Analysis of Centrifugal Pump Performance Characteristics at Various Operating Configurations

D.S. Corpuz, J.L. de Guzman and J.M. Golbin

Department of Chemical Engineering, University of the Philippines-Diliman, Quezon City, Philippines

D.S. Corpuz, J.L. de Guzman and J.M. Golbin, 2008. Centrifugal pumps are one of the widely used pumps in industry because
of their simple design and flexibility of application. The head they produced vary with flow rate and with geometry or
configuration. This experiment analyzed their performance characteristics in single, series, and parallel configuration by making
head vs. capacity plots. It was shown that the over-all head developed decreases as the capacity increases, and that the efficiency
increases as the rotation speed increases.

Keywords: parallel operation, power input, power output, pump capacity, pump efficiency, series and parallel operation, single operation,
total head

OBJECTIVES

The primary objective of this experiment was to determine the

performance characteristics of centrifugal pumps operating
individually and in series and parallel configurations. This
experiment was also done to investigate and establish the
performance curves:
i) Total head vs. Capacity of the pump at constant rotational
speed. Figure 1. Centrifugal Pump
ii) Efficiency vs. Capacity at constant rotating speed. Source: http://www.cheresources.com/centrifugalpumps3.shtml

THEORETICAL BACKGROUND

A pump aids the movement of liquids from one location to

another. (Perry, 1997). Mechanical energy of the liquid,
velocity, pressure and elevation are increased by using pumps.
This mechanical energy is converted to fluid energy. (McCabe,
2001) Pumps can be further categorized to displacement and
dynamic. Under dynamic pumps are centrifugal and special
effect pumps. For centrifugal pump, it is subdivided into (1)
axial flow, (2) mixed flow and radial flow and (3)
peripheral.(Dickenson, 1988) For this study, an open impeller Figure 2. Cross-sectional View of a Centrifugal Pump
centrifugal pump is used. Source:http://www.cheresources.com/centrifugalpumps4.shtml

One of the types of pumps used in the industry is the centrifugal Before operating the pump, priming is performed. It is done to
pump which falls under rotodynamic pump. It can generate high displace the air which is present in the casing of the pump.
rotational velocities and convert the resultant, which is the Since it is airbound, it cannot function until the air present is
kinetic energy, to pressure energy. The centrifugal action is the replaced with liquid. It will not be able draw liquid upward
key in increasing the mechanical energy of the liquid. The liquid from an initially empty suction line and force the liquid through
enters through the suction nozzle which is concentric with the may be influenced by the system used (e.g., single operation and
axis of the impeller, a high-speed rotary element. The liquid multistage operation). Another problem that may be encountered
spreads radially and comes in contact with the vanes. The liquid in the operation of pumps is cavitation, which is the phenomena
leaves the impeller at a higher velocity, passing between the when the liquid flashes to vapor inside the pump. This can be
vanes. Considering an ideal set-up, the spaces between the avoided by the adjusting the pressure at the pump inlet in such a
vanes are completely filled with the fluid in motion without way that it exceeds the vapor pressure by a certain quantity. For
cavitation. The volute, a spiral casing, then collects the fluid multistage operation, the discharge in the discharge in the first
from the impeller. Here, the velocity head of the liquid is stage provides the suction for the second and the same trend is
converted to pressure head. The impeller applies power to the followed for the succeeding pumps. This results to greater head
liquid and the torque of the driveshaft transmits the power to the developed compared to single pump operation. (McCabe, 2001)
impeller. The liquid completely leaves the pump by passing
through the discharge nozzle. If an ideal system is to be PROCEDURE
assumed, friction is negligible and the efficiency of the pump is
100%. It can deliver a definite discharge rate at every given Before performing the analysis of pumps with different set-ups,
developed head. It is also assumed that all liquid flowing across the system was subjected to priming. All of the valves were
the periphery of the impeller is moving at the same speed. This opened except the flow-regulating valve. This is done to ensure
results to the conclusion that the vanes are infinite, has zero the system is free of air column which may result to erroneous
thickness, and are infinitesimally apart. For multistage values for pressure head as read from the height of the gauge.
operation, the discharge in the discharge in the first stage Using the adjustable counterweight, the motor stator is balanced
provides the suction for the second and the same trend is prior to the experiment proper. The tank is filled with water
followed for the succeeding pumps. This results to greater head until it coincides with the apex of the V notch, taking into
developed compared to single pump operation. (McCabe, 2001) consideration the surface tension effect. The point gauge in the
stilling well is zeroed. The system is operated at a speed
ranging from 1000 to 2000 rpm. The nut on the rear of the
pressure gauge is removed and air is expelled. It is followed by
removing the nut on the rear of the vacuum gauge and filling the
tube with water. Priming is concluded by shutting off the Parallel Operation
system. Overall Head vs Capacity

