Sign up Calendar Latest Topics Donate

  Author   Comment   Page 2 of 2      Prev   1   2

Sr. Member
Posts: 76
Reply with quote  #16 
Hi Mniroshan,

could you post some data? Theoretically there are many possibilities, as Shurafa mentioned, and it will be difficult to narrow them without seeing some trends and spectra.

Some vertical pumps sets share the same foundation, that can produce beating and or background vibration.

If you're measuring near the discharge pipe, you may have a lot of flow noise, which can produce unstable readings.

Flow "inside" the machine, such as surge or rotating stall in a compressor, may also produce very unstable vibration readings. I think that flow recirculation in a centrifugal pump may produce unstable readings as well.

Being this a vertical pump, resonance is another clear candidate.

Instabilities will show as vibration at the natural frequency, although the frequency is quite stable the amplitude may be change a lot.

Rubbing can produce very unstable vibrations even if looseness is not present. Unstable would mean in this case changes in terms of minutes or hours. Coke rubs in turbines and compressors are a common example of this and very well documented in the literature.

So to sum up, there are many possibilities and data is needed to provide more accurate answers.

Avatar / Picture

Posts: 1,810
Reply with quote  #17 
Take a reading on the pump housing, away from the bearing.  Vibration should be fairly low, and the Crest Factor (CF) should be low when looking at the acceleration time waveform (CF = Peak/RMS).  If you have a high CF with impacting in the waveform, this could indicate something hung in the pump inlet, something rubbing the impeller, or a plugged or partially plugged inlet screen. (see example below)

One of the best indicators of pump flow is motor amps, and also actual running speed.  For constant flow, the amps will be constant.  On my route setup I have a point on the pump housing, and I also record the actual amp draw for large vertical pumps.  Also, I record the actual running speed for *every* machine, every data collection, whether variable speed, or constant speed (speed will vary slightly with load... if the speed increases, then the flow (work done) has decreased).

I trend the Maximum Peak Waveform value (I save a TWF for every collection, in acceleration) and Crest Factor.

This was a large vertical pump with a piece of wood wedged in the pump inlet.

Attached Images
jpeg VP1.jpg (85.73 KB, 17 views)
jpeg VP2.jpg (115.96 KB, 17 views)
jpeg VP3.jpg (135.84 KB, 18 views)
jpeg VP4.jpg (86.33 KB, 18 views)
jpeg VP5.jpg (87.66 KB, 17 views)

"The trend is your friend"

Previous Topic | Next Topic

Quick Navigation:

Easily create a Forum Website with Website Toolbox.