Hydraulics II CHAPTER -FOUR 2012/2013

4. FLOW THROUGH PIPES

4.1 Introduction
Pipes were introduced in the earliest days of the practice of hydraulics. Their common
place use today makes it of great importance that the laws governing the flow in them
should be fully understood.

Water is conveyed from its source, normally in pressure pipelines, to water treatment
plants where it enters the distribution system & finally arrives at the consumer. In
addition oil, gas, irrigation water, sewerage can be conveyed by pipeline system.

Some loss of energy is inevitable in the flow of any real fluid. In the case of flow in a
horizontal uniform pipeline, this is evidenced by the fall of pressure in the direction of
flow. Predicting the energy loss per unit length is essential to efficient pipeline design.

The prime concern in the analysis of real flows is to account for the effect of friction.
The effect of friction is to decrease the pressure, causing a pressure ‘loss’ compared to
the ideal, frictionless flow case. The loss will be divided into major losses (due to
friction in fully developed flow in constant area portions of the system) & minor losses
(due to flow through valves, elbow fittings & frictional effects in other non-constant –
area portions of the system).

Department of WREE lecture note Page 1

Hydraulics II CHAPTER -FOUR 2012/2013

Figure 4.1 Flow in the pipes (circular pipe)

P V2 P V2
z1 + 1 + 1 = z2 + 2 + 2 + hL ……………………………………….4.1
γ 2g γ 2g
hL = Head loss (major +minor)
4.2 Laminar and Turbulent Pipe flow
Experiments show that a fluid moving along a channel, (e.g., a pipe) can flow in parallel paths (laminar
flow) or there can be a varying amount of mixing (turbulent flow). These two types of flow are
described below for the simplest case of a straight channel with a round cross-section, e.g., a pipe.
Figure below illustrates the difference between the two types of flow with respect to the velocity
profile.

In the laminar flow, the fluid behaves as if it were flowing in a concentric nest of thin cylinders one
inside another. Due to fluids viscosity, the cylinder next to the wall is stationary, while the one in the
centre of the pipe is moving with the highest velocity. There is no mixing between the layers of fluid
and velocity does not pulsate.
In the turbulent flow, there is random mixing between the layers of fluid. Due to the mixing, the
velocity distribution is much more uniform across the pipe cross section. The mixing has a positive

Department of WREE lecture note Page 2

Hydraulics II CHAPTER -FOUR 2012/2013
effect on heat transfer. The negative effect of this mixing is velocity and pressure pulsations. They can
be transferred to piping and equipment and cause vibrations.

The type of flow (laminar versus turbulent) is determined by the fluid properties (viscosity and
density), velocity and the geometry of the channel through which the fluid is flowing. Laminar flow is
promoted when:
• Dynamic viscosity is large (frictional forces resist mixing);
• Density is small (a given volume of fluid has less kinetic energy to overcome friction);
• Velocity is small (less kinetic energy to overcome friction), and
• The channel width is small (wall proximity makes mixing, i.e., movement across channel, more
difficult).
In power plants, most fluids (water, steam, and compressed gases) have low viscosities, and they
usually flow through relatively wide channels at high velocity. Also, they often change flow direction
(e.g., in a pipe elbow or around heat exchanger tubes), which promotes mixing of layers.

For these reasons, turbulent flow is very common in power plants. Laminar flow is extremely rare. For
example, it occurs in the oil film in a plain bearing, where the tight clearance between the shaft and
the bearing surface (a few tenths of a millimeter at most) does not allow mixing of layers.

ρuD
Re =
µ
Where, ρ is density, D is diameter of pipe, v is fluid velocity and μ is viscosity.
• The flow is laminar for Re < 2000
• The flow is transitional between 2000 < Re < 4000
• The flow is turbulent for Re > 4000
Values depend on shape of pipe, roughness, shape of pipe inlet. The limits are also soft numbers.

4.3. Major Losses

Referring to Figure 3.1 and for equilibrium in steady flow, the summation of forces
acting on any fluid element must be equal to zero, i.e. ∑ F = 0,
p1 A − p2 A + W sin α − τ o ( pL) = 0
( z1 − z2 )
sin α =
L
τ o - average shear stress (average shear force per unit area) at the conduit wall,
is defined by:
P
1
τo = τ o dP ……………..………………………… (4.2)
P ∫0

τ o - is the local shear stress1 acting over a small incremental portion dP of the
wetted perimeter.

