Principles of Geotechnical Engineering

Seepage
Lecture Note
Seepage Through Soil Version-2.1
CE341

MIST
May-June 2022

Course Teacher
Lt Col Mohiuddin Ahmed, PEng

V 2.1
Updated – May-June 2022
Welcome to the Lecture on Seepage
 Ref Books

Topics: Part-1: One Dimensional Flow Part-2: Two Dimensional Flow

1. Seepage Force and Velocity 1. Concept of flow net
2. Stress Due to Flow (Seepage Pressure) 2. Laplace Eqn of Continuity
3. Quick Conditions 3. Seepage under dam
4. Critical Gradient 4. Flow net Problems
5. Problem Solve 5. Filters
 Seepage
Importance of study
The principles of water flow through soils are
used to determine seepage quantities,
pressures or stress, and forces.

Permeability: The ease at which

water water can move through soil.
(Ability)
Seepage: The flow of water through
the soil under a hydraulic gradient.
(Quantifier)
Percolation (from Lat. percōlāre, to filter or trickle through) refers to the movement
and filtering (separation of solids from fluids) of fluids through porous materials.
Infiltration is the process by which water on the ground surface enters
the soil. Infiltration is governed by two forces: gravity and capillary action.
 Seepage force
Where the ground water is nearly stationary no
seepage force occurs. If the ground water is moving, it
imparts a drag force, called a seepage force on the
solid particles. Remember : When there is flow (Up or
Downward), the hydraulic gradient (i)
The seepage pressure,Ps at any point is given by h/L. h Water
in the soil mass is given by, Ps=hγw
(Flow)
Where h is the total head at that point.
If Z is the length of flow over which Z
the proportionate head loss is iZ, L
then X

water
Ps=(h/L).Z. γw= i.Z. γw soil

The seepage force J over total cross

sectional area, A of flow in the soil proportionate
mass is, J = PsA = i.Z.γwA head loss at Z i  h
depth, is iZ L
So, Seepage force per unit volume,
j=J/(AZ)=iZγW/(AZ)= i γW j =iγw=0.15x9.8kN/m3= 1.5kN/m3
If water is flowing through a soil with hydraulic gradient of 0.15 what will be the seepage force,j?
Seepage force per unit volume is equal to the product of iγw
Ref : Paragraph 8.4, Das 5th ed.
Seepage force(P’) increase or decrease the effective
stress , σ1’ at a point in a layer of soil. Fig. -2
When there is no seepage, effective stress at depth z
measured from the surface of the soil layer (Fig.1) is Zγ’ Zγ’A
So, Effective force on an area A, P1’= Zγ’A Z

If there is an upward seepage of water (Fig.2) in the Downward force

vertical direction through the soil layer , then effective +
force on area A, at depth Z, P2’ = Zγ’A-iZγwA iZγwA
Z
Fig. -1 Hence, decrease in the total force
Soil Volume = due to seepage is, P1’- P2’ = Zγ’A - Upward seepage force
Z x Area, A
Zγ’A+ iZγwA = iZγwA =
P1’=σ1’A= P2’=σ2’A=
Thus seepage force per unit Zγ’A - iZγwA
Zγ’A Z volume of soil is, P = Z
(P1’- P2’ )/(Volume of soil, Z.A)
No Seepage =(iZγwA )/(Z.A) = iγw Upward Seepage
 Stress Due to ‘Flow through soil’
In the case of vertical flow through soil mass the
effective stress(σ’v) at a section will be increased
or decreased according to the flow in the upward
or downward direction. Thus effective stress at any
depth Z is given by : Upward

Downward Ps = hγw
σv’ = γ’Z ± Ps Ps = i.Z.γw
J = Ps A
= γ’Z ± iZγw j = J/(AZ)=iγW
Static
In Static Situation case (No flow or when there is
no seepage), Ps=0, thus σ’=γ’Z
 Stress due to flow A

Flow takes place due

No head difference B
to head difference, hL h h hL
No Flow i A B 
LAB L
LAB=
Length of
flow path,
AB through
soil
particles

When there is flow (Up or Downward), the hydraulic gradient (i) is given by h L/L.
Three different scenarios, of identical soil specimens subjected to different flow
conditions, are shown in Figure- (a) there is no flow and the water is static. (b)
flow takes place due to a head difference of hL across the specimen and the flow is
upwards through the specimen.(c) the flow through the specimen is downwards,
again due to a head difference of hL.
When the flow is upwards in the soil, pore water pressure increases and effective
stress decreases. Higher the hydraulic gradient, higher the increase or decrease
in the values of pore pressure and effective stress.
 Effective vertical stress(σ’v ) When there is flow, the
and Pore water Pressure (u) hydraulic gradient i is
at X=? given by hL/L.

γw γw γw
For all three situations, the total

γsat γsat γsat

vertical stress is the same.

When the flow is upwards in the soil, pore water pressure increases and effective stress
decreases. When the flow is downward, the pore water pressure decreases and the
effective stress increases. Higher the hydraulic gradient, higher the increase or
decrease in the values of pore pressure and effective stress.
Example – Upward Seepage

In the figure, the inflow water pressure is 10 kPa.

The soil sample is 0.5m high and the total soil unit
weight is 18 kN∕m3. The water above the soil sample
is 0.3m high. Calculate the effective vertical stress at
the mid height of the sample.
 Solution

Inflow pressure

This is a problem of upward flow

Initial Inflow water Pressure, P=10 kPa
P=10=hγw , So, h=10/9.81=1.02m
hL=1.02-(0.3+0.5)=0.22m i=hL/L=0.22/0.5=0.44 izγw=1.08

u z= 9.81(0.3+0.25) + 1.08= 6.47kPa σv=9.81x0.3+18x0.25= 7.44kPa

σ'v=7.44-6.47=0.97kPa For inflow pressure of 10kPa (Ans)
Problem : Effective By : Lt Rayhanur Rashid
vertical stress at mid Solution CE 21, Sec-A
height of the soil June 2021
sample at C=? (assume
inflow pressure
=24.6kPa)
Effective stress , v ' = Total stress v –Pore water pressure, u or uw Stresses due to Flow
Static Situation (No flow or when there is no seepage)
We know
At X, v ' = v –u u = w (hw + z)
At any depth Z on a point X of a soil column
normal or total stress σ, pore water pressure, u
hw water
and effective stress, σ’ is given by :

v ' = whw + satz - w (hw + z)

z
L = (sat - w ) Z = ' z
water

X
soil Where, ‘=(sat- w )=
 v ' = ' z submerged unit weight

Thus, it is concluded that effective stress, v ' is

independent of the depth, hw of water above
submerged soil or rock.

