CHEMICAL ENGINEERING LAB

Flow Measurement in Closed Conduit

& Centrifugal Pump Characteristics

Part I
Flow Measurement in
Closed Conduit

Summary
The objectives of this experiment were to get familiarized with various flow
measuring devices, such as venturi meter, orifice plate, rotameter, etc., for measurement
of discharge in closed conduits and to determine the loss coefficients for the various
fittings in the system.
The loss coefficient, K was found by making use of the following expression.
g c P
V2
K

where P is the pressure drop, is the density of fluid and v is the velocity of the fluid.
The pressure drop across a particular device was measured at different flow rates to
obtain the various loss coefficients of the various devices.
Once the various loss coefficients were obtained, the flow rate across each line
when all three lines were opened was calculated, and the sum of the flow rates was found
to be close to the actual flow rated used which showed that the loss coefficients obtained
were rather accurate as well.

CONTENTS
Part I: Flow Measurement in Closed Conduit
Sections

Page

Title

Summary

Introduction

II

Theoretical Background

III

Experimental set up and procedures

IV

Results and Analysis

Discussion

24

VI

Conclusion

33

VII

References

33

Appendices

Part II: Centrifugal Pump

I Introduction
The objectives of the experiments were to:
1. To get familiarized with various flow measuring devices, such as venturi meter,
orifice plate, rotameter, etc., for measurement of discharge in closed conduits.
2. To determine the loss coefficients for the various fittings in the system.
3. To determine to individual flow rates in each line when all 3 lines were opened.
Flow rate measurements are important not only in the laboratory but also in the
industrial plants, serving as a basis for proper monitoring and control. As a result, we
must be able to account for the losses due to pipe material, pipe fittings as well as the
flow measuring devices in the system. By measuring pressure, elevation and sometimes
velocity, measurement of losses can be accomplished.
The aim of this experiment is to get familiarized with the different flow measuring
devices, such as the venturi meter, orifice meter and rotameter, which are often used for
the measurement of discharge in closed conduits. The choice of a flow meter is
influenced by the accuracy required, range, cost, ease of reading and service life. The
simplest and cheapest device that gives the desired accuracy is then appropriately chosen.
The various fittings loss coefficients are being investigated. The flow in a piping
system may be required to pass through a variety of fittings, bends or abrupt changes in
area. Additional head loss arises as a result of flow separation. For flow through pipe
bends and fittings, the loss coefficient, K is found to vary with pipe size (diameter) in
much the same manner as the friction factor, f, for flow through a straight pipe. Pressure
at different points of the closed conduits is measured and the results are then used for the
determination of the loss coefficient, K for the various fittings in the system
For a good plant design, it is essential to know the values of head losses as well as to
account for them in the design. Given the wide range of flow meters available nowadays,
it is also important for engineers to possess knowledge of the relative advantages and
disadvantages, as well as limitations, of the different types of flow measuring devices
available. Often, the simplest and most economical device that will be able to cope with
the desired accuracy range will be preferred.
The study of loss coefficients for the various fittings in a piping system is an essential
requirement for understanding the flow behavior within the system.

II Theoretical Background
Water is used as the flowing fluid in this experiment and is pumped from the
holding vessel into the piping system consisting of three parallel lines with different
components.
Line 1 consists of:
Smooth bore pipe, rough bore pipe, gradual expansion, sudden contraction, a 90o
bend, a 180o bend, and a 90o elbow.
It can be closed or opened by the control of the isolation valve.
Line 2 consists of:
Diaphragm valve, ball valve, globe valve and needle valve.
It can be closed by the control of any one valve or opened by all the valves.
Line 3 consists of:
Orifice meter and venturi meter.
The closure and open of this line is controlled by the isolation valve.
The combined flow passed through a rotameter and then flowed back to the
holding vessel.
In pipe lines, the total head loss is made up of both the frictional head loss and the
head loss due to fittings (e.g. valves, elbows)
(hL)total = (hL)frictional + (hL)fittings
where (hL)frictional = frictional head loss = 2ffLV2/Dg
(hL)fittings = head loss due to fittings = P/ = KV2/2 gc
In this experiment, we are only concerned about the head loss due to fittings.
(hL)fittings = P/ = KV2/2gc
The loss coefficient, K is defined as follows:
g c P
V2
K

where
P

Pressure drop

Density of fluid

Velocity of the fluid

Loss coefficient

For a particular device, if the pressure drop and the flow rate are known, the loss
coefficient can be found by plotting (gc P/) against V2/2). The slope is then the loss
coefficient, K.
Once the various loss coefficients are obtained for the various devices, the flow
rates in each individual line can be calculated by finding out V, the velocity of the fluid
using the same equation above. The total flow rates of the 3 lines should be equal to the
actual flow rate used if the loss coefficients are accurately obtained.

III Experimental Procedure

Flow Chart of Experimental Procedures
Before starting the pump, ensure that
the bypass is opened
Switch on the power for the digital
pressure gauge and the pump.
Open the gate valve for line 1, and
close the valves for lines 2 and 3
Close the bypass valve partially, and
adjust the knob of the rotameter to set
a desired flowrate
Take the reading for all the pressure
tap points in line 1
Vary the flow rate to make at least 4
different runs
Repeat the same procedure for line 2
(with lines 1 and 3 closed) and line 3
(with lines 1 and 2 closed)

Open all 3 lines and take pressure

readings for 4 different flow rates

3.1 Detailed Procedures

The different apparatus and the apparatus under investigation is listed below.
Pipings
1.

Smooth bore pipe

2.

rough bore pipe

3.

diaphragm valve

4.

ball valve

5.

globe valve

6.

needle valve

7.

isolation valve

Valves

Other pipe fittings

8.

gradual expansion

9.

sudden contraction

10.

bend

11.

elbows

12.

orifice meter

13.

venturi meter

14.

rotameter

15.

holding vessel

16.

pump

17.

digital pressure gauge

Meters

Others

The apparatus was set up as shown in the diagram below.

Figure 1: Experimental Set-up

Switch on the power for the digital pressure gauge and the pump. Open the gate valve for
1. The bypass was ensured to be opened and the pump started.
2. The gate valve for line 1 was opened while the valves for line 2 and 3 were
closed.
3. The bypass valve was closed partially and the knob of the rotameter adjusted to a
flow rate of 10 l/min.
4. The pressure readings at tap points 1-10, 21 and 22 were taken down and
recorded.
5. The flow rate was varied covering the range of flow rates from 10 l/min to
65 l/min to obtained 4 different runs.
6. The same procedure for line 2 (with line 1 and 3 closed) was repeated and
pressure readings at tap points 11-15, 21 and 22 were taken down and recorded.
7. The same procedure for line 3 (with line 1 and 2 closed) was repeated and
pressure readings at tap points 16-22 were taken down and recorded.

8. All three lines were opened and pressure readings at all tap points were taken
down and recorded.
9. The flow rate was varied covering the range of flow rates from 20 l/min to
75 l/min to obtain 4 different runs.
3.2 Precautions Taken

All valves in the lines other that the one to be studied were ensured to be
closed.

The flow rate through lines 2 and 3 were kept below 30 l/min to prevent
damage to the valves in the lines.

