I often talk to pump owners/operators who question why the function of a given pump does not conform to the function curve published by the manufacturer. Of course, there is a long list of reasons for presenting this problem. I will explain the list of potential problems in this column. One possible answer is that the pump functions well, but you may not have the correct measurement function. ALLWEILER

Initially, it is important to understand that pump function tests (field tests in owner/operator systems) performed on site are rarely as accurate as factory tests. Separation is generally caused by the accuracy, calibration plan and placement of the instrument. Other relevant elements are the size and geometry of the pipe. In addition, there is a long list of items, including correct calculation, method and compensation/available readings (NPSHa) for torque speed, atmospheric pressure, acceleration head, temperature, specific gravity, viscosity, air entrapment, net positive suction head, critical submergence, Newtonian characteristics, friction loss and elevation calibration for specific reference points.

In addition, the pumps tested in the test facility will have the correct settings, clearances and smoothness. They also have the active advantage of being new.

Pressure gauge and sensor

Unless otherwise specified, I will replace pressure gauge, digital gauge and sensor in this column. These instruments are sold at various precision levels, and higher precision generally means higher cost. The accuracy of most surfaces used in commercial and industrial applications is significantly lower than that of laboratory, fine or test instruments. The appearance of the laboratory may be accurate to+/- 0.05% in the whole scale of the instrument, while most commercial units do well in the scale of+/- 2% of the reading, and may be worse in the whole scale.

There are differences between precision and fineness and between precision and resolution. Just because the digital surface reads the pressure to two decimal places (resolution) does not mean that it is accurate. Resolution is defined as the minimum increment of pressure change that can be reliably detected by the sensor. Accuracy is defined as the closeness between the instrument reading and the actual pressure. Accuracy is the goal of consistent and repeatable measurement. The person who reads the appearance and the size (resolution) of the dial will also be a factor. The smaller the dial, the lower the resolution, so the greater the possibility of human error when accurately reading the appearance. Many field exterior panels are 1.5 to 2 inches in diameter, while laboratory or test facilities may be 16 inches in diameter. The point of view of appearance will also play a role.

Attention: often install the appearance with incorrect scale. For the pressure you want to measure, it is best to use a medium scale or middle one-third scale pressure gauge. It is difficult and inaccurate to attempt to measure the difference between 2 and 5 psig on a 0 to 200 psi gauge.    

Due to temperature changes, oscillations and pulsations, the appearance in service/operation will lose its accuracy. Pressure measurement will also be affected by "drift", and as time goes by, lag factors, non repeatability and nonlinear factors need to be considered, which is why regular calibration is also required to adhere to accurate readings.

The appearance of the testing laboratory will be calibrated regularly according to the strict schedule, while many appearances on the site will not be subject to any strict calibration or calibration at all. Apologize and praise the equipment that does have a formal calibration schedule and procedure.

Where to measure

The placement of the instrument will affect the reading. In theory and practice, the manufacturer guarantees that the pump performance (head and flow at least) is within the given service scale. The differential pressure data points of the pump are suction and discharge flanges, but these are also positions that cannot be accurately stated by the instrument. In the test equipment, the suction and discharge pressure of the pump is generally monitored at at least two pipe diameters of the spacer flange (the more the better), but we still need to calibrate the pressure (head) reading to explain the friction loss of these short pipe sections and bring the benchmark back to the flange.

Due to recirculation problems, especially at the operating point on the left side of the pump curve, the suction pressure may be affected by recirculation problems, and then lead to wrong readings. No matter where you measure the pressure, the corresponding liquid velocity distribution needs to be symmetrical. Simple work such as the scale and contour of the gauge tap and the position relative to several planes of the pump volute will affect the accuracy.

When calculating the head, the height (positive or negative) of the surface relative to the pump reference plane must also be compensated. In addition, the temperature and density of the liquid as well as the local atmospheric pressure (altitude calibration) must also be considered. For pump testing, the basic standard liquid is water at 68 F or 20 C. If the water temperature is between 50 and 86 F (10 and 30 C), there is almost no need to compensate the temperature, because the error is very small.