1
The local shear stress varies from point to point around the perimeter of all conduits (irrespective of whether the wall is
smooth or rough), except for the case of a circular pipe flowing full where the shear stress at the wall is the same at all points
of the perimeter.

Department of WREE lecture note Page 3

Hydraulics II CHAPTER -FOUR 2012/2013

p1 A − p2 A − γAL
(z2 − z1 ) − τ PL = 0
o
L
p1 p2 PL
− − ( z2 − z1 ) − τ o =0
γ γ γA
p1 p2 PL
+ z1 = + z2 + τ o …………………………………… (4.3)
γ γ γA
Form the above equations (3.1) and 3.3
PL  p1  p 
hL = τ o =  + z1  −  2 + z 2 
γA  γ  γ 
L
hL = τ o ………………………………………… (4.4)
Rhγ

This equation is applicable to any shape of uniform cross-sections, regardless of

whether the flow is laminar or turbulent. For smooth-walled conduits, where wall
roughness may be neglected, it may be assumed that the average shear stress τ o is
a function of ρ, µ, ν& some characteristic linear dimension, which will here be
taken as hydraulic radius R. Thus:

τ o = φ(ρ, µ, ν, R)
By dimensional analysis:
 RhVρ 
τ o = ρ V 2φ   = ρ V 2φ (Re) and let φ (Re) = ½ Cf (dimensionless term)
 µ 
V2
τo = Cf ρ ………………………………………………. (4.5)
2
L V2
From equation (4.4): hL = C f ………………………………….. (4.6)
Rh 2 g

(Applied for any shape of smooth walled conduits).

For circular conduits (pipe) flowing full, R= ¼ D, Therefore,

L V2 L V2
hL = C f 4 = f ………………………………… (4.7)
D 2g D 2g
Where, f = 4C f = 8φ (Re ) …………………………..…………… (4.8)

Equation (4.7) is applicable for both smooth-walled and rough walled conduits. It is
known as pipe –friction equation, and commonly referred to as the Darcy-Weisbach
equation. Friction factor, f, is dimensionless and is also some function of Reynolds

Department of WREE lecture note Page 4

Hydraulics II CHAPTER -FOUR 2012/2013

number. The exact form of φ (Re) and numerical values for Cf and f must be determined
by experiments or other means.

For laminar flow

v 64
f = 64 = ( for la min ar flow) ………………….. (4.9)
DV Re
 64  L V
2
Head loss: hf =   …………………………………… (4.10)
 Re  D 2 g

Experimental Investigation on friction losses in Turbulent flow

In fully developed turbulent flow, the pressure drop, ∆p, due to friction in a horizontal
constant area pipe depends upon the diameter, D, the pipe length, L, the pipe
roughness,∈, the average velocity, V , the fluid density, ρ, and the fluid viscosity, µ.

By dimensional analysis ∆p = φ (V , D, ρ , µ , ε )

∆P  µ L ε 
= φ  , , 
ρV 2
 ρvD D D 
hL L  ε
2
= φ  Re, 
V g D  D
hL L  ε
2
= φ1  Re, 
V 2g D  D

 ε 
∴ f = φ1  Re,  ……………………………………… (4.11)
 D

• Blasius had concluded that there were two types of pipe friction in turbulent
flow. The first is the smooth pipes where the viscosity effects predominate so
that the friction factor is dependent solely on the Reynolds number (f=φ (Re).
He deduced the following expression for the friction in smooth pipes:

0.316
f =
4
………………………………………... (4.12)
Re
The second type was relevant to rough pipes where the viscosity & roughness effects
influence the flow & the friction factor (f) is dependent both on the Reynolds number
ε
& a parameter of relative roughness ( /D).L.F Moody prepared a chart for determining

Department of WREE lecture note Page 5

Hydraulics II CHAPTER -FOUR 2012/2013

friction factor for rough pipes experimentally by plotting f versus Re curve for each
ε
value of . (See Moody Chart)
D

Department of WREE lecture note Page 6

Hydraulics II CHAPTER -FOUR 2012/2013

The moody chart, the various flow it represents, may be divided into four zones: the
laminar flow zone; a critical zone where values are uncertain because the flow might
be either laminar or turbulent; a transition zone, where f is a function of both
Reynolds number and relative pipe roughness; and a zone of complete turbulence
(fully rough pipe flow), where the value of f is independent of Reynolds number and
depends solely upon the relative roughness.

