12 – 1

Chapter 12

12.1.1. From example 12.1.1, L = 300 m, n = 0.013, C = 100, Q1 = 1 m3/s, Q2 = 2 m3/s.

D = 1.5 m.
Using Manning’s equation:

(R = D/4 for pipes flowing full)

Using this equation, when Q = 1 m3/s, hL = 0.06 m, and when Q = 2 m3/s, hL =

0.24 m.
Using the Hazen-Williams equation:

When Q = 1 m3/s, hL = 0.089 m, and when Q = 2 m3/s, hL = 0.32 m.

12.2.1. n = 1800 rpm, Q = 180 gpm, H = 80 ft.

When n = 1500 rpm, H - ?, Q - ?
Using the affinity laws,
; thus, 180/Q1500 = 1800/1500.
Q1500 = 150 gpm.

Also, ; thus, 80/H1500 = (1800/1500)2

H1500 = 500/3 = 55.56 ft.

12.2.2. Q = 0.03 m3/s, H = 30 m, n = 1800 rpm = 188.5 rad/sec

The specific speed, ns, is used to select the pump.

Referring to Figure 12.2.5, a radial flow centrifugal pump is appropriate. The

corresponding
Efficiency is about 74 %.

12.2.3. Q = 0.03 m3/s = 1.06 ft3/sec = 475.61 gpm, H = 30 m = 98.43 ft, n = 1800 rpm
 12 – 2

Referring to Figure 12.2.5, a radial flow centrifugal pump is appropriate. The

corresponding efficiency is about 74 %. This result is essentially the same as that
obtained in problem 12.2.2 except that different unit systems are used.

12.2.4. Q = 0.03 m3/s, H = 30 m, n = 2000 rpm.

Referring to Figure 12.2.5, a radial flow centrifugal pump may be appropriate.

The corresponding efficiency is about 76 %.

12.2.5. Q = 0.18 m3/s, C = 140

Suction pipe: D = 300 mm = 0.3 m, L = 5 m
Discharge (delivery) pipe: D = 250 mm = 0.25 m, L = 100 m
Static suction head, Zs = +3 m
patm/ = 9.19 m.
Using equation (12.2.8):

pv/ = 0.24 (from Table 2.1.2)

The Hazen-Williams equation (equation (12.1.1)) is used to compute the losses.

V = 0.894CR0.63Sf0.54, which was expressed more conveniently in example 12.2.4
as
, where is in meters per 1000 m.
Suction piping losses:

= 16.894 m per 1000 m pipe length = 16.894(5/1000) = 0.085 m.

Discharge piping losses:

= 40.959 m per 1000 m pipe length = 40.959(100/1000) = 4.096 m.

Bend loss (discharge piping):
;

0.25(3.67)2/(2(9.81)) (K = 0.25 for bends, Table 12.2.4)

= 0.171 m
Gate valve loss (discharge piping):
= 0.2(3.67)2/(2(9.81)) (K = 0.2 for gate valves)
 12 – 3

= 0.137 m
Gate valve loss (suction piping):

0.2(2.55)2/(2(9.81))
= 0.066 m.
Bell mouth entrance loss:
K = 0.05 (from Table 12.2.4)
= 0.05(2.55)2/(2(9.81))
= 0.017 m
Total minor losses:

Using equation (12.2.8),

NPSHA = 9.19 + 3 – 0.24 – 0.085 – 4.096 – 0.391
= 7.378 m.

12.2.6. Since the only parameter that has changed is Zs, equation (12.2.8) is used to
determine NPSHA.
NPSHA = 9.19 – 3 – 0.24 – 0.0399 – 1.941 – 0.249
= 3.72

12.2.7. H = 60 m, Q = 400 m3/hr, n = 1200 rpm, e = 80%.

Using equation (12.2.11):

From Table 12.2.1, K = 1210

= 18.55

12.2.8. H = 20 m, Q = 400 m3/hr = 1/9 m3/s, n = 1200 rpm, e = 80%

K = 1210 (from Table 12.2.1)

= 42.29
= 3.566

12.2.9. NPSHR = 5.0, = 0.001 m, = 0.15 m, = 1.5 m.

 12 – 4

NPSHR = patm/ + Zs – pv/ – – –

patm/ = 9.19 m, pv/ = 0.24
Thus, 5 = 9.19 + Zs – 0.24 – 0.001 – 1.5 – 0.15
Zs = -2.299 m.

12.2.10. Q = 0.2 m3/s, H = 20 m, e = 80% = 0.8,  = 9810 N/m3.

bhp = QH/e = 9810(0.2)(20)/0.8 = 49050 watts
= 49.05 kW
Note that although the term break horsepower is used, the power is calculated in
kW units.

12.2.11

Using the following equation solve for Zs

Zs = 5.0 – 9.19 + 0.25 + 0.0001 + 0.15 + 1.5

= - 2.29 m (suction lift)

12.2.12

Using equation 12.2.9

Solve for the specific head using

With n = 1200, Q = 400/3600, and H = 60 m to obtain ns = 13.6

Next use equation 12.2.10 to obtain the cavitation constant

with K = 1210 from Table 12.2.1 for a pump efficiency of 80 % and ns =

13.6, to obtain σ = 0.059. NPSHi = σH = 3.566

12.4.1. Figure 12.4.2 and the associated data are reproduced below.

Pipe Diameter (in) Length (ft)

AB 18 7000
BC 9 3000
 12 – 5

BD 12 4000

Z C
D

A
P
B

Using equation (12.1.10),

From example 12.4.1, KAB = 0.373, KBC = 6.436, KBD = 1.852.

QBD = 0.651QAB, QBC = 0.349QAB
hL = 1.158
= 25 + 1.158(8)2 = 99.11 ft

= 62.4(8)(99.11)/(550(0.8)) = 112.45 hp.

12.4.2. HT = DZ + 1.158 = 40 + 1.158(4)2 = 58.53 ft.

= 62.4(4)(58.53)/(550(0.8)) = 33.20 hp.

12.4.3. Again, flow in the fully turbulent range is assumed.

Pipe 1: /D = 0.0015, f1 = 0.0215.

