Embankment Dam Lecture Note
Embankment Dam Lecture Note
3. Embankment Dam
The bulk of mass in an earth fill dam consists of soils while in the rock fill dam it consists of rock
materials. Depending upon the method of construction, earth dam can be divided in two categories:
(i) Rolled fill Dam
(ii) Hydraulic fill Dam
In the Rolled fill Dam, the embankment is constructed in successive, mechanically compacted layers.
The suitable materials are transported from borrow pits to the construction site by suitable earth
moving machineries. It is then spread by Bulldozers, and sprinkled to form layers of limited
thickness having proper water content. They are then thoroughly compacted and bonded with the
preceding layer by means of power operated rollers of proper design and weight.
In the case of Hydraulic fill dam the materials are excavated, transported and placed by Hydraulic fill
method. In this method the flumes are laid at a suitable falling gradient along the outer edge of the
embankment. The material mixed with water at borrow pits, is pumped into these flumes. The slush
is discharged through the outlets in the flume, at suitable interval along their length. The slush thus
flows towards the center of the bank. The course material of the slush settles at the outer edge while
finer material settles at the center. No compaction is done. At present the method is not in general
use.
Rolled fill earth dams can further use subdivided into the following types
(i) Homogeneous embankment type
(ii) Zoned embankment type
(iii) Diaphragm embankment type
Embankment Dam
Accordint to design
According to method of
Constructuion
b) Modified homogeneous
(2) Zoned Earth Dam, however, contains materials of different kinds in different parts of
the embankment. The most common type of an earth dam usually adopted in the zoned
earth dam as it leads to an economic & more stable design of the dam. In a zoned earth
dam, there is a central impervious core which is flanked by zones of more pervious
material. The pervious zones, also known as shells, enclose, support and protect the
impervious core. The U/s shell provides stability against rapid draw downs of reservoirs
while the down stream shell acts as a drain to control the line of seepage and provides
stability to the dam during its construction and operation. The central impervious core
checks the seepage.
Figure 3.3
(3) Diaphragm embankment type: In this the bulk of the embankment is constructed of
pervious material and a thin diaphragm of impermeable material is provided to check
the seepage. The diaphragm may be of impervious soils, cement concrete, bituminous
concrete or other material and may be placed either at the centre of the section as a
central vertical core or at the u/s face as a blanket.
The designation ‘rock fill embankment’ is appropriate where over 50% of the fill material may be
classified as rock pieces. It is an embankment which uses large size rock pieces to provide stability
and impervious membrane to provide water tightness.
Modern practice is to specify a graded rock fill heavily compacted in relatively thin layers by heavy
plants. The constructions method is essentially similar to that of Earth fill Dams. Materials used for
membrane are earth, Concrete steel, asphalt and wood. The impervious membrane can be placed
ether on the upstream face of the dam or as a core inside the embankment. Such a construction
therefore becomes similar to diaphragm type. Rock fill embankments employing a thin u/s
membrane are referred to as decked rock fill dams.
In spite of taking great care in construction of earth dams, some failures have occurred in the past.
However, knowledge of the principal lessons learned from failures and damages in the past is an
essential part of the training of earth dam designer.
On the basis of investigation reports on most of the past into three main classes:
1. Hydraulic failures : 40%
2. Seepage failures : 30%
3. Structural failures: 30%
The preliminary design of an earth dam is done on the basis of past experience and on the basis of
the performance of the dams built in the past. We shall discuss here the preliminary selection of the
following terms:
1) Top width
2) Free board
3) Casing or outer shells
4) Central impervious core
5) Cut-off trench
6) Downstream drainage system.
1) Top width. The crest width of an earth dam depends on the following considerations:
Nature of the embankment materials and minimum allowable percolation distance
through the embankment at the normal reservoir level.
Height of the structure
Importance of the structure
Width of highway on the top of the dam
Practicability of construction
Protection against earthquake forces.
