IRRIGATION AND DRAINAGE

ENGINEERING

LECTURE IV:

DESIGN OF CONVEYANCE SYSTEM

By: Ir. NAHAYO Déogratias

Lecturer, Civil Engineering Department
Fundamental Equations
 To develop equations describing fluid flow, we will
work from relevant fundamental physical laws.

The Law of Conservation of Matter

 Matter cannot be created nor destroyed (except in a
nuclear reaction), but may be transformed by chemical
reaction. In fluids we neglect chemical reactions and
so we deal with the conservation of mass.
Fundamental Equations
(Cont’d)
 The Law of Conservation of Energy
 Energy cannot be created nor destroyed, it can only be
transformed from one form to another.
 For example, the potential energy of water in a dam is
transformed to kinetic energy of water in a pipe.
 Though we will later talk of „energy losses‟, this is a
misnomer as none is actually lost but transformed to
heat and other forms.
Fundamental Equations
(Cont’d)
The Law of Conservation of Momentum
 A body in motion remains in motion unless some
external force acts upon it.
 This is Newton‟s Second Law:

To apply these laws to fluids

poses a problem, since fluid is a
continuum, unlike rigid
bodies.
Hence we use the idea of a
„control volume‟.
Fundamental Equations
(Cont’d)
Control Volume
 A control volume is an imaginary region within a body
of flowing fluid, usually at fixed location and of a fixed
size:
Fundamental Equations
(Cont’d)
 It can be of any size and shape so we
choose shapes amenable to simple
calculations.
 Inside the region all forces cancel out,
and we can concentrate on external
forces.
 It can be picture as a transparent pipe or
tube, for example.
Fundamental Equations
(Cont’d)
The Continuity Equation
 Applying the Law of Conservation of Mass to a control
volume, we see:

For steady incompressible flow, the rate of mass increase

is zero and the density of the fluid does not change.
Hence:
Fundamental Equations
(Cont’d)
The rate of mass change can be
expressed as:
Fundamental Equations
(Cont’d)
Fundamental Equations
(Cont’d)
A typical application of mass conservation is at pipe junctions:

If we consider
inflow to be positive
and outflow
negative, we have:
Fundamental Equations
(Cont’d)
 The Energy Equation
 We apply the Law of Conservation of Energy to a control
volume.
 To do so, we must identify the forms of energy in the control
volume. Consider the following system:
Fundamental Equations
(Cont’d)
The forms of energy in this system are:
 Pressure energy: The pressure in a fluid also does
work by generating force on a cross section which then
moves through a distance. This is energy since work is
energy.
 Kinetic energy: This is due to the motion of the mass of
fluid.
 Potential energy: This is due to the height above an
arbitrary datum.
Fundamental Equations
(Cont’d)
Bernoulli’s Equation
Comments
 From Bernoulli‟s Equation we note several important
aspects:
1. It is assumed that there is no energy taken from or
given to the fluid between the entry and exit. This
implies the fluid is frictionless as friction generates
heat energy which would be an energy loss. It also
implies that there is no energy added, say by a pump
for example.
Fundamental Equations
(Cont’d)
2. Each term of the equation has dimensions of length, L,
and units of metres. Therefore each term is known as a
head:
Fundamental Equations
(Cont’d)
3. The streamtube must have very small dimensions
compared to the heights above the datum. Otherwise
the height to the top of a cross-section would be
different to the height to the bottom of a cross-section.
Therefore, Bernoulli‟s Equation strictly only applies to
streamlines.
 We have derived the equation from energy
considerations.
 It can also be derived by force considerations upon an
elemental piece of fluid.
Application of Bernoulli’s Equation

Flow Measurement – Small Orifices

 Consider the following tank discharge through a small
opening below its surface:
If the head is practically
constant across the diameter
of the orifice ( h > d ) then,
using the energy equation:
Application of Bernoulli’s Equation (Cont’d)

With both pressures atmospheric and taking v1 = 0, we

have:

And so the velocity through the orifice is:

This is Torricelli‟s Theorem and represents the theoretical

velocity through the orifice.
Measured velocities never quite match this theoretical
velocity and so we introduce a coefficient of velocity, C v , to
get:
Application of Bernoulli’s Equation (Cont’d)

 Also, due to viscosity the area of the jet may not be the
same as that of the orifice and so we introduce a
coefficient of contraction, Cc :

Lastly, the discharge through the orifice is then:

Application of Bernoulli’s Equation (Cont’d)

 In which Cd is the coefficient of

discharge and is equal to Cc Cv .
 For some typical orifices and
mouthpieces values of the coefficient
are:
Application of Bernoulli’s Equation (Cont’d)

Flow Measurement – Large Orifices

 When studying small orifices we assumed that the head
was effectively constant across the orifice.
 With large openings this assumption is not valid.
Consider the following opening:
Application of Bernoulli’s Equation (Cont’d)

 To proceed, we consider the infinitesimal

rectangular strip of area b ⋅ dh at depth h.
 The velocity through this area is (2gh)^0.5
and the infinitesimal discharge through it
is:
Thus the total discharge
through the opening is the
sum of all such
infinitesimal discharges:
Application of Bernoulli’s Equation (Cont’d)

 Large openings are common in civil engineering

hydraulics, for example in weirs.
 But in such cases the fluid has a velocity (Va )
approaching the large orifice:

Using the energy equation:

Hence:
Application of Bernoulli’s Equation (Cont’d)

 In which each term in the brackets is a head.

 Given the velocity we can find the discharge through
the strip to be:

And so the total discharge is:

 Application of Bernoulli’s Equation (Cont’d)

Discharge Measurement in Pipelines

 We consider two kinds of meters based on constricting
the flow: the Venturimeter and the Orifice meter, as
shown. ( Show the Venturimeter formula)
Application of Bernoulli’s Equation (Cont’d)

Accounting for Energy Losses

 Consider the following reservoir and pipe system:

The energy equation gives us:

Application of Bernoulli’s Equation (Cont’d)

 Taking there to be zero velocity everywhere, we can

draw this total head on the diagram:

Hence at each point we have an exchange between

pressure head and static head:
Application of Bernoulli’s Equation (Cont’d)

 If we introduce the effect of velocity into the diagram

we know that the pressure must fall by an amount

Since we now have

Application of Bernoulli’s Equation (Cont’d)

 The hydraulic grade line is the line showing the

pressure and static heads only.
 If the velocity varies over the length of the pipe due to
changes in diameter, say, we now have:
 Application of Bernoulli’s Equation (Cont’d)

 Note that the hydraulic grade line rises at the larger pipe
section since the velocity is less in the larger pipe (Q = AV
).
 If we now consider energy to be lost at every point along
the length of the pipe, the total head will reduce linearly:
Application of Bernoulli’s Equation (Cont’d)

 Thus denoting f h as the friction head loss,

we modify the energy equation to take
account of friction losses between two
points:
Flow in pipes
Flow in pipes (Cont’d)
Flow in pipes (Cont’d)
General Concepts
 The real behaviour of fluids flowing is well described by
an experiment carried out by Reynolds in 1883. He set
up the following apparatus:
The discharge is controlled
by the valve and the small
„filament‟ of dye (practically
a streamline) indicates the
behaviour of the flow. By
changing the flow Reynolds
noticed:
Flow in pipes (cont’d)
At low flows/velocities the filament remained intact and almost straight.
This type of flow is known as laminar flow, and the experiment
looks like this:

At higher flows the filament began to oscillate. This is called transitional

flow and the experiment looks like:

Lastly, for even higher flows again, the filament is found to break up
completely and gets diffused over the full cross-section. This is known as
turbulent flow:
Flow in pipes (cont’d)
 Reynolds experimented with different fluids, pipes and
velocities.
 Eventually he found that the following expression
predicted which type of flow was found:

In which Re is called the Reynolds Number; ρ is the fluid density; ¯v is

the average velocity; l is the characteristic length of the system (just the
diameter for pipes), and; μ is the fluid viscosity.
The Reynolds Number is a ration of forces and hence has no units.
Flow in pipes (cont’d)
 Flows in pipes normally conform to the following:
 Re < 2000 : gives laminar flow;
 2000 < Re < 4000: transitional flow;
 Re > 4000 : turbulent flow.
 These values are only a rough guide however. Laminar
flows have been found at Reynolds Numbers far beyond
even 4000.
 Flow in pipes (cont’d)

Hagen-Poiseuille Equation for Laminar Flow

Turbulent Flow
Description
 In turbulent flows there is no simple
description of the shear forces that act in
the fluid.
 Therefore the solutions of problems
involving turbulent flows usually involve
experimental results.
Turbulent Flow (cont’d)
 In his work, Reynolds clarified two
previous results found experimentally:
 Hagen and Poiseuille found that
friction head loss is proportional to the
mean velocity:

Reynolds found that this only applies to laminar

flows
Turbulent Flow (cont’d)
 Darcy and Weisbach found that friction
head loss is proportional to the mean
velocity squared:
Darcy-Weisbach Equation

λ is known as the pipe friction factor and is

sometimes referred to as f in American practice
(dimensionless number)
Turbulent Flow (cont’d)
Laminar Flow
 We can just equate the Hagen-Poiseuille
and the Darcy-Weisbach Equations:

Hence, for laminar flow we have:

Turbulent Flow (cont’d)
 Smooth Pipes – Blasius Equation

Stanton and Pannell confirmed that this equation

is valid for Re <10^5 . Hence it is for „smooth‟
pipes.
Nikuradse’s Experiments
 Nikuradse carried out many experiments
up to Re = 3×10^6 . In the experiments,
he artificially roughened pipes by sticking
uniform sand grains to smooth pipes.
 He defined the relative roughness (ks /D)
as the ration of the sand grain size to the
pipe diameter.
 He plotted his results as logλ against
Nikuradse’s Experiments
(cont’d)
Nikuradse’s Experiments
(cont’d)
There are 5 regions of flow in the diagram:
1. Laminar Flow – as before;
2. Transitional flow – as before, but no clear λ ;
3. Smooth turbulence – a limiting line of turbulence as Re
decreases for all ks /D;
4. Transitional turbulence – λ varies both with Re and ks /D,
most pipe flows are in this region;
5. Rough turbulence - λ is constant for a given ks / D and is
independent of Re.
The von Karman and Prandlt
Laws
 von Karman and Prandlt used
Nikuradse‟s experimental results to
supplement their own theoretical results
which were not yet accurate. They found
semi-empirical laws:
 Smooth pipes:
Rough pipes:
The von Karman and Prandlt Law for smooth pipes better
fits the experimental data than the Blasius Equation.
The Colebrook-White Transition Formula

 The friction factors thus far are the result

of experiments on artificially roughened
pipes.
 Commercial pipes have roughnesses
that are uneven in both size and spacing.
 Colebrook and White did two things:
1. They carried out experiments and
matched commercial pipes up to
The Colebrook-White Transition Formula (cont’d)
The Colebrook-White Transition Formula (cont’d)

2. They combined the von Karman and

Prandlt laws for smooth and rough
pipes:

This equation is known as the Colebrook-White

transition formula and it gives results very
close to experimental values for transitional
behaviour when using effective roughnesses for
commercial pipes.
The Colebrook-White Transition Formula (cont’d)

 The transition formula must be solved by

trial and error and is not expressed in
terms of the preferred variables of
diameter, discharge and hydraulic
gradient.
 Hence it was not used much initially.
Moody
 Moody recognized the problems with the Colebrok-
White transition formula and did two things to remove
objections to its use:
1. He presented an approximation to the Colebrook-White
formula:

He plotted λ against logRe for commercial pipes, this

is now known as the Moody diagram:
Moody (cont’d)
Barr
 One last approximation to the Colebrook-White
formula is that by Barr, who substituted the following
approximation for the smooth law component:

To get :

This formula provides an accuracy of ±1% for Re >10^5 .

