Week-4

WATER DISTRIBUTION:
The whole system which gets water from the source and provides it to the consumer in an
acceptable quality, including the source is termed as the water distribution system.
Water distribution systems are designed to adequately satisfy the water requirements for a combination
of domestic, commercial, industrial and firefighting purposes. A distribution system may therefore
consist of pipelines of various sizes for carrying water to the streets, Valves for controlling the flow in
the pipes. Hydrants are provided for fire protection and may also be used to release air at high points
and blow off sediments at low points in the system. Service connections to the individual homes are
provided and pumps are provided for lifting and forcing water into the distribution mains.
A complete water distribution system includes the following components.

1) Source:
The point from which the water is fed into the system is the source. It may be a well, pond,
lake, reservoir or a river, whichever is suitable.

2) Intake:
Wells and springs may be considered as intakes and special intake structure will be required
for lakes, reservoirs and rivers. The intake normally consists of an opening (frequently screened in
some manner) through which the water enters, and the conduit conveying the water by gravity to a
well or sump. From the well the water is pumped to the mains or treatment plants.

Typical Intake

3) Adequate Pipeline or Open Channel:

It will be needed to convey the water if the city is long distance from the source.
4) Treatment Plant:
If the water from the source consists of some impurities which makes it unfit for its intended
use than it is fed into a treatment plant before provision to the consumer so that the impurities can
be removed from it.
 5) Pumping Station:
To generate sufficient head to force the water through the network of street mains (in most
cases) pumping stations are required. Some distribution systems depend upon the gravity and no
pumps are required.

Pumping Station

6) Pipelines:
Pipelines of different material are used which consists of reinforced concrete, asbestos cement,
Iron pipes (Cast iron, ductile iron), steel and plastic etc. Mostly iron pipes are used. Selection of the type
of material of the pipe and its size depends on carrying capacity, durability, maintenance cost and first
cost.
The stresses which are induced in the pipes are due to the internal pressure of the water, Water
hammer, Change in direction and temperature (thermal stresses).

Pipelines

7) Connections:
Pipes are combined through connections which may be:
Joints: To assemble small pipes to make them long. Types of joints used are bell and spigot.
Fittings: Changed in direction of flow are made through fittings. The standard CI fittings are given below:
1/4 bend used for 90o change in direction.
 1/8 bend used for 45o change in direction.
1/6 bend used for 22.5o change in direction.
Some other types of connections used are Y-branch, Cross, Tee, Reducer.

Connections

8) Valves:
Different types of valves used are given below:
Gate valve: It is used for on/off services.

Check valve: It permit the water to flow in only one direction, is used to prevent the reversal of flow
when the pump is shut off.
Foot valves: They are check valves installed at the end of a suction line to prevent drainage of suction
line and loss of prime when the pump is shut down.
Air Vacuum and Air-Relief Valves: They are provide in long pipe lines to
permit release of air which accumulates at high points and to prevent
negative pressures from building up when the lines are drained. They are
automatic in operation, and open to release the accumulated air while
close when the pipe is full of water.
REQUIREMENTS OF A DISTRIBUTION SYSTEM:
Various requirements for the proper functioning of a distribution system are:
 It should be capable to supply water at all the intended places within the city at a
reasonably sufficient head.
 It should be capable of supplying the requisite amount of water for firefighting during
such needs.
 It should be cheap with least capital construction cost. The economy in cost of
distribution system is important because the distribution system is the costly item in the
entire water supply scheme.
 It should be simple and easy to install, operate and repair in order to keep minimum
capital, operational and maintenance cost.
 It should be safe against any possible pollution of water. This may be achieved by
keeping water pipelines above and away from the sewerage drainage lines by sufficient
amount and also by improving the general sanitary conditions of the area through which
the distribution pipes have to pass.
 It should be safe as not to cause the failure of pipelines by bursting etc.
 It should be water tight as to keep the water losses due to leakage to minimum.

