Basic Design Double Track
Basic Design Double Track
Basic Design Double Track
150
No
Load
Train
Load
300
Train
Train
No load
No load
Load
Load
Horizontal pile cap force (kN/m)
0.00
0.00
0.00
0.00
600
No load
Train
Load
0.00
0.00
0.00
0.00
-1
0.05
0.44
0.07
0.40
0.10
0.22
0.14
0.11
-2
0.14
1.18
0.23
1.67
0.32
1.72
0.42
1.66
-3
0.27
2.10
0.52
3.56
0.73
4.55
0.95
5.59
-4
0.37
2.87
0.71
5.20
1.01
6.80
1.34
8.37
-5
-6
-7
0.40
0.41
0.41
3.34
3.54
3.55
0.76
0.77
0.76
6.05
6.21
6.17
1.08
1.10
1.09
7.55
7.51
7.41
1.47
1.51
1.48
8.98
8.82
8.57
-8
0.40
3.54
0.74
6.11
1.04
7.27
1.40
8.28
-9
0.38
3.50
0.68
5.94
0.92
6.91
Table 1 Horizontal pile cap forces
1.19
7.60
A cross-section with an embankment height of 2.5 m has been selected for the computations. At this
location, about 5 to 6 m of soft silty clay with an SPT-value of 0 to 4 exists directly beneath the
surface. For all cross-sections the pile rows are spaced 1.00 m apart, centre-to-centre, perpendicular
to the track. A sand blanket with a maximum thickness of 1.0 m will be laid down directly on the soil
surface, if the existing surface soil is too soft for the construction loads. At this location, about a metre
of sand will be applied to enable installation of the piles and to construct the geogrid-reinforced
embankment.
3
The bending moments in the piles have basically been determined in two steps. Firstly, for the crosssection with a height of 2.5 m, varying values of the spring constant are used in the PLAXIS
calculation. The different values of the spring constant will result into a horizontal cap reaction force
(and displacement) at each cap location. Secondly, the extremes in the horizontal behaviour of the pile
head are assumed to be very stiff, very flexible or soft, and the associated moments are determined
using a Winkler model. In the calculations, spring constants of 50, 150, 300, 600 are used to adjust the
horizontal resistance of the pile cap to the stiffness of the soil. A value 50 kN/m equals very soft soil.
Program GROND calculates the forces and displacements in a horizontally loaded pile on the basis of
a Winkler-type system. In this case it is assumed that the pile is horizontally loaded at the pile head by
a horizontal force. The calculations are based on a circular pile with a diametre of 0.15 m and a
7
2
modulus of elasticity of the pile concrete of 2.85 x 10 kN/m . In order to determine the range of the
bending moments that may be expected in the piles, extreme cases have been considered.
Two extreme cases have been used in the calculations. (see table 2)
A situation is considered in which the modulus of subgrade reaction and associated parametres for the
clay layer are expected to relatively low, and in another situation they are expected to be relatively
high. In both cases, the layer thicknesses are equal. It is assumed that the pile has penetrated 1 m
into the foundation layer. No information is available concerning the phreatic water level, so that again
two extremes are assumed.
Layer
Sand
Elevation
Arching
Factor
1)
[m]
38.7 37.7
2.5
30
500
18
35
Mod. of
subgrade
reaction
[kN/m3]
15000
clay +
37.7 32.7
1.5
12
15
2
500
15
22
5
5000
silt
clayey
32.7 31.7
2.5
19
30
0
10000
21
35
0
30000
silt
1) ratio of the maximum horizontal soil pressure on pile and the Coulomb horizontal pressure for the plane strain case
With the use of the parametres in table 2, a force-displacement relationship can be generated for both
the soft and the stiff case. The relationships refer to a single pile without tension cracks. The piles
are spaced centre-to-centre at a distance of 1.15 m in the track direction. Table 3 shows selected
values of the calculated spring constants per metre for the two extreme cases.
Table 3 Calculated foundation spring constants (horizontal)
Soft foundation soil
Stiff foundation soil
Horizontal force on pile
Foundation spring
Horizontal force on pile
Foundation spring
head
constant
head
constant
(kN/m)
(kN/m per metre)
(kN/m)
(kN/m per metre)
1.74
75
2.17
833
3.48
75
4.35
758
5.22
75
6.52
683
6.96
73
8.70
590
8.70
71
10.87
529
Combining the spring constants determined for the extreme cases with the relationship of the
maximum pile cap force versus the value of the spring constant, the results in table 4 are obtained by
manual iteration.
Foundation
Soft
Stiff
Maximum
moment
in pile
kNm
4.5
5.4
In the case of soft foundation soil, the spring constant is more or less constant over the full range of
expected forces. In the case of stiff foundation soil, the constant varies with head force. Therefore,
strictly speaking, the PLAXIS calculation should be performed with a variation in the spring constants,
depending on the pile location and horizontal force. However, in most cases, the variation in
horizontal cap force in the range of x= 4 m to 9 m is not large for most cases. Therefore, it is
expected that such a refined calculation would lead to little or no change in the final result.
