1

Design of Diversion Weirs

I. Site Investigation, Topographic Survey, Soil Survey and Hydrologic
Analysis
Many of the rivers are seasonal but with adequate discharges to develop small scale
irrigation schemes. The diversion weir, although apparently a simple structure across a
river, is a piece of engineering work which needs careful design and thorough
hydrological, hydraulic and structural analysis. The view of collapsed hydraulic
structures, including diversions weirs, in not uncommon.

Site Investigation

I- Social and Economic Aspects

The project planner and the designer, during the first investigations, seek answers to a
few questions which are crucial for deciding the scheme’s feasibility. These questions are:

1. Coherence of the farming society.

The existence of a developed social system is an essential element of the project
feasibility. Construction a structure in an area where the majority of the population are
nomads and do not have strong bonds with agricultural lands is definitely unjustifiable.
Experiences in many developing countries show that, whenever decisions for
constructing a vital irrigation scheme are taken for other than socio-economic reasons,
the result always has been a total failure. Therefore, it is important to discover that there
is already in the area a farming community, though on a small scale and that they have a
system of cooperation, be it on the tribal and tradition basis or a more developed
organization such as a water user’s association.

The existence of a kind of farmer’s organization indicates that the scheme, when it is
built, is likely to be operated and well maintained by the farmers. It also indicates the
possibility of the project cost recovery, if so required by the government. Through their
own organization and tradition, annual taxes can be collected and used for the capital
recovery, operation and maintenance costs. Initially a strong farmer organization may not
exist, but a nucleus for such an organization must be there.

Nowadays many weirs in developing countries are constructed by farmer themselves.

Therefore mobilization of the workforce is a serious task which is to be carried out by the
farmers’ representative. It is therefore important to find out whether such a farmer
organization exists or not.

2. Will the proposed structure create any conflict?

 2

One sensitive issue in constructing weirs is the water rights of the downstream water
users. Farmers and the local authorities need to be questioned again and again to establish
the water rights and a plan should be drawn to evaluate these rights quantitatively. The
water rights at the structure directly affects the design of the main canal intake and size of
the irrigated area.

The structure may force the local community to relocate their bathing place and livestock
watering point. Its location may be a sacred place of the local community. In some
rehabilitated irrigation schemes, the farmers need to change their irrigation habits, for
example, from day only to day and night irrigation, in order to expand the irrigated area.

3. The need for miscellaneous structure.

Constructing diversion weirs, sometimes with a little extra cost, can make the farmer’s
life a lot easier. For example, providing a foot bridge, and a washing bay. Constructing a
washing bay where a structure constructed on a seasonal river is almost a necessity. If it
is not provided, in dry seasons the apron is used as a washing platform and deep ditches
are dug downstream of the cut-off walls to bail out water. This cause a serious piping
problem in flood seasons.

The above issues must be discussed with the farmers and points of conflict should be
avoided.

II- Technical Considerations

A design engineer who visits the site, for the first time is seeking answers to the
following vital questions, which directly affect the design.

1. Location of the weir

Initially, it is difficult to decide on the location of the proposed structure without having
topographic maps of the project area and layout of the river course. However, by walking
along the river up and downstream of the location where the existing intake is or where
the farmers believe it is an appropriate location, it is possible to identify a few places for
the proposed structure. The engineer, at this stage, considers the following in selecting
the structure site.

Location of the irrigated area

If the selected structure site is too far away from the agricultural land which it serves, it
means a long main canal is required, and hence the need for a high capital investment. On
the other hand, if the site is too close to the proposed land, some of the area in the upper
reach of the main canal cannot be commanded. In a situation like this, the design
engineer should carry out a rough economic analysis to determine the economic merits of
 3

the two possibilities, although the economic factor is not always the decisive one. In
many cases, the engineer finds himself in a dilemma between considering only the
technical and economical factors, or sacrificing some of these to fulfil all the farmers’
wishes. In many traditional irrigation projects, it is hard to convince farmers, wishes. In
many traditional irrigation projects, it is hard to convince farmers’ who may have been
cultivating lands. If such a problem is encountered, probably the best advice is to consult
the village council and the local authority who may succeed in persuading them to accept
what is best fir the whole community.

Stability of river bank and dimensions of the structure

This factor affects the cost of the structure directly and its operation performance
indirectly. River banks are usually unstable in shallow reaches where its cross-section is
wide. This implies that a larger and costlier structure is needed in this site than when it is
built on a narrower and a more stable section. In a shallow, wide section of a river, where
the flow is sluggish, the velocity is less than that in a smaller cross-section, and hence
there is a higher rate of silt accumulation in the shallow section. This latter fact, of course,
affects the performance of structure and also increases its maintenance cost. The design
engineer, if possible, chooses a location for the structure where the river is straight, has
stable banks and no deposit islands are found in the river. When a straight reach cannot
be found, the weir should be built on the outside bank of a bend where the river, even at
its minimum discharge, can supply the main canal with adequate water and where the
sedimentation is less.

2. Type of Structure
By the structure type is meant, type of the construction materials used, and its shape. In
developed countries, probably the major decisive factor in selecting the construction
materials and shape of the structure is the economic factor. However, this is not
necessarily the case in the developing countries, especially when the beneficiaries
provide labour. On the first site visit, the design engineer usually investigates the
following.
1. What construction materials are available in the locality and their process?
2. What other materials are used for construction, which are not available in
the local market and what would be their costs, if imported?
3. Is there a shortage of any construction material in the market? If yes, what
are they and what is the average waiting time to obtain n order?
4. Is it possible to hire construction machinery in the locality? Where is the
nearest machinery rental station and what are the rates?
5. What is the availability of skilled labour in the area? For the kind of job in
mind can skilled labourers be recruited in the area and what are the
average wages?
6. What will be the construction period? Diversion structures are usually
constructed in dry seasons, when the river is dry of at its lowest level. In
developing countries, where construction technology is not usually
sophisticated, the construction period should be limited, as much as
 4

possible, to the dry season. This factor in many cases surpasses all other
factors in selecting the structure type.

The answers to the above questions usually give the design engineer sufficient
information to decide what materials should be used and what construction technology
followed. However, in constructing large structures, the result of geotechnical
investigation may be a very important factor in selecting the shape and construction
materials of the structure, although in the case of small diversion weirs, this is very
unlikely.

Another important factor in choosing the structure type is the implementation method. If
the project is implemented by farmers themselves one naturally opts materials and low
technology. On the other hand, if it is constructed by constrictors, the economy of the
structure may be only decisive factor.

3. Topography survey

It is necessary to view the cross sections and profile of the river at these sections. Here a
permanent bench marks should be fixed near the site not less than 50 m away from the
structure’s location preferably made of concrete with a steel peg in the middle. The
irrigated area should be surveyed and a contour map prepared. A contour map of the
project site covering location of the structure to a scale of 1:5000 or 1:10000 is ideal
At the proposed location of the structure the survey required is as follows:

• Layout of the river. About one kilometer up and downstream of the structure’s
location is sufficient.
• The topographic survey should extend 250 m to both sides of the river.
• Cross section of the river at each 50 m in straight reaches of the river and at 25 m
in bends.
• A contour map of an interval not larger than 1 m for the land, 0.50 m for the river
bed and of 0.25 m near the structure.
• Layout of the river stretch surveyed should be drawn to a scale not larger than
1:2000. The weir location for a distance 50 m up and downstream of the structure
to be drawn to a scale not less than 1:200.
• The horizontal and vertical scale of the river cross section must be the same and
must not exceed 1:50 or 1:100 depending on the size of the river.

4. Soil Investigation

In the first site visit and when a few locations are proposed for the structure, the engineer
should also visually test the soil and describe its physical properties. Shallow pits should
be dug for initial description of the soil profile and to ascertain the groundwater level at
the weir site. The purpose of this preliminary investigation is to recommend the type of
tools to be used and tests required to be carried out. It should also be ascertained what
facilities are available in the nearest soil laboratories. These observations will help the
 5

team in charge of the soil tests in selecting the right tools and equipment when they visit
the area.
Soil specialists carry out the investigation in response to the designer’s. The
designer should realize that soil investigati0n is an expensive operation requiring
transportation, equipment all the existing data relevant to the design. The data
should be studied carefully to decide what can be achieved without soil tests, by
making realistic assumptions and what cannot be done without it.

It happens very often in developing countries, the moment a decision to construct

a structure is made, soil specialists in the decision to construct a structure is made,
soil specialists in the government and academic institutions show a great interest
in the investigations and rush to carry them out without consultation with the
designer. The result is usually a nice bound report with an expensive bill which
may not be of any use in the design. Soil investigation should be carried out for a
purpose rather than as routine work of the project implementation. The designer
first of all should list what data he needs for the design. In the case of designing
hydraulic structures, as will become clear in the subsequent chapters, the data
given in Table 3.1 will be needed for the purposes indicated.

