Asian Journal of Science and Appelied Technology (AJSAT)

Vol1.No.1 2013 pp 12-17.

available at: www.goniv.com
Paper Received :25-02-2013
Paper Published:18-05-2013
Paper Reviewed by: 1. John Arhter 2. S. Sivaramakrishnan
Editor : Prof. P.Muthukumar

DESIGN, FABRICATION AND PERFORMANCE

EVALUATION OF A TEST RIG OF CENTRIFUGAL
PUMPS IN SERIES AND PARALLEL COMBINATION
Naveen Sharma1* and Lajpat Rai1
1
Department of Mechanical and Industrial Engineering, Indian Institute of Technology, Roorkee (U.K.)-247667
2
Department of Mechanical Engineering, BSA Institute of Technology and Management (BSA ITM), Faridabad,
Haryana-121002, India
1
sharma.naveen28@yahoo.com 2railajpat@gmail.com

ABSTRACT

Centrifugal pumps find vast applications in modern machineries and often works either in series or
parallel. In this experimental work, an attempt has been made to design and fabricate an experimental set-up that
comprises of two centrifugal pumps arranged in such a manner that the experimental set-up works either
individually or in series and in parallel combinations. The maximum value of efficiency estimated is 62.45% for
C1 and 65.30% for C2, 55.10% and 58.15%, when pumps operate individually, in series and in parallel
combination, respectively. The designed and fabricated test rig has efficiently served the purpose of providing a
clear idea about the effectiveness of experimental set-up in transmitting fluid to distant work stations. This study
indicated the methodology could be used to transfer gases or liquid to long distances by using more number of
pumps in series and parallel combinations.

Keywords— Centrifugal Pump; Efficiency; Series Combination; Parallel Combination; Best Efficiency Point
(BEP).

1. INTRODUCTION large impeller eye diameter is required, which caused

This Centrifugal pumps have found enormous high suction recirculation at high flow rates and this
industrial and commercial applications and in became responsible for cavitation, piercing noise and
particular in oil and gas industry, steam power plants, vibration, premature wear and all sorts of
buildings, agriculture and irrigation works, food components failures. Fraser (1981) explains
industry, paper and pulp industry etc. These enable relationship between onsets of the impeller inlet
fluid transmission between two places by utilizing recirculation and average impeller blade inlet angle.
power from some other external source. Centrifugal Li (2000) studied the effect of viscosity on the
pumps belong to dynamic pressure pumps that performance of centrifugal pump and flow pattern
generate head and flow by increasing the velocity of within impeller. The results obtained from this study
the liquid with the help of a rotating vane impeller conclude that high viscosity results in rapid
followed by a diffuser. Specific speed, blade enhancement in the friction and hydraulic losses and
orientation, casing insulation, impeller diameter, correspondingly decrease in the centrifugal pump
number and pitch of blades, surface finish of the performance. The flow pattern near the suction side
internal surfaces, viscosity of fluid etc. are the was observed to be greatly affected, as compared to
parameters that influence the performance of such the outlet side, by fluid viscosity which was
pumps. accurately measured using LDV (Laser Doppler
velocimetry).
Hallam (1982) studied the effect of suction
specific speed on the performance reliability of The common causes of failure of pump during
centrifugal pump. The trend for frequency of operation are briefly discussed in (Hartigan, 2001;
centrifugal pump failure increased above a certain World Pump, 35-39, 2001). For example, a mono-
value of specific suction speed (Ns = 213). The block centrifugal pump often cause serious problems
reason behind was that at high suction specific speed like abnormal noise, leakage, high vibration, and
cavitations etc. due to defect in bearing, seal,

goniv publications Page 12

 Asian Journal of Science and Appelied Technology (AJSAT)
Vol1.No.1 2013 pp 12-17.
available at: www.goniv.com
Paper Received :25-02-2013
Paper Published:18-05-2013
Paper Reviewed by: 1. John Arhter 2. S. Sivaramakrishnan
Editor : Prof. P.Muthukumar

