Stability Analysis

Overall Shear Failure

Overall shear failure modes：

Push In

Basal Heave
Overall Shear Failure Mode: PUSH-IN

wall

strut  settlement
failure surface

wall bottom "kick out"

It is mainly due to the unbalance between passive earth

pressure (inside) and active earth pressure (outside)
Overall Shear Failure Mode: BASAL HEAVE

Strut settlement
It is mainly due to
bearing capacity of the bottom heave
soil beneath the
excavation bottom

failure surface
ANALYSIS OF PUSH-IN

strut
Ms

La
Lp
Pa

Pp

pp pa

Mr Pp Lp  Ms
The factor of safety against push-in: Fp  
Md Pa La
Distribution of earth pressures for cohesive soil：

a  v Ka  2cKac (4.16)

cw
K ac  K a (1  ) (4.17)
c

 p  v K p  2 cK pc (4.18)

K pc  K p (1 
cw
) (4.19)
c
 cw   su (5.6)

1.2
Cast-in-place pile
1.0
API
0.8


0.6

0.4

0.2

0.0
0.0 0.2 0.4 0.6 0.8 1.0
su /  v

FIGURE 4.12 Relation between adhesion and undrained shear

strength of clay
 cw  0 . 67 su （Diaphragm walls）

c w  0 . 5 su （Steel sheet piles）

Factor of safety Fp =1.2~1.3

Cohesionless soil（sandy, gravel）

Distribution of water pressures ：
Gross water pressure distribution ？
Net water pressure distribution？
 dj
Strut

Strut

di

uf uf
(a) (b)
FIGURE 5.7 Distribution of water pressure due to seepage (a)
distribution of water pressure (b) net water pressure
(note: u = water pressure due to seepage)
f
Cohesionless soil（sandy, gravel）

Distribution of earth pressures：

Caquot-Kerisel's or Coulomb's active earth pressure should be adopted
for the active earth pressure.
Caquot-Kerisel's passive earth pressure should be adopted for the
passive earth pressure. When  /2 , Coulomb's passive earth
pressure coefficient is quite close.
Caquot-Kerisel's earth pressure theory's K a ,Kp and  have
some relationship. Section 4.5.3 has summarized some findings on
values of  .

Clough's research：concluded that between concrete (cast in steel

mold) and sand,  is about 0.8 .
 1.0
hm

0.8 He
/ H e (%) hm

0.6
hm

0.4

0.2

0.0
0.8 1.0 1.2 1.4 1.6 1.8
Fp
FIGURE 5.8 Factors of safety against push-in for excavations in
sand (all cases are safe cases;     is assumed)
Conclusion：

Assumption that     seems to be reasonable.

For conservative reason, we usually assume
 =0.5  
Fp =1.2~1.3
Strut Ms
Lan
Pan
Lpn

Ppn

(a) (b)

FIGURE 5 . 9 Analysis of- push in by the net pressure method

( a) distribution of net earth pressure
( b) force equilibrium of the retaining wall as a free body
ANALYSIS OF BASAL HEAVE

The analyses of the basal heave failure are only applicable to clayey
soils.
Like Terzaghi, Bjerrum and Eide,
Tschebotarioff, Terzaghi and Peck,
Clough and O'Rourke, etc.

But the most commonly applied of which are Terzaghi's method,

Bjerrum and Eide, and the slip circle method.
(1) Bearing capacity method
B B1 B1

d d

He
45 o a b c a b c

B1 B1

(a) (b)

B1  B / 2 B1 = B / 2

d d

a b c a b c
45 o

(c) (d)
FIGURE 5.10 Analysis of push-in by bearing capacity method
(a) a B1 wide trial failure surface
(b) a second B1 wide trial failure surface
(c) a third B1 wide trial failure surface
(d) both sides of the excavation produce failure surfaces
He

2.5m  sat  19.62 kN/ m3

FIGURE 5.11 Excavation profile of the assumed excavation case

 3
Bearing capacity method
Negative bearing capacity method
Slip circle method (side strength neglected)
2 Slip circle method (side strength considered)
Fb

