Dependable Centrifugal Pump Operation

By Allan R. Budris

I recently received a poignant question from a reader on a subject near and dear to my heart. He wanted to know whether a low pump Net Positive Suction Head Required (NPSHR) was a good thing -- or a bad thing -- for pump reliability.

On the one hand, the Hydraulic Institute recommends higher NPSH margins (NPSHA/NPSHR) for improved reliability, which means either a higher system NPSH availability or a lower pump NPSH requirement. On the other hand, many pump users and system consultants specify a low suction specific speed (RPM x GPM1/2/NPSHR3/4), which dictates higher NPSHR values. Given these apparent conflicting guidelines, my reader's confusion is understandable.

The simple answer is that it depends -- largely on the pump Suction Energy Ratio (SER). To further clarify this response, I offer the following considerations.

Argument for Low Suction Specific Speed Values (High NPSHR)

Many years ago, a study was published based on the evaluation of a number of installed large split-case pumps, which showed a correlation between higher suction specific speed values and lower pump reliability, especially for specific speed values above 11,000. This study is the basis for many of the industry limits and penalties on high suction specific speed pumps.

However, a similar, less circulated study concerning suction specific speed and pump reliability was published a few years later that focused on smaller end-suction pumps. In contrast to the larger split-case study, this report failed to illustrate a correlation between suction specific speed and pump reliability.

This conflict was one of the reasons I developed the concept of Suction Energy (SE).

Argument for High NPSH Margin

In addition to the recommendations of the Hydraulic Institute for higher NPSH margins, I evaluated over 100 split-case and end-suction pumps and discovered a definite trend of improved pump reliability with increased NPSH margin values (see Fig. 1).

Suction Energy

After extensive investigation, I found that I could predict when a pump will or will not be damaged by pump cavitation by adding the impeller inlet tip speed (diameter x pump speed), plus the liquid specific gravity, to the pump specific speed and then comparing the resulting SE values with gating values for various pump types. Based on data from over 100 axial split-case and end-suction pump installations, pump SE was confirmed to be an effective predictor of pump reliability (see Fig. 2).

In order to determine the Suction Energy Ratio (SE/start of high SE), I developed SE "Gating" and "Very High" values (based on turns and turbulence in the pump inlet passage) for five different pump types (see Table 1). Low SE is any value below the High SE/Gating value. For a more detailed explanation of SE (and NPSH margin), see my http://www.waterworld.com/articles/print/volume-28/issue-12/departments/pump-tips-techniques/key-consider-help-deter-opti-npsh-margin-centri-pump-apps.html Dec. 2012 and http://www.waterworld.com/articles/print/volume-29/issue-1/departments/pump-tips-techniques/key-consider-help-determine-optimum-npsh-2.html Jan. 2013 WW columns.

SE is the product of four key factors:

De x N x S x S.G.

Where:
De = Impeller eye diameter (inches)
N = Pump speed (rpm)
S = Suction specific speed (rpm x (gpm)0.5)/(NPSHR)0.75
S.G. = Specific gravity of liquid pumped

Based on this SE concept, low pump NPSHR values can be beneficial when the pump has low SE, although high NPSH margins are less important with low SE. On the other hand, with high, or especially very high, SE pumps, low pump NPSHR can be detrimental, but higher NPSH margins can be critical for these pumps.

NPSHR

NPSHR is the absolute suction pressure, above the liquid vapor pressure, that will allow a level of cavitation in the impeller, reducing the developed head by 3 percent. One of the key factors impacting the pump NPSHR is the liquid inlet velocity (the slower the inlet velocity, the lower the NPSHR).

There are several ways to reduce the pump inlet velocity/NPSHR. The first method is to increase the impeller eye diameter (which is not included in the formula for suction specific speed). Low NPSH impellers typically have larger eye diameters. However, increasing the impeller eye will also increase the pump SE, which can be detrimental if it moves the pump to high or very high SE.

Reducing the pump flow rate will also reduce the inlet velocity and NPSHR, but it will also probably reduce the pump efficiency and could push the pump into suction recirculation. If available, an inducer (an axial flow impeller) could also reduce the NPSHR without increasing the impeller eye, plus it has a higher Start of High SE gating value. Further, changing the pump type can also increase the Start of High SE by increasing the gating value (see Table 1), which would result in a lower SER.

Finally, the best way to reduce the pump NPSHR is typically to reduce the pump speed, as with a variable speed drive (VSD). This not only reduces the pump NPSHR but it also reduces the pump SE. The pump NPSHR changes by a factor of 1.7 to 2.0 with respect to changes in the pump speed.

Suction Energy Ratio vs. NPSH Margin Ratio vs. Reliability

Based on a review of field repair records from two industrial plants (with more than 100 end-suction and split-case pumps, as referenced above), I was able to combine the separate reliability impacts of the pump SER with the NPSH margin ratio (NPSHA/NPSHR) to demonstrate their combined impact on pump reliability. The results are shown in Figure 3, with a reliability factor of 1.0 equal to a four-year "Mean-Time-Between-Failure" rate.

This reliability represents not only the cavitation erosion of the impeller but also the reduced life of the pump bearings and mechanical seal from cavitation initiated vibration.

In a separate study, laboratory NPSH Margin Ratios vs. Pressure Pulsation results were coupled with more recent impeller cavitation damage data to develop an additional SER vs. NPSH Margin Ratio Reliability graph (see Fig. 4). This was established for controlling very high SE pumps, handling warm water and operating in the allowable operating region. This graph is primarily an indicator of the cavitation erosion life of the impeller.

These two reliability plots can be used as guides for pump and system changes to reduce or eliminate pump damage from cavitation in order to maximize mean time between failures.

Conclusion

NPSHR and pump reliability are complex issues, but determining the SER is an ideal first step towards evaluating the situation. If the pump has high SE, the NPSHR should not be lowered by increasing the impeller eye diameter. However, if the NPSHR can be lowered by a means that does not increase the impeller eye diameter, such as by slowing the pump speed with a VSD, it can become a win-win situation with improved pump reliability.

About the Author: Allan R. Budris, P.E., is an independent consulting engineer who specializes in training, failure analysis, troubleshooting, reliability, efficiency audits, and litigation support on pumps and pumping systems. He can be contacted via email at [email protected].

More WaterWorld Current Issue Articles
More WaterWorld Archives Issue Articles