This statement is known as PRINCIPLE OF EFFECTIVE STRESS

(Terzaghi 1925, 1936, Skempton 1960)
Exercise-1 (DAS & SIVA EX-7) for home practice
Water IN
Water flows under constant
head through the two soil 90mm
samples 1 and 2 as shown in
the figure. The cross sectional
60mm
area of the sample is 2000
30mm
mm2. In five minutes, 650 ml
100mm
water flows though the sample. 2
a. Find the permeability of the
samples, and 40mm
b. In sample 2, find the pore
water pressure at a point
40mm above the bottom. 1 120mm

Ans : 0.145 cm/s, 0.181cm/s, 1.23kPa

 Ans : 0.145 cm/s,
Exercise-1 (DAS & SIVA EX-7) Solution
0.181cm/s,
Here, A=20cm2,t=300sec, Q=650cm3 and, Water IN 1.23kPa
q=Q/t=2.166 cm3/sec, L1= 12cm, L2= A
10cm, h1= 9cm, h2= 6cm
(Remember : for water 650ml=650cm3) 90mm

Sample-1 : i1 = h1/L1= 90/120=0.75

K1=q/(i1A)=2.166 / (0.75x20) = 0.145 cm/s
60mm
Sample-2 : i2 = h2/L2= 60/100=0.60
K2=q/(i2A)=2.166 / (0.60x20) = 0.181 cm/s 30mm
Total Head (TH) at A (let datum is at the bottom
of sample 1) 60mm 100mm
TH=(90+60+30+100+40+120)mm=0.440m X 2
40mm
TH at X (40mm above the bottom of
sample2)=THx=[120+40+40+(60/100)x40+90]mm
= 314mm = 0.314m
40mm
In sample 2,
Applying Bernoulli’s eqn, i.e., Total Head at any find the pore
point is same. So at point X, 0.440 = P/γ + 0.314 water
Or, P = (0.44-0.314)x9.81= 1.236 kPa pressure at a
point 40mm
1 120mm
For Soil 2, i=Δh/L= 60/100=0.6, Let head at X = Y, so above the
bottom.
i=0.6=Y/60, Thus, Y=36m. Total head = 36+90=126mm.
Pore water pressure = 0.126m x 9.81kN/m3=1.236kPa
uw= w(hw+z)+wiz= 9.81(0.03+0.06)+9.81x0.06x9.81=1.236kPa
 Critical Hydraulic Gradient (icr)
A quick condition exits when the upward
force on a soil particle equals the total
particle weight.
Said another way, it is when the seepage
force equals the submerged weight of the
soil.
 If the pore-water pressure is high enough
such that the effective stress is reduced to
zero or less, then it is possible a quick
condition exists.
Remember : γsub/γw= (Gs-1)/(1+e)=icr

icr= i.(γsub/γw) (L/h)

= (h/L).(γsub/γw) (L/h)
= (γ /γ ) =(Gs-1)/(1+e)
Q: Derive an expression for critical hydraulic gradient with its condition of unity.

 As water level is not the same on both sides in the

Remember,
arrangement shown, here an upward flow will occur γsub= γb = γ’ = γsat - γw

Remember
From wt-vol relationship we get : To avoid quick
γsub/γw= (Gs-1)/(1+e)=icr condition:
Gs ≥ 2.67
e ≤ 0.67
If specific gravity of a soil =2.67, and void ratio=0.67, then critical
hydraulic gradient becomes unity, [ icr = (Gs-1)/(1+e) = 1 ] Math
Remember : Boiling is the phenomenon that is misleadingly
called quick sand in old Tarzan Movies, and in many lay
publications. Instability is a condition rather than a material. If
condition is not permitted to develop the same quick sand can be
quite stable as a foundation material. The general approach to
preventing this condition (i.e. Exit gradient ≤0.8) is to force the
water to flow through a longer distance, for example installing or
deepening a cutoff wall.
A saturated sand becomes quick-quick means alive-if upward
flow at a gradient(h/L) of approximately unity occurs at its
Quick Conditions-Prevention
Quick conditions are common in excavations
below the ground water table. This can be
prevented by lowering the ground water
.
elevation by pumping before excavation.

Quick Conditions-Where not?

Quick conditions occur most often in fine sands or silts and
cannot occur in coarse soils.
The larger the particle size, the greater is the porosity. To
maintain a critical gradient of unity, the velocity at which water
must be supplied at the point of inflow varies as the
permeability.
Therefore a quick condition cannot occur in a coarse soil
unless a large quantity of water can be supplied.
 Remember: To avoid quick condition: Gs ≥ 2.67, e ≤ 0.67
If we consider, Gs=2.68, e=0.66, Then unit
Remember : γ'/γw= (Gs-1)/(1+e)=icr
wt of quick sand =19.73kN/m3 or 125pcf

A man can float in fresh water which

weighs about 62.4 lb/cft
Unit wt of quick sand =
[(G+e)/(1+e)]γw= 100 lb/cft (approx)…. So,
as per laws of physics
A man can really avoid floating in it. He is
drowned there due to panic,
 Exercise 2 : Stress due to flow
Example 8.4, DAS 5th ed Remember : γ'/γw= (Gs-1)/(1+e)=icr
a. Consider the upward flow (+ izγw)of water through a layer of sand in a tank as
shown in the Fig. For the sand, e=0.52, Gs=2.67
b. Calculate (i) Total stress, (ii) Pore water pressure, u, (hwγw+ izγw=?) and
(iii) effective stress, σ’ at points A and B
c. What is the upward seepage force (j) per unit volume of soil? ( j=iγw=?)
Solution : Saturated unit weight of Sand, γsat
hL=
hw= Point Total P.W.P, u (kPa), Effective
Stress, σ hwγw+ izγw stress,
=ZA (kPa) σ’=σ-u (kPa)
L=
A 0.7γw+1γsat (1+0.7) γw+ 27.46-24.03=
= 27.46 (1.5/2)(1)γw 3.43
=24.03
B 0.7γw+2γsat (2+0.7+1.5)γw 48.05-41.2 =
= 48.05 =41.2 6.85

