Hi Claudio,

Not sure I understand your question. Can you elaborate a little more, in that way you may get the answer you are looking for.

P

The meaning is simple indeed. Suppose a 2MVA four-pole wound-field generator has around a 9% of maximum rotor center displacement (including both static and dynamic component, the % is on the normal air-gap). May I diagnose that percentage displacement by only using vibrations? And with how much accuracy?

Ahh, now I understand. Th simple answer is no. the vibration of the rotor shaft is only one source of eccentricity. The windings and any 'un fixed' components all expand under rotation, also the vibration in the journal bearings add to the rotor as does expansion due to heat. A final addition is the momentary addition of electromagnetic expansion due to the system itself. Therefore you need to look at texts taking these into account and see if they apply to your system. I hope this helps a little, get back to me if you need more.

P

Thank you Peter. The reason for my question is more practical than academic. I have recently carried out a practical method to assess the percentage displacement of a working machine rotor by measuring split-phase currents in parallel-connected windings (SPCSA). This has been done in the scientific track of "condition monitoring and fault diagnosis of electrical machines by using voltage or current signature analysis", which is much appreciated by electrical engineers worldwide this years. Please refer to a flag-ship conference "IEEE SDEMPED" symposium, completely devoted to CBM of electrical machines by MVSA (machine voltage signature analysis) or MCSA (machine current signature analysis). I know this sounds like electrical engs. try to steal the job to mechanical engineers, but the tools are completely different, i.e. electrical signatures instead of mechanical signatures. Now, I want to compare my eccentricity assessment results with results obtained with classical tools, since the manufacturers (ANSALDO, MARELLI MOTORI, and others) want reference measurements to verify the method (this is not simple with machine with natural eccentricity of 5%-10%, which I can asses quite simply by using SPCSA). So I'm interested in hearing from mechanical engs. like you, but I have received conflicting opinions until now. Other opinions are welcome. Cheers!

In case of eccentricity faults, from the stator current spectrum, it is observed that:

1. The magnitude of the fault component increases with increasing level of eccentricity, 2. It decreases with increasing load. 

The difference in magnitude was also very high when compared the level of displacement. But, again the vibration signals are yet to be analysed.