The question that needs to be asked at this stage is: do we really have to carry out
geotechnical investigations to obtain the design data given in Table 1? If the
proposed construction operation involves only a relatively small investment, as is
the case in constructing small diversion weirs, the designer cannot afford to
include more than a small number of exploratory holes and a few soil
classifications in the classifications in the investigation. The lack of accurate
information concerning the subsoil conditions can be compensated by the use of a
liberal factor of safety in the design.

In the following, the design data given in Table 1 is briefly examined to conclude
what types of soil tests are required to be carried out on the site and in the
laboratory and in the laboratory and what can be avoided Shallow pits should be dug
for initial description of the soil profile and to ascertain the groundwater level at the site.

Table 1 Required soil data for the design of hydraulic structures

Data Purpose
1 Soil profile under to detect impermeable and soft layers under the foundation
the foundation
2 Soil classification to estimate engineering properties of the soil
estimate safe exit hydraulic gradient at the d/s toe of the foundation
design filter under riprap
3 Permeability to estimate seepage rate under the foundation
4 Unit weight, angle to calculate lateral pressure on retaining walls,
of internal friction, estimate the soil bearing capacity
and cohesion of the
soil.
 6

5 Compressibility to estimate settlement of the structure

and coefficient of
consolidation of
soil layers

Soil profile under the foundation

Soil profile under the foundation of structures is required to reveal the depth of
impermeable and soft layers. In designing the foundation of heavy structures, e.g.
a large weir with a bridge for heavy vehicles on the top, investigating the soil
profile is necessary to assess settlement of the construction and the allowable
pressure on the soil layers. For such a case, soil and geotechnical specialists
should be consulted. The depth to which the investigation to be carried out
depends on the load and dimensions of the foundation. In small diversion weirs, it
is hardly necessary to investigate the soil below 3 m, even if it carries a small foot-
bridge. The designer would like to discover a hard rock base near the river bed to
make the cost of the investigation worthwhile by eliminating construction of an
expensive stilling basin for the structure. When construction of small weirs are
concerned, the investigation usually does not require more than a set of hand
operated augers. Sophisticated drilling equipment is expensive and in remote,
underdeveloped areas is not usually available in the locality.

Soil classification
Soil classification and grain size distribution of the foundation soil are needed of
safe hydraulic gradient at the downstream edge of the structure’s foundation; (ii)
design of filter for under riprap wherever it is used; (iii) estimating roughly the
engineering property of the soil such as coefficient of permeability.
To decide whether site or laboratory tests are required to obtain the design data,
one should initially assume some likely undesirable properties of the soil and
roughly design the structure for these properties. Later a second design should be
carried out for a set of assumed better soil properties. If the cost difference
between the two designs in substantial and greatly exceeds the cost of the site and
laboratory tests, the tests should definitely be carried out.

Permeability of the foundation soil

Permeability is of significant value in the determination of seepage and uplift
pressure under the foundation of hydraulic structures. It is needed when a
numerical model or the flow net method is used for the flow analysis under the
foundation. However, where the design of small weirs is concerned, usually
simplified methods are used which assume homogeneous soils and do not require
the use of the soil permeability to determine the uplift pressure. The amount of
 7

water flow from upstream to downstream of a weir is not the designer’s concern.
The permeability is related to the type of soil and its grain size distribution

Unit weight, angle of internal friction and cohesion of soil

Here again, it is the cost of the work which decides whether the field and
laboratory tests should be carried out for determination of the soil properties or
whether to rely on some realistic assumptions. It is known that the active soil
pressure of a soil which has both cohesion and friction is smaller than that of a
purely frictional soil. Therefore when a retaining wall is designed, if it is assumed
that the backfill soil of the structures’ wing walls is cohesion less, the design will
be safer than that which results from a soil which has both cohesion and the
friction. The assumption made may lead to an increase in the dimension of the
wall and hence the cost of its construction. In the case of designing a small
structure where the height of the walls rarely exceed sis meters, the increased cost
will be much less than the cost of the soil tests.

Small hydraulic structures usually do not collapse because the pressure of the
foundation is more than the bearing capacity of the soil. They collapse due to the
scour of the foundation or the differential settlement. Small weirs built on rigid
foundations result in a very small contact pressure: it hardly needs to be checked
against the soil bearing capacity. However, the contact pressure should be

5. Hydrological data

Hydrological data is required at the project area to obtain, after their analysis, the design
discharges. The design discharges, maximum and minimum values, are important figures
which are used by the designer to size the structure. To design the canal network, the
mean discharge of the river is also needed.
The maximum design discharge is the peak river discharge that corresponds to a certain
return period which is usually decided by the designer on economic and engineering
ground. The maximum design discharge is used in the design, to determine the backwater
curve results from constructing the weir in order to predict the highest water level that
occurs, on average once every (T) years, where T is the selected return period of the
discharge.

Mean River Discharge in the Design

To design the intake of the irrigation scheme’s main canal, where it takes off from a
diversion weir, the discharge of the main canal will be needed. In a proposed irrigation
area, the discharge of the feeder canal depends on, (i) types of crops and (ii) area of the
project. However, area of the project is directly linked to the availability of water in the
river and the water rights downstream of the weir. The procedure for determination of
the offtake canal capacity and hence the developed area is as follows:
 8

1. The river mean discharge which has an acceptable probability of occurrence, in

short periods say 10 or 15 days, should be determined. In irrigation projects,
usually 80% flow reliability is taken, i.e. a flow which would be equaled or
exceeded in four out of five years.
2. The water rights for the farmers d/s of the weir should be established. The
maximum available water to the project in each period in the irrigation season is
the calculated mean river discharge minus the d/s water rights.
3. Determine the water requirements per hectare of the project area for the crop
pattern advised by the project agronomist. This value should include all the losses
in the field and the canal network.
4. Values of the available water calculated for each period in Step 2 divided by the
water requirements established in step 3 gives the maximum area which can be
irrigated in a particular period. The smallest area in any period of the season is
the maximum area which can be developed for irrigation relying on the river
discharge.

Minimum River Discharge

The function of the diversion weir is to raise the water surface in the river, at its u/s, to a
level which can command the proposed irrigated area. To command the irrigated area, the
weir crest should be set at a level so that the water afflux at the minimum river discharge
secures the project water demand.
Therefore the minimum discharge directly affects the weir crest level and indirectly the
length of the apron, since in most cases the critical condition for determination of the
percolation length and the exit gradient is when there is no overflow and the water is
ponded behind.
The minimum river discharge should also be determined from the river flow record and
similar to the Qmax related to the return period or probability of its occurrence, e.g. a
drought which occurs once every ten years on average.

The Design Discharge

While designing a weir, provision must be for the flood that is likely to occur during the
lifetime of the structure. For example, length of the apron must be adequate to
accommodate the hydraulic jump and the weir walls and dykes along the river must be
high enough to avoid being overtopped by the floods. However, one can neither choose a
very high nor a very low flood magnitude for the design. Taking a very high flood results
in a costly structure. On the other hand if a low flood magnitude is chosen, it will result in
failure.
It is desired to select a design flow which is not likely to occur during the lifetime of the
structure, it is necessary to use a return period greater than its estimated economic life.
The return periods required for specified risk of occurrence within the life time of the
project can be estimated from

J = 1-(1-P)n
 9

where P = probability of non occurrence of an event in any year, J = risk or probability of

the event occurring in any year period.
For example if a project life (n) is estimated at 50 years, and the designer is prepared to
take a chance (J) of 10%, the probability of the design of flow (P) not occurring within
the project life time calculated from the above equation is 0.21%. This means that the
discharge which he would use in the design corresponds to (1/0.21%) or 475 year return
period.
In designing small weirs, the project life is usually assumed to be between 25 to 50 years.
Determination of the design flood by analyzing the river discharge records (frequency
analysis of flood records) is available in hydrology texts.
 10

II - Hydraulic Analysis of Surface Flow

Surface flow analysis means the determination of the flow condition upstream and
downstream of the weir at different flow rates and to size different parts of the structure
accordingly, so that the structure serves the purpose for which it is built. To design a weir,
all external forces acting on it must be calculated. These forces which include the soil and
uplift pressures can be evaluated whenever the surface water flow profile at the weir
location is drawn. It is the duty of the designer to envisage all the possible water profiles
and find out the critical situation for which the structure should be designed. It will be
shown later in this chapter that for each part of the structure there is one design discharge.
In the subsequent sections different aspects of the weir design and analysis of different
possible situations are discussed.

General Design Considerations of the Weir

The behavior of the surface flow is greatly affected by the geometry of the weir and the
geometry of the weir directly affects the design and economy of the structure. Some
important features of the weir and their effects on the design are given in the following
sections.

