This article discusses factory performance testing of vertical pumps. For guidelines on specific test procedures, calculations and specific requirements, refer to ANSI/HI - 2.6 Standard. A performance test is only one type of test. Tests not discussed in this article include hydrostatic tests, NPSH required tests and mechanical/vibration tests.
A critical phase in the design and construction of a pumping station is a witnessed pump performance test at the manufacturer's facility. The purpose of a pump performance test is to develop the actual pump curve of the assembled pump. Since no two pumps will perform exactly the same, a factory pump performance test will determine the actual flow, head and efficiency of a pump along its operating range. The standard that governs a factory performance test for vertical pumps is the American National Standard for Vertical Pump Tests (ANSI/HI - 2.6). The standard describes certain aspects of vertical pump tests, including types of tests, acceptable methods of measurement, test procedure and acceptable performance tolerances.
The Test Setup
Each manufacturer may use a different setup for testing vertical pumps. Some manufacturers use a wet pit, while some may use a pump can. Different types of meters measure flow, and different methods measure head. Power can be measured as input power to the motor, or input power to the pump using torque and speed. Regardless of the pump test setup, the tested pump should include all major components of the installed pump. These components include the bowl assembly, column pipe assembly, discharge head and motor.
Some concessions can be made to accommodate testing. For example, if the column pipe assembly of the finished pump will be too long to test in the factory, it is acceptable to eliminate sections of the column pipe for the performance test, as long as the performance of the complete bowl assembly is not affected. All components of a pump factor into the overall performance, so the most accurate performance test will include all components of the finished pump.
Performing the test in the factory is of utmost importance for several reasons. First, when the test is performed in the factory, it is typically in close proximity to where the pump was assembled. If the test reveals that adjustments need to be made, it is possible to hoist the pump off the test bench and make the necessary adjustments. For example, it can be determined in the factory if the performance curve is not within the specified tolerances. If this were found after the pump was installed, fixing the problem could be extremely difficult and costly.
Second, the factory presents a controlled atmosphere. Water levels, pipe lengths, pipe dimensions and suction to the pump can all be controlled. These are all possible constraints in the field, but non-issues in a factory.
Third, the manufacturer should have extensive experience in testing pumps at its facility. The manufacturer will have all the measuring instruments calibrated and know all the factors to be considered in the test. In general, a test performed in the factory has significant advantages over testing once the pump has been installed.
The Role of the Witness
The pump test witness must have a thorough understanding of all aspects of the test, the project specifications, the ANSI/HI - 2.6 standard, the tested pump and project details.
The project specifications should state the pump's required performance, as well as other requirements of the performance test. The witness must verify that all requirements of the specification are followed. The standard has several provisions for how the test should be set up, what types of instruments should be used, where the instruments should be located, how the test should be run and acceptance criteria for the results. It is important for the witness to understand how the pump will be used in operation. The witness should use engineering judgment when evaluating results.
The witness must verify the results of the test. Results of the pump performance test include the flow, head and efficiency along the tested pump's operating range. The total dynamic head verification starts with a pressure gauge reading but also requires additional calculations. Efficiency involves comparing power delivered to the pump (pump input) with power delivered to the water (pump output).
The Performance Test
A flowmeter, including venturi, electromagnetic or propeller, measures flow. The flowmeter must be calibrated prior to the pump test, and current calibration reports should be provided to the witness prior to or at the time of the pump test.
Throttling a control valve on the pump's discharge side creates the various points of the pump curve, and the flow is recorded from the flowmeter.
Determining Total Dynamic Head of the Pump
In a pump station design, the design consultant will determine the required design point of a pump through the development of a system curve. The pump supplier will then use the design point along with other factors to choose the optimum pump for the application. The pump test must verify the total dynamic head of the pump.
The total dynamic head of the pump is determined from the read-out of the pressure gauge. However, the witness must remember that the pressure gauge simply displays the pressure (or head) that is "seen" at that point for a given flow. The total dynamic head of the pump includes several other factors. The static head must be determined, which is the difference between the static suction head and static discharge head. For a wet pit test, this is simply the distance from the water level to the pressure gauge. The velocity head must be considered, which uses the velocity in the discharge pipe. Any minor or friction losses must be added in between the pump discharge and the pressure gauge.
Depending on the test setup, each of these factors is accounted for in a different way. To accurately determine the TDH of the pump, the witness must be familiar with all aspects of the test setup and measurement.
Efficiency is simply defined as the ratio of output over input. To determine efficiency, the pump input power and pump output power must be obtained. Pump input power is the power that the driver delivers to the pump. It can be measured using the speed and torque of the pump, or by determining the input power to driver.
Pump output power is determined from the flow and total dynamic head. The efficiency of the pump is defined as the ratio of the pump output power over the pump input power. A manufacturer's pump curve will typically display not only the flow and head performance of the pump, but also the efficiency along the pump curve. The efficiency of the pump is based on the head delivered by the pump (total dynamic head), compared to the power input.
The purpose of the testing procedure is to develop a performance curve of the actual pump being installed on the project, and to use that curve to compare it to the specified curve. The ANSI/HI - 2.6 Standard has established allowable tolerances for accepting a pump performance test. Table 1 shows the acceptance criteria for flow, head and efficiency.
There are a few important things to note about Table 1. First, the acceptance tolerance only applies to the specified design point. Unless it is stated in the project specifications, only the design point must fall within the acceptance tolerance. Second, no negative tolerances are allowed. The flow, head and efficiency of the tested pump must all be equal to or greater than the specified design point, but within the acceptable tolerances. Third, the efficiency requirement must be met, but it is only required to comply with either the flow or the head tolerance at the design point.
After testing is complete and the tested pump curve has been established, results of the test should be discussed with the pump manufacturer. Results should be compared to determine if there are any areas of disagreement. The pump manufacturer often knows aspects about the pump test setup that the witness may not have known. Especially if results differ, all equations, measurements and assumptions should be discussed until the witness is confident with all results and how they were obtained.
The witness should provide a technical memorandum to the client that summarizes the results of the test. If the pump fails to meet the acceptance criteria, it is the manufacturer's responsibility to make the necessary adjustments to the pump, and perform another test until the requirements are met.
1. American National Standards Institute, Inc./Hydraulic Institute - Section 2.6, Vertical Pump Tests.
Pumps & Systems, October 2009