MT was proud to publish this article by one of the most recognized and respected technical experts in the pump world as part of our very first Utilities Manager supplement in February 2006. Almost seven years later, its recommendations are as relevant as ever for reliability- and efficiency-focused users of pumping systems.
Pump systems in industry often are connected to various types of monitoring and control systems. Typically, end users will only monitor one or several parameters, such as flow rate, pressure, power and/or current. That rather limited type of monitoring, though, really doesn’t provide a full view of a pump’s performance. By just monitoring a few individual parameters, a company risks overlooking important information about system performance as a whole.
It is difficult to find useful key performance indicators (KPIs) to track in the quest for improved efficiencies in a pumping system. Tracking maintenance costs by themselves certainly can provide good data on the health of a system. But, in many cases, better information can be obtained by combining several parameters into a one KPI and following it over time. One way to do this is to relate flow and power, rather than to simply measure them independently. Let’s examine how this can be done.
Fluid system relationships
For pump systems, the relationship between fluid power, flow rate, pressure (head) and specific gravity of the fluid can be expressed in the following equations:
In order to determine the electrical power used by the system it is necessary to divide the fluid power by the efficiencies of the different components that produce it: i.e. motor, drive and pump efficiencies.
Specific energy (Es) is good measure to use for calculating the cost of pumping. This is the energy used to move a certain volume through the system. It is measured as Watt hours/gallon, or any other suitable units, and it has the advantage of being a direct measurement of the cost of pumping once you know the cost of energy. (This is the same as measuring the miles per gallon for a car.)
As seen in equation (3), the instantaneous value of the specific energy equals the input power divided by the flow rate. Using the equation for Fluid Power and dividing by the various efficiencies we get Equation 4:
From this we get Equation 5:
If there is no drive in the system, the corresponding term is replaced by 1.
Specific energy is a useful measure for comparing different system solutions and the cost of pumping. In systems where the flow is constant, this is a simple task by using the equations above. In systems with varying flow rates, it becomes a little more complicated.
First, Es needs to be calculated as a function of flow rate, which requires information from pump, motor and drive manufacturers. The pump manufacturer has to provide pump curves for variable speed operation, while the motor and drive suppliers have to provide efficiency curves as a function of load and speed.
Specific energy (Es) is a linear function of the head if the other factors are constant. We can, therefore, plot it as a function of head for different overall efficiencies. See Fig. 1 where the overall efficiency is the product of the different component efficiencies from Equation 5.
Fig. 1. Specific energy as a function of head for different overall efficiencies (click to enlarge)
The lowest line in the diagram represents 100% efficiency, which, of course, is not reachable. If input power, flow rate and head are available at a specific duty point, it is relatively easy to mark the value of Es as a point in Fig. 1, and then interpolate the overall efficiency.
Using a program such as the U.S. Department of Energy’s (DOE’s) Pump System Assessment Tool (PSAT), the best available pump and motor efficiencies for a specific duty point can be found and a lowest possible specific energy can be calculated for the duty point in question, if there is no VSD involved. It is harder to find out what the efficiency of a motor/drive combination is, as it varies depending on how well these two components fit together. The drive, if present, introduces additional losses in the motor that should be accounted for.
Fig. 2. Combined motor/drive efficiency when attached to different drives (DOE)
(Fig. 2 gives an idea about the combined efficiency of a drive and motor. Motors react differently to different drives. It is, therefore, a good idea to buy a drive and motor from the same manufacturer to assure a combination that is well matched. Modern drives have improved considerably compared to those available some years ago.)
Fig. 3. Controlling output of a centrifugal pump
After using the PSAT, a new point below the first can be plotted in the specific energy diagram. (See Fig. 4, where this has been done.) Potential savings from using better-matched equipment at the upper duty point can be calculated from the difference in specific energy. PSAT does it automatically.
Fig. 4. Specific energy at different operating points
Pump control methods
In many applications, pump flow is routinely controlled by throttling a control valve on the pump discharge. The throttled valve can be controlled to maintain a designated flow rate, pressure or any other parameter to satisfy system needs. When a valve is used to control flow, the flow rate is decreased by increasing resistance in the pipe system and moving the operating point up the pump head curve as shown in Fig. 3.
From the information in Fig. 3, it is possible to estimate the head necessary to produce a certain flow in the system if there were no throttling. (One way of getting the pump to operate at such an unthrottled duty point would be to use a variable speed drive to reduce pump output.) Head requirement without throttling can be read on the original unthrottled curve below the throttled operating point. Using PSAT, the best motor and pump efficiencies for this unthrottled operating point can be found and Es can be calculated and plotted in the specific energy diagram. The result is a good graphical representation of where the system is operated from a cost point of view and where it could operate if optimized and without throttling losses (see Fig. 4).
In the Fig. 4 example, the original specific energy is about 4.8 Wh/Gallon at 700 ft. The optimized pump/motor combination is about 2.8, and the specific energy with an optimized pump/motor combination without throttling losses is around 1.2 Wh/Gallon at 300 ft.
Avoiding throttling losses is of the utmost importance when it comes to economic pumping. In the case of varying flow rates, Es has to be calculated for different flow rates. The corresponding head is obtained from the pump curve. The total operational cost can be obtained if the flow distribution as a function of time is known.
Key performance indicators
As indicated above, it is very useful to track the performance of a pump system not only to see how efficiently it is operating relative to optimum, but also to be able to easily and quickly discover deviations from the norm. It is, therefore, suggested that power-divided-by-flow rate would be a very useful KPI if flow rate and power are monitored.
In many industries, motor current is monitored instead of power. The current is roughly proportional to power and can be used as a substitute for power. In both cases, the quotient will be a very good measure of the system efficiency and also sensitive to system changes.
In a recent assessment of a pump station, it was found that one of three parallel pumps was drawing about 100kW, but contributed almost nothing in flow. The pump was practically dead headed by the other two pumps since the wear rings were badly worn and the delivered head, therefore, was lower than for the other pumps. As the system was set up to monitor total flow and motor currents independently, the problem with the pump was hidden. If the current-divided-by-flow rate for the system as a whole had been monitored and tracked, the problem would have caught someone’s attention sooner.
Another very valuable performance indicator is the repair record of a pump. Pumps are generally quite reliable. If they are operated close to their best efficiency points (BEPs), they should last for a long time. If a specific pump deviates from the rest of the pumps in a facility, or shows a rising curve for the cost of maintenance, there is good reason to investigate it further. Remember, there is a very strong relationship between reliability and efficient operation. Thus, an inefficiently operated pump usually costs more all the way around–not only to run, but also to maintain.
Pump systems should be continuously monitored and tracked using key performance indicators (KPIs) to discover performance improvement possibilities.
Maintenance costs are particularly telling. Don’t underestimate the benefits of using maintenance cost records for individual pumps as KPIs.
Power divided by flow rate also can be a very useful KPI for the task. This method gives a number that is directly proportional to the cost of pumping and is sensitive to changes in the system. Furthermore, it can easily be demonstrated graphically, so that operators get a visual picture of how efficiently their system is operating.
Tracking these types of KPIs will provide crucial clues in your quest for improved efficiency. MT