1. Crest Elevation

The crest elevation of the weir affects the water profile in two ways:
1. The height of the crest affects the discharge coefficient and consequently the
waterhead above the weir and the back water curve.
2. The height of the weir affects the shape and location of the jump and the design of
the basin.
Height of the weir is usually decided with the requirement of the canal intake in mind. To
fulfill the objective of the intake structure the following points should be considered in
deciding the crest elevation:
- The crest level should be so set that the waterhead required to deliver the main
canal’s discharge design is obtained
- If the entire flow of the river at low flows is diverted, the crest elevation must be
set at a level so that the ponded water gives the required head to apply the canal
with the design flow.
- If the minimum flow in the river exceeds the discharge of the offtake canals, the
crest level of the water can be set lower than the river water level which is
required to deliver the design flow in the canal, to allow for the downstream water
rights.
- The maximum (allowable) upstream water surface elevation must also be
considered in selecting the crest elevation. The maximum allowable water level
depends on the upstream river bank elevation and the infrastructure such as
bridges, roads, buildings, etc. The minimum freeboard at the maximum design
 11

discharge must not be less than 0.50 m, although this cannot be achieved in many
seasonal rivers which overflow annually.

2. Length of the Weir

Length of the weir depends on the physical features of the given site. The effect of the
weir length on the upstream waterhead and sedimentation behind the weir must be
understood.

- A weir with a long crest gives a small discharge per unit length and hence the
required energy dissipation per metre of the crest is smaller than what is needed
for a shorter crest length.
- Constructing a weir longer than the river width causes formation of islands at the
upstream side of the weir. As a result the canals’ inlets can become cutoff from
the river flow. The formation of the islands upstream of the weir reduces the
effective length of the crest.

As a general rule the crest length should be taken as the average wetted width during the
flood. In taking the average, the upstream and downstream of the proposed location
should be examined and the width at a stable location measured.

3. Shape of the Weir

In deciding the shape of the weir two important factors need to be considered: the
practicality and the economy of the structure. When a weir is designed, the designer
should consider the skill of the people who are expected to implement the structure.
There are weirs of different types and shapes and most of them are constructed to serve
the same purpose. The designer should not impose a weir with a shape which cannot be
constructed easily by the local builders. It is unwise, for example, to impose an Ogee-
shaped weir on farmers in a remote village, just because the discharge coefficient is
slightly better than other alternatives. It is not suggested here that the designer should
sacrifice the efficiency and economy of the structure, altogether, for the sake of
simplicity. However, compromise must not be made on the stability of the structure.

Two types of weirs are very common in small irrigation schemes. These are:
1. concrete weir with vertical upstream and slopes downstream faces,
2. stepped weir, which is usually constructed from gabion boxes.
A commonly constructed, but slightly complicated, weir is the one developed by the US
Army of Engineers known as the WES weirs (developed at its Waterways Experiment
Station).

Discharge over Weirs

Discharge over weirs is generally expressed as follows (Fig. 5.1):

 12

Q = CLH e
3/ 2

where Q = discharge
L = length of the weir
He = height of energy line above the crest
= V2/2g + Hd

C = discharge coefficient

Fig. 5.1 Hydraulics of flow over weirs

When piers are placed on the crest to support a slab for the river crossing or to operate the
controlled gates:

Q = C (Lo − KNH e )H e
3/ 2
(5.2)

Lo = Clear length of the weir

N = Number of pier contraction = 2 x number of piers
K = Pier contraction coefficient. In practice K is assumed equal to 0.05.

Discharge coefficient is related to He/Hd and it varies according to the type of weirs.
Determination of the coefficient C for two common weirs is given in the following
section.

1. WES – standard weir

The shape of the Waterways Experiment Stations (WES) weirs, which are presented in
Fig. 5.3 is expressed by:

X n = K o H dn −1Y (5.3)

Where X and Y are coordinates of the crest profile with the origin at the highest point of
the crest. Hd is the design head excluding the velocity head of the approach flow and Ko
and n are parameters depending on the slope of the upstream face. The values of Ko and n
are given in Table 5.1.
 13

Fig. 5.3 The WES-standard spillway shapes

Table 5.1 Parameters for the WES weir shape formula (Eq. 5.3)

Slope of U/S face Ko N

Vertical 2.000 1.850
3 on 1 1.936 1.836
3 on 2 1.939 1.810
3 on 3 1.873 1.776

Determination of the coefficient C

There are two conditions for which the value of C is determined:

(i) h/Hd > 1.33, h = height of the weir,

Hd = design head excluding the approaching velocity head.
 14

For this case the velocity head is negligible and C = 2.225.

(ii) h/Hd < 1.33
The velocity head should not be neglected. The value of C calculated from the
experimental curve is presented in Fig. 5.4.
The calculation procedures:
1. determine He/Hd and h/Hd
2. from a relevant curve determine C/Cd
3. the discharge coefficient for a vertical upstream face is:

C = C/Cd (from the graph) x 2.225

4. if the upstream face is not vertical, correct the value of C calculated in step 3,
multiplying it by the correction factor for the given h/Hd value.

Fig. 5.4 Head-discharge relation for WES – standard spillway shapes.

 15

Fig. 5.5 Geometry of the WES weir

2. Horizontal Broad Crested Weir

The weir comprises a horizontal crest between vertical abutments and two vertical or
inclined faces. The upstream corner should be rounded in such a manner that flow
separation does not occur. Flow separation can also be avoided by using an upstream ram
which slopes between 2/1 and 3/1 (H/V). The downstream of the horizontal crest may be
a vertical or downward slope face. The dimensions of the weir and it abutments should
comply with the requirement in Fig. 5.6 if the flow over the weir is required to be
measured accurately.
 16

Fig. 5.6 A typical broad crested weir.

Evaluation of the Discharge Coefficient

The hydraulics of the broad crested weir is complicated by the fact that the value of the
discharge coefficient is a function of many parameters. These parameters are the
upstream and downstream height of the energy line, width of crest (b), and height of the
crest above the river bed. It can be shown theoretically that the discharge over a broad
weir is expressed by the following equation:

Q(theoretical) = 1.706LHe3/2

Experiments have shown that for a well rounded upstream edge the discharge is:

Q = 1.673LHe3/2

Bos (1989) concluded that for He/b > 1.5, the weir acts as a suppressed sharp weir and the
discharge formula would be:

Q = 1.84LHe3/2

If, for example the crest width is 0.50 m and the head of water over the crest is 0.75 m,
this last equation applies.

In practice a discharge coefficient C = 1.70 is usually used. Here also the last figure is
recommended. When a weir is designed, engineers deal with the probability and
discharge records that are liable to have many errors. Changing the coefficient C from 1.7
to 1.75, for example, affects the discharge by only 2.9%. While errors in the hydrological
data and the method of estimating the design discharge are undoubtedly much greater.
The equation applied here is:
 17

Q = 1.70LHe3/2

When the crest supports piers, the weir length must be modified according to Eq. (5.2).

Water Profile at the Weir Site

The engineer who designs a weir is interested in the water profile upstream and
downstream of the weir for the following reasons.

Water surface at downstream of the weir:

1. Carry out the stability analysis of the weir.
2. Design the weir structurally.
3. Design the downstream wing wall, and protection works downstream of the apron.

Water surface at upstream of the weir:

The level of the back water curve is required to be determined for the following reasons:
1. to determine the safe level of the embankment and superstructures,
2. to find out whether the canal intakes for which the weir is built, receive water at
all stages of the river and what type of headwork is needed.

In the following, determination of the water profile upstream and downstream of the weir
is discussed.

1. Water Profile Upstream of the Weir

In the previous sections, it was explained how the level of the weir crest is fixed to give
the required water level upstream of the weir. The required water level may be a level
which guarantees a certain discharge for the offtake canal when the river is at its lowest
discharge. Once the crest level is decided, it is required to draw the water profile for few
possible discharges. Usually the analysis is carried out for Qmax which corresponds to 50
and 100 years return period, 2/3 and 1/3 of Qmax and also for a flow which corresponds to
a certain drought level, for example the minimum discharge which has a return period of
ten years.
Construction of weirs elevates the total energy line upstream of the structure. The
difference between the upstream and downstream energy grade line become very high.
Therefore, the energy must be dissipated before it reaches the natural river course,
otherwise it causes damage to the banks and downstream of the apron. The flow over the
weir is in a supercritical state. Therefore, the energy tends to dissipate through a
hydraulic jump downstream of the weir. To control location of the jump, the apron and
stilling basin are designed to suit a range of river discharges.

Hydraulics of the jump

Hydraulic jump on horizontal surface

 18

On the horizontal apron downstream of a weir the following equations are valid (Fig. 5.7):

1 + 8F12 − 1
D2 = D1 (5.11)
2

D1 ⎛⎜ 2q 2 1 2 1 ⎞⎟
D2 = + D1 −
2 ⎜⎝ gD1 4 2 ⎟⎠

HL =
( D2 − D1 )
3

4 D1 D2

q2
dc = 3
g

V1
where F1 = (Froude number) =
gD1
V1 = Velocity at the jump = q/D1
q = discharge per meter width of the channel
HL = head loss or dissipated energy as a result of the jump
dc = critical depth

Fig. 5.7 A typical walls profile over weirs

It should be noted that the friction loss on the weir is ignored, since the distance is short
and friction loss compared to the jump loss is very small. It is also assumed that the jump
loss occurs suddenly at the location of the jump.
To determine the water depths the well known Bernoulli equation is used. The equation,
if applied to sections 0 and 1, can be written as follows:
 19

Vo2 V12
Z0 + h + H d + = Z 1 + D1 +
2g 2g

The above equation is applied just to the left of the jump. For a point just on the right of
the jump the equation is written as:

Vo2 V2
Z0 + h + H d + = Z 1 + D1 + 1 + H L (5.17)
2g 2g

It should be known that the terms on both sides of the above equations represent the
height of the total energy line. In general form the equation is written as follows, when it
is applied between two points 1 and 2:

V12 V2
+ D1 + Z 1 = 2 + D2 + Z 2 + H L
2g 2g

Eq. (5.17) is valid for horizontal surfaces only. If it is applied on sloping surfaces, by
assuming that the stream lines are still parallel, the following modified equation can be
used:

V12 V2
+ D1 cos θ + Z 1 = 2 + D2 cos θ + Z 2 + H L (5.19)
2g 2g

where θ is the angle of the surface with horizontal at the location considered, and HL is
the total head loss between the two points. To demonstrate application of the Bernoulli
and the jump equations, assume for the time being that the jump is known definitely to
occur downstream of the weir.