impeller and insulation. Cavitation deteriorates

hydraulic performance of the pump by decreasing
head-capacity and efficiency, too much erosion of
parts, pump malfunctioning and structural vibration.
Now-a-days, owing to huge applications in
machineries, scholars are challenged to develop a
better system. Henceforth, research related to basics
(World Pumps, Waterworks, 30-31, June 2010),
selection of pump systems for different applications
(Larralde & Ocampo, 2010), commissioning/start-up
and a magnetic drive pump (World Pumps, 24-26,
Dec. 2000) has been paid attention. Recently,
research related to optimization of pump casing
(Golbabaei, Torabi, and Nourbakhsh, 2009), working
under multi-phase flow conditions, improvements in
design of impeller (Šavar, Kozmar, & Sutlović,
2009) and simulation of a low-specific-speed high
speed centrifugal pump has been carried out
(Jafarzadeh, Hajari, Alishahi, & Akbari, 2011).
Derakhshan and Nourbakhsh (2008) have been
estimated best efficiency point (BEP) of reverse Figure. 1 General components of centrifugal
pump based on geometric and hydraulic mode pump
characteristics. The values predicted in this study
using CFD at Ns = 23.5 were found to be slightly
lower than the experimental values. Caridad et al.
(2008) presented biphasic characterization of a
centrifugal pump impeller of an ESP (Electrical
Submersible Pump) carrying air-water mixture at
specific speed (Ns = 2063) by means of CFD
(Computational Fluid Dynamics) tools.

Inspired by the above literature, the present work

experimentally investigates performance of
centrifugal pumps individually, in series and parallel
combinations on one test rig and also discusses their
characteristic curves.

2. EXPERIMENTAL SETUP
DESCRIPTION AND PROCEDURE Figure. 2 Direction of fluid flow in centrifugal
The general components and direction of fluid
pump
flowing in the fabricated centrifugal pump setup are
(http://www.britannica.com/EBchecked/topic/632
presented in Fig. 1 and Fig. 2. The schematic
655/volute-centrifugal-pump)
diagram and actual fabricated experimental set-up is
shown in Fig. 3 and fig. 4, respectively.
It consists of two identical mono-block centrifugal
pumps C1 and C2 (technical specifications of the
centrifugal pump as show in Table 1), piping
network, strainers (St), valves (V1-V5), measuring
instruments, sump (S1) of 1.20 m × 0.54 m × 0.40 m
and a collecting tank (S2) of 0.75 m × 0.61 m × 0.46
m.

A. Instrumentation
1) Pressure Gauge (P2, P4)

goniv publications Page 13

 Asian Journal of Science and Appelied Technology (AJSAT)
Vol1.No.1 2013 pp 12-17.
available at: www.goniv.com
Paper Received :25-02-2013
Paper Published:18-05-2013
Paper Reviewed by: 1. John Arhter 2. S. Sivaramakrishnan
Editor : Prof. P.Muthukumar

Used to measure pressure at inlet and outlet sections

of the CPs. Different categories were available, but
as per the requirement for this test rig 7 kgf/cm2 has M S2
been used.
2) Vacuum Gauge (P1, P3)
T4
Here 760 mm of Hg vacuum gauge has been used for P4
measuring the suction pressure at the inlet of CPs. T3 V5
3) Thermometer (T1-T4)
P3
In the experimental set-up sensor thermometers have CC22
V2
been used ranging from 0C-150C to measure
temperature variation of the test rig at different P2
positions. T1
V3
4) A Graduated Glass Tube With a Linear Scale (M) T2
C1 V4
P1 C1
Used to measure the height of water in the collecting
tank through which discharge has been calculated. V1

Table 1 Technical Specifications Of The

Centrifugal Pump As Per Pump
St2 S1
Manufacturer
S. No Specifications
1 Motor Crompton Greaves
St3
St1
2 Power 0.37 HP/0.5 kW
3 Phase AC, 50 Hz (A) Pressure Gauge (P2, P4)
4 Voltage 220 V (B) Vacuum Gauge (P1 , P3 )
(C) Thermometer (T1-T4 )
5 Speed 2780 rpm
(D) Graduated Glass Tube (M)
6 Current 2.8 A
Figure 3 Schematic diagram of the centrifugal
7 Size (25) x (25) mm
pump test rig
8 Head 28.5 m
9 Discharge 2600/570 lph.
10 Insulation ‘B’ class
The set-up runs to collect the data for head,
discharge, pressure and temperature. Four sets of
experiment were conducted on the test rig. The
positions of operating valves for different
experiments were given in Table 2.

Table 2 Operating Conditions for

Conducting Four Experiments
Centrifugal Valve Position Experiment
Pumps (C1 & Open Close
C 2)
C1(on), C2 (off) V1, V2 V3, V5 Individual
C2 (on), C1 (off) V4 V3, V5 Individual
C1 & C2 (both V1, V3 V2, Series
on) V4,
V5
C1 & C2 (both V1, V2, V3, V5 Parallel Figure. 4 Actual fabricated the centrifugal pump
on) V4 test rig

goniv publications Page 14

 Asian Journal of Science and Appelied Technology (AJSAT)
Vol1.No.1 2013 pp 12-17.
available at: www.goniv.com
Paper Received :25-02-2013
Paper Published:18-05-2013
Paper Reviewed by: 1. John Arhter 2. S. Sivaramakrishnan
Editor : Prof. P.Muthukumar