Hydraulics Research Station Charts

 To derive charts suitable for design, the Colebrook-

White and Darcy-Weisbach formulas were combined to
give:

In which ν = μ/ ρ and is known as the kinematic viscosity

and Sf is the hydraulic gradient, i.e Sf = hf / L . A sample
chart is:
Pipe Design
 Local Head Losses
 In practice pipes have fittings such as bends, junctions,
valves etc.
 Such features incur additional losses, termed local
losses.
 Once again the approach to these losses is empirical,
and it is found that the following is reasonably accurate:
Pipe Design (cont’d)
 In which hL is the local head loss and kL is a constant for a
particular fitting.
 Typical values are:
Pipe Design (cont’d)
Sudden Enlargement
 Sudden enlargements (such as a pipe exiting to a
tank) can be looked at theoretically:
Pipe Design (cont’d)
 From points 1 to 2 the velocity decreases
and so the pressure increases. At 1‟
turbulent eddies are formed.
 We will assume that the pressure at 1 is
the same as the pressure at 1‟.
 Apply the momentum equation between
1 and 2:
Pipe Design (cont’d)
 Using continuity Equation:

Now applying the energy equation from 1 to 2

And so
Pipe Design (cont’d)
Substituting from above:

Multiplying out and rearranging:

Using continuity again and so

Therefore in the case of sudden contraction, the local

head loss is given by:
Pipe Design (cont’d)
Sudden Contraction
 We use the same approach as for sudden enlargement
but need to incorporate the experimental information
that the area of flow at point 1‟ is roughly 60% of that at
point 2.
Pipe Design (cont’d)

Hence:

And so :
Pipe Design (cont’d)
Partially Full Pipes
 Surface water and sewage pipes are designed to flow
full, but not under pressure.
 Water mains are designed to flow full and under
pressure. When a pipe is not under pressure, the water
surface will be parallel to the pipe invert (the bottom of
the pipe).
 In this case the hydraulic gradient will equal the pipe
gradient, S0 :
Pipe Design (cont’d)
 In these non-pressurized pipes, they often do not run
full and so an estimate of the velocity and discharge is
required for the partially full case.
 This enables checking of the self-cleansing velocity
(that required to keep suspended solids in motion
to avoid blocking the pipe).
Depending on the
proportional depth of flow,
the velocity and discharge
will vary as shown in the
following chart:
Pipe Design (cont’d)
This chart uses the
subscripts: p for
proportion; d for
partially full, and; D for
full.
Note that it is possible
to have a higher velocity
and flow when the pipe
is not full
due to reduced friction,
but this is usually
ignored in design.
Flow in Open Channels

Description
 The main difference between what we
have studied so far and open channels is
the existence of the free surface.
 It has great effect as can be seen from
the following comparison:
Flow in Open Channels
(cont’d)
Flow in Open Channels
(cont’d)
 In general, the analysis of channel flow is
more difficult than that of pipe flow as
there are many more variables.
 Some approximate analyses are
possible.
 Natural channels (mainly rivers) are the
most variable whilst man-made channels
are more regular and thus hydrological
DESIGN OF CHANNELS

 Channels must be designed with non-silting and non-

scouring velocity. Why?
 When channel is silted up, its carrying capacity is
reduced and therefore will irrigate less area
 If channel sides and bed are eroded away, the cross-
section increases and the full supply depth decreases
which reduces its command area.
 Bed and side erosion also causes various types of
damages to the canal structures on it as well as
neighbouring areas.
 Aim of channel design: is to arrive at a
cross-section, which can carry the design
flow without either scouring or silting
problems.
Design Parameters