Soln :Part-2: Hydraulic gradient, i = hL/L = 1.5/2=0.75,j =iγw=0.75x9.81=7.36kN/m3

 Exercise 2 : Boiling or Quick Condition or Piping
To prevent piping what would be the minimum
height of soil column? [e=0.52, Gs=2.67] For piping or boiling
Ans : γsat= (Gs+e) γw /(1+e) = (2.67+0.52)x hL= condition to occur, we know,
9.81/(1+0.52) ]= 20.588kN/m3 Lγsub= hLγw . At this condition
L= hLγw/ γsub = (1.5x9.81)/
hw= Water (20.588-9.81) = 1.365 m

ZB
Z A=
Sand Again, For piping or boiling
condition to occur, we know,
ic=i,
L=
Now, ic= (Gs-1)/(1+e)=(2.67-1)/
(1+0.52) = 1.0986.
And, i = hL/L, L=? Now, ic=i,
or, 1.0986=hL/L,
Remember,
γsub= γb = γ’ = γsat – γw So, L= hL/1.0986 =1.5/1.0986 =
= (20.588-9.81 ) kN/m3 1.365m
Remember : γ'/γw= (Gs-1)/(1+e)=icr
SIVA-
To prevent boiling, height of soil column in the tank must be > 1.365m EX7
Example Problem 5 on Stress due to flow Exercise problem 8.11 and 8.12, DAS 5th ed,
Refer to the Figure below (8.4a), given that H1=1.5m, 1. What is the rate of upward
H2=2.5m, h=1.5m, void ratio of sand, e=0.5, Gs=2.68, Area seepage ? (Ans : 360cm3/s)
of the tank= 0.6 sqm, K=0.1cm/s 2. If h=1.5m will boiling occur?
Draw the variation of (i) Total stress; (ii) Pore water Why? (No. Since soil column height, H2 < h
pressure; and (iii) Effective stress with depth for the (head loss)
soil layer shown in the figure with upward seepage. 3. What value of h would cause
boiling? (Ans: 2.8m)
Remember, 4. What is the minimum height of
γsub= γb = γ’ = γsat – γw h (h/H2)Z
soil column to prevent piping ?
= iz
H1 A Water

Z Sand
H2 C

Valve open Effective stress at B, σ’=σ-u =H2 γ’- h γw

Inflow 5. Also calculate the effective stress at points A, B and C.( j=iγw=?)
6. What is the upward seepage force per unit volume of soil ?
Example Problem on Boiling

In the figure, the inflow water pressure is 10 kPa. The soil

sample is 0.5m high and the total soil unit weight is 18 kN∕m3.
The water above the soil sample is 0.3m high. Calculate the
effective vertical stress at the mid height of the sample. What
will happen if the inflow water pressure is increased to 20
kPa?
 Solution

Inflow pressure

This is a problem of upward flow

Initial Inflow water Pressure, P=10 kPa
P=10=hγw , So, h=10/9.81=1.02m
hL=1.02-(0.3+0.5)=0.22m i=hL/L=0.22/0.5=0.44 izγw=1.08

u z= 9.81(0.3+0.25) + 1.08= 6.47kPa σv=9.81x0.3+18x0.25= 7.44kPa

σ'v=7.44-6.47=0.97kPa For inflow pressure of 10kPa (Ans)
Solution

Final Inflow water Pressure, P=20 kPa C

P=20=hγw , So, h=20/9.81=2.04m

hL=2.04-(0.3+0.5)=1.24m i=hL/L=1.24/0.5=2.48

u = 9.81(0.3+0.25) + 2.48x0.25x9.81= 11.48kPa

σ'v=(18x0.25) – (2.48x0.25x9.81) = -1.04 kPa < 0
(As the effective stress reduces to zero, so, Soil boiling will occur)
 Home Practice Problem- EX9.9(DAS, 8th ed)
A soil profile is shown in the Figure. Given: H1=H3=1.83m and
H2 =0.91m Plot the variation of σ, u, and σ’ with depth.
(i) Dry unit weight of Sand, (moisture content, w=0) General eqn of γ
Solution 𝐺 𝑠 𝛾 𝑤 2.65 𝑥 9.81 (1+𝑤)𝐺 𝑠 𝛾 𝑤
𝛾 𝑑(𝑠𝑎𝑛𝑑 )= = =17.33 𝑘𝑁 /𝑚 3 𝛾=
1+𝑒 1+0.5 1+𝑒
(ii) Moist unit weight of Sand,𝑆𝑟 𝑒=𝑤 𝐺 𝑠 ,𝑜𝑟 ,0.5 𝑥 0.5=𝑤𝑥 2.65 ,𝑠𝑜 ,𝑤=9.434%
(1+𝑤) 𝐺 𝑠 𝛾 𝑤 ( 1+0.09434 ) 𝑥 2.65 𝑥 9.81
𝛾 (𝑚𝑜𝑖𝑠𝑡 )= = =18.97 𝑘𝑁 / 𝑚3
1+𝑒 1+ 0.5
(iii) Saturated unit weight of Clay (Sr=100% or 1) A
𝑆𝑟 𝑒=𝑤 𝐺 𝑠 ,𝑜𝑟 ,1 𝑥𝑒=0.42 𝑥 2.71, 𝑆𝑜,𝑒=1.1382 Dry Sand
H1=1.83m
Gs=2.65
(1+𝑤)𝐺𝑠 𝛾 𝑤 ( 1+0.42 ) 𝑥 2.71𝑥 9.81 B e=0.5
𝑎𝑡(𝑐𝑙𝑎𝑦)= = =17.655𝑘𝑁/𝑚3
1+𝑒 1+1.1382 Saturated sand (zone of capillary rise)
variation of σ, u, and σ’ with depth Gs=2.65, e=0.5
σ (kPa) 0 u (kPa) 0 σ' (kPa) H2=0.91m
0 Degree of saturation=50%
36.17
1.83 31.71 -4.46 0 1.83 1.83
C
31.71
48.97 Saturated Clay
2.74 48.97 2.74 2.74
H3=1.83m Gs=2.71
Moisture content,w=42%
17.95 4.57 D
4.57 4.57
81.29 63.34
Depth(m) Depth(m) Depth(m) Rock
Exercise-3 : Stress due to flow
 An excavation was made in a clay stratum. When the depth of excavation
was 7.5m, the bottom of excavation pit heaved, cracked and the pit was
filled by a mixture of sand and water.
 A subsequent boring indicate that the thickness of the clay layer was 10.5m,
and below the clay layer was layer of pervious water bearing sand.
Ground surface
0m
Excavation
Heave
Clay -7.5 m
Artesian pressure head, h=?
γsat = 20kN/m3
-10.5 m
Layer of pervious water bearing sand