IV Results and Analysis

Line

Device
Rough Bore Pipe

Internal Diameter, ID (m)

0.0200

180o Bend

0.0221

Sudden Contraction

0.0317

Gradual Expansion

0.0317

90o Elbow

0.0221

90o Bend

0.0221

Smooth Bore Pipe

0.0221

Diaphragm Valve

0.0166

Ball Valve

0.0166

Globe Valve

0.0166

Needle Valve

0.0166

Orifice Meter

0.0127

Venturi Meter

0.0127

Rotameter

0.0221

Table 1: Diameters of various devices

Set
Number

Flow
rate
(l/min)
Point
1
2
3
4
5
6
7
8
9
10
21
22

10

30

50

65

22.602
22.644
22.678
22.651
22.616
22.617
22.659
22.679
22.724
22.73
22.522
22.223

Pressure (psi)
18.273 12.527
19.073 14.632
19.132 14.866
19.296 15.263
19.227 15.123
19.358 15.285
19.416
15.72
19.45
15.81
19.564 16.082
19.667 16.301
17.783 11.432
17.275 10.593

6.688
10.29
10.72
11.395
11.073
11.333
12.023
12.225
12.655
13.029
4.87
3.687

Table 2: Pressure readings for line 1 open

Set
number

Flow rate
(l/min)

10

20

25

30

Point
11
12
13
14
15
21
22

22.457
21.395
21.331
21.167
19.332
20.335
20.013

Pressure (psi)
20.97
20.507
18.822 18.321
18.554 17.222
17.777 17.209
13.418
9.649
15.236 10.486
15.046 10.045

19.946
15.655
15.58
13.442
1.751
7.922
8.429

Table 3: Pressure readings for line 2 open

Set number

10

Flow rate
(l/min)
Point
16
17
18
19
20
21
22

10

22.308
22.27
22.295
22.108
22.255
22.26
21.959

20

25

Pressure (psi)
21.893 20.288
20.653 18.872
20.723 18.027
20.394 18.604
20.602 17.837
22.593 19.83
20.202 19.378

30

19.778
17.185
19.366
19.085
19.146
19.131
18.604

Table 4: Pressure readings for line 3 open

Set number
Flow rate
(l/min)
Point
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

1
20

2
40

3
60

4
75

Pressure (psi)
20.944
21.009
21.021
21.023
21.008
21.012
21.025
21.037
21.041
21.058
21.077
21.023
21.016
20.98
20.905
21.044
20.895
20.977
20.841
20.944
20.830
20.456

17.683
17.895
17.966
17.984
17.927
17.991
18.025
18.037
18.056
18.082
18.191
18.049
18.033
17.96
17.503
18.225
17.338
17.58
17.141
17.513
17.165
16.502

13.706
14.225
14.32
14.406
14.372
14.425
14.553
14.565
14.619
14.679
14.961
14.609
14.549
14.437
13.502
14.906
13.666
12.802
13.511
12.531
12.731
11.712

9.541
10.48
10.588
10.794
10.707
10.787
10.979
11.044
11.123
11.212
11.714
11.142
11.069
10.842
9.32
11.721
8.403
9.522
8.109
9.283
8.008
8.488

Table 5: Pressure readings for all 3 lines open

Calculation of loss coefficient:

11

The loss coefficient, K, is defined by the following expression:

g c P
V2
K

where P: pressure drop

: density of fluid
V: velocity of the fluid.
There are two methods to calculate the value of the loss coefficient K.
First Method:
Let Q be the flow rate set in the lines.
Hence, V = Q/A where A is the area.
With substituting P, and V, K can be obtained.
Eg:
Rough Bore Pipe (diameter = 0.0200m) with Q = 10 l/min
Q(l/min)
10

Q(m3/s)
0.000167

P1

P (psi)

P2
22.602

22.644

0.042

P (Pa)
289.6

Velocity
0.53

Velocity2
0.2809

V=Q/A
= 0.000167 * 4 / * 0.02002
= 0.53 m/s
K = (gcP * 2 ) / ( * V2 )
= (1*289.6*2) / (1000 * 0.2809)
= 2.062

Second Method:
Plot a graph of P vs V2 .
12

The gradient obtained from the graph = K / 2.

With density known as 1000 kg m-3 , K can be found.
Eg:
Rough Bore Pipe (diameter = 0.0200m) with Q = 10 l/min

From the best fit line, the equation is found be to y = 2094.2x 90.246
Hence, Gradient = 2094.2 = K / 2
Therefore, K = 2094.2 * 2 / 1000
= 4.18

For line 1:
Rough Bore Pipe

13

Q (l/min)
10
30

Q (m3/s)

P1 (psi)

P2(psi)

P (psi)

P (Pa)

V (m/s)

1.67 x 10-4

22.602

22.644

0.042

289.6

0.53

2.062

5x 10-4

18.273

19.073

0.8

5515.8

1.59

4.364

8.33 x 10-4

12.527

14.632

2.105

14513.4

2.65

4.133

1.08 x 10-3

6.688

10.29

3.602

24834.8
3.45
Average K:

4.173
3.683

K (mtd 1)

50
65

1800 Bend
Q (l/min)

Q (m3/s)

P2 (psi)

P3
(psi)

P
(psi)

P (Pa)

V (m/s)

10

1.67 x 10-4

22.644

22.678

22.644

22.678

0.43

2.535

30

5x 10-4

19.073

19.132

19.073

19.132

1.3

0.481

50

8.33 x 10-4

14.632

14.866

14.632

14.866

2.17

0.685

65

1.08 x 10-3

10.29

10.72

10.29

10.72

K (mtd1)

2.82
Average K:

Sudden Contraction
P3 (psi)
P4
P
(psi)
(psi)

P (Pa)

V (m/s)

0.746
1.112

Q (l/min)

Q (m /s)

10

1.67 x 10-4

22.678

22.651

-0.027

-186.16

0.21

-8.443

30

5x 10-4

19.132

19.296

0.164

1130.7

0.63

5.698

50

8.33 x 10-4

14.866

15.263

0.397

2737.2

1.06

4.872

65

1.08 x 10-3

10.72

11.395

0.675

4653.9

Gradual Expansion
P4 (psi)
P5
P
(psi)
(psi)

P (Pa)

V (m/s)

22.651

22.616

-0.035

-241.32

0.21

-10.944

19.296

19.227

-0.069

-475.74

0.63

-2.397

Q (l/min)

Q (m /s)

10

1.67 x 10-4

30

5x 10

-4

1.37
Average K:

K (mtd1)

4.959
5.17633
K (mtd1)

14

50

8.33 x 10-4

15.263

15.123

-0.14

-952.26

65

1.08 x 10-3

11.395

11.073

-0.322

-2220.1
1.37
Average K:

Q (l/min)

Q (m3/s)

P6 (psi)

900 Elbow
P7
P
(psi)
(psi)

10

1.67 x 10-4

22.617

22.659

30

5x 10-4

19.358

50

8.33 x 10-4

65

1.08 x 10-3

1.06

-1.695
-2.366
1.615

P (Pa)

V (m/s)

K (mtd1)

0.042

289.6

0.43

3.133

19.416

0.058

399.9

1.3

0.473

15.285

15.72

0.435

2999.2

2.17

1.274

11.333

12.023

0.69

4757.4

P (Pa)

V (m/s)

2.82
Average K:

1.196
1.519

Q (l/min)

Q (m3/s)

P7 (psi)

900 Bend
P8
P
(psi)
(psi)

10

1.67 x 10-4

22.659

22.729

0.07

482.6

0.43

1.492

30

5x 10-4

19.416

19.45

0.034

234.4

1.3

0.277

50

8.33 x 10-4

15.72

15.81

0.09

620.5

2.17

0.264

65

1.08 x 10-3

12.023

12.225

0.202

1392.7

Q (l/min)

Q (m3/s)

Smooth Bore Pipe

P8 (psi)
P10
P
(psi)
(psi)

P (Pa)

V (m/s)

10

1.67 x 10-4

22.729

22.735

0.006

41.3

0.43

3.803

30

5x 10-4

19.45

19.667

0.217

1496.2

1.3

1.771

50

8.33 x 10-4

15.81

16.301

0.491

3385.3

2.17

1.438

65

1.08 x 10-3

12.225

13.029

0.804

5543.4

K (mtd1)

2.82
Average K:

2.82
Average K:

0.35
0.596
K (mtd1)

1.394
2.102

15

Rotameter
P22
P
(psi)
(psi)

P (Pa)

V (m/s)

Q (l/min)

Q (m /s)

P21 (psi)

10

1.67 x 10-4

22.522

22.223

0.299

2061.5

0.43

22.299

-4

K (mtd1)

