I've noticed a strange phenomenon about 4 or 5 times over the last couple of years and it's always been in fixed speed AC motors.

Typically it starts off with a high spike energy reading in one of the motor bearings. The corresponding demodulated spectrum has a rise in the low frequency end, but no bearing defect frequencies are present. Very often I get these symptoms with a lubrication issue. Usually I'll follow this up by listening to the bearing with an ultrasonic gun. Surprisingly the bearing will usually sound very smooth, but after a period of time a crackling noise will slowly grow and start to sound like a defective bearing. After a short while the noise recedes and eventually disappears.

I've got a couple of theories, but neither seems to fit all the facts.

1. There is a small defect in the outer race of the brg, but due to excessive end float, the shaft is shifting axially and the defect is only being rolled part of the time.

2. There is an outer race defect and the outer race is slowly creeping, so that the defect is moving in and out of the load zone, hence the change in sound from smooth to rough.

Has anyone else experienced this and do you have any other theories as to what may be happening? Feel free to shoot my theories down in flames.

I have heard similar noises from motor bearings also with corresponding raised gSE (HFD) overalls. Usually, adding grease will deaden the noise very quickly. Where I look for defects is in the acceleration spectrum with an fmax of at least 3KhZ for 1800rpm 6KhZ for 3600 rpm. There will usually be rotor bar pass with running speed or 2X line frequency in the higher frequency range but it is also where I have found bearing defects to appear. I'm not exactly sure why they don't always show up in the gSE spectrum but never the less, I have successfully found bearing problems using this method. I think the noise is generated by the cage when there is insufficient lube. Your theory about the creeping outer race is valid in my opinion but it would be a rare event.

I think as the bearing heats up a little, some grease will loosen up and fall into the bearing until it gets flung out and the cycle starts over. How about that for a theory?

Al, the bearing maybe over lubed, with roller spin speed slowing or stopping while out of the load zone and skating back up to speed once back into it generating high freq vibration that’s not necessarily repetitive enough to isolate a fault ,
creating friction heat thins the grease it separates and the oil can sometimes be seen on the outside of the housing, the bearing sounds smooth at first from fresh lube but some oil quickly separates and escapes the bearing, the remaining lube cake dries a little and the rollers still skating really start to squeal eventually the bearing cavity loosens up and rollers spin normally, I check the nipple before taking my reading to get an idea of how fresh or old the blob of grease the oiler left behind is and rough estimate the time since last relube, I find motors that have a lube schedule are more often over lubed than under.

Captnb