 What do you think is the reason for the instability of the excavation base?
How much is the artesian pressure head exerted at the bottom of the clay
stratum, if γsat (clay) = 20 kN/m3 ? What can be done to prevent boiling?
Exercise -3 : Stress due to flow
 Solution : (This is a classical case of boiling condition that has developed as the
water in the sand stratum is flowing upward towards the base of the pit under
artesian pressure )
 When the depth of excavation is 7.5m the downward pressure due to saturated clay
and the upward pressure due to artesian pressure, are equal at X-X.
Ground surface
0m

Note : For piping Excavation

or boiling -7.5 m
condition to occur,
Clay h= ? (6.12 m)
Lγsub=hLγw
γsat = 20kN/m3
-10.5 m
X Layer of pervious water bearing sand X

 Thus, (10.5 -7.5) γsat = h. γw in which h is the artesian pressure head. Hence ,
3x20 = h x 9.8, So, h = 60/9.8 = 6.12m.
 To Prevent boiling certain depth of water may be kept in excavation so as to
neutralize upward seepage pressure.
 Alternately water table may also be lowered just before excavation to minimize
pressure head.
Exercise-4 : Stress due to flow
 Problem : A 8m thick layer of stiff saturated clay (γ=19kN/m3) is
underlain by a layer of sand. the sand is under an artesian pressure of
5m. Calculate the minimum depth of the cut that can be made without
causing a heave.
Ground surface

Excavation h
8m
Stiff Clay 5m
γsat = 19 kN/m3
X Layer of pervious water bearing sand X

 Let h be the required depth of the cut. Heave will occur when the
effective stress, σ’ becomes zero at level X-X.
σ = 19x(8-h)=152-19h ; u = 5x9.81=49.05kN/m2
 For heave to occur at X-X, effective stress, σ’ =0= 152-19h-
49.05, Hence, h = 5.42m (Ans)
 Ans : q=0.59cm3/sec
Exercise-6 i2=0.975, k2=1.2x10-2 cm/sec,
Vs1=0.033cm/s, Vs2=0.029cm/s
Soils of two types 1 and 2 are taken Water IN Quick condition occurs in
the upper soil when total
in a permeameter of diameter 8cm head reaches 30.8 cm.
and water is allowed to flow 30cm Water
through them under head of 30cm Dia=8 cm OUT
as shown in the figure
(i) The permeability of soil 1 is 2.8x10-2
cm/sec and 35% of total head causing 20cm Soil 2
flow is lost during flow through this layer,
calculate the rate of discharge and
permeability for soil 2
(ii) If void ratio is 0.55 for soil 1 and 0.7
for soil 2 compute seepage velocity for Soil 1
flow through each layer. 25cm

(iii) If the total head is increased, determine its

value at which either of the soil will become
quick . Assume, Gs=2.65 for soil 1 and Gs=2.7 for soil2
 Solution: (i) Area, A= (π/4)82= 50.26cm2. Ans : q=0.591cm3/sec
Head loss, hL=30cm, K1= 2.8x10-2 i2=0.975, k2=1.2x10-2 cm/sec,
Vs1=0.033cm/s, Vs2=0.029cm/s
cm/sec Water IN Quick condition occurs in
i1= (35% of hL)/25 = (0.35x30)/25 =0.42 the upper soil when
hydraulic head reaches
iRate of discharge, q =k1i1A=0.591=0.975
cm /sec
3
2= (65% of hL)/20 =1 (0.65x30)/20
30.8 cm.
30cm Water
Rate of discharge, q2=q1=0.591=k2i2A So, Dia=8 cm OUT
K2=0.591 / (0.975x50.26)=1.206x10-2 cm3/sec
(ii) Discharge velocity, v=q/A=0.591/50.26 =0.0118
cm/sec , Porosity of soil-1, n1= e1/(1+e1)=
0.55/(1+0.55)=0.355 for Soil-2, n2=e2/(1+e2)=
20cm Soil 2
0.7/(1+0.7)=0.412, Seepage velocity for soil-1,
Vs1=v/n1=0.0118/0.355=0.0332 cm/sec
Similarly, Vs2=v/n2=0.0118/0.412=0.0286 cm/sec 25cm

(iii) Critical hydraulic gradient, ic= (Gs-1)/(1+e)

ic1= (2.65-1)/(1+0.55)=1.06, ic2= (2.7-1)/(1+0.7)=1 Soil 1
For piping or boiling condition to exist, ic=i, So,
i1= ic1, or, (0.35hL /25)=1.06, or, hL=75.7cm
i2= ic2, or, (0.65hL /20)=1, or, hL=30.77cm So, Quick condition
 Exercise-7 (DAS & SIVA EX-8) for home practice
Water flows through the constant head Ans : k=9.82x10-3 cm/s, 1349Pa, 79Pa

setup in the laboratory as shown in the (Remember : Volume of 1g of

water is 1 cm3 = 1mL )
figure, where two identical dense sand
sample A and B are placed – one water
horizontally and the other vertically. e=0.52
The samples are 50mm diameter and 190mm
100mm in length. The water levels in
the left and right sides are maintained

Dia = 50 mm
in the levels shown, ensuring constant 40mm
head throughout the test. The void ratio
of the sand is 0.52 and the specific B
100mm
gravity of the grain is 2.69. If 165g of
water is collected in the bucket within 100mm
15 minutes, what is the permeability of A 165g
soil ? What are the pore water pressure
and vertical effective stress at the mid Find : K of soil
Find : uw , σ’v of B at mid
height of sample B ?
 Find : K of soil
Exercise-7 (DAS SIVA EX-8)-home practice Find : uw , σ’v of B at
Here, A=πD2/4=π52/4=19.635 cm2 mid height
Q=165g=165cc, L=(10cm+10cm)=20cm, Ans : k=9.82x10-3 cm/s, 1349Pa, 79Pa
t=15min=900sec, h = 190mm=19cm, iB= iB= 190/(100+100)=0.95
190/(100+100)=0.95
For constant head setup, water
QL 165 x(10  10) e=0.52
k   9.82 x10 3 cm / s
Ath 19.635 x900 x19 190mm
Let, at the mid height of sample B, the head is
X

