Pump Motor Vibration

Was it a clean 2.2 spike, or was there a haystack of frequencies around 2N?

What was your Fmax set to and how many lines of resolution?

This may be a figment of the FFT process. If you change Fmax and or increase the number of lines of resolution (say from 400 to 800) it may go away.

This also may be a side band assosiated with the motor speed control, or it may be a torsional being excited.

Finally it is also possible that this is a natural frequency of the motor in the horizontal direction. Try running the motor slower or faster to see if it goes away. Also look for look hold down or bearing bracket bolts, cracked supports, or broken foundation indications.

Hope that helps, let me know what you find.

The horizontal measurements had a haystack around the 2.2N peak (~1450-1530 CPS), but it did not tail out across the 2N frequency (1342 CPS). The vertical and axial measurements also had a smaller haystack at 2.2N peaks, but there was a clean 2N peak with distinct separation between the two.

Fmax is 12,000 CPS with 1600 lines.

I thought about it being a natural frequency, but since it wasn't present in previous measurements and there is no excitation harmonic, the only thing I could think of was some sort of internal looseness. But I would have expected (perhaps erroneously) that looseness would show up on a harmonic frequency. I will inspect the motor closely and try running it at a slower speed to see what happens.

Thanks.

What length of the line from the last bend/elbow or valve to the inlet to the pump.

(pump manufacturer would have the minimum requirement)

If length is not sufficient, may have had slight cavitation that was at first unnoticeable, until at a point that enough of the impeller was eroded away, causing the vibration.

phoenix911

We have about 30 ft of straight 34" pipe on the pump inlet, which is borderline but unfortunately that's the way the plant was built. We do get some minor cavitation under certain flow conditions, possibly due to low NPSHA, but we have never had any issues with impeller erosion on other pumps.

If the vibration originated at the pump due to impeller erosion, wouldn't I see stronger vibration at the pump bearings instead of the motor bearings? I saw much lower vibration at the 2.2N peak on the pump, which I expect was transmitted through the rigid gear coupling.

Like your company, I have design using minimum requirements for inlets before, because of no other way.

That may be an issue, that should be looked at. Remember, erosion due to cavitation can happen immediate, be results or effects can show up later as its eroded away.

phoenix911

Well... I know you are worried because it changed but 0.12 ips is very OK. I like to think in terms of severity, not % of change. Ex: If a 0.005" changed to a 0.01" it is double ... "Vibes went up 100% ... !!!! " , but severity is still low.

Make a rough sketch of the horizontal profile of the unit from the foundation up, like looking at the unit, foundation, grout and machine from the end bell of the motor. Probe it horizontally each 1 inch or at each structure transition (foundation to grout ... grout to skid ... skid to motor mount) and graph the vibe levels at each point. You might just find the trouble area where the vibes jump. The graph helps to visualise 'looseness' or deformation. I drew cute little ppt and made a pdf of it but can't seem to embed it here. I can send it by email.

I agree that the vibration severity is still OK, but I am mostly concerned with the appearance of this particular frequency, which wasn't evident about a month ago. Also, since it jumped up to about 2/3 of the ISO limit of 0.28 in/s I suspect that the condition could rapidly deteriorate and become a problem. I will try taking additional measurements as you suggested and see if I can identify where the problem is located.

Thanks.

Assuming that you have a "flexible" gear coupling, not a "rigid gear" coupling and assuming that you're pumping "cool" raw municipal water that doesn't vary much in temperature or corrosiveness, I don't see erosion at either inlet vane( loss of NPSH ) or discharge tips (creating an imbalance) as being as much an issue. The coupling should have been selected so as to prevent any such transmission of vibration issues to the motor. I would suggest that when the opportunity arises to remove the pump from service, disconnect the pump from the motor and run the motor without any load through your spectrum of desired speeds and see if the vibraton changes are still there. If you have ever thrown a wheel balance weight on your car and noticed that there are certain limited speed ranges at which you have a shimmy then you'll know where this is going.