the sheet
pile wall calculation, then extra anchors must be added.
4.5.9 Anchor piles
Different pile types can be used as anchor piles. The calculation of the support capacity for
tensile force of the usual pile types, like driven pre-fabricated concrete piles or steel piles can
be executed by using the shaft friction, determined as for compressed piles according to NEN
6743; 1991 [NNI; 1991].
The pile part within the active slide plane behind the sheet pile wall is excluded from the calcu-
lation of the support capacity of the anchor pile, as is shown in figure 4.87.
Project number
415660-21 v04
Date
December 2007
Page
239 of 278
Figure 4.87 Effective anchor pile length.
The calculated support capacity must be checked regarding pile interaction, according to para-
graph 4.5.10.
The equilibrium of the sheet pile wall including the anchor piles must be verified for total sta-
bility loss according to paragraph 4.4.
The determination of the design value of the maximum tensile force should go according to the
method for compressed piles given in NEN 6743; 1991. This method has been mentioned in
paragraph 4.5.3.
The tensile loaded MV-piles, which are often used for anchor piles in the construction of quay
walls, are discussed below.
MV-piles
The MV-pile is applied in cases where big anchor forces must be transmitted to the soil. The
pile consists of a steel beam section which is, during the driving, fit with a cement grout cover.
This cover is formed by welding an enlarged foot to the pile point as is shown in figure 4.88a
or by welding steel buckets to the outer side of the flanges, as is shown in figure 4.88b. During
the driving the cavity, formed behind the point or the steel buckets, is immediately filled with
grout cement by means of a pipe. This cement is dosed in suck a manner that the grout level
always equals the surface level.
When calculating the extraction force it must be taken as a starting point that the tensile force
is transmitted to the supporting soil only by the pile part. The calculation must be executed
based on the grout pressure as well as based on the cone resistance.
The extraction force F
a;max
is the lowest of the resulting forces of both the calculation methods.
effective anchor length
L
eff
45
o
|'/2
active
ahear band
rotation point
Project number
415660-21 v04
Date
December 2007
Page
240 of 278
effective stress
rotation point
L
eff
h
h
1
45
o
|/2
o
g
= h
g;d
o
w
= h
1