Several set-ups
ups are required to be analyzed: series, parallel and 8.5
single pump operation. For the series and parallel operations, 8.0

Overall Head (m)

the valves are adjusted as shown in the appendix.
ix. Speed was set 7.5
to 1800 rpm, 2000 rpm and 2200 rpm and four data points for 7.0
flow rates were gathered for each trial. The discharge rate was 6.5
regulated by slowly opening the flow-regulating
regulating valve. The 6.0
values for v notch head, discharge rate, suction head,
head delivery 5.5 1800 rpm
head and the torque mass were directly obtained from the gauges 2000 rpm
5.0 2200 rpm
and weights used. These quantities were recorded at
4.5
incremental discharge flow rates. For the single operation,
pump 1 and pump 2 were operated individually. The same 1.78E-03 1.98E-03 2.18E-03 2.38E-03
values for speed were used. The valves are adjusted to fit the Capacity (m3/s)
operation of the selected pump. The system is shut down every
after operation, before proceeding to the next analysis.
Single Pump Operation (pump 1)
Head vs Capacity
RESULTS AND ANALYSIS 6.0
5.5
y = -4E+06x2 + 11167x - 1.761
Some of the essential pump performance characteristics are 5.0 y = -3E+06x2 + 6547.x + 1.583
R² = 0.999
R² = 0.998
pump capacity, head, power, and efficiency. Capacity is the 4.5

Head (m)
mass rate of fluid through the pump (Perry, 1997). 4.0
3.5
3.0
2.5
best fit, 1800 rpm
2.0 y = 1E+06x2 - 7009x + 9.695 best fit, 2000 rpm
where Q = capacity 1.5 R² = 0.999 best fit, 2200 rpm
he = h + kh 1.0
h = measured head (meters) 1.10E-03 1.30E-03 1.50E-03 1.70E-03 1.90E-03
kh = 0.00085 meters
Ce = 0.5765 Capacity (m3/s)

Head is the expression for the energy of the fluid due to its
elevation above some reference point, its velocity and its Single Pump Operation (pump 2)
pressure. The two fundamental pressure conditions of fluid are Head vs Capacity
the static head, where fluids are at rest, and the dynamic head, 7.5
where fluids are flowing. The total system head is the energy 7.0
required to move this liquid at a given flow rate (Dufour & 6.5 y = 77139x2 - 2691.x + 9.220
Nelson, 1993). 6.0 R² = 0.167
5.5
Head (m)

5.0 y = 1E+07x2 - 47693x + 46.55

4.5 R² = 0.997
4.0
where Hm = manometric total head 3.5 best fit, 1800 rpm
Hd = delivery head 3.0 y = -3E+07x2 + 93753x - 58.50 best fit, 2000 rpm
Hs = suction head R² = 0.963 best fit, 2200 rpm
2.5
2.0
The total head versus capacity of a system are shown below.
below For 1.29E-03 1.49E-03 1.69E-03 1.89E-03
03 2.09E-03
the stable overall head vs. capacity graph, the maximum head is
at zero discharge. Head decreases as discharge increases. For the Capacity (m3/s)
unstable graph, Hm initially increases from the value at zero
discharge and falls with further increase in discharge (McCabe, The graph of the actual developed head drops to zero as the flow
2001). rate increases to a certain value (zero-head
head flow rate).

Series Operation It is customary to calculate the power output when arriving at

Overall Head vs Capacity the performance of the pump, that is
11
10
The power input to the pump is greater than the power output
Overall Head (m)

9
because of the internal losses due to friction, leakage, etc.
8
7
6 1800 rpm
5 2000 rpm
2200 rpm Efficiency is the ratio of fluid power
wer to the total pow
power
4 consumed, that is
1.47E-03 1.97E-03 2.47E-03
Capacity (m3/s)
 the sum of the discharges through each pump at the head
Series Operation concerned (Series/Parallel Centrifugal Pump Test Rig
Overall Efficiency vs Capacity Instruction Manual). Figure 3 and 4 illustrate this.

