THE FEDERAL POLYTECHNIC MUBI

HYDRAULICS LECTURE NOTES (CEC 301)

1
1.0 BASICS OF FLOW IN OPEN CHANNELS

Water travels downhill from points of higher energy to points of lower energy (unless forced

to do otherwise) until it reaches a point of equilibrium, such as an ocean. This tendency is

facilitated by the presence of natural conveyance channels such as brooks, streams, and

rivers. The water’s journey may also be aided by man-made structures such as drainage

swales, pipes, culverts, and canals. Hydraulic concepts can be applied equally to both man-

made structures and natural features.

1.1 Flow Conveyance

Two types of conduits are used to transport water; namely,

 Open channels and

 Closed conduits.

2
The main difference between these channel types is the free surface, which is an interface

between the air and water layer. An open channel has a free surface unlike a closed conduit

where the water must flow full. At the free surface, the pressure is constant, and for many

situations it is atmospheric. Accordingly, in open channels the gravity constitutes the main

force that causes the flow, and hence, a channel bed slope must be maintained

downstream. Thus, an open channel flow is often referred to as a gravity flow. In closed

conduits, the flow results from an external pressure force that may cause the water to flow

against the direction of the bed slope, and is often referred to as a pressurized flow.

The accurate solution of an open channel flow is more difficult and empirical compared to

that of a closed conduit flow. One factor that complicates the solution in open channels is

the variation of the location of the free surface with respect to time and space. Another

factor is that the depth of flow is highly interrelated with factors such as discharge,

boundary roughness, slope and the channel cross section. One additional factor is that the

boundary roughness, slope and channel cross section can also be variable along the reach,

such as for the case of natural rivers and streams.

Open channels can be classified as being:

 Natural or artificial,

 Covered or open at the top,

 Prismatic or non-prismatic, and

 Regular or irregular.

3
Natural channels include rivers, streams and estuaries, while artificial channels include

storm and sanitary sewers, culverts and canals. All natural, and most artificial, channels are

open at the top, but some of the artificial channels are covered for security or for some

other reasons. An example of a closed open channel is the pipe storm or sanitary sewer.

Examples to those that are open at the top are the natural rivers and the artificial canals for

drainage and irrigation. A prismatic channel has a constant bed slope and a cross sectional

shape that do not vary along the reach. Otherwise, the channel is non-prismatic.

Based on shape, artificial channels are usually built with regular geometric cross sections

such as rectangular, trapezoidal, triangular or circular shapes. Natural channel sections have,

in general, irregular shapes.

1.2 Basic Principles

1.21 Classification of Flows

Flow in open channel can be classified based on time and space.

For time criterion, the flow may be

 Steady or

 Unsteady.

Steady flow has a water depth constant with time at a particular point on the channel, while

the depth in unsteady flow changes with time at a particular point on the channel.

For space criterion, the flow may be either

4
  Uniform or

 Non-uniform.

Uniform flow has a constant water depth along the reach of the channel, but the depth of

water in non-uniform flow changes along the reach of the channel.

Non-uniform flows, also termed varied flow, can further be classified as

 Rapidly varied and

 Gradually varied flows.

For rapidly varied flow, the water depth changes significantly over a relatively short distance

such as the case for a hydraulic jump or a hydraulic drop. For gradually varied flow, the

depth changes rather slowly over a relatively long distance such as the case of a reservoir

upstream of a dam.

According to these classifications, four combinations of flows can be considered.

1. Steady uniform flow has a constant water depth that does not change with time and

space, and is found practically only in lab flume experiments. Although this case is

rarely found in nature, it is widely used for the design of open channels due to its

simplicity.

2. Unsteady uniform flow has a depth function of time f (t), which requires the water

surface fluctuating with time while remaining parallel to the channel bottom.

Obviously, this case is very rare and cannot be found in nature. Therefore, the term

uniform flow is used hereafter to refer only to steady uniform flow.

5
 3. Steady non-uniform flow has a depth function of space f (x), such as a reservoir

behind a dam.

4. Unsteady non-uniform flow has a depth function of both space and time f (x , t) and is

very common in nature, such as waves and hydraulic bores.

1.22 State of Flows

The state or behavior of an open channel flow is governed by the effects of gravity and

viscosity forces relative to the inertia force.

3
Gravity force=Mg( ρ L g)
du V 2
Viscous force=μ A=μ L (μVL)
dy L

()
2
L 2 L
Inertial force=Ma= ρ L3 =ρ L = ρ L2 V 2
t2 t

Where M =¿ mass; g=¿gravity acceleration; ρ=¿density; L=¿ characteristic length; μ=¿

viscosity; A=¿area; V =¿velocity; and a=¿ acceleration

Dimensionless parameters measuring the ratios of these forces can be used as a basis for

considering whether or not the variables have a significant influence on the flow.

Two dimensionless parameters are considered,

 The Reynolds number and

 The Froude number.

6
Reynolds number shows the effect of the inertia relative to viscosity forces as

Inertia force ρ L2 V 2 VL VL
Re = = = =
Viscous force μVL
()μ
ρ
υ (1.1)

Where υ=¿kinematic viscosity of fluid

An open channel flow is laminar if Reynolds number is small ( Re <500) and turbulent if it is

large ( Re >1000), while in between the flow is transition or intermittent ( 500< Re <1000). In

laminar flow, the water particles have no significant mixing with each other, flowing in

laminas or layers. In turbulent flow, the movement of water varies irregularly. In transition,

the flow is unstable and difficult to control, and would be either laminar or turbulent.

High values of Reynolds number in turbulent flows indicate small viscous forces relative to

inertia forces and, therefore, the viscosity could possibly be neglected as a variable. Since

the viscosity of water is very low, it is difficult to have laminar flows in open channels.

Usually the flow in open channel is turbulent with high Reynolds numbers. Thus, the viscous

forces would not be needed for the dynamic similarity of open channel flows. However,

when a high viscosity liquid flows in an open channel or when water flows with relatively low

Reynolds numbers, the effect of viscosity should be taken into consideration.

Both gravity and inertia forces emphasize the dynamic similarity in open channel flow. The

effect of gravity as the driving force in open channel flow is present due to the existence of

the free surface. The ratio of inertia to gravity forces is used to develop the Froude number

as

7
F r=
√ Inertia force
Gravity force √
=
ρ L2 V 2
ρL g
3
√
=
V2
=
V
Lg √ Lg
(1.2)

The square root is taken in this expression because it is desirable to have V in the first power

as in the Reynolds number.

For F r >1, the flow has relatively high V and shallow L, which is referred to as supercritical

flow. If F r <1, the flow has relatively low V and deep L, is referred to as subcritical flow. For

F r=1, the flow is critical with critical V and L.

Another reason to take the square root in the above expression is to develop √ Lg, which is

equal in open channel to the celerity c of a gravity wave. This provides an indication about

the wave propagation. When F r=1, V = √ Lg implying a stationary wave(V =c). However if

F r >1, a wave can move upstream ( V <c ). Flow with F r >1 is characterized by a wave moving

downstream (V >c ).

1.23 Channel Geometry

To perform the hydraulic analysis for an open channel with a specific geometry, the

following elements are most commonly used:

 y is the flow depth from the free surface to the bottom of the channel;

 b is the width of the channel bottom; A is the cross sectional area normal to the

flow direction;

 P is the wetted perimeter measured as the length of the line of contact with the

water and the channel; and

 B is the top width at the free surface.

8
Other geometrical elements used to designate the geometry of the channel cross section

are the hydraulic radius R and the hydraulic depth D , defined as

A
R= (1.3)
P
A
D= (1.4)
B

The hydraulic radius is a measure of the efficiency at which the channel can flow. For a given

channel geometry and slope, the greatest hydraulic radius provides the largest flow.

The hydraulic depth constitutes a mean depth of flow, which becomes equal to the actual

depth for the case of a rectangular cross section, where D= y .

Table 1.1 shows the geometric elements for regular channel cross sections. The simplest

channel shape is the rectangular section. Many theoretical developments focus on

rectangular section to simplify the mathematics associated with open channel flow. The

circular section is another important one to consider, since it is used widely for drainage and

sewer systems. A composite cross section is one made up of several subsections.

Table 1.1: Channel geometry for regular channel cross sections

9
Figure 1.1: Dimensional elements defined for some channel types sections

10
1.24 Velocity of Flow

The velocity of a section is not constant throughout the cross-sectional area. Instead, it

varies with location. The velocity is zero where the fluid is in contact with the conduit wall.

The variation of flow velocity within a cross-section complicates the hydraulic analysis, so

the engineer usually simplifies the situation by looking at the average (mean) velocity of the

section for analysis purposes. This average velocity is defined as the total flow rate divided

by the cross-sectional area, and is in units of length per time.

