Debutanizer Bottoms Pump Cavitation After a Rate Increase
A debutanizer bottoms pump runs at 45 m³/h, taking 150°C liquid at 610 kg/m³ from a column bottoms drum to the feed/product exchanger. The drum pressure is 11.10 bara and the liquid vapor pressure at 150°C is 10.85 bara. At the normal 60% level, the liquid surface is 5.0 m above the pump centerline. The 6-inch suction line has a calculated friction loss of 2.2 m at 45 m³/h and 2.9 m at 52 m³/h. The pump NPSHr is 6.5 m at 45 m³/h and 7.2 m at 52 m³/h.
To increase debutanizer throughput, operations raised the rate to 52 m³/h and lowered the drum level to 50%, cutting the static head to 4.2 m. Within a week the pump began making a rumbling noise, the flow indicator swung ±15%, differential pressure fluctuated, and the mechanical seal failed after three weeks. The spare pump shows the same symptoms. The seal flush is product taken from the pump discharge, and the suction strainer has not been cleaned for six months.
1. Calculate NPSHa at the original and current conditions and decide whether cavitation explains the symptoms. State what additional data could change the conclusion.
Read the worked answer
NPSHa = (P_drum − P_v)/(ρg) + z − h_f. The pressure term is (11.10 − 10.85) × 10^5 = 25,000 Pa; ρg = 610 × 9.81 = 5,984 N/m³, so the pressure head is 25,000/5,984 = 4.18 m.
Original: 4.18 + 5.0 − 2.2 = 6.98 m; NPSHr = 6.5 m, margin 0.48 m. Current: 4.18 + 4.2 − 2.9 = 5.48 m; NPSHr = 7.2 m, shortfall 1.7 m. Cavitation is expected at the current condition. Note that even at 45 m³/h and 50% level, NPSHa = 6.18 m, below 6.5 m, so the level reduction alone would have made operation marginal. The rate increase added both friction and a higher NPSHr.
Additional data that could change the conclusion: actual vapor pressure at the measured temperature, dissolved light ends or entrained vapor that raise P_v, suction strainer fouling, and whether the gauge pressure is measured at the pump flange. A high-speed pressure transducer at the suction would confirm vapor collapse.
2. The operations superintendent proposes raising column pressure by 0.3 bar to suppress the cavitation. Evaluate this proposal and give two better immediate actions.
Read the worked answer
Raising pressure by 0.3 bar adds 0.3 × 10^5/5,984 = 5.0 m of NPSHa, which would restore a large margin. But it also raises the bottoms temperature and reboiler duty, reduces relative volatility, and may exceed condenser or reboiler capacity; the column may already be pressure-limited. It also increases the relief load and the seal temperature. It is acceptable only as a temporary move after checking the column hydraulic rating, the PSV set pressure and the relief header.
Better immediate actions: restore the drum level to 60% (+0.8 m) and reduce rate to 45 m³/h, which lowers friction to 2.2 m and NPSHr to 6.5 m, giving 6.98 m. Also clean the suction strainer and vent the pump casing; confirm the seal flush is not vaporizing. If the level cannot be restored, lowering the pump by 2 m or increasing the suction line size would add margin.
3. Recommend a long-term fix and quantify the NPSHa improvement.
Read the worked answer
Lower the pump foundation by 2.0 m (+2.0 m), replace the 6-inch suction line with 8-inch to cut friction from 2.9 to about 0.9 m at 52 m³/h (+2.0 m), and install a low-NPSH inducer or a pump with NPSHr ≤4 m. Keep the drum at 60% level. The new NPSHa is approximately 4.18 + 6.2 − 0.9 = 9.5 m, giving more than 2 m margin over a 7.2 m NPSHr.
Add a low-low level trip and a suction pressure alarm, and include the suction strainer in the preventive maintenance schedule. Until the modifications are complete, operate at 45 m³/h and 60% level. If a temporary rate increase is essential, raising column pressure by 0.3 bar is safer than running the pump dry, but it must be evaluated against the column and relief system limits.