Sudden high vibration in a 3000 RPM motor driving a compressor 1

[edison123]

A 6.6 KV, 1500 HP, 3000 RPM cage horizontal sleeve bearing motor is driving a compressor through an intermediate shaft having 4 flexible plates of 3 mm thick at each end.

The vibration peak velocity (mm/sec) with the motor decoupled was as below

DE - H/V/A - 16/9/7

ODE - 15/8/8

Trim balancing on the decoupled motor was done in situ at 3000 RPM with the following final peak velocity levels

DE - H/V/A - 2.6/0.5/0.5

ODE - 1.9/0.5/0.4

The motor was also found to have an axial play of about 5 to 6 mm in the decoupled state at 3000 RPM. The motor does not have a thrust bearing.

Motor was then coupled to the compressor through the intermediate shaft and the vibration readings of the motor did not change much in the coupled state. The axial float was also reduced to 2 to 3 mm.

After about 15 hours of running on load, the motor vibrations suddenly jumped to

DE - H/V/A - 10/6/8

ODE - 11/6/8

The bearing oil temperature was around 80 deg C, which is normal as per the client.

The motor vibrations dropped to the after-balancing levels the moment the motor supply was cut off.

The motor is found to be electrically ok. I am suspecting that the flexible plates could be the problem for this sudden increase in the motor vibration level (like cracked plates). I also attribute the drop in vibration when the supply is cut off to the fact that the torque transmitted through this flexible plates drops the moment the supply is cut off.

Are my surmises correct ? What else could cause this sudden jump in the motor vibrations and its vanishing when the supply is cut off ? Could the axial float be the cause ? If yes, then how the motor ran for 15 hours with the same axial float without any issues ?

(The rotor run-out, eccentricity, uneven air-gap etc. can be safely ruled out since this was a running motor and the motor sleeve bearings are supported from the end brackets fitted to the stator frame.)

Thanks in advance for your tips.

Muthu

http://www.edison.co.in

 

[electricpete]

What frequency is the vibration?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

Since balance brought it way down, I assume it is 1x.

It has somewhat of resemblance to this one:

http://maintenanceforums.com/eve/forums/a/tpc/f/3751089011/m/174100007?r=174100007#174100007

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

pete

It was 1X in all directions. Balancing at 3000 RPM brought it down to very levels. The spectrum is attached.

Muthu
www.edison.co.in

 

[waross]

I would check the vibration at the compressor input shaft.
I would decouple and check the motor vibration.
I would inspect the flexible plates.
If possible, I would reassemble the intermediate shaft with the flexible plate stacks swapped or better, the shaft also turned end for end, and then recheck the motor vibration and the compressor input shaft vibration.
If jam nuts are used on the disk pack fastenings, count them.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[edison123]

Further feedback from the site received just now

Motor decoupled vibration

DE H/V/A - 2.5/0.5/0.5

ODE H/V/A - 2.1/0.3/0.5

But,

"However we found that the vibration had increased in the NDE side from 2.1 to 6.9 and an abnormal sound was found in the motor and hence we immediately halted the operation of the motor."

Can't make head or tail of it.

I will be at site in another 16 hours. Any pointers ?

Muthu

http://www.edison.co.in

 

[electricpete]

Tough to say. But since you're looking for quick feedback, my vote is similar to the linked motor:
1 - rotor defect.
2 - some problem allowing dynamic eccentricity: excess bearing clearance, rotor bow (possibly thermally-induced)

On-site checks include:
rotor TIR at accessible locations
bearing clearances
bearing inspection
airgap checks 4 positions around stator. Then rotate rotor 90 degrees and repeat.
depending on test equipment availability: rotor influence check (pdma) or single phase test much better.
bearing oil sample

If you see changing behavior over time (without change in load) points to rotor problem of some kind.

Probably nothing you haven't already thought of. Good luck

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

Thanks pete. Good luck indeed. I am going to tons of that tomorrow.

Muthu

http://www.edison.co.in

 

[edison123]

I was at site and did a vib check on the decoupled motor run for 90 minutes. No problem in the motor solo run.

After coupling the compressor, under compressor unloaded state, there was a slight increase in the vibration.

As the compressor was loaded, the vibration continued to rise and at full load, the vibration went beyond the allowable limits.

