1

University of Eldoret
School of Engineering
Department of Civil and Structural Engineering

Course Name: Flow in Non-erodible Open Channels

Course Code: CSEN 315
Units: 3
Year: 3
Semester: 1
Pre-requisite: Introduction to Fluid Mechanics
Lecturer: Dr. C.K. Kiptum

Course Map

Course Goal Learning Objective(s) Assignments Indicative contents

/Activities
Learners will be able to By the end of this Lecture Types of flow: Steady, unsteady,
understand the basic course, students will uniform, non-uniform such as
principles of fluid flow be able to distinguish gradually and rapidly varied flow.
in non-erodible open between various
channels types of open channel Momentum and energy concepts:
flow and geometry sub-critical, critical and super-
critical flow, Froude’s number.
Rapidly varied flow: hydraulic
jump, bore.
Gradually varied flow: basic
equations, back water curves, flow
past islands.
Learners will be able to By the end of this Lecture Design aspects of open channels
learn design, operation course, students will and their engineering structures.
and maintenance of be able to derive
open channels and equations and design
related structures. channels according to
geometry and type of
flow.
By the end of this Tutorials Design aspects of open channels
course, students will and their engineering structures.
be able to design
specific hydraulic
structures.
Learners will be able to By the end of this Practical/ Applications: flumes, weirs, deep
appreciate the course, students will Laboratory and shallow stilling basins.
application and the be able to plan, Field work
practical aspects of organize, install, Demonstration Waves and bores in water
open channel operate and maintain s and
hydraulics, both in the a water resource presentations
laboratory and in the engineering scheme. Group
field. presentations
 2

Course Assessment
Continuous Assessment Tests - 20%
Laboratory Practical - 10%
End of semester Examination - 70%

Course Reference
Fluid Mechanics, 4th Edition- Douglas JF, Gasiorek JM, Swaffield JA.
Mechanics of Fluids, 3rd Edition, Merle C. Potter and David C.Wiggert
Solving Problems in Fluid Mechanics 1-Douglas JF.
Solving Problems in Fluid Mechanics 2 - Douglas JF.
Essentials of Hydraulics 2nd Edition - Dake JMK.
Fluid Mechanics for Civil Engineers - Webber NB, 2002
Applied Fluid Mechanics 7th Edition by Mott and Untener. Global Edition published by Pearson
Education Ltd: Essex, England.
Donald. F. Young, Bruce R. Monson and Theodore H. Okiishi. A brief Introduction to fluid
Mechanics. Second edition. New York: John Wiley and Sons Inc.

1.0 Introduction
Open-channel flow involves the flow of a liquid in a channel or conduit that is not completely
filled. There exists a free surface between the flowing fluid (usually water) and fluid above it
(usually the atmosphere). The main driving force for such is the fluid weight-gravity forces the
fluid to flow downhill.

Fluids flow under the action of forces, deforming continuously for as long as the force is applied.
When a fluid flows over a solid surface or other boundary, whether stationary or moving, the
velocity of the fluid in contact with the boundary must be the same as that of the boundary.
Conditions in a body of fluid can vary from point to point and, at any given point, can vary from
one moment of time to the next.

Free-surface flow or open channel flow is the common flow phenomenon in the world. They can
be natural like ocean waves, river currents, and overland flow of rainfall or can be human-
induced flows in canals and culverts, drainage over impervious materials such as roof gutters and
parking lots, and wave motion in harbours. An open channel is a flow system in which the top
surface of the fluid is exposed to the atmosphere. In free surface flow, there is an interface
between the air and the upper layer of water called free surface where the pressure is atmospheric
and is constant meaning that the hydraulic grade line and the liquid surface coincide (Figure 1.1).
Hydrostatic pressure
distribution assumed 2
V
2g Energy grade line (EGL)

y
Water surface (WL)
V
θ
so 1 z
x

Figure 1.1 Reach of open-channel flow

 3

The channel is so small. Sinθ =h/L= So. Since the angle is so small, it is rarely used to measure
the slope.
Figure 1.1 shows a channel in which the flow is nearly horizontal. In such a case, there is little or
no vertical acceleration of fluid within the reach, and the streamlines remain nearly parallel. Such
conditions are common in open channel flows where streamlines are assumed to be parallel.
Since the vertical acceleration are near zero, one can conclude that in the vertical direction the
pressure distribution is hydrostatic as shown in Figure 1.1. Therefore, the sum of (p + γz)
remains constant at any depth, and the hydraulic grade line coincides with the water surface. In
open channel flow z represents the elevation of the channel bottom, and y as the depth of flow.
Since at the channel bottom p/γ = y, the hydraulic grade line is given by the sum (y + z).

In most civil engineering practice, most open channels convey water as the fluid. The flow is
normally three dimensional; however simplifications are done by using two-dimensional and
one-dimensional analysis. River and channel flows are usually treated as one-dimensional with
respect to the position coordinate along the streambed. In this course we shall deal with one-
dimensional flow Figure 1.2 where it is assumed that the velocity is equal to V everywhere at a
given cross section. Flow in an open channels are most likely to be turbulent, and the velocity
profile can be assumed to be approximately constant.

Figure 1.2: One dimensional model

1.1 Types of flow

Flow in a channel is characterized by the mean velocity, even though a velocity profile exists at a
given section (Figure 1.3). The flow is classified as a combination of steady or unsteady, and
uniform or non-uniform.

y v
x

Figure 1.3 Velocity profile

Uniform flow is flow where the velocity at a given instant is the same in magnitude and direction
at every point in the fluid. Uniform flow is the situation where terminal velocity has been
reached in a channel of constant cross section; not only is the mean velocity constant, but the
depth is invariant as well. Non uniform flow occurs when the velocity changes from point to
point. In practice, when a fluid flows past a solid boundary there will be variations of velocity in
 4

the region close to the boundary and hence a non-uniform flow. However, if the size and shape
of the cross-section of the stream of fluid is constant, the flow is considered to be uniform.

A steady flow is one in which the velocity, pressure and cross-section of the stream may vary
from point to point but do not change with time. If, at a given point, conditions do change with
time, the flow is described as unsteady. In practice, there will always be slight variations of
velocity and pressure, but, if the average values are constant, the flow is considered to be steady.

The four possible types of flow that can occur are:

Steady uniform flow

Conditions do not change with position or time. The velocity and cross-sectional area of the
stream of fluid are the same at each cross-section; e.g., flow of a fluid through a pipe of uniform
bore running completely full at constant velocity.

Steady non-uniform flow

Conditions change from point to point but not with time. The velocity and cross-sectional area of
the stream may vary from cross-section to cross-section, but, for each cross-section, they will not
vary with time; e.g., flow of a liquid at a constant rate through a tapering pipe running
completely full. Steady non-uniform flow is a common occurrence in rivers and in man-made
channels. In such situations, it will be found to occur in two ways. In relatively short reaches,
called transitions, there is a rapid change in depth and velocity; such flow is termed rapidly
varied flow. Examples are hydraulic jump, flow entering a steep channel from a lake or reservoir,
flow close to a free outfall from a channel, and flow in the vicinity of an obstruction such as a
bridge pier or a sluice gate.

Along more extensive reaches of channel the velocity and depth may not vary rapidly, but rather,
change in a slow manner. Here the water surface can be considered continuous and the regime is
called gradually varied flow. Examples of gradually varied steady flow are the backwater created
by a dam placed in a river, the drawdown of water surface as flow approaches a fall. Figure 1.4
illustrates how both rapidly varied flow (RVF) and gradually varied flow (GVF) can occur
simultaneously in a reach of a channel.

Hydraulic Sluice gate Hydraulic

Jump Jump

RVF GVF RVF GVF RVF GVF RVF

Figure 1.4 Steady non-uniform flow in a channel

 5

Open channel flow is classified as uniform flow (UF) if the depth of flow does not vary along the
channel (dy/dx = 0), where y is the fluid depth and x is the distance along the channel.
Conversely, it is non-uniform flow or varied flow if the depth varies with distance (dy/dx ≠0).
Non-uniform flow are further classified as rapidly varying flow (RVF) if the flow depth changes
considerably over a relatively short distance; dy/dx ≈ 1. Gradually varying flows (GVF) are
those in which the flow depth changes, slowly with distance along the channel; dy/dx<<1.

Gradually varied flow is a transition zone between two uniform flow sections and can be due to:
(a) Hydraulic structures- gates, weirs, dams.
(b) Natural objects-drops, waterfalls, channel slope changes, obstructions (e.g. rocks)

Unsteady uniform flow

At a given instant of time, the velocity at every point is the same, but this velocity will change
with time; e.g. accelerating flow of a liquid through a pipe of uniform bore running full, such as
would occur when a pump is started up. Unsteady uniform flow rarely occurs.

Unsteady non-uniform flow

The cross-sectional area and velocity vary from point to point and also change with time: e.g. a
wave travelling a channel. Flood waves in rivers, hydraulic bores, and regulated flow in canals
are other examples.

The possible combinations are shown in Table 1.1; the position coordinate is designated as x.

Table 1.1 Combinations of one-dimensional Free-surface Flows

Type of flow Average velocity Depth
Steady, uniform V = constant y = constant
Steady, non-uniform V= V(x) y = y(x)
Unsteady, uniform V= V(t) y = y(t)
Unsteady, non-uniform V= V(x,t) y = y(x,t)

1.2 Froude Number

The primary mechanism for sustaining flow in an open channel is gravitational force. For
example, the elevation difference between two reservoirs will cause water to flow through a
canal that connects them. The parameter that represents this gravitational effect is the Froude
number. It is the ratio of inertial force to gravity. The Froude number, F r, is the ratio of the
velocity in a channel divided by the speed of propagation, or celerity of a small amplitude wave,
c. In the context of open-channel flow, V is the mean cross-sectional velocity, and L is a
representative length parameter. For a channel of rectangular cross section, L becomes the depth
y of the flow.

