PIPE FLOW

INTRODUCTION

A pipe is a closed conduit carrying a fluid under pressure. Fluid motion in a pipe is subjected to
a certain resistance. Such a resistance is assumed to be due to Friction. Mainly due to the viscous
property of the fluid.
Fluid flow in pipes is of considerable importance in many of the processes, like;
• Animals and Plants circulation systems.
• In our homes.
• City water.
• Irrigation system.
• Sewer water system, etc
To describe any of these flows, conservation of mass and conservation of momentum equations
are the most general forms could be used to describe the dynamic system. The key issue is the
relation between flow rate and pressure drop.

CLASSIFICATION OF PIPE FLOW

Based on the values of a non-dimensional number known as Reynold’s number (Re), the flow in
the pipe can classified.
The Reynold’s Number (Re) is defined as the ratio of Inertia force of a flowing fluid and the
Viscous force. It is mathematically expressed as,
Re=(Inertia force/Viscous force) =( ρ V D/µ )
Where, ρ is mass density
V is average velocity of flow
D is diameter of pipe
µ is dynamic viscosity of flowing fluid

Based on the values of Reynold’s number (Re), flow is classified as Follows:

Laminar flow
Turbulent flow
Trasition flow

Laminar Flow or Viscous Flow

In the laminar flow, the stream lines are practically

parallel to each other or flow takes place In the form
of telescopic tubes.
Occurs when Reynold’s number Re< 2000. Fig. 1 Laminar Flow
Viscous forces are more predominant compared to
inertia Forces when the flow velocity or discharge is
low, i.e, under laminar flow condition
In laminar flow velocity increases gradually from zero at the boundary to Maximum at the center
(Fig. 2).
Laminar flow is regular and smooth and velocity at any point practically remains constant in
magnitude & direction.
Therefore, the flow is also known as stream Line flow.

Fig. 2 Velocity Distribution in Laminar Flow

In the laminar flow conditions, there will be no exchange of fluid particles from one layer to
another and no momentum transmission from one layer to another.
Ex: Flow of thick oil in narrow tubes, flow of Ground Water, Flow of Blood in blood vessels.

Turbulent Flow

In the turbulent flow the fluid flow at higher flow

rates, the streamlines are not steady and straight and
the flow is not laminar. Flow is very much disorder
and there will be violent mixing as shown in Fig.3.
The turbulent flow occurs when Reynolds number
above 4000. Fig. 3 Turbulent Flow
Fluid velocity at a point varies randomly with time.
Generally, the flow field will vary in both space and
time with fluctuations that comprise "turbulence”
(Fig. 4)

Fig.4 Velocity Distribution in Turbulent Flow

Transitional Flow:

In this, the condition is neither the laminar nor turbulent. It is intermittently turbulent flow. The
stream lines get disturbed a little.
This type of flow occurs when 2000< Re < 4000.

Fig. 3 Transitional Flow

A good example of laminar and turbulent flow is the rising smoke from a incense stick (agara
bathi) or cigarette.
The smoke initially travels in smooth, straight lines (laminar flow) then starts to “wave” back
and forth (transition flow) and finally seems to randomly mix (turbulent flow).

Fig. 4 Smoke from Incense Stick

Losses in Pipe Flow

Losses in pipe flow can be two types viz:-
a)Major Loss
b)Minor Loss
a)Major Loss: As the name itself indicates, this is the largest of the losses in a pipe.
This loss occurs due to friction only. Hence, it is known as head loss due to friction (hf)
b)Minor Loss: Minor losses in a pipe occurs due to change in magnitude or direction of
flow.
Minor losses are classified as (i) Entry Loss, (ii) Exit loss, (iii) Sudden expansion
loss (iv) Sudden contraction loss (v) Losses due to bends & pipe fittings.

These losses must be calculated so that, for example:

the proper size and number of pumps can be specified in the design of a municipal water
distribution system;
• the conduit size for a gravity-flow urban drainage project may be determined;
• the optimum size of valves and the radius of curvature of elbows can be stipulated in the
specifications of a pipeline design.

