Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Experiment 9: Centrifugal Pump

 
Introduction:

Pumps fall into two main categories: positive displacement pumps and
rotodynamic pumps.
In a positive displacement pump, a fixed volume of fluid is forced from one
chamber into another. One of the oldest and most familiar designs is the reciprocating
engine, utilising a piston moving inside a cylinder. Steam pumps, the 'nodding
donkey', stirrup pumps and hydraulic rams are all of this type. Animal hearts are also
positive displacement pumps, which use volume reduction of one chamber to force
flow into another chamber.
The FM50 pump is, by contrast, a rotodynamic machine. Rotodynamic (or simply
dynamic) pumps impart momentum to a fluid, which then causes the fluid to move
into the delivery chamber or outlet. Turbines and centrifugal pumps all fall into this
category.
Pumps

Turbo-hydraulic (Kinetic) pumps Positive Displacement Pumps

Centrifugal Propeller Jet Screw Reciprocating

Pump (Radial) (Axial) (Mixed)

 
Description:

The apparatus consists of a tank and pipework which delivers water to and from a
small centrifugal pump. The unit is fitted with electronic sensors which measure the
process variables. Signals from these sensors are sent to a computer via an interface
device, and the unit is supplied with data
logging software as standard.
Pump speed and outlet pressure may
be varied to allow the collection of
performance data over a range of
parameters. The inlet (suction) head
pressure may be adjusted to investigate
the onset of cavitation. An alternative
impeller is also supplied so that the effect
of impeller design may be studied.
For more Details refer to Instruction
Manual FM50.
Figure 1: Centrifugal Pump Demonstration Unit
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Exercise B

Objective
To create head, power and efficiency characteristic curves for a centrifugal
pump.

Theory
One way of illustrating pump characteristics is to construct contour lines of
constant power or efficiency on a graph of pump head plotted against pump
discharge. These allow engineers to see the maximum efficiency of a pump
over a range of operating parameters, which can assist in the selection of an
appropriate pump to suit particular conditions. An example is given in Figure
2.

Figure 2

Equipment Set Up

If the equipment is not yet ready for use, proceed as follows:

Ensure the drain valve is fully closed.
If necessary, fill the reservoir to within 20cm of the top rim.
Ensure the inlet valve and gate valve are both fully open.
Ensure the equipment is connected to the IFD7 and the IFD7 is connected to a
suitable PC. The red and green indicator lights on the IFD7 should both be
illuminated.
Ensure the IFD7 is connected to an appropriate mains supply, and switch on
the supply.

Run the FM50-304 software. Check that 'IFD: OK' is displayed in the bottom
right corner of the screen and that there are values displayed in all the sensor
display boxes on the mimic diagram.
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Procedure

Switch on the IFD7.

Switch on the FM50 pump within the software using the Pump On button.
In the software, rename the current (blank) results table to '50%' (this will be
the only table if results from Exercise A are not available).
On the mimic diagram of the software, set the pump speed to 50%.

The interface will increase the pump speed until it reaches the required setting.
Allow water to circulate until all air has been f1ushed from the system.
Partially closing and opening the inlet and gate valves a few times will help in
priming the system and eliminating any bubbles caught within the valve
mechanism. Leave the inlet valve fully open.

Close the gate valve to give a flow rate Q of 0. (Note that the pump may not
run well with the gate valve closed or nearly closed, as the back pressure
produced is outside normal operating parameters. The pump should begin to
run more smoothly as the experiment progresses).

Select the icon to record the sensor readings and pump settings on the
results table of the software.
Open the gate valve to allow a low flow rate. Allow sufficient time for the
sensor readings to stabilise then select the icon to record the next set of
data.
Open the gate valve in small increments, allowing the sensor readings to
stabilise then recording the sensor and pump data each time.

Create a new results sheet by selecting the icon (you may also wish to
save the results at this time to avoid losing the data in the event of problems).
Close the gate valve.
Set the pump to 60%.