√
2
V V V Q B
F r= = = = …………………………1

√
c √ gL A gA
3
g
B
 6

By having knowledge of the magnitude of Froude number, one can ascertain significant
characteristics regarding the flow regime. For example, if Fr>1, the flow possesses a relatively
high velocity and shallow depth; on the other hand, when Fr < 1, the velocity is relatively low
and the depth is relatively deep. Except in the vicinity of rapids, cascades, and waterfalls, most
rivers experience a Froude number less than unity. Constructed channels may be designed for
Froude numbers to be greater or less than unity, or to vary from greater than unity to less than
unity along the length of the channel.

An open-channel flow is classified according to how the average velocity, V, of the flow
compares with the speed, c, of a small amplitude gravity wave in that channel. The speed of a
small amplitude gravity wave is given by:

c=
√
gA
B
=√ gy
Where g is the acceleration of gravity, A is the cross section of the flow, B is the top width of the
flow and y is the hydraulic depth, (A/B).

2.0 Uniform flow

Though rare, its consideration helps in understanding other types of flow since it is the basis in
which derivations for other types is done.
Channel Geometry
Channel cross sections can be considered to be either regular or irregular. A regular section is
one whose shape does not vary along the length of the channel, whereas an irregular section will
have changes in its geometry. In this course we shall consider only regular channels which are
common.
B B
B

y 1 y 1
b m1 m2
b y α d

Rectangular Trapezoidal
circular

Figure 2.1 Representative regular cross sections

Rectangular section
The simplest channel shape is a rectangular section. The cross sectional area is give by
A=by … … …..2. 1

Where b is the width of the channel bottom.

 7

Additional parameters of importance for open-channel flow are the wetted perimeter, the
hydraulic radius, and the width of the free surface. The wetted perimeter P is the length of the
line of contact between the liquid and the channel.
For a rectangular section, the wetted perimeter is

P=b+2 y … … … ..2. 2

The hydraulic radius R is the area divided by the wetted perimeter, i.e.

A by
R= = … … …..2. 3
P b+ 2 y

Trapezoidal section
The trapezoidal section has the added feature of sloped side walls. If m 1 is the ratio of the
horizontal to vertical change of the wall on one side, and m 2 is the corresponding quantity on the
other wall, the area, wetted perimeter, and free-surface width are given as.

B=b + y ( m1 +m2 ) … … … … 2.4

1 2
A=by + y ( m1 +m 2 ) … … …..2. 5
2
P=b+ y ( √ 1+m12 + √ 1+m 22) … ..2.6

If m1 and m2 are zero, the relations A, P and B become those of a rectangular channel. If b is set
as zero the equations 2.4 to 2.6 describe the geometry of a triangular shaped channel.

Circular
The circular cross section is an important one to consider, since many free-surface flows in
drainage and sewer systems are conveyed in circular conduits. If d is the conduit diameter, the
area, wetted perimeter, and free-surface width are given by. Assignment 1

2∝ d d ∝d 2
Area of sector= ×π× × =
2π 2 2 4

d
Area of Triangle with sides B , ,d /2
2

2
d (
A= ∝−sin ∝cos ∝ ) … … … ..2.7
4

∝
P= × π ×d ×2
2π

P=∝ d … … 2.8
 8

Let x = α-90
Therefore –x = 90-α
y
=cos (−x )=cosx=cos ( 90−∝ )=sin∝
d
2

Therefore
d
y= sin ∝
2
But B = 2y

B=dcos ∝… … … ..2. 9

2.2 Equations for Uniform flow

Uniform flow occurs in a channel when the depth and velocity do not vary along its length, that
is, when terminal conditions have been reached in the channel. Under such conditions, the
 9

energy grade line, water surface, and channel bottom are all parallel. The equation for uniform
flow takes the form
V =C √ R S0 … … … … .2 .10

Where S0 is the slope of the channel bottom and C is the Chezy coefficient, which is independent
of the Reynolds number since the flow is considered completely turbulent. In engineering
practice we relate C to channel roughness and the hydraulic radius by use of the Manning
relation.
1
1 6
C= R … … … … 2.11
n

Combining equations 2.10 and 2.11 and finding discharge results in the Chezy-Manning
equation.

1
Q= A R √ S 0 … … … 2.12
2/ 3
n

In Equation 2.12, one can solve explicitly for Q, n or S 0. In determining y0 trial solutions is used
when the other parameters are given.

Values of the Manning coefficient n are given in Table 2.1. The values in Table 2.1 result in flow
rates too large for hydraulic radii greater than 3m. The manning n should be increased by 10 to
15% for such large conduits.

Table 2.1 Average Values of the Manning n

Wall Material Manning n

Planed wood 0.012
Unplaned wood 0.013
Finished concrete 0.012
Unfinished concrete 0.014
Sewer pipe 0.013
Brick 0.016
Cast iron, wrought iron 0.015
Concrete pipe 0.015
Riveted steel 0.017
Earth, straight 0.022
Corrugated metal flumes 0.025
Rubble 0.03
Earth with stones and weeds 0.035
Mountain streams 0.05

The depth associated with uniform flow is designated y 0; it is called either uniform depth or
normal depth. Uniform flow rarely occurs in rivers because of the irregularity of the geometry. In
manmade channels it is not always present since the presence of controls such as sluice gates,
weirs, or outfalls will cause the flow to become gradually varied. It is necessary to determine y 0
when analyzing gradually varied flow conditions, since it provides in part a basis for evaluating
 10

the type of water surface that may exist in the channel. The design of gravity flow sewer
networks is often based on assuming uniform flow and the use of Chezy-Manning equation, even
though much of the time the flow in such systems may be non-uniform.

Example 1
Water is flowing at a rate of 4.5 m 3/s in a trapezoidal channel whose bottom width is 2.4 m and
side slopes are 1 vertical to 2 horizontal. Compute y0 if n = 0.012 and S0 = 0.0001.
Use excel sheet to calculate or other method.

Example 2
Uniform flow occasionally occurs in a 5 –m- diameter circular concrete conduit, but the depth of
flow can vary. The manning coefficient is n = 0.013, and the channel slope is S o = 0.0005. Plot
the discharge-depth curve. Use excel, MATLAB or any program and come with this figure.

4.5

3.5

2.5
y

1.5

0.5

0
0 5 10 15 20 25 30 35 40 45
q(y)

2.3 Most efficient section

Design of a channel to convey uniform flow typically consists of selecting or specifying the
appropriate geometrical cross section provided that Q, n, and S0 are known. Once chosen, an
optimum sizing of the cross section can be based on the criteria of minimum resistance to flow.
The flow resistance per unit length equals surface shear stress times wetted perimeter. Using a
control volume for uniform flow, and assuming a small slope S0 so that sin θ ≈ S0 (Figure 1.1),
one can show that the resistance per unit length is:
τ 0 P=γA S 0 … … …..2.1 3

Hence a least resistance criterion is equivalent to requiring a minimum cross-sectional area with
respect tothe parameters defining the area. Furthermore, we have to satisfy the Chezy-Manning
equation. Since Q, n, and S0 are given and R=A/P, Equation 2.12 can be written as
5
P=c A 2 … … … … 2.14
 11

In which c is a constant. As an example, consider a rectangular channel with width b and depth
y. The best hydraulic cross section is obtained by rewriting equation 2.14 such that A becomes a
function of b only. Therefore, we express P in terms of A and b, using A=by and P=b+2y:
2A
P=b+ … … … … 2.15
b
Substitution of this equation into 2.14 yields
5
2A
b+ =c A 2 … … …..2.1 6
b
Now, differentiate Eq 2.16 with respect to b, remembering that A is dependent on b:
3
2 dA 2 A 5 dA
1+ − 2 = c A2 … … ..2.1 7
b db b 2 db

Setting dA/db = 0, since the objectives is to find the value of b that minimizes A. The result is

2A
2
=1 … … 2.18
b

Or using A =by,
b=2 y … … 2.19
Thus, if the width of a rectangular channel is twice the depth of the flowing water, the water will
flow most efficiently.
For a trapezoidal cross section, it is simpler to begin with equation 2.14, eliminate b, and express
P as a function of A, y, and m, where m=m1 = m2. Then by considering A as a function of m and
y, the minimum of A is found by setting the gradient vector of A to zero. The result is m =
m=
√ 3 or side slope angle of 60o with the horizontal. The resulting hexagonal shape is a
3
trapezoid that best approximates a semicircle. Exercise for students

3.0 Energy concepts in open-channel flow

The energy at any position along the channel is the sum of the vertical distance measured from a
horizontal datum z, the depth of flow y, and the kinetic energy V2/2g. That sum defines the
energy grade line and is termed the total energy H.
V2
H=z + y + … … ..3.1
2g
The energy balance between two points of a channel carrying a fluid is given by the relation
H 1=H 2 +h L … … 3.2
Where hL is the head loss. The only manner in which energy can be added to an open-channel
flow system is for mechanical pumping or lifting of the fluid to take place.

