ENERGY_PERFORMANCE_ASSESSMENT_FOR_EQUIPMENT AND UTILITY SYSTEMS
(CHAPTER-7:ENERGY PERFORMANCE ASSESSMENT OF PUMPS)
Introduction
Pumping is the process of addition of kinetic and potential energy to a liquid for the purpose of moving it from one point to another. This energy will cause the liquid to do work such as flow through a pipe or rise to a higher level. A centrifugal pump transforms mechanical energy from a rotating impeller into a kinetic and potential energy required by the system. The most critical aspect of energy efficiency in a pumping system is matching of pumps to loads. Hence even if an efficient pump is selected, but if it is a mismatch to the system then the pump will operate at very poor efficiencies. In addition efficiency drop can also be expected over time due to deposits in the impellers. Performance assessment of pumps would reveal the existing operating efficiencies in order to take corrective action.
Purpose of the Performance Test
« Determination of the pump efficiency during the operating condition
« Determination of system resistance and the operating duty point of the pump and compare the same with design
Performance Terms and Definitions
Pump Capacity, Q = Volume of liquid delivered by pump per unit time, m3/hr or m3/sec
Q is proportional to N, where N- rotational speed of the pump
Total developed head, H = The difference of discharge and suction pressure
The pump head represents the net work done on unit weights of a liquid in passing from inlet of the pump to the discharge of the pump.
There are three heads in common use in pumps namely
(i) Static head
(ii) Velocity head
(iii) Friction head
The frictional head in a system of pipes, valves and fittings varies as a function (roughly as the square) of the capacity flow through the system.
System resistance: The sum of frictional head in resistance and total static head.
Pump Efficiency: Fluid power and useful work done by the pump divided by the power input in the pump shaft.
Field Testing for Determination of Pump Efficiency
To determine the pump efficiency, three key parameters are required: Flow, Head and Power. Of these, flow measurement is the most crucial parameter as normally online flow meters are hardly available, in a majority of pumping system. The following methods outlined below can be adopted to measure the flow depending on the availability and site conditions.
Flow Measurement, Q
The following are the methods for flow measurements:
« Tracer method BS5857
« Ultrasonic flow measurement
= Tank filling method
= Installation of an on-line flowmeter
Tracer Method
The Tracer method is particularly suitable for cooling water flow measurement because of their sensitivity and accuracy. This method is based on injecting a tracer into the cooling water for a few minutes at an accurately measured constant rate. A series of samples is extracted from the system at a point where the tracer has become completely mixed with the cooling water. The mass flow rate is calculated from:
The tracer normally used is sodium chloride.
Ultrasonic Flow meter
Operating under Doppler effect principle these meters are non-invasive, meaning measurements can be taken without disturbing the system. Scales and rust in the pipes are likely to impact the accuracy. « Ensure measurements are taken in a sufficiently long length of pipe free from flow disturbance due to bends, tees and other fittings.
« The pipe section where measurement is to be taken should be hammered gently to enable scales and rusts to fall out.
« For better accuracy, a section of the pipe can be replaced with new pipe for flow measurements.
Tank filling method
In open flow systems such as water getting pumped to an overhead tank or a sump, the flow can be measured by noting the difference in tank levels for a specified period during which the outlet flow from the tank is stopped. The internal tank dimensions should be preferable taken from the design drawings, in the absence of which direct measurements may be resorted to.
Installation of an on-line flowmeter
If the application to be measured is going to be critical and periodic then the best option would be to install an on-line flowmeter which can get rid of the major problems encountered with other types.
Determination of total head, H
Suction head (hS)
This is taken from the pump inlet pressure gauge readings and the value to be converted in to meters (1kg/cm2 = 10. m). If not the level difference between sump water level to the centerline of the pump is to be measured. This gives the suction head in meters.
Discharge head (hd)
This is taken from the pump discharge side pressure gauge. Installation of the pressure gauge in the discharge side is a must, if not already available.
Determination of hydraulic power (Liquid horse power),
Measurement of motor input power
The motor input power Pm can be measured by using a portable power analyser.
Pump shaft power
The pump shaft power Ps is calculated by multiplying the motor input power by motor efficiency at the existing loading.
Pump efficiency
This is arrived at by dividing the hydraulic power by pump shaft power
Example of pump efficiency calculation
Illustration of calculation method outlined
A chemical plant operates a cooling water pump for process cooling and refrigeration applications.