Select the icon to record the sensor readings and pump settings on the new
results table.
Repeat as before, opening the gate valve in small increments and allowing the
sensor readings to stabilise then recording the sensor and pump data each time.
Close the gate valve.
Repeat the procedure at 70%, 80%, 90% and 100%. Create a new results sheet
for each setting (and save the results if desired- the same file may be
overwritten each time as more data is added). For convenience, rename each
sheet of results in the software with the pump setting.
Ensure the results are saved after taking the final set of results.
Switch the pump off. If not proceeding directly to another exercise then switch
off the IFD7 and close the FM50 software.
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Results
On the same graph plot Total Head Ht against Flow Rate Q for each setting.
Graphs may be produced using the software graph facility, in which case the
resulting graph with multiple plots must be printed. Alternatively the results
may be imported into a more sophisticated spreadsheet program that allows
the following procedure to be performed.
Select a value for efficiency, for example 40%. On each line plotted, mark the
points at which an efficiency of 40% is achieved (the data is unlikely to
include recorded points at which the efficiency is exactly 40%, so estimate the
points based on the values obtained). Where the pump performance at a
particular setting does not ever correspond to the efficiency chosen, note
whether the efficiency would lie above the line or to the right of the pump
performance curve. Join the marked points to form a smooth curve.
Repeat for other efficiency values. for example 35%.45% and 5090. to give a
family of efficiency curves.
Create and/or print a second head-flow rate graph for all pump frequencies.
Using the same procedure as for drawing contour lines of constant efficiency,
produce curves for constant mechanical power.

Conclusion
Examine and describe the shapes of the efficiency and power curves obtained.
Are the shapes consistent? How do they relate to the head-flow rate
characteristic? How do the efficiency and power curves relate to each other?

Compare the results to the example pump curves presented in the theory
section. How does the pump in the example compare to the pump on the FM50
in terms of capacity, power, and efficiency?
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Calculations
Table 1: Example of data taken from the Software (Setting 50%)
Density
Pump Pump Water Inlet Outlet Motor Flow
Sample Notes of
Setting Speed Temperature Pressure Pressure Torque Rate
Number water
S n T Pin Pout t Q
[%] [rpm] [°C] [kPa] [kPa] [Nm] [l/s]
[kg/m³]
1 50 750 26.7 2.6 18.5 0.62 0.04 997
2 50 750 27.2 2.7 18.3 0.64 0.12 996
3 50 750 26.7 2.3 17.7 0.64 0.21 997
4 50 750 26.9 2.2 17.0 0.66 0.29 997
5 50 750 27.1 1.6 15.4 0.65 0.40 997
6 50 750 27.4 1.3 14.0 0.67 0.49 996
7 50 750 26.7 0.9 13.1 0.66 0.54 997
8 50 750 27.0 0.3 11.0 0.67 0.60 997
9 50 750 26.6 0.2 10.1 0.69 0.64 997
10 50 750 26.7 -0.4 9.5 0.68 0.66 997
11 50 750 26.8 -0.6 8.6 0.68 0.69 997
12 50 750 27.1 -0.9 8.1 0.68 0.72 996
13 50 750 26.7 -1.1 7.1 0.67 0.74 997
14 50 750 27.1 -1.0 7.2 0.70 0.76 996
15 50 750 26.7 -1.1 6.5 0.68 0.76 997
16 50 750 27.5 -1.0 6.2 0.69 0.76 996
17 50 750 27.1 -1.2 6.2 0.72 0.77 996
18 50 750 26.7 -1.2 6.2 0.70 0.77 997
19 50 750 27.4 -1.2 6.4 0.68 0.76 996

Table 1 (Cont.): Example 50% setting (n = 750 rpm)

Inlet Outlet Static Velocity Elevation Total Hydraulic Mechanical Pump Predicted
Velocity Velocity Head Head Head Head Power Power Efficiency Flow
Vin Vout Hs Hv He Ht Ph Pm E Rate
[m/s] [m/s] [m] [m] [m] [m] [W] [W] [%] [l/s]