Q
V= (
A
SEQ Equation ¿ ARABIC1.5)

1.25 Energy Concepts

The energy equation between two sections in an open channel as shown in Figure (1.2) can

be expressed based on the principle of conservation of energy as

2 2
p1 V 1 p2 V 2
+ + z 1= + + z 2+ hL (1.6)
γ 2g γ 2g
p 2
Where p=¿pressure; γ=¿ specific weight of fluid; =¿ pressure head; V =¿ velocity head;
γ 2g

z=¿ potential head; and h L =¿head loss

11
Figure 1.2: Distribution of energy heads in open channel flow between two sections

The head terms in the equation represent different types of energy per unit weight of water

contained in the channel, referred to as head since they have the dimensions of length. The

loss in energy between the two sections designated as h L reflects the major loss due to

boundary friction. If other losses are present, e.g. due to effects of turbulence, they can be

added on the right side of the equation. If pressure distribution is assumed to be

hydrostatic, i.e. p=γy , then the energy equation can be written as

V 21 V 22
y1+ + z 1= y 2+ + z 2 +h L (1.7)
2g 2g
A useful term defined as the energy head referred to the channel bed as datum (i.e. z=0 ) is

the specific energy E , given as

12
 2
V (1.8)
E= y +
2g
This term which is simply defined as the sum of the depth of flow and the velocity head

provides a means to analyze complex flow situations. From the continuity equation with

flow discharge expressed asQ=VA , the specific energy can be rewritten as

Q2
E= y + 2
(1.9)
2gA
For a givenQ : if A=f ( y ) , then E=f ( y). This expression can be used to plot the relation

between E and y as shown in Figure (1.3). It is seen that as y becomes very large, E

approaches y so that the straight line E= y is an asymptote of the upper limit. When y

approaches zero, E becomes very large approaching infinity so that the E axis is an

asymptote of the lower limit. It is also seen that for each value of E there are two positive

flow depths; y 1 and y 2, named the alternate depths. The fact that there can be more than

one possible depth for a given specific energy leads to the question of which depth will

occur in the flow. This can be evaluated by calculating the Froude number F r. The smaller

water depth y 1 corresponds to the supercritical flow ( F r >1), and the larger depth y 2 to the

subcritical flow ( F r <1). It is worth noting that for a givenQ , there is only one minimum

possible value of specific energy, called the critical energy Ec , producing a single critical

depth y c . This statement can be proved true mathematically by showing that if the specific

dE
energy is minimized, i.e. when =0 , then the flow condition will be critical ( F r=1).
dy

13
Figure 1.3: Specific energy diagram

Example 1.1

Show that when the specific energy is minimized, then the flow condition will be critical (

F r=1).

Solution

Differentiating the specific energy given by Equation (1.9) with respect to depth of flow y .

14
 −2
dE Q2 d( A ) Q2 −3 dA Q2 dA
=1+ =1+ (−2 A ) =1−
dy 2 g dy 2g dy g A3 dy

Given thatQ=VA :

2
dE V dA
=1−
dy gA dy

Now consider figure below

When the depth of flow increases bydy , the incremental change in area A=Bdy , which we

dA
can write as =B . Substituting this value into the above equation yields
dy

2
dE V
=1− B
dy gA

A
Now recall that the hydraulic depth is defined by D= , implying that:
B

2
dE V
=1−
dy gD

dE
The specific energy is at minimum when =0 therefore resulting in
dy

15
 2
V
0=1−
gD

V2
⟹ =1 ⟹ F r =1
gD

Example 1.2

Show that for rectangular channels, the minimum specific energy and critical depth become

3
√
2
respectively Ec = y c and y c = 3 q .
2 g

Q
q=¿ Discharge per unit width of channel, defined asq= .
b

Solution

For rectangular channels, hydraulic depth D equals the depth of flow y . At critical flow

Vc
F r= =1. Where V c =¿ critical velocity such that the expression can be written as
√g y c

V 2c y c
=
2g 2

Using the above expression with equation (1.8), the minimum specific energy for rectangular

channels becomes

V 2c yc 3
Ec = y c + = y c + = y c
2g 2 2

To express the critical depth in terms of discharge per unit width of channel,

16
 Q V c yc b q
q= = ⟹ V c=
b b yc

Therefore:

( )
2
q
V 2c yc y
= = c
2g 2g 2

Solving we have

√
2
3 q
yc=
g

Example 1.3

Water flows in a triangular channel with a discharge ofQ=3 m 3 / s. Determine the specific

energy, hydraulic depth, Froude number, and the alternate depth.

Solution

For the channel cross section,

B= y ( 1+1 )=2.5 ×2=5 m

17
 1
A= × 2 y 2=6.25 m2
2

The required information can be determined as

Q2 32
E= y + 2
=2.5+ 2
=2.51 m
2gA 2× 9.81× 6.25

A 6.252
D= = =1.25 m
B 5

Q 3
V A 6.25
F r= = = =0.137
√ gD √ gD √ 9.81 ×1.25

To determine the alternate depth, it should be noted that the specific energy remains the

same, i.e.

Q2
E= y +
2 g¿¿

This equation has two positive roots; one of them is the original depth of y=2.5m . The other

is the alternate depth and is found as y=0.71m .

1.3 Uniform Flow

Uniform flow is considered to be steady only, because unsteady uniform flow is practically

non-existent. In open channels, uniform flow occurs when all the forces exerted on the

water body are in balance and there is no acceleration of flow. Under uniform flow, the

gravity force is balanced with the resistance force. The pressure distribution of water

becomes in this case hydrostatic with p=γy . This flow can be assumed to occur in prismatic

18
open channels with constant cross section and bed slope. The main properties of uniform

flow are the constant water depth and velocity, and the parallel slopes of channel bed S0 ,

water surface Sw and energy grade line Sf (Figure 1.2). It can be shown that for uniform flow

∆z ∆(z + y)
S0 and Sw are equivalent, as S0 = =tanθ and Sw = =tan θ, but Sf is not, defined as
∆x ∆x

hL
Sf = =sin θ . However, if the longitudinal slope angle θ is small enough (θ<10 ° ), then S0, Sw
L

and Sf will be the same and can be considered equal to a common slope S.

The Darcy-Weisbach equation can be used to derive the velocity of flow in an open channel

2
L V (1.10)
h L =f
4R 2g
hL
Assuming a small longitudinal slope angle θ and substituting by ¿ , we have
L

V =c √ RS (1.11)

Where

c=
√ 8g
f
(1.12)

The equation is termed the Chezy equation and c is called the Chezy coefficient. The friction

factor f is a function of channel roughness, hydraulic radius and Reynolds number. However,

for completely turbulent flows, the Reynolds number can be eliminated from the analysis of

open channel flow, and thus the friction factor can be determined explicitly from the

Colebrook equation as

1 k
=−0.86 ln (1.13)
√f 14.8 R

19
Where k =¿roughness element of the channel

The practical application of this equation is limited, because more investigation is required

for the determination of the proper value ofk .

Another expression for the uniform flow is suggested by Manning by proposing that

1
c= R 1/ 6 (1.14)
n

To obtain

2 1
1 (1.15)
V = R3 S2
n

Where n=¿roughness or Manning coefficient of the channel

The Manning equation is widely adopted for the estimation of water velocity in open

channels. In general, the selection of the n value is based on best engineering judgment.

Representative values for various surfaces are given in Table (1.2).

An examination of the Manning equation reveals that it can be solved explicitly for V , n or S,

while a trial and error solution is necessary when it is required to find the flow depth or any

of the remaining parameters. The depth associated with uniform flow is designated y 0 and it

is called uniform or normal depth.

Table 1.2: Typical values of Manning's coefficient n

20
Example 1.4

Estimate the uniform depth of flow y 0 for a rectangular open channel givenQ=6 m3 /s,

n=0.012, S=0.0016, and b=4 m.

Solution

A=b y 0=4 y 0

P=b+2 y 0=4 + y 0

21
 A 4 y0
R= =
P 4 + y0

2 1
1
UsingQ= AR 3 S 2 ,
n

( )
2/3
1 4 y0
6= 4 y0 ( 0.0016 )1 /2
0.012 4+ y 0

The flow depth is determined as y 0=0.7 m.

1.4 Rapidly Varied Flow

Rapidly varied flow is a non-uniform flow with the water depth changing significantly over a

relatively short distance. Accordingly, the loss in energy due to the boundary friction is

generally small and can be neglected. Rather, most of the head loss within the flow is caused

by the high turbulence. For this case, the pressure distribution within the flow cannot be

assumed hydrostatic as the streamlines are highly curved. An example of a rapidly varied

flow is the hydraulic jump.

1.41 Hydraulic Jump

A hydraulic jump occurs in the transition from supercritical to subcritical flow (Figure 1.4).

The depths of flow upstream and downstream of the jump are called conjugate depths. The

transition between supercritical and subcritical flow results with an effective energy loss

that cannot be neglected. Owing to the complex internal flow pattern with energy losses

that initially are unknown, the momentum equation is employed for the formulation and

analysis of the hydraulic jump problem.

22
Figure 1.4: Hydraulic jump

For a system of one input and one output as in Figure (1.5), the momentum equation can be

written as

∑ F x=ρQ (V 2−V 1 )

Here, the subscripts 1 and 2 refer to conditions before and after the jump, respectively.