 For the design of channels, the following data shall be

available:
 i) Design discharge Q; ii) Surface and soil properties
(roughness coefficient); silt factor f
 The design: consists of determining the following four
factors:
1. Area of cross-section, A of the channel
2. Hydraulic mean depth or hydraulic radius R
3. Velocity of flow, v
4. Longitudinal slope of the bed, S.
 To begin the design, the following two
equations are considered (available)
 Q = Av - continuity equation
 V = f(n, R, S) – Flow equation, Manning,
Chezy, Kutter.
 In the above, there are four unknowns
and therefore to solve them, we require
two additional equations.
 To get the additional two equations, the following
procedures may be adopted
 By providing a channel of best discharging section, an
equation between area A and hydraulic mean depth
may be obtained.
 Based on consideration of scouring and silting, a
limiting equation for velocity of flow may be obtained.
 Fixing the longitudinal slope based on available ground
slope.
 Based on experience, fixing a suitable width-depth
(B/D) ratio not necessary that which gives best section.
Properties of open channel
flows
 Depth (y): the vertical distance from the lowest point of the
channel to the free surface;
 Stage (h): the vertical distance from an arbitrary datum to the
free surface;
 Wetted Cross Sectional Area or Area (A): the cross sectional
area of flow normal to the flow direction;
 Wetted perimeter (P): the length of the wetted surface measured
normal to the flow;
 Surface width (B): the width of the channel at the free surface;
 Hydraulic radius (R): the ratio of area to wetted perimeter ( A/ P );
 Hydraulic mean depth: the ratio of area to surface width ( A/ B ).
Properties of open channel flows (cont’d)

The properties on a general channel are

thus:
Properties of open channel flows for various shapes
Basics of Channel Flow
Laminar and Turbulent Flow
For a pipe we saw that the Reynolds Number indicates
the type of flow:

For laminar flow, Re < 2000 and for turbulent flow, Re >
4000 .
These results can be applied to channels using the
equivalent property of the hydraulic radius:

For pipe flowing full, R=D/4, hence

 Laminar channel flow: Channel Re < 500
 Turbulent channel flow Channel Re >1000

Moody Diagrams for Channel Flow

 Using the Darcy-Weisbach equation:

And substituting for channel properties: R = D/ 4 and hf/ L =

S0 where S0 is the bed slope of the channel, we have:

Hence for Channel:

Basics of Channel Flow
(cont’d)
 The λ − Re relationship for pipes is given by the
Colebrook-White equation and so substituting R = D/ 4
and combining with Darcy’s equation for channels
gives:

•A diagram, similar to that for pipes, can be drawn based on

this equation to give channel velocities.
•This is not as straightforward though, since R varies along
the length of a channel and the frictional resistance is far
from uniform.
Basics of Channel Flow
(cont’d)
Friction Formula for Channels
 For uniform flow, the gravity forces exactly balance
those of the friction forces at the boundary, as shown in
the diagram:
Basics of Channel Flow
(cont’d)
 The gravity force in the direction of the flow is ρ
gALsinθ and the shear force in the
 direction of the flow is Ƭ0 PL , where Ƭ0 is the mean
boundary shear stress. Hence: Ƭ0 PL = ρ gALsinθ
 Considering small slopes, sinθ ≈ tanθ ≈ S0 , and so:

To estimate Ƭ0 further we again take it for turbulent flow

Basics of Channel Flow
(cont’d)
Hence we have:

Or taking out the constants gives the Chézy

Eqaution:
In which C is known as the Chézy coefficient which is not
entirely constant as it depends on the Reynolds Number
and the boundary roughness.
From the Darcy equation for a channel we see:

An Irish engineer, Robert Manning, presented a formula to

give C, known as Manning’s Equation:
Basics of Channel Flow
(cont’d)
 In which n is a constant known as Manning’s n.
 Using Manning’s Equation in the Chézy Equation gives:

And the associated discharge is:

Manning‟s Equation is known to be both simple and

reasonably accurate and is often used.
Evaluating Manning’s n
 This is essentially a roughness coefficient which determines the
frictional resistance of the channel. Typical values for n are:
Varying Flow in Open Channels