w
active
shear band
Figure 4.88 Two types of regularly used MV-piles;
a. full point;
b. with buckets welded to the flanges.
Calculation based on the grout pressure
Here is assumed that the maximum friction on the dividing plane between the grout edge and
the soil, determined by the hydrostatic pressure of the liquid grout column on that dividing
plane. This principle is sketched in figure 4.89.
Figure 4.89 Course of grout pressure and water pressure immediately after the making of the grout
anchor. Determination of the effective pressure on the dividing plane grout/soil.
a. b.
Project number
415660-21 v04
Date
December 2007
Page
241 of 278
At this depth this grout pressure equals:
; g g d
h o =
(4.120)
where:
h is the depth of the point of the MV-pile below the surface level;
g;d
is the design value of the volumetric weight of the cement grout, for which 22
kN/m
3
can be assumed.
Also hydrostatic water pressure is present at a depth h below the surface level:
1 w w
h o =
(4.121)
where:
h
1
is the depth of the pile point below the water level;
w
is the volumetric weight of water.
The difference in pressure between both liquids is regarded as an effective pressure between
grout and soil. This leads to a maximum shear stress for the depth h:
t
max
= (o
g
o
w
) tan |
d
(4.122)
where:
|
d
is the design value of the angle of internal friction of the sand.
a;max
is obtained by summing the maximum shear stresses along the di-
viding plane between the hardened cement grout and the sand outside the active zone over L
eff
of figure 4.89.
The design value for the extraction force F
a;max;d
is found from:
;max
;max;
a
a d
m
F
F
=
(4.123)
where:
m
is the partial material factor which can be assumed 1.4 in this case.
Calculation based on cone resistance
From many test loads on MV-piles in the Rotterdam harbor area it was concluded that in the
Pleistocene sand layer a maximum shear stress can be assumed of 1.4 % of the cone resistance,
with a maximum of 250 kN/m
2
[Brassinga; 1987].
This equals:
t
max
= 0.014 q
c
(4.124)
Project number
415660-21 v04
Date
December 2007
Page
242 of 278
Where, in addition, the following must be met:
t
max
_ 250 kPa (4.125)
MV-piles with a full point section, as shown in figure 4.88a also accomplish, by penetration of
the sharp point in the soil, an increase of the pressure in the surrounding soil. For this type the
shaft friction may be assumed to be 1.6 % of the cone resistance instead of 1.4 %. The maxi-
mum remains at 250 kN/m
2
though.
a;max
is obtained by summing the maximum shear stress along the divid-
ing plane between the hardened cement grout and the sand, over L
eff
(see figure 4.89).
For the determination of the design value of the maximum extraction force F
a;max;d
the same
method should be applied as is described in paragraph 4.5.6, for screw anchors.
Test loads
The extraction force of an MV-pile can be strongly influenced by the quality of its execution.
Therefore it is usually the case that some fully installed MV-piles are subjected to test loads.
Due to the related high costs the number of test loads is limited.
4.5.10 Group functioning of anchors and piles
Different aspects which go together with the mutual influence of the in-line anchor struts,
screw anchors, screw injection anchors, grout anchors and piles have already been discussed in
previous paragraphs.
If the anchors are assumed to be individual for the determination of the extraction force F
a;max
by means of the applicability tests or calculations, then the possibility of overlapping of the
individual influence zones should be taken into account for the determination of the representa-
tive value for the extraction force F
a;max;rep
.
The influence zones of screw injection anchors, grout anchors and MV-piles will overlap if the
distance between the anchors is smaller then 7D, where D is the effective diameter of the an-
chor body or the MV-pile. Indications for the determination of the effective extraction force in
case of closely situated anchors are given in paragraph 4.5.8.
Closely situated anchors or piles can be manipulated, by means of varying the driving angle of
the anchors or the piles, so that enough soil can be mobilized to obtain the desired extraction
force.
4.5.11 Variable loads
Variations of the anchor force in time are caused by e.g. tidal influences, changing top loads
and temperature effects.
However, in general the variation is relatively small when compared to the anchor force, and,
additional, usually the frequency is low.
In these cases the variation of the force has no negative effect on the extraction force of the
anchor and can therefore be neglected.
There is hardly anything known about the effect of big variations with high frequencies on the
extraction force and creep force of anchors for sheet pile wall constructions. In general it can
be assumed that negative influences on the extraction force can be neglected.
Project number
415660-21 v04
Date
December 2007
Page
243 of 278
4.5.12 Specific execution aspects
Research has shown that vibrations, that originate in the soil due to pile driving or trilling, have
no effect on the hardening and the quality of fresh concrete [Zandbergen; 1987]. This goes for
strength development as well as for the attachment of steel to concrete. From this point of view
pile driving or trilling of not yet hardened grout anchors or MV-piles in the immediate sur-
rounding is not a problem.
Pile driving or trilling at a relatively small distance of stressed anchors or not sufficiently hard-
ened grout anchors can cause a loss of soil pressure around the anchor. Distances of at least 50
meters from the anchors will cause no negative effects. Is the distance is smaller then that fur-
ther research is necessary.
4.5.13 Designing the wales
The wales can be seen as a continuous beam on a large number of supports, which is uniformly
loaded similar to figure 4.90.
In this figure:
F
a;d
is the design value of the anchor force according to step 9.4 from paragraph
2.4.2;
a is the distance between anchors;
q is the uniformly divided load that follows from the equilibrium with the anchor
forces: q = F
a;d
/ a.
Figure 4.90 Calculation scheme for the wale of a sheet pile wall.
In case of a calculation model where plastic hinges may occur, the design value of the moment
in the wales becomes:
; ;
1
16
gord d a d
M F a =
(4.126)
If the model, where, at the location of the maximum moment, the maximum elastic strain is just
reached in the furthest fiber of the cross section, is assumed then:
; ;
1
12
gord d a d
M F a =
(4.127)
F
a;d
F
a;d
F
a;d
F
a;d
F
a;d
a a a a a
q
Project number
415660-21 v04
Date
December 2007
Page
244 of 278
It is advised in case of steel wales to use the following design value for the moment [EAU;
1990]:
; ;
1
10
gord d a d
M F a =
(4.128)
In relation to the testing of the ultimate limit state the following should be met:
M
gord
_ M
r;d
(4.129)
where:
M
r;d
is the design value of the moment that the wale can support if the
maximum elastic strain is just reached in the furthest fiber of the cross section.
then:
M
r;d
= o
vl;d
W
d
(4.130)
where:
o
vl;d
is the design value of the flow border according to the regarded material stan-
dard;
W
d
is the design value for the resisting moment which can be equaled to the value
given by the producer.
4.6 Sheet pile wall constructions loaded in the plane
4.6.1 Introduction
For building pits, quay walls and anchor constructions the situation occurs where a sheet pile
wall construction is loaded along the direction of the wall [Vonk; 1984].
Examples are corner solutions for quay walls, head walls for construction pits and anchor con-
structions where the anchor force has a component in the direction of the wall. Another exam-
ple is a bolder force on the wall caused by an in the wall integrated bolder.
The load can usually be supported optimally by creating the wall as a sheave. In case of rela-
tively small loads connecting the planks by means of a wales is usually enough.
The external horizontal load component will be transmitted to the subsoil by means of the
shear stresses between the wall and the soil.
Project number
415660-21 v04
Date
December 2007
Page
245 of 278
4.6.2 Failure mechanisms
Three failure mechanisms can occur:
1. An external failure mechanism where the external horizontal load component exceeds the
maximum supported shear stresses between the wall and the soil.
These maximum supported shear stresses also depend on the load situation as a result of ex-
ternal load components in other directions.
2. An internal failure mechanism where the sheet pile wall construction collapses due to an
excessive bending tension, normal tension or shear stress in the construction.
It also goes internally that the tensions due to possible external load components in other di-
rections should be superimposed.
3. A failure mechanism where the sheet pile wall construction collapses due to the
deformation of planks as a result of the so called harmonica effect. This failure mechanism
is complicated and therefore hardly quantifiable.
4.7 Coffer dams
4.7.1 Introduction
Generally, a coffer dam is considered to be a soil or water retaining construction consisting of a
body of soil enclosed by two sheet pile walls, which can transmit horizontal and vertical loads
to the subsoil.
The sheet pile walls are connected by means of anchors on one level or more. The design, but
even more, the calculation of a coffer dam construction is complicated. This is evident from the
many different calculation methods developed over the years. The calculation of coffer dams is
therefore left to specialists.
It is also advised to use the finite element method when designing the sheet pile walls and an-
chors.
This manual, however, discusses some methods to design a kist dam.
4.7.2 Function, application options, construction and parts of coffer dams
Function and application of a coffer dam
As mentioned in the introduction, a coffer dam is considered to be a soil or water retaining
construction.
Coffer dams are applied for temporary and permanent situations; e.g. a quay waall, docking
wall, piers, a part of a sluice or a boundary of a construction pit if anchors cannot be used be-
hind the wall. Coffer dams are also applied as dike reinforcements where traditional reinforce-
ments are not an option due to the surroundings.
Project number
415660-21 v04
Date
December 2007
Page
246 of 278
Figure 4.91 Principle sketch and element names.
Construction and elements of a coffer dam
A coffer dam is a construction where a soil retaining sheet pile wall -the front wall- is con-
nected, by means of a system of anchors and wales, to a wall which functions as an anchor -the
back wall-. The walls however, are located in each others influence zone. The soil retaining
wall, mainly loaded with active pressure, is usually called the front wall. The anchor wall, on
which the soil body enclosed between the walls exercises passive pressure, is usually called the
back wall. The principle of a coffer dam is sketched in figure 4.91.
Usually a coffer dam consists of two -usually steel sectioned- parallel sheet pile wall screens,
which are connected on one or more levels by means of anchors.
A coffer dam can also be constructed from cells of flat steel sheet pile piles. It is then called a
cellular coffer dam.
The coffer dam consisting of cells has not been applied in the Netherlands so far.
Coffer dam or sheet pile wall construction?
To answer this question the force envelope in the ultimate limit state in the soil body enclosed
between the two walls must be considered more detailed. Behind the front wall an active zone
will develop as a result of the loads, and near the back wall a passive zone will develop. If the
distance B between the walls is smaller than approximately 1.5 H, the collapse zones will over-
lap. This prevents the passive resistance from fully developing and causes the construction to
be labeled as a coffer dam. If the passive resistance can fully develop then the back wall can be
seen as an anchor wall for the front wall and the construction becomes a regular sheet pile wall
construction.
Both situations are schematically displayed in figure 4.92.
B - 0.7 1.5 H
anchor
filling
wall behind wall in front
front wall back wall
driving depth - 0.5 1.0 H
H = retaining heigth
Project number
415660-21 v04
Date
December 2007
Page
247 of 278
B B
H
H
anchor anchor
passive
zone
passive
zone
active
zone active
zone
rotation point
rotation point
zones do intersect
= coffer dam
zones do not intersect
= two sheet pile walls
Figure 4.92 Coffer dam or anchored wall.
4.7.3 Loads and testing procedures
List of loads
Here follows a list of loads that play a role in coffer dam constructions and a general descrip-
tion is given of the differences between coffer dams and regular sheet pile wall constructions.
earth pressure (usually effective pressure is meant):
Compared to a regular sheet pile wall construction the active pressure on the front wall can
increase or decrease when the passive zone in front of the back wall and the active zone at
the front wall influence each other. A silo effect can also play a role as well. Dutch consul-
tancy practice does not take silo effects into account. In German consultancy practice the
active pressure on the front wall is sometimes increased by 25 %.
The passive pressure on the back wall will usually be smaller due to that influence then with
a regular anchor wall.
water pressure, phreatic level and groundwater flow:
This does not necessarily deviate from regular sheet pile wall constructions. Special atten-
tion does need to be paid to the groundwater flow behavior and the location of the phreatic
level between the two walls. These strongly depend on the transmissivity of the soil body
enclosed between the two walls and can cause higher water pressures then should normally
be expected. This has a negative effect on the yet to be developed horizontal resistance of
the filling.
top load (traffic load, storage material):
This does not deviate from regular sheet pile wall constructions.
formation of crevices at the back wall:
Due to the displacement of the back wall a crevice can originate in cohesive soil at the outer
side of the wall. If an initial earth pressure, which was lower then the hydrostatic water
pressure, was present against the back wall in the undrained situation, filling the crevice
will cause the value to increase to a hydrostatic water pressure belonging to a groundwater
level in a higher draining soil layer, or a free water level.
initial tension situation in the soil:
Project number
415660-21 v04
Date
December 2007
Page
248 of 278
This does not necessarily deviate from regular sheet pile wall constructions, but special at-
tention must be paid to the method of execution, because it, among others, determines the
initial tension situation.
the handling of the execution:
This does not necessarily deviate from regular sheet pile wall constructions, but the initial
tension situation deserves special attention.
the handling of the phasing:
This does not necessarily deviate from regular sheet pile wall constructions.
the handling of consolidation:
This does not necessarily deviate from regular sheet pile wall constructions either.
settlement of the soil:
These are a result of the so called silo effect. This silo effect influences the active and pas-
sive earth pressures. Silo effects do not occur in regular sheet pile wall constructions.
the handling of special loads:
Special loads can be e.g.: impact loads, top loads, wave loads, ice loads and others. The
handling does not deviate from regular sheet pile wall constructions.
Testing procedure
A test according to the demands regarding the ultimate limit state and the serviceability limit
state, for geotechnical constructions as well as other constructions, is stated in NEN 6740; 1991
[NNI; 1991].
The partial material factor recommended in this manual are determined by means of a probabil-
istic analysis of regular sheet pile walls, which have been designed according to the beam on
an elastic bed theory. Coffer dams have not been researched then and the usage of those fac-
tors for coffer dams has therefore not been judged.
The representative values for the soil parameters according to NEN 6740; 1991 [NNI; 1991]
are recommended.
4.7.4 Force envelope in coffer dams
Anchor force
The main function of the anchors in coffer dams is to ensure the cooperation between the front
and the back wall. Sometimes, in practice, more anchors are placed. This leads, together with
the soil fill, to an analogy with a frame work construction, which increases the rigidity of the
construction.
Function of the filling
The filling of a coffer dam construction delivers the highest contribution to the rigidity, if the
filling consists of sufficiently rigid material, preferably highly consolidated sand and/or gravel.
Wall friction
Wall friction (among others) can cause active and passive silo effects in coffer dams. Dutch
consultancy usually disregards this phenomenon. A calculation model using the finite element
method has this phenomenon implicitly in its method. The walls of coffer dams are relatively
weak, which weakens the increasing trend in the earth pressure due to the silo effect. Pre stress
effects of anchors usually promote silo effects.
The wall friction does (among others) determine the magnitude of the active and passive earth
pressure. Prescriptions for the magnitude of the wall friction are given in paragraph 3.4.
Project number
415660-21 v04
Date
December 2007
Page
249 of 278
4.7.5 Possible failure mechanisms
A coffer dam can collapse in many different manners, depending on the design and the load
situation. That is why all different failure mechanisms must be analyzed and checked. The
following failure mechanisms can occur (see also figure 4.93).
Figure 4.93 Failure mechanisms.
a. internal stability b1. shearing b2. rotation of the front wall
passive pressure passive pressure
yielding
concave
shear band
convex shear band
shear band
anchor
failure
c. turn over d. rotation e. reaching fully plastic moment
f. anchor failure g. total stability
Project number
415660-21 v04
Date
December 2007
Page
250 of 278
internal stability:
By exceeding the maximum supprted shear stresses in the soil enclosed in between the two
walls, this coffer dam is no longer rigid and the cross section will deform to a parallelogram
(a). This is similar to Terzaghis mechanism.
ground breach at the location of the front wall:
The passive force against the front wall is too small to support the present load. This will
cause the coffer dam to shear off (b1) or to overturn (b2).
overturning or rotation of the coffer dam:
Overturning occurs if the sum of the driving moments exceeds the sum of the
resisting moments in relation to a point under or in the coffer dam filling, along a
bent slide plane, which goes convexly or concavely through the undersides of the
walls (c and d).
reaching the fully plastic moment of the sheet pile wall:
This will cause a local flow in the wall. The function of the wall is then lost and the kist
dam collapses. This of course also depends on if the coffer dam construction is statically de-
termined or not (e).
anchor breach:
The anchor force exceeds the maximum supported force, which causes the anchor to break
and the cooperation between the two walls to be lost. The coffer dam will then surely col-
lapse (f)
Total stability of the coffer dam construction:
The coffer dam is considered to be a rigid body. Stability loss can occur due to loss of equi-
librium along a deep slide plane.
4.7.6 Rules of thumb dimensioning
General dimensioning of the soil body
The soil body, and then specifically the filling between both walls, mostly determines the rigid-
ity of the construction. It is therefore necessary to use only rigid, highly transmissive materials
for the filling.
Acceptable width-height proportions
If a coffer dam is not rigid, then the cross section can deform to a parallelogram due to the
load. To prevent this from happening, the distance between the two walls B may not be to small
compared to the retaining height H. A rule of thumb is:
1.5 H > B _ 0.7 H
If B > 1.5 H then the construction is no longer considered to be a coffer dam.
For a bigger width a lot of soil is necessary for filling, and more steel is necessary for the an-
chors. On the other hand, a lighter back wall can be used. A bigger width can therefore lead to
an economically more feasible solution.
Penetration depth of the front wall
If the front wall of a coffer dam is not penetrated deep enough, then there is a possibility that
not enough passive resistance can be obtained in front of the wall. If the penetration depth is
Project number
415660-21 v04
Date
December 2007
Page
251 of 278
approximately 0.5 or 1.0 times the retaining height, then enough passive resistance can usually
be generated. This also goes for cellular coffer dams.
4.7.7 Boundary conditions, starting points and calculation scheme
Schematization
For the testing of the design the dimensions and the following data is necessary:
dimension and location of the sheet pile wall;
dimension, location and distance between anchors;
dimensions of the wales;
soil composition at the location of the coffer dam construction;
strength and rigidity parameters of the sheet pile wall, the anchors and the wales;
strength parameters of the soil layers;
groundwater situation;
earth pressure factors and wall friction angles for the active and passive situation;
partial material factors;
external loads, like ship loads, top loads, wave loads ice loads and others;
retaining height including possible uncertainties, like e.g. excavations due to erosion or
dredging.
The first three parts follow from the design calculations, where obviously assumptions are en-
tered initially. This can be done according to the rules of thumb of paragraph 4.7.6.
If the choice does not pass the test, the dimensions must be altered.
Calculation scheme
The following scheme offers a start for the calculation methods that have to be done.
The decision lies between a traditional calculation method and a finite element method.
If the finite element method is chosen, the program determines the most critical mechanism
for the design.
If the traditional calculation method is chosen then it must be determined in a general way if
the active and passive soil wedge between the sheet pile walls intersects. If this is not the
case, then the dimensioning of the sheet pile walls and anchors can be done by means of e.g.
the classic method according to Blum, or by means of a spring model. This is not a coffer
dam.
If the active and passive soil wedge intersect in the filling and the construction is considered
to be a coffer dam, then the front wall must first be calculated in the same manner as a
regular sheet pile wall. This lead to the anchor force, the moment graph and the shear force
graph.
The so called Ansatzpunkt or starting point is, in case of a fixed wall, the point of zero
shear forces at the bottom of the wall. In case of a non-fixed wall it is the underside of the
wall. It is now possible to determine the active zone. This calculation can easily be done by
means of a sheet pile wall program. With the results the front wall and the anchors can be
dimensioned.
Now that the active zone and the anchor force are known, the maximum passive pressure
envelope of the filling against the back wall can be determined by Hombergs method. This
can be used to check the calculation for the design of the back wall. The external force is
Project number
415660-21 v04
Date
December 2007
Page
252 of 278
the anchor force that was obtained by the calculation of the front wall. The moment graph
can now be constructed, which will be used to dimension the back wall. In addition to
Hombergs method, the internal stability can also be determined with Terzaghis method.
Finally the total stability must be checked.
Project number
415660-21 v04
Date
December 2007
Page
253 of 278
5 Stability of a soil mass
5.1 Introduction
The stability of a soil mass is largely determined by the shear strength of soil, which is the
resistance of the soil against shear deformation. To be able to determine how steep a slope can
be made before failure occurs, the magnitude of the available shear strength of the soil has to
be investigated. In the lectures ,,Soil Mechanics it has already been explained that the shear
strength is related to the effective stress, the cohesion and the angle of internal friction accord-
ing to the formula of Coulomb:
| o t ' ' + ' = tan
n f
c
(5.1)
where
c' = effective cohesion (kN/m
2
)
o'
n
= normal effective stress on the failure or slip surface (kN/m
2
)
|' = effective angle of internal friction (degrees).
f
t = shear strength (kN/m
2
)
For the background of the shear strength is referred to the lecture notes on soil mechanics.
In the case of a fill for a dike, a dam or an embankment for a road, the slope stability does not
only depend on the properties of the fill material, but also on the properties of the underlying
soil layers. For example, in the western part of The Netherlands it has to be taken into account,
that during filling operations excess pore water pressures are introduced in the underlying soft
clay and peat layers. Due to the excess pore pressures, the effective stresses reduce, so that in
turn the available shear strength of the soil reduces, which has a negative effect on the stability.
In the case of an excavation (e.g. building excavations, trenches, etc.), other aspects play a role
too. As the groundwater level in The Netherlands is usually quite high, it should be prevented
that groundwater flows into the excavation. Not only because it makes excavation more diffi-
cult, but also because of the seepage of water result in forces that reduces the stability of the
slopes.
5.2 Slope stability calculations
Loss of stability of a mass of more or less homogeneous soil usually takes place along a slip
surface. Although the stability of a slope can be analysed with e.g. the finite element method,
traditional methods of analysis are often used. These methods are based on assumptions that
transform the problem into one, which is statically determinate. In these lectures traditional
methods will be dealt with, whereby the stability of a soil mass is investigated assuming a
plane or cylindrical slip surface (see Figure 5.1).
Project number
415660-21 v04
Date
December 2007
Page
254 of 278
A
B
C
H
T
R
N
G
0
A
G
d
B
r
r
u
t
The governing factor for the stability of a soil mass is the ratio of the available shear strength
of the soil to the occurring shear stresses along the failure surface. This ratio is sometimes
called the safety factor against sliding. A safety factor higher than 1 indicates a stable situation,
sliding of a part of the soil mass does not occur. In case the safety factor is just higher than 1
deformations of the sloping soil mass will increase. If the safety factor is lower than 1 failure
occurs. Also other definitions of the safety factor exist, such as e.g. the ratio of the sum of the
resisting moments to the sum of the driving moments.
5.3 Stability of an infinite slope
Although infinite slopes do not exist the infinite slope method can be used to evaluate the de-
gree of stability of certain slopes. However, if a sliding plane is relatively long and shallow, the
curved edges of the sliding plane have only a small effect on the stability of the soil mass. For
such a shallow sliding plane (e.g. the interface of a thin clay layer in a sloping sand mass) the
analysis according to the infinite slope method can be used to determine the stability of a slope.
In Figure 5.2 a soil element of a slope with a slope angle | is shown. The depth from the slope
surface to the slip surface is z.
|
t T =
1
cos |
x
o
cos |
n x
' N =
1
A B
D C
Z
W
1
|
Z
Figure 5.1 A plane and a cylindrical failure surface
Figure 5.2 Element of infinite slope
Project number
415660-21 v04
Date
December 2007
Page
255 of 278
The safety factor of the element against sliding can be expressed as:
t
t
f
F = (5.2)
where:

t
f
= the available shear strength
t = the actually occurring shear stress.
For equilibrium, the sum of all the forces acting on the soil element should be zero. The forces
acting on the sides of the element are considered to be of equal magnitude and in opposite di-
rection. The force W (weight of the soil element) and the resultant R of the forces T and N on
the base should also be of equal magnitude and in opposite direction. The thickness of the ele-
ment perpendicular to the plane of the paper is 1 and the width of the element in the plane of
the paper is also 1. The area of the base of the element is therefore.
| | cos
1
cos
1 1
=
The normal stress on the base is '

n
= N / l
DC
and the shear stress on the base is = T / l
DC
. The
shear strength, which can be mobilised along the base, can be expressed according to formula
(5.1) of Coulomb.
In this analysis it will be assumed that the slope consists of dry cohesionless sand (c' = 0). The
weight W of the element is equal to the resultant force on the base and is equal to:
1 1 = = z R W
where:
= unit weight of soil
z = height of the soil element.
Subsequently is:
| cos R N =
| sin R T =
As the base area is
n
o
|
' ,
cos
1
and t can be written as:
|
|
o
2
cos
cos
1
z
N
n
= = '
| |
|
t cos sin
cos
1
z
T
= =
The safety factor is then equal to:
|
|
| |
| |
| |
| o
t
t
tan
tan
cos sin
tan cos
cos sin
tan
2
= =
'
= =
z
z
z
F
n
f
(5.3)
Project number
415660-21 v04
Date
December 2007
Page
256 of 278
Equation (5.3) shows that the factor of safety of a slope in cohesionless material (c' = 0) is
independent of its height and that the slope is stable when < | '.
Although the stability is independent of the depth of the sliding surface, the curved edges of the
sliding plane will have an increasing influence on the stability as the depth increases and that
influence can not be neglected anymore.
If the soil material of which the slope exists exhibits cohesional forces, then this cohesion will
the available shear strength along a sliding plane increases, which in turn will increase the sta-
bility. Since the shear strength provided the cohesion is constant and the shear strength related
'
n
tan|' increases with depth, the relative contribution of the cohesion to the stability decreases
relatively with depth.
Example
Determine the maximum height h (see Figure 5.3) of a vertical slope, which is stable directly
after construction. The fill material consists of fully saturated clay with a low permeability, a
cohesion of c' = 50 kN/m
2
and a wet unit weight of 20 kN/m
3
. It is assumed that the governing
slip surface is a plane, which makes an angle | with a horizontal plane. Directly after construc-
tion of the slope, the angle of internal friction | ' = 0.
A C
B
|
h
Solution
If the safety factor F = 1, the slope will be just stable. The driving shear force along the plane
slip surface is then equal to the resisting shear force.
The resisting shear force along line BC is:
| sin
h c
l c T
BC f
'
= ' =
The driving shear force is caused by the weight of the sliding wedge, which is:
|
|
|

sin
cos
2
1
tan
1
2
1
2
2 2
h h
l l
W
AC AB
= =
=
The driving shear force is:
| | cos
2
1
sin
2
h W T = =
For F = 1 and T
f
= T we obtain:
Figure 5.3 Assumed sliding plane in a soil mass with a vertical slope
Project number
415660-21 v04
Date
December 2007
Page
257 of 278
| | |
|
| 2 sin
4
cos sin
2
cos
2
1
sin
2
c c
h h
h c '
=
'
= =
'
The critical sliding plane is obtained for a minimum value of h. This occurs when sin2|
reaches a maximum value, which is 1 for | = 45. The maximum height of a stable slope is
then:
m
c
h 10
20
50 4 4
=
=
'
=
5.4 Circular slip surfaces