30

5x 10

17.783

17.275

0.508

3502.5

1.3

4.145

50

8.33 x 10-4

11.432

10.593

0.839

5784.7

2.17

2.457

65

1.08 x 10-3

4.87

3.687

1.183

8156.4

2.82
Average K:

2.051
7.738

Table 6: Tabulation of results for Line 1 Devices

Figure 2: Plot of Pressure Drop Vs FLow Rate for line 1 devices. (Only the rough bore pipe is plotted
on the secondary axis)

16

Figure 3 : Plot of Pressure Drop Vs Velocity^2 for line 1 devices

Device

Equation of Lines in Fig 2

Gradient

K (mtd 2)

Rough Bore Pipe

y = 2094.2x - 90.246

2094.2

4.1884

1800 bend

y = 364.65x - 20.32

364.65

0.7293

Sudden Contraction

y = 2560.4x - 119.03

2560.4

5.1208

Gradual Expansion

y = -1053.6x - 65.9

1053.6

2.1072

900 Elbow

y = 619.84x 141

619.84

1.23968

900 Bend

y = 128.55x + 215.39

128.55

0.2571

Smooth Bore Pipe

y = 691.27x + 104.44

691.27

1.38254

Rotameter

y = 775.67x + 2057.4

775.67

1.55134

Table 7: Gradients of the plot for line 1 devices

For Line 2:
Diaphragm Valve

17

Q (l/min)

Q (m3/s)

10

1.67 x 10-4

22.457

21.395

1.062

7322.207

0.77

24.700

20

3.33 x 10-4

20.97

18.822

2.148

14809.89

1.54

12.489

25

4.17 x 10-4

20.507

18.321

2.186

15071.89

1.92

8.177

30

5.00 x 10-4

19.946

15.655

4.291

29585.3

2.31

11.089

P11 (psi)

P12(psi)

P (psi)

P (Pa)

V (m/s)

Average K:

K (mtd 1)

14.114

Ball Valve
Q (l/min)

Q (m3/s)

10

P12 (psi)

P13 (psi)

P (psi)

P (Pa)

V (m/s)

K (mtd1)

1.67 x 10-4

21.395

21.331

0.064

441.2629

0.77

1.488

20

3.33 x 10-4

18.822

18.554

0.268

1847.789

1.54

1.558

25

4.17 x 10-4

18.321

17.222

1.099

7577.312

1.92

4.111

30

5.00 x 10-4

15.655

15.58

0.075

517.105

2.31

0.194

Average K:

1.838

Globe Valve
Q (l/min)

Q (m3/s)

10

P13 (psi)

P14 (psi)

P (psi)

P (Pa)

V (m/s)

1.67 x 10-4

21.331

21.167

0.164

1130.736

0.77

3.814

20

3.33 x 10-4

18.554

17.777

0.777

5357.208

1.54

4.518

25

4.17 x 10-4

17.222

17.209

0.013

89.63153

1.92

0.049

30

5.00 x 10-4

15.58

13.442

2.138

14740.94

2.31

5.525

Average K:

K (mtd1)

4.619

Needle Valve
Q (l/min)

Q (m3/s)

10

P14 (psi)

P15 (psi)

P (psi)

P (Pa)

V (m/s)

1.67 x 10-4

21.167

19.332

1.835

12651.84

0.77

42.678

20

3.33 x 10-4

17.777

13.418

4.359

30054.14

1.54

25.345

25

4.17 x 10-4

17.209

9.649

7.56

52124.18

1.92

28.279

30

5.00 x 10-4

13.442

1.751

11.691

80606.33

2.31

30.212

Average K:

K (mtd1)

31.628

Rotameter
18

Q (l/min)

Q (m3/s)

10

P21 (psi)

P22 (psi)

P (psi)

P (Pa)

V (m/s)

K (mtd1)

1.67 x 10-4

20.335

20.013

0.322

2220.104

0.77

24.014

20

3.33 x 10-4

15.236

15.046

0.19

1309.999

1.54

3.461

25

4.17 x 10-4

10.486

10.045

0.441

3040.577

1.92

5.118

30

5.00 x 10-4

7.922

8.429

-0.507

-3495.63

2.31

-4.137

Average K:

8.148

Table 8: Tabulation of results for Line 2 Devices

Figure 4: Plot of Pressure Drop Vs FLow Rate for line 2 devices.

19

Figure 5: Plot of Pressure Drop Vs Velocity^2 from Line 2 devices

Device

Equation of Lines in
Fig 5

Gradient

K (mtd 2)

Diaphragm Valve

y = 4352.8x + 3653.1

4352.8

8.7056

Ball Valve

y = 348.22x + 1552.3

348.22

0.69644

Globe Valve

y = 2350.2x - 1713.4

2350.2

4.7004

Needle Valve

y = 14468x + 502.97

14468

28.936

Rotameter

y = -3121.5x + 3749.7

3121.5

6.243

Table 9: Gradients of the plot for line 2 devices

20

For Line 3:
Orifice Meter
Q (l/min)

Q (m3/s)

10

1.67 x 10-4

22.308

22.27

0.038

261.9999

1.31

0.305

20

3.33 x 10-4

21.893

20.653

1.24

8549.469

2.63

2.472

25

4.17 x 10-4

20.288

18.872

1.416

9762.942

3.29

1.804

30

5.00 x 10-4

19.778

17.185

2.593

17878.04

3.95

2.292

P16 (psi)

P17(psi)

P (psi)

P (Pa)

V (m/s)

K (mtd 1)

Average K:

1.718

Venturi Meter
Q (l/min)

Q (m3/s)

10

1.67 x 10-4

22.295

22.255

0.04

275.7893

1.31

0.321

20

3.33 x 10-4

20.723

20.602

0.121

834.2627

2.63

0.241

25

4.17 x 10-4

18.027

17.837

0.19

1309.999

3.29

0.242

30

5.00 x 10-4

19.366

19.146

0.22

1516.841

3.95

0.194

Average K:

0.250

P18 (psi)

P20 (psi)

P (psi)

P (Pa)

V (m/s)

K (mtd1)

Rotameter
Q (l/min)

Q (m3/s)

10

1.67 x 10-4

22.26

21.959

0.301

2075.315

1.31

21.439

20

3.33 x 10-4

20.593

20.202

0.391

2695.841

2.63

7.123

25

4.17 x 10-4

19.83

19.378

0.452

3116.419

3.29

5.246

30

5.00 x 10-4

19.131

18.604

0.527

3633.524

3.95

4.300

P21 (psi)

P22 (psi)

P (psi)

P (Pa)

V (m/s)

Average K:

K (mtd1)

9.527

Table 10: Tabulation of results for Line 3 Devices

21

Figure 6: Plot of Pressure Drop Vs FLow Rate for line 3 devices (Only the venturi meter is plotted on
the secondary axis)

Figure 7: Plot of Pressure Drop Vs Velocity^2 from Line 3 devices (only venturi meter is plotted on
the secondary axis)

22

Device

Equation of Lines in Fig 6

Orifice Meter

y = 1198x - 1386.9

Venturi Meter
Rotameter

Gradient

K (mtd 2)
1198

2.396

y = 92.065x + 177.28

92.065

0.184

y = 1036.8x + 1887.9

1036.8

2.074

Table 11: Gradients of the plot for line 2 devices

Table 12: Values of K for various devices

23

Device

K (mtd 1)

K (mtd 2)

K ave

Rough Bore Pipe

3.683

4.188

3.936

1800 bend

1.112

0.729

0.921

Sudden Contraction

5.176

5.121

5.149

Gradual Expansion

1.615

2.107

1.861

900 Elbow

1.519

1.240

1.379

900 Bend

0.596

0.257

0.427

Smooth Bore Pipe

2.102

1.383

1.742

Diaphragm Valve

14.114

8.706

11.410

Ball Valve

1.838

0.696

1.267

Globe Valve

4.619

4.700

4.660

Needle Valve

31.628

28.936

30.282

Orifice Meter

1.718

2.396

2.057

Venturi Meter

0.250

0.184

0.217

ROtameter

7.738

8.148

9.527

1.551

6.243

2.074

5.880

V Discussion
1.
From the pressure drop versus flow rate graphs as illustrated in figure 1, 3 and 5,
it can be observed that generally pressure drop increases as flow rate increases. This is
because pressure drop is directly proportional to the square of the velocity as seen in the
equation

g c P
V2
K
. Increasing flow rate also increases the velocity as the diameter

of respective devices is fixed.