0
0 1 2 3 4
X H e

FIGURE 5.12 Relations between failure circle sizes and factors of safety against basal
heave obtained by the bearing capacity method, negative bearing capacity
method, and the slip circle method
( Su  25 KN / m2 )
 3

2
Fb

1 Bearing capacity method

Negative bearing capacity method
Slip circle method (side strength neglected)
Slip circle method (side strength considered)
0
0 1 2 3 4
X H e

FIGURE 5 . 13 Relations between failure circle sizes and factors of safety

against basal heave obtained by the bearing capacity method ,
negative bearing capacity method , and the slip circle method

( Su / v  0.3)
Terzaghi's method

qs qs

B/ 2 B B/ 2 c
su1 He
a b
45o
su2 B/ 2
D
d

Stiff soil

FIGURE 5.14 Analysis of basal heave using Terzaghi's method

(a ) D  B / 2
When D  B / 2 , the formation of a failure surface is not
restrained by the stiff soil.
B
W  ( He  qs )( B1 × 1)  ( He  qs ) (5.7)
2
B
Qu  5 . 7 su 2 ( B1 × 1)  ( 5 . 7 su 2 ) (5.8)
2
Vertical plane bc can offer shear resistance s u1
He and the
factor of safety against basal heave will be：

Qu 5.7 su 2 B / 2
Fb   (5.9)
W  su1 H e (H e  qs ) B / 2  su1 H e
 qs qs

B
s u1 D D He

su 2 D
D
Stiff soil
(b ) D  B / 2

FIGURE 5.14 Analysis of basal heave using Terzaghi's method

(b) D  B / 2
When D  B / 2 , the failure surface will be restrained by the stiff soil.

Qu 5.7 su 2 D
Fb   (5.10)
W  su1 H e (H e  qs ) D  su1 H e

Clough suggested that, Terzaghi's factor of safety ( F )should be

b

greater than or equal to 1.5.

 B B1 B
d d

He He
e
a b c
a b 90 °
90 °

(b)
(a)

FIGURE 5.15 Relation between the embedded part of the retaining wall
and the failure surface
(a) large penetration depth (b) small penetration depth
 (2) Negative bearing capacity method
qs qs qs qs
B

H e 2 B1 He
2 B1

Assumed failure surface

B1 B1

(a) (b)

qs qs

He

(c)

FIGURE 5.16 Analysis of basal heave by negative bearing capacity method (a) a 2B wide failur 1

surface (b) another 2B wide failure surface (c) Failure surface covers the whole
1

excavation bottom
He

2.5m  sat  19.62 kN/ m3

FIGURE 5.11 Excavation profile of the assumed excavation case

 3
Bearing capacity method
Negative bearing capacity method
Slip circle method (side strength neglected)
2 Slip circle method (side strength considered)
Fb

0
0 1 2 3 4
X H e

FIGURE 5.12 Relations between failure circle sizes and factors of safety against basal
heave obtained by the bearing capacity method, negative bearing capacity
method, and the slip circle method
( Su  25 KN / m2 )
 3

2
Fb

1 Bearing capacity method

Negative bearing capacity method
Slip circle method (side strength neglected)
Slip circle method (side strength considered)
0
0 1 2 3 4
X H e

FIGURE 5 . 13 Relations between failure circle sizes and factors of safety

against basal heave obtained by the bearing capacity method ,
negative bearing capacity method , and the slip circle method
( Su / v  0.3)
 q q
s s

He

FIGURE 5.16c

Bjerrum and Eide's method

N c × su
Fb  (5.12)
 × H e  qs
 9
L B 1
8
2
3
7
Nc 
6

4
0 1 2 3 4 5
He B

FIGURE 5.17 Skempton's bearing capacity factor

Nc (rectangular) = Nc (square) B
(0.84  0.16 ) (5.11)
L
Modified Bjerrum and Eide's method：

su 1 N c , s f d f s
Fb  (5.13)
H e

B
fs  1  0.2 (5.14)
L
 undrained shear strength profile
B s u 2 su 1  1 s u 2 su 1  1

