This case is about a sudden failure of cooling tower fan motor of a copper mine.
The motor failed almost immediately after Planned maintenance, which was just about lubricating the motor bearings.
Electrical department conducted a scheduled PM task on this piece of equipment on 25.05.17. After 3 hrs of running; motor Non Driven End (NDE) bearing was damaged.
When the motor was opened it was observed:
1. One of the poles was severely damaged.
2. Bearing cage was also found damaged and all roller elements were crushed.
Why did this happen?
1. Sudden application of load or abrupt change in load. It happened when the machine was started after PM — i.e. starting the machine from rest under loaded condition.
2. This caused hunting of the motor in which a rotor starts seeking equilibrium position. Such an equilibrium is reached when the load torque is equal to the electromagnetic torque. This equilibrium position gets disturbed if a sudden change occurs in the load torque, which has been the case when the motor was started the after the motor was stopped to lubricate its bearings.
3. In this situation, the rotor slips too much, i.e. — the rotor moves around trying to find its steady state equilibrium state and in this process the rotor and stator touched and shorted — damaging one of the poles.
4. Point 1 to Point 3 describes the root cause of the case. A broad at the base 1N (1 times running speed) peak was observed. This indicates presence of rubbing and resonance.
5. Resonant frequency excited the resonant frequency of the NDE bearing which caused complete collapse (crushing) of the motor NDE bearing.
6. Another point which is important to consider is the time taken for a freshly lubricated bearing to stabilise. After lubrication, an anti-friction bearing generally runs hot (temperature greater than 75 degrees C but lesser than 95 degrees C) for a few hours (5 to 6 hours at times) to stabilise to a normal operating condition; with a temperature around 65 degrees C. This phenomenon can abruptly and adversely affect vibration levels of the bearing.
1. The vibration signature did not indicate lubrication starvation of the bearing.
Hence the question is — why stop a machine for re-lubrication when the activity isn’t needed at all?
2. In the future, if the system is stopped, then during start up it has to be ensured that the load is zero or near zero or it has to start at no-load condition.
If that isn’t possible, the system has to be started at low rpm and then the rpm can be gradually increased, all the while maintaining a steady state. It might take up to 6 hours for the system to stabilise after a bearing is lubricated.