W & L are width and length of the collecting tank

respectively (all in m). Discharge denoted by Q
The most important steps involved in carrying the (m3/s).
operation were as described in “Pumps-principles The total head (H) estimated by the relation
and practices,” by Jaico publishing house, p 89-99. H  Hs  Hd
(4)
where Hs and Hd used for suction pressure and
1) Priming
discharge pressure in m respectively.
Priming involved the removal of air, gas or
vapour from the suction piping network and P0   g Q H
(5)
pump casing. Centrifugal pump must be where , g and P0 represents the density of water,
primed prior to start up. Venting is an acceleration due to gravity and output power (P0)
operation carried out to remove any trapped respectively.
air pockets in the system and may be The efficiency estimated through the relation given:
necessary on the suction pipeline as well as Efficiency () = (Hydraulic power output) ÷ (Input
the pump. Many modern designs of pump are power) (6)
inherently self-primed from shut down to start
up and self-venting. In the experimental set-
4. RESULTS AND DISCUSSION
up two self-primed centrifugal pumps have
A. Application of Performance of an Individual
been used.
Pump
2) Starting The variation of head, input power and pump
Standard start-up procedure was followed performance (efficiency) with discharge have been
Before initial start-up, the drive coupling was evaluated and presented in Fig. 4 for C1 and Fig. 5
disconnected and the rotation of the driver for C2. From the figures, it is clear that as flow rate
checked. An arrow on the pump casing increases from 0.002 to 0.008 m3/s head decreases
indicated the correct direction of rotation. from 8.60 m to 3.20 m, power input to pump
3) Stopping the pumping equipment increases and performance of the pump increases for
A typical sequence of operation for shutting both the pumps. The efficiency varies from 36.35%
down the pump set was: to 62.45% for C1 and from 37.15% to 64.54% for C2.
 First shut down the driver. The best efficiency points located at QBEP=0.0072
 If the pump is not required for and 0.0064 m3/s, matching to the best efficiencies
some time, the suction and 62.45% and 60.98% for C1 and corresponding to C2
discharge isolating valves can be the best efficiency points located at QBEP=0.008 and
closed and any ancillary services 0.0072 m3/s, corollary to the best efficiencies 64.54%
such as cooling or flushing lines and 62.80%.
can be shut down.
 0.9
The following parameters were measured for 9
64
Efficiency 62
each conducted experiments: Suction Pressure, Head 0.8 60
Discharge Pressure and Discharge. 8 58
56
0.7
54
3. DATA REDUCTION 7
H (m)

efficiency (%)

52
The correlation used to calculate total head,
P (kW)

6 0.6 50
discharge, efficiency, specific speed and power has 48

been given by the expressions. 5

46
0.5
Input power of motor to the shaft = Power 44
42
(Number of revolution × 3600) ÷ (time × Z × W). 4
40
0.4
(1) 38

Input power to the pump = (Input power of motor to 3 36

the shaft) × (mechanical efficiency) × (transmission 0.002 0.003 0.004 0.005 0.006 0.007 0.008
0.3 34

efficiency) (2) 3
Q (m /s)

Q
 W  L  h Figure. 4 Variation of head (m), input power (kW)
t (3) and efficiency (%) with flow rate (m3/s) for C1
where h is the height of water in collecting tank and

goniv publications Page 15

 Asian Journal of Science and Appelied Technology (AJSAT)
Vol1.No.1 2013 pp 12-17.
available at: www.goniv.com
Paper Received :25-02-2013
Paper Published:18-05-2013
Paper Reviewed by: 1. John Arhter 2. S. Sivaramakrishnan
Editor : Prof. P.Muthukumar

shown in Fig. 7 and Fig. 8. The combined head

64
9 0.9
62
discharge curve is found by the individual flow rates
Efficiency
Head
60 at the same level. The pump having steep
8 0.8 58
characteristic curves have been operated in parallel to
56
7 54 obtain high flow rates. The best efficiency points
0.7
located at QBEP=0.015 and 0.016 m3/s analogous to
H (m)

efficiency (%)
52

P (kW)
50
6
0.6 48
the best efficiencies 58.45% and 56.23%,
5
46 respectively. The discharge in case of parallel
Power 0.5 44
42
combination is varying as 0.0056-0.016 m3/s.
4
40
0.4
38
3 36
0.3 34
0.002 0.003 0.004 0.005 0.006 0.007 0.008
3
Q (m /s)

