Hi guys,
I came across this spectrum on a 10 stage centrifugal Worthington pump coupled to a gas turbine and i have no idea whether there is something wrong with it. The levels are relatively low but the crest factor on the time waveform are qute high. Plus, on the pump outboard X radial spectrum, i noticed that there are extra peaks (not a running speed harmonic) but are spaced equally with the turning speed frequency. Oh yeah, the pump bearings are sleeve. I have no trending as this is the baseline data. The only thing i can relate to is that these non synchronous peaks are related to the vanes on each of the impeller stages (13 vanes on each stage)or maybe a potential rub. Any insights?

Sleeve_Bearing.doc (56 Kb, 59 downloads) Sleeve Bearing

What is the label on the LHS of the spectra. I can't read upside down. Looks like P-PDspnMcons?

On the TWF's, I see impacting-type waveform periodic at 1x with magnitudes as high as 100g's peak at inboard radial Y and 600 g's peak at outboard radial x. Do I have those units right?

I assume these are housing measurements?

Hi Pete,
It's peak to peak displacement microns. And these units are taken from shaft not housing measurements. this may sound weird but if i am taking measurements with a displacement probe, am i to evaluate the waveform in G's or still in displacement units?

You have misconfigured your data collector. It thinks the input is an accelerometer and the displacement unit is double integrated from acceleration.

If you are using a displacement probe, the waveform units as well as the spectrum units should be microns pp. You need to check your configuration.

Hi Steve,
Good morning...I am pretty sure that my setup is for a displacement probe. It's just that i am unsure on how to look at displacement readings especially on the waveform. Am i supposed to look at waveform in G's or in microns? From the G's point of view, there's a lot of impacting every revolution. Please see the attached file after i have converted the waveform in G's to microns

Pump_Sleeve_Bearing.doc (54 Kb, 31 downloads)

If you are dealing with shaft probes, and not a casing accel you want microns pp for both spectrum and waveform.

With a casing accel, you want g's.

Just out of curiosity where in Malaysia are you. I do some work there occasionally as I work for Shell Global Solutions. If you are at a Shell facility look me up and give me a call, we can chat about it.

Hi Steve...
I am currently in a YTL Power Services (Terengganu power plant somewhere in Paka)somewhere nearby kerteh.

hi WooKP,

About your Route Waveform (last fig.), If the P-P value and PK value is in micro m, my opinion is, that this pump is in good condition.

Hi there GT-X2,
Yeah i think so too... It's just that i messed up on the way i looked at the waveform taken using a displacement probe. I was looking it in acceleration which looks like it has severe impacting. But still i couldn't make any sense of my pump outboard x radial spectrum whereby the peaks after 600hz are not exactly harmonics of the turning speed with 13.4 orders.