3.1 Specific energy

It is convenient in open-channel flow to measure the energy relative to the bottom of the channel;
it provides a useful means to analyze complex flow situations. Such a measure is called specific
energy and is designated as E:
 12

2
V
E= y + … …..3. 3
2g
Therefore specific energy is the sum of the flow depth y and kinetic energy head V2/2g.

Rectangular sections
For a rectangular section, the specific energy can be expressed as a function of the depth y. The
specific discharge q is defined as the total discharge divided by the channel width, that is.
Q
q= =Vy … … … .3.4
b
The specific energy for a rectangular channel can thus be put in the form
q2
E= y + … … ..3.5
2 g y2

y Fr < 1
y
E=y (asymptote) Fr < 1

Alternate yc Fr = 1
depths
Fr = 1
yc
Fr > 1 Fr > 1
qmax q
Ec E

Figure 3.1(a) E versus y for constant q. Figure 3.1(b) q versus y for constant E

The E-y relation is shown in Figure 3.1a. We may observe that a specific discharge requires at
least a minimum energy; this minimum energy is referred to as critical energy, Ec. The
corresponding depth yc is called the critical depth. If the specific energy is greater than Ec, two
depths are possible; those depths are referred to as alternate depths. For constant q, equation 3.5
is a cubic equation in y for a given value of E which is greater than Ec. The two positive solutions
of y are the alternate depths (E=y) and (y=0).

Another way to express Equation 3.5 is to consider E constant and vary q. Equation 3.5 can be
solved for q as.
q=√ 2 g y 2 (E− y )… … … ..3. 6

This relation is shown in Figure 3.1(b). This form of the energy relation is useful for analyzing
flows in which the specific energy remains nearly constant throughout the transition region;
examples are a change in the width of a channel and the variation of depth with discharge at the
entrance to a channel. Note that maximum unit discharge, qmax, occurs at critical depth.
 13

The figure 3.1 provides a useful means of visualizing the solution to a transition problem. Even
though one probably would solve the problem numerically, an assessment of the graphical
solution may provide useful physical insight as well as prevent the incorrect root from being
selected.

The critical depth yc can be evaluated from Equation 3.5 by setting the derivative of E with
respect to y equal to zero:
dE q2
=1− 3 =0 … … 3.7
dy gy

Since q= Vy, the condition of minimum E is

V2 2
1− =1−Fr =0 … ..3. 8
gy
Where the Froude number in a rectangular channel is
V q
Fr= = … …3.9
√ gy √ g y 3
Thus from equation 3.8, the Froude number is equal to unity for minimum energy. Solving for
the depth in equation 3.7 in terms of q

( )
2 1
q 3
y= y c = … … 3.10
g
This relation gives the critical depth in terms of the specific discharge. Note that at critical flow
conditions, Ec can conveniently be expressed by combining Equations 3.5 and 3.10 to eliminate
q, resulting in
3
Ec = y c … … … ..3.1 1
2

On the E-y curve, for any depth greater than y c, the flow is relatively slow or tranquil, and Fr<1;
such a state is termed subcritical flow. Conversely, for a depth less than critical, the flow is
relatively rapid or shooting, Fr>1, and the regime is one of supercritical flow. The E-y diagram is
a representation of the change in specific energy as the depth is varied, given a constant specific
discharge. It is possible for q to vary, as when the width of a rectangular section changes in a
transition region. As q increases, the E-y curve shifts to the right in figure 3.1 (a).

Generalized cross section

For a generalized section the specific energy can be written in terms of the total discharge Q and
the cross-sectional area A:
Q2
E= y + 2
… … .3 .12
2gA
The minimum-energy condition is obtained by differentiating Equation 3.12 with respect to y to
obtain
dE Q2 dA
=1− 3
… ..3.13
dy g A dy
 14

For incremental changes in depth, the corresponding change in area is dA=Bdy. Thus setting
equation 3.13 to zero, the minimum-energy condition becomes
2
Q B
1− 3
=0 … …3.14
gA
By analogy to equation 3.8, the second term in equation 3.14 is the square of the Froude number
since
Q

√
2
Q B A V
Fr= = = … ..3.15
g A √ gA /B √ gA /B
3

The ratio A/B = y is termed the hydraulic depth, and its use allows the definition of the Froude
number to be generalized.

Example 3.1
Water is flowing in a triangular channel with m 1=m2=1.0 at a discharge of Q= 3 m3/s. If the water
depth is 2.5 m, determine the specific energy, Froude number, hydraulic depth, and alternate
depth.
A=6.25m2, B= 5.0 E = 2.51m, Fr= 0.137, hydraulic depth=1.25m, alternate depth=0.71m.

Hint
This results in a cubic function equation of the form
3 2
y −E y +G=0
If we know one nonzero root, y= r, then dividing through by (y-r) yields the quadratic equation
2 G G
y − 2 y − =0
r r
From which the other roots can be obtained by
G
y= 2 ±
2r √( )
G 2 G
2r 2
+
r

3.2 Use of the Energy Equations in Transitions

A transition is a relatively short reach of channel where the depth and velocity change, creating a
non-uniform, rapidly varied flow situation. The mechanism for such changes in the flow is
usually an alteration of one or more geometric parameters of the channel.

3.2.1 Channel Constriction

1 y
2 y1

Fr < 1 h
y2
Flow
yc

Ec E
 15

Figure 3.2 (a) raised channel bottom Figure 3.2 (b) Specific energy diagram

Consider a rectangular channel whose bottom is raised by a distance h over a short region in
Figure 3.2 (a). The change in depth in the transition can be analyzed by use of the energy
equation, and as a first approximation, one can neglect losses. Specific energy between the two
points is related as follows:
E1=E 2+ h … … … 3.15

The depth y2 at the end of the transition can be visualized by inspection of the E-y Figure 3.2(b).
If the flow at location 1 is subcritical, y1 is located on the upper leg as shown in the specific
energy diagram. The magnitude of h is selected to be relatively small such that y2 > yc, hence the
flow at location 2 is similarly subcritical; however, as h is increased still further, a state of
minimum energy is ultimately reached in the transition. The condition of minimum energy is
sometimes referred to as choking condition or as choked flow. Once choked flow occurs, as h
increases, the variations in depth and velocity are no longer localized in the vicinity of the
transition. Influences may be observed for significant distances both upstream and downstream
of the transition.

Let us examine the case of a hump or weir in a rectangular channel. We shall assume the hump is
streamlined and neglect head loss. In Figure 3.3 a. We see a small hump. If the flow upstream of
the hump is subcritical, there will be a slight depression in the water surface over the hump.

yc
yc

Δh Δhcritical

(a) Δh<<Δhcritical (b) Δh<Δhcritical (c) Δh = Δhcritical (d) Δh>Δhcritical

Figure 3.3 Subcritical flow over a hump in a rectangular channel. Flow rate is the same in all the
four cases. Δh is the hump height and Δhcritical is the critical hump height.

The hump (locally raised bed) causes a drop in specific energy E, since the energy line is
unchanged. A decrease in E with unchanged q is accompanied by a decrease in y that exceeds the
decrease in E, so there must be a depression in the water surface over the hump. If the height of
the hump Δh is increased with no change in q, there will be a further drop in the water surface
over the hump (Figure 3.3 b) for the same reason. Further increases in hump height create further
depression of the water surface over the hump until finally the depth on the hump becomes
critical as shown in Figure 3.3 c. We shall call the minimum height of the hump that causes
critical depth the critical hump height. If the hump is made still higher (Figure 3.3 d), critical
depth remains on the hump and the depth upstream of the hump increases until it gains sufficient
energy to be able to flow over at the same rate. The more hump is raised, the greater will the
upstream depths become, and the further upstream will these effects extend. We refer to this
phenomenon as the damming action.
 16

Narrowing of the channel width will create a situation similar to an elevation of the channel
bottom. Narrowing of a channel change q, the flow per unit width. A small contraction will cause
a slight depression in the water surface through the contraction. Upon further contraction, the
depth reaches critical depth in the contraction. Continued contraction, the depth remains critical
in the contracted section and upstream depths increase. However, unlike for humps, the depth in
the contracted sections continues to increase. This is because the critical depth there depends
upon q, and q increases with contraction. Contraction beyond the onset of the critical depth
requires the flow to have extra energy, and this again comes from the damming action, which
causes the upstream water surface to rise for a considerable distance.

The most general transition region is one that possesses a change in both width and in bottom
elevation.

Example 3.2
A rectangular channel 3m wide is conveying water at a depth y1= 1.55m and velocity V1=1.83
m/s. The flow enters a transition region as shown in Figure 3.3, in which the bottom elevations is
raised by h = 0.2m. Determine the depth and velocity in the transition, and the value of h for
choking to occur.
Answers q = 2.84 m2/s, Fr = 0.47, E1 = 1.72 m and E2 = 1.52 m, yc = 0.94, Ec = 1.41, y2 = 1.26 m,
V2 = 2.25 m/s, value of h for critical flow to appear is 0.31 m.
1 y
2 y1
Fr < 1 h
y2
Flow
yc

Ec E
h
Figure 3.3

3.2.2 Channel entrance with critical flow

Consider flow entering a channel from a lake or reservoir over a short rounded crest Figure 3.4.

yc
Fr1=0 y1
z2
Datum
1 2

Figure 3.4 Outflow from a reservoir with critical flow at the channel entrance.

If the channel slope is steep, the flow will discharge freely into the channel and supercritical flow
will occur downstream of the entrance region. Upstream of the crest the flow may be considered
 17

as subcritical. Since the flow regime at the crest changes from subcritical to supercritical, the
flow at the crest must be critical. To determine the discharge, we assume that rapidly varied flow
takes place over the crest in conjunction with the critical flow condition at the crest. An example
will illustrate the procedure.