He

1 D s u1 10
0.2
2 s u2 9
D /B

N c,s
8
0.3
7
s u2 s u1
0.4
0.2 0.4 0.6 0.8 0.5 6
1.0
5.53 5.53
1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
5
s u 2 s u1
4
D /B
3
N c,s

0 (a) N c,s for failure circles passing two soil layers

 N c,s
50
40

30
If s u2 / su1 exceeds the values in the figure, the failure circle
will be tangent to the top of the lower soil layer.
20
2.5 3.0

10
2.0
1.45
1.2 1.25
5
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0
D B

(b) N c,s for failure circles tangent to the top of the lower soil layer
 f d
1.5

1.4

1.3

1.2

1.1

1.0
0 1 2 3 4 5 6
H e / B

(c)

FIGURE 5.18 Extended Bjerrum and Eide's method

(a) N c,s for failure circles passing two soil layers
(b) N c,s for failure circles tangent to the top of the lower soil layer
(c) width modification factor f d

DM7.2 suggested that Bjerrum and Eide's factor of safety Fb

should be greater than or equal to 1.5.
 (3) Slip circle method

A
W

Lowest level of struts

O r

su

FIGURE 5.19 Location of the center of a failure circle for the slip circle method
 e c

W
Lowest level of struts
Ms
O O
b  
a s s
d f X N
d
X
N
su N

g N su
N
N N
(a)
(b)
FIGURE 5.20 Analysis of basal heave by the slip circle method
(a) the failure surface (b) balance of the a free body



2
X s u ( X d )  M s
Fb 
Mr
 0

X
(5.15)
Md
W ×
2
 3
Bearing capacity method
Negative bearing capacity method
Slip circle method (side strength neglected)
2 Slip circle method (side strength considered)
Fb

0
0 1 2 3 4
X H e

FIGURE 5.12 Relations between failure circle sizes and factors of safety against basal
heave obtained by the bearing capacity method, negative bearing capacity
method, and the slip circle method
( Su  25KN / m )
2
 3

2
Fb

1 Bearing capacity method

Negative bearing capacity method
Slip circle method (side strength neglected)
Slip circle method (side strength considered)
0
0 1 2 3 4
X H e

( S u /  v  0.3)
 Lowest level
of struts

FIGURE 5.21 Factor of safety increasing due to the failure circle exceeding
the excavation width
Lowest level of struts

s u1

s u2
S u1  S u2

FIGURE 5.22 Analysis of basal heave in layered soft soils

 3
Bearing capacity method
Negative bearing capacity method
Slip circle method (side strength neglected)
2 Slip circle method (side strength considered)
Fb

0
0 1 2 3 4
X H e

FIGURE 5.12 Relations between failure circle sizes and factors of safety against basal
heave obtained by the bearing capacity method, negative bearing capacity
method, and the slip circle method
( Su  25KN / m )
2
 3

2
Fb

1 Bearing capacity method

Negative bearing capacity method
Slip circle method (side strength neglected)
Slip circle method (side strength considered)
0
0 1 2 3 4
X H e

FIGURE 5 . 13 Relations between failure circle sizes and factors of safety

against basal heave obtained by the bearing capacity method ,
negative bearing capacity method , and the slip circle method
( Su / v  0.3)
(5) Applicability to sandy soils

qs qs

B c
B/ 2 B/ 2
su1 He
a b
45o
su2 B/ 2
D
d

Stiff soil
(a) D  B / 2
 qs qs

B
D D He
su1

su 2 D
D
Stiff soil
(b ) D  B / 2

FIGURE 5.14 Analysis of basal heave using Terzaghi's method

 e c

W
Lowest level of struts
Ms
O O
b  
a s s
d f X N
d
X
N
su N

g N su
N
N N
(a)
(b)

