CEWB4013/CEWB493 Hydropower and Dam Engineering, Lecturer: Dr.

Wong Leong Sing

Module 4: Design Embankment and Gravity Dam:

strength, stability and safety factors against slope
stability, Phreatic line, seepage flow discharge
4.1 Embankment dam engineering
4.2 Gravity dam engineering

1
4.1 Embankment dam engineering
4.1.1 Principles of embankment dam design
[Link] Types and key elements

• In its simplest and oldest form the embankment dam was constructed with low-
permeability soils to a nominally homogeneous profile.
• It was then increasingly recognized that, in principle, larger embankment dams
required two component elements.
1. an impervious water-retaining element or core of very low permeability soil, e.g. soft
clay or a heavily remoulded ‘puddle’ clay and
2. supporting shoulders of coarser earthfill (or of rockfill), to provide structural stability.

2
[Link] Defect mechanisms, failure modes and design
principles

• The principal defect mechanisms and failure modes identifiable with embankment
dams are illustrated in schematic form in Figure 4.1.
• Certain mechanisms are interrelated, e.g. overtopping may result from inadequate
spillway capacity or from a lack of freeboard which may, in turn, be the result of long-
term deformation and settlement.
• The internal erosion and overtopping mechanisms are of particular concern, each
being responsible for c. 30–35% of serious incidents and failures.

3
Figure 4.1:
Illustrative
embankment
defect mechanisms
and failure modes

4
• In drawing attention to the principal risks to be guarded against the schematic diagrams of
Figure 5.1 also highlight the principal design considerations:
1. Overtopping and freeboard. Spillway and outlet capacity must be sufficient to pass the
design maximum flood without overstopping and risk of serious erosion and possible
washout of the embankment. Freeboard, i.e. the difference between maximum watertight
(i.e. core) level and minimum crest level of the dam, must also be sufficient to accept the
design flood plus wave action without overtopping, and must include an allowance for the
predicted long term settlement of the embankment and foundation.
2. Stability. The embankment, including its foundation, must be stable under construction and
under all conditions of reservoir operation. The face slopes must therefore be sufficiently flat
to ensure that internal and foundation stresses remain within acceptable limits.
3. Control of seepage. Seepage within and under the embankment must be controlled to
prevent concealed internal erosion and migration of fine materials, e.g. from the core, or
external erosion and sloughing. Hydraulic gradients, seepage pressures and seepage
velocities within and under the dam must therefore be contained at levels acceptable for the
materials concerned.
4. Upstream face protection. The upstream face must be protected against local erosion as a
result of wave action, ice movement, etc.
5. Outlet and ancillary works. Care must be taken to ensure that outlet or other facilities
constructed through the dam do not permit unobstructed passage of seepage water along
their perimeter with risk of soil migration and piping.
• Details of the more important defect mechanisms in relation to illustrative examples of causes
and preventive measures are set out in Table 4.1.
5
Table 4.1: Embankment dam defect mechanisms and preventive measures

6
Table 4.1 … Continued

7
[Link] Design features and practice

• The considerations summarized in Section [Link] have major implications with regard to certain design features and
good construction practice.
• Some of the more important points are outlined below.
(a) Zoning of shoulder fills
The careful and correct zoning of the available materials is an important aspect of embankment design. The principles are
as follows.
1. The core width should be as great as is economically viable.
2. The downstream shoulder should be underlain by a drainage blanket, or base drain, of free-draining material.
3. Finer shoulder material should be zoned closest to the core, with an intervening vertical drainage zone connecting to
the base drain (see Figure 4.2(b)).
4. Shoulder zones should be of progressively coarser material as the face slopes are approached.
5. Where a major change in the characteristics of material in adjacent zones is unavoidable, interface effects should be
eased by the insertion of an intermediate or transition zone.

8
• The permeability of successive zones should increase toward the outer slopes,
materials with a high degree of inherent stability being used to enclose and
support the less stable impervious core and filter.
• Pervious materials, if available, are generally placed in upstream sections to
permit rapid porewater pressure dissipation on rapid drawdown (Figure 4.2(b)).
• The stability of an embankment and its foundation is determined by their
collective ability to resist shear stresses.
• Embankments constructed with cohesive materials of low permeability
generally have slopes flatter than those used for zoned embankments, which
have free-draining outer zones supporting inner zones of less pervious fill
material.
• Lower quality random fill materials may be satisfactorily employed in areas
within the dam profile where neither permeability nor shear strength is critical
and bulk and weight are the primary requirements.
• Examples include the placing of stabilizing fill at the toes of embankments on
low-strength foundations or so-called ‘random zones’ within the heart of either
shoulder.
9
• General points regarding zoning and core profile include the following.
1. A nominal hydraulic gradient through the core of the order of 1.5–2.5 is
satisfactory; a value greater than 3.5–4 is undesirable.
2. Core geometry is not critical provided that the upstream core slope is
not such as to control overall slope stability. It is preferable that the core
be approximately central and it can ease placing of the downstream
drain and transition/filter zones if that face of the core is kept vertical.
3. The dimensioning of structural zones or intermediate transition zones is
governed by stability and deformation considerations. It is sometimes
assisted by data from construction of special test fills.
4. A compromise must be made between sophistication of design and
zoning and ease of construction. Internal zoning and associated
specification requirements should be kept as simple as possible.

