When an Unusual Noise Is the Only Warning You Get
At the McLellan Pumping Station in Englewood, Colorado, a mechanic reported that one of the station’s pumps was making unusual noises. The pump—part of a system that moves between 1 million and 13 million gallons per day from a reservoir—was old but critical. The engineering team’s response provides a model for how to investigate pump noise complaints systematically.
The Investigation: What the Noise Told Them
Rather than dismissing the noise as “an old pump sounding old” or immediately pulling the pump for a rebuild, the team followed a structured diagnostic process:
- Characterize the noise: Was it a sharp crackle (cavitation), a low rumble (bearing or misalignment), a rhythmic thrum (vane-pass hydraulic pulsation), or a high-pitched squeal (dry-running seal)? Each sound points to a different root cause. In this case, the noise suggested cavitation.
- Check the operating point: Measure flow, suction pressure, and discharge pressure. Compare the actual operating point against the pump curve. Was the pump operating within its POR, or had system changes pushed it into an unfavorable region?
- Verify NPSH margin: Calculate NPSHa using the measured suction pressure, fluid temperature, and barometric pressure. Compare to the pump’s NPSHr at the measured flow. If NPSHa is marginal, cavitation is the likely explanation for the noise.
- Check for system changes: Had the reservoir level changed? Had a strainer become partially clogged? Had a new valve been installed that changed the system curve? Many pump noise problems originate not in the pump itself but in changes to the system it serves.
The Diagnosis and the Fix
The investigation revealed that the pump was operating at a flow point where NPSHa was marginal—the reservoir level was lower than normal, reducing the suction head available. The pump was cavitating. Rather than replacing the pump (unnecessary) or installing a larger suction pipe (impractical), the team addressed the root cause: they adjusted the pump’s operating point to stay within the range where adequate NPSH margin existed, by modifying the control strategy to limit maximum flow during low-reservoir conditions.
The Lesson: Old Pumps Deserve Investigation, Not Assumptions
When an older pump develops a new noise, two responses are common—and both are wrong. The first is to ignore it (“it’s always sounded like that”), which allows a developing problem to progress to failure. The second is to immediately overhaul the pump without diagnosing the root cause, which wastes money and may not fix the actual problem if the root cause is in the system, not the pump.
The correct response: investigate the noise as a symptom, diagnose the root cause using the measured operating data, and fix the actual problem—whether it is in the pump, the piping, the controls, or the system conditions.
Key Takeaways
- Pump noise is a diagnostic signal, not an annoyance. Categorize the noise type (crackle, rumble, thrum, squeal) to narrow the root cause.
- Check the operating point and NPSH margin before pulling the pump. Many noise problems originate in system conditions, not pump defects.
- System changes—reservoir level, strainer condition, valve position, parallel pump operation—are common triggers for new pump noises.
- Investigating a noise complaint costs a fraction of an unnecessary overhaul and is far more likely to fix the actual problem.