FIGURE 5.20 Analysis of basal heave by the slip circle method

(a) the failure surface (b) balance of the a free body
北投自強路 excavation failure case
 5.5.3 Case Study of Overall Shear Failure

The excavation case was located in Taipei. The width of the

excavation was 17.6m； the length was 100.1m ；the depth was
13.45m. The excavation adopted a 70cm thick , 34, deep
diaphragm wall as the retaining wall. There four levels of struts
and the excavation was carried out in 5 stages.
 力霸百老匯
excavation failure
力霸百老匯 excavation failure case
 14.7 kN / m 2

t  20.3kN / m3
SM GL-2.8 m
c 0    33
GL-4.5 m
ML - OL t  15.6kN / m3
cu  0.98kN / m2
u  10
SM - SP GL-8.7 m
t  19.7 kN / m3
GL-10.15 m c   0    32


¡E
GL-10.7 m
 CL t  18.8kN / m3
cu  16.7 kN / m2
GL-13.45 m
u  18

GL-24.0 m
FIGURE 5.23 Stability analysis of an excavation case history
(a) excavation and geological profiles
 cu and uwere the total stress strength parameters of the clay soils,
obtained from the triaxial CU test

adopt by the original designer Fp will be 1.5

Fb will be 2.3
We assume the soil below the lowest level of struts (GL-10.15 m)
to be a clayey layer, the adhesion between the retraining wall and
the soil cw  2su / 3
and the normalized undrained shear strength su / v  0.22

a  v Ka  2cKac (4.16)
 p  v K p  2cK pc (4.18)
At the depth of GL-10.15 m
v  20.3 × 4.5 15.6 × 4.2 19.7 × 1.45  185.5kN / m2
u  ( 10 . 15  2 . 8 ) × 9 . 81  72 . 1 kN / m 2
 v   v  u  113 . 3 kN / m 2

su  0 . 22  v  0 . 22 113 . 3  24 . 9 kN / m
2
×

cw 2
h , a  v Ka  2 su Ka (1   
) 185.5 2( 24.9) 1   121.2 kN / m2
su 3
At the depth of GL-13.45 m
Before excavation─
v  185.5 19.7 × 0.55 18.8 × 2.75  248.0kN / m2
u  (13.45  2.8) × 9.81  104.5kN / m2
 v   v  u  143 . 5 kN / m 2
su  0.22 v  0.22 × 143.5  31.6 kN / m2
after excavation was started, v  0 on the passive side, but su
value stayed unchanged.
Thus,
cw 2
h, p  v K p  2su K p (1  )  0  2(31.6) 1   81.5kN / m2
su 3
At the depth of GL-24.0 m
The active side─
v  248.0 18.8 × 10.55  446.3kN / m2
u  ( 24  2.8) × 9.81  208.0 kN / m2
v  v  u  238 . 3 kN / m2
su  0.22 v  0.22 × 238.3  52.4kN / m2

h,a  v Ka 2su Ka (1  w ) 446.3 2(52.4) 1  2 311.0kN / m2

c
su 3
At the depth of GL-24.0 m

The passive side─

After excavation was start su stayed constant,
v  18 . 8 × ( 24 . 0  13 . 45 )  198 . 3 kN / m 2
cw 2
h, p  v K p  2su K p (1   
) 198.3 2(52.4) 1   333.6kN / m2
su 3
 GL-10.15 m
121.2 kN m 2

81.5 kN m2
GL-13.45 m

333.6 kN m2 GL-24.0 m 311.0kN m2

(b)
FIGURE 5.23 Stability analysis of an excavation case history
(b) distribution of earth pressure for the push-in analysis

The factor of safety against push-in as

81.5 ×10.55 ×(10.55/2 3.3) (333.6 81.5) ×10.55 ×0.5 ×(10.55 ×2/33.3)
Fp  0.89
121.2 ×13.85 ×13.85/2 (311.0 121.2) ×13.85 ×0.5 ×13.85 ×2/3
Compute the factor of safety against basal heave
according to Slip circle method ：
Similarly, assuming the soil below the lowest level of
struts is clay, the average value of the undrained shear
strengths (the active side) of the soil between GL-10.15 m
and GL-24.0 m would be
24.9  52.4
su ,a   38.7kN / m2
2

