INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Jobs

Bearing Race Fit

Bearing Race Fit

(OP)
Question: I have a Porsche transmission pinion bearing race going into an aluminum housing. On top of this roller style bearing is a four point ball bearing and then a steel retainer plate. The problem with these transmissions is that the pinion race creeps over time and sooner the pinion race becomes real loose. Mine had .002 clearance. Some have quite a bit more and can wobble in the bore. The common fix is to machine the housing and press fit a sleeve and then the race. This takes a lot of meat away from the upper bearing, the main shaft bearing. The other alternative is to electro-plate the OD race and then sneak up on a clearance, which is the route I've taken because it was cheap (nickle plating). My question is fit. My housing is 3.151". My bearing was 3.149" OD and is now 3.154" OD. This bearing is FAG, special bearing, impossible to get information, $600 list ala Porsche -- basically a black hole. I believe this bearing takes the radial load from the pinion and the other takes the axial. The bearing race is .904" wide. The other smaller main shaft bearing appears a tight press fit, and I am not touching it to see how tight. The pinion bearing race (to the best of my knowledge) is not a slip fit by design because the factory manual says to heat up the aluminum to 120c to get it out. I think that I have covered all the information needed for the pros to give some advice on the matter. Any info on fit is appreciated.   

RE: Bearing Race Fit

The .002" clearance is just about right for Loctite to set up nicely.

 

Mike Halloran
Pembroke Pines, FL, USA

RE: Bearing Race Fit

.003" isn't much of an interference fit for a 3" diameter steel race into aluminum alloy. As somewhat of a reference, the suggested 100 degree C rise (from ambient, for race removal) would expand the aluminum bore diameter by about .006" [that's from memory- somebody check that, please]. And from hands-on experience, even .006" interference is none-too-tight if the race is of fairly robust dimensions (you'll need to crunch the numbers to assure that the race ID isn't affected by the press fit).

Can you add more plating without concern for trueness of of the OD?

RE: Bearing Race Fit

(OP)
Thanks for the replies guys. I've thought of the Loctite 638 method on the bearing OD cavity but this has proven ineffective on various re-builds of this style transmission. Some of them are just too loose. This bearing has openings on both sides, one for the pinion/ring gear and the other for the shaft. It has two opposing ledges in the aluminum bore to keep it in place by the gear teeth area and, as I said, a steel retaining plate on the other side. The plating was a big gamble with thickness, so I had to wait until I got it back to measure. The inner race was not plated. My concern was with bearing fit research where most tech articles explain that bearings have an interference fit on the rotating ring and a slip fit on the stationary ring. I even thought the loose race syndrome in this transmission to maybe be a design feature -- but that does not make much sense on a pinion shaft were backlash must be maintained with a ring gear. I've built diffs with tapered bearings (this bearing is not tapered) and crush sleeves and when snugged up, those bearings are locked tightly. The 004" increase is almost .005" because it is .0049", almost five thou. The plating may be about as close as I can get because if I re-plate, I would have to turn down the OD to true it up, and I come up blank with any machine shop who is capable of doing this. Bearing rebuild companies blow you off else want a fortune to put it on their OD bearing machines.   

RE: Bearing Race Fit

zcoker,

The pinion gear bearing mounting you describe (cylindrical roller and split race ball bearing) sounds like the one in the drawing below.  The roller bearing takes radial loads and the split race ball bearing takes thrust loads in both directions along with some radial load.

The fit between steel bearing races and aluminum housings is always problematic.  Unless the bearing race has an adequate interference fit at operating temperatures then the race will creep/spin in its bore causing excessive wear.  If there is too much interference fit at RT, then there is the possibility that the bearing will lose its internal clearance at low temps due to CTE mismatch between the race and housing, potentially causing seizure of the bearing.  Thus fits between steel races and aluminum housings must be carefully considered.

Since your pinion gear roller bearing is subject to fairly high radial loads, and these radial loads will change direction as the car accelerates or downshifts, I'd recommend sleeving the housing with a similar material and re-boring to the factory specs.  Unfortunately, this would also be the most costly solution.

Using an anaerobic adhesive (like Loctite) to "bond" the loose fitting race into place would not likely be a reliable long-term solution in this particular instance.  Loctite only has a shear strength of about 3 or 4 ksi and does not work well as an adhesive.  Instead, the most effective use of Loctite is to increase the coefficient of friction at clamped joint interfaces.



Hope that helps.
Terry

RE: Bearing Race Fit

(OP)
Terry,

You nailed it with that drawing, and I thank you for the detailed reply. You also know exactly how the bearings function. Both the bearings on the pinion shaft are clamped down with a steel retaining plate which also clamps down the same bearing setup (though smaller) on the upper main shaft. The problem with the steel insert and re-bore, it takes away the meat between both main shaft and pinion shaft. This is what I am trying to avoid. I have the race plated over-sized and can sneak it down to "a size", so I presumed, naturally, that there was some standard to go by, at least for transmissions. The factory calls for a temp of 120c to get it out, so that can be .002 - .003", theoretically. A temperature number to get it out is all I can get out of the factory manual, no bearing fit specs whatsoever.  