Location of the Jump

From the previous discussions it is possible to determine the sequent (conjugate) depth of
the hydraulic jump, but it did not explain its position downstream of the weir. There are
three alternative patterns that allow a hydraulic jump to form downstream of a weir: these
alternatives are presented in Fig. 5.8.

Case I: It represents a case in which the tail water D3 is equal to the sequent depth D2. In
this case the jump occurs on the apron. For scouring protection purposes, this is an ideal
case. However, in practice this condition is hardly fulfilled since a little difference
between the actual and assumed values of the hydraulic properties of the structure and the
river will cause the jump to move from its calculated position. Consequently, some
device to control the position of the jump is always necessary.

Case II D3, the tail water, is less than the sequent depth D2. In this case the jump will
recede downstream to a point where the flow conditions allow the jump to occur. This
 20

case must be avoided in the design, because the jump moves away from the scour-
resisting apron and will take place either on the loss rubble bed or in an entirely
unprotected channel, resulting in severe erosion.
The remedy for the design is to use a certain control energy dissipating structure in the
apron to ensure that the jump occurs within the protected apron.

Case III D3 is greater than D2. Therefore, the jump will be forced upstream and may
finally be drowned out at the source, becoming a submerged jump. In this case, the
hydraulics of the jump on sloping surfaces would apply. The occurrence of the jump on a
sloping surface is the most stable one of all.

The above discussion reveals that, when a weir is designed, several possibilities have to
be considered regarding the overflowing discharge. The effect of different flow rates on
the tail water and the sequent depth is graphically presented in Fig. 5.9. It is shown that
there are four different possibilities to consider in the design of a weir. Therefore, the
analysis should be considered for various discharge rates.
21
 22

The above discussion on the hydraulic jump is summarized as follows:

1. Determine the jump depth D1.
Height of energy line above the horizontal front floor = D1 + V12 2 g
2. Apply Eq. (5.11) to determine D2.
3. If D2 = D3 case 1 jump occurs at the toe of the weir.
4. If D2 > D3, the jump will recede downstream and its location depends on the slope
and condition of flow in the river.
5. If D2 < D3, the jump occurs on the weir face or may be drowned and become
submerged. The jump on a slopped surface is discussed in the following section.

Jump on sloping surfaces

Hydraulic jump occurs on the sloping face of weirs or sloping aprons when the tail water
exceeds the conjugate depth D2. Studies on hydraulic jump on a sloping surface are
limited, and there is no pure mathematical approach. However, with the help of
experimental data, an approximate solution for the problem has been found. Types of
jump which are likely to occur on the sloping surfaces are shown in Fig. 5.10, C-F. The
flowchart in Fig. 5.10 is designed to identify the type of jump in the problem and
accordingly to choose a suitable solution. Case A and B in Fig. 5.10, which are jumps on
horizontal surfaces, are already covered. This section deals with the case C to F.

Hydraulics of the jump

D2 S =
D1S
2 cos θ
( 1 + 8G − 1)
2
(5.20)

where D2s = conjugate depth when the jump occurs on a sloping surface.

D1S = Pre-jump water depth

θ = angle of the surface
G2 = F12T 2
T1 = (10)0.027θ
F1 = Froude number before the jump

Eq. (5.20) is applied to type D and E. An analytical solution for type C is not yet
developed, but graphical solutions based on laboratory experiments is available. Figs.
5.11 and 5.12 present the graphical solution of case C. It is suggested that for practical
purposes Fig. 5.12 can be used for cases C, D, and E. For case F, the following equation
is suggested:

D2 S 1
=
D1S 2
(
1 + 8G 2 − 1 )
where
 23

F
G=−
KL sin θ
cos θ −
d 2 − d1

d1 = D1Scosθ, d2 = D2Scosθ
K = a correction factor which results from the assumption that the jump profile is linear.
For simplicity the value of K is assumed to be 1, without serious implications.
L = Length of the jump (Fig. 5.12)
Fig. 5.13 is the graphical solution of Eq. 5.23.
24
 25

Prejump water profile

The water profile before a hydraulic jump is the supercritical flow surface. It should be
understood from the previous sections that the jump either occurs on the inclined surface
(front face of the weir or a sloping apron) or on the horizontal surface (the front apron or
downstream channel). The two cases are explained as follows.

(i) The prejump profile on sloping surfaces

The profile on gentle slopes: The slope of a channel can be considered a gentle slope
when the streamlines, practically remain parallel, and the pressure distribution can be
assumed to be hydrostatic. Flow on long sloping aprons, in most cases, has this type of
profile. To determine the water surface, the Bernoulli equation is applied and within a
short reach the friction loss can be ignored. Fig. 5.14 is a case where D2S, the post-jump
depth, is larger than the tail water: therefore, the jump has moved downstream. Knowing
the height of the total energy line, the depth can be determined from:

V x2
H= + Dx + Z x + h f
2g
q
Vx =
Dx

The friction loss hf can be ignored and Z is known. Therefore, the depth can be
determined by the usual trail and error.
26
 27

The water profile on steep slopes: When the stream lines are no more parallel and the
pressure distribution cannot be considered hydrostatic, the energy line and the water
profile can not be determined theoretically. The case of flow over the sloped face of a
weir is of this type of the profile. When the jump draw high up the weir surface, the pre-
jump profile can not be determined, Fig. 5.15. If the weir is high and jump occurs on the
sloping surface, but not far up the toe, the stream lines become parallel again on most
parts of the weir face (Fig. 5.15), and hence the Bernoulli equation can be applied to
determine the water depth. However, as far as the flow over the weir crest is concerned, it
has to be determined experimentally, since the flow is curvilear and depends on the shape
of the crest. The water profile over the WES standard weirs has been investigated and the
result is shown in Fig. 5.16.

(ii) The prejump profile on horizontal surface

If the jump advances downstream of the apron, as it may happen when no energy
dissipater is provided, the water profile between the toe of the weir and the location of the
jump forms what is known as H curve. Computation of this curve can also be done by the
standard step method, when the water depth at the weir toe or at the location of the jump,
is known. The standard step method is explained in the following section.

The post-jump water profile

The water profile after the hydraulic jump, can be calculated with the help of Fig. 5.17,
which is based on experimental work and it is self explanatory. Length of the jump can
be checked from Fig. 5.12, which should closely agree with the value of X in Fig. 5.17,
beyond which, the water depth remains constant.
28
29
 30

Water Profile Upstream of the Weir

Water profile upstream of the weir needs to be known for two purposes:

(i) to determine the height of the river banks upstream of the weir and,
(ii) to find out whether the water surface is high enough to deliver the
required discharge to the offtake canals.

There are several methods to determine the water profile upstream of weirs, but here only
two methods are explained.

The profile by the approximate method

This method gives results accurate enough for the preliminary design in channels of
uniform cross-section and constant hydraulic properties. It assumes that the following
equation fits the profile.

( XS − 2∆ o )2
Y=
4∆ 0

Y = Water rise, at distance X upstream of the weir, above the normal water depth.
X = Distance from the crest to the point where Y is required to be determined.
S = Slope of the river.
∆o = Rise of water above the normal depth at the weir site (X = 0).

This method does not take into account the geometry and the hydraulic properties of the
channel.

Standard step method

This method is a trail and error procedure by which the water profile upstream or
downstream of a structure can be determined. The trial can start at any location, but it is
advisable to carry the computation upstream if the flow is subcritical and downstream if it
is supercritical. The computation stepwise are as follows:

1. Field works
The river cross section must be surveyed, whenever its geometry and hydraulic properties
significantly change. The sections and river profile must be drawn to appropriate scales.
The distance upstream of the weir which should be covered by the survey, depends on: (i)
the maximum afflux over the weir, and (ii) how significant the installations and
infrastructures on both sides of the river banks are. For example, if it is required only to
find out whether the back water curve hits a bridge at some distance upstream of the weir,
then only the distance between the weir and the bridge needs to be surveyed. It is helpful
to realize that in a channel of uniform cross-sections and constant hydraulic properties,
the effect of the back water curve disappears at a distance X from the weir, where:
 31

2∆ o
X =
S

S = gradient of the channel

∆o = afflux over the normal water depth at the weir site.