The average value of the undrained shear strengths of

the soil between GL-13.45 m and GL-24.0 m would be
31.6  52.4
su , p   42.0kN / m2
2
The radius of the failure circular arc would be
24  10 . 15  13. 85 m
The central angle of the failure circular arc on the active side
would be

  1.57
2
The central angle of the failure circular arc on the passive
side would be
  cos 1 ( 3.3 )  1. 33
13.85

The factor of safety against circular arc failure would be

13.85 × 1.33 × 42.0 ×13.85 13.85× 1.57 × 38.7 × 13.85 22370
Fb    0.94
v ( GL 13.45) × 13.85 ×13.85 / 2 23786
Computing the factor of safety against basal heave
following Terzaghi's method：
The width of the excavation B 17.6m , B / 2 was larger than the
penetration depth (10.55 m). Assumed failure surface will pass below the
bottom of the retaining wall.

B
d

He
a b c
90°

(b)

FIGURE 5.15 Relation between the embedded part of the retaining wall and the failure surface
(b) small penetration depth
The average undrained shear strength of soil within the range of
the failure circle can be calculated as follows：

of soil 13 .45  B / 2  25 .9 m deep below the ground surface--

v  248.0 18.8 × ( 25.90 13.45)  482.1kN / m2

v  v  u  482.1  ( 25.90  2.8) × 9.81  255.5kN / m2

su  0 . 22 v  0 . 22 × 255 .5  56 . 2 kN / m2
 The average undrained shear strength within the range of the
failure circle would be
su  (31.6  56.2) / 2  43.9 kN / m2
As computed earlier, the total stress outside the excavation zone
at the depth equaling the excavation surface would be
v  248.0kN / m2
 To simplify the analysis and be conservative, we assume the
soil above the excavation surface is clay and has soil shear
strength expressed as su / v  0.22 . The average undrained shear
strength of the soil outside the excavation zone and the excavation
surface would be

0.22 v( GL 13.45) 31.6

su    15.8kN / m2
2 2

The factor of safety according to Terzaghi's method would be

Qu 5.7 × 43.9× 17.6 / 2 3118
Fb    1.08

W su1 He 248.0 17.6 / 2 15.8 × 13.45 2874
×
The factor of safety following Bjerrum and Eide's method would be

su N c 43 . 9 × 6 . 2
Fb    1 . 10
H e  qs 248 . 0
Undrained shear strength and the depth：
2
s u ( kN m )

0 50 100 150 200

0

5
depth(m)
10
Active side ( c u ,  u )

15

2
0

25
Passive side ( c u ,  u )
s u /  v  0.22

(c)
FIGURE 5.23 Stability analysis of an excavation case history
(c) the undrained shear strength used in the analysis
【Example 5.1】Assume a 9.0 m deep excavation in a sandy
ground and the lowest level of struts is 2.5m above the
excavation surface. The level of groundwater outside the
excavation zone is ground surface high while that within the
excavation zone is as high as the excavation surface. The
unit weight of saturated sandy soils sat  20 kN / m, the
3

effective cohesion c   0 and the effective angle of friction

   30. Because of the difference between the levels of
groundwater , seepage will occur. Assume that the friction
angles ( ) between the retaining wall and soil on both the
active and passive sides are 0.5  and the factor of safety
against push-in, F 1.5 . Compute the required penetration
p

depth( H p ).
 d j
a
H e z
x

di
b ub
z
H p x

uc c

(H p  d i ) w (H p  H e  d j ) w

(a) (b)

FIGURE 4.22 Simplified analysis method for seepage

(a) distribution of water pressure (b) net water pressure
【Solution】
1.determine the coefficient of the earth presure