When the ratio of the length of the pipeline, , to the diameter, , exceeds 2000:1, pipe system
energy losses are predominantly the result of pipe friction.
The energy losses resulting from pipe appurtenances are termed “minor” losses and are usually
neglected in the calculation of pipe system energy losses.
In short lengths of pipe, however, these minor losses can become major sources of energy loss.

Energy loss is usually called power loss or head loss.

Head loss is the measure of the reduction in the total head (sum of elevation head, velocity
head and pressure head) of the fluid as it moves through a fluid system.
Head loss includes friction loss and local loss minor losses)

The Darcy-Weisbach equation is used to express energy loss caused by pipe friction

Head Loss due to Friction

Consider the flow through a straight horizontal pipe of diameter D, Length L, between two
sections (1) & (2) as shown. Let P1 & P2 be the pressures at these sections. Ƭ0 is the shear stress
acting along the pipe boundary.

Fig. 5 Various Components of Head Loss through pipe

From II Law of Newton

Force = Mass x accn. But acceleration = 0, as there is no change in velocity, the reason that pipe
diameter is uniform or same throughout.
Applying Bernoulli’s equation between (1) & (2) with the centre line of the pipe as datum &
considering head loss due to friction hf,.

p1 V12 p V2
Z1 + + = Z2 + 2 + 2 + hf
γ 2g γ 2g

Z1 = Z 2 Q Pipe is horizontal

V1 = V2 Q Pipe diameter is same throughout

P1 − P2
∴ = h f − − − ( 2)
γ
Substituting eq (2) in eq.(1)

From Experiments, Darcy Found that

f
ρ V 2 − − − ( 4)
τ0 =
f=Darcy’s friction factor (property of the8 pipe materials Mass density of the liquid.
V = average velocity
Substituting eq (4) in eq.(3)
f h γD 4 LfρV 2
ρV 2 = f or hf =
8 4L 8γD
But
γ
=g
ρ
 fLV 2 
∴ h f =   − − − (5)
 2 gD 
From Continuity equation, Q= AxV, V= Q/A
4Q
V=
πD 2
Substituting for V in Eq. 5,
 8 fLQ 2 
∴ h f =  2 5  − − − (6)
 gh D 

Equations (5) & (6) are known as DARCY – WEISBACH Equation

Pipes in Series or Compound Pipe
The compound pipe (or pipe n series) is an arrangement made by connecting different diameters
of pipe with a common axis as shown in figure.

Q
D1 D2 D4
L1 D3
L2 L4

Fig. 6 Compound Pipe OR Pipes in Series

Let D1, D2, D3, D4 be the diameters of the pies as shown in figure..
Let L1, L2,L3, L4 are lengths of a number of Pipes connected in series
The (hf)1, (hf)2, (hf)3 & (hf)4 are the head loss due to friction for each pipe.

The total head loss due to friction, hf , for the entire pipe system is the summation of each
of the head loss occurring in all he pipes, which is given by,

h f = hf1 + hf 2 + hf 3 + hf 4
I.e.,

Or
8 fL1Q 2 8 fL2Q 2 8 fL3Q 2 8 fL4Q 2
hf = + + +
gπ 2 D15 gπ 2 D25 gπ 2 D35 gπ 2 D45

PIPES IN PARALLEL
The below figure shows the arrangement of pipes in parallel. As it can be seen from the figure,
the pipes are parallel to each other.

Fig. 7 Pipes in Parallel Arrangement

In this arrangement, the length of the each pipe is same and discharge is distributed in the
parallely connected pipes a shown in figure.

Let D1, D2, D3be the diameter of the pipe.

Let L1=L2=L3= L, is length of Pipe connected in parallel
The (hf)1, (hf)2, (hf)3 are the head loss due to friction for each pipe.
The condition for the parallel pipe is
hf = hf)1= (hf)2= (hf)3
i.e.,

Or

From continuity equation Q= Q1+Q2+Q3

EQUIVIVALENT PIPE
In practice adopting pipes in series may not be feasible due to the fact that
• they may be of not standard size (ie. May not be comemercially available)
• they experience other minor losses.
Hence, the entire system will be replaced by a single pipe of uniform diameter D, of the same
length, L=L1+ L2+ L3. This pipe is called as Equivalent Pipe. The below figure shows
representation of equivalent pipe.

