Busting Ghosts

There are many diagnostic techniques and tools for troubleshooting pumps. Any one of them—or in some cases, all of them—might be used to accurately determine why a pump system is performing poorly. The symptoms are many, including increased component wear, premature bearing failure, reduced mechanical seal life, motor overload, noisy pump operation, or simply not pumping enough fluid.

04/01/2004


There are many diagnostic techniques and tools for troubleshooting pumps. Any one of them—or in some cases, all of them—might be used to accurately determine why a pump system is performing poorly.

The symptoms are many, including increased component wear, premature bearing failure, reduced mechanical seal life, motor overload, noisy pump operation, or simply not pumping enough fluid. Consequently, the pump system troubleshooter always starts with the same question: Where do I begin?

A pump troubleshooting guide might list 24 different causes of pump system problems (see "Pump Troubleshooting and Problems," p. 50). Other guides might have more than 40 different causes. Yet others will even list more than 90. But the first step in pump system diagnostics is always the same: determining exactly—or as close as possible, anyway—where a centrifugal pump is operating on its pump performance curve.

Designers may know how they want the pump to function. But what they really need to know is how the pump is functioning. Going through the initial step of determining this is tedious but will pay huge dividends. And the key to determining the operating performance of this equipment is the pump curve.

Pitching a curve

The pump curve typically includes six operating specifications:

  • Flow. Flow through a centrifugal pump is measured in gals. per min. (gpm).

  • Head. Pump head is expressed in ft. of liquid pumped and is usually determined experimentally by the pump manufacturer.

  • Horsepower. This is determined experimentally and is the power required for the pump to maintain a given flow rate at a given pump head.

  • Pump efficiency. This is calculated and expressed as a percentage of a pump's overall efficiency to pump the given flow rate at the given pump head.

  • NPSHR. This is net positive suction head required to pump a given flow rate at a given speed, in revolutions per minute (rpm) and with a given impeller diameter. This is determined experimentally and expressed in ft. of head. This is the minimum energy required at the pump impeller to ensure proper pump operation.

  • BEP. This is best efficiency point of the pump curve around which most pumps are selected.

Figure 1 (p. 50) shows a basic pump performance curve for a pump operating with a 17-in. impeller at 1,780 rpm. The typical curve plots pump flow capacity (in gpm) against the head (in feet). Some curves will show pump performance at a constant speed with a variety of impeller diameters, while others will show the pump performance with a fixed impeller diameter and variable rpm. To simplify the discussion, only the pump curve for a single impeller diameter and a constant speed is shown. Figure 1 illustrates a pump performance curve with horsepower curve, pump efficiency and NPSHR for various flows and heads.

Horsepower is calculated by measuring torque in the laboratory and converting to horsepower. From the flow and head curve, and the horsepower required, the pump's efficiency at various flows and heads is calculated and added to the overall curve. Finally, the NPSHR, in ft. of head required at various flow rates, is shown.

But it doesn't tell the troubleshooter where on the curve the pump will operate. It only indicates the pump's capabilities, not what the pump is actually doing.

The best source for a reliable pump curve is the pump manufacturer. But one must be sure to get the exact pump curve for the specific pump. Therefore, furnishing the pump manufacturer with the pump's serial number is crucial.

A general pump curve illustrating various impeller diameters or pump speeds is usually satisfactory. Verify certain basic details of the actual pump/driver combination to ensure that the set-up exactly matches the pump performance curve. A suggested check list would include the following:

  • Motor rpm

  • Pump rpm

  • Gearbox ratios

  • V-belt drive ratios

  • Impeller diameter

First of all, one must verify the motor speed using a tachometer or an rpm-measuring strobe gun. There are times when motors are mislabeled with the incorrect rpm rating. Verify gearbox ratios or belt-drive ratios in order to confirm the pump rotational speed.

Finally, verify the pump impeller diameter. This may be a bit difficult without disassembling the pump. But the more confident one is that the diameter is correct, the more reliable the trouble-shooting effort. Pump speed and impeller diameter are crucial for determining as close as possible whether the pump is operating on its pump curve.

System head curve

The pump, however, does not decide on its own where on the pump curve it will operate. The system to which it is connected also determines how the pump functions. So, one must figure out what the system will demand of the pump by determining the system's fluid characteristics and parameters at desired operating conditions, which include the following:

  • Desired gpm

  • Temperature

  • Specific gravity or density of fluid

  • Viscosity of fluid at operating temperature

  • Vapor pressure of fluid at operating temperature

The pumping system is made up of total head (also known as total dynamic head or total differential head) at any given flow rate. TDH is expressed in "feet of head" of the liquid being pumped. The TDH for a specific system is equal to the total discharge head (Hd) less the total suction head (Hs):

TDH = Total Hd %%MDASSML%% Total Hs

Total Hd is what the hydraulic system demands. Total hs is what the system will already furnish to the system's demands. This is basically an energy balance on each side of the pump. The pump will then make up the difference in energy required to meet system demands.

Total Hd is the sum of:

  • static discharge head, which is the height to which the fluid is pumped

  • all piping and friction losses on the discharge side, including pipe, valves, fittings, strainers, control valves, meters and heat exchanger losses

  • pressure, expressed in feet of liquid, required at the ultimate discharge point.

  • losses at sudden enlargements

  • existing losses

  • any other energy losses

Total Hs is the sum of:

  • static suction head, which is the liquid level height above pump suction

  • all piping and friction losses on the suction side including pipe, valves, fittings, strainers, control valves, meters, heat exchanger losses, etc.

  • pressure, expressed in feet of liquid, on the supply tank

  • suction side losses at sudden enlargements.

