Show me a cavitating pump and I'll show you an energy hog.
Most operating facilities in today's marketplace are aware of the effects that cavitation has on mechanical pump reliability. Reduced rotor stability, shorter bearing life and the ever-popular premature mechanical seal failure are just a few of the more common manifestations. If, however, we were to look at the total cost of ownership for a cavitating pump, including its reduced efficiency and subsequent higher utility costs, we would see that this daily operating expense mounts up to a huge waste of both energy and money.
Unraveling the issue
"Best in Class" companies evaluate the purchase of a pump based on Total Life Cycle Cost (TLCC). Pump efficiency will be one of the variables that weigh into this calculation. A specified margin for Net Positive Suction Head Required (NPSHR) and a range for Suction Specific Speed will be specified in their engineering guides. Again, these pump characteristics weigh into the purchasing decision, but can ultimately be overridden during the project development cycle due to delivery schedules and initial purchase price. The TLCC philosophy applies to new pumps being purchased today.
What about the large population of pumps in service today that are 20, 30 or even 40-plus years old. TLCC and pump reliability were not even on the map when these units were purchased and commissioned. Their inefficiencies and diminished reliability are further aggravated by being operated at off-design conditions resulting from process demand changes that occurred after the pump was installed.
Further confusion is added by the term NPSHR. Keep in mind that the design goal of a pump manufacturer is to design a pump that meets the broadest range of operating conditions possible rather than designing a pump to meet your specific hydraulic needs.
A manufacturer's certified performance curve will list the NPSHR for the pump. This curve is not the point at which incipient cavitation occurs in the pump. Rather, it is the point at which cavitation is significant enough that the pump head is reduced by 3%. This is determined by testing the performance of the pump with the suction fully fl ooded. The pump is later retested at known fl ow rates and the suction valve is pinched off. The NPSHR curve is then plotted once the head meets a 3% reduction at the target fl ows. This is accepted in industry because the 3% condition is typically repeatable independent of process conditions (fl uid, temperature, etc.). Consequently, many pumps in service today are being operated within the prescribed NPSHR margin, yet cavitation still exists-as evidenced by the damage found on their impellers during a pump repair.
Reliability teams fight to keep the pump available, but rarely get the opportunity to affect real change, since the cost of a design modification is thought to be too high. Pumps are pulled for maintenance. Cavitation is evident, as seen in Fig. 1. The affected area of the impeller is weld-repaired or, occasionally, the impeller is replaced with a more cavitationresistant metallurgy. The pump is reassembled and placed back in service. If a design change is considered, it usually is dismissed due to price-without the energy savings ever having been considered.
This is cavitation
The published or known efficiency of the pump includes the hydraulic inefficiencies that are sufficient to cause this kind of mechanical damage to the impeller. As the fl uid being pumped drops below the fl uid's vapor pressure, it rapidly fl ashes from a liquid to a gas and back to a liquid. This is cavitation. The subsequent shock waves carry enough energy to literally rip a minute piece of metal from the impeller vane. Over the course of operating, these minute pieces of removed metal compound upon each other, leading to the damage shown in Fig. 1. Additionally, the vibration associated with these shock waves is transmitted down the shaft and its cumulative effect wipes out the mechanical seals and bearings. This is well known and discussed. One common solution is to install larger diameter or stiffer shafts with bigger bearings to try to extend the mean time between repairs (MTBR). API-610 has taken this approach in the last few revisions, which places a greater emphasis on the L/D ratios and other shaft stiffness design criteria.
Dealing with hydraulic inefficiencies
What we often fail to recognize is that hydraulic inefficiency from cavitation is costing us horsepower (HP) every time the pump is placed in service. In other words, pump users often are literally paying to tear up their equipment. With the availability of Computational Fluid Mechanics (CFM) and Computation Fluid Dynamics (CFD), the existing inherent inefficiency in a pump's hydraulic development can be reduced-and in many cases eliminated.
CFM and CFD allows a qualified individual to evaluate the suction characteristics of the impeller before any manufacturing takes place. Adjustments can be made to the inlet eye diameter and/or the inlet vane angles that can dramatically improve these characteristics. Multiple modeling runs can be examined to optimize the impeller geometry around your specific desired hydraulic condition.
There are many different ways to calculate the annual savings, but for this discussion we will use the following equation:
Assigning some values to the above variables, we can use a typical pump efficiency of 69% and assume a modest 4% efficiency increase. Let's say that we have a 200 HP motor with a rate load of 175 HP. Using a unit availability of 96% will give us 8,410 hours of operation. From the Energy Information Administration [Ref. 1], we find that in October 2006, the average retail price of electricity for an industrial user in the United States was 6.12¢ per kilowatt hour. Thus, the annual utility savings would be $6,098. By itself, for a single pump, that's a nice piece of change. Think, though, what this type of savings could add up to for operations with multiple pumps.
What to do with about your hogs
If you have a cavitating pump, don't just upgrade the metallurgy, stiffen the shaft and move on. Instead, eliminate or minimize the cavitation by redesigning the suction characteristics of the impeller. The savings detailed in this article are strictly a reduction in the cost of plant utilities for one pump. They do not take into account the overhead to maintain additional HP consumption. If the impeller needs to be replaced because of cavitation damage, then that cost should be removed from the incremental cost of a design modification. Once you couple in the increased MTBR and reduction in the maintenance budget, the total savings often make design changes practical.
Richard E. Martinez is vice president of operations with Standard Alloys, in Port Arthur, TX, a company he joined in 1989 as director of engineering, following several years working with the Lower Colorado River Authority (LCRA). Under his direction, Standard Alloys developed the capacity to perform custom design of impellers, volutes/diffusers and return guide vanes. Promoted to his current position in 2006, he now is responsible for operation of Standard Alloys Engineering, Pattern Shop, Foundry and Machine Shop/Repair Center. Martinez, who holds a B.S.M.E. from Lamar University, has published a number of articles related to pump performance, modifications and enhancements. For more information, telephone: (800) 231-8240 x 312; e-mail: richardm@standardalloys. com; Internet: www.standardalloys.com