Centrifugal compressor- Vibration 1

[Psycho Dreamer]

Hi folks,

We've a vibration issue issue with the startup of a newly overhauled 4 stage compressor bundle.
Make: MHI
Operating speed : 13K
Minimum governance speed App. 9.5k

On startup, the DE vibration jumped to 24 microns and 22 microns all of a sudden during ramping up (Alarm: 35 microns) and on further loading to 100% it gets stabilized around 35 microns while the NDE probes settled around 20 microns.
Both the DE spectrums came out with a dominant peak at 1x.
The machine stopped and both the probes been replaced with new ones and on manual hand turning the vibration went upto 21 microns!
The residual magnetism at probe area surface was found to be varying between 2.5-7.6 A/cm with a North pole indication.

Can someone put any light on the above? Any data can be shared later as they are not handy now.

Thanks in advance.

 

[Psycho Dreamer]

Hello guys,

Would appreciate a kind response for the above.

Many thanks!

 

[JJPellin]

With the data you provided, the best we could do would be speculate about potential causes. Given the vibration frequency, you should already know the potential causes: Imbalance, misalignment, run-out. Is the shaft straight? Is the rotor balanced? Was the alignment completed with proper thermal offsets using laser alignment tools? I don't know the allowable amount of shaft magnetism, but you could easily do an internet search for that. If in doubt, demagnetize the shaft.

You indicate that the vibration was nearly as high on "manual hand turning" as it was at 13,000 rpm. That would tend to rule out alignment and imbalance. All that is left is magnetism and run-out.

Johnny Pellin

 

[Psycho Dreamer]

@JJPellin,

Thanks for the update.

The rotor was re-furbished 1.5 years back and all checks (Mechanical/Electrical run-out, Magnetic checks) were satisfactory. The high speed balancing also been carried out during re-refurbishment.
The rotor was stored in a storage container (unpressurized) vertically all these time.
Due to the narrow time window, no crosschecks been done on rotor side during assembly.

IS there any possibility for an induced magnetism /run-out during storage or transit via air?

 

[JJPellin]

I have generally not seen run-out change from shipment if the rotor is well supported in a good canister. However, for general purpose machines, we have seen this. For large fan rotors, it is not uncommon for them to exhibit high vibration when they run for the first time. I have attributed this to the fact that the shafts and wheels were not stress-relieved. They were assembled and balance at the repair shop and then shipped to us by truck. I believe that we were seeing changes in balance and run-out as a result of vibratory stress relief during shipment. Do you know if your shaft and impellers were stress-relieved?

For big fans, we simply expect to need to do a full field balance once installed. For your compressor, that will probably not be practical. The vibration levels you listed do not seem excessive to me. You will probably be forced to adjust your alarm settings and live with it until the next overhaul. Complete analysis of the vibration data should confirm if the balance has changed since the at-speed balance. I assume you have a full report from that balance with the final readings.

If there is any chance that magnetism is the main issue, you should be able to demagnetize the shaft in place. I would recommend that as a next step.

Johnny Pellin