In a more or less homogeneous soil mass or embankment, slip surfaces are often curved. When
assuming circular slip surfaces, the safety factor against sliding can be expressed as the
ratio between the sum of the resisting moments and the sum of the driving moments with re-
spect to the centre of the circular slip surface:
=
moments driving
moments resisting
F
As long as F > 1, the slope is stable, if F = 1, the slope is just stable and for F < 1 the slope
fails.
t
A
G
B
C
r
r
0
d
u
From Figure 5.4 it appears that the resisting moment can be expressed as the force formed by
the maximum shearing resistance along the slip surface multiplied by the radius of the slip
circle. The driving moment can be expressed by the weight of the sliding mass of the soil mul-
tiplied by the distance between the line of action of the weight and a vertical line through the
centre of the slip circle.
The | = 0 analysis for a circular slip surface
Stability analysis of a recently constructed embankment of completely saturated clay with a
low permeability can be carried out using the so called (c' = 0) analysis (see Figure 5.5). As
excess pore water pressures will develop during filling operations for the construction of an
embankment, a stability analysis has to be made for the construction stages. For such a con-
struction stage an undrained analysis is suitable, for which it is assumed that | ' = 0.
Figure 5.4 Trial circular slip surface
Project number
415660-21 v04
Date
December 2007
Page
258 of 278
This simplifies the problem, as it is not necessary anymore to know the distribution of the ef-
fective stresses along the rupture surface of the slip circle. In such a case the resisting moment
becomes:
r L c r L c M
a a n f
' = ' ' + ' = ) tan ( | o
where:
c' = cohesion along the slip surface
L
a
= length of circular arc
r = radius of the circular slip circle
The driving moment is:
M = Wd
where:
W = resultant of the forces that act on the soil mass
d = distance between the line of action of W and a vertical line through the centre of the
slip circle.
The safety coefficient can be expressed as:
Wd
r L c
F
a
= (5.4)
0
d
W
L
C
r
u
a
r
Example
An embankment consisting of clay with very low permeability has a slope of 1:1 and a height
of 8 m. The clay has a cohesion (or undrained shear strength) of 65 kN/m
2
and a wet unit
weight of 19 kN/m
3
. Determine the safety factor against sliding along a circular slip surface as
shown in Figure 5.6. The area of ABCD is 70 m
2
and the angle AOC is 89.5. The slip circle
passes through the toe of the slope.
Figure 5.5 Stability analysis for = 0
Project number
415660-21 v04
Date
December 2007
Page
259 of 278
0
W
B
r =
12
.1
0
m
C
D
45
3.50 m
8.00 m
89
4.50 m
A
1
2
Solution, driving moment
For the calculations a soil mass of 1 m thickness perpendicular to the plane of the paper is con-
sidered. The weight of the soil mass ABCD is:
W = area ABCD x thickness = 70 1 19 = 1330 kN
d = 4.5 m
The driving moment becomes:
M = 1330 4.5 = 5985 kN/m
Solution, resisting moment
Angle AOC = 89.5.
Length of slip surface L
a
= 89.5/360 2 t 12.1 = 18.9 m
The resisting moment becomes:
M
w
= 65 18.9 12.1 = 14865 kN/m
The safety factor is:
F = 14865/5985 = 2.48
In the example the safety factor against sliding for one slip circle has been determined. The aim
of a stability analysis is to calculate the safety factor for a series of potential slip surfaces and
to find the slip circle for which the safety coefficient against sliding is a minimal. This circle is
called the critical circle or critical slip surface. To find the critical slip surface involves a con-
siderable amount of calculations. This is usually done by computer.
5.4.1 The method of slices, Fellenius
The stability analysis becomes more complicated, when effective stresses have to be taken into
account. The effective stresses cause normal and shear stresses along the considered slip sur-
face, the magnitude and direction of these shear stresses vary with depth. Furthermore, the
sloping soil mass and the subsoil may consist of several different soil layers. For this case the
method of slices can be used.
Figure 5.6 Stability analysis for toe circle
Project number
415660-21 v04
Date
December 2007
Page
260 of 278
First, a potential circular slip surface is selected for a stability analysis. Subsequently the soil
mass above the slip surface is divided into a number of slices (see Figure 5.7). The width of
each slice can vary. The slice width is made so small, that the base of each slice can be ap-
proximated by a plane. This plane makes an angle o with a horizontal plane. The height of a
slice measured along a vertical line through the centre is h. The thickness of the slices is equal
to 1 [m] and the base width is equal to l[m].
The following forces act on a slice:
- the weight W = b h [kN/m]
- the total normal force N [kN/m] on the base consisting of two components, one due to the
effective stress, N' = o'
n
l [kN/m
2
] and the other one due to water pressure, ul [kN/m
2
] (u =
water pressure at the centre of the base)
- the shear force on the base, T = t l[kN/m]
- normal forces on both sides, E
1
and E
2
[kN/m]
- shear forces on both sides X
1
and X
2
[kN/m].
In addition to the above mentioned forces, other forces, which act on a slice, such as e.g. sur-
charge may be present. As the forces on the sides of a slice are unknown, the problem is stati-
cally indeterminate and assumptions have to be made to find a solution. Several researchers
have shown that little error is introduced by neglecting these side forces. When these forces are
neglected the problem can easily be solved.
0
A
B
C
D
r
b
h
l
o
r sin o
N
'
u
l
N
E
1
E2
X
1
X
2
W
o
o
T
The factor of safety is again expressed as the ratio between the sum of the resisting moments
and the sum of the driving moments. For each slice the resisting moment is obtained by multi-
plication of the shear force T = t l with the radius of the slip circle. The driving moment is
obtained for each slice by multiplication of the weight W of each slice and the distance r sino
between the line of action of W and a vertical line through the centre of the slip circle.
The safety factor against sliding is:
= =
r Wr
lr
moments driving
moments resisting
F
f
) sin(o
t
(5.5)
Figure 5.7 Geometry of the method of slices
Project number
415660-21 v04
Date
December 2007
Page
261 of 278
Using Coulombs law (5.1), these formulae can rewritten:
' +
=
o
| o
sin
) ) ' tan ' (
W
l c
F
n
(5.6)
For the assumption that the side forces on the slices (or the forces in between the slices) are
neglected we obtain:
N' = Wcoso - ul=o'
n
l
and subsequently
+
=
o
o |
sin
) cos ( ' tan '
W
ul W lc
F
(5.7)
As angle o between the base of a slice and a horizontal plane can be measured in the figure, it
is possible to determine the factor of safety against sliding.
Example
A simple numerical example is now given to illustrate how to solve a stability problem with the
method of slices.
Figure 5.8 shows a slope and a circular slip surface. The soil mass above the slip surface has
been divided into 8 slices. The unit weight of the soil = 20 kN/m
3
, the cohesion c' = 10
kN/m
2
and the angle of internal friction | ' = 29. The water pressure at the base of each slice
and further information required to solve the problem is provided in Figure 5.8. Required is the
safety factor against sliding for the indicated slip surface.
Pore water
pressure
(kN/m )
2
6.00 m 2.50 m
3.15 m
1
2
3
4
5
6
7 8
0
0
11
17
19
16
12
6
1.50 m
2
:3
0 1 2 3 4 5 10m
0
Figure 5.8 Geometry for hand solution of method of slices
Project number
415660-21 v04
Date
December 2007
Page
262 of 278
Solution
Formula (5.7) can be used to determine the value of the safety factor. The weight of each slice
is (considering a slice thickness of 1 m):
W = bh 1 = 30h kN
The height h of each slice is drawn from the centre of the base downwards, where after the
normal component hcos(o) and the tangential component hsino can be drawn. These compo-
nents can also be determined by measuring the base angle o of each slice in Figure 5.8 and
subsequently solving hcos(o) and hsin(o) for each slice. With this information Wcos(o) =
30hcoso and Wsin(o) = 30hsin(o) can be computed for each slice.
Table 5.1 shows the data which has been obtained from Figure 5.8.
slice no. hcos(o)
(m)
hsin(o)
(m)
u
kN/m
2
l
(m)
ul
(kN/m)
1
2
3
4
5
6
7
8
0.75
1.80
2.70
3.25
3.45
3.10
1.90
0.55
-0.15
-0.10
0.40
1.00
1.75
2.35
2.35
0.95
6
12
16
19
17
11
0
0
1.55
1.50
1.55
1.60
1.70
1.95
2.35
2.15
9.3
18.0
24.8
30.4
28.9
21.4
0
0
Total 17.50 8.55 L
a
=14.35 132.8
With the data of Table 5.1 the following can be calculated (slice thickness is 1m):
kN W 525 50 . 17 30 cos = =
o
kN W 5 . 253 45 . 8 30 sin = =
= = kN ul W 2 . 392 8 . 132 525 ) cos ( o

Using formula (5.7), the safety factor is calculated as follows:
42 . 1
5 . 253
2 . 392 554 . 0 35 . 14 10
sin
) cos ( ' tan '
=
+
=
+
=
o
o |
W
ul W lc
F
5.4.2 The method of slices, Bishop
The most frequent used method in engineering practice is Bishops method. In contrary to Fel-
lenius, Bishop incorporated that there are forces acting between the slices, and that these forces
working on both sides of a slice (Figure 5.7) can be of different magnitudes.
The Bishop method focuses on the total moment equilibrium of the sliding soil mass and the
vertical force equilibrium of each slice.
Table 5.1 Data of slices
Project number
415660-21 v04
Date
December 2007
Page
263 of 278
) sin( ' ) cos( ) ' ( o o T lu N W + + =
(5.8)
Expressed in stresses:
) sin( ) cos( ) ' ( o t o o l lu l W
n
+ + =
(5.9)
) cos(
=
b
l (5.10)
) tan(
) cos(
) sin(
) cos(
) cos( ) ' (
o t o
o
o t
o
o o
b ub b
b u b
W
n
n
+ + ' = +
+
=
(5.11)
The factor of safety F is defined as the quotient of the available shear stress ( ' tan ' ' | o
n
c + )
and the (for moment and vertical equilibrium) required shear stress (t