Q
D1 D2 D3 == Q D
L1 L3
L= L 1 +L 2 +L 3
Fig. 8 Equivalent Pipe

In the equivalent pipe system the different diameter will be replaced by a single pipe of
uniform diameter D, but of the same length L=L1+ L2+ L3 such that the head loss due to
friction for both the pipes, viz equivalent pipe & the compound pipe are the same.

For a compound pipe or pipes in series.

h f = hf1 + hf 2 + hf3

…………………7
for an equivalent pipe,

…………………..8
Equating (7) & (8) and simplifying;

L L L L
5
= 15 + 25 + 35
D D1 D2 D3
1
Or
 5
 L 
D=
L L L 
 15 + 25 + 35 
 D1 D2 D3 

PROBLEMS

1. Find the diameter of a Galvanized iron pipe required to carry a flow of 40lps of water, if the
loss of head is not to exceed 5m per 1km. Length of pipe is 1km, Assume f=0.02.

Solution:-
D=?, Q=40lps = 40x10-3 m3/s
hf=5m, L=1km = 1000m. f=0.02
Darcy’s equation is
 8 fLQ 2 
∴D =  2 
 gπ h f 
1
 8 x0.02 x1000 x(40 x10−3 ) 2  5
∴D =  
 9.81xπ 2 x5 

D − 0.22m = 220mm
 Problem-2
Two tanks are connected by a 500mm diameter 2500mm long pipe. Find the rate of flow if the
difference in water levels between the tanks is 20m. Take f=0.016. Neglect minor losses.

Solution:-
Applying Bernoulli’s equation between (1) & (2) with (2) as datum & considering head loss due
to friction hf only,

p1 V12 p V2
Z1 + + = Z 2 + 2 + 2 + h f − − − (1)
γ 2g γ 2g
Z1 = 20m, Z2 = 0 (Datum);
V1=V2 = 0 (tanks are very large)
p1=p2=0 (atmospheric pressure)
Therefore From (1)
20+0+0 = 0+0+0+hf
Or hf = 20m.
But
8 fLQ 2
hf =
gπD 5
1
 20 x9.81xπ 2 x0.55  2
Q= 
 8 x0.016 x 2500 

Q = 0.4348m3 / sec = 434.8lps

Problem-3

An existing pipe line 800m long consists of four sizes namely, 30cm for 175m, 25cm dia for the
next 200m, 20cm dia for the next 250m and 15cm for the remaining length. Neglecting minor
losses, find the diameter of the uniform pipe of 800m. Length to replace the compound pipe.

Solution

L=800m
L1=175m D1=0.3m
L2=200m D2=0.25m
L3=250m D3=0.20m
L4=175m D4=0.15m

For an equivalent pipe

L  L1 L2 L3 L4 
= + + + 
D 5  D15 D25 D35 D45 

1
 5
 L 
D=
L L L 
 15 + 25 + 35 
 D1 D2 D3 

1
 5
 800 
∴D =  
  175 + 200 + 250 + 175  
  0.35 0.255 0.25 0.155  

D = Diameter of equivalent pipe = 0.189m less than or equal to 19cm.

Problem-4

Two reservoirs are connected by four pipes laid in parallel, their respective diameters being d,
1.5d, 2.5d and 3.4d respectively. They are all of same length L & have the same friction factors f.
Find the discharge through the larger pipes, if the smallest one carries 45lps.

Solution:-
D1=d, D2 =1.5d, D3=2.5d, D4=3.4d
L1=L2=L3=L4= L.
f1=f2=f3=f4=f.
Q1=45x10-3m3/sec, Q2=? Q3=? Q4=?

For pipes in parallel hf1=hf2=hf3=hf4 ,i.e.