24

The loss coefficient is obtained by the equation based calculation (method 1) and
graphical determination (method 2). In method 1, K was calculated using the equation,
g c P
V2
K
where V2 is determined by the flow rate used. The K is than evaluated at

all the flow rate used and arithmetic K average is found. In method 2, P versus V2 was
plotted using the best fit method (by linear regression) and the gradient of the line was
obtained. With the gradient, K was calculated. Since two different methods are used to
determine K, it is expected to obtain K values which differ slightly.
The loss coefficient K is used to express the amount of head loss associated with
various devices, pipes and fitting in addition to flow separation and mixing as these
obstruct the smooth flow of the fluid. A large K means greater energy loss with the device
or fitting and small K means lesser energy loss. Generally, devices and fittings of smaller
diameter, larger loss coefficients are recorded while devices and fittings of larger
diameter have smaller loss coefficient. With respect to pipes, a rough bore pipe has a
larger loss coefficient as more work has to be done against friction while a smooth bore
pipe. Valves also have a higher loss coefficient. These observations are justified as
devices with small internal diameters and valves by the structural design nature interrupt
fluid flow and rough pipes increase friction which results in the increase of loss
coefficient.
Significance of Reynolds number is ignored as all experiments were conducted
as high Reynolds number and hence this factor will not significantly affect the loss
coefficient.
Relative advantages and disadvantages of various discharge measuring devices Orifice
Meter, Venturi Meter and Rotameter.
Device

Advantages

Disadvantages

25

Orifice

1. Cheap, small and convenient

meter

2. Easy to install and maintain

3. Greater flexibility to change

1. Large power consumption in the

form of irrecoverable pressure loss
2. High permanent pressure loss due to

throat to pipe diameter ratio to

the uncontrolled expansion

measure a larger range of flow

downstream from the metering

rates
4. Extensive industrial use

3. Deposits near the orifice plate may

decrease the lifespan of the meter
4. Has a small and simple geometry,
which cannot be used to measure
large flows.

Venturi
meter

1. Very accurate for large range of

flows and has very little
frictional loss
2. Has a conical diffuser section,

installation more difficult

2. It only has a limited range of

which reduces head loss.

pressure drop because it is only

Therefore, only small pressure

constructed for certain flow rates

drop is required.
Rotameter

1. Heavy and bulky hence making

3. It is expensive

1. Simple device that can be mass

manufactured out of cheap
materials
2. Requires no external power or

1. Heavy, bulky and unsuitable for

large flow rates
2. Due to its use of gravity, a rotameter

fuel, it uses only the inherent

must always be vertically oriented

properties of the fluid, along

and right way up, with the fluid

with gravity, to measure flow

flowing upward

rate
3. Allows the flow rate to be

3. Is not easily adapted for reading by

machine

adjusted
2)
We classify all the pipe fittings into 4 categories: pipes, valves, meters and other pipe fittings.
Comparison of K between Valves, Meters and Pipes
Generally the K values for valves are much greater than the values of meters and
pipes. This can be attributed to the fact that the internal diameter and hence the cross sectional
area of the valve are reasonably smaller than that of meters and pipes (With reference to table 1).
26

With smaller internal diameter, the restrictions to fluid flow increases which results in greater
pressure drop. Given the same flow rate, a smaller diameter for fluid flow increases the velocity
of fluid flow. Frictional losses are proportional to fluid velocity hence increasing fluid velocity
(by increasing flow rate) increases the frictional loss. Hence, all valve devices have high loss
coefficient. However, the sudden contraction has a relatively high loss coefficient of 5.149 as the
sudden contraction of diameter creates eddy currents and vortex rings within the fluid flow.
Comparison of K between pipes
The rough bore pipe has a higher loss coefficient as compared to the smooth bore
pipe. The rough bore pipe has higher friction and hence more work has to be done to overcome
friction as compared to smooth bore pipe.
Comparison of K between bends and elbows
The K value for the 90 elbow was the highest, followed by that of the 180 bend, and
lastly the 90 bend. The more abrupt the directional change in fluid flow, the greater the loss in
energy, giving rise to a higher loss coefficient. These energy losses can be attributed to the
formation of eddy currents. Hence for the 90 degree elbow, the change in fluid flow is the most
abrupt resulting in the higher pressure loss. Similarly, comparing the 180 degree and 90 degree
bend, the 180 degree bend has a larger loss coefficient due to the larger change in magnitude of
fluid flow.
Comparison of gradual expansion and sudden contraction

From theoretical knowledge, it is expected that the loss coefficient for sudden
contraction to be larger than that of gradual expansions as in sudden contraction the fluid
enters a constriction resulting in greater energy loss. This result is also proven in the
experiments.

Comparison of K between valves

The K values of the valves are in increasing order of Ball valve, Globe valve, Diaphragm
valve and Needle valve with the largest K value. The needle valve has a relatively small orifice
with a long, tapered, conical seat. A needle-shaped plunger, on the end of a screw, exactly fits this
seat. This results in the greatest constriction of the fluid flow giving rise to the largest loss
coefficient. The diaphragm valve closes by a diaphragm, which actually sits on top of a

27

saddle-like structure of the valve. Likewise, this causes a greater disturbance in fluid flow
hence having a reasonable loss coefficient. However, both the ball valve and globe valve
have relatively low loss coefficients due to the rounded interiors which do not
significantly obstruct the fluid flow and reduces energy loss.
Comparison of K between discharge measuring devices
The K value for the orifice meter was much higher than the one for the venturi meter. (10
times larger) This result can be attributed to the structural design of the discharging measuring
device.
The cross sectional expansions and contractions are gradual in venturi meter which gives
rises to small loss coefficients as fluid flow is not significantly disrupted and reducing energy
loss.
However, in the orifice meter, the position of the orifice plate collides with the fluid flow
head on. In addition, the formation of eddy currents at the downstream of the orifice meter also
contributes to the head loss.
3.
g c P
V2
K

------------- (1)

Equation 1 can be rearranged to express the equations in terms of V.Therefore,

Vi = (gcP * 2) / (K*)
Qi = Vi * A

where i refers to individual device

Q1 average = ( Qi ) / i
QT = Q1 + Q2 + Q3

QT = 75 l/min = 0.00125 m3/s

Device

P (Pa)

Kave

V2 (m2/s-2)

V(m/s)

A (m2 )

Qi

Line 1
Rough Bore Pipe
180 bend

6474.15

3.936

3.29

1.81 0.000314

5.70 x 10-4

744.63

0.921

1.62

1.27 0.000384

4.88 x 10-4
28

1420.32

5.149

Gentle Expansion

-599.84

1.861 -

90 elbow

1323.79

1.379

1.92

1.39 0.000384

5.32 x 10-4

90 Bend

448.16

0.427

2.10

1.45 0.000384

5.56 x 10-4

1158.32

1.742

1.33

1.15 0.000384

4.42 x 10-4

Smooth Bore pipe

0.55

5.86 x 10-4

Sudden Contraction

0.74 0.000789
-

0.000384 -

Q1 Average : 5.29 x 10-4

Line 2
3943.79

11.41

0.69

0.83 0.000216

1.80 x 10-4

503.32

1.27

0.79

0.89 0.000216

1.93 x 10-4

Globe Valve

1565.10

4.66

0.67

0.82 0.000216

1.77 x 10-4

Needle Valve

10493.78

30.28

0.69

0.83 0.000216

1.80 x 10-4

Diaphragm Valve
Ball Valve

Q2 Average

1.83 x 10-4

Line 3
Orifice Meter

22876.72

2.06

22.24

4.72 0.000127

5.98 x 10-4

Venturi Meter

1647.84

0.22

15.19

3.90 0.000127

4.94 x 10-4

Q3 Average

5.46 x 10-4

Table 13: Calculations of individual flow rates for flow of 75 l/min

QT = Q1 + Q2 + Q3
= 5.29 x 10-4 + 1.83 x 10-4 + 5.46 x 10-4
= 0.001258 m3/s

QT = 60 l/min = 0.001 m3/s

Device

P (Pa)