Example 3.3
Water flows freely from a reservoir into a trapezoidal channel with bottom width b= 5.0 m and
side slope parameters m1 = m2 = 2.0. The elevation of the water surface in the reservoir is 2.3 m
above the entrance crest. Assuming negligible losses in the transition and a negligible velocity in
the reservoir upstream of the entrance, find the critical depth at the transition and the discharge
into the channel shown in Figure 3.5.

yc
Fr1=0
y1
z2
Datum
1 2

Figure 3.5 Example

Solution
The total energy at location 1 in Figure 3.5 is y 1 since the kinetic energy in the reservoir is
negligible (V1≈0). Equating the total energies at location 1 and 2 gives.
y 1=E 2+ z 2

since critical conditions occur at location 2, Equations 3.12 and 3.14 can be combined to
eliminate discharge, with the result
A
E 2= y c +
2B
Elimination of E2 in the two equations yields the expression
1 2
b y c + ( 2+2 ) y c
A 2
y 1−z 2= y c + = yc +
2B 2 [ 5+ ( 2+2 ) y c ]
With the given data we find
1 2
5 y c + (2+ 2 ) y c
2
2.3= y c +
2 [ 5+ ( 2+2 ) y c ]
The relation above is a quadratic in yc. The positive root is chosen, which gives
y c =1.70 m

Subsequently, one can find that A = 14.28 m 2 and B = 11.8 m. Equation 3.14 was used to
calculate the discharge to be

√ √
3 3
gA 9.8 ×14.3 3
Q= = =49.3 m /s
B 11.8

3.2.3 Energy losses

 18

Energy losses in expansions and contractions are known to be relatively small when the flow is
subcritical; however, it may be necessary under certain circumstances that the losses be
considered. The following experimentally derived formulas can be employed. For a channel
expansion use

h L =K e (
V 12 V 22
−
2g 2 g )
… … ..3.1 7

For a channel contraction use

( )
2 2
V 2 V1
h L =K c − … …..3.1 8
2g 2 g

In equation 3.17, Ke stands for an expansion coefficient; it has been suggested to use K e = 1.0 for
sudden or abrupt expansions, and Ke = 0.2 for well designed or rounded expansions.
For the contraction coefficient Kc in equation 3.18, use Kc = 0.5 for sudden contractions, and
Kc=0.10 for well designed contractions.

3.3 Flow measurement

The most common means of measuring discharge in an open channel is to use a weir. A weir is
an obstruction in the channel designed to make measurements of the flow rate in a stream. A
weir is also a device placed in the channel that forces the flow through an opening or aperture
designed to measure the discharge. Specialized weirs have been designed for specific needs.
A properly designed weir will exhibit subcritical flow upstream of the structure, and the flow
will converge and accelerate to a critical condition near the top or crest of the weir. As a result, a
correlation between discharge and a depth upstream of the weir can be made. The downstream
overflow is termed the nappe, which usually discharges freely into the atmosphere. There are a
number of factors that affect the performance of a weir:
 The three dimensional flow pattern.
 The effects of turbulence
 Frictional resistance,
 Surface tension,
 Amount of ventilation beneath the nappe.

The above effects accounted for by modifying the ideal discharge with a discharge coefficient,
Cd; the actual discharge is the ideal discharge multiplied by the discharge coefficient. When
possible, it is advantageous to calibrate a particular weir in place in order to obtain the desired
accuracy.

Notch geometry for different weirs.

Some of the terms used are defined below:

H
crest
 19

Hc

1
4
L L L

Rectangular Weir Contracted Weir Cipolletti Weir

V-notch weir

A rectangular weir, also called a suppressed weir, has a crest length L that extends the full width
of the channel into which it is installed.
The standard design requirements are:
The crest height above the bottom of the channel Hc ≥ 3Hmax
The minimum head above the crest Hmin > 0.2 ft or 60 mm
The maximum head above the crest Hmax < L/3
A contracted weir is a rectangular weir having sides extended inward from the sides of the
channel by a distance of at least 2H max. The fluid stream must then contract as it flows around the
sides of the weir decreasing slightly the effective length of the weir.
The standard design requirements are:
The crest height above the bottom of the channel Hc ≥ 2Hmax
The minimum head above the crest Hmin > 0.2 ft or 60 mm
The maximum head above the crest Hmax < L/3
A cipolletti weir is also contracted from the sides of the stream by a distance at least 2H max and
has the sides that are sloped outward.
The standard design requirements are:
The crest height above the bottom of the channel Hc ≥ 2Hmax
The minimum head above the crest Hmin > 0.2 ft or 60 mm
The maximum head above the crest Hmax < L/3
Triangular weir is used primarily for low flow rates because the V-notch produces a larger head
than can be obtained with a rectangular notch. The angles vary from 35 o to 120o but angles 60o
and 90o are quite common.

Broad-crested weir
A broad-crested weir is shown in Figure 3.6.

V 2c
2g
 20

EGL

Y
yc
Fr1=0
h 2

1
Figure 3.6 Broad-Crested weir

It has sufficient elevation above the channel bottom to choke the flow, and it is long enough so
that the overflow streamlines become parallel, resulting in a hydrostatic pressure distribution.
Then at some position, say location 2, a critical flow condition exists.
Assume a rectangular horizontal channel and let the height of the weir be h. Location 1 is a point
upstream of the weir where the flow is relatively undisturbed, and Y is the vertical distance from
the top of the weir to the free surface at that point. Applying Bernoulli’s equation from location 1
to location 2 at the free surface and neglecting the kinetic energy head at location 1 (V 1=0), one
has the result
2
V c
h+Y =h+ y c + … … … … ..3.19
2g
Solving for Vc,
V c = √ 2 g ( Y − y c ) … … … … 3.20

For a weir whose breadth normal to the flow is b, the ideal discharge is
Q=b y c V c =b y c √ 2 g ( Y − y c ) … … .3 .21

Recognizing that Y=Ec, in Equation 3.11 is used to relate yc and Y, and when substituted into
equation 3.21, the result is

√
3
2 2
Q= g b Y 2 … … .3.22
3 3

For a properly rounded upstream edge on the weir, Equation 3.22 is accurate to within several
percent of the actual flow; hence a discharge coefficient is not applied.

Sharp-Crested Weir
A sharp-crested weir is a vertical plate placed normal to the flow containing a sharp-edge crest so
that the nappe will behave as a free jet. Figure 3.7 shows a rectangular weir with a horizontal
crest extending across the entire width of the channel. Because of the presence of the side walls,
lateral contractions are not present.

η v2 Nappe
Y

h
v1

1 2
 21

1 2

Figure 3.7 (a) Rectangular sharp-crested weir (Ideal flow;) Figure 3.7 (b) Actual flow

Let us define an idealized flow situation: The flow in the vertical place does not contract as it
passes over the crest so that the streamlines are parallel, atmospheric pressure is present in the
nappe, and uniform flow exists at location 1 with negligible kinetic energy (v 1=0). The Bernoulli
equation is applied along a representative streamline (Figure 3.7a) and solved for v 2, the local
velocity in the nappe:
v 2=√ 2 gη … … ..3.23
If b is the width of the crest normal to the flow, the ideal discharge is given as
Y Y 3
2
Q=b∫ v 2 dη=b ∫ √ 2 gη dη=b √ 2 g Y 2 ….3 .24
0 0 3
Experiments have shown that the magnitude of the exponent is nearly correct but that the
discharge coefficient Cd must be applied to accurately predict the real flow, shown in Figure 3.7
(b):
3
2
Q=C d √ 2 g b Y 2 … … … 3.25
3
The discharge coefficient accounts for the effect of contraction, velocity of approach, viscosity
and surface tension. An experimentally obtained formula for Cd has been given in the form

Y
C d=0.564+0.0846 … … … ..3.2 6
h

Normally, for small Y/h ratio, Cd=0.61. if the weir crest does not extend to the side walls but
allows for end contractions to appear, as in Figure 3.8a, the effective width of the weir can be
approximated by (b-0.2Y).
θ

(a) (b)
Figure 3.8 Contracted rectangular and V-notch weirs: (a) rectangular; (b) V-notch.
The V-notch weir Figure 3.8 b is more accurate than the rectangular weir for measurement of
low discharge.
 22

b/2

θ/2 z

b
2
=z tan
θ
2( )
Or
b=2 z tan ( θ2 )
δQideal=bδz √ 2 g ( Y −z )

( θ2 ) √2 g . z . √( Y −z ) δz
δQideal=2 tan

( ) √2 g .∫ z . √( Y −z ) δz
Q=2C d tan
θ
2 0

{ [ ] }
3 Y Y 3
Q=2C d tan ( )
θ
2
2 2
√2 g . − z ( H −z ) 2 + ∫ ( H−z ) 2 dz
3 0 3 0

{ [ ]}
5 Y

( )
Q=2C d tan
θ
2
√2 g
−2 2
× ( Y −z ) 2
3 5
0

()
5
8 θ
Q= C tan √2 g Y 2
15 d 2

In a manner similar to the development of the rectangular weir relation, the idealized discharge is
found by integrating the local velocity throughout the nappe above the crest. A discharge
coefficient is applied to yield

( )
5
8 θ
Q=C d √ 2 g tan Y 2 ….3 .27
15 2
For use with water, and for θ varying from 22.5 to 1200, experiments done by King and Brater,
0

1963 have shown that a value of Cd = 0.58 is acceptable for engineering calculations.