Compute both the active and passive earth pressures following

Caquot-Kerisel's earth pressure theory. When   0.5 , the
coefficients of active and passive earth pressure can be found from
Figure 4.9 and Figure 4.10 to be 0.3 and 4.6 separately. Thus, the
coefficients of the horizontal active and passive earth pressure
would be
Ka , h  0.3 cos   0.3 cos 0.5   0.29

K p,h  4.6 cos   4.6 cos 0.5   4.4

2. Compute the effective active active earth pressure on the wall
At the lowest level of strut (z=6.5m, x=6.5m) --
 a, h   v K a , h v  20 × 6.5 130 kN/m2
According to Eq. 4.51the porewater pressure at x away from upstream
water level would be

2 x ( H p  di )  w 2 × 6 . 5 × H p × 9 . 81 63 . 77 H p
u   
2 H p  H e  di  d j 2Hp 9 H p  4.5

63 . 77 H p 18 . 49 H p
a, h  (   u ) K a , h  ( 130  ) × 0 . 29  37 . 7 
H P  4.5 H p  4.5
At the bottom of the retaining wall( z  9  H p , x  9  H p )─
v  20 × ( 9  H p )  180  20 H p

2 x(H p  di ) w 2 × (9  H p ) × H p × 9.81 9.81H p 88.29H p

2

u  
2H p  He  di  d j 2H p  9 H p  4.5

9 . 81 H p
2
 88 . 29 H p
a, h  ( 180  20 H p  ) × 0 . 29
H p  4.5
2 . 84 H p  25 . 60 H p
2
 ( 52 . 2  5 . 8 H p  )
H p  4.5
3. Compute the lateral effective passive earth pressure on the wall
At the bottom of the retaining wall（ z  H p ）─

v  20 × H p  20 H p

9.81H p  88.29 H p
2

u
H p  4.5

9.81H p  88.29H p
2

p,h  (20H p  ) × 4.4

H p  4.5
43.16 H p  388.48H p
2

 88H p 
H p  4.5
4. Compute the maximum net water pressure (at the
excavation surface)
According to Eq. 4.53, the maximum net water pressure
would be

2(He di dj )(Hp di )w 2× 9× Hp × 9.81 88.29Hp

ub   
2Hp He di dj 2Hp 9 Hp 4.5
5. The effective earth pressure on both sides of the wall and the
distribution of the net water pressure are as shown in Figure
5.24
 a, h u net

Lowest level 18.49H p

9 37.7 
of struts o H p  4.5 6 .5u b
9
2.5  p,h ub
b

H p

2.84H p  25.60H p
2
43. 16 H p  388. 48 H p
2

88 H p  52.2  5.8H p 
H p  4.5 H p  4.5

FIGURE 5.24 Distribution of lateral earth pressure

 6. Compute the driving moment ( Md ) and the resistant moment
( M r) for the free body below the lowest level of struts

Md  Pa,h La
18.49Hp × (Hp 2.5)2 2.84Hp 7.11Hp × 2(Hp 2.5)2
2

 (37.7  ) (14.5 5.8Hp  )


Hp 4.5 2 
Hp 4.5 2 ×3
ub Hp Hp 6.5u 2.52
2.5u 2 × 2.52
 × (2.5  ) b ×
 b ×

2 3 9 2 2 9 2 ×3
0.95Hp 11.62Hp
 (23.68 1.93Hp  )(Hp 2.5)2 [0.17Hp 1.25Hp 2.84]ub
2

Hp 4.5
Mr  Pp,h Lp
43.16H p 388.48H p H p
2
2H p
 (88H p  )× × (2.5  )
H p  4.5 2 3
 3
21.58H p 194.24H p
2
2H p
 (44H p 
2
)(2.5  )
H p  4.5 3
7. determine the penetration depth H p
Mr
Fp   1. 5
Md

