We have a 3600 rpm API overhung pump that on the discharge piping has a 1.25 in/sec vibration at a frequency of 982 Hz. 1X in the spectrum was 58.375 Hz, so this vibration is not synchronous. All the vibration on the motor and pump bearings is reasonable, but the buzz in the piping is rattling off valve handles and small branch connections.

Any ideas?

That corresponds to roughly 20 g's.

Would be interesting to see if it changes with flow.

Would be interesting to look at on a log scale and see if it is part of a harmonic series. (look for peaks at 982/k or peaks at k*982 where k is an integer). Also with high res spectrum is the peak really narrow (might suggest excitation at that frequency) or broad (might suggest resonance.)

Would seem to be some type of resonance (including fluid column resonance) excited by either broadband flow noise or blade pass pattern.

Edited to add - Regarding blade pass frequency, I am allowing a small possibility that your resolution did not allow you to determine running speed accurately enough to assess whether non-sync or not.

Also did the vib just show up? Note the magnitude at this high frequency will probably be sensitive to mounting method of the accelerometer.


That's just a lot of rambling from someone who doesn't know much about pumps. I'm interested to hear others' comments.

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

Steve,

I can't tell you anything you don't know about the vibrations, but my experience has led me to a theory about the situation you describe.

In most cases where I have been asked to find out about a vibration the piping, the solution has been in the piping and valve operations.

The theory is that broad band energy from flow turbulence created by pumping flows in oppostion, surging, pumping against closed valves, etc., excites resonance in something, somewhere.

Most cases have ended without really knowing what was in resonance, but knowing what causes the excitation (usually improper operation) and not doing that.

Good Luck,

Danny

Hey Pete, I used a 1200 line spectrum, and Fmax of 1000 Hz. I think it is not synchronous... I am going to try to set up a test where we can vary the flow. But that is a bit tricky from an operations standpoint.

Danny, I think you are exactly right, but I am having a hard time proving it. There is a caped off tee in the discharge line that I was thinking is probably shedding voticees at that rate, but all my calcs so far indicate that it should generate a frequency much higher than I am seeing. But who knows what other types of flow turbulence is being generated.

Piping support OK. Type? Spring hangers their pins pulled? Sonic velocity from flow?

Try springing with wedges at ???location and/or long pry bar/pipe to put tension on piping. Monitor every so many feet. I know I'm not telling you anything you don't already know but!!!!!!!