For 60o
5
Q=0.79Y 2
 23

For 90o
5
Q=1.37 Y 2

Francis’ formula in SI units is given below for fully-contracted notch:

3
2
Q=1.84 ( L−0.2 H ) H

Francis’ formula in SI units is given below for no end contractions

3
Q=1.84 ( L ) H 2

Further derivations

Y
Z1

1 2

Applying Bernoulli’s equation for the stream line

Pressure head at point two is zero because it is a jet and can be assumed to be atmospheric.
Cancel z1 then you can have.

v 21 v 22
z 1+ ( Y −z 1 ) + =z 2 + ( 0 )+
2g 2g
2 2
v1 v2
( Y −z 2) + 2 g = 2 g
Multiplying by 2g we have

v 2=√ ¿ ¿
Let us take an elementary element of dz so that area is bdz and dQ = bdzv

dQ =bdz √ ¿ ¿
To get Q with integrate
∫xn dx = xn+1/(n + 1) + C.
 24

Y Y 2 1
v1 2
Q=b∫ v 2 dz=b √ 2 g ∫ ( Y −z + ) dz
0 0 2g

[( ( ) ( ) )]
3 3
2 u21 2 u21 2
Q= b √ 2 g Y+ −
3 2g 2g
This is Q ideal that is no losses so there is need to introduce cd.
For small u1 we have
3
2
Q=C d √ 2 g b Y 2
3

Example 3.6
Determine the discharge of water over a rectangular sharp-crested weir, b = 1.25 m, Y = 0.35, h
= 1.47 m, with side walls and with end contractions. If a 90o V-notch weir were to replace the
rectangular weir, what would be the required Y for a similar discharge?
Solution
Cd = 0.63
Q = 0.48 m3/s
With end contractions Q = 0.45 m3/s
Y=0.66 m
Additional methods of flow measurement
Other types of weirs include those whose faces are inclined in the upstream and downstream
direction (triangular, trapezoidal, irregular). In addition, the spillway section of a dam can be
considered as a weir with a rounded crest.
A special type of open flume is one in which the geometry of the throat is constricted in such a
manner to choke the flow, thereby creating critical flow followed by a hydraulic jump. When
manufactured using a particular standardized section, the flume is called a Parshall Flume
(Figure 3.9). Extensive calibrations have established reliable empirical formulas to predict the
discharge. For throat widths from 0.3 to 2.4 m the discharge is given by the formula.
0.026
1.522B
Q=4 B H … … ….3 .28

Where H is the depth measured at the upstream location in Figure 3.9. Note that H and B are
measured in feet, and Q is in cubic feet per second.

2 ft 3 ft

1 1
Flow 5 6
B

L=0.5B+4ft

2L/3
 25

Gaging well for

measurement of depth H
Plan

2 ft.R
A

H
3 inch
9 inch

Elevation

Figure 3.9 Parshall Flume (After Henderson, 1966)

In a natural section of a river it may be impractical to place a weir; in that case stream gaging can
be used to measure the discharge. A control location is established upstream of the gaging site,
and for a given depth, or stage, of the river, the two-dimensional velocity profile is measured
using current meters. Subsequently, the profile is numerically integrated to yield the discharge. A
series of such measurements will produce a stage-discharge curve, which can then be employed
to estimate the discharge by measuring the river stage.
4.0 Momentum Concepts in Open-Channel Flow
So far we have been looking at how energy equation is applied to rapidly varied flow situations,
and in particular, how it is utilized to analyze flows in transition regions. The momentum
equation is also employed to study certain phenomena in those situations. When used in
conjunction with the energy and continuity relations, the momentum equation gives the user a
concise means of analyzing nearly all significant transition problems, including problems that
involve a hydraulic jump.

4.1 Momentum Equation

Consider the open-channel reach with supercritical flow upstream of a submerged obstacle, as
shown in Figure 4.1 (a). It represents a rapidly varied flow with an abrupt change in depth but no
change in width. In general, such a change may result from an obstacle in the flow or from a
hydraulic jump. The upstream flow is supercritical and the downstream flow is subcritical. Other
flow regimes may also be considered; for example, the conditions could be subcritical
throughout the entire control volume. Each situation should be approached as a unique
formulation.
2
Fr1>1

γA1ȳ1 γA2ȳ2
V1 V2

F
 1
26

(a) (b)

Figure 4.1 Channel flow over an obstacle: (a) Idealized flow; (b) control volume

The generalized flow situation may be used to develop the equation of motion for transition
regions. The control volume corresponding to Figure 4.1(a) is shown in Figure 4.1(b). The
pressure distribution is assumed hydrostatic, and the resultant hydrostatic forces are given by
γAȳ, where the distance ȳ to the centroid of the cross-sectional area is measured from the free
surface. The submerged obstacle imparts a force F on the control volume with a direction
opposite to the direction of flow. The linear momentum equation is applied to the control volume
in the x-direction to give

γA1ȳ1 – γA2ȳ2 – F= ρQ(V2-V1)..................4.1

The frictional forces have not been included in equation 4.1 because they are usually small
relative to the other terms, so they can be ignored. Similarly, gravitational forces in the flow
direction are negligible for the small channel slopes considered. Equation 4.1 can be rearranged
in the form:

F
M 1−M 2= … …..4. 2
γ
In which M1 and M2 are terms that contain the hydrostatic force and the momentum flux at
locations 1 and 2, respectively. The quantity M is called the momentum function, and for a
general prismatic section it is given by
Q2
M=A y+ …..4. 3
gA
For a rectangular section, Aȳ =by2/2, and the momentum function is

M=
b y2 b q 2
2
+
gy
=b(y 2 q2
+
2 gy )
… … 4.4

y
M2

Fr<1

Conjugate
depth points
yc Fr=1

M1
 27

Fr>1 M’2

Mc F/γ M

Figure 4.2 Variation of the momentum function with depth

Equation 4.4 is sketched in Figure 4.2. Two positive roots y exist for a given M and q; they are
termed either conjugate or sequent depths. The upper leg of the curve (y > y c) applies to
subcritical flow and the lower leg (y < y c) to supercritical flow. The values of M1 and M2 for the
example of Figure 4.1 (a) are shown, indicating supercritical flow upstream and subcritical flow
downstream of the submerged obstacle. The horizontal distance between M 1 and M2 is equal to
F/γ. The downstream flow would be supercritical if no hydraulic jump occurred; the
corresponding value of the momentum function is indicated by M’2.
The depth associated with a minimum M is found by differentiating M with respect to y in
Equation 4.3
2
dM BQ
=A− =0 … …..4. 5
dy gA
2

Note that d(Aȳ)/dy = A. The condition for minimum M is thus

2 3
Q B=g A … …..4. 6
This result is identical to Equation 3.14. Hence the condition of minimum M is equivalent to that
of minimum energy; critical flow with a Froude number equal to unity.

Momentum equation applied to a transition region

Quite often the momentum equation is applied in situations where it is desired to determine the
resultant force acting at a specific location, or to find the change in depth or velocity when there
is a significant undefined loss across the transition region. It is important to remember that the
energy and continuity equations are also at our disposal, and we must determine which relations
are necessary. In some instances, along with the continuity equation, both energy and momentum
equations must be applied. An example is given below to illustrate the technique.

Example 4.1
In a rectangular 5 m-wide channel, water is discharging at 14.0 m 3/s in Figure 4.3. Find the force
exerted on the sluice gate when y1=2 m and y2 = 0.5 m.

Fr1<1
Q Fr1>1

1 2

Figure 4.3 Example

Solution
 28

Using equation 4.3

Q2 14 2 3
M 1 = A 1 y 1+ =5 ×2 ×1+ =12.0 m
g A1 9.81× 5 ×2

2 2
Q 14
M 2 = A 2 y 2+ =5 ×0.5 × 0.25+ =8.62 m3
g A2 9.81×5 × 0.5

The resultant force acting on the fluid control volume is determined using equation 4.2 to be
F=γ ( M 1−M 2) =9800 ( 12.0−8.62 )=33100 N

Hence the force on the gate acts in the downstream sense with a magnitude of 33.1 kN.

4.2 Hydraulic jump

A hydraulic jump is a phenomenon in which fluid flowing at a supercritical state will abruptly
undergo a transition to a subcritical state. Boundary conditions upstream and downstream of the
jump will dictate its strength as well as its location. An idealized hydraulic jump is shown in
Figure 4.4.

EGL
hj
Fr1>1

V2 Fr1<1
V1
1 2

Figure 4.4 Idealized hydraulic jump

The strength of the jump varies widely, as shown in Table 4.1, with relatively mild disturbances
occurring at one extreme and significant separation and eddy formation taking place at the other.
As a result, the energy loss associated with the jump is considered to be unknown, so the energy
equation is not used in the initial analysis. Assuming no friction along the bottom and no
submerged obstacle, that is, setting F equal to zero, equation 4.2 shows that M 1 = M2. For
rectangular sections, Equation 4.4 can be substituted into the relation, allowing one to obtain.
q2 1 1
(
−
g y1 y2 2
1 2
)
= ( y 2 − y 1 ) … ….4 .7
2

Rearranging, factoring, and noting that Fr12=q2/(gy13) there results

2 1 y 2 y2
Fr 1 =
2 y 1 y1 ( )
+1 … … .4 .8

This equation is dimensionless and it relates the Froude number upstream of the jump with the
ratio of the downstream depths to upstream depths. It can be seen that equation 4.8 is quadratic
with respect to y2/y1 provided that Fr1 is known. Solving for y2/y1, one obtains.
 29

y2 1
= ( √ 1+ 8 Fr 1−1 ) … … 4.9
2
y1 2
The positive sign in front of the radical has been chosen to yield a physically meaningful
solution. It is worth noting that equation 4.9 is also valid if the subscripts on the depths and
Froude number are reversed:

y1 1
= ( √ 1+ 8 Fr 2−1 ) … … 4.10
2

y2 2
The theoretical energy loss associated with a hydraulic jump in a rectangular channel can be
determined once the depths and flows at locations 1 and 2 are known. The energy equation is
applied from 1 to 2, including the head loss h j across the jump. Combining that relation with
equation 4.7, and the continuity equation, one can show after some algebra that
3
( y 2− y 1 )
h j= … … … 4.11
4 y1 y2

Equations 4.8 to 4.11 are useful forms for solving most hydraulic jump problems in a rectangular
channel.
Table 4.1 shows the various forms that a jump may assume relative to the upstream Froude
number. A steady, well-established jump, with 4.5 < Fr < 9.0, is often used as an energy
dissipater downstream of a dam or spillway. It is characterized by the existence of breaking
waves and rollers accompanied by a submerged jet with significant turbulence and dissipation of
energy in the main body of the jump; downstream, the water surface is relatively smooth. For
Froude numbers outside the range 4.5 to 9.0, less desirable jumps exist and may create
undesirable downstream surface waves. The length of a jump is the distance from the front face
to just downstream where smooth water exists; a steady jump has an approximate length of six
times the upstream depth.