Q12 Q22 Q32 Q42

= = =
D15 D25 D35 D45
1
 1.5d 5

−3 2 
2
Q2 =  ( )
 x 45 x10  = 0.124m / sec
3

 d   1
 2.5d  5
−3 
 2
Q2 =   x(45 x10
2
) 3
 = 0.4446m / sec
 d  

1
 3.4d 5 2
2
Q2 =  (
 x 45 x10
−3
) 3
 = 0.9592m / sec
 d  
 MINOR LOSSES
Minor losses in a pipe flow can be either due to change in magnitude or direction of flow.
They can be due to one or more of the following reasons
i)Entry loss
ii)Exit loss
iii)Sudden expansion loss
iv)Sudden contraction loss
v)Losses due to pipe bends and fittings
vi)Losses due to obstruction in pipe.

Fig. 8 Minor Losses

Head loss for inlets, outlets, and fittings will be in the form of:

where K is a parameter that depends on the geometry.

For a well-rounded inlet, K = 0.1, for abrupt inlet K = 0.5
(much less resistance for rounded inlet).
ENTRANCE LOSSES
 Rounded inlet

Sharp-edged inlet
Fig. 9 Enrance Conditions and Losses

Fig. 10

For solving problems : 0.5V 2

hL entry =
2g
Exit Loss

Fig. 10
For solving problems : V2
hL exit =
2g

Losses also occur because of a change in pipe diameter as is shown in Figures. The sharp-edged
entrance and exit flows are limiting cases of this type of flow with either A1/A2 = ∞ or A1/A2 = 0,
respectively.

Fig. 11 Sudden Contraction Fig. 12 Sudden Expansion

Loss coefficient for a sudden contraction, KL=hL/(V2/2g), is a function of the area ratio, A2/A1,

For solving problems : V22

hL = 0.5
2g
Suddden Expansion

Fig. 13

Flow in a sudden expansion is similar to exit flow.

Referring Fig., the fluid leaves the smaller pipe and initially forms a jet-type structure as it
enters the larger pipe.
Within a few diameters downstream of the expansion, the jet becomes dispersed across the pipe.

In this process of dispersion [between sections (2) and (3)], a portion of the kinetic energy of the
fluid is dissipated as a result of viscous effects.
Loss coefficient for sudden expansion can be obtained by means of a simple analysis based on
continuity and momentum equations for the control volume shown in figure

Assumption: Flow is uniform at sections (1), (2), and (3) and the pressure is constant across the
left-hand side of the control volume (pa = pb = pc = p1)
Consider the sections as shown in figure
P1 & P2 are the pressure acting at (1) (1) and (2) (2)
From experiments, it is proved that pressure P1 acts on the area (a2 – a1) i.e. at the point of
sudden expansion.
V1 and V2 are the velocities.
From II Law of Newton Force = Mass x Acceleration.
The forces acting on the control volume (LHS)

∑ forces = + p a 1 1 − p2 a2 + p1 (a2 − a1 ) − − − (i )

or , ∑ forces = a2 ( p1 − p2 ) − − − (ii )

RHS of Neton’s second law,

Mass x acceleration = ρ x vol x change in velocity /time

= ρ xvolume/time x change in velocity

= ρxQx(V1 − V2 ) − − − (iii )

Substitution (ii) & (iii) in newton’s Equation

a2 ( p1 − p2 ) = pQ(V1 − V2 )

Divide both sides by sp.weight

 p − p2  V2 (V1 − V2 )
∴  1  = − − − (iv)
 γ  g

Applying Bernoulli’s equation between (1) and (2) with the centre line of the pipe as datum and
considering head loss due to sudden expansion hLonly.
p1 V12 p V2
Z1 + + = Z 2 + 2 + 2 + hL
γ 2g γ 2g
Z1 = Z2 because pipe is horizontal

 p − p 2  V12 − V22
∴  1  + ( ) = hL
 γ  2g ----------- v

Repalcing (p1-p2)/ɤ by Eq. (iv) in Eq. (v)

hL =
(
2V2 (V1 − V2 ) + V12 − V22 )
2g

2V1V2 − 2V22 + V12 − V22

hL =
2g
2V − 2V1V2 + V12 − V22
2
hL = 2
2g
V22 + V12 − 2V1V2
hL =
2g

hL =
(V1 − V2 )2
2g …..vi
The Equation (vi) represents the loss due to sudden expansion.