Following are some of the empirical expressions for the top width b of the earth dam, in terms of the
height H of the dam:
H For very low dam (H<10m)
b 3
5
b=0.55H1/2 + 0.2H For medium dam (10m<H<30m)
b=1.65(H+1.5)1/3 For large dam (H>30m)
Civil Engineering Department 5
Hydraulic Structure I Embankment Dam
2) Free board. Free board is the vertical distance between the horizontal crest of the embankment
and the reservoir level. Normal free board is the difference in the level between the crest or top
of the embankment and normal reservoir level. Minimum free board is the difference in the
elevation between the crest of the dam and the maximum reservoir water surface that would
result and spillway function as planned. Sufficient free board must be provided so that there is
no possibility whatsoever of the embankment being overtopped.
3) Casing or outer shells. The function of casing or outer shells is to impart stability and protect
the core. The relatively pervious materials, which are not subjected to cracking on direct
exposure to atmosphere, are suitable for casing. Table 10.2 (a) gives recommendations for
suitability of soils used for earth dams as per IS: 8826-1978.
The design slopes of the upstream and downstream embankments may vary widely, depending on
the character of the materials available, foundation conditions and the height of the dam. The slopes
also depend up on the type of the dam (i.e. homogeneous, zoned or diaphragm).
The seeping water flows under a hydraulic gradient which is due only to gravity head
loss, or Darcy’s law for flow through porous medium is valid.
There is no change in the degree of saturation in the zone of soil through which the
water seeps and the quantity flowing in to any element of volume is equal to quantity
which flows out in the same length of time.
(Steady flow)
The hydraulic boundary conditions at entry and exit are known.
∆y
Vx Vx+ (∂Vx/∂x)∆x
∆x
v v y
vx y.1 v y x.1 vx x x y.1 v y x.1
x y
From which
v x v y
0 … (i)
x y
This is the continuity equation.
According to assumption 2:
h
vx K xix K x * … (ii)
x
h
And VY kY IY Ky …(iii)
y
Where h = hydraulic head under which water flows.
Kx and Ky are coefficient of permeability in x and y direction.
Substituting (ii) and (iii) in (i), we get
2 K x h ( K y .h)
2
0 … (3.1)
x 2 y 2
For an isotropic soil,
Ky = Kx = K
Hence we get from eq. (3.1)
2h 2h
0
x 2 y 2
Substituting velocity potential = = K*h , we get
2 2
2 0 … (3.2)
x 2 y
This is the Laplace equation of flow in two dimensions. The velocity potential may be defined as a
scalar function of space and time such that its derivative with respect to any direction gives the fluid
velocity in that direction.
This is evident, since we have
=Kh
h
K K .i x v x
x x
h
Similarly , K K .i y v y
y y
The solution gives two sets of curves, know as equipotential lines and stream lines (or flow lines),
mutually orthogonal to each other, as shown in Fig. below. The equipotential lines represent contours
of equal head (potential). The direction of seepage is always perpendicular the equipotential lines.
The paths along which the individual particles of water seep through the soil are called stream lines
or flow lines.
A network of equipotential lines and flow lines is known as a flow net. Fig.3.6 shows a portion of
such a flow net. The portion between any two successive flow lines is known as flow channel. The
portion enclosed between two successive equipotential lines and successive flow lines is known as
field such as that shown hatched in Fig. 3.6.
h
Then h
Nd
h b Nf b
q q k . .N f kh .
Nd l Nd l
Where Nf = total number of flow channels in the net. The field is square and hence b=l
Nf
Thus, q kh
Nd
This is the required expression for the discharge passing through a flow net, and is valid only for
isotropic soils in which
kx k y k.
This is not a Laplace equation. Hence flow net can not be drawn directly. Rewriting, it we get
kx 2h 2h
0
k y x 2 y 2
ky
Let us put xn x
kx
Where xn is the new co-ordinate variable in the x - direction.
Then the above equation becomes,
2h 2h
0 … (3.3)
xn2 y 2
To plot the flow for such a case, the cross-section through anisotropic soils is plotted to a natural
scale in the y-direction, but to a transformed scale in the x-direction, all dimensions parallel to x- axis
ky
being reduced by multiplying by the factor . The flow net obtained for this transformed section
kx
will now be constructed in the normal manner as if the soil were isotropic. The actual flow net is then
obtained by re- transforming the cross- section including the flow net, back to the natural scale by
kx
multiplying the x- coordinates by factor . The actual flow net thus will not have orthogonal set of
ky
curves. As shown in figure 10.17, field of transformed section will be a square one, while the field of
Kx
actual section (retransformed) will be a rectangular one having its length in x direction equal to
Ky
times the width in y direction.