).
t
| o
t
t
' tan ' '
n
f
c
F
+
= = (5.12)
F
b c
ub b W
n
n
) tan( ) ' tan ' ' (
'
o | o
o
+
+ + = (5.13)
or
F
F
b c
ub W
b
n
' tan tan
1
tan '
- -
'
| o
o
o
+
=
(5.14)
The safety factor against sliding is:
= = =
r W
lr
r W
Tr
moments driving
moments resisting
F
f
) sin( ) sin( o
t
o
(5.15)
Replacing t
f
according to formula (5.1), replacing l according to formula (5.10) gives:
' +
=
o
o
| o
sin
cos
' tan ) ( '
W
b b c
F
n
(5.16)
substituting (o
n
'b) (5.14) in (5.16) leads to the final equation for Bishops method:
Project number
415660-21 v04
Date
December 2007
Page
264 of 278
|
|
|
|
.
|
\
|
|
.
|
\
|
+
+
=
o
| o
o
|
sin
' tan tan
1 cos
' tan ) ( '
W
F
ub W b c
F
(5.17)
In formula (5.17) the factor of safety F figures on both sides of the equation. This implicit ex-
pression can be solved iteratively successive approximations. In most cases the convergence is
rapid. Bishops method is in general seen as better than Fellenius because it is more consistent.
For | ' = 0 the Fellenius and Bishops methods are the same
5.4.3 The method of slices, Bishop with Earthquake effect
Earthquakes influence the stability of slopes, dikes, earth dams, excavations etc. considerably.
This is be caused by increase in the shear stresses, but also by a lowering of the shear strength
of the soil due to the cyclic loading character of the earthquake. The shear strength can be re-
duced because the movements within the grain skeleton reduce the effective stresses. Deforma-
tions, e.g. in earth dams, due to the influence of the earthquake may also increase the pore
pressures; the excess pore pressures affect the stability. If a dam e.g. is made of loosely packed
fine sand, consisting of more or less round grains of equal grain size, an earthquake force could
cause densification, which in turn causes excess pore pressures that induce liquefaction.
Famous, severe, earthquakes occurred in Lisbon (1755), San Francisco (1906), Tokyo (1923),
Kobe (1995) and Sumatra (2005) and after that many others.
When building a structure, one first has to ascertain the probability of an earthquake and the
possible intensity. This can be done by studying seismograms that have been registered in the
region. It is advisable, if not necessary, to call in an expert geologist for that purpose, as an
extrapolation of the records for the future has to be made. A study of the geologic conditions is
necessary. There are two methods of calculating the behaviour of a dam, slope, etc. under vi-
brational motions of an earthquake. One is dynamic analysis. For that analysis, not only the
magnitude of the vibrations should be known, but also the frequency. The dynamic behaviour
of the soil mass can be studied with the finite element method. It is an elaborate approach but
the method describes the physical processes quite well. The other method, called pseudo-static,
is a rather rough simplification of the physical processes. In this method which it is assumed
that the effect of the earthquake can be expressed in a horizontal thrust, perpendicular to the
dam, or other earth structure, which is a certain percentage of the weight of the dam. The idea
is that a certain horizontal acceleration can be assumed which defines the severeness of the
earthquake. This acceleration can be expressed as a = A.g in which g is the gravity acceleration
and A, a seismic coefficient. A depends on the severeness of the earthquake. Usually A is taken
to be 0.10 till 0.15 for normal to moderately severe earthquakes. For light earthquakes A = 0.05
can be taken. This method entails that on a slice a horizontal force AW acts due to the potential
earthquake, if W is the weight of the slice of a potential slip circle. If A = 0.1, the additional
horizontal force will be 0.1 Wr.
Taken the effect of an earthquake load in account in the Bishop method is adopted by adding a
horizontal load in each slice (see Figure 5.9): the mass of the slice time the horizontal earth
quake acceleration (a).
Project number
415660-21 v04
Date
December 2007
Page
265 of 278
N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

AW
W
r
N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

AW
W
r
The driving moment (M
o
) now becomes:
r AW W moments driving M
o
)) - ) ((cos( ) sin( ( o o + = =