Similarly, K2 = 0.0776, K3 = 0.579 (from example 12.4.2)

; that is,
; that is,
; that is,
From the above three equations,
; ;
The continuity equation is
Q1 – Q2 – Q3 = 0
Thus, – – =0

Solving for hj gives hj  80.3 ft.

 12 – 6

12.4.4. Similar to the solution to problem 12.4.3,

, ,

Thus, – – =0

Solving for hj gives hj  64.0 ft.

12.5.1. Like in example 12.5.1, use the energy equation from reservoir surface A to
reservoir surface B.
0 + 0 + 20 + hp = 0 + 0 + 50 +

where L is the total pipe length (both suction and discharge pipes)

= 30 + 0.571Q2 (system head curve equation)

To obtain the operating point, plot a few points of the system head curve on the
pump characteristic curve.

Q (cfs) hp (ft)
2 32.28
3 35.14
4 39.14
5 44.28
6 50.56
7 57.98

After plotting this data on Figure 12.5.13, the operating point is obtained at Q 
7.1 cfs, H  59 ft and e  36 %
The break horsepower required to derive the pump is
Bhp = HP
 12 – 7

12.5.2. Again, using the energy equation,

0 + 0 + 20 + hp = 0 + 0 + 50 +

= 40 + 0.126Q2 + 1.482Q2
= 40 + 1.61Q2
To obtain the operating point, plot a few points of the system head curve on the
pump characteristic curve.

Q (cfs) hp (ft)
3 54.49
4 65.76
5 80.25
5.5 88.70
6 97.96

After plotting this data on Figure 12.5.13, the operating point is obtained at Q 
5.4 cfs, H  87 ft and e  78 %
The break horsepower required to derive the pump is
Bhp = HP

12.5.3. n = 1500 rpm. The Z-discharge relationship from example 12.5.3 is reproduced
below.

Z (ft) -20 0 40 80
Q1800 (gpm) 216 192 150 80

Using the affinity laws (equations (12.2.4) and (12.2.5)), Q1500 and H1500 are
computed.
Q1500 = Q1800(1500/1800) = (5/6)Q1800
H1500 = H1800(1500/1800)2 = (25/36)H1800
The resulting pump-head relationship is given below.

H1500 (ft) 69.4 62.5 55.6 41.7 27.8 13.9

Q1500 (gpm) 0 91.7 150.0 208.3 250.0 283.3
 12 – 8

This relationship is plotted in the following Figure, which is used to derive the
Z-discharge relationship.

Z (ft) -20 0 40 80
Q1500 (gpm) 190 165 100 --

12.5.3 (Second solution)

Part A
1500 rpm
H (ft) Q (gpm)
69 0
62.1 91
55.2 149
41.4 208
27.6 249
13.8 282

1800 rpm
H (ft) Q (gpm) @ z=-20 @ z=0 @ z=-40 @ z=80
100 0 -20 0 40 80
90 110 4.2 24.2 64.2 104.2
80 180 44.8 64.8 104.8 144.8
60 250 105 125 165 205
40 300 160 180 220 260
20 340 211.2 231.2 271.2 311.2

2160 rpm
H (ft) Q (gpm)
144 0
129.6 132
115.2 216
86.4 300
57.6 360
28.8 408
 12 – 9

Part B
Delivery Rate @ 1500 Delivery Rate @ 1800 Delivery Rate @ 2160
delta z rpm rpm rpm
-20 195 216 250
0 175 192 232
40 100 150 195
80 0 80 158
12 – 10
 12 – 11

12.5.4. Since the elevation of reservoir 2 is changed, the total head HT is modified.
HT = 200 – 100 + = 100 +
HT = 100 + 2.514Q 2

Next the system-head curve is plotted on the combined pump curve (Figure
12.5.17) to find the point of operation.

Q (gpm) Q (cfs) HT (ft)

1500 3.345 128.1
2000 4.460 150.0
2500 5.575 178.1
3000 6.690 212.5

(a) The operating point is a flow of approximately 2450 gpm. Thus, the total
discharge to the reservoir is about 2450 gpm (5.464 cfs).
Flow in line A, QA = 0.380Q3 = 0.380(2450) = 931 gpm
Flow in line B, QB = 2450 – 931 = 1519 gpm.
(b) The percentage of discharge in line A is (931/2450)(100) = 38%
(c) From the plot of the combined curve drawn, the discharge components can be
determined by projecting the operating point horizontally on to the individual
curves. The discharge delivered through pump A is approximately 1500 gpm and
through pump B is approximately 950 gpm.
(d) Efficiency is computed using

Hpump A = HT = 100 + 2.514Q2 = 100 + 2.514(5.464)2

= 175.06 ft.
Qpump A = 1500 gpm = 3.345 cfs.
= 83%.

12.5.5. Again, HT = 250 – 100 +

= + = +
=
Now, for pipe (line) A,

Then total head is

HT = 150 + 0.489 + 4.604 = 150 + 0.489 + 5.262
Thus, 4.604 = 5.262
 12 – 12

QA = 1.069QB or QB = 0.935QA
Q3 = QA + QB = QA + 0.935QA = 1.935QA
QA = 0.517Q3
Now, HT = 150 + 0.489 + 4.604(0.517Q3)2
= 150 + 1.720 = 150 + 1.720Q2
The following system-head curve relationship is obtained.