There is no sharp line of demarcation between the transition zone and the zone of
complete turbulence. The dashed line that separates the two zone was suggested by R.
J. S. Pigott; the equation of this line is Re= 3500 . On the other hand side of the
(ε D)
equation of this line is corresponding to the curve and not the grid.

The Colebrook has developed the formula:

∈ 
= −0.809 ln  D +  …………………………………… (4.13)
1 2.523
f 3.7 Re f 
 

A simplified form of this equation is provided with restriction placed on it:

 −6 ∈ −12 
1.325 10 ≤ ≤ 10 
f = ⇒  D  …………………………… (4.14)
∈
2
 5.74  5000 ≤ Re ≤ 10 
ln( 3.7 D + R 0.9 
8
 

(For Rough pipes)

∴ Head loss in pipes is given by:

L V2
hL = f (For all pipes rough, smooth, laminar, & turbulent)
D 2g

4.4 Minor losses in the pipes

Loss due to the local disturbances of the flow conduits such as changes in cross-
section; bend, elbows, valves, joints, etc are called minor losses. In case of a very long
pipe, these losses may be insignificant in comparison with the fluid friction in the
length considered.
Whenever, the velocity of a flowing stream is altered either in direction or in magnitude
in turbulent flow, eddy currents are set up and a loss of energy in excess of the pipe

Department of WREE lecture note Page 7

Hydraulics II CHAPTER -FOUR 2012/2013

friction in that same length is created2. Head losses in decelerating (i.e., diverging)
flow is much larger than that in accelerating (i.e., converging) flow.
The most common minor losses can be represented in one of two ways. It may be
expressed as kv2/2g, where the loss coefficient k must be determined for each case. Or
it may be expressed as an equivalent length of a straight pipe, usually in terms of the
number of pipe diameters, N. Since,
V2 f ( ND ) V 2
k = , it follows that k = Nf .
2g D 2g

i. Loss of head at entrance

A poorly designed inlet to a pipe can cause an appreciable head loss. Referring to
Figure 3.2 it may be seen that, a cross section with maximum velocity and minimum
pressure at B. This minimum flow area is known as the vena contracta.

Figure 4.2 Condition at

entrance

It is seen that the loss of energy at entrance is distributed along the length AC, a
distance of several diameters. The increased turbulence and vortex motion in this
portion of the pipe cause the friction loss to be much greater than in a corresponding
length where the flow is normal, as it is shown by the drop of the total-energy line. Of
this total loss, a small portion hf would be due to the normal pipe friction (See Figure
4.2). Hence the difference between this and that total, or he' is the true value of the
extra loss caused at entrance.
The loss of head at entrance may be expressed as
2

h = ke
'
e
V …………………………………….. (4.15)
2g
Where V is the mean velocity in the pipe, and ke is the loss coefficient

2
In laminar flow these losses are insignificant, because irregularities in the flow boundary create a minimal disturbance to
the flow and separation is essentially nonexistent.

Department of WREE lecture note Page 8

Hydraulics II CHAPTER -FOUR 2012/2013

Figure 5.4 Entrance

Figure 4.3 Entrance Loss
Loss Coefficients
Coefficients

ii. Loss of head at submerged discharges: (leave of pipe), (hd’)

When the fluid with a velocity V is discharged from the end of a pipe in to a large
reservoir, (v=0), the entire kinetic energy of the coming flow is dissipated.

This may be shown by writing an energy equation between (a) and (b) in Figure 4.4
Taking the datum plane through (a) and recognizing that the pressure head of the
fluid at (a) is y, its depth below the surface, H a = y + 0 + V 2 2 g and H c = 0 + y + 0 .
Therefore,
V2
hd' = H a − H c = ……………………………….. (4.16)
2g

Figure 4.4 Submerged

Discharge Loss

iii. Loss due to contraction (hc)

a) Sudden contraction
There is a marked drop in pressure due to increase in velocity and to the loss of energy
in turbulence. The loss of head for sudden contraction may be represented by
V22
h = kC
'
c …………………………………. (4.17)
2g

