Sidebands arround the BPFI and BPFO

Remember that the vibration you measure is a combination of all the forces acting on the machine.

First, as the damaged rolling element rolls in and out of the load zone, the force of each impact rises and falls (greater in the load zone, minimal at the top of the horizontally-oriented bearing). Any time a vibration amplitude (vibration at the ball spin frequency in this case) rises and falls periodically you get sidebands. In this case, sidebands of the cage frequency (fundamental train frequency).

Second, as the damaged area on the inner race moves in and out of the load zone, again, the force of the impact rises and falls. So now the vibration at the ball-pass-inner-race frequency will have sidebands of turning speed - because it goes in and out of the load zone once per revolution.

The ball-pass-outer-race vibration does not have sidebands because it lives in the load zone (normally), so all its impacts are the same.

Now, if there is beating, or any other vibration, the waveform and spectrum will include this vibration (beating is just the interaction between two other periodic forces). If that force does not affect the force of impact that I just described, then the bearing vibration amplitudes and patterns won't change.

Have you been watching my Cat III prep videos?


Just to add to Jason's answer.

Sidebands around the outer race are theoretically possible in unusual circumstances. If the inner race is static and the outer race rotates (think wheel bearing), then any damage on the outer race would pass in and out of the load zone, and generate sidebands at 1x RPM.

I've never seen it myself, but I've been told that a severe unbalance could give a rotating load zone, which means that even with a static outer race the impacts on the defect could change in amplitude as the load zone rotates. Again Sidebands of 1x RPM.

Add Reply