23
Figure 1.5: A system of one input and one output in non-uniform open channel flow

Considering the forces

F 1−F2−F f + W sinθ=ρQ (V 2 −V 1)

Where F 1 and F 2=¿ hydrostatic forces at two sections; F f =¿ boundary friction force; and

W =¿ weight of water

Since F f and W sin θ can be neglected for a short distance and for a small θ , respectively,

then

F 1−F2= ρQ( V 2−V 1)

For a rectangular cross section of widthb :

1 2 1 2
γ y b− γ y 2 b=ρQ
2 1 2
Q
(−
Q
y2 b y1 b )
This equation can be simplified to obtain

y2
(−1+ √ 1+8 F r 2 )
2
y 1= (1.16a)
2
y1
y 2= (−1+ √1+8 F 2r 1 ) (1.16b)
2
V1 V2
Where F r 1= and F r 2=
√ g y1 √ g y2

The headloss ∆ E through the hydraulic jump can be computed as

24
 (
∆ E=E1−E 2= y 1+
Q2
)(
2 g y 21 b2
− y 2 +
Q2
2 g y 22 b2 )
This will yield,

3
( y 2− y 1 )
∆ E=
4 y1 y2
(1.17)

1.42 Length of a Hydraulic Jump

The length of a hydraulic jump L can be estimated from figure (1.6). The location of the

hydraulic jump is usually determined from water surface profile computations both

upstream and downstream of the jump.

Figure 1.6: Length of hydraulic jump for a rectangular open channel

1.43 Applications of Hydraulic Jump

25
  The main application of hydraulic jump is to provide energy dissipation in hydraulic

structures such as dams and weirs in order to prevent scouring downstream of the

structures

 They can also be applied to maintaining high water levels in channels for irrigation or

other water distribution purposes

 The knowledge of the surface profiles of a hydraulic jump is desirable in the case for

designing the freeboard of a channel

 The hydraulic jump is also important in the design and analysis of storm sewer

systems as the increase in downstream water depth may result in surcharged pipe

flow conditions.

Example 1.5

A hydraulic jump occurs in a 4 m wide rectangular channel. The discharge in the channel is

Q=7.5 m /s and the depth upstream of the jump is y 1=0.2 m. Determine the downstream
3

depth y 2, the energy loss∆ E and the length of the hydraulic jump L.

Solution

Given that

Q 7.5
V1 A1 (4 × 0.2)
F r 1= = = =6.7
√ g y 1 √ g y 1 √9.81 ×0.2

y1 0.2
y 2= (−1+ √ 1+8 F r 1 )=
2
(−1+ √ 1+8 ×( 6.7)2)=1.8 m
2 2

26
The energy loss is determined from,

3
( y 2− y 1 ) (1.8−0.2)3
∆ E= = =2.74 m
4 y1 y2 4 ×0.2 ×1.8

L
Lis determined from Figure (1.5) with F r 1=6.7 to have =6.13
y2

Hence L=6.13 y 2=6.13× 1.8=11.034

2.0 HYDROSTATIC FORCE ON A PLANE SURFACE

When a surface is submerged in a fluid, forces develop on the surface due to the fluid. The

determination of these forces is important in the design of storage tanks, ships, dams, and

other hydraulic structures. For fluids at rest we know that the force must be perpendicular to

the surface since there are no shearing stresses present. We also know that the pressure will

vary linearly with depth if the fluid is incompressible. For a horizontal surface, such as the

bottom of a liquid-filled tank (Fig. 2.1), the magnitude of the resultant force is simply F R =pA

Where p=¿ the uniform pressure on the bottom and A is the area of the bottom

27
The pressure for the open tank p=γ h . Note that the resultant force on the bottom is simply

due to the liquid in the tank since the effect of atmospheric pressure cancels out. Since the

pressure is constant and uniformly distributed over the bottom, the resultant force acts

through the centroid of the area as shown in Fig. 2.1.

Figure 2.1: Pressure and resultant hydrostatic force developed on the bottom of an open tank

For the more general case in which a submerged plane surface is inclined, as is illustrated in

Fig. 2.2, the determination of the resultant force acting on the surface is more involved. For

the present we will assume that the fluid surface is open to the atmosphere. Let the plane in

which the surface lies intersect the free surface at O and make an angle θ with this surface

as in Fig. 2.2,

28
Figure 2.2: Notation for hydrostatic force on an inclined plane surface of arbitrary shape

The x− y coordinate system is defined so that O is the origin and y is directed along the

surface as shown. The area can have an arbitrary shape as shown. We wish to determine the

direction, location, and magnitude of the resultant force acting on one side of this area due

to the liquid in contact with the area. At any given depth, h , the force acting on dA (the

differential area of Fig. 2.2) is dF=γ h dA and is perpendicular to the surface. Thus, the

magnitude of the resultant force can be found by summing these differential forces over the

entire surface. In equation form,

29
 . .
F R =∫ γ h dA=∫ γy sin θ dA
A A

Where h= y sinθ and for a constant γ andθ ,

.
F R =γ sin θ ∫ y dA (2.1)
A

The integral appearing in Eq. (2.1) is the first moment of the area with respect to the x axis,

so we can write,

∫ y dA= y c A
A

Where y c is the y coordinate of the centroid measured from the x axis which passes through

O . Equation (2.1) can thus be written as,

F R =γA y c sin θ

Or more simply as

F R =γ h c A (2.2)

Where h c is the vertical distance from the fluid surface to the centroid of the area. Note that

the magnitude of the force is independent of the angle θ and depends only on the specific

weight of the fluid, the total area, and the depth of the centroid of the area below the

surface.

In effect, Eqn. (2.2) indicates that the magnitude of the resultant force is equal to the

pressure at the centroid of the area multiplied by the total area. Since all the differential

30
forces that were summed to obtain F R are perpendicular to the surface, the resultant F R

must also be perpendicular to the surface.

Although our intuition might suggest that the resultant force should pass through the

centroid of the area, this is not actually the case. The y coordinate, y R of the resultant force

can be determined by summation of moments around the x axis. That is, the moment of the

resultant force must equal the moment of the distributed pressure force, or

. .
F R y R =∫ y dF=∫ γ sinθ y dA
2

A A

And, therefore, since F R =γA y c sin θ

∫ y 2 dA
y R= A
yc A

The integral in the numerator is the second moment of the area (moment of inertia), I x , with

respect to an axis formed by the intersection of the plane containing the surface and the

free surface ( x axis). Thus, we can write

Ix
y R=
yc A

Use can now be made of the parallel axis theorem to express I x as

2
I x =I xc + A y c

Where I xc is the second moment of the area with respect to an axis passing through its

centroid and parallel to the x axis. Thus,

31
 I xc
y R= +y (2.3)
yc A c

Equation (2.3) clearly shows that the resultant force does not pass through the centroid but

I xc
is always below it, since >0.
yc A

The x coordinate, x R for the resultant force can be determined in a similar manner by

summing moments about the y axis. Thus,

.
F R x R =∫ γ sin θ xy dA
A

And, therefore,

∫ xy dA I xy
A
xR = =
yc A yc A

Where I xy is the product of inertia with respect to the x and y axes. Again, using the parallel

axis theorem, we can write

I xyc
xR= +x (2.4)
yc A c

Where I xyc is the product of inertia with respect to an orthogonal coordinate system passing

through the centroid of the area and formed by a translation of the x− y coordinate system.

If the submerged area is symmetrical with respect to an axis passing through the centroid

and parallel to either the x or y axes, the resultant force must lie along the line x=x c since

I xyc is identically zero in this case. The point through which the resultant force acts is called

the center of pressure. It is to be noted from Eqns. (2.3) and (2.4) that as y c increases the

32
 hc
center of pressure moves closer to the centroid of the area. Since y c = , the distance y c
sin θ

will increase if the depth of submergence, h c increases, or, for a given depth, the area is

rotated so that the angle, θ , decreases. Centroidal coordinates and moments of inertia for

some common areas are given in Fig. 2.3.

Figure 2.3: Geometric properties of some common shapes

33
Recall that the parallel axis theorem for the product of inertia of an area states that the

product of inertia with respect to an orthogonal set of axes ( x− y coordinate system) is

equal to the product of inertia with respect to an orthogonal set of axes parallel to the

original set and passing through the centroid of the area, plus the product of the area and

the x and y coordinates of the centroid of the area. Thus, I xy =I xyc + A x c y c

Example 2.1

The 4-m-diameter circular gate of FigQ.2a is located in the inclined wall of a large reservoir

containing water ( γ=9.80 kN /m3). The gate is mounted on a shaft along its horizontal

diameter. For a water depth of 10 m above the shaft determine: (a) the magnitude and

location of the resultant force exerted on the gate by the water, and (b) the moment that

would have to be applied to the shaft to open the gate.

Figure Question 2.1

34
Solution

a)

To find the magnitude of the force of the water we can apply the eqn. (2.2)

F R =γ h c A

And since the vertical distance from the fluid surface to the centroid of the area is 10 m it

follows that

F R =(9.80 × 103 N /m3 )(10 m)(4 π )=1.23 MN

To locate the point (center of pressure) through which F R acts, we use the eqns. (2.3) and

(2.4).