Loss of Power
The loss of power in overcoming the head loss in the transmission of fluid is given by

P = γQh f − − − (vi)
PROBLEMS
Problem-1

A 25cm diameter, 2km long horizontal pipe is connected to a water tank. The pipe discharges
freely into atmosphere on the downstream side. The head over the centre line of the pipe is
32.5m, f=0.0185. Find the discharge through the pipe

Solution:

Applying Bernoulli’s equation between (A) and (B) with (B) as datum & considering all losses.

PA VA2 p v2
ZA + + = Z B + B + B + entryloss + frictionloss + exitloss
γ 2g γ 2g
The tank surface and the outlet are exposed to atmospheric condition and hence, PA = PB .
When tank area is compared with the pipe area, it is very much grater than the pipe and hence the
variation of velocity in the tank can be neglected. Therefore, VA = 0.
The above equation now can be written as,

V 2 0.5V 2 fLV 2 V 2
32.5 + 0 + 0 = 0 + 0 + + + +
2g 2g 2 gD 2 g

V2  0.0185 X 2000 
32.5 = 1 + 0.5 + + 1
2g  0.25 

32.5 = 7.67V 2
V = 2.06m / s
The discharge is calculated using continuity equation.
πD 2
Q= V
4
2
πx 0.25 4 x 2.06 = 0.101m 3 / sec
Q = 101lps
Problem-2

The discharge through a pipe is 225lps. Find the loss of head when the pipe is suddenly enlarged
from 150mm to 250mm diameter.

Solution:
D1=0.15m, D2 = 0.25m Q=225lps = 225m3/sec

Head loss due to sudden expansion is

hL =
(V1 − V2 )
2g
Writing the above equation in terms of discharge,,

2
 4Q 4Q  1
hL =  2 −  X
2 
 πD1 πD2  2g
2
16Q 2  1 1 
hL =  2 − 2 
2 
2 gπ  D1 D2 

2
16 x0.225 2  1 1 
2  
= −
2 x9.81xπ  0.15 0.252 
2

hL = 3.385m
Problem-3
The rate of flow of water through a horizontal pipe is 350lps. The diameter of the pipe is
suddenly enlarge from 200mm to 500mm. The pressure intensity in the smaller pipe is 15N/cm2.
Determine (i) loss of head due to sudden enlargement. (ii) pressure intensity in the larger pipe
(iii) power lost due to enlargement.

Solution:
Q=350lps=0.35m3/s
D1=0.2m, D2=0.5m,
P1=15N/cm2
hL=?, p2=?, P=?

From continuity equation

4Q 4 x0.35
V2 = = = 1.78m / s
πD22 πx0.52

hL =
(V1 − V2 ) = (11.14 − 1.78)2 = 4.463mofwater
2g 2 x9.81

Applying Bernoulli’s equation between (1) (1) and (2) (2) with the central line of the pipe as
datum and considering head loss due to sudden expansion hL only,

p1 V12 p V2
Z1 + 150+ 11=.14 Z 2 2+ 2 + p2 + h1L.782
0 + γ 2+g =γ0 + 2 g2 + + 4.463
9.81 19.62 9.81 19.62
Z1 = Z 2 = 0( pipehorizontal )
p2 = 166.68kN / m 2 = 16.67 N / cm 2

Power Loss;

P = γQhL
= 9.81x0.35 x 4.463
P = 15.32kW
Problem-4
Two reservoirs are connected by a pipe line which is 125mm diameter for the first 10m and
200mm in diameter for the remaining 25m. The entrance and exit are sharp and the change of
section is sudden. The water surface in the upper reservoir is 7.5m above that in the lower
reservoir. Determine the rate of flow, assuming f=0.001 for each of the types.

