Centrifugal pump troubleshooting guide
Why Is My Pump Off Curve and Cavitating?
Reader question: “How are we so far off the curve and why are we cavitating”
Low flow, a low discharge-pressure reading and pump noise do not prove one cause. Reconstruct the operating point, calculate NPSH available and separate vapor cavitation from recirculation, gas and measurement error.
Diagnostic sequence
Six checks turn pump noise into evidence
Do not change the pump until the measured operating point and suction condition agree with the applicable curve.
Identify
Size, impeller, speed, rotation and curve revision.
Measure
Flow plus synchronized suction and discharge pressure.
Plot
Pump curve, system curve and control-valve loss.
Calculate
Absolute suction head minus vapor-pressure head.
Differentiate
Recirculation, gas, blockage, wear and instrumentation.
Correct
Act on the measured root cause and verify the result.
Short answer: the pump may be operating far from its intended point, but the reported numbers do not yet prove cavitation. Verify the correct pump curve, reconstruct the actual differential head, and calculate NPSH available using the glycol mixture's real vapor pressure at 127°C.
In the original Reddit question published on July 4, 2026, the reported best efficiency point was 600 GPM, while measured flow was approximately 300 GPM.
That is 50% of the stated BEP flow. It warrants checking the manufacturer's preferred and allowable operating regions, but it does not by itself prove that vapor cavitation is occurring. This guide analyzes the published information; it is not a remote diagnosis of the installation.
What the Reported Measurements Prove
| Observation | What it indicates | What it does not prove |
|---|---|---|
| 300 GPM versus a 600 GPM BEP | Operation far to the low-flow side of the stated BEP | The cause of the low flow |
| PCV maintains 50 psig | Pressure is controlled at one measurement point | Pump differential head |
| Curve shows 180 psig near BEP | A performance value exists on the selected curve | That it is directly comparable with one discharge gauge |
| Glycol is at 127°C | Vapor pressure and mixture properties are critical | NPSHA without glycol type and concentration |
| Pump is noisy | Hydraulic instability is possible | Cavitation rather than recirculation, gas or a mechanical fault |
The Hydraulic Institute defines BEP as the point of maximum efficiency for a specified speed and impeller diameter. Operation away from BEP increases the potential for recirculation, vibration and hydraulic loading, but the permitted range is pump-specific. See Hydraulic Institute, December 7, 2022.
Why 50 psig Is Not the Pump Head
A discharge pressure reading alone is not pump head. Synchronized suction and discharge measurements are required. The pump head is reconstructed as:
Hpump = (Pd - Ps)/(ρg) + (vd² - vs²)/(2g) + (zd - zs)
The calculation must use the actual liquid density and account for velocity and gauge-elevation differences. If the suction is under positive pressure, a discharge-only reading omits part of the energy balance. If the suction is under vacuum, the same shortcut creates a different error.
Cavitation and Low-Flow Recirculation Are Not the Same Problem
True vapor cavitation occurs when local absolute pressure falls below the liquid's vapor pressure. NPSH available at the pump suction can be expressed as:
NPSHA = Ps,absolute/(ρg) + vs²/(2g) - Pv/(ρg)
NPSHA is a system property. NPSHR is supplied for a particular pump, speed and flow. Adequate margin must be applied rather than treating the published NPSH3 point as a guaranteed damage-free operating condition. See Hydraulic Institute, March 18, 2025.
At 127°C, do not substitute water data or pure-glycol data unless the circulating liquid is actually that substance. The glycol type, concentration and contaminants determine the property data.
Low-flow operation can also create inlet or discharge recirculation while NPSHA appears adequate. It may produce noise, pressure fluctuation and vibration that operators describe as cavitation.
Diagnose the Problem in This Order
- Confirm the pump and curve. Record the manufacturer, model, size, impeller diameter, operating speed, rotation, curve revision and curve units.
- Verify the flow measurement. Check calibration, units, straight-run requirements, pipe-full condition and possible entrained gas.
- Measure suction and discharge pressure together. Use calibrated instruments near the nozzles and record their elevations.
- Calculate differential head. Correct for density, velocity and measurement elevation before plotting the point.
- Reconstruct the system curve. Include tank pressure or elevation, pipe friction, exchangers, strainers, fittings and the control valve.
- Calculate NPSHA. Use absolute suction pressure and the correct vapor pressure at the operating temperature.
- Compare the full operating range. Review minimum, normal and maximum cases against BEP, POR, AOR and NPSHR.
- Check alternative causes. Inspect for blockage, gas entrainment, wrong speed, wear, recirculation and mechanical defects.
How to Separate the Possible Failure Modes
| Possible cause | Evidence to collect |
|---|---|
| Insufficient NPSH margin | Absolute suction pressure, vapor pressure, suction loss and performance as suction pressure changes |
| Low-flow recirculation | Position relative to the manufacturer POR/AOR and vibration or pressure behavior at low flow |
| Suction blockage | Pressure loss across strainers, valve position, internal lining condition and deposits |
| Gas entrainment | Source vortexing, air leaks, bubbles and unstable flow indication |
| Wrong impeller or speed | Nameplate, tachometer reading, impeller measurement and correct curve revision |
| Control-valve restriction | Valve pressure drop and controller output at the operating point |
| Wear or damage | Internal clearances, impeller condition and a controlled performance test |
A 2017 experimental study found measurable vibration changes before the conventional three-percent head-drop cavitation point. Noise or vibration should therefore be treated as diagnostic evidence rather than proof of one failure mode. See Lu et al., April 18, 2017.
What Should Be Changed?
Do not change the pump until the cause is established. Depending on the evidence, corrective work may involve removing a suction obstruction, correcting air entry, increasing source pressure or level, reducing suction losses, revising the control strategy, changing speed or impeller diameter, or selecting a pump whose required operating range is closer to BEP.
Never throttle the suction valve as a routine method of moving a centrifugal pump's operating point.
Where an ANSI Chemical Process Pump Fits
For a characterized process liquid, YSM can review the duty against the G196 or D Mark III ANSI process-pump families. A model should only be proposed after confirming the operating range, differential head, glycol properties, temperature, suction conditions, NPSH margin, material and seal requirements.
ASME B73.1 provides the dimensional and design framework for horizontal end-suction chemical process pumps. Compliance with the standard does not prove that a particular size fits this duty. See ASME B73.1-2026.
Sources and Dates
- Original Reddit question — July 4, 2026
- Hydraulic Institute: NPSH and operating regions — December 7, 2022
- Hydraulic Institute: NPSH-margin guideline update — March 18, 2025
- U.S. DOE, Improving Pumping System Performance — May 2006
- Lu et al., cavitation experiment — April 18, 2017
- ASME B73.1 — 2026 edition
Related Technical Guides
System Curve, BEP and Low-Flow Operation
Review how the pump curve and system curve establish the actual operating point.
Read the operating-point guidePump Selection Data Beyond Flow and Head
Prepare the liquid, suction, operating-range and construction evidence required for selection.
Open the selection checklistNeed an Application-Specific Review?
Send the pump identity, selected curve, minimum/normal/maximum flow, synchronized suction and discharge readings, glycol data, suction layout and control-valve information.