Dia = 50 mm
Hence, X / 50 = 190 / (100 +100), So, X= 47.5mm 40mm
Total Head = (47.5 + 50 + 40)mm =137.5mm
= 0.1375m, Pore water pressure, uB = 9.81x X
0.1375 = 1.349kPa =0.95
i=hL/L = 190/(100+100)= 1349PaUw at X 100mm B
UB= w(hw+z)+wiz= 9.81(0.04+0.05)+9.81x0.95x0.05=1.3488kPa 100mm
v ' = ' z - wiz = (20.71-9.81)0.05 – 9.81x0.95x0.05=0.079kPa

 sat
(G  e) w (2.69  0.52) x9.81
 s   20.71 _ kN / m3 A 165g
1 e 1  0.52
At mid height of B total stress, σv= γsat z + hwγw = 20.71x0.05+0.04x9.81=1.428 kN/m
2

Effective stress at mid height of B= σ’v = σv - uB =(1.428-1.349)kPa=0.079kPa=79Pa

 Quick Review
So far we have learned :
 Permeability and Darcy’s law.
 Darcy’s law can not be solved directly for two-
dimensional flow as i and A varies in the flow regime.
 Quantification of flow through soil, Q=AVD or (Darcy)
Seepage velocity, VS = VD / n =VD
Terzaghi’s PRINCIPLE OF EFFECTIVE STRESS
concept.
Development of seepage pressure, (wiz) as a result of
the frictional drag of the flowing water on soil grains.
 Assignment Problems on Seepage
Problem-1. The specific gravity of solids (Gs) of a sand is 2.66. The
porosity(n) of the soil in its loosest and densest state is 45% and 37%,
respectively. Determine the critical hydraulic gradient (ic) for these two
states.
Problem-2. A 10m thick layer of stiff saturated clay is underlain by a layer of
sand as shown in the Figure-1. The sand is under artesian pressure. A 5.75m
deep cut is made in the clay. (a) Determine the factor of safety against heaving
at point A. (b)What would be the maximum permissible depth of cut before
heaving would occur?

Saturated Clay

Sand

Figure -1
 Assignment Problems on Seepage
Problem-3. Refer to Figure-2 in which upward seepage is taking place through
a granular soil contained in a tank. Given: H1=1.5m; H2=2.5m; h=1.5m; area
of the tank=0.62m2; void ratio of the soil, e =0.49; Gs=2.66; and hydraulic
conductivity of the sand (k)= 0.21 cm/sec. (a). What is the rate of upward
seepage of water (m3/min)? (b). Will boiling occur when h=1.5 m? Explain.
(c). What would be the critical value of h to cause boiling? (d). If the point C is
located at the middle of the soil layer, then what is the effective stress at C?

(e) Draw the variation of (i) Total stress; (ii) Pore water
pressure; and (iii) Effective stress with depth for the soil
layer shown in the figure with upward seepage.

Figure -2
Home Practice Assignment
Figure-A shows two saturated soil samples that are connected in series in a
tube with three vertical standpipes. Soil samples 1 and 2 can have different
coefficients of permeability (k1 and k2), lengths (L1 and L2), and cross-
sectional areas (A1 and A2). Suitable screens are provided to maintain the
samples intact.
(1) Consider the case where the same soil is used for both samples in the
device. For h1 = 60.0 cm, h3 = 6.0 cm, L1 = 45 cm, L2 =55 cm, A1= 28
cm2, and A2 = 32 cm2, determine the value of h2.
(2) Finally, consider the case where two different soils are tested. If the
discharge is equal to 0.20 cm3/s, and h1=75 cm, h2 =38 cm, and H3=10
cm, compute the permeability of soils 1 and 2.
Solution Hints :
The hydraulic
gradient for soil 1
=i1=(h1-h2)/L1,
Q1 = k1i1A1, Now,
Q1=Q2, For same
soil, k1=k2, find
h2=(h1+h3)/2

Figure-A
Two Dimensional
Fluid Flow
 Concept of Flow Net
Fig(a) shows a tube 1.2m x 1.2m in cross section by 4.8m high through which steady
–state vertical flow(laminar) is occurring. Rate of seepage through the soil by Darcy’s
law, Q=kiA=(0.05x10-2m/sec)x(4.8/3.0)x(1.2mx1.2m)=0.00115 m3/sec
Elev.
- 4.8 Fig(c): If we placed a dye on the top of the soil (el.-3.6) and traced on a
Tube microscopic scale the movement of dye through the soil will present
1.2m x - .
numerous flow lines(red vertical line)
1.2m in - 4.2
cross - . 4 flow channels(0.3m wide)
section
- 3.6 H

e
,h
- . 9/10 H

ead
hp
Elevation(m)

- 3.0 h 8/10 H
d,

nh
ea
ad ,
- . 7/10 H
h

Equipotential line
tio
al
H

Soil - 2.4
e

t 6/10 H
re h

va
n =1/3 - . Ele To
5/10 H
ssu

K=0.05 - 1.8 4/10 H

Pre

cm/sec - .
3/10 H
- 1.2 2/10 H
- . 1/10 H
- 0.6 . . . . . . . . . . 0
- . -0.6 0.0 0.6 1.2 1.8 2.4 3.0 3.6 4.2 4.8 Flow/Stream line
- 0.0 Head (m)
Datum Fig(a) Fig(b) : Different heads Fig(c) : Flow net
 Flow Net For One Dimensional Flow
From a Flow net soil engineer can determine three useful information
(1) Rate of flow b
(2) Head l aA=lxb
(3) Gradient
In isotropic soil the flow line and 4 flow channels(0.3m wide)
equipotential lines intersects at right H
9/10 H
angle. They form an orthogonal net A 8/10 H
consisting of squares, A (A = l x b). 7/10 H