Due to the change of elastic modulus of bourdon tube, the accuracy of pressure gauge will be affected by the ambient temperature. Generally, it is not a problem, but in case of extreme temperature change, please verify the scale deviation value with the manufacturer.

Pump curve

The pump function curve (also known as factory curve or price manual curve) issued by the manufacturer represents typical hydraulic function. Due to the reasons now clarified and the existence of casting and fabrication service in the general equation, your pump function on site (better or worse) will be different. The manufacturer tested the pump and released the results as a verification and guarantee for the function and accuracy of the pump model. You should also know that most commercial and industrial pumps are not function tested alone. The manufacturer only used the published curve to clarify that similar pumps with the same design, type and scale will operate within certain regular service scale of the test pump.

Most producers will periodically retest the pump to revalidate functionality. The service scale will be determined by the expected pump service and some other factors including power scale. These general services are typically between+/- 2% and 6%, but may be higher depending on the pump test level. As for the agreed guarantee points, the public service will be lower. Details are defined in ANSI/HI Specification 14.6.

Making public service

The pump volute/casing and impeller are castings. The casting process is inherently a service of scale and surface finish. Compared with simple sand casting, investment casting will produce significantly better dimensions and finish. Of course, as the word "investment" implies, they also need higher costs. Compared with parts with rough surface finish, impellers and housings with lubricating surface finish have higher power due to less conflict. In addition to casting service, there are also manufacturing/processing service and finish. If you want to make thousands of pumps every year, the ability to keep your service and finish within the same standard scale over time and in large quantities is the most important, and it is also a sign of excellent manufacturers.

Other elements

Air entrainment as low as 1% will negatively affect the function of the pump. Even if the viscosity value is as low as 15 centipoises, it will affect the full function of the pump. You will need to correct the flow, head, and power based on the viscosity of the liquid. See HI specification 9.6.7 ("Influence of viscosity on pump function") for details.

If you do not use the correct specific gravity to convert pressure to head, the calculation will not be accurate. The placement of the flowmeter and the type/technology selected will affect the accuracy of the instrument. Follow the manufacturer's instructions clearly, understand the number of straight pipes before appearance, and pay attention to the sensitivity of some appearances to direction. The portable appearance will warn against correct placement; Your understanding of the velocity flow curve and the presence of solids in the liquid flow will be key to a successful installation.

The pipe size and geometry usually hinder the measurement accuracy. ALL Authorized agent of WEILER In the name of reducing the floor area, I often see that the 90 degree pipe elbow next to the pump suction and the suction filter with low valve flow coefficient (Cv [high pressure drop]) will not only affect the function, but also lead to inaccurate pressure and flow readings. I also witnessed the suction side velocity exceeding 10 to 15 feet per second (3 to 4.6 meters per second), and even higher in a few cases, which will lead to inaccurate pressure readings, because the surface/sensor cannot sense the component of the velocity pressure head.

The function of the pump depends on the correct operating clearances and settings. Refer to the pump manual and/or manufacturer for guidance in this regard.

Speed is a key element of pump function. I don't often see this problem, but when I look at it, it will vividly stay in my mind. For example, the VFD is set or programmed incorrectly, the belt and pulley dimensions are set incorrectly, the gear dimensions are incorrect, the single-phase induction motor, the turbine with faulty governor, and in remote areas, I have seen that the diesel generator frequency switch is set to run 50 cycles in 60 cycle applications. If in doubt, please measure the speed to eliminate the elements in the list of non peak or non functional causes. ALLWEILER Screw pump

If the pump shows cavitation, the function will be affected by bad luck. Measure or calculate the NPSH margin to ensure that the pump operates at a satisfactory scale. The pump will run where the system curve intersects the pump curve.

In some equipment, the common practice is to operate the pump at a closed head, even if it is to "check the health of the pump" for a short time. Due to recirculation problems, rapid temperature changes and inability to compensate the velocity head, it is difficult to accurately measure the pump function when the pressure head is closed (no flow). I suggest that perhaps a small number of traffic will allow more accurate testing, of course, only for a short time. The more energy I touch, the less I recommend this practice. It's a good decision to check with the producer.