High Vibes Trip during startup

[REVenezuela]

Hoping to get some advice on this.

This is what is happening:

1. While trying to start up a centrifugal compressor we experienced high vibration trips on the high speed shaft outboard bearing of the gear box (both X-Y probes, dual voting), inboard bearing went to alarm as well (X only).
2.- Although the first time this situation was noticed it occurred three times consecutively, after that occurred randomly until we managed to start the unit, no vibration issues at full speed or during the ramp up were observed then.
3.- During the failed startup events the motor barely reached 100 rpm, while on the compressor (and GB high speed shaft) reached 420 rpm.

The train configuration and some background info has been attached to this post.

Additional info:

1.- Ops reported a loud noise during the first two failed startups attempts. Since the startup occurred around 4:30 am the Rotating Engineer was not present while these trips happened. This "noise" seem to have dissapeared.
2.- The unit was brought down normally to fix an inadequate installation of a flow instrument located at the discharge of the compressor.
3.- The compressor was overhauled on July 2010, the gear box was opened and inspected and found to be ok during the same time.
4.- After the three first trips it was decided to open the top cover of the gear box and inspect for damage or wear on the gears, nothing (evident) was found. Then it was coupled to the motor and ran with no load (compressor uncoupled) and everything went normal.
5.- Instrumentation assure the probes are ok (due to the random nature of the events I still have my doubts). Since the trips occurred a few seconds after pushing the startup button the response of the vibration monitoring system didn't collect valuable data to be analized, it barely captured the vibration spikes on the GB, the probes went overrange after every trip.
6.- Shaft centerline plots suggests coupling misalignment high speed shaft-compressor. This was confirmed after reviewing the maintenance records from the last overhaul, however in the previous months startups this was never an issue.
7.- The thrust bearing on the compressor was inspected last September because of relatively high axial displacement readings. This was a condition that had been followed up, the axial float was corrected and the DBSE (GB-compressor) was adjusted accordingly.
8.- A serie of additional events and trips happened during this messy start up but we could find a reasonable explanation for them, it is not the case for the GB trips.
9.- I noticed while correlating process and vibration data that all four vibration probes spike up to 1 mil p-p when any of the lube oil pumps are started. This doesn't happen in any other probe of the train. I don't know whather or not this relevant but considering the amplitude it seems odd to me.

We are kind of stuck in the RCA, I wouldn't expect high vibes at such a low speed unless a significant damaged is present but obviously I'm missing something, that's why the post.

I have put several plots and spectrum at different operating states and dates so you have a better idea about the case.

Any advise or suggestion will be highly appreciated.

Thanks in advance

 

[electricpete]

Since you mentioned a loud noise was heard... is it possible that liquid was getting into the compressor?

=====================================
(2B)+(2B)' ?

 

[REVenezuela]

Yeah....that was considered during the analisys but 1st Ops will never admit it and 2nd once we brought the compressor back online for a few hours we had a new trip on low suction pressure (CCC was tuning the antisurge controller) a few minutes we reset, met all the pstart permissives and again a new trip on high vibration (being this event a few minutes later after the machine was online there's no chance to accumulate liquids again n the system) so we kind of ruled it out.

Thanks for the suggestion though

 

[jbattershell]

something fishy going on. loose probe on gearbox high speed pinion?
any chance that during inspection of gearbox, oil flow passage was blocked or piping disconnected that feeds the bearing with oil?
hs pinion on gearbox has trouble starting on Nov2, but is ok to start nov4? how did it get to speed nov4 if two days earlier, it couldn't get past 100rpm?

 

[REVenezuela]

Hi jbattershell,

As I mentioned before E/I ruled out a probe related problem. During last September inspection on the compressor thrust bearing neither the GB or pipelines were touched on this equipment. The GB was last serviced on June-July 2010 and no vibration issues since then. And yes, vibration trips on Nov. 2, but then we were able to start on Nov. 4 although we have some other issues to bring the plant back online and in between these issues we had again trips on high vibes.

For your information the machine is still up and running with no vibration issues. Last thing we are considering is a posible axial misalignment between the compressor and the high speed shaft, this after reviewing the maintenance records following September job, it looks like the distance between coupling flanges could have been altered which may have been reducing the axial backlash and possible becoming bound on startups, as the machine goes to full speed / load and the pinion centered itself the high vibration dissapear. It is just an hypothesis, nothing confirmed yet.

Thanks anyhow for your comment

 

[REVenezuela]

Hi community,

Wanted to give you a feedback on this case.

Axial misalignment it was found to be the cause of these trips, after reviewing the last maintenance report it was found that 40 mils were removed from the shim pack used to adjust the axial alignment of the coupling spacer, the reason why this was done has not been explained yet, although most likely followed what the installation drawing recommends however actual conditions were not taking into consideration. This is quite a big change which for sure is disturbing the pinion alignment against the bull gear especially during transient conditions were the axial mesh force can be less than external forces.

The compressor is running now with no vibations issues so we'll schedule a job to fix this problem at first opportunity available.

Lesson learned; OEM Reps are not mistaken proof and we should always keep an eye on their job, as left conditions must be always discussed and understood before and not after the fact. In this case obviously the effect of removal such amount of shims was not discussed. Axial misalignment is not very common that's why sometimes it's overlooked, we normally pay more attention to correct the paralell / angular misalignment. This case I think proves how important both are especially when there's a gearbox involved.

Saludos!