I xc I xyc
y R= + y c and x R = +x
yc A yc A c

For the coordinate system shown, x R =0 since the area is symmetrical, and the center of

pressure must lie along the diameter A−A . To obtain y R

4
πR
I xc = and y c is as shown in Fig 2.1, thus,
4

π 4
( )(2 m)
4 10 m
y R= +
( 10 m
sin 60 ° )
(4 π m 2 )
sin 60°

¿ 0.0866+11.55=11.6 m

35
And the distance (along the gate) below the shaft to the center of pressure is

y R− y c =0.0866 m

We can conclude from this analysis that the force on the gate due to the water has a

magnitude of 1.23 MN and acts through a point along its diameter A−A at a distance of

0.0866 m (along the gate) below the shaft. The force is perpendicular to the gate surface as

shown.

b)

The moment required to open the gate can be obtained with the aid of the free-body

diagram of FigQ.2.1c. In this diagram W is the weight of the gate and O x and O y are the

horizontal and vertical reactions of the shaft on the gate. We can now sum moments about

the shaft

∑ M c =0

And, therefore,

M =F R ( y R− y c ) =( 1230 ×10 ) ( 0.0866 ) =1.07× 10 N . m

3 5

2.1 Pressure Prism

An informative and useful graphical interpretation can be made for the force developed by a

fluid acting on a plane area. Consider the pressure distribution along a vertical wall of a tank

of widthb , which contains a liquid having a specific weight y . Since the pressure must vary

36
linearly with depth, we can represent the variation as is shown in Fig. 2.4a, where the

pressure is equal to zero at the upper surface and, equal to γ h at the bottom.

Figure 2.4: Pressure prism for vertical rectangular area

It is apparent from this diagram that the average pressure occurs at the depth h/2, and

therefore the resultant force acting on the rectangular area A=b h is

F R =p av A=γ ( h2 ) A
Which is the same result as was obtained from Eq. (2.2). The pressure distribution shown in

Fig. 2.4a applies across the vertical surface so we can draw the three-dimensional

representation of the pressure distribution as shown in Fig. 2.4b. The base of this "volume"

in pressure-area space is the plane surface of interest, and its altitude at each point is the

pressure. This volume is called the pressure prism, and it is clear that the magnitude of the

37
resultant force acting on the surface is equal to the volume of the pressure prism. Thus, for

the prism of Fig. 2.4b the fluid force is

1
()
F R =Volume= ( γ h ) ( b h ) =γ
2
h
2
A

Where b h is the area of the rectangular surface, A .

The resultant force must pass through the centroid of the pressure prism. For the volume

under consideration the centroid is located along the vertical axis of symmetry of the

h h
surface, and at a distance of above the base (since the centroid of a triangle is located at
3 3

above its base). This result can readily be shown to be consistent with that obtained from

Eqns. (2.3) and (2.4).

This same graphical approach can be used for plane surfaces that do not extend up to the

fluid surface as illustrated in Fig. 2.5a.

38
Figure 2.5: Graphical representation of hydrostatic forces on a vertical rectangular surface

In this instance, the cross section of the pressure prism is trapezoidal. However, the

resultant force is still equal in magnitude to the volume of the pressure prism, and it passes

through the centroid of the volume. Specific values can be obtained by decomposing the

pressure prism into two parts, ABDE and BCD, as shown in Fig. 2.5b. Thus,

F R =F1 + F 2

And the components can readily be determined by inspection for rectangular surfaces.

The location of F R can be determined by summing moments about some convenient axis,

such as one passing through A . In this instance

F R y A =F1 y 1+ F 2 y2

And y 1 and y 2 can be determined by inspection.

39
For inclined plane surfaces the pressure prism can still be developed, and the cross section

of the prism will generally be trapezoidal as is shown in Fig. 2.6. Although it is usually

convenient to measure distances along the inclined surface, the pressures developed

depend on the vertical distances as illustrated.

Figure 2.6: Pressure variation along an inclined plane area

The use of pressure prisms for determining the force on submerged plane areas is

convenient if the area is rectangular so the volume and centroid can be easily determined.

Example 2.2

A pressurized tank contains oil ( SG=0.9) and has a square, 0.6 m by 0.6 m plate bolted to its

side, as is illustrated in Fig. E2.8a. When the pressure gauge on the top of the tank reads

50 kPa , what is the magnitude and location of the resultant force on the attached plate? The

outside of the tank is at atmospheric pressure.

40
Figure Question 2.2

SOLUTION

The pressure distribution acting on the inside surface of the plate is shown in Fig.Q2.2. The

pressure at a given point on the plate is due to the air pressure, ps at the oil surface, and the

pressure due to the oil, which varies linearly with depth as is shown in the figure. The

resultant force on the plate (having an area A ) is due to the components, F 1 and F 2, with

F 1 = ( ps + γ h 1 ) A

¿ [ 50 ×103 N /m2+(0.9)(9.81 ×103 N /m3)(2 m) ]( 0.36 m2)

3
¿ 24.4 ×10 N

And

41
 F 2=γ ( h2−h 1
2
A )
3
¿(0.9)( 9.81× 10 N /m )
3
( 0.62 m )(0.36 m )
2

¿ 0.954 × 103 N

The magnitude of the resultant force, F R , is therefore

F R =F1 + F 2=25.4 ×103 N

The vertical location of F R can be obtained by summing moments around an axis through

point O so that

F R y O=F 1 (0.3 m)+ F 2 (0.2 m)

Or

3 3 3
25.4 × 10 Ny O=24.4 × 10 N (0.3 m)+0.954 × 10 N (0.2m)

y O =0.296 m

Thus, the force acts at a distance of 0.296 m above the bottom of the plate along the

vertical axis of symmetry.

Note that the air pressure used in the calculation of the force was gage pressure.

Atmospheric pressure does not affect the resultant force (magnitude or location), since it

acts on both sides of the plate, thereby canceling its effect.

42
2.2 Hydrostatic Force on a Curved Surface

The equations developed in for the magnitude and location of the resultant force acting on a

submerged surface only apply to plane surfaces. However, many surfaces of interest (such

as those associated with dams, pipes, and tanks) are non-planar. Although the resultant fluid

force can be determined by integration, as was done for the plane surfaces, this is generally

a rather tedious process and no simple, general formulas can be developed. As an

alternative approach we will consider the equilibrium of the fluid volume enclosed by the

curved surface of interest and the horizontal and vertical projections of this surface.

For example, consider the curved section BC of the open tank of Fig. 2.6a. We wish to find

the resultant fluid force acting on this section, which has a unit length perpendicular to the

plane of the paper.

Figure 2.6: Hydrostatic force on a curved surface

43
We first isolate a volume of fluid that is bounded by the surface of interest, in this instance

section BC, the horizontal plane surface AB, and the vertical plane surface AC. The free-body

diagram for this volume is shown in Fig. 2.6b. The magnitude and location of forces F R and

F R can be determined from the relationships for planar surfaces.

The weight,W , is simply the specific weight of the fluid times the enclosed volume and acts

through the center of gravity (CG ) of the mass of fluid contained within the volume. The

forces F H and F V represent the components of the force that the tank exerts on the fluid.

In order for this force system to be in equilibrium, the horizontal component F H must be

equal in magnitude and collinear with F 2 and the vertical component F V equal in magnitude

and collinear with the resultant of the vertical forces F 1 andW . This follows since the three

forces acting on the fluid mass ( F 2 , the resultant of F 1 andW , and the resultant force that

the tank exerts on the mass) must form a concurrent force system. That is, from the

principles of statics, it is known that when a body is held in equilibrium by three nonparallel

forces they must be concurrent (their lines of action intersect at a common point), and

coplanar. Thus,

F H =F 2

F V =F 1+ W

And the magnitude of the resultant is obtained from the equation

F R =√ (F H )2 +( FV )2

44
The resultant F R passes through the pointO , which can be located by summing moments

about an appropriate axis. The resultant force of the fluid acting on the curved surface BC is

equal and opposite in direction to that obtained from the free-body diagram of Fig. 2.6b.

The desired fluid force is shown in Fig. 2.6c.

Example 2.3

The 6-ft-diameter drainage conduit of Fig E2.9a is half full of water at rest. Determine the

magnitude and line of action of the resultant force that the water exerts on a l-ft length of

the curved section BC of the conduit wall.

SOLUTION

We first isolate a volume of fluid bounded by the curved section BC, the horizontal surface

AB, and the vertical surface AC, as shown in Fig. E2.9b. The volume has a length of 1 ft.

45
The forces acting on the volume are the horizontal force, F 1 , which acts on the vertical

surface AC, the weight,W , of the fluid contained within the volume, and the horizontal and

vertical components of the force of the conduit wall on the fluid, F H and F V , respectively.

The magnitude of F 1, is found from the equation

F 1=γ hc A=¿

46
3.0 Dimensional Analysis, Hydraulic Similitude and Model Investigation

3.1 Introduction

Dimensional analysis is a mathematical technique which makes use of the study of

dimensions as an aid to the solution of several engineering problems. Each physical

phenomenon can be expressed by an equation, composed of variables (or physical

quantities), which may be dimensional or non-dimensional quantities. Dimensional analysis

helps in determining a systematic arrangement of the variables in the physical relationship

and combining dimensional variables to form non-dimensional parameters.