  • suction side existing losses

  • any other suction side energy losses

Note: All suction side losses—friction, energy or other—are expressed as a negative number in the S(hs) equation. A system head is determined for a given liquid flow rate through the system of piping, valves and fittings. Values for other system heads are determined for a variety of other flow rates through the system. The overall system curve can be drawn once sufficient pairs of flow rates and corresponding system heads are determined.

Figure 2 (p. 50) is the generic pump curve from Figure 1 but with a typical system curve. For this example, assume a "static head" difference of liquid levels on the system suction and on the system outlet of 50 ft. The rest of the system curve consists of "friction head".

The pump diagnostician must determine what method of friction loss calculations to use in order to estimate system demands on the pump. There are two choices: either the Hazen Williams formula or the Darcy-Weisback equation. The former produces reliable results when pumping a liquid with suspended solids. Most operating plants utilizing this empirically based formula have predetermined their acceptable "C" factor for various pipe types and slurry consistencies. The Darcy-Weisback equation, on the other hand, recognizes that pipe friction is dependent on roughness of the pipe's interior surface and the internal pipe diameter. Common friction loss tables using this equation are available in such industry resources as Cameron Hydraulic Data or the Standards of Hydraulics Institute. The theoretical system curve derived from friction loss tables or calculations using a "C" factor should be compared with actual system performance data if at all possible. Calculate the estimated discharge head converted to pressure, and compare that to the actual discharge pressure as measured by an accurate pressure gauge. One may find that even though calculations are correct, as one understands the system, yet the system may have fouled heat-exchanger pipe fittings, a blocked elbow, failed check valve, or some other factor that alters the actual system curve and is driving the pump to operate at some undesirable point on its pump curve.

Hydraulic systems

Let's turn now to how a pump functions with the particular hydraulic system in which it is installed. Like a train on its track, a pump can only operate on its pump curve, with the installed impeller diameter and at the speed it runs. To hitch a ride on the train, someone must decide where they want to get on—front, middle or rear.

Pumps and pumping systems work the same way. The pump doesn't decide where on the curve it operates. The system determines where the pump will hitch a ride on the curve. Figure 2 shows the system curve overlaid and intersecting with the pump curve. Unless the pump or system is modified, that intersection is where the pump will operate.

Let's assume the pump will produce 6,000 gpm at a TDH of 240 ft. Figure 2 speaks volumes as to what could potentially be causing problems. From here, one can make some preliminary determinations about the performance of the pump.

Of course, this is just the first step in troubleshooting pump ghosts. It doesn't mean that one can ignore other diagnostic efforts. For example, if there is premature bearing failure, one should definitely bring in the bearing supplier to analyze the bearings. And the same is true for mechanical seal failure and motor overloads. By all means, use all resources at your disposal.

But what happens if the generic system head curve is not exactly what one originally thought? What if a different system curve drives the pump too far to the left of BEP? This system will force the pump to operate at ~ 2,300 gpm, not the 6,000 gpm it was originally designed for. Depending on the type of pump, it could experience some of the following symptoms:

  • Radial shaft loads increase.

  • Internal recirculation.

  • Slurries settling.

  • Increased horsepower.

  • Excess thrust.

These are just a few of the problems that might occur when the system drives a pump too far to the left of its BEP. But what might happen if the head curve drives the pump too far to the right of BEP? Operating too far right of the pump's BEP could result in the following:

  • Radial shaft loads.

  • Cavitation.

  • Motor overload.

  • Discharge recirculation.

  • Rapid component wear.

Forcing the pump to pump much more than it is designed for can result in discharge recirculation causing increased internal pump component wear. Also, rapid wear can result from simply pumping too much fluid, especially abrasive slurry.

Once again, these are just a few unpleasant pump and system symptoms that might occur when the system drives a pump too far to the right of its BEP. But no matter what the problem, there are general trouble-shooting principles to follow.

All-purpose troubleshooting

It doesn't really matter what type of pump application, the diagnostic principles are the same. Whether it's an industrial process, HVAC or commercial plumbing pump system, the primary troubleshooting tool is to determine the exact pump performance curve and true system head curve. Then, unite the two curves and see where the pump is operating. In many cases, the troubleshooter will find that the "ghost" is a mismatched pump and system.

(For a look at some actual pump ghostbusting case histories, see "Pump Ghost Stories" by visiting www.csemag.com and clicking on the blue Plumbing button on the left.)





No comments
Consulting-Specifying Engineer's Product of the Year (POY) contest is the premier award for new products in the HVAC, fire, electrical, and...
Consulting-Specifying Engineer magazine is dedicated to encouraging and recognizing the most talented young individuals...
The MEP Giants program lists the top mechanical, electrical, plumbing, and fire protection engineering firms in the United States.
Combined heat and power; Assessing replacement of electrical systems; Energy codes and lighting; Salary Survey; Fan efficiency
Commissioning lighting control systems; 2016 Commissioning Giants; Design high-efficiency hot water systems for hospitals; Evaluating condensation and condensate
Solving HVAC challenges; Thermal comfort criteria; Liquid-immersion cooling; Specifying VRF systems; 2016 Product of the Year winners
Driving motor efficiency; Preventing Arc Flash in mission critical facilities; Integrating alternative power and existing electrical systems
Putting COPS into context; Designing medium-voltage electrical systems; Planning and designing resilient, efficient data centers; The nine steps of designing generator fuel systems
Designing generator systems; Using online commissioning tools; Selective coordination best practices
As brand protection manager for Eaton’s Electrical Sector, Tom Grace oversees counterfeit awareness...
Amara Rozgus is chief editor and content manager of Consulting-Specifier Engineer magazine.
IEEE power industry experts bring their combined experience in the electrical power industry...
Michael Heinsdorf, P.E., LEED AP, CDT is an Engineering Specification Writer at ARCOM MasterSpec.
click me