10
(b) Spillway location
• Geotechnical and hydraulic design considerations require that to
minimize the risk of damage to the dam under flood conditions the
spillway and discharge channel are kept clear of the embankment.
• Spillways are therefore generally built on natural ground with the channel
bypassing the flank of the dam and discharging to a stilling basin clear of
the downstream toe.
• The alternative is to use a dropshaft-type spillway located within the
reservoir and discharging via an outlet tunnel or culvert.
• In the latter case it is preferable to tunnel the outlet through the natural
ground of the abutment wherever possible.
• The alternative is a concrete culvert if founded on incompressible rock.
• In a number of instances auxiliary or emergency spillways have been
constructed on embankment crests using suitable proprietary grass
reinforcement techniques.
11
(c) Freeboard
• The provision necessary for long-term settlement within the overall
minimum freeboard is determined by the height of dam and the depth of
compressible foundation at any section.
• It is therefore customary to construct the crest of the dam to a
longitudinal camber to accommodate the predicted consolidation
settlement.
• A proportion of the design freeboard is sometimes provided by
construction of a continuous concrete wavewall along the upstream edge
of the dam crest.
• This can also be done when it is necessary to uprate the freeboard of
older dams following reassessment of the design flood.
• The overall minimum freeboard from spillway sill to dam crest (or top of a
structural wave wall) should be at least 1.0 m on the smallest reservoir
embankment, and it will be very much greater for larger embankments
and/or reservoirs.
12
(d) Foundation seepage control
• Seepage flows and pressure within the foundation are controlled by
cutoffs and by drainage.
• Cut-offs are impervious barriers which function as extensions of the
embankment core into the foundation.
• They are generally located directly under the core, but can also be
located a short distance upstream and connected to the core by an
impervious horizontal blanket under the shoulder.
• The cut-off may penetrate to impervious strata (a ‘fully penetrating’ cut-
off) or, if pervious material occurs to considerable depths, may terminate
where the head loss across the cut-off is sufficient to effect the required
degree of control (a ‘partially penetrating’ cut-off).
• Older cut-offs were frequently constructed as very narrow clay filled
‘puddle trenches’, many proving vulnerable to hydraulic fracture,
seepage damage and erosion. The principal variants of cutoff now
employed are illustrated schematically in Figure 4.2.
13
Figure 4.2: Cut-offs
and control of
under seepage

14
(e) Outlet works: tunnels and culverts
• Outlet works should where practicable be constructed as a tunnel driven
through the natural ground of the dam abutment.
• Where this is difficult or uneconomic a concrete culvert founded on rock is a
satisfactory alternative, provided that care is taken to ensure that, if not
founded in a shallow rock excavation, a relatively rigid projecting culvert does
not promote shear cracking of the embankment fill as the latter settles.
• The culvert cross-section should therefore be ogival, with transverse joints at
intervals of 10–15 m and an external slip coating to assist settlement of the fill
relative to the culvert.
• It is potentially dangerous to construct an outlet culvert on a compressible
foundation within the fill itself owing to the effects of differential deformations
and possible cracking.
• Concrete culverts sometimes incorporate a number of external transverse
seepage collars or plates at intervals along their length to inhibit preferential
seepage and possible erosion at the culvert-fill interface (USBR, 1987). The
effectiveness of such features can be questioned.
15
((f) Upstream face protection
• Several options are available for protection of the upstream face against wave
erosion, ranging from traditional stone pitching with grouted joints through concrete
facing slabs to the use of concrete blockwork and rock armouring.
• A heavy and thick protective layer is necessary between crest level and the minimum
operating or drawdown level, with reduced protection typically in the form of
beaching, i.e. smaller uniform rocks, provided down to the reservoir bed.
• Pitching is expensive and, while durable, it is not an efficient dissipator of wave
energy.
• Concrete slabs have similar limitations, allowing considerable wave run-up which has
to be included within the design freeboard.
• Open-jointed heavy concrete blockwork bedded on gravel and a granular filter is
preferable, and is now widely employed.
• Where durable rock is available in large, angular sizes, rock armouring and dumped
rock riprap provide the most effective protection.
• The wave protection facing plus under-layer interacts with the immediately
underlying embankment fill to perform as a composite system.
• The performance of the overall protection is particularly dependent upon the under-
layer fulfilling the following functions:
16
• filtration, to resist movement of embankment fill by wave-induced under-
layer water flows;
• drainage, to relieve wave-induced uplift forces under the facing;
• erosion protection at the interface between under-layer and embankment
fill;
• separation, to prevent the protection layer penetrating the fill.
The principal characteristics of the outer face protection system which
influence the under-layer are:
• flexibility, controlling the degree to which inter-layer contact can be
maintained;
• permeability, determining the extent to which hydraulic conditions are
transferred into the under-layer.

17
(g) Embankment crest
• The crest should have a width of not less than 5 m, and should carry a
surfaced and well-drained access road. (With old dams the latter
provides valuable resistance to the erosive effect of occasional limited
and short term overtopping.)

