Retrofit fix for motor end shield resonance

[MS0000]

I am now spamming two topics in a row.

I have a problem with end shield resonance vibration in an industrial motor. It 4000 HP 1500 rpm machine with newly modified bearing system.

I cannot change the excitation source which is however new anti-friction bearing(DGBB) in drive end. Bearings are not spring-preloaded because there is cylindrical roller in the other end. Motor is a little bit inclined because of the application so that is basically all axial load what comes from rotor mass.

I think there might be two options:

1. End shield needs to stiffened or loosened or the damping of the shield needs to be increased. In practice, what kind retrofit modifications have you done when facing this kind of issue?

2. Preloading DE bearing with the coupling (axially elastic lamellar coupling is used)... Anyone tried this?

All lessons learned stories and rules of thumbs are highly appreciated!

 

[Skogsgurra]

I have used that sticky black tar-pitch-rubber thing used for cars on other parts of motors and that reduces the Q value so you do not get any resonance. It doesn't look nice but if you can open the motor and put it inside the shield, it may be acceptable.

There is also a material consisting of rubber and lead laminated together in several plies. Good resonance damper and looks better.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[MS0000]

Interesting. For how large machines have you done that?

 

[Skogsgurra]

One pump motor at a power plant and one at a paper mill. Must have been 400 - 600 kW. Siemens motors with cast iron pecker box cover that screamed so you couldn't stay close to it. Both motors fed from VFDs.

That rubber-lead laminate was used to reduce vibrations in a supercalander.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[MS0000]

In what frequency range was your noise/resonance xkHz? My problem is at about 100 to 150 Hz.

 

[Skogsgurra]

It was much higher. Probably 400 - 600 Hz. Didn't measure, but that was what it sounded like. Stiffer end-shield is probably a better solution if you are that low in frequency.

The calander was lower, though. More like your frequency.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[electricpete]

Assuming the excitation is a single constant frequency, some options to address resonance include:

1 - rigidly attaching mass
2 - attempt to stiffen
3 - create a tuned dynamic absorber

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

 

[electricpete]

Attached is a presentation of dynamic absorbers including very successful application to reduce resonant vibration on a horizontal motor generator set at our plant.

I had to delete most slides in the 3rd part (stress estimation) to get it small enough to upload. If anyone wants that part, let me know.

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

 

[electricpete]

By the way, how did you conclude you have a resonance?

A few ways to test for resonance:
attempt temporary bracing... often by wedging wooden timber between machine and nearby support ... and check for vibration reduction.
Observe behavior during speed changes.

"My problem is at about 100 to 150 Hz."

I'm curious why you say "about". The frequency changes... or is spread out among multiple peaks.. or you don't have a vib analyser ?

If you know the exact frequency and the bearing part number, it would be interesting to see how this frequency relates to the bearing fault frequencies such as outer race ball passing frequency.

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

 

[MS0000]

Hey. Thanks alot for your efforts pete. Very interesting presentation.

I forgot to add that it is variable speed drive(0 to 1500 rpm). I dont have the vibration data now with me, to see exact resonance frequency. I am at home on a sick leave. But yes, it is one clear peak. It seems to correspond 2 x outer race bypass freq.

Resonance noise was first judged by bearing specialist who has been dealing a lot with similar motors. And he verified that it is "absolutely" bearing exciting the end shield.

 

[electricpete]

For variable speed drive, the dynamic absorber is not an option to resolve resonance. The reason is as shown on the 19th slide of my attachment where it shows the effect of dynamic absorber is to split the resonant peak into two other peaks at higher and lower frequencies..... works well for fixed speed, but you might be running on those other frequencies with vfd.





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

 

[electricpete]

Also stiffening and adding mass may not be great options for vfd for similar reason. To be successful you have to move the resonance out of the range where it can be excited by 2*BPFO, which varies with machine speed.

For vfd I'm thinking damping devices like Gunnar's tar thing are the best option.

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

 

[Skogsgurra]

I would have loved to do what you did, Pete. But my customers/problems are spread out and never close to my office. So I have to use quick and dirty (yes, really dirty) solutions.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.