Department of WREE lecture note Page 9

Hydraulics II CHAPTER -FOUR 2012/2013

Figure 4.5 Loss due to

sudden contraction

Table 4.1 Loss coefficients for sudden contraction

D2 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
D1
KC 0.50 0.45 0.42 0.39 0.36 0.33 0.28 0.22 0.15 0.06 0.00
b) Gradual contraction

In order to reduces high losses, abrupt changes of cross section should be avoided.
This is accomplished by changing from one diameter to the other by means of a
smoothly curved transition or by employing the frustum of a cone. With a smoothly
curved transition a loss coefficient kc as small as 0.05 is possible. For conical reducers, a
minimum kc of about 0.10 is obtained, with a total cone angle of 20-400. Smaller or
larger total cone angle results in higher values of kc.

A nozzle at the end of a pipe line is a special case of gradual contraction. The head loss
through a nozzle at the end of a pipeline is given by equation (4.17), where kc is the
nozzle loss coefficient whose value commonly ranges from 0.04 to 0.20 and vj is the jet
velocity. The head loss through a nozzle cannot be regarded as a minor loss because
the jet velocity head is usually quite large.

iv. Loss due to Expansion (he)

a) Sudden Expansion
Both the figures in Figure 3.6, drawn to scale from test measurements for the same
diameter ratios and the same velocities, and show that the loss due to sudden
expansion is greater than the loss due to a corresponding contraction. This is so
because of the inherent instability of flow in an expansion where the diverging paths
of the flow tend to encourage the formation of eddies within the flow. Moreover,
separation of the flow from the wall of the conduit induces pockets of eddying
turbulence outside the flow region. In converging flow there is a dampening effect on

Department of WREE lecture note Page 10

Hydraulics II CHAPTER -FOUR 2012/2013

eddy formation, and the conversion from pressure energy to kinetic energy is quite
efficient.
After the flow enters expanded pipe, there is excessive turbulence and formation of
eddies which causes loss of energy. The loss due to sudden enlargement in a pipe line
system can be calculated with the application of energy and momentum equations by
neglecting the small shear force exerted on the walls of between sections 1 and 2
(figure 4.6) for steady incompressible turbulent flow.

Figure 4.6 Loss due to sudden enlargement

Rate of momentum between section (1) & (2)

γ
p1 A2 − p2 A2 = ( A2V22 − A1V12 )
g
Energy relation between section (1) & (2)
p1 v12 p v2
+ = 2 + 2 + he
γ 2g γ 2g

p1 − p2
Solving for in each equation and equating the results
γ
v22 − v2v1 v22 − v12
= + he
g 2g
And noting that from continuity equation A1V1=A2V2 and that
A1V21=(A1V1) V1=(A2V2)V1

Substituting in the above equation

2 2
(V − V ) 2  D 2  V 2  D 2  V 2
∴ h = 1 2 = 1 − 12  1 =  22 − 1 2 …………………………….. (4.18)
'
x
2g  D2  2 g  D1  2g

Department of WREE lecture note Page 11

Hydraulics II CHAPTER -FOUR 2012/2013

b) Gradual Expansion
To minimize the loss accompanying a reduction in velocity a diffuser may be used.
Diffuser is a curved outline, or it may be a frustum of cone. In figure (4.8) the head loss
will be some function of the angle of divergence and also of the ratio of two areas, the
length of the diffuser being determined by these two variables.

Figure 4.7 Loss due to gradual enlargement

In flow through a diffuser the total loss may be considered as made up of two
components. One is the ordinary pipe-friction loss, which may be represented by
f V2
hle = ∫ . dL .
D 2g
In order to integrate, it is necessary to express the variables f, D, and V as functions of
L. For our present purpose, it is sufficient, however, merely to note that the friction
loss increases with the length of the cone. Hence, for given values of D1 and D2, the
larger the angle of the cone, the less its length and the less the pipe friction.
The other is turbulence loss due to divergence. Turbulence loss increase with the
degree of divergence, if the rate of divergence is great enough then there may be a
separation at the wall and eddies flowing backward along the walls.
The total loss for gradual expansion pipe is the sum of these two losses, marked k ' . It
has been seen that the loss due to a sudden enlargement is very nearly represented by
(V1 − V2 )2 2 g . The loss due to a gradual enlargement is expressed as

h' = k '
(V1 − V2 )2 ……………………………………..(4.19)
2g
.
Where K’ loss coefficient which is a function of cone angleα.
Table 4.2 Loss coefficients for gradual expansion
K’ 0.4 0.6 0.95 1.1 1.18 1.09 1.0 1.0
α 200 300 400 500 600 900 1200 180

v. Loss in pipe fittings

Department of WREE lecture note Page 12

Hydraulics II CHAPTER -FOUR 2012/2013

V2
The loss of head in pipefittings is expressed as h f = k f where v is the velocity in a
2g
pipe of the nominal size of the fitting. Typical values are given below.
Table 4.3 Values of “Kf” based on the type of fittings.