Example 4.2
A hydraulic jump is situated in a 4-m wide rectangular channel. The discharge in the channel is
7.5m3/s, and the depth upstream of the jump is 0.20m. Determine the depth downstream of the
jump, the upstream and downstream Froude numbers, and the rate of energy dissipated by the
jump.
Answers
q=1.88m2/s, Fr1 = 6.71,
y2 from equation 4.9 = 1.80m, Fr2 = 0.25
Head loss from equation 4.11 is 2.84m
Hence the rate of energy dissipation is γQhj= 9800 x 7.5 x 2.84 = 209 kW.

Table 4.1 Hydraulic jumps Horizontal Rectangular Channels

Upstream Type Description Shape

Fr
 30

1.0-1.7 Undular Ruffled or undular water

surface; surface rollers
form near Fr = 1.7

1.7-2.5 Weak Prevailing smooth flow;

low energy loss

2.5-4.5 Oscillating Intermittent jets from Oscillating jet

bottom to surface, causing Roller
persistent downstream
waves

4.5-9.0 Steady Stable and well-balanced;

energy dissipation
contained in main body of
jump
>9.0 Strong Effective, but with rough,
wavy surface downstream

Source: Adapted from Chow, 1959

Translating hydraulic jump

ω
V2 V2 + ω
V1 V1 + ω
1 2 1 2
(a) (b)
Figure 4.5 Translating hydraulic jump: (a) front moving upstream; (b) front appears stationary by
superposition.

A translating hydraulic jump, alternately termed a surge or a hydraulic bore, is shown in Figure
4.5 (a); it is termed a positive surge wave in the sense that it maintains a stable front as it
propagates at speed ω into an undisturbed region. Such a wave can be generated by abruptly
lowering a downstream gate or by rapidly releasing water at an upstream location into a channel.
This unsteady rapidly varied flow situation can be conveniently analyzed as a steady-state
problem by superposing the bore speed ω in the opposite sense on the control volume Figure 4.5
(b). The front appears stationary and the relative velocities at locations 1 and 2 are equal to V 1 +
ω and V2 + ω, respectively.
Assume a bore translating in a rectangular, frictionless, horizontal channel equation 4.9 can be
applied by substituting V1 +ω for V1 in the definition of Fr1 to obtain.
 31

[√ ]
2
y2 1 ( V 1 +ω )
= 1+8 −1 … … … 4.12
y1 2 g y1

The continuity relation applied to the control volume in Figure 4.5 (b) is
y 1 ( V 1+ ω ) = y 2 ( V 2 + ω ) … … .4 .13
Equations 4.12 and 4.13 contain five parameters: y 1, y2, V1, V2 and ω. Three of them must be
known to solve for the remaining two. Depending on which variables are known, the solution of
Equations 4.12 and 4.13 will be either explicit or based on a trial procedure.

Example 4.3
Water flows in a rectangular channel with a velocity of 2.5m/s and a depth of 1.5m. A gate is
suddenly completely closed, forming a surge that travels upstream. Find the speed of the surge
and the depth behind the surge.
Solution
Since the gate is closed, the downstream velocity is V 2=0. Equations 4.12 and 4.13 contain the
two unknowns ω and y2. Combining them to eliminate y2 and substituting V1 = 2.5 m/s and
y1=1.5 m results in the relation

[√ ]
2
ω ( 2.5+ ω )
2.5+ω= 1+ 8 −1
2 9.81 ×1.5
Solving yields ω=3.41m/s and using equation 4.13 to compute y2

V 1 +ω 2.5+3.41
y 2= y 1 =1.5 × =2.60 m
V 2 +ω 3.41

Drag on submerged objects

If an object is submerged in the flow, it is possible to describe the drag force in the manner.
V2
F=C D Aρ … … 4.14
2
In which CD is the drag coefficient and A is the projected area normal to the flow. In open
channel flow, since a free surface is present, the drag coefficient must be modified to account for
wave drag as well as drag due to friction and separation. In practice equation 4.14 is seldom
employed to analyze or design a stilling basin, since other factors such as additional
appurtenances, approach velocity, scour and cavitation must be considered. Instead, design
standards have been established based on observations of existing basins and laboratory model
studies. Examples of submerged objects in open channel flow include pipelines and bridge piers;
in these situations, a hydraulic jump may not be present.

b Q

y1
 32

y2

2
1 h

Figure 4.6 Stilling Basin with baffle blocks

A design application is illustrated in Figure 4.6. Baffle blocks are devices placed in a reach of
channel known as a stilling basin to stabilize the location of a hydraulic jump and to aid in the
dissipation of flow energy. Typically they are used for Fr 1 greater than 4.5. When water from
elevated water bodies like reservoirs flows over spillways, through outlet works, or down chutes,
it can have tremendous energy. This energy can cause great damage to the structure itself, to the
ground supporting structure (and therefore ultimately to the structure), and to the downstream
channels. The hydraulic jump is one of the most effective methods of dissipating such flow
energy. But hydraulic jumps themselves are can also be very destructive. The objective of a
stilling basin is to initiate a jump for energy dissipation and to contain it within a structure that
will minimize damage. The basin floor may be recessed or have an adverse slope to provide
sufficient tail-water depth. To increase the turbulence and to encourage the jump to form
immediately as the water enters the basin, rows of chute blocks and or staggered baffle blocks
are commonly arranged across the basin. Sometimes those blocks with an end sill can reduce the
length of the jump, and so the basin.
Example 4.4
In the flow example presented in example 4.2, a series of baffle blocks is placed in the channel as
shown in Figure 4.7. Laboratory experimentation has shown that the arrangement has an
effective drag coefficient of 0.25, provided that the blocks are submerged in the flow. If the
blocks are 0.15 m high, and if the discharge and upstream depth remain the same, determine the
depth downstream of the jump and the rate of energy dissipated by the jump.

b Q

y1
y2

2
1 h

Figure 4.7 Example 4.4

Solution
It is necessary to use equation 4.2 since obstacles (i.e., the baffle blocks) are placed within the
control volume. The upstream velocity is
V1 = Q/A1= 7.5/4 x 0.2 = 9.38 m/s
The force F due to the presence of the baffle blocks is computed from equation 4.14
F=CDAρV12/2 = 0.25 x (4 X 0.15) x 1000 x 9.382/2=6600N.
 33

Note that the frontal area is the width of the channel multiplied by the height of the blocks.
Substituting known conditions into equation 4.2, making use of equation 4.4 which defines M for
a rectangular channel, and noting that q = 7.5/4 = 1.88 m2/s, we find

b
(
y 12 q 2
+
2 g y1
−b
) (
y 22 q2
+
2 g y2
=
F
γ )
( ) ( )
2 2 22
0.2 1.88 y 1.88 6600
4 + −4 2 + =
2 9.81 ×0.2 2 9.81 y2 9800
The relation reduces to
2 0.721
y 2+ =3.31
y2

The solution for y2 is approximately 1.70 m. The change in specific energy between locations 1
and 2 is

( )
2 2
q q
E1−E 2= y 1 + 2
− y 2 + 2
2 g y1 2 g y2

( )
2 2
1.88 1.88
¿ 0.2+ 2
− 1.70+ 2
=2.94 m
2× 9.81× 0.2 2 ×9.81 ×1.70

The rate of energy dissipation, therefore is

γQ ( E1−E2 ) =9800 ×7.5 ×2.94=216 kW

4.3 Numerical solution of the momentum equation

For non rectangular channels the momentum relation can be used directly to analyze the
hydraulic jump or other problems requiring the momentum equation; the technique is
demonstrated as follows. Consider a trapezoidal channel with conditions known at location 1
upstream of the jump. Consequently, M1 and F are evaluated as constants and Equation 4.2 can
be written in the form
F
M 2−M 1+ =0 … .4 .15
γ
Where M2 is a function of y2. Introducing the trapezoidal geometry at location 2, with m 1=m2=m,
the relation above can be written as
2
y2 Q
2
F
6
( 2 m y 2+ 3 b ) + −M 1 + =0 … … 4.16
γ
g ( b y 2 +m y 2 )
2

This can be solved for y2 by a suitable numerical technique such as interval halving, false
position, or Newton’s method found in any book of numerical maths. Note that by setting F = 0,
it becomes the relation for finding the conditions downstream of a hydraulic jump, and that by
additionally letting m = 0, a rectangular hydraulic jump problem can be solved.

Example 4.5
 34

A hydraulic jump occurs in a triangular channel with m 1 = m2 = 2.5 m. The discharge is 20 m3/s
and yc = 1.67 m. Upstream of the jump the following parameters are given: y 1 =0.75 m, Fr1=
7.42, and M1 = 29.35 m3. Determine the conjugate depth y2 downstream of the jump.
Solution
Use equation 4.16 with F=0:
2 2
y2 20
6
( 2× 2.5 y 2) + 2
−29.35=0
9.81 ×2.5 y 2

The relation reduces to

319.58
f ( y 2) = y 2 + −35.23=0
y 22
The result after nine iteration is y2=3.22 m rounded off to three significant figures.
One can use MATLAB and the solution is done as follows
clear
>> %input side slope, bottom width, discharge, critical depth,
>> %upstream depth, and gravitational constant
>> m=2.5;
>> b=0;
>> Q=20;
>> yc=1.67;
>> y1=0.75;
>> g=9.81;
>> %set approximated upper and lower limits
>> y=yc:.001:3*yc;
>> %reducing equation 4.16 with F=0 yields
>> f=y.^3+19.58./y.^2-35.23;
>> %Find value of y where f crosses zero
>> [s,t]=min(abs(f));
>> y2=y(t)

y2 =

3.2190

5.0 Non-uniform gradually varied flow

The evaluation of many open-channel flow situations must include accurate analyses of
relatively long reaches where the depth and velocity may vary but do not exhibit rapid or sudden
changes. In rapidly varied flow, losses may be ignored without severe consequences, whereas for
gradually varied flow, it is necessary to include losses due to shear stress distributed along the
channel length. The shear stress is the primary mechanism that resists the flow.
Gradually varied flow is a type of steady, non-uniform flow in which y and V do not exhibit
sudden or rapid changes, but instead vary so gradually that the water surface can be considered
 35

continuous. As a result, it is possible to develop a differential equation that can describe the
incremental variation of y with respect to x, the distance along the channel. An analysis of this
relation will enable one to predict the various trends that the water surface profile may assume
based on the channel geometry, magnitude of the discharge, and the known boundary conditions.
Numerical evaluation of the same equation will provide engineering design criteria.