58
Overall Efficiency

53

48

43 1800 rpm
2000 rpm
38
2200 rpm
33
Figure 3. Pumps in Parallel
1.47E-03 1.97E-03 2.47E-03 Source: www.engineeringtoolbox.com
Capacity (m3/s)

Parallel Operation
Overall Efficiency vs Capacity

87
82
Overall Efficiency

77
72 Figure 4. Pumps in Series
67 Source: www.engineeringtoolbox.com
62
1800 rpm
57 2000 rpm CONCLUSIONS AND RECOMMENDATIONS
52 2200 rpm
47 Head loss factors include fluid friction in passages and channels
1.78E-03 1.98E-03 2.18E-03 03
2.38E-03 and shock losses (the sudden change in direction of liquid
leaving the impeller and joining the stream of the liquid flowing
Capacity (m3/s) circumferentially around the casing. Friction is highest at the
maximum flow rate (McCabe, 2001).

Single Pump Operation (pump 1) Power lost happens because of (1) the effec effect of friction and
Efficiency vs Capacity shock losses i.e., conversion of mechanical energy to heat, (2)
98
leakage – unavoidable reverse flow from the impeller discharge;
93 reduces the volume of the actual discharge from the pump per
88 unit of power expended, (3) disk friction – friction between the
Efficiency (m)

83 outer surface of the impeller and the liquid in the space between
78 the impeller and the inside of the casing,, and (4) bearing losses –
73 power required to overcome mechanical frictionriction in th
the bearing
68 1800 rpm and seals of the pump (McCabe, 2001).
63 2000 rpm Smoother impeller finishes will increase the efficiency of the
58 2200 rpm pump. Generally, the head produced decreases as the
53 amount of water pumped increases. The shape of the
curve varies with pump’s specific ific speed and impeller
1.10E-03 1.30E-03 1.50E-03 1.70E-03 1.90E-03
design. Usually, the highest head is produced at zero
Capacity (m3/s) discharge and it is called the shut-off
off head. The efficiency of a
pump
steadily increases to a peak, and then declines as Q
increases further (http://www.aces.edu).
Single Pump Operation (pump 2)
Efficiency vs Capacity
78 To improve thehe performance characteristics of the pump, the
73 pump system must be primed before use. The system must be
68 completely air free
ee so that reliable data can be obtained. Also, it
63 is recommended that the experiment be performed well within
Efficiency (m)

58 acceptable operating ranges. The upper and lower limits are set
53 by the allowable speed variations. The efficiency is sacrificed if
48 the experiment is performed at extreme operating conditions.
43 Although not seen in the experiment, pumps are more efficiefficient
1800 rpm
38 2000 rpm at higher speeds. Improving the equipment such that readings
33 2200 rpm such as torque and pressures are more accurately read would
28 certainly improve the performance curves for the pumps. The
pumps are already old. Also, the effect of temperature was not
1.29E-03 1.49E-03 1.69E-03 1.89E-03 2.09E
2.09E-03
taken into effect. Installing temperature
rature sensors would give
Capacity (m3/s) more accurate data.

For a series operation, the same discharge passes through both

pumps but the developed heads supplements each other. Total
developed head is calculated by simply adding the heads of each
pump. For a parallel set-up, up, similar head across the pump is
observed
ed but it is to be noted that the individual discharges may
be unlike unless the identical pumps are used. Total discharge is
REFERENCES

Dufour, J.W. and Nelson, W. E. (1993). Centrifugal Pump

Sourcebook. USA: McGraw-Hill, Inc. pp. 7-8, 25

Foust, A.S. (1980). Principles of Unit Operations. Singapore:

John Wiley & Sons (Asia) Pte Ltd. pp. 596

McCabe, W.L. (2001). Unit Operations of Chemical

Engineering. Singapore: McGraw-Hill Book Co. pp. 195

http://www.aces.edu/dept/fisheries/education/ras/publications/fa
cility_design/Selection%20of%20Centrifugal%20Pumping%20
Equipment.pdf - Retrieved March 6, 2008

Engineering Toolbox.
http://www.engineeringtoolbox.com/centrifugal-pumps-
d_54.html - Retrieved March 5, 2008

Centrifugal Pumps.
http://www.tpub.com/content/doe/h1012v3/css/h1012v3_69.htm
Retrieved March 6, 2008
Pumps in Series
Pump 1 Pump 2 Pump 1 Pump 2 Overall
Rotational Discharge
Flow Rate Vnotch Head Torque Mass
Speed (capacity) Delivery Delivery
Suction Head Suction Head Total Head
Head Head
N Q h Q Hs1 Hd1 Hs2 Hd2 Hm1 Hm2 HmT M
rpm L/min mm m3/s m m m m m m m g