The Energy Equation : Assuming that the channel bed is has a very
small slope, the energy lines are:
Varying Flow in Open Channels (cont’d)

Hence Bernoulli‟s Equation is:

To avoid the arbitrary datum, we use a quantity called the

specific energy, Es:

For steady flow we can write this as:

This is a cubic equation in y for a given q:

In which q is the mean flow per

meter width of the channel
Cubic Equation in y for a
given q
Flow Characteristics
 In this graph we have also identified the Froude Number, Fr:

In which L is the characteristic length of the system.

The different types of flows associated with Fr are:
Fr <1: Subcritical or tranquil flow;
Fr =1: critical flow;
Fr >1: Supercritical or rapid flow.
Flow Characteristics (cont’d)
 The Froude Number for liquids is analogous to Mach
number for the speed of sound in air.
 In subcritical flow, a disturbance (waves) can travel up
and down stream (from the point of view of a static
observer).
 In supercritical flow, the flow is faster than the speed
that waves travel at and so no disturbance travels
upstream.
Flow Characteristics (cont’d)
Associated with the critical flow, as shown
on the graph, we have the critical depth:

A change in flow from subcritical to supercritical

is termed a hydraulic jump and happens
suddenly.
Flow Characteristics (cont’d)
Flow Transition
 Consider the situation shown where a
steady uniform flow is interrupted by the
presence of a hump in the streambed.
 The upstream depth and discharge are
known;
 it remains to find the downstream depth
at section 2.
Flow Characteristics (cont’d)
Flow Characteristics (cont’d)
Applying the energy equation, we have:

In addition we also have the continuity equation:

Combining we get:

Which gives:
Flow Characteristics (cont’d)
 Which is a cubic equation in y2 which mathematically
has three solutions, only one of which is physically
admissible.
 At this point refer to the specific energy curve. We see:
Flow Characteristics (cont’d)
Also we see:
 point A on the graph represents conditions at section 1
of the channel;
 Section 2 must lie on either point B or B‟ on the graph;

 All points between 1 and 2 lie on the Es graph between

A and B or B’;
 To get to B‟ the river would need to jump higher than
Δz (since Es1 − Es 2 > Δz between B and B‟). This is
physically impossible (rivers jumping?!) and so section
2 corresponds to point B.
Example

The discharge in a rectangular channel of width

5 m and maximum depth 2 m is 10m3/s. The
normal depth of flow is 1.25 m. Determine the
depth of flow downstream of a section in which
the river bed rises by 0.2 m over 1.0 m length.
 KENNEDY’S THEORY (METHOD)

 To keep silt in suspension, the silt supporting power is

directly proportional to the bed width of the stream and
does not depend on the wetted perimeter.
 The limiting velocity which does not produce silting or
scouring may be called critical velocity. The relation
between critical velocity and depth is
 vc = 0.55xD0.64

And after modification we will arrive

 v = 0.55 x m x D0.65 where m = v/vc
 v - velocity of flow in channel, m = critical velocity
ratio (C.V.R.), m varies from 0.7 for fine sand to 1.3
for very coarse sand
 To determine the mean velocity of the flow in the
channel, Kennedy used the Kutter‟s formula: v = C
(RS)1/2 where
 The difficulty with Kennedy‟s formula is
that he did not give any formula to fix the
slope S but has to be fixed based on the
natural ground slope which makes it
difficult to obtain the best section for the
given discharge.
Design of channel by Kennedy’s Method
The design procedure:
i) Assume a trial value of depth, D, in metres and determine Kennedy’s
velocity
v = 0.55 x m x D0.64. ----------- (K-1)
ii) Calculate the cross-sectional area, A, from the continuity: A = Q/v
iii) Calculate the bed width, b, since A and D are known. For eg for a
trapezoidal channel, one selects the side slope of channel based on material of
channel. Eg. Assume 1/2H:1V; then the cross-sectional area is given by:
A = bxD + 1/2xD2 from this, b can be determined.
iv) Calculate the wetted perimeter and the hydraulic radius:
The wetted perimeter, P = b + 2x[D2 + (D/2)2]1/2 = b + D(5)1/2
The hydraulic radius R = A/P = (bD +1/2D2)/(b + D(5)1/2
i) Calculate the actual mean velocity by using the Kutter’s formula;
i.e v = C(RS)1/2 where C is computed by the formula above. ----- (K-2)
The value of v in (K-2) must tarry with the v in (K-1) for the assumed depth. If the
two values do not tarry, the trial D must be changed and the whole steps repeated
till the two values of v are the same.
The design is done for a given value of S
Design an irrigation channel to carry a discharge of 50m3/s at a slope of 1/5000.
Take Kutter’s n = 0,0225 and m = 0.9.