Figure. 5 Variation of head (m), input power (kW)

and efficiency (%) with flow rate (m3/s) for C2

B. Performance of Centrifugal Pumps in Series

Combination
As the centrifugal pumps operated in series, input
power required to operate two centrifugal pumps
increased. Head produced in this case also increases,
nearly becomes double as compared to single Figure. 7 Variation of head (m) of centrifugal
centrifugal pump. The characteristics curve in series pumps with discharge (m3/s) in parallel
operation as shown in Fig. 6. The best efficiency II.
points located at QBEP = 0.008 and 0.0094 m3/s, 5. CONCLUSION
corresponding to the best efficiencies are 55.10% and The present work was undertaken with the
53.22%, respectively. Head varies from 5.5 m to objective of detailed investigation of the centrifugal
15.42 m corresponding to discharge of 0.0094 m3/s pumps performance when operated in different
to 0.0035 m3/s, respectively. conditions either individually or in combinations of
series or parallel. The evaluated results have been
58 presented in graphical form for variation of Head,
16
1.0
56 Input power and Efficiency with flow rate
Head Efficiency 54
52
(Discharge). From this study, an insight about the
14
50 characteristics of centrifugal pump and their
0.9
48 behaviour corresponding to different conditions and
12 46
combinations has been examined.
H (m)

Power
efficiency (%)

44
P (kW)

0.8 42
10
40 60
1.0
38
8 0.7 36 55
Efficiency
34 0.9

6 32 50
efficiency (%)

0.6 30
Power
P (kW)

28 0.8 45
4
0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.010
3 40
Q (m /s)
0.7

Figure. 6 Variation of head (m), input power (kW) 35

and efficiency (%) with flow rate (m3/s) in series 0.6

30

C. Performance of Centrifugal Pumps in Parallel 0.004 0.006 0.008 0.010 0.012 0.014 0.016
3
Combination Q (m /s)

If the flow rate of one pump is insufficient, the Figure. 8 Variation of input power (kw) and
desired flow rate can be obtained by operating two or efficiency (%) with flow rate (m3/s) for centrifugal
The characteristics curve for parallel operation is pump in parallel

goniv publications Page 16

 Asian Journal of Science and Appelied Technology (AJSAT)
Vol1.No.1 2013 pp 12-17.
available at: www.goniv.com
Paper Received :25-02-2013
Paper Published:18-05-2013
Paper Reviewed by: 1. John Arhter 2. S. Sivaramakrishnan
Editor : Prof. P.Muthukumar

[14] http://www.britannica.com/EBchecked/topic/63
The result shows that in series combination head 2655/volute-centrifugal-pump
increases whereas in parallel flow rate increases. [15] “Pumps-principles and practices.” by Jaico
This experimental study has been proved very crucial publishing house, p 89-99.
for industries in transferring fluid form one place to
another and in common day life as well.

REFERENCES

[1] Hallam JL. “Centrifugal pumps: Which suction

specific speeds are acceptable?.” J of
Hydrocarbon Processing, 195-197, April 1982.
[2] Fraser WH. “Recirculation in centrifugal
pumps.” ASME Winter Annual Meeting, 65-86,
1981.
[3] Li WG. “Effects of viscosity of fluids on
centrifugal pump performance and flow pattern
in the impeller.” International J of Heat and
Fluid Flow, vol. 21, 207-212, 2000.
[4] Hartigan K. “Inside monitoring to extend pump
life.” World Pumps, vol. 502, 28–30, 2008.
[5] “Optimizing centrifugal pump operation.”
World Pumps, 35-39, 2001.
[6] “Back to basics with centrifugal pumps,” World
Pumps, Waterworks, 30-31, June 2010.
[7] Larralde E and Ocampo R. “Centrifugal pump
selection process.” World Pumps, Pump
Selection, 24-28, Feb. 2010.
[8] “Successful centrifugal pump commissioning
and start-up.” World Pumps, 24-26, Dec. 2000.
[9] Golbabaei M, Torabi R, Nourbakhsh SA.
“Experimental and FEM failure analysis and
optimization of a centrifugal-pump volute
casing.” Engineering Failure Analysis, vol. 16,
1996–2003, 2009.
[10] Šavar M, Kozmar H and Sutlović I. ”Improving
centrifugal pump efficiency by impeller
trimming.” Desalination, vol. 249, 654–659,
2009.
[11] Jafarzadeh B, Hajari A, Alishahi MM and
Akbari MH. ” The flow simulation of a low-
specific-speed high-speed centrifugal pump.”
Applied Mathematical Modelling, vol. 35, 242–
249, 2011.
[12] Derakhshan S and Nourbakhsh A. “Theoretical,
numerical and experimental investigation of
centrifugal pumps in reverse operation.”
Experimental Thermal and Fluid Science, vol.
32, 1620–1627, 2008.
[13] Caridad J, Asuaje M, Kenyery F, Tremante A
and Aguillón O. “Characterization of a
centrifugal pump impeller under two-phase
flow conditions.” Journal of Petroleum Science
and Engineering, vol. 63, 18–22, 2008.

goniv publications Page 17