Solution;

From continuity equation

πx0.1252 πx0.2 2
V1 = V2
4 4
∴V1 = 2.56V2
pA VA2 p B VB2
ZA + + = ZB + + + entryloss + frictionloss + sudden exp ansionloss + frictionloss + exitloss
γ 2g γ 2g
The tank surfaces are exposed to atmospheric condition and hence, P1 = P2 .
When tank area is compared with the pipe area, it is very much grater than the pipe and hence the
variation of velocities in the tanks can be neglected. Therefore, V1 = V2 =0.
The above equation now can be written as,

 0.5V12 fL1V12 (V1 − V2 )

2
fL V 2 V 2 
7.5 + 0 + 0} = 0 + 0 + 0 + + + + 2 2 + 2 
 2g 22g 2g 2g 2g
(
 0.5 2.56V22 )
0.01x10 x(2.56V2 ) (2.56V2 − V2 )2 0.01x 25 x(V2 )2 V22 
7.5 + 0 + 0} =  + + + + 
 2g 2g 2g 2g 2g 

147.15 = 3.2768 (V2)2 + 0.65536(V2)2 + 2.4336 (V2)2 + 0.25(V2)2 + (V2)2

147.15 = 7.61576 (V2)2

V2 = (147.15/7.61576)0.5

V2 = 4.4 m/s
Q= ( π(0.2)2/4)x4.4 = 0.138 m3/s
WATER HAMMER
Water Hammer Phenomenon in pipelines
A sudden change of flow rate in a large pipeline (due to valve closure, pump turnoff, etc.)
involve a great mass of water moving inside the pipe. The force resulting from changing the
speed of the water mass may cause a pressure rise in the pipe with a magnitude several times
greater than the normal static pressure in the pipe. This may set up a noises known as knocking.
This phenomenon is commonly known as the water hammer phenomenon

The excessive pressure may fracture the pipe walls or cause other damage to the pipeline system
Has a very high speed (called celerity, C ) which may reach the speed of sound wave and may
create noise called knocking,
Magnitude of this pressure depends on
(i) The mean pipe flow velocity
(ii) The length of the pipe
(iii) The time taken to close the valve and
(iv) The elastic properties of the pipe material and that of water.

Some typical damages

Burst pipe in power Pipe damage in

sation Big Creek #3, USA power station Okigawa
Sudden rise in pressure in the pipe due to the stoppage of the flow generating a high pressure
wave, which will have a hammering effect on the walls of the pipe, is known as Water Hammer.

Water Hammer Phenomenon in pipelines

Consider a long pipe AB: Connected at one end to a reservoir containing water at a height H from
the center of the pipe. At the other end of the pipe, a valve to regulate the flow of water is
provided.

Kinetic energy of the water moving through the pipe is converted into potential energy stored in
the water and the walls of the pipe through the elastic deformation of both.
The water is compressed and the pipe material is stretched.
The following figure illustrates the formation and transition of the pressure wave due to the
sudden closure of the valve
Propagation of pressure wave due to valve closure

a. Steady state prior to valve closure

\\

b. Rapid valve closure – pressure increase,

pipe walls expand, liquid compression;
transient conditions propagate upstream

c. End of step 1 transient process

d. Pipe pressure > tank pressure water

flows from pipe to tank relieving
pressure in pipe

e. Process starts at tank and continues

to valve, time L/c, total time 2L/c =>
water hammer period
\

f. Wave of backwater cannot go past

the valve, starts wave of negative
pressure toward tank
 g & h. Pressure difference causes
water to flow toward valve

i. One full cycle, 4L/c

Analysis of Water Hammer Phenomenon

The pressure rise due to water hammer depends upon:
(a) The velocity of the flow of water in pipe,
(b) The length of pipe,
(c) Time taken to close the valve,
(d) Elastic properties of the material of the pipe.
The following cases of water hammer will be considered:
Gradual closure of valve,
Sudden closure of valve and pipe is rigid, and
Sudden closure of valve and pipe is elastic.

The time required for the pressure wave to travel from the valve to the reservoir and back to the
valve is:
2L
t=
C
Where:
L = length of the pipe (m)
C = speed of pressure wave, celerity (m/sec)
If the valve time of closure is tc , then

If 2L the closure is considered gradual

tc f
C
If 2L the closure is considered sudden
tc ≤
C

The speed of pressure wave “C” depends on :

the pipe wall material.
the properties of the fluid.
the anchorage method of the pipe.