0.090 0.162 1.627 0.001 0.075 1.70 0.7 48.4 1.3 0.03
0.275 0.495 1.596 0.009 0.075 1.68 2.0 50.5 3.9 0.08
0.491 0.885 1.570 0.028 0.075 1.67 3.5 50.3 6.9 0.14
0.675 1.218 1.516 0.052 0.075 1.64 4.7 52.1 9.0 0.20
0.919 1.657 1.413 0.097 0.075 1.58 6.2 50.8 12.2 0.27
1.135 2.046 1.302 0.148 0.075 1.52 7.3 52.7 13.9 0.33
1.256 2.266 1.250 0.181 0.075 1.51 8.0 51.8 15.5 0.36
1.378 2.485 1.097 0.218 0.075 1.39 8.1 52.6 15.5 0.40
1.468 2.647 1.020 0.247 0.075 1.34 8.4 54.4 15.4 0.42
1.531 2.761 1.012 0.269 0.075 1.36 8.8 53.2 16.6 0.44
1.594 2.875 0.935 0.292 0.075 1.30 8.8 53.3 16.5 0.46
1.653 2.980 0.919 0.313 0.075 1.31 9.2 53.4 17.2 0.48
1.716 3.094 0.837 0.338 0.075 1.25 9.1 52.8 17.2 0.50
1.747 3.151 0.839 0.350 0.075 1.26 9.4 54.9 17.1 0.51
1.747 3.151 0.777 0.350 0.075 1.20 8.9 53.1 16.8 0.51
1.747 3.151 0.733 0.350 0.075 1.16 8.6 54.2 15.9 0.51
1.774 3.199 0.757 0.361 0.075 1.19 9.0 56.2 16.0 0.51
1.774 3.199 0.754 0.361 0.075 1.19 9.0 55.1 16.2 0.51
1.747 3.151 0.775 0.350 0.075 1.20 8.9 53.8 16.5 0.51
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Table 1 (Cont.): 50% setting (n = 750 rpm)

Predicted Vapour Net +ve Pipe Pipe Coefficient Coefficient System Walkthrough
Total Pressure Suction Head Length Diameter k C Head Questions
Head Pv Available L d Loss Score
[m] [kPa] [m] [m] [m] [-] [-] [m] [%]

0.757 36.64 6.73 0.916 0.032 4.9 140 0.06

0.746 37.33 6.68 0.916 4.9 0.21
0.743 36.64 6.72 0.916 4.9 0.40
0.730 36.89 6.71 0.916 4.9 0.58
0.704 37.14 6.67 0.916 4.9 0.86
0.678 37.58 6.64 0.916 4.9 1.13
0.669 36.70 6.72 0.916 4.9 1.29
0.618 37.08 6.66 0.916 4.9 1.46
0.597 36.57 6.73 0.916 4.9 1.60
0.603 36.64 6.69 0.916 4.9 1.69
0.578 36.82 6.68 0.916 4.9 1.79
0.581 37.20 6.63 0.916 4.9 1.88
0.555 36.64 6.69 0.916 4.9 1.98
0.562 37.20 6.65 0.916 4.9 2.04
0.534 36.70 6.69 0.916 4.9 2.04
0.515 37.65 6.60 0.916 4.9 2.04
0.530 37.20 6.64 0.916 4.9 2.08
0.529 36.64 6.70 0.916 4.9 2.08
0.533 37.58 6.59 0.916 4.9 2.04

Pump Curves for different velocities (rpm)

8.00

7.00

6.00
Total Head Ht [m]

750
5.00
900
4.00
1050
3.00
1200
2.00 1350
1.00 1500
0.00
0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60
Flow Rate Q [l/s]

Figure 3: Pump Curves for different velocities

Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Exercise C

Objective
To investigate the use of the affinity laws in predicting the head-flow
characteristic for a pump.