If for example, ∆o = 1.50 m and S = 0.001, the effect of constructing a weir on the water
profile diminishes at 3000 m away from the structure.

2. Foe each cross section surveyed, determine its area (A) wetted perimeter (P) and the
hydraulic radius (R) at different water depths. Plot the water depth D against A and D
against R. (Table 5.2).

3. Prepare a table with 21 columns as it is shown in Table 5.3. Number of the rows
initially is not known, it depends on the number of river cross sections and initial value of
the water surface assumed in the trial.

The heading of the columns is self explanatory, with the possible exception of column 13
which contains the energy coefficient:

∑ (α K )
i=N
3
i i / ai2
α= i =1

⎛ i= N
⎞
⎜ ∑ K i ⎟ / A2
⎝ i =1 ⎠

K = hydraulic conveyance of the section K = AR2/3/n which is calculated in column 12.

Wherever the river cross section is a composite one, i.e. consists of main and minor
channels such as in river plains, the section is divided into subsections separately. The
index (i) refers to the subsections number.

a = area of subsection (i)

A = total cross section area of the river
αi = the subsection’s energy coefficient.
32
 33

In the computation, αi is not known: it is assumed to be equal to one. Therefore the above
equation becomes

∑ (α K )
i=N
3
i i / ai2
α= i =1
3
⎛ i= N ⎞
⎜ ∑ K i ⎟ / A2
⎝ i =1 ⎠

Where K = hydraulic conveyance of the section K = AR2/3/n which is calculated in

column 12. Wherever the river cross section is a composite one, i.e. consists of main and
minor channels such as in river plains, the section is divided into subsections and the
conveyance is calculated for each subsection separately. The index (i) refers to the
subsection’s number.
a = area of subsection (i)
A = total cross section area of the river
αi = the subsection’s energy coefficient.
 34

In the computation, αI is not known: it is assumed to be equal to one. Therefore the above
equation becomes

∑ (K )
i=N
3
i / ai2
α= i =1
3
⎛ i=N ⎞
⎜ ∑ K i ⎟ / A2
⎝ i =1 ⎠

4. Start of the computation: The computation starts by assuming a water level at the first
section if the actual value is not known. Insert this Figure in column 5 of Table 5.3. From
the given water and bed levels, the water depth is calculated. The figure in columns 7 and
8 are obtained from the graphical relationships of D, A, and R, explained in step 2 above,.

6. The figures in columns (14-21) are then calculated systematically by applying the
formulae given in the column headings. The average slope of the energy line S is
the arithmetic mean of the value calculated for the present station (X) and its
previous one (X-∆X). The total head in column 21 is equal to the total head at the
previous station ((X-∆X) and the friction loss hf at the current station X. If the
value of H calculated in column 21 does not agree with the H value in column 16,
a new water level is assumed and entered in column 5. The procedure must be
repeated until the two figures in column 16 and 21 agree closely.

Example
A small weir is designed for a river discharge of 25 m3/s. The water and river bed levels
at the weir site are 103.00 and 100.00 m respectively. Four river sections at 500 m
intervals are surveyed. Their cross sections and hydraulic properties are presented in Figs.
5.19 and 5.20. Determine the water level at these sections when a flood of a magnitude
equal to the design discharge passes over the weir and show whether the water within the
first 2000 m overtops the banks or not.

Solution
Presented in Table 5.3 and the determined profile in Fig. 5.21. It can be seen that the
expected water level at section 4 is 103.12 m, which is well below the river banks.
35
36
 37

Determination of the tail water depth

Throughout this chapter the tail water depth and its effects on the jump location and the
profile are mentioned several times. How the depth is calculated is now explained.
To determine the tail water depth, the rating curve at the location of the water needs to be
known. The curve is usually constructed from the river stage and discharge measurement
records. Usually, this data is not available at the location of the weir; therefore, a
theoretical rating curve should be constructed. If the stream flow record is not available at
the weir site, the following steps can be taken to draw a rating curve accurate enough for
the design purpose.
1. Surveys should be carried out at the location of the weir to draw the cross section
of the river.
2. The cross section needs to be drawn to a suitable scale. Assume different water
depths in the river, and determine the water area and wetted perimeter
corresponding to each depth.
3. For each depth, determine the flow velocity and discharge by using the Manning
equation (Fig. 5.22).
4. Plot the assumed depths and their corresponding discharges to present the rating
curve.

The rating curve is used to determine the water depth for known values of the river
discharge. Fig. 5.22 is an example on constructing the rating curve.
38
 39

Canal Head Regulator

The purpose of the weir is to create sufficient head to supply the main irrigation canal
with the design discharge. The canal head regulator is usually gated to control the amount
of flow into the canal. The most common types of head regulators are:
1. open intake and
2. closed or culvert intake.
Each of these cases is explained separately.

Open intake

This consists of the intake mouth, which acts as an orifice or weir, a transition and the
main canal. Fig. 5.25 presents a typical open intake. The intake is usually designed as an
orifice. Discharge through the intake is calculated by using the orifice formula as follows

Q = CA 2 gh

where A = area of the gate opening

h = difference between the upstream and downstream water level
C = discharge coefficient
The outflow of the gate may be either free or submerged, depending on the tail water
depth in the main canal (Fig. 5.23). The coefficient C is therefore variable and depends
on the flow condition. The following two equations may be used for evaluation of C

C = 0.615 + 0.007 x 2 5− h

for h < 5 ft and

C = 0.7201 + 0.0074W

where h < 5 ft and w < 10 ft, where w is the width of the gate. For h < 1 ft both give
nearly the same answer. In practice C = 0.6 and 0.7 are used for rectangular and circular
orifices respectively.
The hydraulics of the downstream floor of the intake should be treated in the same way as
for the main weir downstream. The floor must be long enough to accommodate the jump
and its weight balance the uplift pressure.

Design of Culverts

Hydraulic design of a culvert is primarily influenced by head water depth. The factors
that affect the discharge from the culvert are headwater, pipe size, tail water, roughness
of the pipe, pipe slope, pipe length and inlet geometry.
The flow in culverts is classified as being under either inlet or outlet control, that is the
discharge is controlled by either the outlet or inlet characteristics. The major factors
affecting the inlet control flow are: cross sectional area of the culvert, inlet geometry and
 40

headwater depth. In the outlet control flow, in addition to the mentioned factors, the
following ones are also significant: tail water depth, slope roughness and length of the
culvert. The characteristics of flow in culverts are very complicated since the flow is
controlled by many factors. In designing a culvert for the intake, it is convenient to
assume that the pipe is fully flowing with both ends submerged and to include all the
head losses in the orifice coefficient C ( Q = CA 2 gh ). To obtain the full flow condition,
the pipe inlet must be submerged to a depth not less than the sum of the velocity head and
the head losses in the pipe. Experiments on a large number of pipes of various diameter
and length have resulted the following equations for evaluation of the orifice constant C

Concrete pipe beveled-lip entrance,

−0.50
⎡ 0.026 L ⎤
= ⎢1.1 +
⎣ (D )1.2 ⎥⎦
Concrete pipe, square-cornered entrance,

−0.50
⎡ 0.026 L ⎤
C = ⎢1 + 0.31D 0.50 +
⎣ (D )1.2 ⎥⎦
for vitrified-clay pipe, bell and upstream,

−0.50
⎡ 0.022 L ⎤
C = ⎢1 + 0.023D 1.9 +
⎣ D ⎥⎦

For corrugated-metal pipe

−0.50
⎡ 0.106 L ⎤
C = ⎢1 + 0.16 D 0.60 +
⎣ (D )1.2 ⎥⎦
For concrete-box culvert with round lip entrance

−0.50
⎡ 0.0045 L ⎤
C = ⎢1.05 +
⎣ (R )1.25 ⎥⎦
For concrete-box culverts with square-cornered entrance,

−0.50
⎡ 0.0045 L ⎤
C = ⎢1 + 0.40 R 0.30 +
⎣ (R )1.25 ⎥⎦
 41

where D is diameter of culvert pipe, R is hydraulic radius for box culvert and L is length
of the culvert. All are in feet. Fig. 5.26 presents some standard culvert inlets.
42
 43

Culvert discharges may also be computed by considering the culvert as a short pipe by
applying Bernoulli’s equation to points 1 and 2 in Fig. 5.27. Taking the datum’s plane at
the downstream water surface gives the following

H = he + hf + ho

Where H = total energy head

he = entrance loss
hf = friction loss
ho = loss at the outlet

V2
he = k e
2g
V2
ho = k o
2g
n 2V 2
hf = 4/3 L Manning Eqn.
R

The following values of Ke, ko, and n can be used in calculating the head losses (Table 5.5)

Table 5.5 Coefficient of the head losses Ke, Ko, and n

Type of pipe Ke Ko Manning (n)
Concrete, flush inlet 0.10 1.00 For concrete pipe n =
Concrete projected inlet 0.15 1.00 0.011
Metal flushed inlet 0.50 1.00 For corrugated metal
Corrugated inlet 0.90 1.00 pipe n = 0.025

Design of De-silting Basin – Dimension of the basin

To avoid accumulation of sediments in the canal and costly maintenance, construction of

a de-silting basin, downstream of the intake is recommended. Length of the basin can be
calculated as follows (Fig. 5.28):

Db Lb
= (5.49)
Vs V

where Lb = length of the basin, Db = water depth in the basin, Vs = settling velocity of the
particles, V = water velocity in the basin.