Kave

V2 (m2/s-2)

V(m/s)

A (m2 )

Qi

Line 1
Rough Bore Pipe
180 bend

3578.37

3.94

1.82

1.35 0.000314

4.24 x 10-4

655.00

0.92

1.42

1.19 0.000384

4.57 x 10-4

29

Sudden Contraction
Gentle Expansion

592.95
-234.42

5.15

0.23

1.86 -

3.79 x 10-4

0.48 0.000789
-

0.000384 -

90 elbow

882.53

1.38

1.28

1.13 0.000384

4.34 x 10-4

90 Bend

82.74

0.43

0.39

0.62 0.000384

2.39 x 10-4

786.00

1.74

0.90

0.95 0.000384

3.64 x 10-4

Smooth Bore pipe

Q1 Average : 3.83 x 10-4

Line 2
2426.95

11.41

0.43

0.65 0.000216

1.41 x 10-4

Ball Valve

413.68

1.27

0.65

0.81 0.000216

1.75 x 10-4

Globe Valve

772.21

4.66

0.33

0.58 0.000216

1.25 x 10-4

Needle Valve

6446.58

30.28

0.43

0.65 0.000216

1.41 x 10-4

Diaphragm Valve

Q2 Average

1.45 x 10-4

Line 3
Orifice Meter

8549.47

2.06

8.31

2.88 0.000127

3.65 x 10-4

Venturi Meter

1868.47

0.22

17.22

4.15 0.000127

5.26 x 10-4

Q3 Average

4.46 x 10-4

Table 14: Calculations of individual flow rates for flow of 60 l/min

QT = Q1 + Q2 + Q3
= 3.83 x 10-4 + 1.45 x 10-4 + 4.46 x 10-4
= 0.000974 m3/s

QT = 40 l/min = 0.000667 m3/s

Device

P (Pa)

Kave

V2 (m2/s-2)

V(m/s)

A (m2 )

Qi

Line 1
1461.68

3.94

0.74

0.86 0.000314

2.71 x 10-4

180 bend

489.53

0.92

1.06

1.03 0.000384

3.96 x 10-4

Sudden Contraction

124.11

5.15

0.05

0.22 0.000789

1.73 x 10-4

Rough Bore Pipe

30

Gentle Expansion

-393.00

1.86 -

0.000384 -

90 elbow

234.42

1.38

0.34

0.58 0.000384

2.24 x 10-4

90 Bend

82.74

0.43

0.39

0.62 0.000384

2.39 x 10-4

310.26

1.74

0.36

0.60 0.000384

2.29 x 10-4

Smooth Bore pipe

Q1 Average : 2.55 x 10-4

Line 2
Diaphragm Valve

979.05

11.41

0.17

0.41 0.000216

8.96 x 10-5

Ball Valve

110.32

1.27

0.17

0.42 0.000216

9.03 x 10-5

Globe Valve

503.32

4.66

0.22

0.46 0.000216

1.01 x 10-4

Needle Valve

3150.89

30.28

0.21

0.46 0.000216

9.87 x 10-5

Q2 Average

9.48 x 10-5

Line 3
Orifice Meter

6115.63

2.06

5.95

2.44 0.000127

3.09 x 10-4

Venturi Meter

461.95

0.22

4.26

2.06 0.000127

2.61 x 10-4

Q3 Average

2.85 x 10-4

Table 15: Calculations of individual flow rates for flow of 40 l/min

QT = Q1 + Q2 + Q3
= 2.55 x 10-4 + 9.48 x 10-5 + 2.85 x 10-4
= 0.000635 m3/s

QT = 20 l/min = 0.000333 m3/s

Device

P (Pa)

Kave

V2 (m2/s-2)

V(m/s)

A (m2 )

Qi

Line 1
448.16

4.19

0.21

0.46 0.000314

1.45 x 10-4

180 bend

82.74

0.73

0.23

0.48 0.000384

1.83 x 10-4

Sudden Contraction

13.79

5.12

0.01

0.07 0.000789

5.79 x 10-5

Rough Bore Pipe

Gentle Expansion

-103.42

2.11 -

0.000384 -

31

90 elbow

89.63

1.24

0.14

0.38 0.000384

1.46 x 10-4

90 Bend

110.32

0.26

0.86

0.93 0.000384

3.55 x 10-4

Smooth Bore pipe

117.21

1.38

0.17

0.41 0.000384

1.58 x 10-4

Q1 Average : 1.74 x 10-4

Line 2
372.32

8.71

0.09

0.29 0.000216

6.33 x 10-5

48.26

0.70

0.14

0.37 0.000216

8.06 x 10-5

Globe Valve

248.21

4.70

0.11

0.32 0.000216

7.03 x 10-5

Needle Valve

517.10

28.94

0.04

0.19 0.000216

4.09 x 10-5

Diaphragm Valve
Ball Valve

Q2 Average

6.38 x 10-5

Line 3
Orifice Meter

1027.32

2.40

0.86

0.93 0.000127

1.17 x 10-4

Venturi Meter

227.53

0.18

2.47

1.57 0.000127

1.99 x 10-4

Q3 Average

1.58 x 10-4

Table 16: Calculations of individual flow rates for flow of 20 l/min

QT = Q1 + Q2 + Q3
= 1.74 x 10-4 + 6.38 x 10-5 + 1.58 x 10-4
= 0.000396 m3/s
QT ( m3/s)

Q1 (m3/s)

Q2 (m3/s)

Q3 (m3/s)

QT (Calculated) (m3/s)

% error

0.00125

5.29 x 10-4

1.83 x 10-4

5.46 x 10-4

0.001258

0.64

0.001

3.83 x 10-4

1.45 x 10-4

4.46 x 10-4

0.000974

-2.67

0.000667

2.55 x 10-4

9.48 x 10-5

2.85 x 10-4

0.000635

-5.04

0.000333

1.74 x 10-4

6.38 x 10-5

1.58 x 10-4

0.000396

15.90

Since the percentage errors are relatively small, the approximate individual flow rate in
each line can be taken as the summarized values in the table above. The relative low
percentage errors suggest that the K values obtained earlier on were rather accurate.

32
Table 17: Comparison of actual flow rates and calculated flow rates

Error Analysis
1. Assumptions
-

Fully developed laminar flow

Flow is incompressible

No shaft work is done on or by the system

2. The pressure reading recorded by the electronic reading has a precision up to 3

decimal place. The pressure readings do not reach a stable value and it keeps
fluctuating. Hence the average of the maximum and minimum value is determined.
However, this is largely dependent on human judgment.
3. Calculation of K is assumed as constant density but in reality, its a function of
temperature and there is a possibility of temperature variation due to frictional forces
acted upon on the fluid by the pipes.

