We have a boiler feed pump ("A") which is powered by a small turbine. Speeds range from 5100-5350 RPM depending on load requirements. The IB(coupling side)bearing cap is approaching .465in/sec. while the prox probe reading is a steady 1.8 mils. I know that prox probe readings tells me what the shaft is doing or how it is rotating, does the bearing cap readings tell me everything else that needs to be addressed? The highest vibration is the IB horizontal with most of the energy concentrated at 1X and 5X. I have also noticed that the horizontal reading on the other pump ("B") IB bearing is also the highest on that particular pump, but not as high as "A". The OB Pump bearing on "A" is .2in/sec. Could this be resonance?

Thanks in advanced

Rafael,
Have you analysed the spectral data from the prox probe or just viewed the overall value? With some systems the value displayed on the panel is only for the 1x shaft vibration, confirm what is actually being displayed. The waveform and spectral prox data will also help determine what the shaft is doing within the bearing.
The high velocity reading is possibly related to a looseness. Other possibilities are for a misalignment, out of balance and pump blade pass (if impellers have 5 vanes).
Suggest you check to following:
1. shaft orbits (if possible) or at least waveform and spectral data of prox probe.
2. phase of 1x Axial, Horz, Vert from NDE of turbine to NDE of pump.
3. If you can shut the pump down for a physical inspection I would check the DE bearing clearance, housing seat condition and any secondary mating housing seat conditions. Check the alignment of the unit, ensure thermal growth allowances are included. Check the coupling condition, if grease lubricated ensure high viscosity coupling grease is used and the condition of the mating parts are good.
What is the shutdown set point for these turbines?

Hope this helps.
Tony S
Disclaimer: These opinions are my own and not necessarily those of my employer.

Thanks for the response Tony. The shutdown set point is 5mils at the vibration panel(prox. readings). The high vibraiton at the DE bearing cap has been there for a few months now. I'm going to take phase reading across the coupling and see what I find.
Thanks again.

Rafael

Phase across the coupling has a 10°difference. Here's something to think about, when I said that "B" pump is lower it's only slightly lower. It's reading .45in/sec at the same place as "A" pump (DE horizontal) and showing the same peaks 1x&5x, but 5x is higher on "B" than is on "A". These cap readings are an overall value. Haven't had a chance to get orbits on the prox. Could it be a flow issue? Both pumps are on the same unit.
Thanks

Rafael,
Do you have spare impellers for these pumps in your store? If yes count the number of vanes. I have previously experienced high vane pass at 5X with boiler feed pumps, but not with the highest on the turbine DE. It may be that the turbine de has a structural resonance close to 5X in the horizontal direction. Ususally if it resonance related there will be a "rasied noise floor" around the peak.
To help confirm if it is a flow related issue compare the flow and head of the pumps now to when the vibration levels were ok. Are there any other harmonics of running speed present or harmonics of 5X? Sometimes looseness can exhibit dominant running speed harmonics. What does the acceleration waveform look like for the DE vibration. If it is impacting then likely looseness occuring at turbine bearing, if 5X sine wave then more likely flow issue from pump. The amplitudes on the pump may be reduced due to being of heavier/stiffer construction but the vibration pattern should be present if the source of vibration is from the vanes. Is the pattern of 5X on the pump bearings or casing?
Was any maintenance performed between the vibration being ok and now?
I suggest the you determine what is creating the 5X first then if you can not remove it determine if the turbine structure has a resonance at 5X. If it is then determine if you need to remove the 5x or just monitor it. If the vibration level is too high then you need to determine how you change the resonant frequency of the casing without creating another problem.

Hope this helps.
Tony S
Disclaimer: These opinions are my own and not necessarily those of my employer.

Tony:
Had a chance to look into the bearings and couplings of both pumps. Both pump DE bearings looked fine, the coupling nut on "A" was tightened 1/2 a turn, the coupling on "B" pump was fine. I took vibration readings before the unit came down and will take again after the unit comes back up again. If the high vibratin continues, I'll probably have to look at internal issues even though the vibration on the prox probes are readings a steady 1.5 mils. What do you think?
Thanks in advance

Rafael,
Were you able to inspect the turbine DE bearings? I would suspect these to be the possible loose items based on the information above. I would check for the shaft loose in the bearing and for the bearing loose in the housing.
Were you able to perform any of the previous suggested checks? If yes what were those results? I will provide additional suggestions when you are able to provide more data/inspection results.

Regards
Tony s
Disclaimer: These opinions are mine and not necessarly those of my employer.

This might no apply to your unit but I collect data and have the same frequencies showing up on a turbine driven boiler feed pump in a power plant. The case vibration is high but the prox probes show the shaft deflection is within limits. The source of this vibration was found to be coming from the coupling guard which is rigidly mounted to both the turbine inboard bearing and the pump which is transfering the vibration.