The readings are attached.

I am suspecting thermal growth and the resultant misalignment. In the cold alignment, the motor was off-set by 0.10 mm (up) as per the client.

Any ideas ? The motor is running with these high vibrations as of today.

Muthu
www.edison.co.in

 

[wolf39]

Assuming a piston type compressor it may be that the compression of the cylinders is unequal.

Wolf

http://www.hydropower-consult.com

 

[electricpete]

If I understand correctly, you had a dramatic jump in vibration with motor uncoupled. This suggests to me problem with the motor, unrelated to compressor and unrelated to alignment changes from thermal growth.

Also you have not reported compressor vibration.... I assume it remains low/normal? Again points more to the motor rather than to compressor / alignment problem.

I lean toward rotor problem of some kind. That is about the only thing I know of that causes increasing vib over time other than alignment changes (which seem ruled out as discussed above)

Did you get high-res spectrum and check for pole pass sidebands?



=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[electricpete]

The basic types of rotor problems I know of are the following three types:
1 - electrical high resistance due to crack at end-joint etc.
2 - mechanically loose bars contribute to shifts in unbalance
3 - mechanical effects of rotor bar movement causes thermal bow.

I don't think type 1 shows up during uncoupled run. I think type 2 or 3 most likely 3.



=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

Further feedback.

The motor rotor was balanced in the hot condition immediately following a full-load run. The rotor required further addition of about 350 gms very near to the same balancing locations of the cold balancing. The vibration levels were brought down to about 5 mm/sec peak in both bearing housings by this hot balancing.

But after about 8 hours of running, the vibration started rising again to 10 mm/sec peak and is staying there for the past 12 hours.

A very strange behavior so far from this motor and even the OEM of the motor is unable to reach a conclusion about the reasons for this.

Muthu

http://www.edison.co.in

 

[TonyCro]

Hello Muthu,

sorry for coming in late to this.


Can the OEM supply a cross sectional drawing of the motor ?

What is the cooler, Air or water and is it functioning ?

When the vibration step changes,
is there a change in phase angle ?
is there a change in process or adjacent machinery ?


Have the bearings been checked ?



Given that the rotor was balanced, run and ok till a sudden spike and then balanced again - this suggests that something is moving on the rotor; Fans, balance disks, cage, core ? but it may only be appearing after a prolonged heat soak and the loose component expands and 'releases' the rotor causing a change in balance condition. As the rotor cools after switch running the component then shrinks back on and 'grabs' the rotor in a subtly different way.


It may the cooling is no longer adequate and the rotor getting hotter than ideal ( blocked stator/rotor ducts or dirty/blocked cooler tubes - water on ? )

Can you run the motor uncoupled for an extended period, 12to 24hrs to simulate the time taken for it to change ? If the air/winding temperature doesn't reach 'normal' running levels then consider blocking the cooler to 'boost' the internal temperature - clearly this requires a degree of caution and close monitoring.

tony.

 

[wolf39]

Muthu:

Check the rotor design. In case there is a laminated or solid rotor rim shrunk onto a spider shaft then there is a possibility that after some time the temperatures are high enough to separate the rim from the spider.

Wolf

http://www.hydropower-consult.com

 

[wolf39]

Tony mentioned lots of possible causes but the most likely is the one I mentioned yesterday.

Wolf

http://www.hydropower-consult.com

 

[edison123]

Thanks Tony.

I don't have the OEM drawing neither does the client. This is an open air cooled motor, only it has a top mounted intricate air guide with inlet filters. The air is taken in from the sides of the motor and is guided through the DE by internal centrifugal fan, which then blows out the hot air, again through the same air guide towards DE. The whole air guide unit is a light weight structure and removing it did not change the vibration levels.

The fan, coupling and the balancing disc in ODE are keyed and tight fit on the shaft. No looseness there. About core looseness, I do not have any idea.

The client is not permitting a long uncoupled run due to production issues.

Muthu

http://www.edison.co.in

 

[edison123]

Thanks wolf. This being, a high speed motor, does not have a spider shaft. The rotor core + winding assembly is shrunk onto the shaft directly. It is keyless shrink fit of 0.3 mm as per OEM.

Muthu

http://www.edison.co.in