Babbits in hydroelectric generator bearing 2

[Skogsgurra]

This is somewhat off my line of expertise. So I hope that someone has some experience and can answer this question:

A segment bearing in a hydroelectric 5 MW generator/turbine combination has been severely damaged after ten years service. Babbitts metal has broken loose from the segments and there are several seemingly unrelated positions on different segments where this has happened.

I was asked to check if there is any electrical activity in the bearing and there is some very weak discharges with a rather low frequency. The discharge frequency is lower than 1 PPS (pulse per second) and the discharge voltage is about 1 V. This is normally regarded as totally harmless for a roller or ball bearing (needs 1000+ PPS at about 10 V to cause any trouble) but I can guess that the capacitance of the rather large (around 1/2 square meter) bearing with a thin oil film can be high enough to cause some EDM damage when it discharges through the bearing.

I have, for the time being, told the generating people that there is no cause for alarm regarding the EDM activity in the bearing. But I would like to have a second opinion from someone out there that has more experience with this kind of problem.

There are pictures showing the damages and I can mail them to anyone that is interested.

 

[unclesyd]

I wouldn't think that your have arc damage from your description and the voltages involved. Saying this and as you state the potential is always there so the potential for the other is there also.
I would appear that you have problem with “fatigue” in the babbitt. This is actually a bond problem but given the age of your bearing it not very likely. 10 years is long time for a bearing in this type service.

Can you see any damage on the edge of the babbitt?
This where this type damage, if I am reading you right, usually starts.

Is your shaft straight?
Not bent or running in a orbit.

Have you changed lubes with the last year or so?

 

[Skogsgurra]

Dear Uncle,

Yes, the shaft is straight.

No, the damages are not on the edges.

Yes, it looks like a bonding problem, now when you spell it out. The babbitts has separated from the substrate and you can see the ribbed steel below it. The flakes are irregular and 10 - 30 millimeters wide. There is some discoloration and "moon landscaping" on the surface of the loose flakes.

I do not know about lubrication. It was mentioned that there could have been som contamination this summer. No one seems to know for sure. I will go back to the operators about that.

Would a couple of pictures help?

 

[unclesyd]

Take a look at this site and see if it helps you.
I'll get back with you on the pictures.

http://www.machinerylubrication.com/article_detail.asp?articleid=587&relatedbookgroup=Lubrication2

 

[unclesyd]

Here is another article on this website (Machinery Lubrication).

http://www.machinerylubrication.com/article_detail.asp?articleid=638&relatedbookgroup=Lubrication

 

[Skogsgurra]

Dear Uncle,

These segments are plain segments without a leading edge groove. The damages look a little like figure 4 (thermal ratcheting) in reference

http://www.machinerylubrication.com/article_detail.asp?articleid=587&relatedbookgroup=Lubrication2

but with coarser details (larger chunks).

I will make the generator people have a look at

http://www.machinerylubrication.com.

Thanks.

 

[unclesyd]

Larger chunks are not unusual.
You could have either thermal racheting or fatigue, they are caused essentially by the same conditions.

Did you look at the second URL as it is a different article and has a picture of babbitt fatigue. A lot of time you will not see many cracks as the pieces may have come out prior to inspection. In the top fatigue photograph the pieces have been welded to the remaining babbitt, this is the norm if the bearing is allowed to run.

I’ll have to agree that a look at the magazine is a good idea. Any article by William Strecker is usually quite informative.

This is the second post.

http://www.machinerylubrication.com/article_detail.asp?articleid=638

 

[Skogsgurra]

Yes, I looked at both articles. There is no frosting. So I guess that I can leave this problem to "someone else".

 

[hoyle]

The main problem here is the ribs beneath the whitemetal. Its an old idea, and goes on the principle the whitemetal "grabs" onto the ridges. In some bearings, there are dovetails machined into surface for the whitemetal to locate to
In reality all the ridges/dovetails do is induce stress concentrations due to thermal and mechanical strains. A few hours of FEA on this model will convince you just how bad an idea this is.
Machine the ridges off, and get the surfaces tinned and whitemetal-ed properly. The bearing will last forever, as long as the oil is clean, plentiful and of the correct grade and temperature.

 

[Skogsgurra]

Thanks, hoyle!

That is a new insight. Do you have any references? As an electric engineer, I have very little knowledge about this. I do not even understand what FEA is. Failure Effect Analysis?

 

[Carburize]

FEA = Finite Element Stress Analysis

 

[Skogsgurra]

Thanks,

Shouldn't it be FESA? I mean; the TLA FEA is already taken. So a new FLA: FESA seems to be in order ;-)

 

[Carburize]

The failure mode analysis acronym is usually FMEA