Hello, I am aware of FFT, when use velocity and when acceleration but could you please help me to learn reed acceleration spectra, I am sending case of one measurement and I co not understand high acceleration reading far from mill...

Thanks...

case.pdf (43 Kb, 101 downloads)

A few questions first!

The data shown is it from the motor or the support bearings for the fan? I,m guessing it is from the motor Non drive end position as it states that it is point #1. It is a little difficult to analyze your machine off what you have displayed so many people will not respond as the limited info can make them hesitant to put themselves out there.

If I were you I would not focus on spectra completly when doing your analysis. Have a look at the waveform data as well.If you only use spectra to analyze then it may be time to learn time waveform analysis.

As for the data you have attached it seems to have a run speed vibrations & multiples. The acceleration spectra also shows impacting at the rate of the run speed & also has a raised noise floor.(Out of balance, looseness and bearing wear are all possibility's,just guessing at this stage).
The above is just my opinion on your limited data supplied, this is in no way a full analysis on the machine condition!
The next case you send you may want to put the cursor on the run speed for example and a time waveform.

Hooch

Looking at accel - you're fine but may be in need of lube or at lube interval (a lot of noise or elevated floor).

More data though would be helpful; but, all magnitudes presented are low.

Hi Enrique,

I have some doubts about your data posted!! I think both of your spectrums are acceleration spectrums. The unit says mm/s2 and it is written as velocity. Anyway, your g's are in very low levels as Sam mentioned. Also the information available is not adequate..

Have a nice day!!

It was my mistake here are more detail information about the case...thanks a lot!!

The_machine_is_Beater_wheel_mill.pdf (49 Kb, 52 downloads)

I may be way off the theme of the questions but are you saying you are trying to determine the condition of the actual rotor wheel/beater plates rather than the bearing condition by using acceleration and/or velocity on data from the bearing?

Using imbalance, as you said you do, to determine the condition might be ok, but I do not think you will be able to determine the actual surface condition of the wheel from acceleration analysis.

The first spectrum looks like the amplitude might be close to 4 mils. What was it when new plates were installed?

Of course I do not have a clue as to what the wheel looks like or how it does the job it is designed to do or anthing like that. Is the milling done on the front face or on the outer circumference of the wheel?

The data you have shown, is it with the milling actual being done or in an non milling mode? What does it "mill"?

The acceleration data, like others have said, looks ok as far as the bearing goes, but if you are indeed trying to determine the plates condition through this process, you may be chasing a ghost. What other surface is affected by the milling, just the wheel?

Hi Ralph,

Yes you are right about everything, you are reading my mind...

yes I am trying to determine the condition of the actual rotor wheel/beater plates wear condition and bearing condition by using acceleration and/or velocity.

I am trying to determine actual surface condition of beater plates/wheel and in the process of milling there is wear of plates and wheel, but wear can by symmetric on every plate or part of wheel and vibration unbalance will not happen...but it is aim to not allow to damage wheel that much, stop the mill and replace beater plates...

The second case can be that wear of plates in asymmetric and after short period of time rotor wheel needs to be replaced... this is due unbalance.

Spectrum is just before replacing the wheel, because the mill was opened and diagnose to much of wear on wheel and plates, although vibration amplitude was not to high...

Wheel design: is is beater wheel mill, is is to disks conested by beater plates in between and due rotation and centrifugal force it milling the coal and prepare it to combustion in fire plant...

The data are during the milling process...

Other surface are not that important, the wheel and plates are because it takes time to replaced and during that time combustion process using another source of power...

Thank a lot to all, thanks Ralph...

Is there maybe a way to determine what you are trying by examining and comparing the size of the "crushed coal" when new to when worn? Seems like when the system wears there would be a significant change in the size of the crushed coal. Or maybe it goes straight into the boiler and can not be measured, huh?

Have you taken any data as close to the actual wear and work area, looking for something resembling cavation in a fluid system? Comparing when new to when worn?

Good Luck on this project.

No, it goes straight into the boiler and based on quality of milling we have quality of combustion...yes it is hard to figure out...thanks a lot...