Table 3.3 values of k f loss for pipe fittings

Fitting K
Globe valve, wide open 10
Angle valve, wide open 5
Close –return bend 2.2
T-through side outlet 1.8
Short-radius elbow 0.9
Medium radius elbow 0.75
Long radius elbow 0.60
Gate valve, wide open 0.19
Half open 2.06
Pump foot value 5.60
Standard branch flow 1.80

vi. Losses in bend & Elbow

In flow around a bend or elbow, because of centrifugal effects, there is an increase in
pressure along the outer wall and a decrease in pressure along the inner wall. Most of
the loss of head in a sharp bend may be eliminated by the use of a vaned elbow. The
vane tends to impede the formation of the secondary flow that would other wise
occur.

The head loss produced by a bend or elbow is:

V2
hb = k b . ………………………….(4.19)
2g
kb-depends on the ratio of curvature r to pipe diameter D.

Figure 4.8 secondary flows in bend

Department of WREE lecture note Page 13

Hydraulics II CHAPTER -FOUR 2012/2013

Figure 4.9 Vaned elbow

4.5 Pipeline system

4.5.1 Pipes in Series
When two pipes of different sizes or roughness are so connected that the fluid flows
through one pipe& then through the other, they are said to be connected in series. A
typical series pipe problem, in which head H may be wanted for a given discharge or
the discharge wanted for a given H, is illustrated in figure 4.12 and the continuity
equations establish the following two simple relations that must be satisfied.

Q = Q1 = Q2 = Q3 = ... = Qn .
hL = hL1 + hL 2 + hL 3 + ...

Figure 4.11 pipes connected in series.

Applying the energy equation from A to B, including all losses, gives:

Department of WREE lecture note Page 14

Hydraulics II CHAPTER -FOUR 2012/2013
2
PA VA p V2
+ ZA + = B + B + Z B + hi + h f 1 + he'+ h f 2 + hd '
γ 2g γ 2g
V2 L V V − V2
H + 0 + 0 = 0 + 0 + 0 + ki 1 + f1 1 1 + 1
2
( ) 2
L V
2
V
+ f2 2 2 + 2
2

2g D1 2 g 2g D2 2 g 2 g
V1D1 = V2 D2
2 2
From continutyequation. :
2 4 4
L1   D1  2     
H =
V12
{ki + f1 + 1 −  D   + f 2 L2  D1  +  D1  
2g D1   2  D2  D2   D2  

4.5.2. Equivalent pipes

Series pipes can be solved by the method of equivalent lengths. Two pipe systems are
said to be equivalent when the same head loss produces the same discharge in both
systems. From Equation (4.7)

2
L 1 8Q1 f 2 L 2 8Q 2 2
hf 1 = f 1 5
for a second pipe hf2 = 5
D1 Π 2 g D2 Π 2g

For two pipes to be equivalent,

hf1 = hf2 , Q1 = Q2
f L f L
∴ 1 5 1 = 2 52
D1 D2

5 ……………………………..(4.21)
f  D 
L 2 = L 1
1
 2

f 2  D 1 

4.5.3 Pipes in parallel

A combination of two or more pipes connected as in figure 4.12 so that the flow is
divided among the pipes and then is joined again, is a parallel – pipe system. In series
pipe system the same fluid flows through all the pipes and the head losses are
cumulative, but in parallel pipe – system the head losses are the same in each of the
lines the discharge are cumulative.

Department of WREE lecture note Page 15

Hydraulics II CHAPTER -FOUR 2012/2013

Fig.4.12 Parallel Pipes system

PA P 
hf1 = hf2 = hf3 =+ Z A −  B + Z B 
γ  γ 
Q = Q1 + Q2 + Q3

Two types of problems occur:

1) If the head loss b/n A & B is given, Q is determined.
2) If the total flow Q is given, then the head loss & distribution of flow are
determined.