5.1 Differential equation for gradually varied flow

For channels of regular cross-section it is possible to derive analytically an expression for the
variation of depth from point to point and so to determine, theoretically, the profile of the free
surface if it is assumed that:
 The channel is rectangular, straight and of constant roughness.
 The bed slope is small, so that the depth measured normal to the bed can be assumed
equal to the vertical depth.
 Flow is steady and streamlines are approximately parallel, so that pressure distribution is
hydrostatic.
A representative non-uniform gradually varied flow is shown in Figure 5.1. Over the incremental
distance Δx, the depth and velocity are known to change slowly. The slope of the energy grade
line is designated as S. In contrast to uniform flow, the slopes of the energy grade line, water
surface, and channel bottom are no longer parallel. Since the changes in y and V are gradual, the
energy loss over the incremental length Δx can be represented by the Chezy-Manning equation.
Assumptions include regular cross section, small channel slope, hydrostatic pressure distribution,
one-dimensional flow.

2 hL =SΔx
V1 S
1 EGL
2g
V 22
2g
y1 V1
y2
V2
S0
1
z1
z2
Δx

Datum

Figure 5.1 Non-Uniform gradually varied flow

Applying energy equation from location 1 to location 2, with the loss term h L given by SΔx. If
the total energy at location 2 is expressed as the energy at location 1 plus the incremental change
in energy over the distance Δx we find that

dH
H 1=H 2 + S ∆ x=H 1 + ∆ x + S ∆ x … … ..5.1
dx
Substitute H=y+z+V2/2g and dz/dx = -S0 into this relation and rearrange to find
 36

( )
2
−d V
S−S 0= y+ … … 5.2
dx 2g
The right-hand term is –dE/dx, and it is transformed to
dE dE dy 2 dy
= =( 1−Fr ) … … 5.3
dx dy dx dx
Recall that dE/dy=1-Q2B/gA3 = 1-Fr2

Finally, upon substitution into the energy relation and solving for the slope of the water surface,
dy/dx, one finds that

dy S0−S
= … …..5. 4
dx 1−Fr2
This is the differential equation for gradually varied flow and is valid for any regular channel
shape.

Or Total energy
SΔx
S

Δy
y
V
y V+ΔV
S0 Δx S0

Δx

2
v2 ( v+ ∆ v )
S0 ∆ x + y + = y + ∆ y+ +S ∆ x
2g 2g

Ignoring products of small quantities

 37

v ∆v
S0 ∆ x=∆ y+ +S ∆x
g
So that

∆y v ∆v
=S 0−S−
∆x g ∆x

Assuming a constant width of channel, for continuity of flow the discharge per unit width is
constant from section to section
vy=( v+ ∆ v ) ( y + ∆ y)
vy=vy+ v ∆ y + y ∆ v + second terms
∆ v=−v ∆ y / y
Therefore
∆y v v∆ y
=S 0−S+
∆x g y∆x

( )
2
∆y v
1− =S0−S
∆x gy
Therefore
∆y S 0−S
=
( )
∆x v2
1−
gy

Example 5.1
Using an appropriate control volume Figure 5.2 for gradually varied flow, show that the slope S
of the energy grade line is equivalent to τ0/γR

S
EGL
1

γAȳ
γAΔx
d (γA y ) ∆ x
γA y +
dx
τ0 PΔx θ
Δx
x

Figure 5.2 Example

Solution
 38

The control volume is shown in Figure 5.2. The resultant force acting on the control volume is
due to the incremental change in hydrostatic pressure (γd(Aȳ)/dx)Δx, the component of weight in
the x-direction γAsinθΔx, and the resistance term τ0PΔx. Using the momentum equation
∑ F x =ṁ ( V 2 x −V 1 x )
With V2x – V1x = (dV/dx)Δx results in

d dV
−γ ( A y ) ∆ x + γAsinθ ∆ x−τ 0 P ∆ x=ρVA ∆x
dx dx

This relation can be simplified by noting that

d ( A y) d ( A y ) dy dy
= =A
dx dy dx dx
And P = A/R. Substitute and divide the equation by γAΔx, the weight of the control volume, and
find that

dy dV
−γA ∆x ρVA ∆x
dx γAsinθ ∆ x τ P ∆ x dx
+ − 0 =
γA ∆ x γA ∆ x γA ∆ x γA ∆ x
−dy τ 0 V dV
+ sinθ− =
dx γR g dx
Since sinθ = S0 for small θ, the equation above can be rearranged in the form
τ0
γR
−S 0=
−dy V dV −d
dx
−
g dx dx
= ( y+ )
V2
2g
Upon comparison with equation 5.2, it is seen that the right hand side is equivalent to S-S 0, and
consequently
τ0
−S 0=S−S 0
γR
Or
τ0
S=
γR

5.2 Water Surface Profiles

It is possible to identify a series of water surface profiles based on an evaluation (Bakhmeteff,
1932) of equation 5.4. Essential to the development is the determination of normal and critical
depths. Note that y0 and yc are uniquely determined once the channel properties and discharge are
established. Table 5.1 shows the classification of the water surface profiles. Associated with y c is
a critical slope Sc which is found by substituting yc into the Chezy-Manning equation and solving

( )
2 1
q 3
for the slope as in the equation; y c = . The channel slope can be determined as:
g
Mild slope S0 < Sc and y0 > yc.
Steep slope is where S0 > Sc and y0 < yc.
Critical is where S0 = Sc and yn = yc.
A horizontal slope exists when S0 = 0 and
An adverse slope occurs when S0 < 0 that is uphill.
 39

There are five categories of slope and three categories of depth resulting in 15 possible profiles.
Inspection of Table 5.1 shows that there are 12 possible profiles. For horizontal and adverse
slopes, uniform flow is impossible, thus eliminating the H 1 and A1 curves; while for critical
slope, critical depth and normal depth coincide, thus eliminating the C 2 curve. Profiles that get
deeper downstream are called backwater curves, and those that decrease in depth downstream are
called drawdown curves.

Each profile is classified by a letter/number combination. The letter refers to the channel slope,
M for mild, S for steep, C for critical, H for horizontal, and A for adverse. The numerical
subscript designates the range of y relative to y 0 and yc. Flow can occur at depths above or below
yc and at depths above or below y0.

The number refers to the relation between the actual depth of flow y, the normal depth at uniform
flow y0 and the critical depth yc as follows:
1. Free surface of stream lies above both normal and critical depth lines.
2. Free surface of stream lies between normal and critical depth lines.
3. Free surface of stream lies below both normal and critical depth lines.
 40

The variation of y with respect to x for each profile in Table 5.1 can now be found. For a given
Q, n, S0, and channel geometry, the analysis reduces to the determination of how S and Fr vary
with y. An inspection of the Chezy-Manning equation will reveal that S decreases with
increasing y; similarly, the Froude number decreases as y increases. The numerator in equation
5.4 will assume the following inequalities: (S 0-S) > 0 for y > y 0, and (S0-S) < 0 for y<y0. In
addition, the denominator varies in the manner (1-Fr2) > 0 for y > yc and (1-Fr2) < 0 for y < yc.
With these criteria, the sign of dy/dx can be evaluated. In addition, with equation 5.2, the sign of
dE/dx is revealed.

The boundaries of the profiles can be established as follows.

As y tends to y0, S tends to S0. Hence dy/dx approaches zero, or in other words, the water surface
approaches y0 asymptotically. This applies to curves M1, M2, S2 and S3.
As y becomes large, the velocity becomes small and S and Fr tend to zero, so that dy/dx
approaches S0. Hence the surface approaches a horizontal asymptote; curves M1, S1, and C1 are of
this type.
As y approaches yc, dy/dx becomes infinite, a limit that is never reached. For supercritical flow,
as the M3, H3 and A3 curves approach yc, a hydraulic jump will form and create a discontinuity in
the water surface; at the beginning of the curve, rapid acceleration occurs with nonparallel
streamlines. When the flow is subcritical (M2, H2, A2), rapid drawdown takes place close to yc,
and the streamlines are no longer parallel. For all of these equation 5.4 is invalid, since the flow
is no longer one-dimensional.

