Boiler Feed Pump Cavitation with No Standby Available
At a 2×210 MW coal-fired station, Unit 2 is at 180 MW. One motor-driven boiler feed pump is running; the standby pump is out for overhaul. The running pump shows discharge pressure 158 kg/cm2, flow 240 t/h, vibration 6.8 mm/s against an alarm of 7.5 mm/s, a low-frequency growling noise, and suction strainer differential pressure 0.35 bar against 0.12 bar after cleaning. Deaerator pressure is 6.5 bar absolute, water temperature 158°C, deaerator level 65%, and the pump suction static head is 12 m. Suction-line friction loss is estimated at 4 m at current flow, and the pump NPSH required is 9 m. The unit can be reduced to 150 MW without grid objection, but a trip would cost a restart and delayed supply. The control room asks the JE whether to continue at 180 MW, reduce load, or trip.
1. Calculate the approximate NPSH available and decide whether the pump is cavitating. Show the calculation and state your immediate operating decision.
Read the worked answer
At 158°C, the saturation pressure of water is about 6.18 bar absolute. The deaerator pressure is 6.5 bar absolute, so the pressure margin is 0.32 bar.
Pressure head = 0.32 × 100 kPa / (1000 × 9.81) ≈ 3.3 m. NPSH available = pressure head + static head - friction loss = 3.3 + 12 - 4 = 11.3 m. NPSH margin = 11.3 - 9 = 2.3 m.
A margin of only about 2.3 m is low, and the abnormal strainer differential is significant: 0.35 bar instead of 0.12 bar is an extra 0.23 bar, equivalent to about 2.4 m of head. The growling noise, vibration at 6.8 mm/s and low margin together point to cavitation or incipient cavitation.
My immediate decision is to reduce unit load to 150 MW, not to continue at 180 MW and not to trip immediately if the pump remains stable. Reducing load lowers feed flow, lowers NPSH required and reduces strainer pressure drop. I would trip if vibration exceeds 7.5 mm/s, if flow or discharge pressure oscillates, or if the noise becomes severe.
2. Identify likely root causes and the sequence of checks you would authorise, given that no standby pump is available.
Read the worked answer
The most likely root cause is a choking suction strainer, because the differential pressure is nearly three times the clean value. Other contributors could be low deaerator pressure or level, air ingress through a gland or vent, a partly closed suction valve, high feedwater temperature, recirculation-valve passing, or internal wear-ring clearance.
With no standby pump, I would avoid opening the pump or strainer immediately. The sequence should be:
- Confirm the suction valve is fully open and the deaerator pressure and level are stable.
- Trend strainer differential, pump flow, discharge pressure, vibration and noise every 15 minutes.
- Check the recirculation valve and minimum-flow line for unintended passing.
- Check for air ingress by venting and inspecting gland sealing.
- Review vibration spectrum for cavitation and bearing defects.
- Prepare the standby pump and a short controlled shutdown for strainer cleaning only after load is reduced and the unit is stable.
If the strainer differential continues to rise, the risk of losing the only running pump becomes unacceptable, and a controlled unit shutdown is safer than a forced trip.
3. Explain the trade-off between continuing at 180 MW, reducing to 150 MW and tripping. What monitoring would justify continuing?
Read the worked answer
Continuing at 180 MW preserves generation but keeps the pump at the highest risk point: flow and NPSH required are high, strainer differential is already abnormal, and there is no standby. A pump seizure or impeller failure could damage the pump and force a sudden unit trip, with a longer outage than a controlled reduction.
Reducing to 150 MW is the best compromise. It reduces feed flow and suction losses, improves NPSH margin and buys time to prepare the standby pump or a planned strainer cleaning. The cost is about 30 MW of generation, which is preferable to a multi-day forced outage.
Tripping is the safest action for the pump but is not automatically the best action for the plant if the pump is still stable at reduced load. I would continue only while all of the following hold: vibration below 7.5 mm/s and not rising, strainer differential stable below about 0.5 bar, no flow or pressure oscillation, deaerator level above 50%, and no abnormal bearing temperature. Any worsening trend would justify a controlled shutdown rather than waiting for a trip.