(γ s − γ w )
Vs = xgxd 2
18ν

γs, γw = density of the settled particles and water respectively

 44

d = diameter of particle
ν = kinematic viscosity.
G = acceleration due to gravity
For an assumed γs = 2.65 g/cm3 and water at 20oC, Eq. (5.49) will be reduced to

Vs = 0.9d 2

where Vs in m/s and d in mm.

Suspended particles consist of silt (d = 0.06-0.002 mm) and clay (d<0.002 mm).
 45

III - Uplift Pressure under Weir Foundation

Determination of the water uplift pressure and its distribution under hydraulic structures
is a significant part of their engineering design. Without knowing the uplift pressure at
different points along the foundation, the stability analysis of the structure cannot be
carried out and the structure cannot be dimensioned properly to give the required safety
factor.
Failure of weirs on permeable foundations occurs as a result of one or more of the
following:

1. Subsurface flow actions

Piping or undermining soil under the foundation can cause collapse of the apron and
eventual overturning of the structure.
Uplift pressure. A weir can fail when the uplift pressure creates an overturning moment
in excess of the superstructure’s balancing moment. To avoid this happening, the uplift
pressure must be estimated correctly and the structure dimensioned properly.

2. Surface flow actions

This is caused by scouring of the downstream floor of the structure. It is due to
unbalanced pressure in the hydraulic jump.
Analysis of uplift pressure under structures built on impervious foundations is simplified
by the fact that the head dissipates by friction when the water percolates through cracks
and fissures in the foundation. The uplift pressure is usually assumed to vary linearly
from the upstream head to the tail water.
Weirs constructed on impervious foundations are rare since most irrigation projects
locate at or near the alluvial stage of rivers. In the subsequent sections some approaches
for assessing the uplift pressure in pervious foundations and limitations on their
application are discussed.

Methods of the Seepage Analysis

1. Bligh’s creep theory

The theory assumes that the water upstream of the weir, creeps to its downstream along
the contact base of the structure with the soil. The residual uplift pressure at any point
along the base is proportional to the distance of the point from the upstream of the
foundation.
 46

Fig. 6.1 Treatment of vertical forces in the Bligh method

One of the shortcomings of Bligh’s method is that it does not discriminate between the
horizontal and vertical creeps in estimating the exit hydraulic gradient Fig. 6.1. shows
how a foundation with vertical seepage faces is dealt with according to the Bligh method.

The uplift pressure according to the Bligh at point X is determined from the pressure
diagram as follows.

hx H
=
Leq − Lx Leq

Lx = 2a + t1 + L1 + 2b + d

Leq = (t1 + 2a ) + L1 + 2b + L2 + (2c + t2 )

H (Leq − Lx )
hx =
Leq

H
Exit gradient ie = < is
Leq

Where Leq = equivalent creep length according to Bligh theory

H = actual headwater
hx = uplift pressure along the base.

Other variables are as they are shown on Fig. 6.1. To avoid undermining the foundation
by the piping action at the toe of the foundation, the calculated exit gradient ie must be
less than the recommended safe hydraulic gradient is given in Table 6.1. Despite the
inaccuracy in estimating the uplift pressure by Bligh’s method, the method is popular
among the designers because of its simplicity and high safety factor in simple, flat-base
hydraulic structures.
 47

Table 6.1 Recommended safe hydraulic gradient

Material Safe exit gradient is
Lane’s method Bligh’s method
Very fine sand or silt 1/8.5 1/18
Fine sand 1/7.0 1/15
Medium sand 1/6.0 -
Coarse sand 1/5.0 1/12
Fine gravel 1/4.0 -
Medium gravel 1/3.5 -
Gravel and sand 1/3.0 1/19
Coarse gravel including cobbles 1/3.0 -
Boulders with some cobbles and gravel 1/2.5 -
Boulders, gravel and sand - 1/4 - 1/6
Soft clay 1/3.0 -
Medium clay 1/2.0 -
Hard clay 1/1.8 -
Very hard clay or hard pan 1/1.6 -

2. Lane’s weighed creep theory

In Lane’s method, the equivalent creep length is calculated as follows:

N
Leq = +V
3

N = Sum of all the horizontal contacts and all the sloping contacts less than 45o
V = Sum of all the vertical contacts and all sloping contacts greater than 45o

To safeguard against piping, the exit gradient must be less than the safe gradient given in
Table 6.1. To demonstrate the method, the pressure at point X, Fig. 6.1 is calculated.

To horizontal distances N = L1 + L2
Total vertical distances V = t1 + 2a + 2b + 2c + t2
1
Leq = (L1 + L2) + (t1 + 2a + 2b + 2c + t2)
3
Horizontal distances from left to (X) = L1 + d
Vertical distances from left to (X) = t1 + 2a + 2b
1
Lx = (L1 + d) + (t1 + 2a + 2b)
3

H ( Leq − Lx )
hx =
Leq
 48

3. Flow nets

This method is similar to the previous two and assumes that the soil layer under the
foundation is homogenous. The method is the graphical solution of the Laplace equation
for the steady state flow. To illustrate the method, assume a weir retains water to the
height (H) above the ground surface at it is shown in Fig. 6.2. Water will percolate from
the upstream side of the weir to its downstream through the soil along imaginary paths
known as flow lines. Due to the soil resistance, the waterhead decreases as the water
travels downstream. Lines connecting points of equal waterheads on the flow lines are
known as the equiponential lines. The flow and equipotential lines intersect at right
angles, as is demonstrated in Fig. 6.2. Be definition, if two piezometers are placed at two
different levels on the same equipotential lines, as shown in Fig. 6.2, the water level in
both of them will be the same.
The flow nets are constructed by dividing the soil profile under the foundation into
arbitrary numbers of equipotential and flow lines. In drawing the lines, the following
criteria must be considered.
(i) the flow and equipotential lines intersect at right angles;
(ii) the flow and equipotential lines are parallel among themselves;
(iii) the side of the squares which will be formed by intersecting the equipotential
and flow lines are equal.
These conditions usually dictate the number of the flow equipotential lines. The larger the
number of the lines, the smaller the size of the square and hence there is a better chance
to fulfill the required criteria. The solution is achieved by trial and error. To draw a flow
net for a weir which has three or four sheet piles, the trial may take several hours until the
required criteria are fulfilled.
There are several boundary conditions which assist in drawing the flow net. The up and
downstream ground surface are equipotential lines. Flow lines are parallel along
impermeable surface and equipotential lines meet these surfaces at right angles.

Fig. 6.2 Flow net for a water storage structure

By understanding the characteristics of the flow nets, one can reach the following
equations:
 49

b1
∆Q1 = K∆H 1
l1

b2
∆Q2 = K∆H 2
l2

Since b/l = 1 and ∆H1 = ∆H2, therefore ∆Q1 = ∆Q2

If the number of chosen equipotential drops is Nd, we have:

H
∆H =
Nd

and

H
∆Q = K
Nd

In the number of flow channels is Nf, the total flow or seepage per unit width of the weir
is:

Q = N f ∆Q

Nf
Q = KH (6.8)
Nd

The hydraulic gradient across any field is

∆H H
i= =
l lN d

and the average seepage per unit volume across any field parallel to the direction of flow
is:

H
Ps = iγ w = γw
Ndl
 50

If the side slope of the last square at the exit is le, the exit hydraulic gradient is:

H
ie = < is
le N d

Theoretically when ie is less than is, the structure is safe against piping. As a safety factor
against unexpected increase in the head, the calculated exit gradient should not exceed
the safe hydraulic gradient given in Table 6.2.

Table 6.2: Safe exit gradient for different types of soils

Type of material Safe gradient (is)
Shingle 1/4-1/5
Coarse sand 1/5-1/6
Fine sand 1/6-1/7

From the above description of the method one may note that flow lines are constructed by
trial and error until a satisfactory picture is obtained.
Determining the uplift pressure in a foundation with an irregular base and several lines of
cutoff walls by this graphical method can be very time consuming and the flow net
becomes complicated.
While the method is accurate, since it represents the solution of the flow equation, the use
of the method in designing small weirs is not common, because of the involvement of
lengthy trial and error methods in finding the acceptable shape of the flow net.
A question which arises often is: what is the appropriate size of the flow net squares? The
answer is, the smaller the size of the squares the more accurate the result is. The size of
the squares also depends on the shape of the foundation. A foundation which has many
corners, as is the case when several piles are attached to it requires very small squares in
order to fulfill the flow net criteria described above.
Fig. 6.3 is the flow net for the problem in Example 6.1, which was solved by the Bligh
and Lane methods.
The pressure at point X = H-n(∆H/Nd) = 6-13(6/23) = 2.61 m
n = number of squares from upstream to the point considered
N = Total number of the squares between two flow lines:

The exit gradient ie = ∆H/le = (6/23)/1.8 =1/6.9.