Equipotential line
6/10 H
Here, H = Total Head lost in the flow 5/10 H
Nd = Number of head drops in the net 4/10 H
H/Nd = Total head loss in square A 3/10 H
l = vertical dimension of square A 2/10 H
aA= Area of square A = l x b 1/10 H
0
Nf =Number of flow channel Flow/Stream line
s
∫ = Shape factor = Nf /Nd (Characteristics of flownet)
Fig : Flow net
i = Hydraulic gradient = H/(N x l)
 Flow Net For One Dimensional Flow
Rate of Flow qA = k. iA . aA H =Total Head lost in the flow
nd =Total number of head drops
= k.{(H/(Ndxl)}.(lxb) aA= Area of square A = l x b (b=l)
- 4.8 nf = Number of flow channel
Tube
- .
Flow per unit length(L) s
1.2m x
1.2m in- 4.2 ┴ to the paper, ∫ = Shape factor = Nf /Nd
cross
section
- . Q/L=qANf =k. H(Nf /Nd) b 4 flow channels(0.3m wide)
- 3.6 H
- . = 0.0005 m/s x 4.8m x l a =lxb 9/10 H
A
Elevation(m)

A
4/10=0.00096 m /sec/m
3

Ten Equal Head Drops

- 3.0 8/10 H
- . 7/10 H
Head at el. 3m

Equipotential line
H

Soil - 2.4 6/10 H

n =1/3 - . At el. 3, Total head, 5/10 H
K=0.05 -
cm/sec -
1.8 h= (8/10).H= 3.84m, where, H=4.8m 4/10 H
. 3/10 H
-
Pressure head = total head –
1.2 2/10 H
- .
elevation head=3.84m-3m =0.84m
1/10 H
- 0.6 Pressure head, hp=0.84x9.81=8.24kN/m3 0
- . Gradient Flow/Stream line
- 0.0 All squares are same, So, for any square,
Datum gradient, i=H/(Nd xl)=(4.8x/(10x0.3)=1.6 Fig : Flow net
 TWO-DIMENSIONAL FLOW
Two dimensional flow occurs when all the velocity
vectors are confined to a single plane.
The ground
water flow in to
the long t i on
ca va
excavation E x
o ng
shown in the L
figure could be
evaluated using a
two dimensional
analysis in a
vertical plane.
 TWO-DIMENSIONAL FLOW
Flow net and Laplace’s Equation of Continuity
In general, the flow of water through a natural
deposit of soil is not uniform over the entire
area perpendicular to the flow and the flow does
not occur in one direction only. In such case
ground water flow is generally found by use of
graphs referred to as FLOW NETS.
Pierre-Simon_Laplace
(1749-1827)
The concept of flow net
based on Laplace’s Equation
of continuity which
describes the steady flow
condition for a given point
in the soil mass.
Flow net and Laplace’s
Equation of Continuity
S
To derive Laplace’s h
e
Equation of continuity e
let us take a single row of H1
t h
sheet piles driven in to a p H2
permeable soil layer as i
l
shown in the figure. The e
steady state flow of water dz A
from upstream to the dx
down stream side through Impermeable layer
the permeable layer is a
two dimensional flow. Fig : Single row sheet piles
driven in to permeable layer
Fig : A two dimensional flow
sheet pile
flow of water from
upstream to the down
stream side

Permeable layer
Flow Nets
Flow Nets
Flow Nets
Flow Nets
Flow Nets
Flow Nets
Flow Nets
Flow net and Laplace’s
Equation of Continuity
S
For flow at point A, we h
e
consider an elementary e
soil block with H1
t h
dimensions dx,dy and dz p H2
(length dy is i
l
perpendicular to the plane e dz
of paper). Let Vx and Vz dz A dy

be the components of dx dx
discharge velocity in the
Impermeable layer
horizontal and vertical
direction respectively. Fig : Single row sheet piles
driven in to permeable layer
Laplace’s Continuity Equation
Exit velocity
Rate of out flow
Total Rate of inflow =
Rate of out flow [Vz+(δ Vz/δz)dz]dx.dy
Vx.dz.dy+Vz.dx.dy =

[Vx+(δ Vx/δx)dx]dz.dy
[Vz+(δ Vz/δz)dz]dx.dy +
[Vx+(δ Vx/δx)dx]dz.dy
Vx.dz.dy
So,
dz
dx
dy
But, = Volume ≠ 0 Component of the
velocity at entry to
Hence, the element

Vz.dx.dy
(Rate of inflow)
Mass inflow rate -
mass outflow rate Exit velocity
= change in storage
with time

dz
Components of dy
velocity of
flow at entry to dx
the element

Mass in=Mass out + Change in storage

If i in a given direction is iz δh/δz then at opposite face dx away i
will be, iz + (δi/δz)dz
 dz
dy
So, dx
Now, using Darcy’s law, v=ki, the discharge velocities can be
expressed as : Vx=kxix= kx and Vz=kziz= kz
Where kx and kz are the coefficients of permeability in the
horizontal and vertical directions. I = hydraulic gradient=δh/δx

Now,
𝜕❑
𝜕𝑥 ( kx
𝜕h
𝜕𝑥
+
𝜕❑
𝜕𝑧 ) kz
𝜕h
𝜕𝑧 (
=0 )
Or, k x
𝜕2h
𝜕𝑥 2 (
+ kz
𝜕2h
𝜕𝑧 2
=0 )( )
If soil is isotropic with respect to the permeability coefficients,
i.e., kx=kz , the above equation simplifies to
 Laplace continuity Equation
 Solution of Laplace Equation
 Laplace equation for two dimensional
flow can be solved analytically which is
much complex for seepage problem due
to complex boundary conditions.

So! What can

be done ? Geotechnical engineers widely rely
on alternative solution methods,
most notably :
a ble for us
om for t
Most c
 Flow nets
 Electrical analogy models
 Numerical solution
 Solution of Laplace Equation

 Laplace equation

This differential equation represents two families of

orthogonal curves (meeting mutually at 900 )

 One family represent the flow line in a flow net

 The other family

represents
equipotential lines
in the flow net.
Steady-state seepage under a dam
Seepage Terminology
Stream line is simply the path of a water molecule.
From upstream to downstream, total head (TH) steadily
decreases along the stream line.

hL

datum
concrete dam
TH = hL TH = 0

soil

impervious strata
Seepage Terminology
Equipotential line is simply a contour of constant total
head (TH).

hL
datum
concrete dam
TH = hL TH = 0

soil
TH=0.8 hL
impervious strata 67
Flownet
A network of selected stream lines and equipotential
lines.

concrete dam

curvilinear
square

90º
soil

impervious strata 68
 Quantity of Seepage (Q)
Nf # of flow channels

Q  khL ….per unit length normal to the plane

Nd
# of equipotential drops

head loss from upstream to

downstream
hL

concrete
dam

impervious strata
 Heads at a Point X
Total head = hL - # of drops from upstream x h
Elevation head = -z hL