Q (gpm) Q (cfs) HT (ft)

1500 3.345 169.2
2000 4.460 184.2
2500 5.575 203.5
3000 6.690 227.0

The above data is plotted on the combined pump curve that was given in Figure
12.2.17.
(a) The operating point is a flow of approximately 2100 gpm. Thus, the total
discharge to the reservoir is about 2100 gpm (4.683 cfs).
QA = 0.517Q3 = 0.517(2100) = 1088 gpm (2.421 cfs).
QB = 2100 – 1088 = 1012 gpm (2.257 cfs)
(b) The percentage of discharge in line A = (1088)(100)/2100 = 51.8%
(c) Referring to the component curves on which the combined curve is plotted, the
discharge delivered through pump A is approximately 1350 gpm and through
pump B is approximately 750 gpm.
(d) Qpump A = 1350 gpm = 3.011 cfs.
Hpump A = 150 + 1.720Q2 = 150 + 1.720(4.683)2
= 187.7 ft.
= 80%.
 12 – 13

12.6.1

10.0 cfs 2.0 cfs

5000 ft of 24” 5000 ft of 12”
A B C

4000 ft of 18” 4000 ft of 18” 4000 ft of 18”

D E F
5000 ft of 18” 5000 ft of 18”
4.0 cfs 4.0 cfs

n= 1.85 n-1= 0.85        

   
KAB 0.0866 QAB 6.2 QCF 0.2
KBC 2.5322 QBC 2.2 QDE 3.8
KAD, KBE,
KCF 0.2812 QAD 3.8 QEF 3.8
KDE, KEF 0.3515 QBE 4  
   
   
-
Q1= 0.54054 1.291845 = -0.21225  
  3.289938  
   
-
Q2= 0.54054 -3.09407 = 0.237973  
    7.027977          

n= 1.85 n-1= 0.85        

   
KAB 0.0866 QAB 6.412252 QCF 0.037973
KBC 2.5322 QBC 1.962027 QDE 3.587748
KAD, KBE,
KCF 0.2812 QAD 3.587748 QEF 4.037973
KDE, KEF 0.3515 QBE 4.450225  
   
   
-
Q1= 0.54054 -0.42236 = 0.069295  
  3.294659  
   
-
Q2= 0.54054 0.289126 = -0.02347  
    6.659525          
 12 – 14

n= 1.85 n-1= 0.85        

   
KAB 0.0866 QAB 6.342957 QCF 0.014506
KBC 2.5322 QBC 1.985494 QDE 3.657043
KAD, KBE,
KCF 0.2812 QAD 3.657043 QEF 4.014506
KDE, KEF 0.3515 QBE 4.357462  
   
   
-
Q1= 0.54054 0.04363 = -0.00714  
  3.303768  
   
-
Q2= 0.54054 -0.12627 = 0.01023  
    6.671953          

n= 1.85 n-1= 0.85        

   
KAB 0.0866 QAB 6.350095 QCF 0.024736
KBC 2.5322 QBC 1.975264 QDE 3.649905
KAD, KBE,
KCF 0.2812 QAD 3.649905 QEF 4.024736
KDE, KEF 0.3515 QBE 4.374831  
   
   
-
Q1= 0.54054 -0.01863 = 0.003048  
  3.304333  
   
-
Q2= 0.54054 0.012531 = -0.00102  
    6.662306          

QAB 6.347047 QCF 0.025753

QBC 1.976281 QDE 3.652953
QAD 3.652953 QEF 4.023719
QBE 4.370767    

12.6.1

Determine the flow rate in each pipe for the simple network. Assume that fully turbulent flow exists
for all
pipes use the Hazen Willams eq. with C=140
[in] [ft]
Step 1: LAB= 5000 DAB= 24 2
LBC= 5000 DBC= 12 1
Using EQ 12.1.5 LAD= 4000 DAD= 18 1.5
LBE= 4000 DBE= 18 1.5
 12 – 15

K=L/C1.85D4.87 = 4.73 LCF= 4000 DCF= 18 1.5

LDE= 5000 DDE= 18 1.5
1.85 4.87
K=4.73L/140 D LEF= 5000 DEF= 18 1.5

KAB = 0.087

KBC = 2.532

KAD = KBE = KCF = 0.281

KDE = KEF = 0.352

Step 2: The initial guesses of flow for each pipe are shown below

One
5000 ft of 24" 5000 ft of 12"
A 6.2   B 2.2 C
       
       
       
4000 ft of
4000 ft of 18" 3.8 4000 ft of 18" 4.0 0.2 18"
       
       
       
D 3.8   E 3.8 F
5000 ft of 18" 5000 ft of 18"

Step 3, 4, 5 Consider loop 1 first and compute Q1: hL= KQ1.85

Q1= (-1)[KADQAD1.85 +KDEQDE1.85 - KBEQBE1.85 - KABQAB1.85]

1.85[KADQAD0.85 +KDEQDE0.85 + KBEQBE00.85 + KABQAB0.85 ]

Q1= -0.212

Step 6 Next consider remaining loop, loop 2, to compute Q2:

Q2= (-1)[KBEQBE1.85 +KEFQEF1.85 - KCFQCF1.85 - KBCQBC1.85]

1.85[KBEQBE0.85 +KEFQEF0.85 - KCFQCF0.85 - KBCQBC0.85 ]

Q2= 0.238

Step 7 Define new flows for the next iteration 2nd Iteration

QAD= 3.588 cfs

 12 – 16

QDE= 3.588 cfs

QBE= 4.450 cfs
QAB= 6.412 cfs
QBC= 1.962 cfs
QEF= 4.038 cfs
QCF= -0.012 cfs

5000 ft of 24" 5000 ft of 12"

A 6.412   B 1.962 C
       
       
       
4000 ft of
4000 ft of 18" 3.588 4000 ft of 18" 4.450 0.012 18"
       
       
       
D 3.588   E 4.038 F
5000 ft of 18" 5000 ft of 18"

Q1= (-1)[KADQAD1.85 +KDEQDE1.85 - KBEQBE1.85 - KABQAB1.85]

1.85[KADQAD0.85 +KDEQDE0.85 + KBEQBE00.85 + KABQAB0.85 ]

Q1= 0.069 cfs

Q2= (-1)[KBEQBE1.85 +KEFQEF1.85 + KCFQCF1.85 - KBCQBC1.85]

1.85[KBEQBE0.85 +KEFQEF0.85 + KCFQCF0.85 + KBCQBC0.85 ]

Q2= -0.024 cfs

Third Iteration
QAD= 3.657 cfs
QDE= 3.657 cfs
QBE= 4.357 cfs
QAB= 6.343 cfs
QBC= 1.986 cfs
QEF= 4.014 cfs
QCF= 0.012 cfs

5000 ft of 24" 5000 ft of 12"

 12 – 17

A 6.343   B 1.986 C
       
       
       
4000 ft of 18" 3.657 4000 ft of 18" 4.357 0.012
       
       
       