1800 170 71 1.88E-03 1 0.75 -1.5 2 0.17 3.92 4.08 2287.18

160 68.4 1.72E-03 0.9 1.8 -0.75 3 1.25 4.10 5.34 2110.18

150 66.4 1.60E-03 0.8 2.1 0.3 4.5 1.60 4.50 6.10 2103.18

140 64.5 1.49E-03 0.75 2.7 1 5.5 2.21 4.76 6.97 2006.18

2000 200 75 2.16E-03 1.2 1.5 -2 2.75 0.85 5.30 6.14 2948.18

190 73 2.02E-03 1.1 2 -1 3.5 1.38 4.98 6.36 2845.18

180 70.8 1.87E-03 1 2.5 0 5 1.91 5.41 7.32 2693.18

170 69.3 1.77E-03 1 3 1 6 2.37 5.37 7.74 2615.18

2200 225 79.1 2.46E-03 1.4 1 -2.5 3 0.31 6.21 6.52 3666.77

215 77.6 2.35E-03 1.25 2.5 -1.5 4.5 1.90 6.65 8.54 3518.27

205 76 2.23E-03 1.25 3 -0.5 6 2.33 7.08 9.42 3341.27

190 74 2.09E-03 1.1 4 1 7.5 3.41 7.01 10.42 3000.77

Pump 1 Pump 2 Pump 1 Pump 2 Combined

Rotational
Torque
Speed
Power Output Power Input Power Output Power Input Efficiency

N T Po1 Pi1 Po2 Pi2 η1 η2 ηT

rpm Nm W W W W
1800 5.61 3.07 110.68 72.36 110.68 2.78 65.38 34.08
5.17 21.00 102.11 69.02 102.11 20.56 67.59 44.08
5.16 25.04 101.77 70.45 101.77 24.60 69.22 46.91
4.92 32.20 97.08 69.37 97.08 33.17 71.46 52.31
2000 7.23 17.89 158.52 112.04 158.52 11.29 70.68 40.98
6.97 27.26 152.98 98.50 152.98 17.82 64.39 41.10
6.60 35.05 144.81 99.27 144.81 24.21 68.56 46.38
6.41 41.23 140.61 93.44 140.61 29.32 66.45 47.89
2200 8.99 7.48 216.87 149.86 216.87 3.45 69.10 36.28
8.63 43.64 208.09 152.96 208.09 20.97 73.51 47.24
8.19 50.99 197.62 154.83 197.62 25.80 78.35 52.08
7.36 69.80 177.48 143.48 177.48 39.33 80.84 60.09
Pumps in Parallel
Pump 1 Pump 2 Pump 1 Pump 2 Overall
Rotational Speed Flow Rate Vnotch Head Discharge
Suction Head Delivery Head Suction Head Delivery Head Total Head

N Q h Q Hs1 Hd1 Hs2 Hd2 Hm1 Hm2 HmT

rpm L/min mm m3/s m m m m m m m

1800 200 74.1 2.09E-03 0.7 4 -1.5 4 3.81 6.01 4.91
190 73.8 2.07E-03 0.65 4.2 -1.5 4.2 4.05 6.20 5.13
180 71.7 1.93E-03 0.65 4.2 -1.3 4.5 3.99 6.24 5.11
170 69.7 1.80E-03 0.65 4.2 -1.1 4.5 3.93 5.98 4.95
2000 245 78.6 2.42E-03 0.75 5.5 -1.8 5.2 5.44 7.69 6.56
215 77.5 2.34E-03 0.7 5.5 -1.5 5.5 5.44 7.64 6.54
205 74.6 2.13E-03 0.7 5.8 -1.5 5.8 5.63 7.83 6.73
190 73.5 2.05E-03 0.65 5.9 -1.5 5.8 5.74 7.79 6.77
2200 255 79.9 2.52E-03 0.8 7 -2 6.8 6.95 9.55 8.25
240 79.6 2.50E-03 0.75 7 -1.8 7 6.98 9.53 8.26
230 78.2 2.39E-03 0.75 7.2 -1.5 7.2 7.12 9.37 8.25
220 77.4 2.33E-03 0.75 7.2 -1.5 7.2 7.09 9.34 8.21
Pump 1 Pump 2 Pump 1 Pump 2 Overall
Rotational Speed Torque Mass Torque Total Q
Power Output Power Input Power Output Power Input Efficiency