18
4.1.2 Materials and construction
[Link] Earthfill materials
Three principal categories of fill material are necessary for earthfill
embankment dams to fulfil the requirements for core, shoulder and
drainage blankets, filters, etc.
1) Core fill should have low permeability and ideally be of intermediate to
high plasticity to accommodate a limited degree of deformation without risk
of cracking.
• It is not necessary, and possibly disadvantageous, to have high shear
strength. The most suitable soils have clay contents in excess of 25–
30%, e.g. glacial tills etc., although clayey sands and silts can also be
utilized.
• The core is the key element in an embankment and the most demanding
in terms of material characteristics and uniformity, the properties of the
compacted clay core being critical to long-term watertight integrity.
• The principal performance characteristics of the more important groups
of soils suitable for rolled cores are summarized in Table 5.2. 19
Table 4.2 Characteristics of core soils

20
• Representative values for the more important engineering parameters of
compacted earthfills, to be utilized in core or shoulders as appropriate,
are summarized in Table 4.3.
• Note that the comparable parameters for compacted rockfill are also
tabulated.

21
Table 4.3 Indicative engineering properties for compacted earthfills

22
2) Shoulder fill requires sufficiently high shear strength to permit the economic
construction of stable slopes of the steepest possible slope angle.
• It is preferable that the fill has relatively high permeability to assist in
dissipating porewater pressures.
• Suitable materials range across the spectrum from coarse granular material to
fills which may differ little from the core materials.
• The shoulder need not be homogeneous; it is customary to utilize different fills
which are available within predetermined zones within the shoulders.
3) Drain/filter material must be clean, free draining and not liable to chemical
degradation.
• Processed fine natural gravels, crushed rock and coarse to medium sands are
suitable, and are used in sequences and gradings determined by the nature of
the adjacent core and/or shoulder fills.
• The cost of processed filter materials is relatively high, and the requirement is
therefore kept to a minimum.
23
[Link] Construction
The construction operations which follow initial site development fall into
four principal groups of activities, relating to (1) material source
development, (2) foundation preparation and construction, (3) fill
construction and control and (4) ancillary works construction.
1) Material source development activities involve the opening out of
borrow areas or quarries, including the installation of fixed plant, e.g.
crushers, conveyors, etc. Access and haulage roads are also constructed
between the various borrow areas and the embankment site, and
excavation and haulage plant is mobilized.

24
2) Foundation preparation activities, including river diversion, can
proceed concurrently with the development of the fill sources. Temporary
river diversion is commonly effected by driving a flanking tunnel, which in
most cases subsequently houses the outlet works. (Where an outlet culvert
through or under the embankment is planned rather than a flanking tunnel
the culvert may be used temporarily for river diversion purposes.)
Topsoil and weathered surface drift deposits etc. are removed. In the case
of a soft, compressible foundation, strength can be enhanced and
construction accelerated by preconsolidation and/or the installation of sand
drains, as for the Derwent dam, constructed on a difficult site in the UK
(Ruffle, 1970).
Foundation instrumentation is also installed at this stage to monitor pore
pressures and cut-off performance. Foundation construction is completed
with the laying of the drainage blankets which will underlie the downstream
shoulder.
25
3) Fill construction is an exercise in efficient plant utilization within the terms of
the specification requirements as to materials compliance and compaction
technique. Placing operations may be subject to the influence of weather
conditions and to subtle changes in material characteristics.
Control of placing is centred upon supervision of water content, layer thickness
and compaction procedure. The quality and uniformity of the compacted core fill
are critical. Recent UK practice is to employ a statistical approach to testing and
quality control and to require that the undrained shear strength, cu, should lie
within specific limits. A specification requirement of cu = 55–100 kNm2 is typical
for UK clays (Kennard et al., 1979). Control of construction porewater pressures
and acceleration of consolidation in cohesive fill materials of low permeability
can require the installation of horizontal blanket drains in both shoulders, at
vertical intervals of 3–6 m (Figure 4.3 (a) and (b); Gibson and Shefford (1968)).
Knight (1989) reviews aspects of fill construction and control in terms of recent
practice on a number of projects.
The installation of instrumentation in the core and shoulders proceeds in parallel
with the placing of fill. Fill construction is concluded with the completion of
upstream rock armouring or other face protection works.
26
Figure 4.3:
Representative UK
rolled clay core
embankment
dams: (a) Kielder
(1982); (b) Lower
Lliw (1978) (after
Binnie, 1981)

27
4) Ancillary works construction embraces the construction of spillway and
stilling basins, culverts or tunnels for outlet works etc., valve towers and
similar control works. It also includes completion of crest details, e.g.
roadway, drainage works, wavewall etc. and, where climatic conditions
allow, grassing of the downstream face slope.