3.2 Some Uses of Dimensional Analysis

1. Testing the dimensional homogeneity of any equation of fluid motion

2. Deriving equations expressed in terms of non-dimensional parameters to show the

relative significance of each parameter

3. Planning model tests and presenting experimental results in a systematic manner in

terms of non-dimensional parameters; thus making it possible to analyze the

complex flow phenomenon.

3.3 Dimensions

The various physical quantities used by the engineers and the scientists to describe a given

phenomenon can be described by a set of quantities which are in a sense independent of

each other. These quantities are known as fundamental or primary quantities. The primary

47
quantities are mass, length, time and temperature, designated by the letters M, L, T and θ

respectively. Temperature is especially useful in compressible flow. All other quantities such

as area, volume, velocity, acceleration, force, energy power etc. are termed as derived

quantities or secondary quantities, because they can be expressed in terms of the primary

quantities. The expression for a derived quantity in terms of the primary quantities is called

the dimension of the physical quantity. For example the dimension of force can be

expressed as

[ Force ] =[ Mass× Acceleration ]

Since

[ ]
[ Acceleration ] = L2
T

[ ]
[ Force ] = ML
T
2
=[ ML T ]
−2

The rectangular brackets signify that the dimensions of the quantity are being considered.

Some engineers prefer to use force instead of mass as fundamental quantity because the

former is easier to measure. The system is then represented by symbols F-L-T instead of M-L-

T. The dimensional form of any quantity is independent of the system of units (SI, metric or

English) chosen and enables us to convert the units readily from one system of

measurements to the other.

48
3.4 Dimensional Homogeneity

Fourier’s principle of dimensional homogeneity states that an equation which expresses a

physical phenomenon of fluid flow must be algebraically correct and dimensionally

homogenous.

An equation is said to be dimensionally homogenous, if the dimensions of the terms on its

left hand side are same as dimensions of the terms on the right hand side. To illustrate this

consider the equation for the time of swing of a simple pendulum,

t=2 π
√ l
g

The dimensions of the left hand side are[ T ]. Since 2 π is a pure number, i.e., it has no

[]
1/ 2
L

√ l =[ T ]
dimension, the dimensions of are L . Thus the dimensions of the terms on both
g
T2

sides of the equation are same and hence it is dimensionally homogenous. Some of the

other examples of the dimensionally homogenous equations are as noted below:

2
Q= C d √ 2 g L H
2/ 3
3

8 θ 5/ 2
Q= C d √ 2 g tan H
15 2

32 μVL
∆ p= 2
D

49
 2
l ρV
∆ p=f
D 2

A dimensionally homogenous equation has the unique characteristic of being independent

of the units chosen for measurement i.e., if an equation is dimensionally homogenous, it can

be used without any modification with either system of units.

It is always possible to reduce a dimensionally homogenous equation to a non-dimensional

form as explained below:

[√ ] ] [ ][ ]
t ∆p l
l
g
=2 π
[
;
Q
=
√ g L H 3 /2 3 ][
2 √2
Cd ;
(∆ p)D
μV
=32
L
D
; ρV
2
2
=f
D

Further from the study of these equations it will be observed that in general the number of

dimensionless groups that can be made is equal to the number of variables constituting the

equation minus the number of primary quantities involved.

In addition to these, there are several equations in hydraulics which are dimensionally non-

homogenous, but still well applicable to flow system within their limited ranges. For

example

1 2/ 3 1 /2
V= R S
n

Is Manning’s formula which relates mean velocity V, in channels to the hydraulic mean depth

R, bed slope S and the roughness coefficient n . If n is Assumed to be a dimensionless factor,

then on analyzing the dimensions of both sides of the equation it will be found that

50
 []
L
T
≠ [ L2/ 3 ]

Since the equation is dimensionally non-homogenous, it can be used only in SI or metric

system of units, with V in m/s and R in meters. However, if this equation is to be

dimensionally homogenous then n must have the dimension [ ]

T
L
1/ 3 .

Some of the other examples of the dimensionally non-homogenous equations are

V =C √ RS

0.64
V =0.84 y

3.5 Methods of Dimensional Analysis

The following two methods of dimensional analysis are generally used:

1. Rayleigh Method

2. Buckingham π -Method

3.51 Rayleigh Method

This method of dimensional analysis was proposed by Lord Rayleigh in 1899 for determining

the effect of temperature on the viscosity of a gas. In this method a functional relationship

of some variables is expressed in the form of an exponential equation which must be

dimensionally homogenous. Thus if X is some function of variables X 1 , X 2 , X 3 ,… . X n; the

functional equation can be written in the following general form

51
X =f (X 1 , X 2 , X 3 , … . X n ) (3.1)

In this equation X is a dependent variable, while X 1 , X 2 , X 3 ,… . X n are independent variables.

A dependent variable is the one about which information is required while independent

variables are those that govern the variation of dependent variable.

The above noted equation may be expressed as

a b c m
X =C( X 1 X 2 X 3 … … .. X n ) (3.2)

In which C is a dimensionless constant which may be determined either from the physical

characteristics of the problem or from experimental measurements. The exponents

a , b , c , … ., n are then evaluated on the basis that the equation is dimensionally

homogenous. The dimensionless parameters are then formed by grouping together the

variables with like powers.

The method is now illustrated by means of an example.

Consider the problem of fluid flow through a small orifice discharging freely into the

atmosphere under a constant head H . Also let ρ be the mass density and μ the dynamic

viscosity of the fluid flowing through the orifice. The discharge Q may be assumed to

depend on the variables viz., d , H , ρ , μ and the gravitational acceleration gsince the flow is

freely into atmosphere. Therefore the functional relationship for Q may be written as

Q=f (μ , ρ , d , H , g)

Of which by Rayleigh’s method may be expressed in an exponential form as

a b c d e
Q=C ( μ ρ d H g )

52
Where C is a dimensionless constant

Substituting the proper dimensions for each variable in this exponential equation in M-L-T

system,

3 a b e
L M M L
( )=( M 0 L0 T 0 ) ( ) ( 3 ) (L)c ( L)d ( 2 )
T ¿ L T

For dimensional homogeneity the exponents of each dimension on the both sides of the

equation must be identical. Thus

For M : 0=a+b

For L :3=−a−3 b+ c+ d+ e

For T :−1=−a−2e

Since there are five unknowns in three equations, three of the unknowns must be expressed

in terms of the other two

b=−a

1 a
e= −
2 2

5 3a
c= − −d
2 2

Therefore,

Q=C [ μ ρ ]
a −a
d
( − −d )
5 3a
2 2 d ( 2 2)
H g
1 a
−

53
 [( )( 5 1
¿ C d 2 g 2 μ a ρ−a d
−3 a
2
g
−a
2 )( H dd −d
) ]

[ ( )( )( )]
1 1 a d
μ H
¿ C d2d 2 g 2
ρd
3 /2
g
1/ 2
d

( ρd )( )
a 1
μ H d− 2 1/ 2 1 /2
Q=C 2
d H g
3/ 2
g1 /2
d

( )( )
a 1
C μ H d− π 2
¿ 2
d √ 2 gH
π 2 /3 1/ 2
d 4
√ 2 ρd g
4

¿ ( a √ 2 gH ) f 1
[( ρd
μ
2/ 3
g
1 /2 )( Hd )]
This expression may be written in the usual form viz

Q=C d a √ 2 gH

Where C d is the coefficient of discharge of the orifice which can be expressed as

C d=f 1
[( ρd
μ
2 /3
g
1/ 2 )( Hd )]
It may be pointed out that both the terms in the bracket are dimensionless and C d is also a

non-dimensional factor.

3.52 Buckingham π -Method

The Buckingham π -method is based on the π -theorem which states that if there are n

dimensional variables involved in a phenomenon, which can be completely described by m

54
fundamental quantities or dimensions (such as mass, length, time etc.), and are related by a

dimensionally homogenous equation, then the relationship among the n quantities can

always be expressed in terms of exactly (n−m) dimensionless and independent π -terms.

Mathematically, if any variable Q 1 depends on the independent variables, Q2 ,Q3 , Q4 , … .Q n;

the functional equation may be written as

Q1=f ( Q2 , Q3 ,Q 4 , … .Qn ) (3.3)

And which can be transformed to another functional relationship

f 1 ( Q 1 , Q2 , Q3 , …. Qn ) =C (3.4)

And C is a dimensionless constant

This is as if x=f ( y )= y 2 +C ; whence x− y 2=f 1 ( x , y )=C . In accordance with the π -theorem, a

non-dimensional equation can thus be obtained in the following form

f 2 ( π 1 , π 2 , π 3 , … … . π n−m )=C 1 (3.5)

Wherein each dimensionless π -term is formed by combining m variables out of the total n

variables with one of the remaining (n−m) variables. These m variables which appear

repeatedly in each of the π -terms, are consequently called repeating variables and are

chosen from among the variables such that they together involve all the m fundamental

quantities (or dimensions) and they themselves do not form a dimensionless parameter.