33

VI Conclusion
The loss coefficients for the various fittings and devices were calculated from the
experimental data. The K values obtain were then used to calculate the flow rate through
each fitting. The results obtained were then compared with the flow rate recorded by the
rotameter. As discussed in question 3 of the discussion section, the flow rates calculated
are sufficient close to the recorded values. The percentage errors were found to be rather
low. Thus we can conclude that the experiment was fairly accurate.
From this experiment, we learned how the geometry of a fitting affects the head
loss. In general, it was observed that if there was a sudden increase or decrease in the
cross sectional area of the pipe, the loss coefficient would be greater than if these changes
were gradual. The needle valve was found to be the device which posses the highest K
value.
A better understanding of the various devices and the various flow meters was
achieved. Familiarization of the applications and limitations of the flow meters was also
obtained.
VII References
1. Fox, R.W., McDonald, A.T., Pritchard, P.J., Fundamentals of Fluid Mechanics,
6th Ed., Wiley (2004)
2. Welty, J.R., et al, Fundamental of momentum, heat and mass transfer, 4th ed.
Wiley (2001)
3. Gerhart, PM & Gross, RJ, "Fundamentals of Fluid Mechanics", Addison-Wesley.
4. Munson, BR, Young, DF & Okiishi, TH, "Fundamentals of Fluid Mechanics",
John Wiley & Sons.
5. Mays, L. W. editor. 1999. Hydraulic design handbook. McGraw-Hill Book Co

34

Appendix

Needle Valve

From the above picture, it can be seen that the needle like plunger gives rise
to significant constriction fluid flow giving rise to the largest loss coefficient.
(http://www.tpub.co
css/h1018v2_54.htm

m/content/doe/h1018v2/
)

Diaphragm Valve

35

The diaphragm valve closes by a diaphragm, which actually sits on top of a

saddle-like structure of the valve which obstructs does not obstruct the fluid
flow as significantly as by the needle valve.
(http://www.roymech.co.uk/Related/Valves.html )
Globe valve

The rounded

structure reduces

energy loss.
(http://www.valve

diagnostics.com/medi

a/pictures/globe.

gif)

Ball Valve
36

Rounded surface of the ball reduces energy loss hence giving rise to a small
loss coefficient.
(http://www.spiraxsarco.com/resources/steam-engineering-tutorials/pipelineancillaries/isolation-valves-rotary-movement.asp)

Venturi Meter

The gradual expansions and contractions in the venturi meter can be

observed above.

37

(http://www.flowmeterdirectory.com/images/flowmeter_prof_001.gif)

Orifice Meter

The formation of eddy currents gives rise to a larger loss coefficient.

(http://instrumentation.co.za/Articles/SA%20Instrumentation%20&%20Control
%20-%20Published%20by%20Technews/77c123b.png )

Part II
38

Centrifugal Pump
Characteristics

39

Summary:
The objective of this experiment was to determine the performance curve of the pump and
to verify pump laws. In this experiment, centrifugal pump was selected as the experimental
system and water was used as the fluid medium. Experiment was carried out at 6 different pump
speeds. At each speed, flow rate of water through the pump was varied between the minimum and
maximum possible flow rate. The recorded reading was then used to plot the performance curve
of the pump and to verify the pump laws.
For our experiments, we have concluded that centrifugal pump behaves ideally according to
Pump Affinity Law at higher pump speed.

40

CONTENTS
Part II: Centrifugal Pump
Page
Title

34

Summary

35

Introduction

37

II

Theoretical Background

38

III

Experimental set up and procedures

44

IV

Results and Analysis

46

Discussion

53

VI

Conclusion

55

VII

References

55

Appendices

41

I. Introduction:
In chemical industry, engineers often face the problems of pump failures. Thus in order to
protect the pumps and to choose the right system for use in the industry, it is important for
engineers to have a good understanding of the process as well as having thorough knowledge of
the mechanics of the systems. On top of that, they also need to have the ability to observe the
performance of the system over times. This can be done by conducting experiment to obtain the
performance curve of the pump, which is one of our main objectives of the experiment.
Another objective of this experiment was to verify pumps laws. Pumps law is the general
law that the centrifugal pump is obeying. This law states that the flow rate of liquid is directly
proportional to the pump speed, thus the pump speed in the industry is adjusted according to this
law. Therefore it is very important this law is verified so that the system designed in industry
would not be under or over-estimated from the ideal system.
In our experiment, we varied the pump speed for six times. For each pump speed, we
varied the flow rate six times as well using the control valve. Readings recorded were then used
to plot performance curve and to verify the pump law. Water was used as the fluid medium in this
experiment.

42

II. Theoretical background:

Centrifugal pump
A centrifugal pump is made up of two main components:
1. A rotating component made up of an impeller and a shaft.
2. A stationary component made up of a casing, casing cover and bearings.

Figure 8:General component of a Centrifugal Pump

43

Figure 9: General component of a Centrifugal Pump

Centrifugal pump converts the input power from turbine or electric motor into kinetic
energy in the liquid. This is done by accelerating the liquid with turbine or electric motor which
acts as the revolving device, also known as impeller. The kinetic energy is then changed into
pressure energy of liquid that is being pumped. Kinetic energy in the liquid is obstructed by
creating a resistance in the flow. The first resistance is created by the pump volute (casing) which
serves as an obstruction to the flow of liquid and thus slows the flow of liquid. The kinetic that
decreased as liquid flow decreases is converted into pressure energy.
Mechanisms

Figure 10: Liquid flow path in a centrifugal pump

Generation of Centrifugal Force:

The fluid medium enters through the suction nozzle into the center of the impeller. As the
impeller rotate, fluid is being rotated in between the vanes as well and thus centrifugal

44

acceleration resulted. As fluid leave the center of impeller, a region of low pressure is then
created. This will induce more fluid to flow in from a higher pressure region. This will thus
ensure the continuous flow of fluid into the impeller. As the impeller blades are curved, the fluid
is made to move in a tangential and radial direction due the centrifugal force. Energy of liquid
generated by this centrifugal force is termed as kinetic energy. When non-viscous liquid such as
water is being used as the fluid medium, head is used to measure the kinetic energy generated.
Head is measurement of the height of the liquid column with the kinetic energy of the liquid
created by the pump.

Conversion of Kinetic energy to Pressure energy

The kinetic energy coming out from the impeller is obstructed by a restricted created to
the flow and as mentioned during the earlier discussion, the first resistance is created by the pump
volate. Following that, the fluid further decelerates in the discharge nozzle. By Bernoullis
principle, the kinetic energy which decreased as flow of fluid decreases is being converted into
pressure energy.
Therefore head, which is a measure of kinetic energy, is approximately equal to the
velocity energy at the periphery of the impeller. Equation representing this relation is given by:
H = v2/ (2g), with
H= total head developed
v = velocity at periphery of impeller
g = acceleration due to gravity

Pump Performance Curve:

Pump performance curve relates pressure (head) and flow rate developed by the pump at
different impeller sizes and pump speed. This curve can be obtained by plotting total head H
against pump delivery Q.
The operation of the centrifugal pump can be studied by considering the angular
momentum of the fluid. With sufficient number of blades, it is assumed that the fluid flows in and

45

out of the impeller with relative velocity tangential to the blades. The velocity triangles for the
fluid particles entering and leaving the impeller are as shown.

Figure 81

Angular momentum of fluid entering and leaving the impeller:

Hi = r1m(u1cos1) and
Ho = r2m(u2cos2)

respectively

where
m : mass of fluid flowing through the impeller
u1 : the fluid velocity at the impeller inlet
u2 : the fluid velocity at the impeller outlet
1 : the angle between u1 and the tangential direction at the blade
2 : the angle between u2 and the tangential direction at the blade
The torque acting on the fluid, , is the rate of change of angular momentum with time,
= Ho - Hi
= m(r2 u2cos2 - r1 u1cos1)
Therefore, the power required is

46

P =
= m(r2 u2cos2 - r1 u1cos1)
where is the (angular speed) rotational speed of the impeller.

The fluid output power is:

Pf = m g h
where h is the theoretical head of the fluid
Compare P with Pf :
h = (1/g)(r2 u2cos2 - r1 u1cos1) - (Eulers Equation)

Assuming there is no pre-rotation at the impeller inlet, r 1=0, the above equation is further
simplified to:
h = r2 u2 cos 2 / g

(1)

From the velocity triangle at the outlet of the impeller,

u2 cos 2 = ut2 - ur2 / tan (2)
and since ur2 is directly proportional to the flow rate Q,
u2 cos 2 = ut2 - C Q

(2)

where C is a constant for the pump.

