Included is a velocity spectrum of newly overhauled 3566 RPM ball bearing vertical greased lubricated 2 stage pump process pump(electric motor driven)

The dominant frequency is at 1.8x with a few harmonics.A spectrum was also taken with high resolution to verify the non- syncronous freq. Normally when I see non-sync. with harmonics I suspect a bearing problem.

Demod readings and HFD readings were also taken but did not show abnormal amplitudes. Also the 1.8x harmonics do not appear at higher frequencies as I woul normally see.

By the way,this pump was pulled for vibration initially but the issue is still to be resolved.
The pump has run satisfactory in the past since initial installation.

Any thoughts would be greatly appreciated. I can post additional spectrums if interested.

Thanks in advance,
Robbie

G-106_velocity.doc (39 Kb, 37 downloads)

Robbie,
Something is wrong with your freq ID. The peak you have marked as 1.83x is the same distance from '0' as the 2x,3x,4x are from the 1.83 peak. Isn't that truly your 1x peak? In that case, the sidebands surrounding the 1x peak look like they should be your issue, along with the 1x. They could be cage frequency modulation.

Ron... Thanks for the reply.
The 1.83x is actually that. The small peak to the right is approx. 2.56x followed by a few smaller harmonics of the non-sync. 1.83x.

Thanks,
Robbie

Robbie,

Could you take hi res data, say Fmax=30,000cpm and 6400 lines? It looks like there are some other strange peaks and you also want to make absolutely sure that harmonics of 1.83x are present or not.

David

This message has been edited. Last edited by: David_G,

Dave,

I am unable to collect data at this point because the machine is out of service.
I had previousely collected high resolution out to 12000 cpm and 6400 lines to confirm. Notice that the frequency of interest is not quite the same but is definitely still non-sync.

Thanks,
Robbie

g-106_vel_hi_res.doc (39 Kb, 24 downloads)

Robbie,

Unfortunately 12000 cpm is little bit too small to see if there are harmonics of 1.79x in order to rule out a resonance. Although unclear how to interpret, but numerically 1x and 2.57x look like SBs of 1.79x spaced at 0.79x. Is there any frequency at 0.79x which may be modulating resonance?

If feasable I'd check for resonance of various components.

David

Dave,
I did do some simple bump testing at various locations and did not see any related frequencies.I also have not seen any frequencies around .7x.

The first spectrum sent does appear to me to have some harmonics of the fundemental 1.8x but does not have high resolution(800 lines). But this is just a guess at this point.

Thanks,
Robbie