(5.18)
Where r is the vertical distance from the bottom of the slice to the centre of gravity of the
slice.
As described before A is the ratio between the horizontal earthquake acceleration (a) and the
acceleration of gravity (g). The adopted Bishops equation then becomes:
)) (cos sin (
' tan tan
1 cos
' tan ) ( '
o o
| o
o
+
|
|
|
|
.
|
\
|
|
.
|
\
|
+
+
=
AW W
F
ub W b c
F
(5.19)
In Annex B a worked-out case of a slip circle analysis for a dam is given.
5.5 Special cases
5.5.1 Stability of clay slopes
As clay has a low permeability, the stability of slopes in clay due to excavation or filling opera-
tions is time dependent. It has already been explained that during the construction of an em-
bankment on a subsoil of clay, excess pore water pressures develop. This means that the effec-
tive stresses in the subsoil do not increase directly in proportion to the increasing weight of the
fill. Since the effective stresses do not increase quickly enough during filling operations, the
shear strength in a potential slip surface increases only very slowly during filling.
Figure 5.9 Bishops simplification with earthquake force
Project number
415660-21 v04
Date
December 2007
Page
266 of 278
Therefore, the degree of stability during filling operations depends largely on the initial effec-
tive stress and cohesion, just before filling. This is called the short-term stability. The stability
during filling operations is often determined with a | ' = 0 analysis. The shear strength related
to the initial effective stress is then called the undrained shear strength (C
u
) and the angle of
internal friction is assumed to be zero (| ' = |
u
= 0). In the period after execution excess pore
water pressures will dissipate, the shear strength along a potential slip surface will increase and
subsequently the safety factor against sliding will increase. This factor reaches a maximum
value when the excess pore water pressures are zero. The stability at that particular state is
called the long-term stability. This long-term stability is larger than the short-term stability.
When an excavation is made in clay, negative pore water pressures develop. In the short-term
effective stresses are larger due to the unloading than in the long-term, as the effective stresses
decrease due to the dissipation of the negative pore water pressures. For an excavation in clay
the shear strength along a potential slip surface will slowly decrease in the period after excava-
tion. In this case the long-term stability will be smaller than the short-term stability. This can
be an advantage if e.g. a trench is dug for a pipeline, which is filled again the same day. For
one day the slope can then can steeper than in the case that the trench has to stay open for sev-
eral weeks.
During the construction of an embankment, dike or dam on soft soil two elements are of special
interest for the design and execution:
- The increase of strength in time. During the consolidation period the soil strength increases
due to the dissipation of the excess pore pressures.
- The change in geometry in time due to the deformation of the subsurface caused by con-
solidation (and creep).
These two aspects make the design for the execution of an embankment an iterative process.
The increase of strength in time
In some cases the cohesive soft subsoil is not strong enough to withstand the full load of the
embankment in one time (C
u
analysis). Then the choice is made to construct the embankment
in several heightening steps. First the maximum height under fully undrained conditions is
determined. Secondly the next heightening steps are designed, for instance steps of 0.5 or 1.0
m. For each step it is calculated how much of the excess pore water pressure there has to be
dissipated before it is possible to apply a next heightening step under stable conditions. This
procedure is repeated until the full embankment height is reached. Based on a consolidation
calculation a prediction is made how much waiting time between the heightening steps is re-
quired. During the execution of the work this time schedule is only indicative. In the field usu-
ally the observational method is applied. The decision of applying a next heightening step is
based on measurement of the excess pore pressure with pore pressure transducers in the soft
soil layers. The next heightening step is applied when the pore water pressure measurements
indicate that the excess pore water pressures are reduced after some time to the level that was
calculated in the stability calculations for this next step. The placing of the pore pressure trans-
ducers has to be on the most critical area: below the slopes of the embankment in the underlay-
ing soil layer. This method is called the red and the green lines method. The red line shows the
line maximum allowable excess pore pressure at a certain embankment height and should never
be exceeded. The green line is the prediction of the excess pore pressures based on a certain
heightening schedule (stages), see Figure 5.10. If the line with measured excess pore pressures
approaches the maximum value, the probability of instability increases and further heightening
should be delayed.
Project number
415660-21 v04
Date
December 2007
Page
267 of 278
Change in geometry due to settlements
Settlement due to the heightening can be very high (sometimes several meters), so that during
the stability calculation process for the heightening steps an updated geometry has to be used.
This change geometry can be calculated by using a normal settlement calculation. Fortunately
now days there are computer programs that do this updating of the geometry automatically. For
the fact that the soil settles compensation height has to be applied. This compensation height
has to be incorporated in all the stability and settlement analyses. For dykes in the Netherlands
usually the settlement for a life span of 50 years is incorporated.
Speed up of pore water pressure dissipation
Sometimes there is not enough time in a project to wait and let the excess pore water pressures
dissipate in a natural way. Several methods are available to speed up this process: This usually
is accomplished by using some kind of vertical drainage in the form of strip drains or sand
columns. The principle of this system based on a reduction of the distance for the dissipation of
water with an excess pressure. The shorter dissipation distance causes a higher hydraulic gradi-
ent which in turn leads to a higher velocity of the water flow en hence a shorter consolidation
period and the process will speed up.
5.5.2 Stability of slopes in fine sand
Phenomena as have been discussed for clay, can also occur to a lesser extent in sand. When
dense sand is made to shear, dilatancy (increase in volume) will occur and as a consequence
negative pore water pressures will develop. These negative pore water pressures will dissipate
quicker than in clay. Therefore the temporary increase in stability due to these negative pore
water pressures does not last very long. The reverse effect can be experienced with loosely
Figure 5.10 Excess pore pressures due to heightening, limit, prediction and measured values
Project number
415660-21 v04
Date
December 2007
Page
268 of 278
packed sand. When such sand is made to shear, there will be a tendency for a decrease in vol-
ume (contraction). A decrease in volume means that pore water has to flow out of the pores.
If the sand is loaded rapidly, the tendency to volume increase will cause excess pore water
pressures. The effective stresses will decrease and the sand mass becomes in a state of lique-
faction. The soil will then behave as a heavy fluid. If this happens at the edge of an underwater
slope, a flow slide will occur and a very flat slope (often flatter than 10) will be the result.
This phenomenon used to occur regularly along the coast of the Eastern Scheldt in Zeeland,
The Netherlands.
At first a steep underwater slope is created by scouring seawater. Subsequently the slope col-
lapses locally, but the failure zone expands rapidly due to progressive failure. Such a flow slide
can involve large amounts of sand.
5.5.3 Composite slip surface
The critical circular slip surface will not always yield the minimum safety factor. Figure 5.11
shows an example. In the depicted embankment is a thin layer, which will act as a slip surface.
A stability analysis with circular slip surfaces will provide a too high safety factor, while the
critical circular slip circle will cut the thin clay layer, which has relatively low shear strength,
over a short length. In the case of the slope of Figure 5.11 a more realistic approach is to apply
a composite slip surface, consisting of a circular and a plane part.
thin soft clay layer
start of plastic zone
5.6 Influence of groundwater on the stability of a soil mass
5.6.1 General
The groundwater condition is of eminent importance for the stability of a soil mass. The height
of the groundwater table determines the groundwater pressure along a potential slip surface.
Since (o'
n
tan |) in Coulombs law and as Terzaghis law states that o' = o - u, higher water
pressures lead to lower effective stresses and hence a lower available shear strength of the soil.
Figure 5.11 Composite slip surface
Project number
415660-21 v04
Date
December 2007
Page
269 of 278
5.6.2 Groundwater flow and stability of dams
In the simplest situation there is a hydrostatic pore pressure distribution present. In other situa-
tions, a difference in groundwater head along the potential slip surface causes groundwater
flow. For the determination of the groundwater pressure distribution a long a potential slip
surface computer programs based on finite element of finite difference can be used. Also
graphical methods are available. In annex A the graphical method for the determination of pore
pressures with the aid of a flow net is explained.
It is not intended to discuss in any full measure the design of dams and dikes. Only a few as-
pects will be mentioned. The main purpose is to indicate in what way low pore pressures
which, as was shown, are essential in improving stability can be obtained.
Figure 5.12 (a) shows the cross-section of a dam, consisting of homogeneous soil. The pore
pressures acting in the slip circle are shown; the resultant force is U.
Impermeable layers
A method to reduce the pore pressures is illustrated in Figure 5.12 (b). A clay blanket with a
low permeability is applied upstream. As a result, the phreatic line takes a dip (the dotted line
indicates the phreatic line if there has been no clay blanket). The pore pressures are reduced
and so is the resultant U.
Sometimes a clay core is applied, see Figure 5.12 (d) and (e), instead of a clay blanket. Such a
core is less effective in reducing the pore pressures. However it is less vulnerable to damaging
influences than a clay blanket. The damage might be caused by animals, men, the drying out of
the clay where there is a varying water table and a hot dry climate, etc.
Drainage
A further reduction in pore pressures can be obtained by providing the downstream side of the
dam with a horizontal drainage, as is shown in Figure 5.12 (c).
The drain at the downstream part of the dam may be designed as a chimney drain, as shown in
Figure 5.12 (f). The drain can be so located that the pore pressures in the interior part of the
dam are low under all circumstances, e.g. at heavy rainfall. In case the horizontal permeability
of the dam soil is high a chimney drain is appropriate.
Care should be taken that the chimney drain is not too long, because the flow of water through
the dam to the drain might then be too greatly increased. That could lead to piping, especially if
the dam were to show weak spots in the clay blanket, due to bad construction or damage after
construction.
Another aspect of the use of chimney drains is the possible increase of the quantity of water
flowing through the dam. In hot, dry, climates where water is very scarce, the loss of water
from a reservoir caused by seepage through the dam should be reduced as far as possible. This
aspect should not be disregarded; for each different design the water losses should be com-
puted. Of course a well-designed clay blanket or clay core, sufficiently thick, well-placed and
possessing a low permeability helps to reduce the seepage losses. If water is extremely scarce it
should be tried to drain off and catch the seepage water.
It should be emphasized that in any event the drains should be able to carry the quantity of
water which flows through the dam. Especially if heavy rainfalls occur which can penetrate the
interior of the dam, the drains should be wide and pervious enough to carry the total quantity of
water and at the same time keep the pore pressures low.
Project number
415660-21 v04
Date
December 2007
Page
270 of 278
The influence of clogging of drains on the stability of a dam
The grading of the drain material and the placing of the different grades should be such that
penetration of fine materials into the drains is prevented as much as possible. This is necessary
to protect the drains against clogging. Even if this has been taken care of, it must be accepted
that in the long run, e.g. after 50 years, drains might have stopped functioning properly because
of the occurrence of clogging. The stability of a dam should therefore also be analysed, assum-
ing that the drains do not function. In that case, however, a factor of safety F
S
= 1 can be ac-
cepted, as these conditions will occur after a long period in which the dam has settled and ac-
quired additional strength. With full functioning of the drains, the factor of safety should be 1.3
1.5, depending on the importance and size of the dam and the risks run with a view to people
and properties. These factors can only be used as a guideline when the soil conditions and
properties have been ascertained accurately and the design is of a high standard.
Figure 5.12 Pore pressures along a circular slip surface for different pore pressure reduction measures
5.6.3 Effect of rapid draw down of reservoirs on dam stability
Sometimes the upstream water table may fall rapidly (rapid draw-down condition). Usually this
is caused by a failure of a secondary dam, which empties the reservoir in a short time. Such
rapid draw-down conditions are very dangerous for the stability of a dam. Figure 5.13 illus-
trates this.
The horizontal dotted line indicates the initial upstream water level; when this water level
drops suddenly, the water pressure against the dam falls away. The pore pressures in the dam
remain high as they have hardly time to change in such short time. Figure 5.13 indicates the
phreatic line which possibly might occur when the drawdown has been completed. In this case
the existing pore pressures are only moderately influenced by the rapid drawdown. If the draw-
down occurred in zero time (t = 0), there would be no change at all in the, pore pressures. This
situation will not occur, because even an extremely rapid drawdown will take time. The actual
pore pressures will depend on the lowering of the water table as a function of time. It should be
noted that during the period in which the water table drops, the water flow is not steady. A
flow net can still be constructed but it will change with time. The point is to find out for which
d
e
f
Project number
415660-21 v04
Date
December 2007
Page
271 of 278
flow net the pore pressures will be the most dangerous. This can only be answered by thorough
investigations, theoretical or in a model test in which the drawdown period will be a major
factor.
In normal cases, drawdown lasting 5 to 10 days, an approximate method might be followed. In
that method the upstream water level is assumed to be as low as ground level in the reservoir
near the dam slope. The phreatic line is maintained in the downstream part of the dam, but is
curved down to the upstream slope of the dam. It touches the upstream slope at its lowest point;
the slope acts as an asymptote (see Figure 5.14). For that phreatic line a flow net can be con-
structed (see Annex 1).
If one goes into a detailed study of the stability conditions due to a rapid drawdown the situa-
tion is quite complicated.
In Figure 5.14 the phreatic line is sketched during the total rapid drawdown period for ten in-
tervals. In interval 1 the initial steady flow condition is given. Interval 2 shows the phreatic line
after a certain drop of the upstream (reservoir) water level. At the end, in interval 10, the
phreatic line has gone totally down to what can be described as normal ground water level.
What phreatic line corresponds with the most critical stability conditions depends, as was
stated, on the speed of the lowering of the upstream water level. A major element is the posi-
tion of the level at any given time relative to the position of the phreatic line and corresponding
flow net at that time. Up to the upstream water level the water pressure acting against the dam
will provide a counter force against loss of stability.
The influence of the speed of the rapid drawdown on the pore pressures inside the dam is