Size of pipes, properties, and roughness are assumed to be known. Since this type of
problem is more complex, as neither the head loss nor the discharge for any one pipe
is known. The procedure is:
1) Assume discharge Q’1 through pipe 1,
2) Solve for h’f1, using assumed discharge,
3) Using h’f1, find Q’2 & Q’3
4) With the three discharges for a common head loss, now assume that the given
Q is split up among the pipes in the same proportion as Q’1, Q’2 & Q’3, Thus,

Q1 ' Q2' Q3'

Q1 = Q, Q2= Q, Q3 = Q
∑ Q' ∑ Q' ∑ Q'
5) Check the correctness of these discharges by computing hf1, hf2, & hf3 for the
computed Q1 , Q2 & Q3

→ Q –Q1 – Q2 – Q3 = 0

4.5 Branching pipes

Let us consider three pipes connected to three reservoirs as in fig. below & connected
together or branching at the common junction point J. We shall assume that all the
pipes are sufficiently long that minor losses & velocity heads may be neglected. The
continuity & energy eqn. require that the flow entering the junction equal the flow
leaving it& that the pressure head at J (with open piezometer tube water at elevation
P) be common to all pipes.
There being no pumps, the elevation of p must lie b/n the surfaces of reservoirs A& C.
If p is level with the surface of reservoir B then water must flow in to B & Q1 = Q2 + Q3
If P is below the surface of reservoir B then the flow must be out of B & Q1 +Q2 = Q3
So for the situation of the following fig, we have the following governing conditions:
1) Q1 = Q2 + Q3
2) Elevation of p is common to all.

Department of WREE lecture note Page 16

Hydraulics II CHAPTER -FOUR 2012/2013

a. Length, diameter, &friction factors are required.

b. The flow is steady & minor losses neglected
c. Three basic equations to solve these problems are:-
i. Continuity equations
ii. Bernoulli’s equation
iii. Darcy- weisbech equation

• Total rate of in flow at junction = total rate of out flow (continuity equation)

Pipe 1 Pipe 2 Pipe 3

D1, L1, V1, Q1 hf1 D2, L2, V2, Q2, hf2 D3, L3, V3, Q3, hf3
Elevation, Z1, Reserv, A Z2, Reserv, B Z3, Reserv. C
Junction of elevation

Zj, pressure head p j r = total head at junction = ( Zj + p j r )

Applying Bernoulli’s eqn b/n the junction point & each of reservoirs

Z1 = ( pj r + Zj) + hf1 - - - - - - - - -- -- ----------------- (*) (1)

=>
Z2+hf2 = ( pj r + Zj ) -------------------------- (**) (2)

Z3 + hf3 = ( pj r + Zj ) --------------------------- (***) (3)

=> If the head of reservoir A is grater than head at junction, the flow is in to the
junction from A & out of the junction to B&C

=> Q1 = Q2 + Q3 -------------------------- * (4)

π π π
D12 V1 = D22 V2 +
D32 V3 ------------- (5)
4 4 4
=> D12 V1 = D22 V2 V2 + D32 * V3 ------------- (6)

There are three types of problem fouling of branching pies :-

Case 1: Given all pipes data (L, D, E, Z1 & Z2 Q1 or Q2, find Z3 ?
L
=> Solution: first hf1 can be calculated directly ( hf1 = f1 1 v12 2 g )
D1
Then ( p j r + Zj ) piezometric head at junction can be determine
 From eqn ( 2 ) h f2 & Q2 can be determined
 Q3 can be detrained from eqn (4) continuity eqn

Department of WREE lecture note Page 17

Hydraulics II CHAPTER -FOUR 2012/2013

 Then from eqn (3) hf3 and finally Z3 can be determined

Case 2: Given a pipe data, the surface elevation of two reservoirs (A& C) and the flow
to or from the second, find Z3 and Q1, Q3?
 From eqn (1) & iii) (hf1 + hf3) = (Z1 - Z3) (hf1 + hf3) is known & (Q1 - Q3) or (Q3 –
Q1) is known.
 Assume trail values of hf1 & hf3 & from these compute the discharge Q1 + Q3 &
compare with (Q1 - Q3)
 Repeat the procedure until the two values are equal.
 From they, piezometric head at junction can be determined
 From h +2 & ( pj r Zj )  Z2 can be determined.