Theory
When selecting a pump for a system, it is seldom practical to test the
performance of every size of pump in a manufacturer's range at all speeds at
which it may be designed to run. It is therefore useful to have a mathematical
solution that allows assumptions can be made about operating characteristics
of a pump running at one speed, impeller size, etc. from experimental results
taken at another.
The multiple curves obtained from plotting measured pump characteristics on
dimensional axes can be reduced to a single curve if appropriate dimensionless
groups are used. Provided the effects of t1uid viscosity on pump performance
are small, and that cavitation is not occurring, the characteristic of a given type
and shape of pump may be represented by:
∫[]

where n is the pump speed (rpm or Hz), and D is the impeller diameter (m)

For a single curve of the type suggested by this equation to represent more
than one operating condition of the particular type of pump, the criterion of
dynamic similarity must be fulfilled. That is, all fluid velocities at
corresponding points within the machine are in the same direction and
proportional to impeller speed. When this is the case, as for a particular pump
operated at different speeds, a simple graph of data is formed, as depicted in
Figure 4:

Figure 4: Dimensionless head-discharge characteristic of a particular centrifugal pump

operated at different speeds
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

The dimensionless equation given previously is the basis from which the
affinity laws are derived. The affinity laws allow the performance of
geometrically similar pumps of different sizes or speeds to be predicted
accurately enough for practical purposes.

The methods used for deriving the affinity laws will not be presented here, but
the laws are as follows:
Power Coefficient ̅

Flow Coefficient
Head Coefficient

These Laws are most often used to calculate changes in now rate, head and
power of a pump when the size, rotational speed or fluid density is changed.
The following formulae are derived from the above considerations, and allow
calculation of total head H, and power Pm at one speed n. to be deduced from
those measured at another speed n2:

More generally, the relationship between two geometrically similar machines

with characteristic diameters D1 and D2 operating at rotational speeds n1 and
n2 is shown in Figure 5. For any two points at which values of (gH / n2D2) and
(Q / nD3) are the same, it follows that:
( )( )

and
( )

These are termed corresponding points.

The power coefficient and the resulting efficiency E can be compared in a
similar manner.

Figure 5: Relationship of performance characteristics for geometrically similar machines

operating at different speeds
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Equipment Set Up

If the results from Exercise B are available then no further data is required.
Ensure you understand the Theory section then proceed directly to the results.
This experiment may be undertaken directly following another experiment, in
which case the equipment will already be prepared and need only be switched
back out of standby mode again.
If the equipment is not yet ready for use, proceed as follows:
Ensure the drain valve is fully closed.
If necessary, fill the reservoir to within 20cm of the top rim. Ensure the inlet
valve and the gate valve are both fully open.
Ensure the equipment is connected to the IFD7 and the IFD7 is connected to a
suitable PC. The red and green indicator lights on the IFD7 should both be
illuminated.
Ensure the IFD7 is connected to an appropriate mains supply, and switch on
the supply.
Run the FM50-304 software. Check that 'IFD: OK' is displayed in the bottom
right corner of the screen and that there are values displayed in all the sensor
display boxes on the mimic diagram

Procedure
The results from Exercise B may be used to perform the calculations and to
create the graphs for this exercise. Where these results are available, no further
data is required. Proceed directly to the Results section. If results are not
available, proceed as follows:
Switch on the IFD7.
Switch on the FM50 pump within the software.
In the software, set the pump to 50%.
Allow water to circulate until all air has been flushed from the system. Close
the gate valve to give a flow rate Q of 0.
Select the icon to record the sensor readings and pump settings on the
results table of the software.
Open the gate valve to give a very low flow rate. Allow sufficient time for the
sensor readings to stabilise then select the icon to record the next set of
data.
Open the gate valve in small increments, allowing the sensor readings to
stabilise then recording the sensor and pump data each time.
Create a new results sheet by selecting the icon (you may also wish to
save the results at this time to avoid losing the data in the event of problems).
Set the pump to 70%.
Close the gate valve.
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Select the icon to record the sensor readings and pump settings on the new
results table.
Open the gate valve to give a very low flow rate. Allow sufficient time for the
sensor readings to stubilise, then select the icon to record the next set of
data.
Repeat, opening the gate valve in small increments and allowing the sensor
readings to stabilise, then recording the sensor and pump data each time.
Ensure the results are saved using 'Save' or 'Save As .. .' from the software File
menu after taking the final set of results.
Switch off the FM50 within the software using the Power On/Standby button.
Switch off the IFD7.
Results
The results taken at 70% will be used with the affinity laws to give predicted
results at 50%. This will then be compared to the actual results at 50%. The
software uses the affinity laws
and
to calculate the predicted values of Ht2 at predicted flow rates Q2 and 50%
setting from the measured values of Htl and Q1 and the values n1 = 70 and n2
= 50.
Plot a graph of Predicted Head against Predicted Flow Rate.