D 3.657   E 4.014 F
5000 ft of 18" 5000 ft of 18"

Q1= (-1)[KADQAD1.85 +KDEQDE1.85 - KBEQBE1.85 - KABQAB1.85]

1.85[KADQAD0.85 +KDEQDE0.85 + KBEQBE00.85 + KABQAB0.85 ]

Q1= -0.007 cfs

Q2= (-1)[KBEQBE1.85 +KEFQEF1.85 + KCFQCF1.85 - KBCQBC1.85]

1.85[KBEQBE0.85 +KEFQEF0.85 + KCFQCF0.85 + KBCQBC0.85 ]

Q2= 0.010 cfs

The above corrections Q1 and Q2 are insignificant, the final flows are those shown in the 3rd
iteration
 12 – 18

12.6.2.

n = 0.012

10.0 cfs 2.0 cfs

5000 ft of 24” 5000 ft of 12”
A B C

4000 ft of 18” 4000 ft of 18” 4000 ft of 18”

D E F
5000 ft of 18” 5000 ft of 18”
4.0 cfs 4.0 cfs

Using equation (12.1.9),

hL = KQ2, where from equation (12.1.10)
C = 140
Step 1. Determining the K values.
;

Step 2. Initial guesses: Again, the solution obtained in example 12.6.1 is used as
the initial guess.

10.0 cfs 2.0 cfs

A 6.3 cfs B 2.0 cfs C

3.7 cfs 4.3 cfs 0.0 cfs

D 3.7 cfs E 4.0 cfs F

4.0 cfs 4.0 cfs

Step 3. Compute Q1

 12 – 19

= -0.056
Step 4. Compute Q2

= 0.079
Step 5. Refine the flows for new iteration
QAD = 3.7 + (-0.056) = 3.644 cfs; QDE = 3.7 + (-0.056) = 3.644 cfs;
QAB = 6.3 – (-0.056) = 6.356 cfs; QBE = 4.3 – (-0.056) + 0.079 = 4.435 cfs;
QBC = 2.0 – 0.079 = 1.921 cfs; QCF = 0 – 0.079 = -0.079 cfs = 0.079 in the
opposite direction;
QEF = 4.0 +0.079 = 4.079 cfs.

10.0 cfs 2.0 cfs

A 6.356 cfs B 1.921 cfs C

3.644 cfs 4.435 cfs 0.079 cfs

D 3.644 cfs E 4.079 cfs F

4.0 cfs 4.0 cfs

Repeat step 3

= 0.025
Step 4. Compute Q2

= -0.007
These corrections, Q1 and Q2, may be considered practically insignificant.
Thus, the flows shown above may be taken.
 12 – 20

10.0 cfs 2.0 cfs

A 6.356 cfs B 1.921 cfs C

3.644 cfs 4.435 cfs 0.079 cfs

D 3.644 cfs E 4.079 cfs F

4.0 cfs 4.0 cfs

12.6.3. The head loss is computed along three possible paths. Each should result in
practically the same head loss.
(a) A  D  E  F

From the solution to problem 12.6.1, KAD = 0.281, KDE = 0.352, KEF = 0.352; and
QAD = 3.657, QDE = 3.657, QEF = 3.973 cfs.
Thus, = 0.281(3.657)1.85 + 0.352(3.657)1.85 + 0.352(3.973)1.85
= 11.49 ft.
(b) A  B  E  F

From the solution to problem 12.6.1, KAB = 0.087, KBE = 0.281, KEF = 0.352; and
QAB = 6.343, QBE = 4.316, QEF = 3.973 cfs.
Thus, = 0.087(6.343)1.85 + 0.281(4.316)1.85 + 0.352(3.973)1.85
= 11.37 ft.
(c) A  B  C  F

From the solution to problem 12.6.1, KAB = 0.087, KBC = 2.532, KEF = 0.281; and
QAB = 6.343, QBC = 2.027, QCF = 0.027 cfs.
Thus, = 0.087(6.343)1.85 + 2.532(2.027)1.85 + 0.281(0.027)1.85
= 12.01 ft.
The small differences between the values computed above may be attributed to
round off errors.
Average = (11.49 + 11.37 + 12.01)/3 = 11.62 ft.

12.6.4. Step 1. From the solution to example 12.6.1, KAB = 0.079, KBC = 2.520, KAD =
KBE = KCF = 0.265, KDE = KEF = 0.332.
Step 2. The flows obtained in the solution of example of 12.6.1 are used as the
initial guesses with the modification that the extra 2.0 cfs at node A is allowed to
go through pipelines AB and BC, as shown in Figure below.
 12 – 21

12.0 cfs 4.0 cfs

A 8.3 cfs B 4.0 cfs C

3.7 cfs 4.3 cfs 0.0 cfs

D 3.7 cfs E 4.0 cfs F

4.0 cfs 4.0 cfs

Step 3. Compute Q1

= 0.271
Step 4. Compute Q2

= 1.180
Step 5. Refine the flows for new iteration
QAD = 3.7 + 0.271 = 3.971 cfs; QDE = 3.7 + 0.271 = 3.971 cfs;
QAB = 8.3 – 0.271 = 8.029 cfs; QBE = 4.3 – 0.271 + 1.180 = 5.209 cfs;
QBC = 4.0 – 1.180 = 2.820 cfs; QCF = 0 – 1.180 = -1.180 cfs = 1.180 in the
opposite direction;
QEF = 4.0 + 1.180 = 5.180 cfs.
 12 – 22

12.0 cfs 4.0 cfs

A 8.029 cfs B 2.820 cfs C

3.971 cfs 5.209 cfs 1.180 cfs

D 3.971 cfs E 5.180 cfs F

4.0 cfs 4.0 cfs

Repeat step 3.

= 0.327
Repeat step 4.

= 0.170
Repeat step 5. Refine the flows for new iteration
QAD = 3.971 + 0.327 = 4.298 cfs; QDE = 3.971 + 0.327 = 4.298 cfs;
QAB = 8.029 – 0.327 = 7.702 cfs; QBE = 5.209 – 0.327 + 0.170 = 5.052 cfs;
QBC = 2.820 – 0.170 = 2.650 cfs; QCF = 1.180 + 0.170 = 1.350 cfs;
QEF = 5.180 + 0.170 = 5.350 cfs.