Methods of Distribution system:

1) Gravity Distribution System:
 This is possible only when the source of supply is located substantially above the level of
the city so that sufficient pressure can be maintained in the main for domestic and fire
services.
 This is the most dependable technique, provided there are multiple well-protected
conduits carrying the flow to the community.
 High pressure for firefighting may require use of motor pumping trucks of fire
department.
 Low lying areas may be separated to prevent excessive pressures.

2) Pumping Without Storage:

 In this method the pump forces water directly into the mains.
 It is the least desirable method, for a power failure will mean complete interruption in
the water supply.
 Since the flow must be constantly varied to match an unpredictable demand, hence
sophisticated control systems are required (require constant attendance).
 Peak water use and thus peak power consumption are likely to coincide with period of
already high power use thus increasing power cost.
 Systems of this kind have the advantage of permitting increased pressure of firefighting
hence individual users must be protected by pressure reducing valves.

3) Pumping With Storage:

 In this method the excess water pumped during the period of low consumption is stored
in elevated tanks and reservoirs.
  During periods of high consumption, the stored water is drawn from the tank to
augment to that pumped.
 This method allows fairly uniform rate of pumping.
 Since the water stored furnishes a reserve to care for fires and pump breakdown, this
method of operation is fairly reliable.
 Motor pumps must ordinarily be used for high firefighting pressures.
 It may be economical in some cases to pump only during off-peak hours to minimize
power cost.
 It is also known as fill and draw system.


Pipe System
 The network of pipes which makes up the distribution system consists of primary or
arterial lines (Supply mains), secondary lines (sub mains), and distribution mains
(branches and laterals).
 The water mains and sub mains are usually laid sloping from the high level to the low
level areas, so as to achieve the maximum advantage of the available head, and thus
keep their size minimum.
 The small distribution mains form a grid over the entire service area, supplying water to
every user and the fire hydrants.
 The distribution pipes are generally laid on the one side of the road and the streets,
usually below the footpaths and 2m above and 3m away from the sewer so as to avoid
any future contamination of water through the leaky joints.
 Valves are placed along the straight length of the pipes at suitable intervals, and also at
all the junctions and branching of points so as to control the flow of water into different
sections.
 Drain valves are placed at all the low points while air valves are placed at all the high
points.
 Velocities at maximum flow including fire flow normally do not exceed 1m/s (3ft/sec) in
the immediate vicinity of the large fires.
Sizes:
Sizes of small distribution mains is seldom less than 150mm (6in) with cross mains
located at interval of more than 180m (600ft).
In high value district main size is 200mm (8in) with the same spacing. Major streets are
provided with mains not less than 12in in diameter. Lines for domestic flow are provided as
small as 4in in diameter but should not exceed 1300ft in length for dead end system or 2000ft
for systems connected at both ends.
For small communities 2in to 3in diameter pipes are used with length not to exceed
300ft if dead ended and 600ft if connected at both ends.
LAYOUT OF DISTRIBUTION SYSTEM:
Generally following are the layouts of the distribution networks.

1) Dead End System (Tree system):

It consists of:
 One main supply pipe.
 A number of sub mains originate from the main pipe.
 Each sub main divides into a number of branches called laterals.
 From laterals service connections are provided to the consumers.
It is suitable for localities which expands irregularly and where the water pipes have to be laid
randomly due to the absence of any planned full-fledged road network.
 Advantages:
1) Analysis design is easy.
2) Discharge and pressures are calculated at different places easily and correctly.
3) Less number of cut off valves (sluice valves) is required.
4) Shorter pipe lengths are needed and pipes are easily laid.
5) Cheap and simple, and can be extended or expanded easily.
 Disadvantages:
1) Any damage or repair will disrupt supply in the area fed by that pipe.
2) Stagnant water at dead end as there is no free circulation of water, causing color, taste and
odor. To check, periodic removal of water is required.
3) Supply from one direction cannot be increased for fire flow.

