Reciprocating compressor tripping on oil failure after an outage
At a food processing plant, a 110 kW reciprocating compressor using R-404A and POE oil tripped on its oil-failure switch three times in one morning. The compressor had been shut down for a three-day maintenance outage while the oil was changed (55 L of the specified oil) and the oil filter replaced. The crankcase heater was switched off at the breaker for the outage, and the suction service valve was left open. The machine room is at 12 °C. On restart the compressor ran for about 95 seconds and tripped; the oil-failure switch has a 90-second delay. The crankcase sight glass is above the normal level and full of foam. A differential gauge on the oil pump reads 90 kPa; before the outage it read 220 kPa. Discharge and suction pressures are normal for the load, the oil is warm, and the oil cooler outlet is 35 °C. The maintenance foreman wants to fit a higher-capacity oil pump.
1. Define net oil pressure, explain how it is measured, and why 90 kPa is worse than it looks.
Read the worked answer
Net (or 'useful') oil pressure is the oil pump discharge pressure minus the pressure in the crankcase - for a suction-cooled crankcase, essentially the suction pressure. It is the pressure that actually forces oil through the bearings, so it is the quantity the oil-failure switch is protecting.
It must be measured as a differential: a differential gauge across the pump and the crankcase, or two gauges read at the same instant. A single gauge read against atmosphere is meaningless because suction pressure changes with load and the switch never sees it.
On many machines the switch is set around 100–140 kPa net with a 90–120 s time delay. A 90 kPa differential with a 90 s delay is already at or below the trip point, and the delay - which exists so the compressor can ride through a brief start-up dip - cannot cover a persistent condition. Three trips in one morning confirms that. Before drawing conclusions from the number, verify the gauge and the switch are calibrated; but the physical evidence in the sight glass says the reading is real.
2. Diagnose the cause. Why did this start after the outage, and how do you rule out the pump, strainer or cooler?
Read the worked answer
The timing is the strongest evidence. The outage combined three things: crankcase heater off, suction valve open, three days of time, and a 12 °C machine room. Refrigerant migrates to the coldest part of the system - the unheated crankcase - and condenses into the oil. POE absorbs R-404A readily. The level rises above normal (as the sight glass shows), and on start-up the fall in crankcase pressure flashes the dissolved refrigerant out of solution, which is the foam. Diluted, foamy oil has low viscosity, the pump loses its liquid seal and net pressure collapses.
Ruling out the alternatives:
- Oil pump or blocked strainer: would give low pressure with a normal level and no foaming, and usually a progressive fall rather than a trip on the first start.
- Blocked oil cooler: would give hot oil; the outlet is 35 °C with warm oil, so flow exists.
- Wrong oil: would show wax or sludge; the specified oil was used.
- Stuck oil-separator float: can overfill a crankcase, but would not produce immediate foaming on a cold start with the heater off.
Foaming plus elevated level plus the outage makes refrigerant dilution the primary cause.
3. What is the corrective sequence, and what should change in the maintenance procedure and records?
Read the worked answer
Stop resetting into the same trip. Isolate the compressor, pump it down and close the suction valve, then keep the crankcase heater energised and let the oil warm and the refrigerant boil off - typically a day - venting or pumping off per the manufacturer's procedure. Recheck level and foaming. Drain and replace the oil if it is diluted or smells of refrigerant, replace the oil filter and clean the strainer, verify the oil type against the nameplate, and confirm the oil-failure switch set point and delay against the manufacturer's data before restarting. On restart, watch net oil pressure and the sight glass through the first hour, then recheck. Fitting a larger oil pump would mask a dilution problem and is the wrong fix.
Procedure changes: crankcase heaters stay energised during outages - add it to the shutdown checklist; close the suction service valve or pump down for any outage longer than a few hours; record oil charge, oil type, filter change and outage duration in the maintenance record; and log the three trips, the diagnosis and the corrective action in the plant log.