Plot the measured Total Head at 50% against measured Flow Rate at 50% (if
the data is exported into a dedicated spread sheet package or similar then it
may be possible to plot both graphs on the same axes).

Conclusion
Compare the predicted results at 50% with the measured results. How accurate
were the values obtained using the affinity laws? Discuss the advantages and
disadvantages of this technique for pump system design
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Calculations

Table 2: Data for 50% Setting and 70% setting from software

Practical
70% = 1050 rpm 50% = 750 rpm

Flow Total Flow Total

Sample
Rate Head Rate Head
Number
Q Ht Q Ht
[l/s] [m] [l/s] [m]

1 0.08 3.41 0.04 1.70

2 0.15 3.38 0.12 1.68
3 0.27 3.26 0.21 1.67
4 0.43 3.26 0.29 1.64
5 0.56 3.11 0.40 1.58
6 0.66 2.99 0.49 1.52
7 0.76 2.88 0.54 1.51
8 0.82 2.79 0.60 1.39
9 0.89 2.68 0.64 1.34
10 0.93 2.63 0.66 1.36
11 1.00 2.58 0.69 1.30
12 1.01 2.52 0.72 1.31
13 1.04 2.42 0.74 1.25
14 1.04 2.33 0.76 1.26
15 1.06 2.44 0.76 1.20
16 1.05 2.34 0.76 1.16
17 1.06 2.34 0.77 1.19
18 1.08 2.38 0.77 1.19
19 1.08 2.35 0.76 1.20
20 1.06 2.34

Practical
4.00
3.50
3.00
Total Head Ht [m]

2.50

2.00 70% = 1050 rpm

1.50
50% = 750 rpm
1.00
0.50
0.00
0.00 0.50 1.00 1.50
Flow Rate Q [l/s]

Figure 6
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Using Similarity Laws to calculate Q and Ht for 50%

and
Table 3: Data for 70% setting from software and 50% Setting from Similarity Laws

Similarity Laws
70% = 1050 rpm Calculated 50% = 750 rpm

Flow Total Flow Total

Sample
Rate Head Rate Head
Number
Q Ht Q Ht
[l/s] [m] [l/s] [m]

1 0.08 3.41 0.06 1.74

2 0.15 3.38 0.10 1.72
3 0.27 3.26 0.19 1.66
4 0.43 3.26 0.30 1.66
5 0.56 3.11 0.40 1.59
6 0.66 2.99 0.47 1.53
7 0.76 2.88 0.54 1.47
8 0.82 2.79 0.59 1.42
9 0.89 2.68 0.64 1.37
10 0.93 2.63 0.66 1.34
11 1.00 2.58 0.71 1.32
12 1.01 2.52 0.72 1.29
13 1.04 2.42 0.74 1.23
14 1.04 2.33 0.74 1.19
15 1.06 2.44 0.76 1.24
16 1.05 2.34 0.75 1.19
17 1.06 2.34 0.76 1.19
18 1.08 2.38 0.77 1.22
19 1.08 2.35 0.77 1.20
20 1.06 2.34 0.76 1.19

Affinity Laws
4.00
3.50 70% = 1050
3.00 rpm
Total Head Ht [m]

2.50
2.00 Calculated
1.50 50% = 750
1.00 rpm
0.50
0.00
0.00 0.50 1.00 1.50
Flow Rate Q [l/s]

Figure 7
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Exercise D

Objective
To investigate the effect of changing inlet head on pump performance.