K
C= if the pipe is rigid
ρ

Ec
C=
ρ if the pipe is elastic

1 1 Dk
and = +
Ec K E p e
Where:
C = velocity (celerity) of pressure wave due to water hammer.
ρ = water density ( 1000 kg/m3 ).
K = bulk modulus of water ( 2.1 x 109 N/m2 ).
Ec = effective bulk modulus of water in elastic pipe.
Ep = Modulus of elasticity of the pipe material.
e = thickness of pipe wall.
D = diameter of pipe.
k = factor depends on the anchorage method:
= 5 for pipes free to move longitudinally,
( −ε )
4

= (1−ε 2 ) for pipes anchored at both ends against longitudinal movement

= (1−0.5ε ) for pipes with expansion joints.

where ε = poison's ratio of the pipe material (0.25 - 0.35). It may take the value = 0.25 for
common pipe materials.
Maximum pressure created by the water hammer
Instantaneous rise in pressure in a pipe running full due to Gradual closure of
valve
Consider a pipe AB of length L connected to a tank at A and a valve at B with water flowing in it
as shown in Fig.
Let V be the mean flow velocity
a is the flow cross-sectional area,
p the instantaneous rise in pressure due to gradual closure of valve
t be the actual time of closure of valve.

From Newton’s second law of motion,

the retarding force generated against the flow direction is given by the rate of change momentum
of the liquid along the direction of the force.

Retardation of water = Change in velocity / Time =

( V − 0) = V
t t
 V
Retarding force = Mass of water x Retardation = ρaL
t
The force generated due to pressure wave = Pressure intensity x area = pi × a ……….…2

From Eqs. 1 and 2, we get

V
pi a = ρ a L
t
Hence the increase in pressure rise due to gradual closure of valve:
ρ LV
pi =
t
where,
V = initial velocity of water flowing in the pipe before pipe closure
t = time of closure.
L = length of pipe.
ρ = water density.

The pressure head caused by the water hammer is

pi ρ LV LV
H= = =
ρ g ρ gt gt

OR
∆P ρ LV LV
∆H = = =
γ ρ gt g t

Instantaneous rise in pressure in a pipe running full due to Sudden

closure of valve when the pipe is rigid tc ≤ 2L
C
When the valve provided at the downstream end is closed suddenly and the pipe is rigid, then the
converted pressure energy from the kinetic energy due to closure is to be absorbed by the fluid
due to its compressibility only.

i.e. Ek = Ew ….1
Consider the pipe AB of length L
cross-sectional area a
water of mass density ρ,
weight density γ
bulk density of water K
mean velocity of flow V
be suddenly stopped due to closure of valve provided at B.

The kinetic energy of flowing water before closure of valve will be converted to strain energy,
when the effect of friction and elasticity of pipe material are ignored.
Loss of kinetic energy Ek
= ½ x mass of water x V2
As mass = ρ x volume = ρ x aL
Loss of kinetic energy, Ek = ½ x ρ a L x V2 …..2
1  pi2  1  p2 
Gain in strain energy Ew =   × Volume =  i  × aL …………3
2 K  2 K 
From Eqs. 2 and 3, we get
 p2 
ρ a L V 2 =  i  × aL
K 
OR
p i2 = ρ K V 2
OR
pi = V ρ K

But Celerity K
C=
ρ
Substituting for the value of C in the above equation for pressure rise, we get
pi = ρ V C

CV
∆H =
g
 K V K
But for rigid pipe C= so ∆H =
ρ g ρ

Instantaneous rise in pressure in a pipe running full due to Sudden

closure of valve when the pipe is elastic
When the valve provided at the downstream end is closed suddenly and the pipe is elastic, then
the converted pressure energy from the kinetic energy due to the valve closure is to be absorbed
by both the fluid due to its compressibility and the elasticity of the pipe.
i.e. Ek = Ew+Ep …………….1

Ek and Ew can be computed as in the previous derivation.

