Rust on bearing journal?

[AshleyK]

As a quick overview, we're commonly finding that our alternator bearings (6334, so a 170mm ID) have rust on the journal underneath the inner races when removed. The alternators are in a plant room and not exposed to any rain. Has anyone experienced something similar before?

The grease system is working fine. The only thing I can think of is if they were assembled in the factory after being kept in poor conditions, or whether they got wet during transport. One of the explanations given is that 'damp air creeps underneath the race when the bearing gets hots and expands', but I'm not overly impressed with that as an engineering diagnosis. The bearing would be spinning on the journal long before 'damp air' worked its way in.

Any help would be great.

 

[JJPellin]

It is more likely that you are seeing fretting corrosion caused by inadequate interference fit. What ISO grade fit is specified? Water is not necessary for corrosion to occur.

Johnny Pellin

 

[electricpete]

Got any pictures?
Was the lockwasher tab groove rusted? Seems like that area would be a lot worse if it was moisture from environment (which I agree shouldn't get into the fit). Perhaps if the groove is better than the rest might suggest fretting as JJP said. I don't recall much inner ring fretting (as we know they usually have enough interference to prevent motion). I did see one inner ring fit recently that was fretted in select portions. Particularly in the area immediately around the area exposed from an oversized lockwasher groove in shaft. Someone had filed down the burrs on the shaft around that groove and created shaft low spots in the process.... there was lots of fretting in/around this area and it was black/grey/mottled appearance. I haven't seen any fretting that I'd call rust-colored (brown/red) except outer ring fretting with a very loose fit in an outdoor environment.

I wonder if some unusual compound was used to help the heated bearing slide onto the shaft during initial assembly. Oil wouldn't be a problem but maybe someone used something else which either helped create corrosion or left a colored residue that has been mistaken for corrosion.




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

 

[electricpete]

Also what did the bearing ID's look like?

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

 

[Tmoose]

The rust wouldn't happen to resemble a fingerprint? If so, Mariska Hargitay could send it to the lab to be run through the Integrated Automated Fingerprint Identification System and get a name and picture of the tech at the alternator factory who is eating Cheetos on the job.

 

[spciesla]

Another vote for fretting corrosion. Was the shaft diameter verified (for the proper fit) prior to the last bearing replacement?

 

[AshleyK]

I've been doing some calculations for the fit.

Our journal tolerances are between 170.015 and 170.033

The bearing coefficient of expansion is 11.7x10^-6

Therefore, 170mm x 11.7x10^-6 x 90 degrees = 170.179mm

90 degrees is the temp change, from the ambient 25 to 115.

 

[JJPellin]

What material is the shaft? If it is carbon steel, the coefficient of expansion of the shaft should be very close to the expansion of the bearing. And the shaft temperature should be very close to the bearing temperature. We would use an ISO k5 fit for this bearing. We would allow up to 0.020 mm interference at the high end of the range. You could consider increasing your interference fit.

With a shaft this large, I assume that the rotor is quite heavy. Do you know the expected loads on these bearings from rotor weight or other sources?


Johnny Pellin

 

[AshleyK]

I'm not sure on the shaft material but I'll try and find out. As far as I know the rotor weighs 6t.

Of the shafts I've measured, they've all been near the top end of that range (170.033), and although I haven't measured any of the bearing bores I'm assuming their tolerances would be pretty tight, so our fit is probably higher than 0.02 anyway.

 

[Tmoose]

pictures of rust might be telling

 

[JJPellin]

I am sorry. When I try to convert between English and metric I tend to have problems. ISO k5 fit criteria would allow up to 0.045 mm interference. This is above your maximum allowable shaft diameter and so you do have some room to increase the interference fit. Sorry for the math error.

Johnny Pellin

 

[BearingDave]

Another vote for fretting corrosion.

Follow EASA or bearing manufacturer fit standards for the bearing size and type.

 

[electricpete]

Some questions
Got any pictures?
Discoloration only on shaft or on shaft and on bearing bore ?
If on bearing, what was appearance of bearing bore under the shaft lockwasher tab groove?
On the shaft, what is the appearance of the bottom of the shaft lockwasher tab groove?
Are we talking grey/black or red/brown?

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