Thus the different π -terms maybe established as shown on page 54

55
π 1=Qa1 Qb2 Q3c … .Q mm Qm +1
1 1 1 1

a2 b2 c2 m2
π 2=Q1 Q2 Q3 … . Qm Qm +2 (3.6)

⋯ ⋯⋯⋯ ⋯⋯ ⋯⋯⋯ ⋯⋯⋯ ⋯⋯⋯

a n−m b n−m c n−m m n−m

π n−m=Q1 Q2 Q 3 … .Q m Qn

In equation (3.6) each individual equation is dimensionless and the exponents a , b , c , d … m

etc., are determined by considering dimensional homogeneity for each equation so that

each π -term is dimensionless.

The final general equation for the phenomenon may then be obtained by expressing any one

of the π -terms as a function of the others as

π 1=f 1 (π 2 , π 3 , π 4 , … . π n−m)

π 2=f 2 (π 1 , π 3 , π 4 , … . π n−m)

Or any other desired relationship may be obtained.

The use of Buckingham’s π -method may now be illustrated by means of an example.

Consider the same problem of flow through a small orifice as discussed under Rayleigh’s

method, the following steps may be adopted to solve it by Buckingham’s π -method.

Step 1

56
The physical quantities involved in the phenomenon are Q , d , H , g , μ∧ρ. The functional

equation for the discharge Q may be expressed as

Q=f (d , H , g , μ , ρ)

For which its most general form may be written as

f 1 ( Q , d , H , g , μ , ρ )=C

Thus the total number of variables n=6 and all these variables may be completely described

by the three fundamental dimensions of M-L-T system. Hence m=3 . Therefore there are

(n−m)=3 dimensionless π -terms, so that

f 2 ( π 1 , π 2 , π 3 )=C 1

Step 2

In order to form these π -terms, we chose 3 repeating variables, since in this case m=3 . As

stated earlier these variables should be such that they, among them, contain all three

fundamental dimensions and they themselves do not form a dimensionless parameter. Thus

let us choose, ρ(M / L3 ), d ( L) and g(L/T 2) as repeating variables, since the above noted

requirements are fulfilled by these.

Step 3

Since physical quantities of dissimilar dimensions can neither be added nor subtracted the

terms are expressed as products as follows:

a1 b1 c1
π 1=ρ d g Q

57
 a2 b2 c2
π 2=ρ d g μ

π 3=ρ a d b gc H
3 3 2

Step 4

Expressing π 1 dimensionally in terms of M-L-T system

( ) ( )( )
a1 c1
M b1 L L3
π 1=M 0 L0 T 0= ( L)
L3 T2 T

Equating for exponents of M , L∧T , we get

For M ; 0=a1

For L :0=−3 a1+ b1 +c 1+ 3

For T :0=−2 c 1−1

From these equations, we get

−5 −1
a 1=0 ; b1= ; c 1=
2 2

( )
Q
Hence π 1= 1 5
2 2
g d

Similarly, we have

a2 b2 c2
π 2=ρ d g μ

( ) ( ) ( M¿ )
a2 c2
M
M 0 L 0 T 0= 3
( L )b L2
2

L T

58
Equating for exponents of M , L∧T , we get

For M ; 0=a2+ 1

For L :0=−3 a2+ b2 +c 2−1

For T :0=−2 c 2−1

From these equations, we get

−3 −1
a 2=−1 ; b2= ; c2 =
2 2

( )
μ
Hence π 2= 1 3
ρ g2 d 2

Lastly we have

π 3=ρ a d b gc H
3 3 3

( ) ( ) ( L)
a3 c3
0 0 0 M ( L )b L2
M L T = 3
3

L T

Equating for exponents of M , L∧T , we get

For M ; 0=a3

For L :0=−3 a3 +b3 +c 3 +1

For T :0=−2 c 3

From these equations, we get

59
 a 3=0 ;b3 =−1; c3 =−0

H
Hence π 3=
d

Step 5

Since

f 2 ( π 1 , π 2 , π 3 )=C 1

By substitution, we have

( )
Q μ H
f2 , , =C1
1
2
5
2
1
2
3
2
d
g d ρg d

Or

( )
Q μ H
1 5
=C 2 f 3 1 3
,
2 2 2 2
d
g d ρg d

From which the expression in its usual form may be explained as earlier.

3.6 Model Investigation

For the efficient design, construction and working of the various types of hydraulic

structures or machines, it is wise to try to find out in advance how the structure or machine

would behave when it is actually constructed. For this purpose the engineers have to resort

to experimental investigation. Such experiments are also necessitated in the case of

60
problems which cannot be solved completely simply by theoretical analysis. Indeed the

experiments cannot be carried out on the full-size hydraulic structure or machines which are

proposed to be erected. It is then wise to construct a small scale replica of the structure or

machine and perform the tests on it to obtain the desired information. The small scale

replica of the actual structure or machine is known as its model while the actual structure or

machine is called the prototype. The prototype may sometimes be smaller than the model.

The model tests are quite economical and convenient, because the design, construction and

operation of the model may be altered several times if necessary (without incurring much

expenditure), till all the defects of the model are eliminated and the most suitable design is

obtained. On the basis of the final results obtained from the model tests the design of the

prototype may be modified and also it may be possible to predict the behavior of the

prototype. However, the model test results can be utilized to obtain in advance useful

information about the performance of the prototype only if there is a complete similarity

between the model and prototype.

3.7 Types of Similarities (Similitude)

There are in general three types of similitude to be established for complete similarity to

exist between a model and its prototype these are:

a. Geometric similarity

b. Kinematic similarity, and

c. Dynamic similarity

61
3.71 Geometric Similarity

Geometric similarity exists between a model and the prototype if the ratios of

corresponding length dimensions in the model and the prototype are equal. Such a ratio is

defined as scale ratio and may be written as (in which subscript m and p correspond to

model and prototype respectively):

L m b m dm
Length Scaleratio , Lr = = =
Lp bp d p

A m Lm × bm
Area scale ratio , A r= = =L2r
A p Lp× bp

V m Lm × bm ×d m
Volume scale ratio ,V r= = =L3r
V p L p ×b p × d p

3.72 Kinematic Similarity

Kinematic similarity exists between the model and the prototype if:

1. the paths of the homologous moving particles are geometrically similar, and

2. If the ratio of the velocity as well as the acceleration of the homologous particles are

equal.

Homologous point means corresponding points in the model and the prototype. Since the

velocity and acceleration have magnitude and direction, kinematic similarity implies that

velocities and accelerations at corresponding points are parallel and have a constant ratio of

62
magnitude between all the corresponding sets of points. A few useful scale ratios are as

follows:

Tm
Time scale ratio ,T r =
Tp

Lm
v m T m Lr
Velocity scale ratio , v r= = =
v p Lp Tr
Tp

Lm
am T m 2 L r
Acceleration scale ratio , ar = = =
ap L p Tr 2
2
Tp

Exercise 3.1

Define the scale ratio for discharge.

3.73 Dynamic Similarity

Dynamic similarity exists between the model and the prototype which are geometrically and

kinematically similar if the ratios of all the forces acting at homologous points in the two

systems, that is, the model and the prototype, are equal. Thus for flows to be dynamically

similar, the ratios of the various forces acting on the fluid particles in one flow system should

be equal to the ratios of similar forces at corresponding points in the other flow system. In

fluid flow problems, the forces acting may be one or a combination of the several of the

following forces:

63
 1. Inertia forces, F i=ρA V 2

u
2. Friction or viscous forces, F υ =μ A
L

3. Gravity forces, F g= ρALg

4. Pressure forces, F p=PA

5. Elastic forces, F E =KA

6. Surface tension forces, F s=σA

Inertia force is the force of resistance offered by an inert mass to acceleration. According to

Newton’s second law of motion, the magnitude of inertia is equal to the product of the

particle mass and particle acceleration and its direction is opposite to the direction of the

acceleration of the particle. The conditions required for complete dynamic similarity are

developed from Newton’s second law of motion. If in a certain system of flowing fluid, a

fluid particle of mass M is subjected to accelerationa , the inertia force F , of the particle

equals Ma . Again, if all the above noted forces exist in the system under consideration, then

the resultant force ∑ F acting on the particle, which is the vectoral sum of all the forces
acting on the particle, will be equal to the inertia force of the particle, that is,

∑ F=F υ+→ F g +→ F P+ → F E +→ F s=Ma

For complete dynamic similarity to exist between the model and its prototype, the ratio of

inertia forces of the two systems must be equal to the ratio of the resultant forces. Thus the

following relationship between the forces acting on model and prototype develops:

64
 ∑ ( F)m
( F ¿ ¿ υ+ → F g +→ F P +→ F E + → F s )m
∑ (F ¿)p= ( Ma )m
¿¿
(F ¿ ¿υ+→ F g+ → F P +→ F E +→ F s) p= ¿
(Ma) p
(3.7)

In addition to the above noted conditions for complete dynamic similarity, the ratios of the

inertia forces of the two systems must also be equal the ratio of individual component

forces.

It may thus be mentioned that when the two systems are geometrically, kinematically and

dynamically similar, then they are said to be completely similar, or complete similitude is said

to exist between the model and the prototype.