28
4.1.3 Seepage analysis
[Link] Seepage
• The phreatic surface of the seepage régime, i.e. the free surface, must
be kept well clear of the downstream face to avoid high porewater
pressures which may promote slope instability.
• In the extreme case of the seepage line emerging on the face, local
softening and erosion will occur and may initiate sloughing as a prelude
to instability.
• Seepage pressures and velocities must also be controlled to prevent
internal erosion and particle migration, particularly from the core.
• Seepage control is effected by the incorporation of vertical chimney
drains and horizontal drainage layers, protected by suitable filters and
transition layers.
• In this section a basic knowledge of seepage theory and flownet
construction is assumed, including entry and exit conditions.
29
The fundamental relationships applicable to flownets for two dimensional
seepage are summarized below.
For anisotropic soils, with the coefficient of horizontal permeability kh > kv,
the coefficient of vertical permeability, the horizontal scale transform factor,
λ, and the effective permeability, k’, are respectively given by

(Equation 4.1)
and (Equation 4.2)

The seepage flow, q, is defined by

(Equation 4.3)

where H is the head differential and the ratio Nf/Nd is the flownet shape
factor, i.e. the number of flow channels, Nf, in relation to the number of
decrements in potential, Nd (Nf and Nd need not be integers). 30
• For the unconfined flow situation applicable to seepage through a
homogeneous embankment the phreatic surface is essentially parabolic.
• The curve can be constructed by the Casagrande–Kozeny
approximation, defined in the references previously detailed, or from
interpretation of piezometric data (Casagrande, 1961).
• In the case of a central core and/or zoned embankment, construction of
the flownet is based upon consideration of the relative permeability of
each element and application of the continuity equation:
(Equation 4.4)
• An illustrative flownet for seepage under an embankment is given in
Figure 4.4. (The figure forms the transform scale solution to worked
example 4.1.)
• In Figure 4.5 is shown the flownet for a simple upstream core two-zone
profile, where piezometric data have been interpreted to define the
phreatic surface within the core (Figure 4.5 is the flownet solution to
worked example 4.2).
31
• The thickness of horizontal blanket drain, td, required to discharge the
seepage flow and shown in Figure 4.5 can be estimated from
(Equation 4.5)
• where L is the downstream shoulder width at drain level and kd and kc
are the drain and core permeabilities respectively (the factor 1.5 in
equation (4.5) is derived from a representative embankment geometry).

32
Figure 4.4:
Idealized flownet
for foundation
seepage (Worked
example 5.1 should
also be referred to)

33
Figure 4.5: Flownet
for internal
seepage in
embankment core
(Worked example
5.2 should also be
referred to)
In Worked example 5.2:

34
[Link] Core and cut-off efficiency
The effectiveness of a core or cut-off may be empirically defined in terms
of two criteria (Telling, Menzies and Simons, 1978):
1. head efficiency,
EH = h/H (Equation 4.6)
where h is the head loss across the core or cut-off and H is the overall
differential head, as shown in Figure 4.4;
2. flow efficiency,
EQ = 1 - Q/Q0 (Equation 4.7)
where Q and Q0 are, respectively, the seepage flows with and without the
core or cut-off in position.
EH may be determined from the piezometric levels upstream and
downstream of core or cut-off, and EQ from flow measurements. Both may
be approximated from flownet studies. 35
For the special case of a fully penetrating cut-off and a parallel boundary
flow régime efficiencies EH and EQ are related thus:

(Equation 4.8)

where k1 and k2 are the coefficients of permeability of natural foundation

and cut-off zone respectively.
Efficiencies EH and EQ recorded for well constructed and effective cut-offs
are normally both in excess of 50–60%.

36
[Link] Filter design
• The design of filters and transition layers to prevent seepage-induced
migration of fines is discussed in soil mechanics texts and, in greater
detail, in Mitchell (1983) and Sherard and Dunnigan (1985).
• They are required to be sufficiently fine to prevent migration of the
protected soil (piping criterion) while being sufficiently permeable to
freely discharge seepage (permeability criterion).
• The essential principle of design is that any change from fine to coarse
material must be effected gradually in staged filter or transition zones,
e.g. clay core → sand → coarse sand → pea gravel → coarse
shoulder, i.e.

37
• A widely employed empirical approach to defining appropriate filter
material grading envelopes is given in the form of ratios for specified
particle passing sizes as typified by the expressions

(Equation 4.9a)

(Equation 4.9b)

(Equation 4.9c)

38
• where D15 etc. refer to the 15% passing size etc. as determined from
particle size analysis. Expressions (4.9a) and (4.9b) set out piping and
permeability criteria respectively; expression (4.9c) further defines
permeability ratio.
• A recent development is the suggestion that, in view of the potential
problem of hydraulic fracturing and cracking etc., with risk of progressive
erosion, rational filter and transition design should be based on
considerations of relative permeability (Vaughan and Soares, 1982).
• This approach introduces the concept of basing specification of the filter
to protect a clay core on considerations of seepage water chemistry and
clay floc or particle size in relation to filter void size and effectiveness.
• It is considered to define the way ahead in filter design for the protection
of very fine soils such as clays, the empirical approach based on particle
size ratios having proved unsafe in a number of instances (Vaughan et
al., 1970).
39
Worked Example 4.1
The outline profile of an earthfill embankment is illustrated in Figure 5.6.

Figure 4.6: Worked example 4.1 (Figure 4.4)

40
GEOTECHNICAL DATA
The foundation alluvium has a permeability (antisotropic) of kh = 4kv= 1 ×
10-5 ms-1. The design effective permeability of the grouted cut-off zone, kc
= 2.5 × 10-7 ms-1.
1. Draw the foundation flownet without a grouted cut-off in position and
calculate seepage flow, Q0, in m3 m-1 day-1.
2. Determine the width, t, of grouted cut-off zone required to achieve a
design EQ of 70% and compute the corresponding value of EH.
3. Determine the corresponding porewater pressure at point X in m head.