Case:3 Given a pipe lengths , diameters, and the elevation of all the three reservoirs ,
find Q1 Q2, Q3,

 In this case the direction of the flow is not known clearly.

 Assam the elevation of B (Z2) is equal to the piezometric head (Zp) & ( i.e an flow in
pipe 2)
 From Zp the head losses hf1 & hf3 determined, and then Q1 & Q3 can be obtained
 If Q1 > Q3, then Zp must be increased to satisfy continuity eqn at J, causing water to
flow into reservoir B, and we will have Q1= Q2 + Q3
 If Q1<Q3, then Zp must be lowered, causing water to flow out of reservoir B, & we
will have Q1 + Q2 = Q3

4.6. Pipe networks (Grid Pattern)

A group of interconnected pipes forming several loops or circuits is called a network of pipes.
Such networks of pipes are commonly used for municipal water distribution systems in cities.
The main problem in a pipe network is to determine the distribution of flow through the various
pipes of the network such that all the conditions of flow are satisfied and all the circuits are
then balanced.
The conditions to be satisfied in any network of pipes are as follows:
1. According to the principle of continuity the flow into the junction must be equal to the
flow out of the junction.
2. In each loop, the loss of head due to flow in clock wise direction must be equal to the loss
of head due to flow in anti-clock wise direction.
3. The Darcy – Weisbach equation must be satisfied for flow in each pipe.
→ Minor losses may be neglected if the pipe lengths are large. However, if the minor
losses are large , they must be taken into account by considering them interims of the
head lost due to friction in equivalent pipe lengths
→ according to Darcy- Weisbach equation the loss of head hf though any pipe
discharging at the rate of Q can be expressed as:

H L = KQ n -------------------- (1)

Department of WREE lecture note Page 18

Hydraulics II CHAPTER -FOUR 2012/2013
Where K = proportionality factor which can be determined for each pipe, knowing the friction
factor f, length L, and the diameter D of the pipe.
fL
K= ----------------------- (2)
12.1D 5
n = an exponent having a numerical value ranging from 1.72 to 2.0.
For any pipe if Qo is the assumed discharge and Q is the corrected discharge, then:

Q = Qo + ∆Q ---------------------- (3)
and the head loss for each pipe is

H L = KQ n = K (Qo + ∆Q) n ---------------------- (4)

Thus for the complete circuit:

∑ H L = ∑ KQ n = ∑ K (Qo + ∆Q) n ------------------------- (5)

By expanding the terms the terms in the brackets:
∑ KQ n = ∑ K (Qo n + nQo n −1 ∆Q + ...) ------------------------- (6)
If ∆Q is small, compared with Qo, all terms of the series after the second one may be dropped.
Thus,
∑ KQ n = ∑ KQo n + ∑ K nQo n−1∆Q -------------------------- (7)
For the correct distribution the circuit is balanced and hence ∑ KQ n = 0 .
Therefore,
∑ KQo n + ∆Q ∑ K nQo n−1 = 0 --------------------- (8)
In the above expression ∆Q has been taken out of the summations as it is same for all the pipes
in the circuit. Solving for ∆Q:

∆Q = −
∑ KQo n

=−
∑ hL ------------------ (9)
∑ KnQo n −1
hL
n∑ ( )
Q
 In the above expression for the correction the denominator is the sum of absolute terms
and hence it has no sign.
 Further if the head losses due to flow in the clock wise direction are more than the head
losses in the due to flow in the anti-clock wise direction, then according to the sign
convention adopted, ∆Q will be negative and hence it should be added to the flow in the
anti-clock wise direction and subtracted from the flow in the clock wise direction.
 For pipes common to two circuits or loops a correction from both the loops will be
required to be applied.
 With the corrected flow in all the pipes, a second trial calculation is made for all the
loops and the process is repeated till the correction becomes negligible.

Procedures can be expressed as follows:

1. Assume any internally consistent distribution of flow. The sum of the flows entering
any junction must equal the sum of the flows leaving
2. Compute the head losses in each pipe by means of an equation or diagram.
Conventionally, clockwise flows are positive and produce positive head losses.
3. with due attention to sign, compute the total head loss around each circuit:

Department of WREE lecture note Page 19

Hydraulics II CHAPTER -FOUR 2012/2013

∑hL = ∑KQn.
4. Compute, without regard to sign, for the same circuit, the sum of: ∑KnQn-1.
5. Apply the corrections obtained from equation (9) to the flow in each line. Lines
common to two loops receive both corrections with due attention to sign.