3.8 Model Laws

Results obtained from model tests may be transferred to the prototype by model laws

which may be developed from the principles of dynamic similarity. Refer to the conditions

necessary for complete dynamic similitude as denoted by equation (3.7). It is noteworthy

that in almost all hydraulic problems, not all the forces noted above are simultaneously

predominant. Therefore various model laws have been developed depending upon the

significant influence of each of the forces on the different phenomena. As a consequence of

the foregoing, we shall be examining only two model laws, namely:

1. Reynolds model law, and

2. Froude model law.

3.81 Reynolds Model Law

65
For the flows where in addition to inertia, viscous force is the only other predominant force,

the similarity of flow in the model and its prototype can be established if the Reynolds

number is same for both the systems. This is known as Reynolds model law, according to

which:

( Re )model=( R e ) prototype

ρ m V m Lm ρ p V p L p
= (3.8)
μm μp

Some of the phenomena for which the Reynolds model law can be a sufficient criterion for

similarity of flow in the model and the prototype are

1. Flow of incompressible fluid in closed pipes

2. Motion of submarines completely under water

3. Motion of air planes and

4. Flow around structures and other bodies immersed wholly under moving fluids

3.82 Froude Model Law

When the force of gravity can be considered to be the only predominant force which

controls the motion in addition to the force of inertia, the similarity of the flow in any two

such systems can be established if the Froude number of both the systems is the same. This

is known as Froude model law according to which:

( F r )model=( F r ) prototype

66
 Vm Vp
= (3.9)
√ g m Lm √ g p L p

3.9 Types of Models

Hydraulic models are classified generally under two broad categories:

1. Undistorted Models, and

2. Distorted Models

3.91 Undistorted Models

An undistorted model is that which is geometrically similar to its prototype, that is, the scale

ratios for corresponding linear dimensions of the model and its prototype are same. Since

the basic condition of perfect similitude is satisfied, prediction in the case of such models is

relatively easy and many of the results obtained from the model tests can be transferred

directly to the prototype.

3.92 Distorted Models

Distorted models are those in which one or more terms of the model are not identical with

their counterparts in the prototype. Since the basic condition of perfect similitude is not

satisfied, the results obtained with the help of distorted models are liable to distortion and

have more qualitative value only.

A distorted model may have either geometrical distortion, or material distortion, or

distortion of hydraulic quantities or a combination of these. In geometrical distortion, the

67
distortion can either be of dimension or that of configuration. When different scale ratios

are adopted for the longitudinal, transverse and vertical dimensions then it is said to be a

distortion of dimensions. Distortion of dimensions is frequently adopted in river models

where a different scale ratio for depth is adopted.

Such models with vertical scales greater than the horizontal scales are called Vertically

Exaggerated models. The distortion of configuration results when the general configuration

of the model does not bear a resemblance with its prototype. for example a river model will

have distortion of configuration if it is constructed with a bed slope different from the one

given by vertical exaggeration. The material distortion takes place when the physical

properties of the corresponding material in the model and in the prototype do not satisfy

the similitude conditions. The material distortion may have to be adopted in the river models

constructed for studies of sediment transport. Also it may not be possible to obtain

similitude in respect of certain uncontrollable hydraulic quantities such as time, discharge

etc., which may lead to distortion of hydraulic quantities.

Distorted models are required to be to be prepared for the following hydraulic structures:

 Rivers

 Dams across very wide rivers

 Harbors

 Estuaries etc.

For all the above mentioned, their horizontal dimensions are large in proportion to their

vertical dimensions.

68
3.92.1 Reasons for Adopting Distorted Models

1. To maintain accuracy in vertical measurements

2. To maintain turbulent flows

3. To obtain suitable bed material and its adequate movement

4. To obtain suitable roughness condition

5. To accommodate the available facilities such as space, money, water supply and time

3.92.2 Merits of Distorted Models

1. The vertical exaggeration results in steeper water surface slopes and magnification

of wave height in models, which can therefore be measured easily and accurately

2. Due to exaggerated slopes, the Reynolds numbers of a model is considerably

increased and the surface resistance is lowered. This assists in simulation of the flow

conditions in the model and the prototype

3. In case of distorted models sufficient tractive force can be developed to produce

adequate bed movement with a reasonably small model’

4. Model size can be sufficiently reduced by its distortion, thereby its operation is

simplified and also cost is lowered considerably

3.92.3 Limitations of Distorted Models

1. The magnitude and distribution of velocities are incorrectly reproduced because

vertical exaggeration causes distortion of lateral distribution of velocity and kinetic

energy

69
 2. The pressures may not be correctly reproduced in magnitude and direction

3. Some of the flow details may not be correctly reproduced because distortion

increases longitudinal slopes of model streams thus tending to upset flow regime at a

point where artificial model roughness is required to restore it

4. Slopes of river bends, earth cuts and dikes are often so steep that they cannot be

molded satisfactorily in sand or other erodible material.

5. A model wave may differ in type and possibly in action from that of the prototype

Example 3.1

a. Show by method of dimensional analysis that the resistance R to the motion of a

sphere of diameter D moving with uniform velocity V through a fluid having a

density ρ and viscosity μ may be expressed as

R=( ρ D2 V 2 ) ϕ ( ρVD
μ
)
Also show that the above expression reduces to R=kμVD when the motion is through

viscous fluid at low velocity, where k is a dimensionless constant.

b. Find the viscosity in poise of a liquid through which a steel ball of diameter 1 mm falls,

with a uniform velocity of20 mm/ s. The specific gravity of the liquid is 0.91 and that of

the steel is7.8. Given thatk =3 π .

70
Solution

a. The functional relationship for R may be expressed as

R=f ( μ , ρ , D ,V )

This by Rayleigh’s method may be expressed in an exponential form:

R=k ( μ a ρb Dc V d )

Where k is a dimensionless constant

Now considering the dimensions for each variable on both sides of the equation in M-L-T

system, we get

[ ] [ ][ ] []
a b d
ML 0 0 0 M M c L
=[ M L T ] [L]
T
2 ¿ L
3
T

For dimensional homogeneity the exponents of each dimension of both sides of the

equation must be identical. Thus

For M ; 1=a+b ¿> b=(1−a)

For L :1=−a−3 b+ c+ d ¿> c=(2−a)

For T :−2=−a−d ¿> d=(2−a)

Therefore, by substitution we get,

R=k [ μ ρ ]
a (1−a) (2−a) (2−a )
D V

71
 ( )
a
μ
¿k (ρ D V )
2 2
ρVD

This may be expressed in the functional form as

R=( ρ D2 V 2 ) ϕ ( ρVD
μ
)
For viscous flow, R V , hence taking a=1, we get

R=k ( ρ D V )
2 2
( ρVDμ )=kμVD
b. Given D=1 mm ; V =20 mm/s ; k=3 π

For uniform velocity of the sphere, the resistance R is equal to the gravitational

force; therefore by substitution we get,

( )
3
−3 −3 1 −3
3 π × μ× 20 ×10 ×1 ×10 =π ×10 ( 7.8−0.91 ) 9810
2

4 ( 0.5 )3 (6.89× 9.81)

¿> μ=
3× 3 ×20

2
¿ 0.1878 N . s /m

¿ 1.878 poise

Example 3.2

Assuming that the rate of discharge Q of a centrifugal pump is dependent upon the mass

density ρ of the fluid, pump speed N (rpm), the diameter of the impeller D , the pressure p

and the viscosity μ, show using the Buckingham’s π -theorem that it can be represented by

72
 Q=( N D ) ϕ
3
([ NgHD )( NυD )]
2 2 2

Given that H=head and υ=kinematic viscosity of the fluid .

Solution

From the data given, we have

Q=ϕ ( ρNDpμ )

Thus there are 6 variables involved in the phenomena which are expressible in terms of

three fundamental dimensions. Hence the above fundamental relationship may be replaced

by another one involving only 3 dimensionless π -terms.

Thus taking ρ , N and D as repeating variables, we have

a1 b1 c1
π 1=ρ N D Q

a2 b2 c2
π 2=ρ N D p

a3 b3 c3
π 3=ρ N D μ

For π 1=ρa N b D c Q , expressing it dimensionally in terms of M-L-T system

1 1 1

( )( ) ( )
a1 b1 3
0 0 0 M 1 c1 L
π 1=M L T = ( L)
L
3
T T

And equating exponents of M , L, and T , we obtain

For M : 0=a1 ¿> a1=0

73
For L :0=−3 a1+ c 1+3 ¿> c1 =−3

For T :0=−b 1−1 ¿> b1=−1

π 1=
( )
Q
ND
3

For π 2=ρ a N b D c p
2 2 2

( ) ( ) ( L ) ( LMT )
a2 b2
M 1 c2
π 2=M 0 L0 T 0= 3 2
L T

For M : 0=a2+ 1 ¿> a2=−1

For L :0=−3 a2+ c 2−1 ¿> c2 =−2

For T :0=−b 2−2 ¿> b2=−2

π 2=
( ρ Np D )
2 2

For π 3=ρ a N b Dc μ
3 3 3

( ) ( ) ( L) ( M¿ )
a3 b3
0 0 0 M 1 c3
π 3=M L T = 3
L T

For M : 0=a3+ 1 ¿> a3=−1

For L :0=−3 a3 +c 3−1 ¿> c3 =−2

For T :0=−b 3−1 ¿> b3=−1

74
 π 3=
( μ
ρN D
2 )
Therefore

Q
ND
2
=ϕ
[(
p
2
ρN D
2 )( ρNμD )] 2

μ
Since p=γh=ρgH and =υ, by substituting these expressions, we get
ρ

Q=( N D ) ϕ
3
([ NgHD )( NυD )]
2 2 2

Example 3.3

Water at 15 ℃ flows at 4 m/s in a 150 mm pipe. At what velocity must oil at 30 ℃ flow in a

75 mm pipe for the two to be dynamically similar? Take υ for water at 15 ℃ as

1.145 ×10 m /s and that of oil at 30 ℃ as 3.0 ×10 m /s .