Combining both equation (1) and (2):

h = r2 (ut2 - C Q) / g

(3)

A plot of the above equation will results a linear relationship between H and Q (Euler Line).
However the actual characteristics curve of the centrifugal pump shows a sharper decline of the
head H over the increase of flow rate Q. This is shown in the figure below:

47

Figure 12: Expected plot of H vs Q

The possible factors that might bring about the sharp decrease of the head H are:
1. Prerotation of fluid on entering the impeller
2. Interblade rotation of the fluid
3. Losses at entrance of the impeller and in the subsequent diffusing process
4. Leakage through the impeller

Pump Affinity Laws

The performance of a centrifugal pump is affected by the change in speed of pump as well
as the diameter of impeller. The technique dimensional analysis is applied to the study of the
characteristics of the centrifugal pump operation to produce useful results. The basic quantities
involved in the pump operation with similar geometrics are:
Q : Flow rate
N : Rotational Speed
D : Impeller Diameter
: Fluid Density
: Viscosity of the fluid
H : Increase in fluid head

48

The general function of the pump operation will have the form:
f ( Q, N, D, r, , H) = 0
Applying dimensional analysis, we obtain the following relations:
Q = C1 x N x D3
H = C2 x N2 x D2 x
and since power required for the pump is directly proportional to the product of Q and H :
P = C3 x N3 x D5 x

With the diameter of the impeller (D) and density of fluid () kept constant, the 3 affinity
laws can be written as
1.

Q2/Q1 = N2/N1

2.

H2/H1 = (N2/N1)2

3.

P2/P1 = (N2/N1)3

Hence the Pump Affinity Laws states that

1. The flow rate or capacity is directly proportional to the pump speed
2. The discharge head is directly proportional to the square of the pump speed
3. Power required by the pump motor is directly proportional to the cube of the pump speed.

49

III Experimental Procedure

Flow Chart of Experimental Procedures
Close the control valve at the discharged
side and fully open the control valve at
the suction side of pump

Set the initial reading of the

load cell to zero

Set the frequency output of the

inverter to a ratio of approximately
50%

Set the control switch to of

position and switch on the power
supply

Set the control switch of the

inverter to run position.

Fine tune the inverter output so that the

Control the
flow of
the pump
tachometer
display
indicates
a speed of 1450
once the pump has accelerated to a steady
Repeat the above steps with diferent
state
pump speeds

50

51

3.1 Detailed Procedures

1. The control valve at the discharge side was closed and the control valve at the suction
side of pump was fully opened.
2. The initial reading of the load cell was set to zero by pressing the mode and zero key
simultaneously.
3. The frequency output of the inverter was set to a ratio of approximately 50%.
4. The control switch was set to off position.
5. The power supply to the electrical panel was switched on.
6. The control switch of the inverter was set to run position.
7. Once the pump had accelerated to a steady state, the inverter output was fine tuned so
that the tachometer display indicated a speed of 1450 rpm.
8. The flow of the pump was controlled by gradually opening the control valve at the
discharge side at small steps.
9. The readings of the measured values were taken down.
10. The above steps were repeated with pump speed changed to 1750, 2050, 2350, 2650 and
2900 rpm respectively.
Apparatus used:
1. Water tank
2. Pump
3. Motor
4. Control valve
5. Decorder
6. Torque Arm
7. Local Cell
8. Rotameter
9. Pressure Gauges
10. Electrical Panel

52

3.2 Precautions Taken

1.

Ensure that the control valve at the suction side of the pump was fully open and the valve at
discharge side was closed before you switch on the pump.

2.

Stand clear of the motor and pump when setting the control switch of the inverter to run
position. If the pump does not run after switch on, switch it off immediately and inform the
lab demonstrator.

3.

Do not touch any moving parts when the pump is running.

4.

If water is spilled around the electrical points, inform the lab demonstrator immediately.

IV Results and Calculation

Impeller diameter, D = 0.11m
Torque arm length, L = 0.14m
gc = 1 kgm m / N / s2 = 9.81 kgm m / kgf / s2

N(rev/min) = 1450

F
(kg)

Q (m3/h)

P0
(kgf/cm2)

Pi
(mmHg)

Hs
(m)

H
(m)

Pf
(W)

(N-m)

(rad/sec)

Pp (W)

E (%)

0.40

3.60

0.200

-100

3.36

3.36

32.95

0.55

151.90

83.55

39.44

0.38

3.00

0.225

-100

3.61

3.61

29.50

0.52

151.90

78.99

37.35

0.36

2.00

0.250

-100

3.86

3.86

21.03

0.49

151.90

74.43

28.26

0.34

1.00

0.275

-100

4.11

4.11

11.20

0.47

151.90

71.39

15.68

(N-m)

(rad/sec)

Pp (W)

E (%)

Table 188: Measured and calculated values at 1450rpm

N(rev/min) = 1750

F
(kg)

Q (m3/h)

P0
(kgf/cm2)

Pi
(mmHg)

Hs
(m)

H
(m)

Pf
(W)

53

0.54

5.00

0.300

-125

4.70

4.70

64.02

0.74

183.33

135.66

47.19

0.50

4.00

0.350

-100

4.86

4.86

52.96

0.69

183.33

126.50

41.87

0.60

3.00

0.400

-100

5.36

5.36

43.81

0.82

183.33

150.33

29.14

0.44

2.00

0.425

-100

5.61

5.61

30.57

0.60

183.33

110.00

27.79

Table 19: Measured and calculated values at 1750rpm

N(rev/min) = 2050

F
(kg)

Q (m3/h)

P0
(kgf/cm2)

Pi
(mmHg)

Hs
(m)

H
(m)

Pf (W)

(N-m)

(rad/sec)

Pp (W)

E (%)

0.66

6.00

0.425

-150

6.29

6.29

102.82

0.91

214.76

195.43

52.61

0.62

5.00

0.500

-125

6.70

6.70

91.27

0.85

214.76

182.55

50.00

0.58

4.00

0.550

-100

6.86

6.86

74.76

0.80

214.76

171.81

43.52

0.54

3.00

0.600

-100

7.36

7.36

60.16

0.74

214.76

158.92

37.86

Table 20: Measured and calculated values at 2050rpm

N(rev/min) = 2350

F
(kg)

Q (m3/h)

P0
(kgf/cm2)

Pi
(mmHg)

Hs
(m)

H (m)

Pf (W)

(N-m)

(rad/sec)

Pp (W)

E (%)

0.84

7.00

0.575

-150

7.79

7.79

148.57

1.15

246.19

283.12

52.47

0.74

5.00

0.725

-125

8.95

8.95

121.93

1.02

246.19

251.11

48.56

0.66

3.00

0.825

-100

9.61

9.61

78.55

0.91

246.19

224.03

35.06

0.58

1.00

0.900

-75

10.02

10.02

27.30

0.80

246.19

196.95

13.86

(N-m)

(rad/sec)

Pp (W)

E (%)

Table 21: Measured and calculated values at 2350rpm

N(rev/min) = 2650

F
(kg)

Q (m3/h)

P0
(kgf/cm2)

Pi
(mmHg)

Hs
(m)

1.00

8.00

0.750

-200

10.22

10.22 222.75

1.37

277.62

380.34

58.57

0.90

6.00

0.925

-150

11.29

11.29

184.57

1.24

277.62

344.25

53.61

0.80

4.00

1.050

-100

11.86

11.86

129.26

1.10

277.62

305.38

42.33

H (m)

Pf (W)

54

0.70

2.00

1.125

-100

12.61

12.61

68.72

0.96

277.62

266.52

25.78

Table 22: Measured and calculated values at 2650rpm

N(rev/min) = 2900

F
(kg)

Q (m3/h)

P0
(kgf/cm2)

Pi
(mmHg)

Hs
(m)

H (m)

Pf (W)

(N-m)

(rad/sec)

Pp (W)

E (%)

1.18

10.00

0.800

-250

11.40

11.40

310.58

1.62

303.81

492.17

63.11

1.10

8.00

1.000

-200

12.72

12.72 277.25

1.51

303.81

458.75

60.44

0.98

6.00

1.175

-150

13.79

13.79 225.44

1.35

303.81

410.14

54.97

0.86

4.00

1.300

-100

14.36

14.36 156.51

1.18

303.81

358.50

43.66

Table 23: Measured and calculated values at 2900rpm

Calculation
Sample calculation of first experimental run
D = 0.11 m, Q = 10.00 m3/hr (max value)
Fluid Velocity, v = 4Q / D2
= 0.292 m/s
Fluid Velocity Head, Hv = v2 / 2g
= 4.35x 10-3 m (negligible compared to Hs)