Compare the above results to the ones obtained by the Bligh and Lane methods.
According to the result of the flow net method, the structure is safe on fine sand.
It should be noted that in the given example, the tail water depth is assumed to be zero.
The same principle can be used for other values of tail water. In that case the head (H) is
the difference between the upstream and the downstream water level.
 51

When it is necessary to estimate the amount of water which percolates through the
subsoil from one side of the structure to another, for example, when a structure is built to
create a reservoir, the seepage rate is calculated by applying Eq. (6.8).
The approximate value of K is given in Table 6.3. For the given example, assume that the
soil is coarse sand, which has particle size D20 = 0.6 mm, where D20 is the size which
corresponds to 20% of the soil sample finer in the sieve analysis. For the soil

K = 1100 x 10-4 cm/s = 1.1 x 10-3 m/s

Q = 1.1 x 10-3 x 6 x 6/23 = 1.72 x 10 m3/s per meter width of the structure.

It can be shown by using the flow net that an efficient way to reduce the exit gradient is
to provide a cutoff wall at the end of the apron.

Table 6.3 Approximate coefficient of permeability

 52

Soil classification 20% size* Coefficient of permeability Note

(mm) K x 10-4 cm/s
Coarse clay 0.005 0.030 The figures in this
Fine silt 0.010 0.105 table represents a
Coarse silt 0.020 0.40 very rough
0.030 0.85 approximation of
0.040 1.75 average conditions
0.050 2.80 in the field.

Very fine sand 0.060 4.60

0.070 6.50 A difference in
0.080 9.0 density,
0.090 14.0 temperature or
0.100 17.5 porosity may
account for a wide
difference in the
Fine sand 0.120 26.0
coefficient of
0.140 38.0
permeability. *
0.160 51.0
The 20% size is
0.180 68.5 that size where
0.200 89.0 20% of the sample
0.250 140.0 is smaller and 80%
coarser.
Medium sand 0.300 220.0
0.350 320.0
0.400 450.0
0.450 580.0
0.500 750.0

Coarse sand 0.600 1100

0.700 1600
0.800 2150
0.900 2800
1.000 3600

Fine gravel 2.000 1800

Energy Dissipater and its Effect on the Apron Length

It was explained earlier that construction of a solid apron is required to accommodate the
hydraulic jump. The relation between the exit gradient and length of the apron was also
established. Constructing ordinary horizontal flat aprons to match the required hydraulic
jump and existing gradient criteria usually results in a massive costly structure. The
length of apron can be reduced by introducing cutoff walls, but there is a limit however
for this reduction, since the hydraulic jump must not be allowed to occur outside the
apron. To control location of the jump and reduce the construction cost, energy
 53

dissipaters should be introduced. Design of the energy dissipaters depends on values of

Froude number in pre- and post- jump and the water depths.
Fig. 6.10 presents the design criteria for various types of energy dissipating structures and
stilling basins.
54
 55

The Design Procedure

1. Determine the pre- and post- hydraulic jump water depth.

2. Determine the pre-jump velocity V1 and Froude number, F1.
3. From Fig. 6.10 select a suitable basin to satisfy the flow condition, taking into
consideration the practicality in constructing it. The figure is self explanatory;
however, with the additional explanation given Table 6.5, the basin selection
should not be a problem.
4. The basin selected in step three is adequate to disperse the flow energy and
accommodate the jump, but it does not necessarily give an adequate creep length
to guarantee a safe exit hydraulic gradient. Therefore, the exit gradient should be
calculated for the given basin length. If it is greater than the safe gradient for
particular soil, the basin length must be increased at its end, or intermediate cutoff
walls introduced. However, the distance between the basin blocks and the sill
must not be changed.

Table 6.5 Selection of the Stilling Basin Criteria

Type of Froude no. F1

basin
- Not economic
I All ranges - The jump entirely depends on the tail water and it may
sweep away from the basin if D2 > D3.
- The basin length is smaller than basin I by 33% and
II > 4.5 disperse the energy within the basin.
- Its construction is a little complicated because of the form
work of the dentated sill and chute blocks.
- Suitable for cases where V < 15 m/s
III < 2.5 - The basin length is smaller than basin I by 60%, but it is
more difficult to construct because of the form works of the
end sill.
- The basin length is the same as the length of basin I, but it
IV 2.5 < F1 < 4.5 guarantees the occurrence of the jump within the basin and
reduces waves result from imperfect jumps.

Protection Work for the Structure

The structure must be well protected from the river overflow and water creep along the
walls must be prevented. Construction of wing walls at both ends of the structure is
essential to anchor it into the embankment. The upstream wings are usually constructed at
an angle of 30 – 40o with the river bank, and it should be extended beyond the top line of
the embankment by at least 1.00 m. Type of required protection and height of
embankment, depend on the frequency of the river overflow and the investment cost of
both the structure and the protection work. Fig. 6.11 presents a typical layout of a
structure.
 56

To protect the channel bed from being eroded by the current leaving the solid apron,
usually riprap, is placed. The length, size of stones used, and thickness of the layer must
be carefully chosen according to certain criteria.

Length of the Protection Work

On the basis of Lacy’s work on the channel scouring the following two formulas obtained
for the scouring depth up and downstream of the solid apron:

Upstream scour Ru = 1.7q2/3 (6.11a)

Downstream scour Du = 2q2/3 (6.11b)

Where q is the discharge per metre width of the channel.

The formulae include a safety factor of 1.25 and 1.50 for the up and downstream scouring
depth respectively. In the two formulae the soil is assumed to be silt and the silt factor is
one.
The scouring depth below the channel bed, Ds is:

Ds = R – D (6.12)

Where D = the water depth just off the apron.

An apron length of 1.5Ds is adequate to protect the structures ends from scouring. By
substituting Eq. (6.11) into Eq. (6.12) and applying the factor 1.5 to the scouring depth,
the resulting apron length would be:

Lu = 2.55q 2 / 3 − 1.5Du (6.13a)

Ld = 3.00q 2 / 3 − 1.5Dd (6.13b)

Where Lu and Ld = length of the protection work up and downstream of the solid
apron respectively.
Du and Dd = the water depth corresponding to the design discharge just of
the solid apron at up and downstream of the weir respectively.

The greater of the following also applies:

L > 4D, (6.14a)

Where D is the tail water.

L = length of earth transition (in canal structures)

L > 1.50 m (6.14b)

 57

Obviously since Eq. (13) is based on the scouring depth, they must always be fulfilled.
Therefore, the designer should use the greatest value results from the two Eq. (6.13) and
(6.14).

Size of riprap stones

The minimum size of stones used in the riprap is related by different investigators to the
bottom velocity. The relation presented in Fig. 6.12 is the result of experimental work
carried out by the USBR. To use the figure the bottom velocity is needed. In natural
rivers, it is not possible to calculate the bottom velocity by a formula. To use the graph it
 58

is recommended to use the average velocity at the basin’s sill. For a structure without the
stilling basin, the following equation is recommended:

Vb = 2 g∆z

where Vb = velocity in m/s

∆z = the drop (height of the crest above the d/s floor in meters).

Thickness of the layers

Riprap is usually placed on layers of gravel and sand. The minimum thickness of each
layer should not be less than 0.10 m. The riprap on the top of the filter should have a
minimum thickness of 1.5 to 2 of the largest stone diameter.

Grain size distribution of the filter materials

To determine the grain size distribution of the filter layers, the following steps need to be
taken:

1. Determine the minimum size of the riprap stones as was described above and
select a range of stone sizes between the calculated minimum diameter and the
maximum size. The maximum size is the size of stones that can be carried by an
average man without difficulty, if it is placed manually.
2. Soil samples at about 0.50 m below the river bed are to be taken for the sieve
analysis and the result is drawn from the usual sieve analysis curve, Fig. 6.13.
3. Apply the following criteria, to determine the limits of the grain sizes in each
layer.
a. To secure adequate soil permeability

d15 upperlayer
L1 < < L2 (6.16)
d15lowerlayer

where d15 is the particle size which corresponds to 15% passing on the
sieve curve, and where L1 and L2 are the upper and lower limits of the ratio. The limit
values are as follows.

Type of soil L1 L2
Homogeneous grains (gravel) 5 10
Homogeneous angular grains 6 20
(broken gravel rubble)
Well graded grains 12 40

(b) To prevent the filter from clogging by soil particles in each layer should
satisfy
d5 > 0.75 mm
 59

d5 is the particle size which corresponds to 5% passing on the sieve curve.