Solution: i). Assume a depth of 2.0 m and determine Kennedy’s velocity

v = 0. 55mD0.64 = 0.55 x 0.9 x 20.64 = 0.77m/s
ii) Determine the cross-sectional area; A = Q/v = 50/0.77 = 65m2
iii) Assume side slope of 0.5H:1V
then A = bxD + 1/2D2
by putting D = 2.0, b = 31.5m
iv) Determine the wetted perimeter P = b + D(5)1/2 = 35.97m
 v The hydraulic radius R = A/P = 65/35.97 = 1.806m

 
 1 0.00155 
 23  0.0225  1 
 5000 
The Kutter’s constant C     49.63
 0.00155  0.0225
  
 1   23  1  1.806 
 
  5000  

vi) The actual velocity v = C (RS)1/2 = 49.63(1.806 x1/5000)1/2 = 0.94m/s

The actual velocity is far greater than the Kennedy velocity. Hence we assume a new
depth of 2.6 and repeat the whole procedure
Kemmedy’s velocity v = 0. 55mD0.64 = 0.55 x 0.9 x 2.60.64 = 0.91m/s
ii) Determine the cross-sectional area; A = Q/v = 50/0.91 = 54.94m2
iii) Assume side slope of 0.5H:1V
then A = bxD + 1/2D2
by putting D = 2.6, b = 19.83m
iv) Determine the wetted perimeter P = b + D(5)1/2 = 25.65m
v The hydraulic radius R = A/P = 54.94/25.65 = 2.14m

 
 1 0.00155 
 23  0.0225  1 
 5000   51.05
C  
 0.00155  0.0225 
 
 1   23  1  2.14 

  5000  
The actual velocity v = C (RS)1/2 = 51.05(2.14 x1/5000)1/2 = 1.06m/s
This value is near to the Kennedy’s critical velocity, however more accuracy
could be achieved by choosing another trial depth. The channel dimensions may
be taken as
Bed width = 19.83; Depth D = 6.6