N M T Po1 Pi1 Po2 Pi2 η1 η2 ηT

rpm g Nm W W W W m3/s

1800 1613.09 3.95 78.32 178.06 123.50 178.06 43.99 69.36 53.83 4.19E-03

1562.09 3.83 82.41 175.59 126.12 175.59 46.94 71.83 56.77 4.15E-03

1512.09 3.71 75.47 173.17 118.06 173.17 43.58 68.18 53.17 3.86E-03

1461.09 3.58 69.37 170.70 105.56 170.70 40.64 61.84 49.05 3.60E-03

2000 1940.59 4.76 129.19 204.34 182.65 204.34 63.22 89.38 74.06 4.85E-03

1889.59 4.63 124.85 201.60 175.33 201.60 61.93 86.97 72.34 4.68E-03

1766.09 4.33 117.58 194.96 163.51 194.96 60.31 83.87 70.16 4.26E-03

1706.09 4.18 115.60 191.73 156.86 191.73 60.29 81.81 69.42 4.10E-03

2200 2440.68 5.98 171.86 244.35 236.18 244.35 70.33 96.66 81.42 5.05E-03

2407.18 5.90 171.15 242.37 233.66 242.37 70.62 96.40 81.52 5.00E-03

2236.18 5.48 167.05 232.26 219.83 232.26 71.92 94.65 81.74 4.78E-03

2054.68 5.04 162.10 221.52 213.55 221.52 73.17 96.40 83.20 4.66E-03
Pump 1
Rotational Suction Delivery Power Power
Flow Rate Vnotch Head Discharge Total Head Torque Mass Torque Efficiency
Speed Head Head Output Input
N Q h Q Hs Hd Hm M T Po Pi η1
rpm L/min mm m3/s m m m g Nm W W
1800 140 65 1.52E-03 0.8 2.5 1.97 1088.5 2.67 29.26 52.67 55.54
130 63.4 1.42E-03 0.8 2.8 2.24 1039.5 2.55 31.27 50.30 62.16
120 61.2 1.31E-03 0.8 3.3 2.70 993 2.43 34.58 48.05 71.95
110 57.6 1.12E-03 0.75 4 3.40 942 2.31 37.48 45.58 82.22
2000 160 68.9 1.75E-03 0.9 3.5 2.96 1336.09 3.28 50.76 71.84 70.66
150 67.8 1.68E-03 0.9 3.8 3.23 1285.09 3.15 53.28 69.10 77.11
140 64.9 1.51E-03 0.8 4.5 3.97 1208.09 2.96 58.72 64.96 90.40
130 61.6 1.33E-03 0.75 5 4.46 1105.09 2.71 57.99 59.42 97.60
2200 180 72.5 1.98E-03 1.1 4.2 3.56 1562.09 3.83 69.29 92.39 75.00
170 71.9 1.94E-03 1.1 4.5 3.84 1562.09 3.83 73.25 92.39 79.28
160 69.4 1.78E-03 1 5.2 4.57 1461.09 3.58 79.85 86.41 92.40
150 64.9 1.51E-03 0.9 6 5.37 1398.59 3.43 79.44 82.72 96.03

Pump 2
Rotational Suction Delivery Power Power
Flow Rate Vnotch Head Discharge Total Head Torque Mass Torque Efficiency
Speed Head Head Output Input

N Q h Q Hs Hd Hm M T Po Pi η1
rpm L/min mm m3/s m m m g Nm W W
1800 160 66.9 1.63E-03 -2.5 0.15 2.96 1101.59 2.70 47.22 153.31 30.80
150 66.4 1.60E-03 -2.5 0.17 2.97 1101.59 2.70 46.49 153.31 30.33
140 64.5 1.49E-03 -2.3 2.2 4.76 1004.09 2.46 69.37 148.59 46.69
130 61.2 1.31E-03 -2.1 2.8 5.10 1039.5 2.55 65.31 150.30 43.45
2000 175 70.7 1.86E-03 -3 2 5.41 1382.59 3.39 98.87 174.34 56.71
160 65.2 1.53E-03 -3 2.5 5.77 1331.59 3.26 86.43 171.60 50.37
150 63.9 1.45E-03 -2.5 3.55 6.30 1234.09 3.03 89.70 166.35 53.92
140 63.9 1.45E-03 -2 4 6.25 1156.09 2.83 88.99 162.16 54.88
2200 200 75.4 2.19E-03 -4 2.5 7.06 1738.59 4.26 151.35 202.83 74.62
190 71.9 1.94E-03 -3.5 2.8 6.74 1691.09 4.15 128.52 200.02 64.26
180 69.4 1.78E-03 -3.2 3.5 7.07 1609.59 3.95 123.51 195.20 63.27
170 66.9 1.63E-03 -2 4.5 6.81 1461.09 3.58 108.64 186.41 58.28
Schematic Diagram for Parallel Flow

M, H2O M, H2O RPM M, H2O M, H2O

L/S
C

D
B
PUMP PUMP
1 2

A A

To V

Notch
Schematic Diagram for Series Flow

M, H2O M, H2O RPM M, H 2O M, H 2O

C L/S

B D
PUMP PUMP
1 2

A A

To V