How to diagnose Cracked Inner Ring and Bearing Looseness over shaft

It is an uncommon fault that bearing inner ring is cracked axially. However bearing getting loose over shaft is not uncommon. In this article I will address the issue and share a case study in which there two problem were simultaneously diagnosed. As we know that typically when we talk about bearing loose over shaft , what comes in our mind, yes we comes up in our mind with 3X peak in FFT spectrum. But keep in mind as per my experience 3X peak only and only comes when bearing is loose over shaft and secondly

that looseness is that much considerable that bearing inner ring starts rotating over shaft as well. That rotation can be intermittent but very considerable. Unless and until there is no considerable repetitive sliding movement between shaft and inner ring there will be no 3X peak in FFT Spectrum.

Here in this case, in the routine vibration analysis survey, it was observed that there is some abnormal sound in Operator End (Non-Drive End) bearing of Size Press Bottom Roll at a paper machine.The overall vibration levels were very low 0.469 mm/s RMS. How G's level were increased up to 1.73 G's which in past remains below 0.3 G's.

When we looked at the Acceleration Time waveform we found that there is severe impacts at a period of 180 CPM which was the RPM of the roller. So it was a conclusion that 1X severe impacts are present.Moreover there was high frequency impacts superimposed in Time waveform. When I looked up at the frequency of those high frequency impacts by zooming out the time waveform the frequency of impacts was 1960 CPM which was clearly matching the Ball Pass frequency Inner Ring BPFI of the spherical roller bearing # 23240 CCK W33. In FFT Spectrum there was BPFI harmonics but 1X BPFI was dominant and each harmonics of BPFI was surrounded by family of 1X side bands which indicated that there is no pit or flaking but a cracked at inner ring raceways.

Whenever there will be flaking or pits at inner ring there will be BPFI harmonics and when fault gets severe BPFI harmonics are surrounded by 1X sidebands. But please note it that family of 1X side bands always come when there is a sharp crack on the inner ring. If we would have allowed this bearing to run further. The sharpness of the crack would have been vanished and there family of side bands would also have been vanished.

So whenever there will be crack in the inner ring of bearing there will be family of 1X side bands across BPFI harmonics. When there is pit or flaking at inner ring raceways then there will be BFPI harmonics with only one side band across BPFI harmonics.

When we look at the demodulation time waveform we see severe impacts of mixed nature some are low frequency and some are high frequency. Low frequency impacts are 180 CPM or 1X apart which indicate that bearing is loose over shaft. high frequency impacts are BPFI related which indicate cracked inner ring.

Now in the picture I have attached axial vibration FFT Spectrum which clearly has a peak at 3X which indicate that bearing is loose over shaft and the looseness is severe because unless and until looseness in not severe 3X peak mostly not comes in FFT Spectrum. Moreover sharp impacts in Demodulation time waveform also verify looseness over shaft.

bearing pics are attached with crack shown and bearing looseness over shaft can be visually see.