Sign up Calendar Latest Topics Donate

  Author   Comment  

Posts: 2
Reply with quote  #1 

I have a problem understanding the FFT or Order Analysis.
I have measured the structure-borne noise with the help of a accelerometer which was attached on top of a electric motor. The measurement was while the motor was accelerting from 0 rpm to 3000 rpm, so a ramp function.
When I want to analyse the signal I got, I have problems understanding it.
The FFT shows peaks at 50Hz, 100 Hz, 150 Hz etc. When I have a look at the order analysis it's the same thing.
I know that 50 Hz are 3000 rpm, but why do I just see that frequency? The motor accelerated from 0 rpm to 3000 rpm, so shouldn't I see all frequencies?
And maybe the FFT is just for stationary operating modes, but shouldn't the order analysis show me speed-depended frequencys (because of the 8 poles of the motor etc) and also speed-independed frequencies (eigenfrequencies)?
My aim is to figure out the eigenfrequencies. But the only speed-independed frequencies I can see in the order analysis are 50 Hz, 100 Hz and so on or also 12,5 Hz.
I hope I made clear what my problem is.
Can anyone help?

Thanks! And sorry for my bad English!
John from PA

Sr. Member
Posts: 963
Reply with quote  #2 
Why don't you attach a representative plot, and then we can gives some specifics as to interpretation? 
Ron Stiemsma

Sr. Member
Posts: 84
Reply with quote  #3 
It is difficult to take run-up data on a motor because they get to speed so fast.  Unless this is on a VFD.  My guess is if it just took one spectrum and it was already at speed.  The 50 and 100 Hz are the electrical frequencies and for a 8 pole motor the 12.5 is running speed.  I would need to know a lot more on how the test equipment was setup to be any more help.

Posts: 2
Reply with quote  #4 
Thanks for the comments. Unfortunately at the moment I cannot uplod a plot.
The motor is connected to a load machine which has a VFD.
I collected the time waveform of the accelerometer and calculated the spectrogramm with an analysing software.
I used 50 rpm/s to get to the 3000 rpm.
When I would like to do an operational modal analysis, how do I do it, if the operation mode is the run-up?
Walt Strong

Sr. Member
Posts: 889
Reply with quote  #5 
"operational modal analysis"

Can't do it with one accelerometer, and it is best not to attempt it during transient (variable speed) operating conditions.


Avatar / Picture

Sr. Member / Supporter
Posts: 599
Reply with quote  #6 
I imagine you are getting the 50 Hz harmonics because of a ground loop on your sensor. I get this all the time regardless of the accelerometer manufacturers telling you they are fully isolated. There is an easy solution that works 90% of the time. As long as you are not expecting to resolve too high frequencies and can use a magnet mount for the accelerometer, you can simply place a normal piece of paper between the magnet and the contact surface. If you are using stud mounting, you will have to use special insulating studs/pads or find a way to isolate the sensor some other way.

Can't do it with one accelerometer, and it is best not to attempt it during transient (variable speed) operating conditions.

Walt says this because modal analysis generally indicates that you want not only the natural frequencies, but also the mode shapes. You will not really be able to measure mode shapes with your set up, but natural frequencies should be easy enough, granted that you will have quite a bit of uncertainty in those values. If your equipment can handle dropping the ramp rate to 10 rpm/s, I would recommend it.
Previous Topic | Next Topic

Quick Navigation:

Easily create a Forum Website with Website Toolbox.