2) Grid-Iron System:
It consists of:
  M = One main supply pipe.
 S = A number of sub mains.
 L = Laterals
It is also known as interlaced or reticulation system. The M, S, and L are all interconnected with
one another. It is suitable for well-planned cities.
 Advantages:
1) The water reaches at the same point through more than one route. The discharge is carried
by each pipe hence the friction losses and the size of the pipe gets reduced.
2) During any damage or repair very small area will be devoid of supply.
3) It is not liable to pollution due to stagnation.
4) During fire, more water can be diverted to a particular direction.
 Disadvantages:
1) More length of pipelines and more sluice valves are required.
2) Its construction is costlier.
3) Design is complicated.

3) Ring System (circular system):

It consists of:
 A closed ring either circular or rectangular.
 A main pipe is formed around the area to be served.
 Distribution area divided into circular or rectangular bocks, on the periphery of which is
the main pipe.
It is suitable for cities and towns having well-planned roads. It has the same advantages and
disadvantages as that of grid-iron system.
4) Radial System:
It consists of:
 Water mains.
 Water is pumped from the water mains into the distribution reservoirs.
 Which is then supplied through radially placed distribution pipes.
It is suitable for cities and towns having radial roads emerging from different centers.
 Advantages:
1) Ensure high pressure and efficient water distribution.
2) Calculations for the design sizes are also simple.
DESIGN OF WATER DISTRIBUTION SYSTEMS:
By the design we mean to calculate pressures that exist at various points in the system under
conditions of operation and to provide adequate size of the pipes. The size of distribution pipes mainly
depends upon the amount of flow to be carried and the permissible losses in the pressure head.
Pressures maintained in the distribution system are to be adequate to cope with the demand of
community as well as the fire demand.
Pressure in the range 150 to 300 Kpa (20 to 40 lb/in 2) are adequate for normal use and may be
used for fire supply in small towns.
AWWA recommends a normal head of 400 to 500 Kpa (60 to 75 lb/in2) for buildings up to 10
stories in height for residential/ commercial districts where direct streams of flow are used for fire-
fighting.
Factors that cause the loss of head or pressure include:
1) Size of pipe
2) Rate of flow
3) Friction (C value)
The detailed design of water distribution systems are affected by Local topography, existing and
expected population densities, and commercial and industrial demand.
Following considerations should be kept in mind while designing a distribution system:
1) The flow must be disaggregated to individual subareas of the system
2) The disaggregated flows are then assigned to individual nodes of the system.
3) A system of interlocking loops is laid out.
4) Determine the size of various pipes to ensure that the pressure and velocities desired in the
system are maintained under a variety of conditions. These design conditions are based on
the maximum daily flow rate plus fire demand.
The usual approach for the design of looped pipe system involves:
1) Layout of the network.
2) Assignment of estimated pipe sizes.
3) Calculation of resulting flows and head losses.
4) The pipe sizes are then adjusted as necessary to ensure that the pressure at the various
nodes and the velocities in the various pipes meet the criteria needed for the community.
 The different methods used for the design of water distribution systems are described below

1) Hazen-Williams Formula:
It is used to predict head losses.
0.63 0.54
V =KC R S
Where
K is a factor dependent on units= 0.849 for SI
= 1.318 for fps
C is a constant dependent on the relative roughness of the pipe