equally important. In Figure 5.14 various phreatic water levels are indicated. If the drawdown
has been slow, the phreatic level, critical with a view to stability, might correspond with the
phreatic line 5. If, however, the drawdown has been quick this level might coincide with for
instance phreatic line 2.
In this case the pore pressures not only would be much higher, but water would flow out of the
dam upstream surface, which is particularly dangerous to stability.
Depending on climatic conditions it is possible that a rapid drawdown occurs when there is a
heavy downpour of rain at the same time. In that case, the phreatic line can rise and cause pore
pressures even higher than the original pore pressures which existed before the rapid draw-
down started. This effect has to be evaluated carefully. By covering the, dam with an imperme-
able blanket the rain effect can be avoided. For dikes stretched out over long distances this is
more difficult and will certainly be more expensive.
Figure 5.13 The phreatic line just after rapid draw down condition
Figure 5.14 Phreatic lines (sketched) during the total rapid drawdown period for ten intervals
Project number
415660-21 v04
Date
December 2007
Page
272 of 278
So far homogeneous soil conditions have been assumed. In Figure 5.15 the situation is sketched
in which a dam rests on a pervious subsoil. In that case a different flow net will occur. As the
subsoil is permeable, much water will seep through underneath the dam if no counter measures
are taken. The seepage not only causes water losses from the reservoir, but will also indirectly
increase the pore pressure in the dam. In an extreme case, piping may occur underneath the
dam: due to the high water velocity, small soil particles are transported until in the end a con-
tinuous open "pipe" is caused in the soil through which a strong water current finds its way.
Finally downstream a blow-up in the permeable sand soil may occur, due to the force of the
water. To prevent this, the pervious layer may be screened off by sheet piling or a trench filled
with clay or other impermeable material, as is indicated in the two upper diagrams in Figure
5.15.
These screens are best placed near the upstream side of the dam, so as to reduce the pore pres-
sures underneath the dam. The best possible solution is when the screens can be driven to an
impermeable layer, as is the case in the upper two cases of Figure 5.15. But even if this is not
possible or feasible to reach the layer with a low permeability the screens will have a reducing
effect on the pore pressures. This can be ascertained by drawing a flow net for such case.
The lower diagram in Figure 5.15 shows relief wells placed in the pervious layer, intended to
lower the pore pressures underneath the dam. This method can be applied in case of an existing
dam, when pore pressures appear to become higher than acceptable and no other possibilities
of reducing the pore pressures are available.
5.6.4 The influence of a horizontal filter on the pore pressure and stability
To give some idea about for instance the influence of a filter in a dam on the pore pressure and
stability some calculations have been made for a dam with the specific dimensions shown in
Figure 5.16. The calculation has been done for certain soil parameters. These are not mentions,
as the intention is to illustrate the main aspects of the influence of the filter length.
Figure 5.15 Examples of piping prevention measures in an underflow situation
Project number
415660-21 v04
Date
December 2007
Page
273 of 278
Six cases for the filter length were considered. Flow nets for these cases are given separately in
Figure 5.17. For each case the stability has been calculated, according to Bishops method. The
outcome is given in
case Filter length
[m]
L/W
1)
[-]
F
s
[-]
1
2
3
4
5
6
0
30
40
50
60
70
0
0.13
0.17
0.22
0.26
0.30
0.96
1.21
1.38
1.52
1.61
1.61
1)
W is the base width of the earth dam (2.32 m)
Figure 5.16 Cross section of dam
Figure 5.17 Flow nets for different filter lengths
Figure 5.18 Overview phreatic lines for different filter lengths
Project number
415660-21 v04
Date
December 2007
Page
274 of 278
As to the filter length, the conclusion is that beyond L = 60 m there is no further improvement
of F
s
. It is fair to conclude that in each individual case there will be a maximum value beyond
which there is no additional improvement of the stability.
By carefully dimensioning of the filter a noticeable improvement of the stability can be ob-
tained: in this particular case up to a factor 1.6. It may be clear that this will be different in
other cases, depending on the geometry of the dam, the water levels upstream and downstream
and the soil properties. Each case has to be investigated in its own right.
The length of the filter has also an effect on the quantity of seepage. For the given dam Figure
5.20 illustrates the effect.
5.6.5 The influence of a filter on the stability in homogeneous soil with anisotropic
permeability
For the same cross-section of the dam shown in Figure 5.16 flow nets have been constructed
for different degrees of anisotropy. Some cases, for varying filter length, are shown in Figure
5.21. The influence of the anisotropy on the stability of the dam, however with different soil
properties, for varying filter length is given in Figure 5.22. It shows that with increasing hori-
zontal permeability the working of the filter is reduced. Beyond a certain value of k
h
/k
v
., which
will be different for different cases, a horizontal filter has no effect. In that situation a chimney
filter would be better.
Figure 5.19 Influence of filter length on the factor of safety
Figure 5.20 Quantity of seepage as a function of filter length
Project number
415660-21 v04
Date
December 2007
Page
275 of 278
5.6.6 Rapid drawdown: pore pressure and stability
In case of a rapid drawdown the water flow is unsteady. Pore pressures depend, apart from
data, on the speed in drop of the upstream water level. Calculations of pore pressures for these
unsteady situations are not simple. An approximate method is to draw quasi-static flow nets for
several time intervals a corresponding upstream water level.
This has been done for the dam of Figure 5.16, assuming isotropic soil with a permeability of k
= 10
-5
m/s. This calculation has been done for five time intervals 50 h, 100 h, 150 h, 200 h and
250 h. Figure 5.23 gives the corresponding phreatic lines and also the equipotential lines are
shown for time interval T = 300 h.
The factors of safety, as explained earlier, depend on the speed in drop of the upstream water
level. The flow net for the indicated time intervals and the resulting pore pressures, enable to
calculate the corresponding factors of safety against loss of stability of the upstream slope. The
result is given in Figure 5.24. As could be expected the factor of safety increases with increas-
ing emptying time. For example after 50 h, F
s
=1.1 and after 300 h, F
s
= 1.23. Once again: the
examples serve to give the general trend.
Figure 5.21 Examples of flow nets in soil anisotropic permeability
Figure 5.22 Influence of the anisotropy in permeability on the relation between filter length and factor of safety
Figure 5.23 Phreatic lines for different time intervals of rapid drawdown
Project number
415660-21 v04
Date
December 2007
Page
276 of 278
Finally Figure 5.25 gives an indication of the influence of anisotropy. The factor of safety
against loss of stability, for the given cross-section of the dam, is plotted against the degree of
anisotropy k
h
/k
v
. The emptying time t = 50 h. If k
h
is much lager than k
v
excess pore pressures
dissipate quicker from the interior of the dam. This leads to higher safety factors. F
s
is about
1.1 for the isotropic case and if k
h
/k
v
= 25 the factor F
s
= 1.4.
5.7 Magnitude of the required factor of safety
In the foregoing paragraphs methods have been described to compute the factor of safety when
the dimension of a slope and soil properties have been given. This calculated factor has to be
compared the required factor of safety to see if the conditions or the design are satisfactory.
For the determination of the required magnitude the required factor of safety in general four
aspects are to be considered.
1. First of all, maybe most important, is the aspect of allowable deformation. As can be
seen in stress-strain curves of laboratory tests it takes a certain amount of deformation
before the maximum shear stress that is the shear strength, has been mobilized. The de-
formation should not be more than the allowable in the earth structure. This means that
only a corresponding part of the available shear strength can be allowed.
Allowing the full shear strength to be developed would also mean that a failure with
very large deformations would be immanent. To avert that danger or reduction of the
allowable stress by introducing a factor of safety is necessary.
2. There also is an element of inaccuracy, first of all in the way |' and c' are determined.
It is impossible to take fully undisturbed samples and also during handling and testing
of the samples disturbance might take place. Apart from taking every precaution to en-
sure a reliable test result, it will be necessary to take a sufficient number of samples to
secure that the result can be treated in a statistical and probabilistic way.
Inaccuracy occurs also in the calculation methods as approximations are not flawless,
as is amply demonstrated by the fact that various methods give different factors of
safety for the same profile and soil conditions. It should be borne in mind that the sta-
bility calculation methods, more of which have been discussed are empirical methods,
Figure 5.24 Change in factor of safety as function of the emptying time of a rapid drawdown
Figure 5.25 Factor of safety as a function of the kh/kv for an emptying time T= 50 hours
Project number
415660-21 v04
Date
December 2007
Page
277 of 278
although they are not devoid of certain theoretical elements. Their accuracy can only
be measured in the field of completed earth works. This has been done frequently, pro-
ducing useful information but not enough to give an unequivocal estimate of the accu-
racy of the various methods.
3. Further there is an element of uncertainty about the loading conditions and another cir-
cumstances that might influence the stability such as more intensive earthquakes than
taken in account in the design, excessive rainfall in combination with failures in the
drainage system, unexpected loading and so on.
4. Finally there is the element of impact of a slope instability on people and property; in
another words the cost of lost of lives and economical damage. Especial the part of
weighing the cost of lives is a very difficult aspect concerning the answering of ethical
questions.
It is very difficult to catch all these elements in one factor of safety. Careful determination of
soil properties, assessment of pore pressures and loading conditions, careful deign and execu-
tion of the works contribute fundamentally to allowing for a lower factor of safety. As ex-
plained earlier it might vary, roughly speaking between 1.3 1.5. Much depends on the insight
of the designing and executing engineers.
Project number
415660-21 v04
Date
December 2007
Page
278 of 278
6 List of references
- Beyers, G.M.H. (1982)
Wegenbouw, Ned. Ver. Wegenbouwers.
- Craig, C.F. (1978)
Soil Mechanics, 2
nd
Ed., Van Nostrand Reinholt.
- Douwes Dekker, D.M. (1985)
Lecture book, Applied Soil Mechanics and Foundation Engineering,
Delft Geotechnics.
- Molenaar, A.A.A. (1974)
De aardebaan, collegedictaat T.U. Delft.
- Van der Veen, C., Horvat, E., Van Kooperen, C.H. (1981)
Grondmechanica, Waltman Delft.
- Verruijt, Arnold (2001)
Soil mechanics, Delft University of Technology, 2001
- TUDELFT
the lecture notes series CT5331 Sheet pile walls
Appendix
1
Date
December 2007
Page
1
Appendix 1 Determination of pore pressures with the aid
of a flow net
Appendix
1
Date
December 2007
Page
2
Appendix
1
Date
December 2007
Page
3
Appendix
1
Date
December 2007
Page
4
Appendix
1
Date
December 2007
Page
5
Appendix
1
Date
December 2007
Page
6
Appendix
1
Date
December 2007
Page
7
Appendix
1
Date
December 2007
Page
8
Appendix
1
Date
December 2007
Page
9
Appendix
1
Date
December 2007
Page
10
Appendix
1
Date
December 2007
Page
11
Appendix
1
Date
December 2007
Page
12
Appendix
1
Date
December 2007
Page
13
Appendix
2
Date
December 2007
Page
14
Appendix 2 Example of the calculation of the stability of
a dam
The stability of an earth dam will be considered using the cross section as given in Figure 6.1. The dam is
made of compacted clayey sand, on top of a sandy clay layer with an underlaying stratum of hard clay.
The dam is provided with a downstream filter.
The soil properties are as follows
clayey sand: c'
la
= 20 kN/m
2
|'
l a
= 33
s a t 1
=18.5 kN/m
3
sandy clay: c'
2a
= 30 kN/m
2
|'
2 a
= 28
s a t 2
=19.5 kN/m
3
p.m.
s a t 1
is the average unit weight of the clayey sand; this is an average value of dry, partially saturated and
completely saturated clayey sand within the dam. In real engineering cases it should be investigated what
part of the dam is saturated and what part is partially saturated or dry. The momantane distribution within
the dam depends on capillary rise, rainfall and permeability of the surface of the dam.
k
1
= k
2
= 5.10
-8
m/s (the same for clayey sand as for sandy clay).
The phreatic line is should be determined as part of the calculation. The upstream and downstream water
tables are assumed to be constant in time. Under normal conditions the upstream water level will show
certain variations. For assessment of the stability of the downward slope of the dam the highest water
table is taken that is maintained for an extended period.
In case of large variations in the upstream water table, the stability of the upstream slope has to be as-
sessed. This example considers the so-called rapid drawdown of the upstream water level, which is the
most critical condition for the stability of upward slope of the dam.
In the area of this example earthquakes occur. The seismic coefficient is A = 0,08.
Figure 6.1 Cross section of the dam
Appendix
2
Date
December 2007
Page
15
This example considers the calculation of the stability for one potential slip surface (indicated in the cross
section in Figure 6.2). The radius (r) equals 60 m.
In engineering practice the calculation has to be repeated for many other potential slip circles to find the
most critical one. The method of calculation is exactly the same. To repeat the calculation for a high num-
ber of circles a computer should be used, as it enables to perform a complete analysis of all possible slip
circles in a short time.
Steady conditions
The first calculation is for constant upstream and downstream water levels. As a result water from the
upstream water reservoir will seep through the dam as a steady flow.
The stability analysis will be done by Fellenius and Bishop methods. First of all, for the chosen slip circle,
the forces working on the soil mass above the slip circle should be determined. The calculation is done for
a 1 m thick part of the dam. The weight of the soil masses is determined by multiplying volumes with unit
weight. The earthquake force is found by multiplying the slice weights with the earthquake coefficient A.
These two forces times their arms relative to the centre determine the driving moment.
1
2
3
4
5
6
7
8
9
R
C
The resisting moment is made up from the shear strength of the soil in the slip circle. As the shear
strength is a function of the effective stresses it is necessary to split the tangential stresses in the slip cir-
cle into effective and pore pressures. Only the effective stresses contribute to the shear strength. To find
out the pore pressures a flow net is constructed. The pore pressures in the slip circle are derived from the
equipotential lines (Figure 6.2).
Next the soil mass above the slip circle is divided in 9 slices with vertical boundaries (Figure 6.2). The
width of the first 3 slices is 11 m; the next four slices are 10 m wide, then follow 7 and 13 m. For each
slice the forces are computed acting on it. For instance in case of slice 3:
W
3
= 42.5 m (average height) 18.5 kN/m3 11 m 1 m = 8648.8 kN
AW
3
(the average horizontal earthquake force) = 0.08 x 8648.8 = 691.9 kN
c'l
3
= 13 m (l
3
) 20 kN/m
2
1 m = 260.0 kN
ul
3
= 102.5 kN/m
2
(average pore pressure) 13 m (I
n
) 1 m = 2665 kN
In slice 1 the pore pressure is acting only along a part of the slip circle; this part is 20 m. Therefore: ul
1
=
35 kN/m
2
(average pore pressure) 20 m 1 m =700 kN
Figure 6.2 Slip circle divided in 10 slices
Appendix
2
Date
December 2007
Page
16
Table 2 gives the calculated magnitude of the forces acting on each side. The direction of some the
forces is indicated in Figure 6.3.
slice no. W
[kN]
AW
[kN]
ul
[kN]
c'l
[kN]
1 3968 317 700 660
2 7733 619 2475 300
3 8649 692 2665 260
4 7420 594 2420 330
5 6433 515 2310 330
6 5199 416 1850 300
7 3621 290 1540 330
8 1129 90 680 240
9 761 61 390 420
Fellenius Method:
According to Fellenius:

+
=
o
o
sin
) cos ( ' tan '
W
ul W lc
F
(6.1)
The angle o is between the base of a slice and a horizontal plane. Because also earthquake forces are act-
ing, the above formula becomes:
)) (cos sin (
) cos ( ' tan '
o o
o
+
+
=

AW W
ul W c l
F
(6.2)
where r is the half of the average height of a slice.
Table 2 Forces on each slide
Appendix
2
Date
December 2007
Page
17
N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

AW
W
r
N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

N
'
u
l
N
E
1
E
2
X
1
X
2
W
o
o
T

AW
W
r
Table 3 gives the result of the calculation according to formula (6.2) for the slip circle of this example
with radius 60 m.
I II III IV V VI VII VIII IX X XI XII XIII XIV XV
no. sin cos r
'
]
tan|
'
W
[kN]
AW
[kN]
ul
[kN]
c'l
[kN]
tan
'*(W*
cos -ul)
W* sin AW*
(cos - )
1 68.2 0.93 0.37 20.0 0.33 33 0.65 3968 317 700 660 502.4 3684 12.1
2 45 0.71 0.71 19.0 0.32 33 0.65 7733 619 2475 300 1943.7 5468 241.7
3 32 0.53 0.85 21.2 0.35 33 0.65 8649 692 2665 260 3032.6 4583 341.8
4 21.8 0.37 0.93 20.0 0.33 28 0.53 7420 594 2420 330 2376.4 2756 353.5
5 11.3 0.20 0.98 17.2 0.29 28 0.53 6433 515 2310 330 2125.9 1261 357.0
6 3.4 0.06 1.00 13.8 0.23 28 0.53 5199 416 1850 300 1775.8 308 319.1
7 -5.7 -0.10 1.00 9.6 0.16 28 0.53 3621 290 1540 330 1097.0 -360 242.0
8 -16.7 -0.29 0.96 4.3 0.07 28 0.53 1129 90 680 240 213.4 -324 79.8
9 -24.2 -0.41 0.91 2.00 0.03 28 0.53 761 61 390 420 161.7 -312 53.6
sum 3170 13229 17064 2001
The numerator of the Fellenius formula becomes: 3170+132299 = 16399
The denominator is: 17064+2001 = 19064
Hence the safety factor is: F = 16399/19064 = 0,86
Bishop Method:
According to Bishop:
Figure 6.3 Forces on a slice including an earthquake force
Table 3 Fellenius
Appendix
2
Date
December 2007
Page
18
)) (cos sin (
' tan tan
1 cos
' tan ) ( '
o o
| o
o
|
+
|
|
|
|
.
|
\
|
|
.
|
\
|
+
+
=
AW W
F
ub W b c
F
(6.3)
With Bishop we now see the factor of safety F appearing on both sides of the equation. This implicit ex-
pression can be solved iterative by successive approximations. However the convergence is rapid. And is
shown in the following table. For convenience the factor
)
' tan tan
1 ( cos
F
m
| o
o
o
+ =
is introduced.
As example, the Bishop safety factor is calculated by iteration for the same circle with radius 60 m as in
the previous case with Fellenius. The process is shown in the table below. The columns I to X are the
same as in the Fellenius case. Columns XI and XII show the quantities ub and c'b instead of ul and c'l as
for Fellenius these are calculated by using the relation c'b=c'l*cos.
Appendix
2
Date
December 2007
Page
19
I II III IV V VI VII VIII IX X XI
no. sin cos r
'
] tan '
W
[kN]
AW
[kN]
ul
[kN]
1 68.2 0.93 0.37 20.0 0.33 33 0.65 3968 317 700
2 45 0.71 0.71 19.0 0.32 33 0.65 7733 619 2475
3 32 0.53 0.85 21.2 0.35 33 0.65 8649 692 2665
4 21.8 0.37 0.93 20.0 0.33 28 0.53 7420 594 2420
5 11.3 0.20 0.98 17.2 0.29 28 0.53 6433 515 2310
6 3.4 0.06 1.00 13.8 0.23 28 0.53 5199 416 1850
7 -5.7 -0.10 1.00 9.6 0.16 28 0.53 3621 290 1540
8 -16.7 -0.29 0.96 4.3 0.07 28 0.53 1129 90 680
9 -24.2 -0.41 0.91 2.00 0.03 28 0.53 761 61 390
sum
The table is continued below:
XII XIII XIV XV XVI XVII XVIII XIX XX XXI XXII
F
input
=0.9 F
input
=1.02 F
input
=1.05 F
input
=1.06 (W-ub)*
tan'
W*
sin
AW*
(cos - )
m
M
r
/r m
M
r
/r m
M
r
/r m
M
r
/r
2408.0 3684.2 12.1 1.04 2548 0.96 2761 0.94 2810 0.94 2821
3885.3 5468.1 241.7 1.22 3366 1.16 3545 1.14 3584 1.14 3592
4149.0 4583.3 341.8 1.23 3551 1.18 3689 1.17 3719 1.17 3725
2750.6 2755.5 353.5 1.15 2663 1.12 2726 1.12 2739 1.11 2742
2216.0 1260.5 357.0 1.10 2316 1.08 2346 1.08 2352 1.08 2354
1782.4 308.3 319.1 1.03 2015 1.03 2023 1.03 2025 1.03 2025
1110.5 -359.6 242.0 0.94 1537 0.94 1525 0.94 1523 0.95 1522
254.0 -324.4 79.8 0.79 614 0.81 598 0.81 595 0.81 595
215.5 -312.0 53.6 0.67 893 0.70 856 0.71 849 0.71 847
sum 17064 2001 19504 20070 20197 20224
Results: F
output
=1.02 F
output
=1.05 F
output
=1.06 F
output
=1.06
with
The sum of M
r
/r, (the resisting moment divided by the circle radius), is the numerator in the Bishop for-
mula which is calculated using the introduced quantity m
.
The denominator in the Bishop formula is the same as in Fellenius: = 17064+2001 = 19064.
To calculate the safety factor, as a first guess, a safety factor of F
input
= 0.9 is used. The resulting m
u
and
Mr/r quantities are given in columns XV and XVI of the table. The sum of the Mr/r is 19504 divided by
the denominator 19064 gives a safety factor of F
output
=1.02. Using this safety factor as a new estimate
(F
input
) again the safety factor is calculated from columns XVII and XVIII in the table. Continuing the
iteration, the safety factor resulting from column XXII is 20224/19064 = 1.06 which rounded to two digits
is the same as the input used to calculate columns XXI and XXII. Hence the iteration is stopped and the
safety factor is 1.06.
Table 4 Bishop
Appendix
2
Date
December 2007
Page
20
Comparison of the methods
In these analytical method the same elements prevail:
s
, c', |' and u have to be determined with the same
degree of accuracy. In Fellenius' method the sum of horizontal forces (E
1
and E
2
) is taken to be zero. This
introduces an element of simplification, which must lower the accuracy. Fellenius' method therefore is not
recommended. Bishop's method takes the difference of horizontal forces (E
1
and E
2
) into account and
therefore arrives at a higher accuracy. Bishop's method has become the most used, together with similar
methods, like those of Morgenstern and of Janbu.
Rapid drawdown condition
When at the upstream side of a dam the water table draws down rapidly, the stability analysis can be done
in exactly the same way. Of course, the numerical values of the various dimensions and forces will be
different. Most important are the pore pressures. As the water table upstream will arrive at new, lower,
positions, it will take time for the water inside the dam to accommodate for this new situation Eventually
all the water will have left the dam, leaving at both sides now. In between the pattern of water flow will
change with time. There is a non steady situation. It will depend on the speed with which the upstream
water level will drop what flow conditions will occur. The higher this speed the more chance of a loss of
stability.
To find out the most critical situation of pore pressures is an intricate job. As explained now a days these
situation can be numerical computer codes are available. In Appendix A there is also graphical quasi-
static approach possible, in which the continuous drop of the upstream water level is replaced by a num-
ber of static water levels separated by equal time intervals. For each time interval a flow net can be con-
structed; the various flow nets can be placed in a sequence, just as static pictures make up a film. In this
case it is assumed that the critical situation occurs when downstream the phreatic line has not changed and
when upstream it touches the lower end of the dam slope. This situation is drawn in Figure 6.4.
In the same figure the pattern of flow lines and a equipotential lines is drawn. This enables for a chosen
slip circle to calculate the pore pressures in this slip circle. It is then possible to calculate the factor of
safety against loss of stability.
Figure 6.4 Flow net with an example of resulting groundwater pressures along a slip circle at rapid drawdown conditions

Retaining Structures and Stability of Soilmass

Uploaded by

Copyright:

Available Formats

Retaining Structures and Stability of Soilmass

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Retaining Structures and Stability of Soilmass

Uploaded by

Copyright:

Available Formats

Project number

The additional length A t

steel sheet pile wall: F

concrete sheet pile wall: F

wooden sheet pile wall: F

take and calculate: x = l; (4.13)

The field moment is maximal where the shear force Q = 0.

take and calculate: x = l; (4.17)

calculate t = u + 1.3 x; (4.18)

the field moment is maximal where the shear force Q = 0;

The normal stress on the base is '

5.4 Circular slip surfaces

= = kN ul W 2 . 392 8 . 132 525 ) cos ( o

You might also like