Then we have H p  7. 25 m
【Example 5.2】An excavation in clay goes 9.0m in to the
ground ( He  9.0m) . The groundwater outside the
excavation zone is at the ground surface level while
that within the excavation zone is at the level of the
excavation surface. sat =17.0 kN / m3. The
undrained shear strength su =45 kN / m2 .
Suppose the excavation width B =10m and the
excavation length L =30m. Compute the factor of
safety against basal heave according to Terzaghi's
method and Bjerrum and Eide's method,
respectively.
【Solution】
In this example, the surcharge qs  0
According to Terzahi's method,
1 5.7 ×su 1 5.7 × 45
Fb  ×   2.7
H e   su 9  45
17
0.7 B 0.7 ×10
According to Bjerrum and Eide's method,

L 30
  3.0
B 10
He 9
  0. 9
B 10
According to Figure 5.17, we have
Nc  7.1
su Nc 45 × 7.1
Fb    2.09
t He 17 × 9
 Upheaval

 t1
This image cannot currently be display ed.

h1

h2  t2 Impermeable layer

Water pressure H Permeable layer

w

FIGURE 5.31 Analysis of upheaval

  ti ×hi
Fup  i (5.17)
H w × w

The factor of safety against upheaval Fup should be larger than or equal to 1.2
 Case Study

  ti ×hi
 ti  hi
Upheaval failure: Fup  i
Fup  i
H w × w Hw  w
Sand Boiling --- Factor of safety and failure mechanism

h hz
H2  u 
H1 H1 w H1 w
A 0
z
C H2
Saturated soil
B

H1 w  H 2 sat ( H1  H 2  h) w H 2    h w

Upward water flow

FIGURE 5.32 Total stresses, effective stresses, and change of porewater pressure in
sandy soils acted on by an upward water flow
Besides, according to the phase relationship of soil, the
submerged unit weight is
 Gs 1 

   w (5.24)

1 e 

The critical hydraulic gradient is then

Gs 1 (5.25)
icr 
1 e

icr
Fs  (5.26)
imax(exit )
 Wall

Sand boiling zone

H w

Hp

Impermeable layer

FIGURE 5.33 Seepage in soil below sheetpiles

icr
Fs  (5.26)
imax(exit )
 Terzaghi's method：

1 2

U =(the volume of the soil column) (iav w )
×  H p iavg w (5.27)
2
1 2 1 2 
W   
H p ( sat   w )
 Hp  (5.28)
2 2

The factor of safety is

W   
Fs   (5.29)
U i avg  w
Provided the computed factor of safety is too small, we can
consider placing filters at the exits of seepage. Assuming the
weight of filters is Q, the factor of safety will be

W Q
Fs  (5.30)
U

In general, the required Fs for the above equation should

be greater than of equal to 1.5
 Marsland's method：
DM7.1 suggested that the reasonable factor of safety against piping
in an excavation be around 1.5 to 2.0.
2.0
B
Hw
1.5
Retaining wall HP
H p / Hw

2.0
1.0
1.5
2.0
1.5
0.5
1.0
Loose sand 1.0
Dense sand Fs
0
0.5 1.0 1.5 2.0
B /(2H w )
(a)

FIGURE 5.34 Relations between wall penetration depths and factors of safety against sand boiling
(a) dense and loose sands with the impermeable layer located at the infinite depth
 2.0

B
Hw
1.5
H p / Hw

HP
D
1.0
Impermeable layer

2.0
0.5 2.0
1.5
1.5
D Hw  2 1.0
D Hw  1 Fs 1.0
0
0.5 1.0 1.5 2.0
B /( 2 H w )

(b)

FIGURE 5.34 Relations between wall penetration depths and factors of safety against sand boiling
(b) dense sand with the impermeable layer located at a finite depth
One dimension seepage method：

a d j

He
H w

di
d Sandy soil
H p

c b

FIGURE 5.35 Analysis of sand boiling

If we assume the datum is at the downstream level, the total
head at the elevation of downstream (point d) will be

ht ,d  he  hp  0  0  0 (5.31)

The total head at the upstream elevation (point a) will be

ht ,a  he  hp  He  di  d j  0  He  di  d j (5.32)