41
Solution:
Figure 5.4:

(Equation 4.3)

42
N’d

Fig. 4.4

43
Worked Example 4.2
The profile of a zoned earthfill embankment founded on impervious clay is
illustrated in Figure 4.5, and steady-state piezometric levels recorded
within the embankment are indicated.
DATA
Measured seepage at the V-notch, q = 0.1m3 day-1per m length. The core
zone consists of soil A. The shoulders are of soil B downstream and
coarse rockfill upstream. The drains are of fine gravel.
1. Interpret the piezometric levels and construct a flownet representative of
the steady-state seepage régime.
2. From the flownet estimate the effective permeability k’, of soil A and the
head efficiency, EH of the core.

44
Solution:
1. See the flownet in Figure 4.5, noting (1) the interpretation of zone
permeabilities, and (2) the construction of the phreatic surface, with
subdivision of the head, H, into increments, Δh.
2. From the flownet: Nf = 3.6, mean Nd ≈ 4.9, and shape factor Nf/Nd =
0.73.

(Equation 4.3)

45
4.1.4 Stability and stress
[Link] Stability analyses
• Embankment dam stability must be assessed in relation to the changing
conditions of loading and seepage régime which develop from
construction through first impounding into operational service, including
reservoir drawdown.
• Two-dimensional limit-equilibrium analysis is based on consideration of
the static equilibrium of the potentially unstable and ‘active’ mass of soil
overlying a conjectural failure surface.
• The factor of safety, F, is defined by

(Equation 4.4)

46
• where τf and τ are, respectively, the unit shear resistance which can be
mobilized and unit shear stress generated on the failure surface.
• The analysis is applied to all conceivable failure surfaces, and the
supposed minimum factor of safety Fmin is sought.
• The form of the critical failure surface for Fmin is controlled by many
factors, including soil type and the presence of discontinuities or
interfaces, e.g. between soft soil and rock.
• A number of failure surfaces representative of different embankment
and/or foundation situations are illustrated schematically in Figure 4.7.
• For most initial analyses involving relatively homogeneous and uniform
cohesive soils, two dimensional circular arc failure surfaces are
assumed.

47
Figure 4.7: Stability analysis:
failure surface schematics

48
• The following critical conditions must be analyzed (see Figure 4.8):
1. end of construction (both slopes);
2. steady state, reservoir full (downstream slope critical);
3. rapid drawdown (upstream slope critical);
4. seismic loading additional to 1, 2 and 3, if appropriate to the location.

• Analysis is in terms of effective stress shear strength parameters c’ and

Φ’, with porewater pressures, uw, or the pore-pressure ratio, ru.
• Total stress parameters, c and Φ, are suitable only for a short-term and
approximate analysis, e.g. stability at an intermediate stage during
construction.

49
• Values of Fmin determined in a comprehensive stability analysis must
always be regarded as relative and not as absolute.
• The expressions determining F are of varying rigour and are inexact, a
reflection of the complexity of the stability problem, where measured
shear strength parameters can be subject to a variance of up to 30–40%.
• Economic considerations require acceptance of relatively low values of F
for embankment slopes.
• The factors of safety for embankment dam slopes are time dependent,
varying significantly in accordance with changes in loading
corresponding to construction and the subsequent operational cycle. This
is illustrated schematically in Figure 4.8.

50
Figure 4.8: Schematic of the
variation of embankment
stability parameters
during construction and
operation (after Bishop and
Bjerrum, 1960)

51
• Representative guideline values of Fmin corresponding to the major
loading conditions are presented in Table 4.4.
• Values considered acceptable must always reflect the confidence
attaching to geotechnical data and design parameters.
Table 4.4: Guideline factors of safety: effective stress stability analysis

52
• The expressions for F corresponding to the analytical methods most
commonly employed are as follows.
(a) Swedish circle (Fellenius) solution: circular arc surface

(Equation 4.5)

• where La is the overall length of the failure surface, W and l are,

respectively, the weight and base length of the slices into which the
active mass is subdivided for analysis and α is the angle of inclination of
the slice base to the horizontal. The latter is considered positive if
upslope from the lowest point on the failure arc.
• The Fellenius solution is, in practice, conservative by comparison with
more rigorous analyses, and may underestimate F by 5–15%. This
margin is generally unacceptable in terms of the cost implications.

53
(b) Bishop semi-rigorous solution: circular arc surface
• The Bishop semi-rigorous solution (Bishop, 1955) differs from the
Fellenius solution in the assumptions made with regard to the interslice
forces required for static equilibrium:

(Equation 4.6a)

• In the above iterative expression b is the width of any slice.

• Alternatively, expressing porewater pressure uw in terms of predicted
pore pressure ratio, ru, for convenience in initial analysis, with ru = uw/γz =
uwb/W for any slice:
(Equation 4.6b)

54
• In applying this method an appropriate trial value of F is first selected,
subsequent iteration resulting in the expression converging rapidly to a
solution.
• The Bishop expression may, with discretion, be applied to a non-circular
arc failure surface, as shown in Figure 4.9, which refers also to Worked
Example 4.3.
• Charts of ma = cos α [1(tan α tan Φ’)/ F] for use with Equation (4.6) are
presented in Figure 4.10.