During the performance testing the following operating parameters were measured;
Measured Data
Pump flow, Q 0.40 m3/s
Power absorbed, P 325 kW
Suction head (Tower basin level), h1 +1 m
Delivery head, h2 55m
Height of cooling tower 5m
Motor efficiency 88 %
Type of drive Direct coupled
Density of water 996 kg/ m3
Pump efficiency
Flow delivered by the pump 0.40 m3/s
Total head, h2 -(+h1) 54m
Hydraulic power 0.40 x 54 x 996 x 9.81/1000 =211 kW
Actual power consumption 325 kW
Overall system efficiency (211 x 100) /325= 65%
Pump efficiency 65/0.88 = 74 %
Determining the System resistance and Duty point
Determination of the system resistance curve and imposing the pump curve over it will give an idea of the operating efficiency of the pump and also the drop in efficiencies when the system curve changes from normal / design. The example following from the earlier example outlines the method of constructing a system curve.
Example:
Location of equipments
The Refrigeration plant is located at +0.00 level and the Process plant condensers are located at +15
m level. One cooler having a design pressure drop of 1.9 kg/cm’ is located at the 0.00 level (ground
level). Other relevant data can be inferred from the earlier section. See schematic in Figure 7.1.
The step-by-step approach for determining system resistance curve is given below.
Step-1 Divide system resistance into Static and dynamic head
Find static head;
Static head (Condenser floor height) = 15 m
Find dynamic head;
Dynamic Head = Total Head — Static Head
Dynamic head = (54-15) =39 m
Step-2 Check the maximum resistance circuit
Resistance in the different circuits is as under
It can be noted that at full load the condenser and cooler circuits offer the maximum resistance to flow.
Step-3 Draw system resistance curve
Choose the condenser loop as it offers maximum resistance and is also having a static head component
Static head: 15 m
Dynamic head at full load: 39 m
Compute system resistance at different flow rates
Step-4 Plot the system resistance against flow in the pump efficiency curves (see Figure 7.2) provided by the vendor and compare actual operating duty point and check whether it operates at maximum efficiency. In the example provided it is found that the pump system efficiency is lower by 4 % due to change in operating conditions.
Solved Examples:
1) Ina municipality pumping system, water is pumped from the river to an underground circular sump of 8 metre dia in the intermediate booster station. Flow measurements were carried out by level difference in the sump. Pump takes 10 minutes to fill 1 metre level of circular sump. Pressure gauges are not available in the pumping system. The discharge pipe is horizontal, 300mm dia and 8 km long.
Friction factor for the pipe is 0.006. The pump has a negative suction of 2 metre.
The details of power measurements at motor are:
3 phase voltage: 415 V, line current: 93 A and power factor: 0.89. The efficiency of the Motor is 0.90.
As an energy auditor, work out the following:
a) Flow rate of the pump in m*/hr
b) Power drawn by the motor in kW
c) Total head developed by the pump (ignore friction losses in suction piping)
d) Operating efficiency of the pump
2) A centrifugal clear water pump rated for 800 m3/hr was found to be operating at 576 m3/hr with discharge valve throttled. The pumps speed is 1485 RPM. The discharge pressure of the pump before the throttle valve is 2 kg/cm’g. The pump draws the water from a sump 4 metres below the centerline of the pump. The input power drawn by the motor is 124 kW at a motor efficiency of 92%.
(1) Find out the efficiency of the pump.
(ii) If the normal required water flow rate is 500 m?/hr to 700 m?/hr, what in your opinion
should be the most energy efficient option to get the required flow rate variation?
(iii) | And what would be the pump shaft power for that most energy efficient option if the
pump is delivering the flow rate of 550 m3/hr.
3) a) Acentrifugal water pump operates at 30 m?/hr and at 1440 RPM. The pump operating efficiency
is 65% and motor efficiency is 89%. The discharge pressure gauge shows 3.4 kg/cm”. The suction is
3 m below the pump centerline. If the speed of the pump is reduced by 25 %, estimate the following:
i) pump flow
il) pump head and
iii) motor power
Assume motor and pump efficiency remains same at the reduced speed.
b) Calculate pressure drop in meters when pipe diameter is increased from 250 mm to 300 mm for a
length of 600 meters. Water velocity is 2 m/s in the 250 mm diameter pipe and friction factor is 0.005.
4) In a commercial building, an energy auditor recommended to bring down the cooling tower from
the terrace to the ground with a view to save energy in the pump. Details are given in the sketch below.
Ignoring the friction losses, will this measure save energy? Explain with reason.
Answer:
No, because the pressure differential across the pump will be same as friction losses are ignored.
----------------------------------
Comments