The difference of the total heads between upstream and

downstream levels will be
H w  ht ,a  ht ,d  He  di  d j (5.33)
 Suppose the seepage is one dimensional and the hydraulic
gradients for each depth along the flow path abcd are equal. the
hydraulic gradient will be

Hw Hw
iavg  
He  d j  di  2( H p  di ) He  2H p  di  d j

The factor of safety against boiling will be

ic  ( He  2 H p  di  d j )
Fs  
iavg w Hw

The required Fs for the above equation should be greater than

or equal to 1.5.
Case Study
23.25m
EL.+104.5m
Sungshan VI formation CL
EL.+98.0m EL.+88.0m
Groundwater level EL.+98.0m
EL.+95.0m
SM
Sungshan V formation
EL.+87.2m
Sand boiling
CL
Sungshan IV formation
EL.+81.8m
. . . . .. . . . . .

...
.

..

..
. . .

..
.

..
..
.
.

.
EL.+80.0m
SM
EL.+72.5m
Sungshan III formation CL
EL.+69.4m
SM
EL.+63.4m
Sungshan II formation CL
EL.+59.6m
Sungshan I formation SM
EL.+56.4m

Chingmei formation

(a)

FIGURE 5.36 Excavation of Siemen Station of Taipei Rapid Transit System

(a) excavation and geological profiles
 N

23.25m

Diaphragm wall

6.2m
bore


5.0m

sump

(b)

FIGURE 5.36 Excavation of Siemen Station of Taipei Rapid Transit System

(b) plan view
 Casing

5m 6m
9m 8m
. . .. . . . . .

..

..
...

..
.

..

.
. . . . . .. . .
. . .

..
.

.
.

..
.

.
EL.+80.0m 4m
SM
EL.+72.5m Sump
Sungshan III CL
EL.+69.4m
SM
EL.+63.4m
Sungshan II CL
EL.+59.6m
Sungshan I SM
EL.+56.4m

(c)

FIGURE 5.36 Excavation of Siemen Station of Taipei Rapid Transit System

(c) process of sand boiling
TABLE 5.2 Stability analysis methods for strutted walls and the required minimum factors of safety

Overall shear failure

Sand boiling Upheaval
Push-in Basal heave
Harza’s method {Eq. 5.26,
Fs 2.0}(4)
Sand or grave

Terzaghi’s method {Eq. 5.30

Gross pressure method
ˍˍ Fs 1.5}(1, 5)
{Eq.5.5, Fp  1.2, Marsland’s method {Fig. 5.34, ˍˍ
F  1.5~2.0}
(4)
assuming Ms=0}(1) s
Simplified 1-D seepage
method {Eq. 5.35,
Fs  2.0 }(1, 5)

Terzaghi’s method

Gross pressure method

{Eq. 5.5, Fp  1.2｝(1) Bjerrum and Eide’s
ˍˍ
Clay

ˍˍ

assuming Ms=0}(1) Slip circle method {Eq. 5.15,

Fb  1.2 }(1)

Terzaghi’s method {Eq. 5.9 or Short term behaviors can

Alternated layers of sand

Gross pressure method 5.10, Fb  1 . 5}(1,2,3) be ignored while long

{Eq. 5.5, Fp  1.2 ｝(1)
(or gravel) and clay

term behaviors may need {Eq. 5.17, Fup≧1.2}(1)

Bjerrum and Eide’s method{Eq.
5.12 or 5.13, Fb 1.2 }(4,3)
consideration. The
assuming Ms=0}(1) analysis methods are the
same as those for sand
Slip circle method {Eq. 5.15,
and gravel.
Fb  1.2}(1,3)
NOTE:

(1) The methods and factors of safety are suggested by TGS (2001)
and JSA (1988)

(2) The factor of safety is suggested by Mana and Clough (1981)

(3) It is only when clay is the dominant soil layer that the analysis of
basal heave is required

(4) The factor of safety is suggested by NAVFAC DM 7.1 (1982)

(5) TGS (2001) and JSA (1988) suggest the conservative value
obtained by Terzaghi's method or the simplified 1-D seepage be
adopted for design.