55
Figure 4.9: Stability analysis: non-circular arc failure surface (Worked Example 4.3 should also
be referred to)

56
Figure 4.10: Stability analysis (Bishop semi-rigorous method): curves of function mα = cos α [1
+ (tan α tan Φ’)/F]
57
Worked Example 4.3
Figure 4.9 illustrates the section of a rolled clay embankment dam at an
intermediate state of construction.

Determine the factor of safety, F, with respect to the hypothetical failure

surface indicated using the semi-rigorous Bishop analysis (Equation
(4.6b): slice dimensions may be scaled from the figure).
58
Solution:
For an assumed value of F = 1.5, equation (4.6b) yields F = 1.71; iteration
(F = 1.71) confirms F = 1.73; iteration (F = 1.73) confirms F = 1.74.

Tabulated below are the effects on F of changing major parameters

involved in the analysis; e.g. c’, Φ’ and ru:

1.74

59
Trial and error
Bishop’s Semi Rigorous Method.
First trial, Try F = 1.5.

A B C
Slice no. b (m) h dry (m) W = γhb h w (m) α° l a (m) c ' (kPa) φ ' (°) [c 'b + (W -ub
1/[cos αAφ
) tan + x'] B φ' sinW α/F)}
(tan sin α
1 2.92 4.17 382.81 2.08 59.0 5.7 5.0 30.0 201.18 1.21 244.30 328.13
2 5.83 3.33 1173.96 6.25 32.0 6.9 5.0 30.0 500.46 0.97 483.46 622.10
3 5.83 1.25 1327.08 9.58 23.0 6.3 5.0 30.0 478.74 0.95 452.60 518.53
4 7.08 0.00 1425.52 9.58 1.8 7.1 5.0 30.0 473.97 0.99 469.06 44.78
5 7.08 0.00 1177.60 7.92 0.0 7.1 5.0 30.0 397.70 1.00 397.70 0.00
6 7.50 0.00 853.13 5.42 0.0 7.5 5.0 30.0 299.96 1.00 299.96 0.00
7 7.29 0.00 446.61 2.92 0.0 7.3 5.0 30.0 173.86 1.00 173.86 0.00
8 5.42 0.00 142.19 1.25 0.0 5.4 5.0 30.0 70.83 1.00 70.83 0.00
Total ∑ 2591.76 1513.55

F 1.71

60
Second trial, Try F = 1.71.

A B C
Slice no. b (m) h dry (m) W = γhb h w (m) α° l a (m) c ' (kPa) φ ' (°) [c 'b + (W -ub
1/[cos αAφ
) tan + x'] B φ' sinW α/F)}
(tan sin α
1 2.92 4.17 382.81 2.08 59.0 5.7 5.0 30.0 201.18 1.28 256.71 328.13
2 5.83 3.33 1173.96 6.25 32.0 6.9 5.0 30.0 500.46 0.99 494.65 622.10
3 5.83 1.25 1327.08 9.58 23.0 6.3 5.0 30.0 478.74 0.96 460.00 518.53
4 7.08 0.00 1425.52 9.58 1.8 7.1 5.0 30.0 473.97 0.99 469.69 44.78
5 7.08 0.00 1177.60 7.92 0.0 7.1 5.0 30.0 397.70 1.00 397.70 0.00
6 7.50 0.00 853.13 5.42 0.0 7.5 5.0 30.0 299.96 1.00 299.96 0.00
7 7.29 0.00 446.61 2.92 0.0 7.3 5.0 30.0 173.86 1.00 173.86 0.00
8 5.42 0.00 142.19 1.25 0.0 5.4 5.0 30.0 70.83 1.00 70.83 0.00
Total ∑ 2623.39 1513.55

F 1.73

61
Third trial, Try F = 1.73.

A B C
Slice no. b (m) h dry (m) W = γhb h w (m) α° l a (m) c ' (kPa) φ ' (°) [c 'b + (W -ub
1/[cos αAφ
) tan + x'] B φ' sinW α/F)}
(tan sin α
1 2.92 4.17 382.81 2.08 59.0 5.7 5.0 30.0 201.18 1.28 257.78 328.13
2 5.83 3.33 1173.96 6.25 32.0 6.9 5.0 30.0 500.46 0.99 495.59 622.10
3 5.83 1.25 1327.08 9.58 23.0 6.3 5.0 30.0 478.74 0.96 460.63 518.53
4 7.08 0.00 1425.52 9.58 1.8 7.1 5.0 30.0 473.97 0.99 469.74 44.78
5 7.08 0.00 1177.60 7.92 0.0 7.1 5.0 30.0 397.70 1.00 397.70 0.00
6 7.50 0.00 853.13 5.42 0.0 7.5 5.0 30.0 299.96 1.00 299.96 0.00
7 7.29 0.00 446.61 2.92 0.0 7.3 5.0 30.0 173.86 1.00 173.86 0.00
8 5.42 0.00 142.19 1.25 0.0 5.4 5.0 30.0 70.83 1.00 70.83 0.00
Total ∑ 2626.08 1513.55