Pressure head = Total head – Elevation head Nd
If hL=10ft
No. of drops at, X = 5.5
(from up to down stream) UP hL=10ft
Total drops = 7
X is 13ft below the datum datum
Total head = ? TH = hL concrete dam TH = 0
Pressure head = ? 1 Z =13ft
Ans : ( up to down stream) Down
Total Head (TH)=10 – (5.5).(10/7)= 2.143ft
Elevation Head (EH)= -13ft h 2 7
3 X
Pressure head=(TH-EH)= 2.134+13=15.134ft 4 5 6
Down to upstream ( Drop at x=1.5)
Pressure head =1.5*10/7 (TH) - (-13) (EH)=15.14 impervious strata
Pressure head = 23- (5.5).(10/7)=15.14ft, Total head = pressure head + elevation head = 15.14-13ft=2.14ft
Seepage Calculation from flow net
q  kiA
Flow Nets and Seepage
h  drop in head For anisotropic soil, k’=√(kmaxKmin)
q  flow rate For isotropic soil, k’=kx=ky
q  k xh y  kh xy 1
h  h
N Flow Line
d
N  Number of  Dh

y
 d 
Flow Line

 equipotent ial 

 increment 
 
  Dq
x
q  kh  k h Flow Line
N

nel
d
q  q

h an
Eq
N uip

wC
f ot ent

Flo
ial
khN Lin
q  kh  q  f es
N N N
f d d
 Anisotropic soil

( )( )
dz
𝜕2h 𝜕2h
kx + kz =0 dy
𝜕𝑥 2 𝜕𝑧 2 dx
If soil is not isotropic with respect to the permeability
coefficients, i.e., kx≠ kz , the above equation represents two
families of curves that do not meet orthogonally (900 to each
other). We can rewrite the equation as under :

(( ) ( )
𝜕2h 𝜕2h
+ =0

)
𝑘𝑧 𝜕𝑧2
𝜕 𝑥2 Substituting, x’=√(kz/kx)
𝑘𝑥

𝜕2h 𝜕2h Laplace continuity

+ =0 Equation in transformed
𝜕𝑥 2 𝜕𝑧2
′
coordinate.
𝐻 𝑁𝑓
Rate of seepage per unit length, 𝑞=√ 𝑘𝑥 𝑘𝑧
𝑁𝑑
Flownet Example-1 K=4.8x10-3 cm/s
Potential
Flow channel, Nf drops, Nd
q=? 5 4 3 2 1

Nd = ? 9 1
8 2
Nf = ? 7 6 4 3
5
Flow boundary Line
khN
f
q
N Example-2
d
k  (4.8  10 3 cm / s )(1in / 2.54cm)(1 ft / 12in)  1.57  10 4 ft / s
h  12 ft  3 ft  9 ft
N  55
f
N  99
d
(1.57  10 4 ft / s )(9 ft )(5)
q  7.85  10 4 ft 3 / s perfoot of sheet pile
9
Flownet Example-2 H/Nd =

Sheet Pile
A flow net for seepage 4m 8/12=0.667
under a sheet pile wall is 8m
shown in the figure. The D
saturated unit weight of Datum
soil being 20kN/m3. 11m 1m
E
F C Z
Determine the values of
effective vertical stress 6m
A
(σ'A and σ'B )at A and B G B
H
Total No. of drop, Nd=12
Total head, H = 8m, So,
H/N =0.667
P dElevation No. of Dissipation Head or Pressure head (m), Pore water pressure,
oi (Z) w.r.t. drop for Head loss, Hd(m) = H1= Hd+Z u = (H1x9.81) kN/m2
nt Datum points, A H/Nd x No. of drop
(m) &B

A 7 8.2 from R 0.667x8.2=+5.47,or +5.47+7=12.47, or, 12.47x9.81=122.30

3.8 from L 0.667x(-3.8)= -2.54 -2.54+15=12.47
B 7 2.4 0.667x2.4 =1.601 1.601+7=8.601 8.601x9.81=84.37 ≈ 84

σ'A= (4x9.81+11x20) – 122.30 = 137kN/m2. σ'B = (1x9.81+6x20) – 84.37 = 45.81=46 kN/m2

 Flownet Example-3

Find
q=?
Total Head at P = ?
(b)
Pore water pressure at P =?
 Flownet Example-3

Total head at P

8.4m=up/γw
 Uplift Pressure under Hydraulic Structures
Flow nets can be used to determine the uplift pressure at the base of a hydraulic
structure. This general concept can be demonstrated by a simple example below.
There are seven equipotential drops (Nd ) in
the fow net, and the difference in the water
levels between the upstream and
=H downstream sides is H=7 m.
The head loss for each potential
drop is H/7 = 7/7 = 1 m.
1 7
2
3 6
4 5

The uplift pressure at a (from left corner) = (Pressure

head at a) x (γw)=[(7+2)-1] γw =8γw
Similarly, the uplift pressure at b = [9 - (2)(1)] γ w = 7 γ w
the uplift pressure at c = [9 - (3)(1)] γ w = 6 γ w
……. uplift pressure at f = [9 - (6)(1)] γ w = 3 γ w
The uplift pressures are plotted in scale. Uplift force per unit
length measured along the axis of the weir can be calculated by
finding the area of the pressure diagram. - Home(Practice
Practice
Example problem-6)
Uplift force diagram
 Piping in Granular Soils
At the downstream, near the dam,
h
the exit hydraulic gradient iexit 
l
Head drop in each
square = hL/Total drop
hL
datum
concrete
dam l
h = total head drop

soil
impervious strata
 Piping in Granular Soils
If iexit exceeds the critical hydraulic gradient (ic), firstly
the soil grains at exit get washed away.
This phenomenon progresses towards the upstream, forming
a free passage of water (“pipe”).

hL
datum
concrete
dam
no soil; all water

soil
impervious strata
 Piping in Granular Soils
Piping is a very serious problem. It leads to downstream
flooding which can result in loss of lives.
Therefore, provide adequate safety factor against piping.

ic
Fpiping 
iexit

concrete typically 5-6

dam

soil
impervious strata
 Piping Failures
Baldwin Hills Dam after it failed by
piping in 1963. The failure occurred
when a concentrated leak developed
along a crack in the embankment,
eroding the embankment fill and
forming this crevasse. An alarm was
raised about four hours before the
failure and thousands of people were
evacuated from the area below the
dam. The flood that resulted when the
dam failed and the reservoir was
released caused several millions of
dollars in damage.
82
Piping Failures

Fontenelle Dam, USA (1965)

Flownet Example-4 (Parcher – Example 3-2)
Assume a low concrete dam as shown in the figure on
a cohesionless soil having a coefficient of permeability
(k) of 20x10-4 cm.s-1
Requirements (d) Determine the seepage force of the flowing water at the toe
of the Dam. Is there danger of piping at this point?