Phreatic line / seepage line / Saturation line is the line at the upper surface of the seepage flow at
which the pressure is atmospheric.
Civil Engineering Department 12
Hydraulic Structure I Embankment Dam
Figure below shows a homogeneous earth dam with horizontal drainage blanket FK at its toe. The
phreatic line in this case coincides with the base parabola ADC except at the entrance. The basic
property of the parabola which is utilized for drawing the base parabola is that the distance of any
point p from the focus is equal to the distance of the same point from the directrix. Thus
Distance PF = Distance PR where, PR is the horizontal distance of P from the
Directrix EG
Figure 3.9
Graphical method
Steps:
Starting point of base parabola is @ A AB = 0.3L
F is the focal point
Draw a curve passing through F center @ A
Draw a vertical line EG which is tangent to the curve
EG is the directrix of the base parabola
Plot the various points P on the parabola in such a way that PF = PR
Analytical method
PF = PR
Civil Engineering Department 13
Hydraulic Structure I Embankment Dam
x y
2 2
x yo
From point A (known), x = b and y = h
yo b 2 h 2 b
2
d ( y 0 2 xy o ) 2
qk ( y o 2 xy o )
dx
yo 2
q k( )( y o 2 xy o )
2
y o 2 xy o …………. (3.7)
q ky o
Figure below shows a homogeneous dam with no horizontal drainage filter at the d/s side. The
focus in this case will be the lowest point F of the d/s slope.
a
Table for the value of with slope angle
a a
a
a a
300 0.36
600 0.32
900 0.26
1200 0.18
1350 0.14
1500 0.10
1800 0.0
Figure 3.11
a. Analytical Solution of Schaffernak and Van Iterson for < 300 (Fig.3.9)
In order to find the value of a analytically, Schaffernak and Van Iterson assumed that the energy
gradient
dy
i tan . This means that the gradient is equal to the slope of the line of seepage, which is
dx
approximately true so long as the slope is gentle (i.e. <300).
and
h 2 a 2 sin 2 a
a sin tan (b a cos )
2
Figure 3.12
dy
Thus q kiA k A (3.10)
ds
dy
At J, s= a and y = a sin then, sin
ds
Where s = distance measured along the curve.
Substituting in (3.10), we get
h 2 a 2 sin 2
a sin 2 .( S a )
2
h2
or a 2aS
2
0
sin 2
h2
From which aS S 2
…. (3.12)
Sin 2
a b 2 h 2 b 2 h 2 cot 2 … [3.13]
Figure 3.13
0
Figure 3.14
Based on the above discussions, the characteristics of the phreatic line may be summarized below:
1. At the entry point, the phreatic line must be normal to the upstream face since the
upstream face is a 100% equipotential line. For other entry condition (Fig.3.15), the
phreatic line starts ta11ngentially with the water surface.
2. The pressure along the phreatic line is atmospheric. Hence the only change in the head
along it is due to drop in the elevation of various points on it. Due to this, the successive
equipotential lines will meet it at equal vertical intervals.
3. The focus of the base parabola lies at the break out point of the bottom flow line, where
the flow emerges out from relatively impervious medium to a highly pervious medium.
4. When horizontal filter or drainage toe is provided, the phreatic line would tend to
emerge vertically.
5. In the absence of any filter, the seepage line will cut the downstream slope at some point
above the base. The location of this point, and the phreatic line itself, is not dependent
on the permeability or any other property, so long as the dam is homogeneous. The
geometry of the dam alone decides these.
After having located phreatic line in an earth dam the flow net can be plotted by trial and error by
observing the following properties of flow net (Fig 3.16), and by following the practical suggestions
given by A Casagrande.
Arthur Casagrande gives the following excellent hints for the beginner in flow net sketching:
1. Use every opportunity to study the appearance of well constructed flow nets. When the picture
is sufficiently absorbed in your mind, try to draw the same flow net with out looking at the
available solution: repeat this unit you are able to sketch this flow net in a satisfactory manner.