Various other formulas for pipe flow are available to predict head losses as a function of velocity
in pipes but Hazen Williams is mostly used in design of circular pipes.
0.63 0.54
V =0.849C R S
0.63 0.54
D H π 2 D H
As R= and S= Then Q= AV = D (0.849C ( ) ( ) )
4 L 4 4 L
¿ ≫ H=10.68 ¿ For SI units
And H=4.72 ¿ for fps Units
Design Procedure for dead end systems:
1) A number of pipes are connected to the main pipe to form the topology of a tree assuming
that the flow is from the main pipe into the laterals.
2) Calculate the flow rate in any pipe as the sum of downstream flow rate by proceeding from
the outmost branches towards the main pipe.
3) From the flow in each pipe calculate the head loss in them by using the Hazen William’s
formula of head loss.
4) By subtracting the head losses from the available head at the input point, the head or
pressure from any point In the system ca be determined by:
Min. height at the input point required=Max. Head loss + Min. residual pressure at the
outlet.
Example: Analysis of dead end system is shown in the figure. Input
head at point 1 is 21m, C value in 100, Minimum residual pressure is
14m.

Pipe Dia. L(m) Flow Velocity Head loss

(mm) (m2/s) (m)
5-4 203 305 1.515 0.0252 1.586
9-4 203 305 1.135 0.0189 1.372
4-3 254 305 2.65 0.0441 1.676
8-3 203 305 1.135 0.0189 1.372
3-2 305 457 3.785 0.0630 1.829
6-2 203 305 1.514 0.0253 1.586
7-2 203 305 1.514 0.0252 1.586
2-1 405 610 6.813 0.1135 1.829

Head at point A = 21m

Head at point B = head at point A – Head loss in BA
Head at point B = 21-1.829 = 19.171 m
Head at point C = 19.171 – 1.829 = 17.342 m
Head at point D = 17.342 – 1.676 = 15.667 m
Head at point E = 15.667 – 1.586 = 14.080 m
Head at point F = 19.171 – 1.586 = 17.585 m
Head at point G = 19.171 – 1.586 = 17.585 m
Head at point H = 17.342 – 1.372 = 15.970 m
Head at point I = 15.666 – 1.372 = 14.294 m

2) Hardy Cross Method:(The Hardy Cross method is an iterative

method for determining the flow in pipe network systems
where the inputs and outputs are known, but the flow inside
the network is unknown. ... The introduction of the Hardy
Cross method for analyzing pipe flow networks
revolutionized municipal water supply design).
 It is used to analyze the pipe network of Grid-Iron System.
 Assuming a distribution of flow in the network.
 Pipe flow formulas i.e. Chezy, Manning, and Hazen Williams are used to predict the head
losses and successive corrections are made in the flows until the heads are essentially
balanced.
 The loss in any element of the system may be expressed as
x
hi =k i Qi −−−−−−−−1
Where
hi = energy loss in the element i.
k i= constant depending on the pipe diameter, length type and condition
Qi= flow in the element
x = 1.85 to 2.0
For any pipe in a loop of the system, the actual flow will differ from an assumed flow by
an amount∆ ,
Qi=Qio +∆−−−−−−−2
 Where
∆ = Required correction
Qio= Assumed flow
Qi= Actual flow in pipe elements
Form equation 1 and 2
x x
hi =k i Qi =k i ( Qio + ∆ ) =k i ¿

The remaining terms in the equation can be neglected if ∆ is small compared to Q i.

For any loop the sum of the head losses about the loop must be equal to zero. Thus for
any loop:
n

∑ k i Qix =0−−−−−−−−−4
1

Where n is the number of pipes in the loop.

Thus from equation 4 we get:
n n n n

∑ hi=∑ k i Qix =∑ k i Qiox+∑ x k i Qiox−1 ∆=0−−−−5

1 1 1 1

Equation 5 may then be solved for corrections.

n n
−∑ k i Qiox − ∑ hi
1 1
∆= n
= n
hi
∑ x k i Qiox−1 x∑
Qio
1 1

Procedure for the design of pipe networks by Hardy Cross method:

1) Disaggregate the flow to the various blocks of other subareas of the community.
2) Concentrate the disaggregated flows at the node of the system.
3) Add the required fire flow at appropriate nodes.
4) Select initial pipe size.
5) Assume any internally consistent distribution of flow. The sum of the flows entering and
leaving each node must be equal to zero.
6) Compute the head in each element of the system by H=10.68 ¿ . Conventionally clockwise
flows are positive and produce positive head loss and vice versa.
n n
7) With due attention to sign, compute the total head loss around each loop. ∑ hi=∑ k i Qio
x

1 1
n
8) Compute without regard to the sign, the sum ∑ x k i Qio
x−1

1
9) Calculate the correction (∆ ¿ for each loop and apply it to each line in the loop. Lines
common to two loops receive two corrections with due attention to sign.
 10) Repeat the process until the corrections calculated in step 9 are less than some stipulated
maximum (i.e. 0.8m3/min or 50gpm) the flows and pressure in the initial network are then
known.
11) Compare the pressures and velocities in the balanced network to a criteria (1-2 m/s). Adjust
the pipe sizes to reduce or increase velocities and pressures until a satisfactory solution is
obtained.
12) Apply any other fire flow conditions which may be critical and re-evaluate the velocities and
pressure distribution. Adjust the pipe sizes as necessary.

Example: Analyze the following pipe network by Hardy-Cross method.

Solution:

 Flows for the area has been

disaggregated to the nodes.
 Major fire flow added at node G.
 Water enters the system at node A.
 Pipe diameter is based on flows and
criteria.
 Calculations are continued until the
corrections are less than 0.2m2/min..
n

∑ hi
1
∆= n
hi
x∑
1 Qio

Hardy-Cross analysis 1st correction:

Loop-1
 Line Flow Dia. Length S hi h/Q
(m3/min) (m) (m) (m)
AB 13 0.4 1250 0.011 13.75 1.058
BN 2 0.25 1100 0.0033 3.63 1.815
HI -9.8 0.3 1000 -0.026 -26.0 2.653
IA -12 0.3 1000 -0.038 -37.80 3.150
Sum -46.42 8.676
−(−46.42 )
∆ 1= =2.9
(1.85 × 8.676 )

Loop-2

Line Flow Dia. Length S hi h/Q

(m3/min) (m) (m) (m)
BE 7.5 0.35 400 0.0075 3.0 0.400
EF 7 0.35 600 0.0066 3.96 0.566
FG 4.7 0.3 1000 0.0067 6.68 1.423
GH -9.3 0.3 1250 -0.0236 -29.54 3.177
HB -12 0.25 1100 -0.0033 -3.63 1.815
Sum -19.53 7.381
− (−19.53 )
∆ 1= =1.4
(1.85 × 7.381 )
Loop-3

Line Flow Dia. Length S hi h/Q

(m3/min) (m) (m) (m)
BC 1.5 0.2 500 0.0058 2.91 1.937
CD 1.0 0.2 400 0.0028 1.10 1.110
DE -0.5 0.2 500 -0.0008 -0.38 0.762
EB -7.5 0.35 400 -0.0075 --3.00 0.400
Sum 0.63 4.209
−0.63
∆ 1= =−0.1
(1.85 × 4.209 )
 Column 5: Slope from Hazen-Williams equation with C=100 and
0.63 0.54
V =0.849C R S
0.63
π 2 D 0.54
Q= AV = D (0.849C ( ) (S) )
4 4
 Column 6: Head loss in each pipe from:
 H
S=
L
¿ ≫ H=S × L
 Column 6 and Column 7 summed and correction calculated from.
n

∑ hi
1
∆= n
hi
x∑
1 Qio
 2nd and 3rd correction should be applied so that the system is balanced.
 The balanced network must be reviewed to ensure that the velocity and pressure criteria are
satisfied.
 If the velocity is different from any suggested criteria than sizes of pipes are revised.
 Pressure at each node must be checked. If it is less than required, revise the pipe sizes
 For pressure calculation if elevation is different at various nodes, the static head must be
included in the calculation.
 Procedure is repeated until adequate pressure and velocity are obtained.