4.7. Energy and Hydraulic Grade lines

Engineers find it useful to employ the "energy grade line" (EGL) and the "hydraulic grade line"
(HGL) in working with the pipe systems. These imaginary lines help the engineers find the
trouble spots in the system (usually points of low pressure).

The Hydraulic Grade Line:

HGL in a piping system is formed by the locus of points located a distance p/γ above the center
of the pipe, or p/γ + z above a pre-selected datum.

The Energy Grade Line:

EGL is formed by the locus of points a distance V2/2g above the HGL, or the distance V2/2g +
p/γ + z above the datum. The following hints are useful for drawing HGL and EGL.
1- As the velocity goes to zero, the HGL and the EGL approach each other. Thus, in a
reservoir, they are identical and lie on the surface.
2- The EGL and HGL slope downward in the direction of the flow due to the head loss in the
pipe. The greater the loss per unit length, the greater the slope. As the average velocity in
the pipe increases, the loss per unit length increases.
3- A sudden change occurs in the HGL and the EGL whenever a loss occurs due to a sudden
geometry change as represented by a valve or a sudden enlargement, or reduction.
4- A jump occurs in the HGL and the EGL whenever useful energy is added to the fluid as
occurs with a pump, and a drop occurs if useful energy is extracted from the flow, as in the
presence of a turbine.

Department of WREE lecture note Page 20

Hydraulics II CHAPTER -FOUR 2012/2013

5- At points where the HGL passes through the centerline of the pipe, the pressure is zero. If
the pipe lies above the HGL, there is a vacuum in the pipe, a condition that is often avoided,

if possible, in the design of piping systems

4.8. Water Hammer

 In pipe flows the change in pressure above or below the normal working
pressure caused by change in velocity has the effect of Hammering
Action on the walls of pipe and this action is called Water Hammer.
• Causes:
 In general, the change in velocity are generated by valve operation, power
failure, sudden start or shut down of pumps, rupture of the pipe line, loose pipes
and faulty valves sometimes caused by shock waves etc.
Signs
 Joints loosening and resulting in leaks.
 Noisy sound
 Damaged valves
 Ruptured pipe.
 Vibration of pipe when not clipped
• Solutions:
 Hydropower system

Department of WREE lecture note Page 21

Hydraulics II CHAPTER -FOUR 2012/2013

• Surge Tank
 Water supply system
• Pressure reducing devices
• Elevated reservoir
A water supply system is subjected to water hammer whenever there is a change in steady state
condition such as pump stoppage, pump startup or valve closure. As a system changes from one
steady state condition to another there is a transient change in flow and pressures as the system
settles to the final steady state condition. The magnitude of transient pressures (or water
hammer) and the time duration of the transient condition depends on the flow rate velocity,
pipeline material and the system boundary conditions such as tanks, pumps, air valves, control
valves and changes in pipeline diameter.

PROTECTION DEVICES

Every water supply system is unique in relation to water hammer effects. The most effective
solution to a potential water hammer problem may be a single or combination of protection
devices. The relative merits of various devices should be compared and the best solution
evaluated during the design phase of a new project.

A number of commonly used protection devices are described:

Flywheel
An effective device attached to pumps for generally shorter pipeline lengths. They help to
dampen surges by slowly decelerating the pump speed on pump stoppage.

Air Vessel
A pressure vessel containing air and water. It is a very effective device for controlling both
positive and negative pressure surges and is often used as a last resort because of high capital
costs.

One Way Surge Tower

Is an open ended device which is connected to the pipeline by a check valve. It allows water to
enter the pipeline when the pipeline is subjected to vacuum pressures.

Non Return Valves

These are often used on steeply rising pumping mains. They help to prevent the pipeline length
of water falling back on the pump’s check valve following pump stoppage.

Standpipe
Can be used in low pumping head systems where the height of the standpipe does not become
excessive. They are often used on gravity systems.

Control Valves
They are often fitted on pump discharges. They are opened and closed slowly to minimize
water during pump stoppage and startup. They are not effective during a sudden pump
stoppage.

Department of WREE lecture note Page 22