Ek = ½ x ρ a L x V2 ……………2

1  pi2  1  pi2 
Ew =   × Volume =   × aL
2 K  2 K 
Computation of Ep can be done by simulating the situation to the thick cylinder subjected to
internal fluid pressure.
Let t be the thickness of the elastic pipe wall and assume that it is small compared to its diameter
D.
Let f1 be the hoop or circumferential and stress
f2 be the longitudinal stress as shown in figure.

Let the Young’s modulous of the pipe material be E and poisons ratio 1/m
Let the instantaneous fluid pressure be pi.
From the knowledge oaf Strength of materials, we can write that
pi D pD
f1 = and
f2 = i Hence f1 = 2 f2
2t 4t
Further, the strain energy stored in pipe per unit volume is given by
Ep 1  2 2f f 
=  f1 + f 22 − 1 2 
V1 2E  m 
Substituting f1 = 2 f2, we get
Ep 1  2 2 4 f 22 
= 4
 2 f + f − 
V1 2E 
2
m 

Ep f 22  4 
= 5−
V1 2 E  m 

Substituting for f2, and V1 = πDtl we get

pi2 D 2 1  4  ……………..4
Ep = 5− π Dtl
16 t 2 2 E  m 
Substituting Eqs. 2,3 and 4 in Eq. 1,

1 1  p2  p2D2 1  4 
× ρ a L ×V 2 =  i  × aL + i 2 5− π Dt L
2 2 K  16 t 2 E  m 
Simplifying, we get
1 1  pi2  pi2 D 1  4  π D 2
× ρ a L ×V =   × aL +
2
5− L
2 2 K  4 t 2 E  m  4
π D2
But a = and aL/2 gets canceled on both sides, now the equation takes the form as,
4
 p2  p2D 1  4  1  D  4 
ρ V 2 =  i  + i 5 −  = pi2   +  5 − 
K
  4 t E  m   K  4 t E  m 

Solving for pi ;

ρ ρ
pi = V =V
 1  D  4   1  D  5 1 
 K  + 4 t E  5 − m   K  + t E  4 − m 
       
The above expression gives the instantaneous rise in pressure in an elastic pipe due to sudden
closure of Valve.

If the Poisons ration is not given, it can be assumed as ¼. Then Eq. 5reduces to

ρ
pi = V
1 D 
K + t E  ………………6
 
Applying the water hammer formulas we can determine the energy gradient line and the
hydraulic gradient line for the pipe system under steady flow condition.

So the total pressure at any point M after closure (water hammer) is

PM =PM ,beforeclosure +∆P

Or
H M =H M ,beforeclosure +∆H

PROBLEM
A steel pipe 1524 m long laid on a uniform slope has an 45.72 cm, diameter and a 5.08 cm wall
thickness. The pipe carries water from a reservoir and discharges into the air at an elevation
45.72 m below the reservoir free surface. A valve installed at the downstream end of the pipe
allows a flow rate of 0.708 m3/s. If the valve is completely closed in 1.4 sec, calculate the
maximum water hammer pressure at the valve. Neglect the longitudinal stress.
Take K= 2.07 x 1011 N/m2, Ep = 1.93 x 1011 N/m2

Solution:
Effective bulk modulus of water In elastic pipe, Ec , is given by

1/Ec =(1/ 2.07 x 1011 ) + (0.4572/(1.93 x 1011 x0.0508))

Hence
Ec = 1.888 x 109 N/m2
Ec
C= = (1.888 x 109 /1000)0.5 = 1374.045m/s
ρ

The required for the wave to return to the value, tc = (2L/C)

tc = (2x 1524/1374.045) = 2.22 sec
But the time of closure of valve = 1.4 sec
There fore, t< tc . Hence it is a case of instantaneous closure.
Since the longitudinal stresses are neglected the case is of rigid pipe.
The instataneous increase in pressure is given by pi = ρ V C.
The velocity, v= Q/A = 0.708/(π x(0.4572)2/4) = 4.312 m/s.
Therefore,maximum water hammer pressure = 1000x4.312x1374.045 = 59235508 N/m2