Method

By varying the pressure at the inlet to the pump using a manual valve to
control the available flow.

Theory

In both the design and operation of a rotodynamic machine, careful attention

has to be paid to the fluid conditions on the suction side. In particular, it is
important to check the minimum pressure that can arise at any point to ensure
that cavitation does not take place.

Cavitation

If the pressure at any point is less than the vapour pressure of the liquid at the
temperature at that point, vaporisation will occur. This is most likely to arise in
the suction side where the lowest pressures are experienced. The vaporised
liquid appears as bubbles within the liquid, and these subsequently collapse
with such force that mechanical damage may be sustained. This condition,
known as cavitation, is accompanied by a marked increase in noise and
vibration in addition to the loss of head.

.‫ جيرذحالب‬suction pipe ‫يف دىجىوال ماوصال قالغإ كيزط يع تاعامفال ىىكج ةيؤر يكوي‬

FM 51 ‫ةظداله‬: ‫زخالا ساهجال يف خضوا ىىكج تاعامفال‬

Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal Pump

Exercise E

Objective
To obtain a Head - Flow curve for the piping system through which the fluid is
to be pumped. To determine the operating point of the pump.

Theory
System analysis for a pumping installation is used to select the most suitable
pumping units and to define their operating points. System analysis involves
calculating a head - flow curve for the pumping system (valves, pipes, fittings
etc.) and using this curve in conjunction with the performance curves of the
available pumps to select the most appropriate pump(s) for use within the
system.
The system curve is a graphic representation of the flow rate in the system with
respect to system head. It represents the relationship between flow rate and
hydraulic losses in a system. Such losses are due to the system design (e.g.
bends and fittings, surface roughness) and operating conditions (e.g.
temperature).
Assuming that
  Flow velocity is proportional to volume now rate
 Losses in the system are proportional to the square of the now velocity

it follows that system head loss must be proportional to the square of the
volume flow rate, and the system head - now graph will therefore be parabolic
in shape.
‫ساهجالب صاخال جىالحكال يف دىجىه حزشال يالب‬

Calculations

Table 4
System Curve
Pump Pump Flow Total
Sample
Setting Speed Rate Head
Number
S n Q Ht
[%] [rpm] [l/s] [m]
1 100 1500 1.49 4.28
2 90 1350 1.36 3.78
3 80 1200 1.22 3.03
4 70 1050 1.08 2.33
5 60 900 0.92 1.71
6 50 750 0.77 1.16
7 40 600 0.61 0.74
8 30 450 0.46 0.38
9 20 300 0.30 0.15
10 10 150 0.13 -0.01
11 0 0 0.00 -0.05
Hydraulics Lab - ECIV 3122 Experiment (9): Centrifugal
Pump

Table 5
Pump Curve
Total Flow Pump Pump
Sample
Head Rate Speed Setting
Number
Ht Q n S
[m] [l/s] [rpm] [%]
2.28 1.08 1050 70 1
2.38 1.02 1050 70 2
2.55 1.00 1050 70 3
2.60 0.97 1050 70 4
2.69 0.92 1050 70 5
2.74 0.85 1050 70 6
2.86 0.80 1050 70 7
2.89 0.69 1050 70 8
3.10 0.61 1050 70 9
3.16 0.49 1050 70 10
3.28 0.35 1050 70 11
3.34 0.24 1050 70 12
3.39 0.13 1050 70 13
3.41 0.09 1050 70 14

Operating Point
5.00
4.00
3.00 [m]HtHeadTotal
Operating Point
Pump Curve 2.00
System Curve 1.00
0.00
2.00 1.50 1.00 0.50 0.00
-1.00

Flow Rate Q [l/s]

Figure 8

‫ جن اخذها عٌذ‬system curve، ‫ ألى لزاءات‬pump curve‫ ًمطة الحشغيل في الزسوة كاًث اخز ًمطة في‬:‫هالدظة‬
flow( outlet valve fully opened‫أعلى‬