−6 2 −6 2

Solution

Since for the flow of fluids through pipes only viscous and inertia forces are predominant,

Reynolds model law is the criterion for similarity. Thus

75
 ( ) ( )
Vd
υ m
=
Vd
υ p

By substitution we get,

( )( )
−3 −3
4 × 150 ×10 V ×75 ×10
−6
= −6
1.145 ×10 3.0 ×10

V =20.96 m/s

Example 3.4

A 1 :10 scale model of a submarine moving far below the surface of water is tested in a

water tunnel. If the speed of the prototype is 8 m/s , determine the corresponding velocity of

the model in the tunnel. Also determine the ratio of the drag for the model and the

prototype. take υ seawater =1.121×10−6 m2 / s; υ water =1.0× 10−6 m2 /s; ρ seawater =1027 kg /m3 ;

3
ρwater =1000 kg /m .

Solution

Since the submarine is moving far below the surface of water, it is totally submerged and

hence the similarity between the model and its prototype can be established on the basis of

reynolds model law according to which

( VLυ ) =( VLυ )
m p

L p υm
V m =V p × ×
Lm υ p

76
 −6
1.0 ×10
¿ 8 ×10 × −6
1.121× 10

¿ 71.365 m/s

If the drag force is assumed to depend on the relative velocity V , between the machine

(model or the prototype) and the fluid, the dynamic viscosity μ, and some characteristic

length L, then it can be expressed as the functional relationship

F=ϕ ( V , ρ , μ , L )

Hence by dimensional analysis the functional relationship can be expressed in terms of

dimensionless parameters as

F=ρ L2 V 2 ϕ ( R e )

And for true dynamic similarity between model and prototype, ϕ ( Re ) must be the same.

Therefore, the ratio of the drag for the model and the prototype is

( ) ( )
Lm 2 V m 2
F m ρm
= × ×
F p ρp Lp Vp

( ) ( )
F m 1000 1
2
71.365
2
= × ×
F p 1027 10 8

¿ 0.775

77
HYDRAULIC TURBINES AND PUMPS

Hydraulic Turbines

Hydraulic or water turbines are machines which use the energy of water (hydro-power) and

convert it into mechanical energy. As such these may be considered as hydraulic motors or

primemovers. The machanical energy developed by the turbine is used in running an electric

78
generator which is directly coupled to the shaft of the turbine. The electric generator thus

develops electric power, which is known as hydroelectric power.

Classification of Turbines

Hydraulic turbines may be classified according to the followingconsiderations

 According to the action of water flowing through the turbine runners

 According to the main direction of flow of water

 According to the head and the quantity of water required

 According to the specific speed

 According to the disposition of their shaft

A. According to the action of water flowing through the turbine runners, turbines may

be classified as:

 Impulse turbines and

 Reaction turbines

a) Impulse Turbine

In an impulse turbine, all the available energy of water is converted into kinetic energy or

velocity head by passing it through a contracting nozzle provided at the end of the

penstock. The water coming out of the nozzle is formed into a free jet which impinges on a

series of buckets of the runner thus causing it to to revolve freely in the air. The water is in

contact with only a part of the runner at a time, and throughout its action on the runner and

79
in its subsequent flow to the tail race, the water is at atmospheric pressure. A cover is

however provided on the runner to prevent splashing and to guide the water discharged

from the bucket to the tail race. Examples of impulse turbines include;

I. Pelton Wheel Turbine

II. Turgo-Impulse Wheel

b) Reaction Turbine

In a reaction turbine, at the entrance to the runner, only a part of the available energy of

water is converted into kinetic energy and a substantial part part remains in the form of

pressure energy. As water flows through the runner the change from pressure to kinetic

energy takes place gradually. As such the pressure at the inlet to the turbine is much higher

than the pressure at the outlet and it remains through out the passage of water through the

turbine. For this gradual change of pressure to be possible the runner in this case must be

completely enclosed in an aie-tight casing and the passage in either full of water throughout

the operation of the turbine. The difference of pressure between the inlet and the outlet of

the runners is called reaction pressure, and hence these turbines are known as reaction

turbines. Examples of reaction turbines are:

I. Francis Turbine

II. Thompson

III. Fourneyron, and

IV. Kaplan

80
B. According to the main direction of flow of water in the runner, turbines may be

classified as:

 Tangential flow turbines

 Radial flow turbines

 Axial flow turbines, and

 Mixed flow turbines

a) In tangential flow turbine, the water flows along the tangent to the path of rotation

of the runner. Example of a tangential flow turbine is the Pelton wheel turbine.

b) In radial flow turbine the water flows along the radial direction and remains wholly

and mainly in the plane normal to the axis of rotation as it passes through the runner. A

radial turbine may either be of the inward radial flow type or the outward radial flow type.

In an inward radial flow turbine the water water enters at the outer circumference and flows

radially inwards towards the centre of the runner example include the Francis turbine,

thomson turbine and Girard turbine. In an outward radial flow turbine, water enters at the

centre and flows radially outwards towards the outer periphery of the runner example is the

Fourneyron turbine.

c) In an axial flow turbine the flow of water through the runner is wholly and mainly

along the direction parallel to the axis of rotation of the runner eaxamples are Girard axial

flow turbine, kaplan turbine etc.

d) In mixed flow turbine, water enters the runner at the outer periphery in the radial

direction and leaves it at the centre in the direction parallel to the axis of rotation of the

runner. Modern francis turbine is an example of the mixed flow type turbine.

81
C. On the basis of head and the quantity of water required, turbines may be classified

as:

 High head turbine

 Medium head turbine, and

 Low head turbine

a) High head turbines are those which are capable of working under very high heads

ranging from several hundred metres to few thousand metres. These turbines thus require

relatively less quantity of water. In general impulse turbines are high head turbines and in

particular Pelton wheel has so far been used under a highest head of about 1770 m.

b) Medium head turbines are those which are capable of working under medium heads

ranging from about 60 m to 250 m. These turines require relatively large quantity of water.

Modern francis turbine may be classified as medium head turbines.

c) Low head turbines are those which are capable of working under the head less than

60 m. These turbines thus require a large quantity of water. Kaplan and other propeller

turbines may be classified as low head turbines.

D. On the basis of the specific speed the various turbines may be considered in the

following three groups where the specific speed of the turbine is the speedof a

geometrically similar turbine that would develop one kilowatt when working under a head

of one meter:

82
 a) Specific speed varying from 8.5−30: pelton wheel with single jet, pelton wheel with

double jet has up to 43 .

b) Specific speed varying from 50−340: Francis turbine.

c) Specific speed varying from 255−860: Kaplan and other propeller turbines

The turbines may be disposed with either vertical or horizontal shafts and hence these may

be classified as turbines with vertical disposition of shaft and turbines with horizontal

disposition of shaft. Vertical disposition turbines are found in common use.

Selection of Turbines

The selection of suitable type of turbine is usually governed by the following factors:

1. Head and Specific Speed: it has been found that there is a range of head and specific

speed for which each type of turbine is most suitable. A summary is given in table

S/No Head in metres Type of turbine Specific speed(SI units)

1 300 or more Pelton wheel single or multiple jet 8.5 to 47

2 150 to 300 Pelton or Francis 30 to 85

3 60 to 150 Francis or Deriaz (Diagonal) 85 to 188

4 Less than 60 Kaplan or propeller or Deriaz or 188 to 860

Tubular

83
However, as a general rule, it may be stated that as far as possible a turbine with highest

permissible specific speed should be chosen, which will not only be the cheapest in itself but

its relatively small size and high rotational speed will reduce the size of the generator as well

as the power house.

2. Part Load Operation: the turbines may be required to work with considerable laod

variations. As the load deviates from the normal working load, the efficiency would also

vary. At part load the performance of the Kaplan and Pelton turbines is better in comparison

to that of Francis and Propeller turbines. The variability of load will influence the choice of

type of if the head lies between 150 m−300 m or lies below 30 m. For higher range of heads

Pelton wheel is preferable for part load operation in comparison to Francis turbine, though

the former involves higher initial cost. For heads below 30 m, Kaplan turbine is preferable for

part load operation in comparison to Propeller turbine.

Pump Types and Definitions

84
85