For the profiles shown in Table 5.1, there is no physical significance to the theoretical limit of y
approaching zero, since a finite depth is necessary for the existence of flow.
 41

Table 5.1 Classification of surface profiles

Channel Profile Depth range Fr dy dE Shape
slope type
dx dx y0 M1 Horizontal
Mild M1 y > y0 > yc <1 >0 >0 asymptote
S0 < Sc M2 y0 > y > yc <1 <0 <0
M2
y0 > yc M3 y0 > yc > y >1 >0 <0 yc

M3

Steep S1 y > yc > y0 <1 >0 >0

yc
S0 > Sc S2 yc > y > y0 >1 <0 >0 S1
y0 < yc S3 yc > y0 > y >1 >0 <0 y0 S2

S3

Critical C1 y > yc or y0 <1 >0 >0

S0 = Sc C3 yc or y0 > y >1 >0 <0
y0 = yc C1
y0 = yc

C3

Horizontal H2 y > yc <1 <0 <0

S0 = 0 H3 yc > y >1 >0 <0 H2
y0 → ∞ yc

H3

Adverse A2 y > yc <1 <0 <0

S0 < 0 A3 yc > y >1 >0 <0
y0 A2
undefined

yc
A3
 42

Example 5.2
 43

By assuming a wide rectangular channel, develop the right-hand side of equation 5.4 to show
that dy/dx varies with y.
Solution
For a wide rectangular channel, assume that b>>y, so that wetted perimeter is approximated by P
≈ b. The hydraulic radius then becomes R = A/P ≈ (by)/b=y. Noting that Q = qb, the Chezy-
Manning equation, used to evaluate S, simplifies to

( qbn )2 ( qn )2
S= =
(b y )
5 2 10
3
y3
It is assumed that in the Chezy-Manning equation c1 = 1. For a rectangular section the square of
the Froude number is
2
2 q
Fr = 3
gy
Substituting into equation 5.4 gives
( qn )2
S 0− 10
dy y3
= 2
dx q
1−
( gy )3

Since (qn)2y0-10/3=S0 and Frc2 = q2/(gyc3)=1, the relation can be written as

dy
=S 0
1− ( )
y 0 103
y

( )
dx yc
3

1−
y
This equation can be used as an alternative to Equation 5.4 to evaluate the water surface profiles
shown in Table 5.1

5.3 Controls and critical Flow

The existence of the various profiles shown in Table 5.1 depend on the boundary conditions that
are specified at given locations in the channel. Quite often, a control will define the boundary
condition. A control exists when a depth-discharge relationship can be established at a section.
The manner in which the control section affects the water surface away from its specific location
can be studied by examining flow near the critical state in a rectangular channel.
Equations 4.12 is the expression for a surge of finite magnitude translating in the upstream
direction in a rectangular channel. As y 2 approaches y1, the magnitude of the surge becomes
infinitesimal; these waves may be generated by the presence of control structures and other
transition sections that tend to “disturb” the flow. In Equation 4.12, one can replace y 1 and y2 by
y, and V1 by V. In addition, let the wave travel in the downstream direction, so that the equation
becomes
 44

[√ ]
2
1 8 ( V −ω )
1= 1+ −1 …..5. 5
2 gy
Solving for ω there results
ω=V ± √ gy=V ± c … …5.6
In which c= √ gy , termed the celerity. The celerity is the speed at which an infinitesimal wave
will travel into an undisturbed region, that is, a region with zero velocity. For a nonrectangular

√ A
channel, c= g . Thus if a disturbance is created at a midstream location in a channel, two
B
infinitesimal waves are generated. One wave front will tend to propagate upstream at the speed
V-c and the second will move downstream at the speed V + c. These observations are made
relative to an observer in a fixed position, that is, to one who is standing on the shore observing
the wave motion.
V
In a rectangular channel with critical flow conditions, Fr= =1 or V = √ gy=c.
√ gy
Hence at critical flow, the first wave front generated by the disturbance would not move
upstream but would appear stationary and become a so-called “standing wave”. The opposite
front would be swept downstream at the speed 2c. If Fr < 1, the first wave would travel
upstream, and the opposite wave would travel downstream at a speed less than 2c. For Fr> 1 in
the channel, since V > c, both waves are swept downstream. Thus, since it contains a mechanism
that disturbs the flow, a control will influence upstream conditions only when the flow is
subcritical (Fr < 1). Similarly, when the flow is supercritical ( Fr > 1), the control can influence
only the downstream flow conditions. This is illustrated in Figure 5.3 a, where a sluice gate is
positioned in a channel with subcritical flow upstream and supercritical flow downstream. An M1
profile is generated above the gate and an M3 profile exists below it. Any movement of the gate
would influence the nature of the two profiles; lowering the gate would lengthen their range, and
raising it would produce the opposite effect.

M1
M3

y0,1 M2
S2
Fr < 1 yc y0,2
Fr > 1 Fr < 1
Fr > 1
S0,1
S0,2
(a) (b)

S2
M2
yc
yo
Fr > 1 y0
Fr < 1 yc

≈ 3.5yc
(c) (d)
Figure 5.3 Representative controls: (a) sluice gate; (b) change in slope from mild (S 01) to steep
(S02); (c) entrance to a steep channel; (d) free outfall
 45

The profiles shown in Table 5.1 are all influenced by the presence of controls. In Figure 5.3,
several controls are shown which create a variety of profiles. Critical depth often is associated
with an effective control. Examples include sluice gates, weirs, dams, and flumes, all of which
force critical flow to occur somewhere in the transition region. In addition, a control can be
located at a break in channel slope from mild upstream to steep downstream (Figure 5.3b). The
entrance to a steep channel ( Figure 5.3c) is an example of an upstream control, with critical flow
occurring at the crest. Critical flow will occur a short distance upstream from a free outfall on a
mild slope (Figure 5.3d). Other controls, not shown in Figure 5.3, are a channel constriction
acting as choke, and for a mild slope, the existence of uniform flow at some location.

5.4 Profile synthesis

Identification of controls and their interaction with possible profiles is a requirement for
successful understanding and correct design and analysis of open-channel flow. Since controls
are essentially transition sections, the rapidly varied flow principles presented in sections 3.0 and
4.0 can be used to determine the necessary depth-discharge relations. Once the controls have
been indentified, the profiles can be selected and the range of influence of the controls
established.
As an example, consider the situation shown in Figure 5.4. Flow enters the steep channel from a
reservoir, so that critical flow exists as the channel entrance.
Both the magnitude of the discharge and the S 2 profile are influenced by the depth at the
entrance; therefore, the depth acts as a control. At the downstream end of the channel, the lower
reservoir acts as a control to establish an S 1 profile that projects upstream. At some interior
location, a hydraulic jump occurs to allow the flow to pass from a supercritical to a subcritical
state. The location of the jump can be found by plotting a curve of depth conjugate to the S 2
profile and finding its point of intersection with the S 1 curve. A lowering of the lower reservoir
elevation would cause the jump to move downstream and ultimately be swept out of the channel.
Increasing the lower reservoir elevation would move the jump upstream; if it were to move into
the entrance region, the upstream control would no longer exist.

Conjugate depth
curve S1
S2

yc
y0

Figure 5.4 example of profile synthesis

Example 5.3
In a rectangular channel, b= 3m, n=0.015, S0 = 0.0005, and Q = 5m3/s. At the entrance to the
channel, flow issues from a sluice gate at a depth of 0.15 m. The channel is sufficiently long that
uniform flow conditions are established away from the entrance region. Figure 5.5a. Find the
nature of the water surface profile in the vicinity of the entrance.
 46

5 m3/s y0
0.15 m

(a)
M3 Profile

y0 = 1.39 m yc = 0.66 m

y1 = 0.25 m
(b)

Figure 5.5 Example

Solution
First find y0 and yc to determine the type of channel. To find y0, follow the method. Substitute
known data into the Chezy- Manning equation:
5
(3 y0)3 0.015 ×5
= =3.354
2
√ 0.0005
( 3+2 y 0 ) 3

Solving gives y0 = 1.39 m, next, the critical depth is computed to be

( ) ]
[
2 1

yc=
2 1
q 3 ()
=
5 3
3
=0.66 m
g 9.81

Since y0 > yc, a mild slope condition exists. The gate is a control and there will be an M3 profile
beginning at the entrance, terminated by a hydraulic jump. Downstream of the jump, the
condition of uniform flow acts as a control, so at that location the depth y0 and the Froude
number is
5
q 3
Fr 0= = =0.325
√ g y 0 √ 9.81× 1.393
3

The depth before the jump is

y0 1.39 (
y 1= ( √ 1+8 Fr 0−1 )= √ 1+8 ×0.325 −1 )=0.25
2 2
2 2

The depths yc and y1 have been calculated to two significant figures, since the Manning
coefficient is known to only two significant figures.

Design of open channels in the field

 47

The catchment concerned in terms of area. If it is near measure the area and if it is far use the
topographical maps.
Determine the inlet and outlet. Find the longest route where water follows.
Determine the geological characteristics of the catchment.
Determine the amount of flow arising from rain. Use the highest daily rainfall amount for 50%
chance of occurring.

Computer applications in Hydraulics

It is essential for students to fully understand the underlying physical and mathematical concepts
when doing hydraulic computations. This reduces the chance for errors in any hydraulic design
or analysis problem. Once the underlying concepts are understood, however, the solution
processes can become repetitive and tedious, even with computational aids such as nomographs.
Repetitive computational problems are very well suited to computer analysis. The use of
commercial software packages avoids the need to program the calculator and generally offers a
more extensive set of options for doing the calculations.
There are several advantages of using computerized solutions for performing common hydraulic
calculations. The amount of time to do the calculations is greatly reduced. As long as there are
on “bugs” in the program, the solution process will be less prone to algebraic and unit conversion
errors. Computer solutions are readily documented and reproducible.
Using computers to perform engineering calculations does not relieve the user of the software
from liability for the accuracy of the results. In other words, an engineer or technician who uses
software with bugs and produces an erroneous analysis or poor-quality design cannot blame the
software for the mistakes. The engineer is still legally responsible for the mistakes. Before using
any software for the first time, it is best to test it out first by comparing the computer solutions to
“hand” calculated solutions for a few typical problems.
One example of a powerful, easy-to-use software program for hydraulic calculations is called
FlowMaster(R) others include EPANET, WATERCAD and so on.