12.0 cfs 4.0 cfs

A 7.702 cfs B 2.650 cfs C

4.298 cfs 5.052 cfs 1.350 cfs

D 4.298 cfs E 5.350 cfs F

4.0 cfs 4.0 cfs

Repeat step 3.

= 0.047
Repeat step 4.
 12 – 23

= 0.047
Repeat step 5. Refine the flows for new iteration
QAD = 4.298 + 0.047 = 4.345 cfs; QDE = 4.298 + 0.047 = 4.345 cfs;
QAB = 7.702 – 0.047 = 7.655 cfs; QBE = 5.052 – 0.047 + 0.047 = 5.052 cfs;
QBC = 2.650 – 0.047 = 2.603 cfs; QCF = 1.350 + 0.047 = 1.397 cfs;
QEF = 5.350 + 0.047 = 5.397 cfs.

12.0 cfs 4.0 cfs

A 7.655 cfs B 2.603 cfs C

4.345 cfs 5.052 cfs 1.397 cfs

D 4.345 cfs E 5.397 cfs F

4.0 cfs 4.0 cfs

Repeat step 3.

= 0.013
Repeat step 4.

= 0.006
These corrections, Q1 and Q2, may be considered practically insignificant.
Thus, the flows given above may be taken.

12.6.5. Step 1. From the previous solutions, KAB = 0.079, KBC = 2.520, KAD = KBE =
KCF = 0.265, KDE = KEF = 0.332.
Step 2. Initial guesses: Instead of making wild guesses, the incoming flows to any
node are split in the outgoing pipes in the same proportion as the solution
obtained in problem 12.6.4. For instance,
QAD = (3.7/10) x 8 = 3.0 cfs, QAB = (6.3/10) x 8 = 5.0 cfs. This procedure may not
result in a correct solution, but it may give a reasonable initial guess.
 12 – 24

8.0 cfs 2.0 cfs

A 5.0 cfs B 1.6 cfs C

3.0 cfs 3.4 cfs 0.4 cfs

D 3.0 cfs E 3.4 cfs F

3.0 cfs 3.0 cfs

Step 3. Compute Q1

= -0.054
Step 4. Compute Q2

= -0.040
Step 5. Refine the flows for new iteration
QAD = 3.0 + (-0.054) = 2.946 cfs; QDE = 3.0 + (-0.054) = 2.946 cfs;
QAB = 5.0 – (-0.054) = 5.054 cfs; QBE = 3.4 – (-0.054) + (-0.040) = 3.414 cfs;
QBC = 1.6 – (-0.040) = 1.640 cfs; QCF = 0.4 + (-0.040) = 0.360 cfs;
QEF = 3.4 + (-0.040) = 3.360 cfs.

8.0 cfs 2.0 cfs

A 5.054 cfs B 1.640 cfs C

2.946 cfs 3.414 cfs 0.360 cfs

D 2.946 cfs E 3.360 cfs F

 12 – 25

3.0 cfs 3.0 cfs

Repeat step 3.

= -0.012
Repeat step 4.

= -0.007
These corrections, Q1 and Q2, may be considered practically insignificant.
Thus, the flows given above may be taken.

12.6.6. Step 1. From the previous solutions, KAB = 0.079, KBC = 2.520, KAD = KBE =
KCF = 0.265, KDE = KEF = 0.332.
Step 2. Initial guesses: Initial guesses are made in a similar fashion as in the
solution of problem 12.6.5.

14.0 cfs 2.0 cfs

A 8.8 cfs B 2.8 cfs C

5.2 cfs 6.0 cfs 0.8 cfs

D 1.2 cfs E 3.2 cfs F

4.0 cfs 4.0 cfs 4.0 cfs

Step 3. Compute Q1

= 0.987
Step 4. Compute Q2
 12 – 26

= 0.352
Step 5. Refine the flows for new iteration
QAD = 5.2 + 0.987 = 6.187 cfs; QDE = 1.2 + 0.987 = 2.187 cfs;
QAB = 8.8 – 0.987 = 7.813 cfs; QBE = 6.0 – 0.987 + 0.352 = 5.365 cfs;
QBC = 2.8 – 0.352 = 2.448 cfs; QCF = 0.8 – 0.352 = 0.448 cfs;
QEF = 3.2 + 0.352 = 3.552 cfs.

14.0 cfs 2.0 cfs

A 7.813 cfs B 2.448 cfs C

6.187 cfs 5.365 cfs 0.448 cfs

D 2.187 cfs E 3.552 cfs F

4.0 cfs 4.0 cfs 4.0 cfs

Repeat step 3.

= 0.082
Repeat step 4.

= 0.188
Step 5. Refine the flows for new iteration
QAD = 6.187 + 0.082 = 6.269 cfs; QDE = 2.187 + 0.082 = 2.269 cfs;
QAB = 7.813 – 0.082 = 7.731 cfs; QBE = 5.365 – 0.082 + 0.188 = 5.471 cfs;
QBC = 2.448 – 0.188 = 2.260 cfs; QCF = 0.448 – 0.188 = 0.260 cfs;
QEF = 3.552 + 0.188 = 3.740 cfs.

14.0 cfs 2.0 cfs

A 7.731 cfs B 2.260 cfs C

6.269 cfs 5.471 cfs 0.260 cfs

D 2.269 cfs E 3.740 cfs F

4.0 cfs 4.0 cfs 4.0 cfs

 12 – 27

Repeat step 3.

= 0.059
Repeat step 4.

= 0.019
Step 5. Refine the flows for new iteration
QAD = 6.269 + 0.059 = 6.328 cfs; QDE = 2.269 + 0.059 = 2.328 cfs;
QAB = 7.731 – 0.059 = 7.672 cfs; QBE = 5.471 – 0.059 + 0.019 = 5.431 cfs;
QBC = 2.260 – 0.019 = 2.241 cfs; QCF = 0.260 – 0.019 = 0.241 cfs;
QEF = 3.740 + 0.019 = 3.759 cfs.
 12 – 28

14.0 cfs 2.0 cfs

A 7.672 cfs B 2.241 cfs C

6.328 cfs 5.431 cfs 0.241 cfs

D 2.328 cfs E 3.759 cfs F

4.0 cfs 4.0 cfs 4.0 cfs

Repeat step 3.