(c) To prevent fine bed materials from being washed away

d15 upperlayer
<5
d 85lowerlayer

and

d 50 upperlayer
L1 < < L2 (6.19)
d 50lowerlayer

d85 and d50 are particle sizes corresponding to 85% and 50% passing on the sieve curve
respectively.
The lower and upper limits of Eq. (6.19) are as follows:

Type of soil L1 L2
Homogeneous grains (gravel) 5 10
Homogeneous angular grains 10 30
(broken gravel rubble)
Well graded grains 12 60

The boundaries of the filter layers are calculated by applying the above criteria to the
riprap stones and the subgrade (river bed) material and poting the results on the same
sheet of the sieve analysis curves of the soil sample.
60
 61

IV - Structural Analysis of Diversion Weirs

Diversion weirs are constructed from a variety of materials. The most commonly used
materials are reinforced concrete, masonry, and gabions. However, whatever materials
are used, the structural analysis remains almost the same.

Acting Forces on Weirs

All external forces acting on a weir are the result of flowing water in the canal or river on
which the structure is constructed. A typical force system of a weir consists of the
following components:

1. static water pressure of the surface water

2. uplift water pressure
3. soil reaction at the weir base
4. friction forces at the base which develop to balance the horizontal forces
5. weight of the weir and water wedges.
Usually in structural analysis of weirs the dynamic force is neglected, since water behind
the weir is built up gradually, and the uplift pressure which results from the arrival of a
new wave does not develop instantly. In seasonal rivers there is a little or no uplift
pressure when the first wave hits the weir, thus the force system which occurs at this
moment, is not the most critical one, especially in weirs which are constructed
monolithically with the apron.
Fig. 8.1 presents typical sloped face weir, showing all the forces acting on them. The
components of the force system are discussed in the following:
1. Weight of the weir
This is calculated simply by multiplying unit weight of the weir by its volume.

2. Weight of the water wedges

The water wedges present the weight of water that is on the weir body and act, either
against or in favor of the weir stability, it depends on the slope of the weir and the water
surface downstream.

3. Upstream water pressure

Its value can be easily calculated if the effect of changing the static pressure, upstream, to
the dynamic one downstream is neglected. In high dams this effect can be substantial and
it causes an over estimate of the design if it is ignored, i.e. the structure will have a higher
factor of safety against overturning.

4. Downstream water pressure

If a weir is designed to match the lower profile of a free water overflow, theoretically, the
water pressure on the face of the weir should be nil, as it is the case in WES ogee-shaped
weir. However, in practice this is hardly the case since the structured weir may not match
the designed shape 100%. Therefore the water curve downstream of the weir cannot be
determined theoretically since the combined effects of the weir geometry and the
condition of the tail water are unpredictable. Theoretical determination of the water
 62

pressure on the face of the weir and weight of the water wedges downstream is not a
straightforward process. The designers, therefore have two choices:
(i) Either to ignore pressure on the d/s face of the weir, in this case increase
slightly the toppling safety factor, or
(ii) Approximately draw the water surface and calculate the water weight. Fig. 8.1
shows a typical weir with the rapidly variable flow at their d/s side.

General Stability Conditions

For a structure to remain stable the following conditions must be fulfilled.

1. Summation of all moments about a point must be equal to zero:

∑M a =0

2. Summation of all horizontal forces must be equal to zero:

∑H f =0

3. Summation of all vertical forces must be equal to zero:

∑V f =0

The above conditions need to be explained in relation to diversion weirs.

For a structure to remain stable, the moments which tend to topple it must be equal to the
moments which balance it. In practice, this condition does not satisfy design engineers,
since unpredictable situations are likely to occur and cause the toppling moment to
exceed the balancing one and hence the structure fails. Usually a safety factor of about
1.5 to 2 is applied.

∑M balance
> 1 .5
∑M topple

in order to avoid lifting up the structure’s heel and tension occurrence at the base, the
forces must pass through the middle third of the structures base, i.e.

B
eccentricity e < or
6

B B
e= −X <
2 6

where
 63

X =
∑M
∑V
f

∑ M = Summation of all moments about the structure toe (Fig. 8.3)

∑V =Summation of vertical forces excluding the base reaction.
f

X = Distance of the resultant of the forces from the toe

B = Width of the weir base
64
 65

The structure may slide in the flow direction if there is not enough grip between the base
and the foundation. To prevent this happening, the following condition should be
satisfied.

Horizontal external forces

< f
Vertical external forces

where f is the friction factor between the base and the foundation.

It should be noted that, when the net vertical force is calculated, the foundation reaction
should not be included (Fig. 8.1). The foundation reaction occurs as a result of the
unbalanced moment and the vertical forces, i.e. weight of the structure and the water
uplift pressure. If the soil reaction is included, sum of the vertical forces will be zero.
The friction coefficient (f) depends on the materials used in the construction and nature of
the foundation. The USBR suggest 0.35 for concrete structures on common soils and 0.65
is recommended for the friction between masonry and concrete

Critical Cases to be considered

It has been observed over the years that diversion weirs collapse, initially not because of
the unbalanced moment, but mainly due to the foundation scouring. The stability analysis
becomes important where the structure and the apron are of two different materials and
act as two independent units, i.e. nonmonolithically built, or when the structure is built on
rock foundation without apron. Two cases are considered:

1. weirs built non-monolithically with the foundation, and

2. weirs constructed monolithically with apron.

Weirs Constructed Non monolithically with the Foundation

A sloped face weir is considered. Fig. 8.3 shows this case and Table 8.1 presents the
acting forces and moments under the following assumptions:

1. The base of the weir is not sealed, i.e. water seeps from the upstream side freely
through the base to the downstream side.
2. No water is flowing over the weir and no water ponding is in its downstream.

Stability Analysis

The moments and forces given in Table 8.1 are first evaluated and the following steps
followed:

1. The toppling moments ∑M a + = (sum of all positive moments without

moments of the base reaction)
 66

2. The balancing moment ∑M a − = sum of all anticlockwise moments. The

positive and negative signs are taken as arbitrary for clockwise and anticlockwise
respectively. Of course, if the sign is reversed systematically it does not affect the
result.

3. Safety factor =
∑ M a − If the safety factor is acceptable, proceed to the next
∑Ma +
step otherwise revise dimensions of the structures.
4. Determine the eccentricity (e)

Net moment Ma = ∑M a − +∑ M a +

Mn
X =
∑V f
∑V f = Summation of all vertical forces excluding the foundation reaction

B
e= −X
2
If e < (B/6) proceed to the next step. Otherwise revise the dimensions of the weirs and
repeat the procedure from step 1.

5. Determine the contact pressure on the foundation.

P1 , P2 =
∑V f
+
e∑ V f
A 1
xDx(W f ) 2
6

where A = area of the base. In most cases engineers take one meter across, i.e. the area
equal width of the weir x 1.00 m.
Wf = width of the weir at floor level.
D = cross dimension of the base = 1.00 m.
The largest value of P1 and P2 must not exceed the bearing capacity of the foundation
material, otherwise the geometry of the base should be changed. P1 and P2 can be
determined by applying the stability equations explained previously, as follows:

W f ( P1 − P2 ) D
∑V f − P2 xW f xD −
2
=0

Wf Wf
P2 xW f xD (P1 − P2 ) + ∑Ma− = 0
2 3

Eqs. (8.10) and (8.11) can be solved simultaneously to find P1 and P2.
67
 68

Weirs constructed monolithically with the Foundation

On many occasions weirs are constructed as one integrated unit with the apron. In this
case, the structure is more stable and it would not fail as a result of unbalanced moments.
This is certainly true in low weirs constructed from RC. A typical case is shown in Fig.
8.5 and the acting forces are presented in Table 8.3.

Design of the Apron

There are two approaches for determination of the apron thickness. The first one assumed
that the apron consists of individual unit volumes which are structurally not linked, and
the weight of each individual unit balances the uplift pressure. While this assumption
leads to an increase in the structure’s cost, the computations involved are very easy and
 69

result in a structure with a high safety factor. Thickness of the apron is calculated as
follows.
The uplift pressure under the apron must first be calculated, assume here it is linearly
distributed. From Fig. 8.7 it can be established that the required thickness of the apron (t)
= uplift pressure divided by γ fb , where γ fb = submerged density of the apron material.
In the equilibrium condition, the weight of the floor must balance the uplift pressure, i.e.

γ w ( h x + t ) = tγ f
hx γ w hγ
t= = x w
(γ f − γ w ) γ fb
hx
t=
(γ f − 1)

hx = uplift pressure at point x, (m)

t = thickness of the floor (m)
γ f = density of the floor materials, kg/m3
γw = densityofwater (kg / m3)
γ fb = submergeddensityofthefloormatearils = γ f − γ w
Fig. 8.7 Uplift pressure due to unbalanced head of water (H)

Assignment to be submitted:
Apply Table 8.3 to determine the following for the weir presented in Fig. 8.5:

a. the safety factor against overturning,

b. the factor of safety against sliding
c. contact pressure on the foundation
d. Also determine the thickness of the apron.
Given: H = 2.00 m, Lu = 6 m, Ld = 10 m, b = 0.50 m, S = 0.50 m, t1 = 0.80 m, t2 = t3 =
0.40 m, γc = 2.25 t/m, h = 0.