Exercise:
Design an irrigation channel by Kennedy’s theory to carry a
discharge of 5m/s.
Take m=1.0, n=0.0225 and B/D ratio=4.40
LACEY’S REGIME THEORY
 If for Kennedy‟s theory, a channel is said to be
in regime, when there is neither scouring nor
silting, then for Lacey, a different view was
held.
 Lacey observed that even when there is
neither silting nor scouring, the channel may
or may not be in regime.
 Lacey defined two regime conditions:- i) initial
regime and ii) final regime.
 Initial Regime: refers to the state of the channel that has
formed its cross- section but not formed its longitudinal
slope. Such a channel appears to be in regime
outwardly, as there may be no visible silting or scouring.
Lacey‟s regime theory does not apply to such channels.
 Final Regime: when a channel is constructed with an
improper slope, it tries to remove the incoherent silt on
its bed and increase its slope. It first forms its section
and then it forms its final slope so that a stable condition
is established. When the channel has finished its task
of forming a stable cross-section and stable slopes, it is
said to be in “final regime”.
 Permanent Regime: This condition exists in a channel which is
protected both on its slopes by suitable protecting materials. Due
to these protections, the channel cannot change its cross-section
or slope. Lacey‟s regime theory is not applicable to channels in
permanent regime.
 True regime conditions: A channel will be in regime only when
there is no silting or scouring. To satisfy this condition, the silt
load in the channel water must be efficiently transported by the
channel cross-section. There can be only one channel with a
particular cross-section and slope for a particular silt load, which
can produce regime conditions. For a channel to be in true
regime, for any sediment load brought to it, it must adjust its
cross-section and slope to be able to transport the given load.
 However artificially constructed channels with fixed cross-
section and slope can attain regime conditions when the
following conditions are satisfied:
 flow in channel is uniform
 discharge in the channel is constant
 channel is flowing through incoherent alluvium.
Incoherent alluvium is that type of soil which can be
scoured and deposited with equal ease. Also the material
(silt) transported by the channel should be the same as
the material through which the channel is flowing.
 The amount and type of silt in the channel (silt charge and
grade) are constant.
SHAPE OF REGIME CHANNELS

 There is always only one cross-section

and one slope for a channel with a
particular discharge carrying a particular
grade of silt.
 Natural silt transporting channels
assume a semi-elliptical cross-section.
 The coarser the silt particles, the flatter
will be the semi-ellipse and the larger will
be the width of water surface.
 As the grade of silt becomes finer and finer,
the shape of the section becomes narrower.
Therefore when a channel is constructed with
small cross-section, and steeper slope than
necessary, scour starts and continues till final
regime is established.
 Similarly, in a channel with a bigger cross-
section and flatter slope, silting starts and
continues till final regime is established.
 LACEY’S REGIME EQUATIONS

 According to Lacey, silt is kept in suspension by

vertical forces, which are the vertical component of
the forces generated by eddies from the bed and
sides of the channel.
 As such the hydraulic radius R becomes a variable.
 Lacey gave two equations, one of which relates
velocity to the hydraulic radius and the other which
relates velocity to area.
 Silt grade is considered important and therefore
becomes a function in both regime equations.
Lacey’s design equations
Design of Channels by Lacey’s Theory
The following formulae shall be employed:
a) Data necessary
i) Channel discharge
ii) the silt factor
b) Formulae to be used
i) v = (f2Q/140)1/6
ii) R = 5/2 (v2/f)
iii) Q = v.A
iv) P = 4.75 (Q)1/2
v) S = (f5/3)/(3340Q1/6)

Procedure for design

1). Calculate the velocity from (i) above
2. Calculate the hydraulic radius R from (ii) above
3. Calculate A
4. Calculate wetted perimeter (P)
5. Calculate b and h
6. Calculate longitudinal slope.
Example
Design a regime channel for a discharge of 50m3/s, with silt factor f = 1.0 using
Lacey’s theory.

Solution: Given: Q = 50m3/s, f = 1.0

1) Calculate velocity , v = (f2Q/140)1/6 = {50 x (1.0)2/140}1/6 = 0.842 m/s

2. Calculate the hydraulic radius, R = 5/2 (v2/f) = 5/2(0.8422/1.0 = 1.774

3. Calculate the area A = bxD + ½ xD2 = Q/v = 50/0.842 = 59.38 m2

therefore A = b x D + 0.5 D2 = 59.38 (**)

4. Calculate the wetted perimeter P = 4.75 (Q)1/2 = 4.75 (50)1/2 = 33.6

But P = b + 2D [1 + (1/2)2]1/2 = b + (5)1/2D =33.6 (***)
Solving equations (**) and (***)
B = 29.34 and D = 1.968m
5. Slope S = (f5/3)/(3340Q1/6) = (15/3)/[3340(50)1/6] = 1/6409
S = 1: 6409

Therefore channel parameters: b = 29.34m; D = 1.97m; Q = 50 m3/s

V = 0.842m/s; S = 1:6409
Thank You For Your
Attention