= 0.006
Repeat step 4.

= 0.010
These corrections, Q1 and Q2, may be considered practically insignificant.
Thus, the flows given above may be taken.

12.6.7. The head loss is computed along two possible paths. Both should result in
practically the same head loss. Note the solution to the network problem of
example 12.6.1 when n = 0.012 is used was obtained in problem 12.6.2.
(a) A  D  E  F

From the solution to problem 12.6.2, KAB =0.083, KBC = 3.333, KAD = KBE = KCF
= 0.307, KDE = KEF = 0.384; and QAB = 6.356, QBC = 1.921, QAD = 3.644, QDE =
3.644, QBE = 4.435, QEF = 4.079 in cfs unit.
Thus, = 0.307(3.644)2 + 0.384(3.644)2 + 0.384(4.079)2
= 15.56 ft.
(b) A  B  E  F

= 0.083(6.356)2 + 0.307(4.435)2 + 0.384(4.079)2

= 15.78 ft.
The small differences between the values computed above may be attributed to
round off errors.
Average = (15.56 + 15.78)/2 = 11.67 ft.
 12 – 29

12.6.8. To develop the system of equations for the flows in the network in Figure 12.6.3
using the linear theory method, the pipes and nodes are first labeled as shown
below.

10.0 cfs 2 2.0 cfs

1 A [1] B [2] C 3

(2) [ ] – Pipe
– Node
[3] (1) [4] (2) [5] ( ) – Loop

D [6] E [7] F

4 5 4.0 cfs 6 4.0 cfs

Let Q1, Q2, …, Q7 represent the flow in pipe 1, pipe 2, …, pipe 7, respectively.
Conservation of flow at each node.
Node 1: Q1 + Q3 = 10
Node 2: Q1 – Q2 – Q4 = 0
Node 3: Q2 + Q5 = 2
Node 4: Q3 – Q6 = 0
Node 5: Q4 + Q6 – Q7 = 4
Node 6: Q7 – Q5 = 4
Conservation of energy (loop constraints)
Loop 1:
Loop 2:
Using , these two conservation of energy equations can be made
linear.
Loop 1: k3Q3 + k6Q6 – k1Q1 –k4Q4 = 0
Loop 2: k4Q4 + k7Q7 + k5Q5 –k2Q2 = 0
Thus, the resulting system of equations is
Q1 + Q3 = 10
Q1 – Q2 – Q4 = 0
Q2 + Q5 = 2
Q3 – Q6 = 0
Q4 + Q6 – Q7 = 4
k3Q3 + k6Q6 – k1Q1 –k4Q4 = 0
k4Q4 + k7Q7 + k5Q5 –k2Q2 = 0
Note that one of the conservation equations is dropped, since it is redundant.
 12 – 30

12.6.9
12 – 31
12 – 32
12 – 33
 12 – 34

12.6.10
12 – 35
12 – 36
12 – 37
 12 – 38

12.7.1. KYPIPE2 input for problem 12.7.1

*******************************************************************************

0 0 7 3 0 0 0 0 0 0.005 0 0.000001 0 0

0 0 2 5000.0 24.0 2.0 0 0 100.0 1

0 0 4 4000.0 18.0 2.0 0 0 100.0 2
0 2 3 5000.0 12.0 2.0 0 0 0 3
0 2 5 4000.0 18.0 2.0 0 0 0 4
0 3 6 4000.0 18.0 2.0 0 0 0 5
0 4 5 5000.0 18.0 2.0 0 0 0 6
0 5 6 5000.0 18.0 2.0 0 0 0 7
2.0 0 3
4.0 0 5
4.0 0 6
0
0 0 0 0 0
-2.
KYDATA

********************************************************************************

KYPIPE2 output for problem 12.7.1

********************************************************************************

FLOWRATE IS EXPRESSED IN CFS AND PRESSURE IN PSIG

A SUMMARY OF THE ORIGINAL DATA FOLLOWS

THE DARCY WEISBACH HEAD LOSS EQUATION IS USED, THE KINEMATIC VIS. = .0000010

PIPE NO. NODE NOS. LENGTH DIAMETER ROUGHNESS MINOR LOSS K FIXED GRADE
(FEET) (INCHES)
1 0 2 5000.0 24.0 2.0 .00 100.00
2 0 4 4000.0 18.0 2.0 .00 100.00
3 2 3 5000.0 12.0 2.0 .00
4 2 5 4000.0 18.0 2.0 .00
5 3 6 4000.0 18.0 2.0 .00
6 4 5 5000.0 18.0 2.0 .00
7 5 6 5000.0 18.0 2.0 .00

THE NUMBER OF JUNCTION NODES = 5 WHICH DOES NOT AGREE WITH INPUT DATA

JUNCTION NUMBER DEMAND ELEVATION CONNECTING PIPES

2 .00 .00 1 3 4
3 2.00 .00 3 5
4 .00 .00 2 6
5 4.00 .00 4 6 7
6 4.00 .00 5 7

OUTPUT SELECTION: ALL RESULTS ARE OUTPUT EACH PERIOD

THIS SYSTEM HAS 7 PIPES WITH 5 JUNCTIONS , 1 LOOPS AND 2 FGNS

THE RESULTS ARE OBTAINED AFTER 4 TRIALS WITH AN ACCURACY = .00003

PIPE NO. NODE NOS. FLOWRATE HEAD LOSS PUMP HEAD MINOR LOSS VELOCITY HL/1000
1 0 2 6.33 3.11 .00 .00 2.02 .62
 12 – 39

2 0 4 3.67 3.79 .00 .00 2.08 .95

3 2 3 1.94 11.19 .00 .00 2.47 2.24
4 2 5 4.39 5.41 .00 .00 2.48 1.35
5 3 6 -.06 .00 .00 .00 -.03 .00
6 4 5 3.67 4.73 .00 .00 2.08 .95
7 5 6 4.06 5.78 .00 .00 2.30 1.16