F 1.74

62
Fourth trial, Try F = 1.74.

A B C
Slice no. b (m) h dry (m) W = γhb h w (m) α° l a (m) c ' (kPa) φ ' (°) [c 'b + (W -ub
1/[cos αAφ
) tan + x'] B φ' sinW α/F)}
(tan sin α
1 2.92 4.17 382.81 2.08 59.0 5.7 5.0 30.0 201.18 1.28 258.31 328.13
2 5.83 3.33 1173.96 6.25 32.0 6.9 5.0 30.0 500.46 0.99 496.06 622.10
3 5.83 1.25 1327.08 9.58 23.0 6.3 5.0 30.0 478.74 0.96 460.93 518.53
4 7.08 0.00 1425.52 9.58 1.8 7.1 5.0 30.0 473.97 0.99 469.76 44.78
5 7.08 0.00 1177.60 7.92 0.0 7.1 5.0 30.0 397.70 1.00 397.70 0.00
6 7.50 0.00 853.13 5.42 0.0 7.5 5.0 30.0 299.96 1.00 299.96 0.00
7 7.29 0.00 446.61 2.92 0.0 7.3 5.0 30.0 173.86 1.00 173.86 0.00
8 5.42 0.00 142.19 1.25 0.0 5.4 5.0 30.0 70.83 1.00 70.83 0.00
Total ∑ 2627.40 1513.55

F 1.74

63
4.1.5 Settlement and deformation
[Link] Settlement

(Equation 4.7)

(Equation 4.8)

(Equation 4.9)
64
(Equation 4.10)

65
(Equation 4.11)

(Equation 4.12)

66
[Link] Deformation

(Equation 4.13)

(Equation 4.14)

67
[Link] Performance indices for earthfill cores
• Progressive deterioration or inadequate performance of an earthfill core
may be indicated by excessive and possibly turbid seepage and leakage.
• Other indicators include localized crest or upstream face depressions,
excessive general settlement or a high phreatic surface in the
downstream shoulder.
• Empirical performance indices have been suggested for certain critical
parameters.

68
(Equation 4.15)

69
(Equation 4.16)

70
(Equation 4.17)

71
Worked Example 4.4
The outline profile of an embankment dam and its foundations are
illustrated in Figure 5.11. The dam has a central rolled clay core flanked by
shoulders of compacted rockfill.

Figure 4.11: Worked example 4.4 72

73
Solution:

(Equation 4.8)

(Equation 4.10)

(Equation 4.9)

74
(Equation 4.10)

(Equation 4.9)

75
76
2.0 × 104

77
4.2 Gravity dam engineering
4.2.1 Principles of gravity dam design

• Concrete dams are quasi-elastic structures and are intended to remain so at their
design level of seismic acceleration.
• They should also be designed to withstand an appropriate maximum earthquake,
e.g. CME (controlling maximum earthquake) or SEE (safety evaluation earthquake)
(Charles et al., 1991) without rupture.
• The possibility of structural resonance must also be investigated for higher dams,
although the risk of serious resonance is considerably reduced in practice by
damping effects.
• Seismic ground motions are in any event irregular in their magnitude, periodicity and
direction.
• They are therefore unlikely to sustain resonance for durations much exceeding a
few seconds.

78
Design and Stability of concrete dams
Dr. M. Hafez

Gravity Dam

79
Imagine that sink basin as a catchment area,
and the sink hole as a lake or a river.
Therefore, a catchment area or drainage basin
is an area of land, where the surface water
from rainfall, snowmelt, or ice flow into a
lower elevation to form a single body of water,
that may be a lake or a river.

80
you need to study the contours!

81
Level of the spillway

82
Gravity Dam

83
84
85
86
If the bedding planes, lamination or joints
are NORMAL to resultant . It should not be
any problem

If the bedding planes, lamination or joints

are PARALLEL to resultant . It will be a
problem

87
88
89
90
91
92
93
 Components of Dam
Crest: The top of the Dam. These may
in some cases be used for providing a
roadway or walkway over the dam.

Parapet walls: Low Protective walls on

either side of the roadway or walkway
on the crest.

Heel: Portion of Dam in contact with

ground or river-bed at upstream side.

Toe: Portion of dam in contact with

ground or river-bed at downstream
side.

94
 Components of Dam
Sluice way: Opening in the dam near the base,
provided to clear the silt accumulation in the
reservoir.

Free board: The space between the highest level of

water in the reservoir and the top of the dam.

Dead Storage level: Level of permanent storage

below which the water will not be withdrawn.

95
96
97
S=Sc

98
Free Board –IS 6512, 1972 specifies that the free board will be
1.5 times the wave height above normal pool level.
Top Width – The top width of the dam is generally fixed
according to requirements of the roadway to be provided. The
most economical top width of the dam is 14 % of its height.
Base Width – The base width of the dam shall be safe against
overturning, sliding & no tension in dam body.
For elementary profile –

When uplift is considered,

When uplift isn’t considered,

99
100
101
102
Horizontal Loads

Headwater (H1) - For the basic calculation of stability the level in the reservoir will be assumed
at or above the level required for the passage of the design flood. In many instances the dam is
designed for the highest level of watertightness, e.g. a concrete parapet.