4
If the height of the embankment changes, the pile distance also changes. This means that the piles
are not in line at the transition of areas with different pile distances. The load transfer from the geogrid
onto the pile cap is not according to the theoretical design. In the design a rectangular pattern is
chosen, with a standard pile distance perpendicular to the track of 1 m. The forces in the geogrid are
mainly dependent on the arching in the embankment. The higher the embankment, the lower the
forces in the geogrid. Various calculation methods have been developed to determine the arching rate.
The method used is according to BS8006. However, due to the fact that a rectangular pile system is
projected, the calculation method was
Load, ws
slightly modified.
Four different load configurations will be
calculated:
Fill: ,
Serviceability
state
construction
Ultimate limit state
construction
Final serviceability state
Final ultimate limit state
during
p'
c
p'
g
WT
during
Tr p
Tr p
s
Serviceability
limit state
const.
final
kN/m
31
127
24
99
2
66
12
283
2
51
WT =
WT =
28
21
21
22
16
s1.s2 / a =
p'
c / '
v=
Vertical load between pile caps // embankment
13
2
13
2
13
2
32
29
29
25
22
12
220 kN/m
Thickness first layer
0.5
-1.0
m
0.7(s1-a)<H<1.4(s1-a)
H>1.4(s1-a)
16 kN/m
1.0
m
13
2
Thickness first layer
0.6
-1.2
m
0.7(s2-a)<H<1.4(s2-a)
22 kN/m
H>1.4(s2-a)
1.2
m
WT =
WT =
68
50
50
63
47
46
kN/m
68
layer 1
0.5
1.0
m
m
75
55
1.43
171
126
55
69
51
1.43
144
107
51
kN/m
1.43
122
122
layer 1
layer 1
0.5
1.0
m
m
Max.
Max.
171
144
kN/m
kN/m
1.43
144
144
95
85
84
88
79
78
kN/m
95
layer 1
0.6
1.2
m
m
104
93
1.43
237
213
92
86
kN/m
1.43
243
243
97
87
1.43
200
180
1.43
205
205
layer 1
layer 1
0.6
1.2
m
m
Max.
Max.
243
243
kN/m
kN/m
0.4
5
0.4
94
0.4
4
0.4
73
Max.
94
kN/m
175
131
238
238
148
110
195
195
Max.
Max.
238
238
kN/m
kN/m
From the calculation results, it is concluded that the average tensile force in the geogrid perpendicular
to the track is 68 kN/m. Along the track, the average tensile force is 95 kN/m.
According to the PLAXIS calculation, the tensile distributions in the bottom geogrid and the top
geotextile are summarised in table 7. The reinforcement consists of three different layers of geogrid. A
layer geogrid 250/50 along the track and a second layer geogrid 250/50 perpendicular to the track are
placed directly over the pile caps. A layer of 110 kN/m geogrid is placed on top of the 600 mm gravel
layer perpendicular to the track
Table 6 Results PLAXIS calculations perpendicular to embankment
Lower geogrid 250
geogrid 110
Spring constant
Lower
layer
Upper
Layer
Max. force
kN/m
kN/m
kN/m
kN/m
50
150
300
600
33.5
24.3
20.1
14.7
1.3
1.0
0.8
0.5
18.1
12.2
10.1
7.3
Required force
without train load
kN/m
6.9
5.2
4.2
3.2
Figure 5. Force in 250 kN/m geogrid over cross-section of a half embankment using a spring constant of 50 kN/m
90
80
M axim um ten s io n fo rce in g eo g rid F o rtrac 2 5 0 (kN /m )
70
la
h bal
s w it
P la x i
60
s t+ tr
ad
a in lo
w it
x is
P la
50
hou
t ba
+ tr
las t
a in
lo a
40
30
20
10
6 .0
CONCLUSIONS
The BS8006 has been used to determine the design of a piled
embankment with basal reinforcement consisting of a geogrid. A
check of the tensile forces in the geogrid using numerical
analyses indicate that, depending on the embankment height,
the force at the edge of the pile cap may be substantially higher
than that which is determined with the use of the BS8006.
The effect of lateral sliding on the tensile forces can be
determined more accurately using numerical methods which
take into account the proper boundary conditions and the stressstrain behaviour of the embankment material and the geogrid.
For example, an extreme load configuration may result in very
high horizontal forces on the pile cap and can cause large
horizontal displacements of the pile head.
The BS8006 does not directly determine the horizontal forces on
the pile caps although they seem to be of much more importance
than lateral sliding.
REFERENCES
1. British Standard BS8006. (1995) Code of Practice for Strengthened Reinforced Soils and
Other Fills, British Standard Institution, London, 162.
2. Chris Lawson (2001), Basal Reinforced Piled Embankments with Steep Reinforced Side
Slopes, Symposium 2001 on soft ground improvement and geosynthetic applications. AIT,
Bangkok
3. Han, J., Gabr, M.A., (2002) Numerical Analysis of Geosynthetic-Reinforced and PileSupported Earth Platforms over Soft Soil. Journal of Geotechnical and Geo-Environmental
Engineering. 44-53
4. Cortlever, N.G., (2001) Design of Double Track Railway on AuGeo Piling System.
Symposium 2001 on soft ground improvement and geosynthetic applications. AIT, Bangkok