JUNCTION NUMBER DEMAND GRADE LINE ELEVATION PRESSURE

2 .00 96.89
3 2.00 85.70
4 .00 96.21
5 4.00 91.48
6 4.00 85.70

THE NET SYSTEM DEMAND = 10.00

SUMMARY OF INFLOWS(+) AND OUTFLOWS(-) FROM FIXED GRADE NODES

PIPE NUMBER FLOWRATE

1 6.33
2 3.67

THE NET FLOW INTO THE SYSTEM FROM FIXED GRADE NODES = 10.00
THE NET FLOW OUT OF THE SYSTEM INTO FIXED GRADE NODES = .00

**********************************************************************

12.7.2. A summary of the solution of this problem obtained using EPANET is given
below.

Node Elevation Demand Head Pressure

ID ft CFS ft psi
Junc 1 0 -10.00 68.87 29.84
Junc 2 0 0.00 63.98 27.72
Junc 3 0 2.00 47.47 20.57
Junc 4 0 0.00 63.16 27.37
Junc 5 0 4.00 56.01 24.27
Junc 6 0 4.00 47.47 20.57
Junc 7 0 0.00 100.00 0.00

Link Length Diameter Rough- Flow Velocity Head- Friction Status

ID ness loss factor
ft in CFS fps ft/Kft
Pipe 1 5000 24 100 6.35 2.02 0.98 0.031 Open
Pipe 2 5000 12 100 1.98 2.52 3.30 0.034 Open
Pipe 3 4000 18 100 -0.02 0.01 0.00 0.068 Open
Pipe 4 5000 18 100 -4.02 2.28 1.71 0.032 Open
Pipe 5 5000 18 100 -3.65 2.07 1.43 0.032 Open
Pipe 6 4000 18 100 -3.65 2.07 1.43 0.032 Open
Pipe 7 4000 18 100 4.37 2.47 1.99 0.031 Open
Pipe 8 1000 12 100 0.00 0.00 0.00 0.000 Closed
 12 – 40

********************************************************
* EPANET *
* Hydraulic and Water Quality *
* Analysis for Pipe Networks *
* Version 2.0 *
********************************************************

Link – Node Table:

--------------------------------------------------------
Link Start End Length Diameter
ID Node Node ft in
--------------------------------------------------------
1 1 2 5000 24
2 2 3 5000 12
3 3 6 4000 18
4 6 5 5000 18
5 5 4 5000 18
6 4 1 4000 18
7 2 5 4000 18
8 1 7 1000 12

Node Results
---------------------------------------------------------
Node Demand Head Pressure Quality
ID CFS ft psi
---------------------------------------------------------
1 -10.00 68.87 29.84 0.00
2 0.00 63.98 27.72 0.00
3 2.00 47.47 20.57 0.00
4 0.00 63.16 27.37 0.00
5 4.00 56.01 24.27 0.00
6 4.00 47.47 20.57 0.00
7 0.00 100.00 0.00 0.00 Reservoir

Link Results:
--------------------------------------------------------------
Link Flow Velocity Headloss Status
ID CFS fps ft/Kft
--------------------------------------------------------------
1 6.35 2.02 0.98 Open
2 1.98 2.52 3.30 Open
3 -0.02 0.01 0.00 Open
4 -4.02 2.28 1.71 Open
5 -3.65 2.07 1.43 Open
6 -3.65 2.07 1.43 Open
7 4.37 2.47 1.99 Open
8 0.00 0.00 0.00 Closed
 12 – 41

8
1 1 2 2 3

6 7 3

4 5 5 4 6

12.7.3

No solution
 12 – 42

12.7.4
 12 – 43

12.7.4 (Second solution)

12 – 44
 12 – 45

12.8.1. D = 20 cm = 0.2 m, tw = 15 mm = 0.015 m, Ew = 2.17 x 109 N/m2, Ep = 16 x 1011

N/m2, Q = 50 L/s = 0.05 m3/s.
(a) p = Vvc

p = 1000(1.59)(1473) = 2.342 x 106 N/m2.

(b) vc = 1460 m/s (from the solution to example 12.8.1)
p = 1000(1.59)(1460) = 2.321 x 106 N/m2.
(c) vc = 1462 m/s (from the solution to example 12.8.1)
p = 1000(1.59)(1460) = 2.325 x 106 N/m2.

12.8.2. D = 15 cm, tw = 15 mm = 0.015 m, Ew = 2.17 x 109 N/m2, Ep = 16 x 1011 N/m2

Q = 40 L/s = 0.04 m3/s
(a) vc = 1473 m/s

p = Vvc = 1000(2.26)(1473) = 3.329 x 106 N/m2.

(b)

= 1463 m/s.
p = 1000(2.26)(1463) = 3.306 x 106 N/m2.

(c)

= 1464 m/s.
p = 1000(2.26)(1464) = 3.309 x 106 N/m2.

12.8.3. L = 2000 m, D = 0.5 m, tw = 5 cm = 0.05 m, Z1 – Z2 = 50 m, Q = 0.8 m3/s,

t = 1.4 s, Ew = 2.17 x 109 N/m2, Ep = 1.9 x 1011 N/m2

= 5.50 m
p2 = 5.50(9810) = 5.4 x 104 N/m2.
At mid-length (point 3)
 12 – 46

= 2.33 m
p3 = 2.33(9810) = 2.3 x 104 N/m2.
vc = 1395.6 m/s (from the solution to example 12.8.2)
tc = 2L/vc = 2(2000)/1395.6 = 2.87 s.
Since the time of closure of 1.4 s is less than tc of 2.87 s, the valve closure is
rapid; hence, the maximum increase of pressure generated by water hammer is
p = Vvc = 1000(4.074)(1395.6) = 5.69 x 106 N/m2 (same as before).
The maximum pressure at the valve is p2 + p = 5.4 x 104 + 5.69 x 106 = 5.74 x
106 N/m2.
The maximum pressure at mid-length is p3 + p = 2.3 x 104 + 5.69 x 106 = 5.71 x
106 N/m2.