Suction Pressure, Pi = -100 mmHg

= -100 x 1/760 x 1.01325 x 105 / 100 / 100 / 9.81
= -0.136 kgf / cm2

Discharge Pressure, Po = 0.200 kgf / cm2

55

Po Pi
( = 1000 kg/m3 for water, g = 9.81 m/s2)
g

Fluid Static Head, Hs =

= [0.200 (-0.136)] x 9.81 x 104 / (1000 x 9.81)

= 3.36 m >> Hv
Fluid Total Head, H = Hs + Hv Hs = 3.36 m

g H Q
3600

Fluid Power, Pf =

= 1000 x 9.81 x 3.36 x 3.60 / 3600

= 32.96 W

L = 0.14 m, F = 0.40 kg
Torque,

= FxL
= (0.40 x 9.81) x 0.14
= 0.55 N-m

N = 1450 rpm
Angular velocity, = 2 x 22/7 x N

1 rps
60 rpm

= 2 x22/7 x 1450 / 60
= 151.90 rad / sec
Pump Power, Pp = x
= 0.55 x 151.90
= 83.55 W

56

Pump Efficiency, E =

Pf
Pp

100%

= 32.95 / 83.55 x 100%

= 39.44%

Figure 19: Plot of H against Q

Pump Affinity Laws

57

Tabulation and Discussion

N=1450rpm

N=1750rpm

N=2050rpm

Q/N

H/N2

Q/N

H/N2

Q/N

H/N2

0.0024828

1.59764E-06

0.00285714

1.5343E-06

0.00292683

1.49639E-06

0.002069

1.71655E-06

0.00228571

1.58663E-06

0.00243902

1.59401E-06

0.0013793

1.83546E-06

0.00171429

1.74989E-06

0.00195122

1.63213E-06

0.0006897

1.95436E-06

0.00114286

1.83153E-06

0.00146341

1.75111E-06

Table 24: Tabulation of Q/N and H/N2

N=2350rpm

N=2650rpm

N=2900rpm

Q/N

H/N2

Q/N

H/N2

Q/N

H/N2

0.0029787

1.41033E-06

0.00301887

1.45505E-06

0.00344828

1.35525E-06

0.0021277

1.62043E-06

0.00226415

1.60749E-06

0.00275862

1.51226E-06

0.0012766

1.73998E-06

0.00150943

1.68872E-06

0.00206897

1.63954E-06

0.0004255

1.81427E-06

0.00075472

1.79552E-06

0.00137931

1.70738E-06

Table 25: Tabulation of Q/N and H/N2

58

Figure 110: Plot of Q/N against H/N2

Discussion

As can be observed from Figure 6 above, the curves did converge but did not converge to
one curve. The curves are relatively closer to each other at lower values of H/N 2.
This deviation at higher values could be due to systematic and experimental error.

The graph of Q/N against H/N2 should converge due to the following reason:
Q = C1 x N x D3
H = C2 x N2 x D2 x , according to the affinity laws.

59

Rearranging the above equations:

Q/N = C1 x D3
(C1 is the pump constant and D is the impeller diameter)
H/N2= C2 x D2 x
(C2 is the pump constant, D is the impeller diameter)

Therefore, a plot of Q/N and H/N2 should converge as they are independent of pump
speed. Figure 6 ,hence, verified the first and second pump affinity laws. The third pump affinity
law (P = C3 x N3x D5 x ) is thus also verified as it is directly proportional to the product of Q and
H which are the first and second pump affinity laws.
Q = C1 x N x D3
H = C2 x N2 x D2 x
P = C3 x N3x D5 x

The above 3 pump affinity laws are useful as they allow us to predict the effect of varying
pump speed, fluid density and etc on the flow rate, head and power of the pump. For example, if
pump speed was increased by 10% (keeping all other variables constant)
Flow rate, Q will increase by 1.1 times
Head, H will increase by: (1.1)2 = 1.21 times
Power, P will increase by: (1.2)3 = 1.331 times.

Error Analysis
60

The experimental data deviates from the ideal case due to the following inevitable systematic and
experimental errors.
1. The pump efficiency laws are based on ideal fluid which has no viscosity. Inaccuracy of
data could have arised from the non-ideality of water.
2. Constant fluctuations in the readings of N and flow rates were observed. This could be
due to the turbulent flow in the system which affected the sensitivity of the tachometer
and rotameter. An average reading was taken and hence this could compromise the
accuracy of the data collected.
3.

It was prone to parallax error while taking the readings of pressure and flow rates. Also,
the scale of the pressure gauge was large and it could only be used to read for a multiple
of 25. These errors could have been minimized with the usage of digital pressure meters.

4.

It was also noted when the flow rate was varied, the pump speed also changed.
These fluctuations might have affected the accuracy of the collected data.

V Discussion

From the experimental plot of H against Q, it is observed that the curve of H against Q
concaves downwards. Hence the curve deviates from the theoretical Eulers linear line. This is
because experimental total heat loss includes the loss due to pre-rotation of the fluid entering
impeller, inter-blade rotation and frictional losses during diffusion of the fluid whereas the
theoretical head loss does not account for such losses. The following figure shows the types of
losses within the centrifugal pump which accounts of the deviation.

61

Figure 111: Effects of losses on the pump head against flow rate curve

These losses were resulted due to the following reasons.

1. Shock losses arise from the turbulence created by the impact of the fluid against the
blades, friction between the fluid and the boundaries, recirculation of a small fluid
amount after leakage through the clearance spaces outside the impeller.
2. Eddies may result in some back flow into the inlet pipe, causing the fluid to have a whirl
before entering the impeller.
3. Hydraulic losses due to pipe friction and pipe connection such as valves and meters.
It is also observed that as the rotating speed of the pump increases, the curves shift outwards.
This is because higher rotating speed results in fluid with higher angular speed and momentum,
this in turn, result in higher fluid output power and hence a larger head value.
When the points with the same efficiency were joined, the resulted curves are convoluted.
Theoretically, the efficiency curves should result in similar two halves for each efficiency value
as shown in the figure below.

62

Figure 112: Pump efficiency curves

This is not the case in our experimental plot as the range of flow rates was not large
enough to form the complete curves. Only half of the efficiency curves was obtained during our
experiment and the best efficiency point could not be determined. We should have used a larger
range of flow rates so that the best efficiency point could be estimated.

VI Conclusion

From this experiment, we were able to determine the performance curve of the pump at different
pump speeds and fluid flow rates. The shape of the actual pump curves were then compared with
the Euler line. It was found that as the speed increases, kinetic energy also increases, which was
indicated by H. However, as flow rate increases, H decreases due to increased resistance by the
first casing on water. From the performance curve, we also concluded that at higher pump speeds,
the process is more efficient and more consistent with the theoretical pump.

The Pump Affinity Laws were verified by plotting Q/N against H/N 2 for various pump speeds.
From our experimental curves, the law is verified when pump speed is at high value. Hence, only
at higher pump speed would the centrifugal pump tend towards ideality according to the Pump
Affinity Law. The curves were observed to converge to a single curve, which shows that Q/N and
H/N2 are constant for different pump speeds as the pump constant C and impeller diameter D

63

were kept constant for all experimental readings. Thus, the 3 pump affinity laws were verified and
found to be accurate.

VII References

1. Welty, Wicks, Wilson, Rorrowe, Fundamental of momentum, heat and mass

transfer, 5th ed. Wiley(2008)
2. Fox, R.W., McDonald, A.T., Pritchard, P.J., Fundamentals of Fluid Mechanics, 6th
Ed., Wiley (2004)
3. Munson, BR, young, DF & Okiishi, TH, Fundamentals of Fluid Mechanics, John
Wiely& Sons.
4. Massey, B.S., Mechanics of Fluids, 3rd ed., Van Nostrand (1975)

64