103
104
Silt (H2) - A changed land usage as a result of a dam may well result in increased
erosion, causing a deposition of silt. Unless very deep deposits of silt are likely it is
adequate to assume a triangular load allotting an appropriate relative density to
the fluid.
Reservoir Behaviour (H3) - Wind and other natural causes will induce movement in
the reservoir as waves, reservoir set-up or seiche effect.

Ice Loading (H4) - It is assumed that ice will not form and exert pressure on the dam
at times of maximum flood. The slope of the upstream face of the dam as well as
the slope and roughness of the valley walls will influence the magnitude of ice
loading. Even wind blowing down the reservoir at 50km/hr may increase the ice
loading by 4-5 t/m of exposed face.

Tailwater (H5) - In some cases water is ponded downstream from the dam.
Assistance from this may be assumed but it must not be overlooked that, in the
case of an overflow dam, flood waters passing over the dam might well evacuate
such water from the face of the dam.
Seismic Force (H6) - Force acting on dam in horizontal plane.
Seiche effect (H7) - Is an undulation in the reservoir water due to natural causes,
intermittent wind, variation in atmospheric pressure, earthquake and motion of the
Earth. Usually less than 0.5m.
105
106
107
108
109
110
 illustrates the uplift pressure on a
concrete gravity dam‟s non
overflow section through two
planes – one at the base and the
other at the horizontal plane
which is above the tail water
level.

the drainage holes either in the

body of the dam, or within the
foundation has not been
considered. If the effects of the
drainage holes are considered,
then the uplift pressure diagram
gets modified as shown in

111
If there is crack at any plane
of the dam, or at the
base then the uplift pressure
diagram gets further modified
as shown in

112
As such, the uplift pressure is
assumed to act throughout
the base area. Further it is
also assumed that they
remain unaffected by
earthquakes

113
114
115
116
 F) EARTHQUAKE FORCE (Fd):

Acting horizontally and vertically at the center of gravity

k (earthquake coefficient): Ratio of earthquake acceleration to

gravitational acceleration.

117
118
 G) DYNAMIC FORCE (Fw) :

In the reservoir, induced by earthquake as below

 Acts at a distance 0.412 h1 from the bottom

• Fw : the force per unit width of dam

• C : constant given by

'

• θ’ : angle of upstream face of the dam from vertical (oC)

• For vertical upstream face  C = 0.7

119
H) FORCES ON SPILLWAYS (∑F):

Determined by using momentum equation btw two successive

sections:

• ρ : the density of water

• Q : the outflow rate over the spillway crest
• ΔV: the change in velocity between sections 1 and 2 (v2-v1)

 Momentum correction coefficients can be assumed as unity.

120
 Concrete Gravity Dams

 For the dam dimensions:

 Check out the safety for

• Overturning

• Shear & sliding

• Bearing capacity of foundation

• No tensile stresses are allowed in the dam body

121
122
Concrete Gravity Dams

Design Criteria:

123
Overturning Check

1/md
H

124
Overturning Check

125
Overturning Check

126
Overturning Check

127
Overturning Check

128
Overturning Check

129
Sliding Check

1/md
H

130
Sliding Check

131
Sliding Check

132
Sliding Check

133
Bearing Capacity Check

1/md
H

134
 3.5.2 STABILITY CRITERIA

 Dam must be safe against

 (1) Overturning for all loading conditions

Mr  resisting moments

FSO  
Mo overturning moments

Safety factor:

 [Link]  2,0 (usual loading)

 [Link]  1,5 (unusual loading)

135
136
STABILITY
CRITERIA

 (2) Sliding over any horizontal plane

f V
FS s 
H
 f = friction coef. btw any two planes

Safety factor:
 FSS  1,5 (usual loading )
 FSS  1,0 (unusual or severe loading)

137
138
STABILITY
CRITERIA

 (3) Shear and sliding together

f åV + 0.5A t s
FSss =
åH
A : Area of shear plane (m²)
τs : Allowable shear stress in concrete in contact with foundation

Safety factor:
 FSss  5,0 (usual loading)
 FSss  3,0 (unusual or severe loading)

139
140
141
 STABILITY
CRITERIA
 (4) Between foundation and dam contact stresses (σ) > 0
at all points

There are two cases for the base pressure:

Mr - Mo B
x= e= -x
åV 2

142
Base Pressure Check
• CASE 1: e  B/6
æ åV ö æ ö
B
Pt = ç ÷ ´ ç1+ 6e ÷ Pt  s
ç B ÷ è Bø
è ø DAM BASE

æ åV ö æ ö
Ph = ç ÷ ´ ç1- 6e ÷ Ph  s Ph
ç B ÷ è Bø Pt
è ø
e x

ΣV

143
Base Pressure Check

B
CASE 2: e > B/6
DAM BASE

æ ö Pt
æ åV ö ç ÷
1
Pt = ç ÷´ç ÷ e x
ç B ÷ ç æ 3 ö æ 1 e ö ÷ Pt  s
è ø çç ÷´ç - ÷÷
èè 2 ø è 2 B øø
ΣV

144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159