Reading Motor Vibration Data

[bellman]

We have had some motors built to a customer's specification. and recently received the actual performance test results on theses motors. The customer is questioning the vibration analysis of the motors. The spec reads "Bearing housing vibration velocity shall not exceed the unfiltered values listed below" which for this 4 pole 1800 RPM motor is given as 0.12 in/s. The test results give 3 different readings in these units, Cursor Amplitude, Spectrum Overall Amplitude, and Band Overall Amplitude. Each of these is given in 3 different axises and all are well below the specification. Which of these relates to the customer's specification? The test speed was given as 3600 RPM if that makes a difference.

 

[BobM3]

My guess is that the spec writer wants no more than .12 in/sec (peak including all frequencies present) at 1800 RPM. For that I'd argue the you would look at the maximum velocity in time domain, not frequency domain ("spectrum" and "band").

 

[bellman]

Thanks!
That seems to make sense.

 

[electricpete]

I would say the operative word is "unfiltered" which would imply an overall measurement. Generally same as spectrum overall, although the true "analog" overall can be a little bit higher if there is more content beyond Fmax.

The terminology maximum velocity in the time domain from previous reply is ambiguous to me. If it was meant true overall or analogue overall, I agree. But the terminology could be interpretted as true peak, which is not what the spec is talking about imo.

Note 0.12 ips is the level identified in NEMA MG-1 for new motors tested on rigid mounting. There is some discussion there about the meaning.


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[bellman]

Thanks again,
The actual readings for spectrum overall from the test is as follows: Vert. 0.017 ips, Horiz. 0.074 ips,
Axial 0.024 ips.

 

[Tmoose]

what is the max frequency of the spectrum? Ought to be 1000 Hz or maybe more

I'm wondering why the test speed of an 1800 rpm motor might be 3600 rpm. Possibly planning for overspeed when used with a VFD? Or, a default setting in the software.