(a) Draw flow net

Water
(b) Determine the quantity
10ft
10ft 26ft Tail Water
of water flowing under
the dam per foot of width
per day 3ft Concrete Dam
Toe
(c) Determine the uplift
pressure on the bottom of 14ft
the Dam 5ft upstream Sand
from the toe
Impervious layer
 How to find head at any point? See
Flownet Example-4 the slide –
(4) seepage force of
the flowing water at
(Head at point X) N
Solution (1) Flow net : See figure q  kh f
Nd
the toe of the Dam,
j = itoe x γw
(2) Quantity of flow, Q=?, Here, Nf=4,Nd=17, k=20x10-4 cm.s-1 iat toe = Δh/L
So, Q=[(20x10-4x602x24)/(2.54x12)]x10x(4/17)=13.4ft3/day/ft
Δh=10/17=0.59
(3) Uplift pressure on the bottom of the dam 5ft L of square at toe =
upstream from toe =[62.4][13-(12/17)x10]=371 lb/ft2 2ft , i at toe = 0.59/2=
From toe to upstream[0.59x5+3 ]*62.4=371 psf 0.29, J=0.29x62.4pcf =
18.1pcf
Comments : Water
Δh
The average Nd from toe 10ft
hydraulic gradient in 10ft 26ft Tail Water
the large square next 5ft
to the toe indicates
that there would be 3ft Concrete Dam toe
no piping at the toe,
but the large square
should be subdivided 14ft
to investigate the Sand
possibility of piping
immediately adjacent Impervious layer
to the toe.
Scale : 1/8” = 1’-0”
A concrete weir shown in Figure with a sheet pile of 6 m depth at the
heel rests on a 16 m thick sandy stratum. Length of the base is 20 m and
it has to retain 8 m of water on up stream. The base has been put 2 m
below ground. The value of K for the soil on which the weir is resting is
4.6x 10-4 cm/sec and specific gravity of the soil grains is 2.65.

Flownet 2m

Example-5 6m

(i) Calculate the amount of water seeping under the weir per meter width.
(ii) Find out seepage force per unit volume (per m3) at the toe of the foundation
and ascertain if there is any chance of piping to occur.
(iii) Calculate the uplift pressure on the bottom of the weir 10 m from the heel.
(Note that point A is at toe side)
(iv) Find out the seepage velocity at toe of the weir if void ratio is 0.72.
 Flownet
Example-5
2m
6m

Solution
(i) Calculate the amount of water seeping under the weir per
meter width.
No. of flow channel, Nf=4, K = 4.6x 10-4 cm/sec
No. of potential drop, Nd=12, h = 8m
Amount of discharge=q=k(Nf/Nd)xh = 2.266 x10-4 cm/sec
 Flownet
2m
Example-5
6m

(ii) Seepage force per unit volume (per m3) at the toe of the foundation
and ascertain if there is any chance of piping to occur. ( G=2.65, e=0.72)
No. of potential drop, Nd=12, h = 8m
Head drop in each sqr, Δh=8/12=0.66m (same for last square also)
Length of smallest side AB of last square = 1.2m (from fig)
i = Δh/L=0.66/1.2=0.55
ic for piping = (G-1)/(1+e)=(2.65-1)/(1+0.72)=0.96
since i < ic, No chance of piping
Seepage per unit volume (per cum) = iγw= 0.55x(1t/cum)=0.55t/cum
(per cum) = 0.55 tonne
 Flownet
2m
Example-5
6m

(iii) Calculate the uplift pressure on the bottom of the weir 10 m from the heel
(upstream).
(iv) Find out the seepage velocity at toe of the weir if void ratio is 0.72.
You can also calculate Head from down to upstream side as under
Head at P=2(elev of P) +0.667(Δh) x[(12-7.3)drop count from B]=5.13

Δh=8/12=0.667m, Head at P = 8+2-0.66x7.3 = 5.13m.

Uplift pressure = γwh = 1 t/cum x 5.13 = 5.13 t/sqm

 Velocity, V = ki= 4.6x 10-4 cm/sec x 0.55 = 2.63x10-4 cm/sec

 n=e/(1+e) = 0.72/(1+0.72)=0.418
 Seepage velocity, Vs=V/n=(2.63x10-4) / 0.418= 6.29x10-4 cm/sec
Flownet Example-6
 Flownet Example-6

H=5m-1m=4m
Nd=15
Nf=4.7
H/Nd=4/15=0.27
Flownet Example-6

(H/Nd)x1

Pressure head at 1
Pressure head at 2

(H/Nd)x2

(H/Nd)x7.5
Flownet Example-7
 Filters
Used for:
 facilitating drainage

 preventing fines from being washed away

Used in: Filter Materials:

 earth dams  granular soils

 retaining walls  geotextiless

Granular Filter Design
Two major criteria: granular filter

(a) Retention Criteria

- to prevent washing out of fines

 Filter grains must not be too coarse

(b) Permeability Criteria

- to facilitate drainage and thus avoid

build-up of pore pressures

 Filter grains must not be too fine

Granular Filter Design
Retention criteria: Permeability criteria:
D15, filter < 5 D85, soil D15, filter > 4 D15, soil
average filter pore size - after Terzaghi & Peck (1967)

D15, filter < 20 D15, soil

- after US Navy (1971)

D50, filter < 25 D50, soil

GSD Curves for the soil and filter must be parallel

End of permeability and seepage
Home assignment
1. Solve Flownet example -4 by drawing graph
on scale.
2. DAS – Problem 7.3 – 5th edition (Ans:
2.1m3/m/day
Home Practice
1. Example 18.3 – Lambe-Whitman flownet
example -2 by drawing graph on scale.
2. DAS – Example 7.2, Fig 7.13, 5th Ed.
3. Example 8.6, Coduto (1999, Reprint 2007)
Let us close now