12.8.4. Z1 – Z2 = 100 m

= 66.4 m
p2 = 66.4(9810) = 6.5 x 105 N/m2.
At mid-length (point 3)

= 32.8 m
p3 = 32.8(9810) = 3.2 x 105 N/m2.
tc = 2.15 s (from the solution to example 12.8.2). The valve closure is rapid.
p = 5.69 x 106 N/m2 (from the solution to example 12.8.2)
The maximum pressure at the valve is p2 + p = 6.5 x 105 + 5.69 x 106 = 6.34 x
106 N/m2.
The maximum pressure at mid-length is p3 + p = 3.2 x 105 + 5.69 x 106 = 5.85 x
106 N/m2.

12.8.5. D = 18 in = 1.5 ft

V = 12.54 ft/s
Check the Reynolds number:
From Moody diagram, for Re = 1.55 x 106 and /D = 0.000075, f  0.013. Hence,

V = 11.80 ft/s
 12 – 47

Check the Reynolds number:

From Moody diagram, for Re = 1.45 x 106 and /D = 0.000075, f  0.013.
Thus, V = 11.80 ft/s is OK.

= 3423 ft/s.
tc = 2L/vc = 2(2)(5280)/3423 = 6.17 s.
(a) For t = 4 s, this is a rapid closure; then,
p = Vvc = 1.94(11.80)(3423) = 78359 lb/ft2 = 544 psi.
The pressure at the valve is
pvalve = pstatic + p = gh + p = 1.94(32.2)(200)/144 + 544 = 87 + 544 = 631 psi.

= 28395 psi.
(b) The 60-second closure is in excess of 2L/vc. Let vc = 2L/tc.
= 2(2)(5280)(1.94)(11.8)/60 = 8058 lb/ft2 = 56 psi.
pvalve = pstatic + p = gh + p = 87 + 56 = 143 psi.
= 6435 psi.

(c) a = a – static = 16000 – = 16000 – 87(9)/0.2 = 12085 psi.

= 6.25 s.

12.8.6. D = 30 in = 2.5 ft

V = 16.12 ft/s.
Check the Reynolds number:
From Moody diagram, for Re = 3.31 x 106 and /D = 0.000075, f  0.012. Hence,

V = 15.79 ft/s.
Check the Reynolds number:
 12 – 48

From Moody diagram, for Re = 3.24 x 106 and /D = 0.000075, f  0.012.
Thus, V = 15.79 ft/s is OK.

= 2984 ft/s.
tc = 2L/vc = 2(2)(5280)/2984 = 7.08 s.
(a) For t = 4 s, this is a rapid closure; then,
p = Vvc = 1.94(15.79)(2984) = 91408 lb/ft2 = 635 psi.
pvalve = pstatic + p = gh + p = 1.94(32.2)(200)/144 + 635 = 87 + 635 = 722 psi.

= 54150 psi.
(b) The 60-second closure is in excess of 2L/vc. Let vc = 2L/tc.
= 2(2)(5280)(1.94)(15.79)/60 = 10783 lb/ft2 = 75 psi.
pvalve = pstatic + p = gh + p = 87 + 75 = 162 psi.
= 12150 psi.

(c) a = a – static = 16000 – = 16000 – 87(15)/0.2 = 9475 psi.

= 35.56 s.

12.8.7. D = 30 in = 2.5 ft, A = (2.5)2/4 = 4.91 ft2, As = 12.57 ft2 (from the solution to
example 12.8.7)
Vo = Q/A = 20/4.91 = 4.07 ft/s.

= 34.72 ft.

= 34.3 s

12.8.8. A = 3.14 ft2 (from the solution to example 12.8.6), Ds = 6 ft, As = (6)2/4 =
28.27 ft2
Vo = 6.37 ft/s (from the solution to example 12.8.6)

= 28.98 ft.
 12 – 49

= 64.3 s

12.8.9. L = 4000 ft, A = 3.14 ft2, As = 12.57 ft2, Vo = 6.37 ft/s (from the solution to
example 12.8.6)

= 35.48 ft.

= 35.0 s

12.8.10

First compute the velocity of flow in the pipe for the 220 gpm, V = 1.41 ft/s

Next compute the pressure head for the 80 psi just upstream of the valve, P/γ =
184.6 ft

Next compute the headloss coefficient for the valve using the energy equation,

P/γ = 184.6 = KV2/2g = K(1.41)2/[2(32.2)] solving K = 5979.

Next the angle of opening of the valve is found using the K in Figure P12.8.10.

The opening is approximately 7o.

12.8.11

To compute the actual time of time of closure for the linear scheme 7o/90o = x/45,
solving then x = 3.5 sec.

The time of closure would be computed using tc = 2L/vc with vc = 4720 ft/s

tc = 2(2x5280)/4720 = 4.5 sec.

Because the actual time of closure is less than the 4.5 sec the full pressure wave

will be realized.

Velocity of flow in the pipe for the 220 gpm, V = 1.41 ft/s
 12 – 50

Water hammer pressure is then Δp = ρVvc = 1.94(1.41)(4720) = 12900 lb

/ft2 = 89.7 psi

12.8.12
First compute the velocity of flow in the pipe for the 220 gpm, V = 1.41 ft/s

Next compute the pressure head for the 90 psi just upstream of the valve, P/γ =
207.7 ft
Next compute the headloss coefficient for the valve using the energy equation,

P/γ = 207.7 = KV2/2g = K(1.41)2/[2(32.2)] solving K = 6728.

Next the angle of opening of the valve is found using the K in Figure P12.8.10.

The opening is approximately 7o.

The time of closure would be computed using tc = 2L/vc with vc = 4720 ft/s

tc = 2(3x5280)/4720 = 6.7 sec.

Solving xo/90o = 6.7/45 for x = 13.4o so the safe closure would be for valve

openings of 13.4o and larger.

12.8.13

vc = (E/ρ)0.5 = 1473 m/s

V = Q/A = 0.566 m/s

Water hammer pressure is then Δp = ρVvc = 8.34 x 105 N/m2

Time of closure is tc = 2L/vc = 2(1000)/1473 = 1.3 s