I have limited information about using Vibration analysis to detect loose stator coils (number of stator coils x rpm) is it feasible.
This is a very large variable speed machine approx 20-30Mw.white metal bearings
As there could be a large distance between the stator and the outside of the motor any suggestions about where to collect the data?
There is a possibility that the motor core pack could be accessed, do you think the stator magnetism would cause problems?
Any help or advice greatly appreciated
Mike

I posted a presentation on a similar subject awhile back. In short, we found very strong repeatable correlation between low-level harmonics of 2*LF (i.e. 4*LF, and expect 6*LF 8*LF). These were low level vibration but on log scale, the difference in pattern between confirmed loose coils and the same motors with known tight coils after rewing or rewedged coils was very evident. There was a caveat about other possilbe sources of these frequencies so it's not definitive. On a horizontal sleeve bearing motor, it is not too tough to get a look at the endwindings. Evidence of movement there would be telling. Lack of evidence at that point would not be as conclusive since I have seen very loose coils/wedges when the endwinding showed no evidence of movement.

http://maintenanceforums.com/eve/forums/a/tpc/f/3751089...051039102#9051039102

EPete
A case history is just what I needed, I am trying to advise some non vibration folk who have misread or have ill advised.
Many thanks Mike.