2. Four or five flow channels are usually sufficient for the first attempt; the use of too many flow
channels may distract the attention from the essential features.
3. Always watch the appearance of the entire flow net. Do not try to adjust details before the
entire flow net is approximately correct.
4. The beginner usually makes the mistake of drawing too sharp transitions between straight and
curved sections of flow lines or equipotential lines. Keep in mind that all transitions are
smooth; of elliptical or parabolic shape. The size of the squares in each channel will change
gradually.
Stability analyses under the following four heads are generally needed:
1. Stability analysis of down stream slope during steady seepage.
2. Stability of up stream slope during sudden Draw down.
3. Stability of up stream & down stream slope during and immediately after construction.
4. Stability of foundation against shear.
It is one of the most generally accepted methods of checking slope stability. In this method the
potential surface is assumed to be cylindrical, and the factor of safety against sliding is defined as the
ratio of average shear strength, as determined by Coulomb’s equation
S = C + tan to the average shearing stress determined by static’s on the potential sliding surface.
In order to test the stability of the slope, a trial slip circle is drawn, and the soil material above
assumed slip surface is divided in to a convenient number of vertical strips or slices. The trail sliding
mass (i.e. the soil mass contained with in the assumed failure surface) - is divided in to a number
(usually 5 to8) of slices which are usually, but not necessarily, of equal width. The width is so
chosen that the chord and arc subtended at the bottom of the slice are slice passes through material of
one type of soil.
The forces between the slices are neglected and each slice is assumed to act independently as a
column of soil of unit thickness and width b. The weight W of each slice is assumed to act at its
centre. If this weight of each slice is resolved in normal (N) and tangential (T) components, then the
Civil Engineering Department 21
Hydraulic Structure I Embankment Dam
normal component will pass through the center of rotation (O), and hence does not cause any driving
moment on the slice. However, the tangential component T cause a driving moment
= T (T*r), Where r is the radius of the slip circle. The tangential components of the few slices at the
base may cause resisting moment; in that case T is considered negative.
If c is the unit cohesion and L is the curved length of each slice, then resisting force from Column’s
equation is = c L + N tan
For the entire slip surface AB, we have
Driving moment Md = rT
Resisting moment Mr = cL tan N Where T = sum of all tangential components
N = sum of all normal components
2r
L= L= length AB of slip circle
3600
Fellenius gave the method of locating the locus on which probable centers of critical slip circle may
lie. He gives direction angles to be plotted at heel measured from the outer slope and to be
plotted from horizontal line above the top surface of the dam. These two lines plotted with given
direction angle intersect at point P. Point P is one of the centers. To obtain the locus we obtain point
Q by taking a line H m below the base of the dam and 4.5 H m away from toe. When the line PQ is
obtained, trial centers are selected around P on the line PQ and factor of safety corresponding to each
centre calculated from Equation given above as ordinates on the corresponding centers, and a smooth
curve is obtained. The centre corresponding to the lowest factor of safety is then the critical centre.
Critical condition for d/s slope occurs when the reservoir is full and percolation is at its maximum
rate. The directions of seepage forces tend to decrease stability. In other words, the saturated line
reduces the effective stress responsible for mobilizing shearing resistance.
cL tan ( N U )
F .S . ….. (3.15)
T
When U is the total pore pressure on the slope surface
The pore-water pressure at any point is represented by the piezometric head (h w) at that point. Thus
the variations of pore water pressure along a likely slip surface is obtained by measuring at each of
its intersections with an equipotential line, the vertical height from that intersection to the level at
which the equipotential line cuts the phreatic line. The pore pressure represented by vertical height so
obtained are plotted to scale in a direction normal to the sliding surface at the respective point of
intersection. The distribution of pore water pressure on the critical slope surface during steady
seepage is shown hatched in fig.3.19.The area of U- diagram can be measured with help of a
planimeter.
In the absence of a flow net, the F.S of the d/s slope can approximately be from the equation
cL tan N '
F .S . … (3.16)
T
The following unit weights may be used for the calculation of N ' and T when pore pressure are
otherwise not included in the stability analysis, however the Phreatic line needs to be drawn.