RE: Bearing Race Fit

".002 - .003" theoretically"? Are you saying my quick rough calculation was way off?

RE: Bearing Race Fit

If your worried about strength, do a rework on all the bores. Use a stronger aluminum as replacement and weld it in, then rebore.

RE: Bearing Race Fit

Quote:



If your worried about strength, do a rework on all the bores. Use a stronger aluminum as replacement and weld it in, then rebore.


That is about the last thing I would consider doing.

The aluminium between the bores is only under minimal stress from spliting as the main load is compression from end thrust and tension as the shafts try to spread apart.

The most likely mode of failure is the end thrust.

The shafts spreading won't really load the area between the shafts, but will load the whole bulkhead.

The steel plate will take some of that load. Is there room to substantially increase it's thickness and maybe machine off some of the face it bolts to so some of the bearing or the steel sleeve if used can locate in the plate.

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Bearing Race Fit

patprimmer is correct regarding the "septum" area between the bearing bores.  The opposing gear forces will mostly cause the two roller bearings to separate, and the septum area will be lightly stressed.  The only issue would be if the septum gets excessively thin it might yield during the thermal fitting of the races, but that would be an extreme case.

I would stay away from welding except as a last resort, due to potential thermal distortions.  However, there are companies that specialize in thermal spraying metals to salvage slightly worn component surfaces exactly like yours. Thermal spraying doesn't put as much heat into the part as welding, so distortion is minimized.

http://www.flamespray.us/

If there is concern about the thickness of the septum with over-boring/sleeving/re-boring, and the existing bore is only a couple thou' oversize, you also might be able to find someone that can lightly knurl the bore surface.  This would raise the surface slightly and then it could be align bored/honed back to spec.  The knurled & honed bore surface would probably even give slightly better wear than the original aluminum surface due to the mechanical work hardening effect from the knurling.

Good luck,
Terry


 

RE: Bearing Race Fit

Metal spraying is certainly a good idea.

Knurling will work harden the surface, but also reduce contact area, so I don;t know how that might net out.

If this was a recuring problem, I would investigate the possibility of boring to take sleeves but not using them, and removing the septum area entirely then making a steel insert to slip in. That way the separating force between the shafts would be restrained by steel and the bearings would be held by steel. A lot depends on how much metal there is to work with.

Of course if this is a repair to an old car that lasted years and will be returned to similar duty, spraying sounds the go to me.

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Bearing Race Fit

patprimmer,

All of these ideas sound great unless you're the one that has to pay the bill!

Terry

RE: Bearing Race Fit

Yes I know.

It all depends on how seriously you need to do the repair and fix the problem.

If it was my DD then I probably would have slapped some locktite on it.

If it was my 1912 model that broke the same part AGAIN then serious fabrication to improve it would be in order. Same for some race cars I work on.

I don't recall the OP stating if it was for normal duty use or being pushed past design limits.

Regards
Pat
See FAQ731-376: Eng-Tips.com Forum Policies for tips on use of eng-tips by professional engineers &
http://eng-tips.com/market.cfm
for site rules
 

RE: Bearing Race Fit

If I were paying the bills, I'd try to catch it before the housing got wallowed out to more than .002" clearance, and I'd use Loctite (and the appropriate cleaner and primer).

Once the clearance gets worse than that, chances are the housing hole is not round anymore, and not in the right location either, so I'd be looking at boring the housing on the correct center and installing a thin bronze sleeve, and boring that to size.

Given that the housing appears to have zero extra meat, one might face the prospect of buying a new housing.  If that happened, I'd take a hard look at grinding some half-round notches in the bearing race OD, broaching mathing notches in the housing bore and installing some Dutch pins to keep the bearing from rotating.



 

Mike Halloran
Pembroke Pines, FL, USA

RE: Bearing Race Fit

For that, I'd clamp it in a lathe over a mandrel, and lap the nickel plate to size.
 

Mike Halloran
Pembroke Pines, FL, USA

RE: Bearing Race Fit

Hello... hello... someone please tell me whether my posts are visible to anyone other than myself.

RE: Bearing Race Fit

Pat- You've confirmed my worst fears- that I'm invisible and voiceless...

RE: Bearing Race Fit

That old bearing tech publication "Ask Walt" had a picture of a bearing that had reportedly been run in a housing that had been "repaired" by knurling.  The knurled pattern had shivered and fretted right into the race, most likely because the forces that made the micromotions or creeping and wore out the housing the first time were opposed by about 1/10 as much metal after knurling. The accuracy (location, roundness and circularity) of the surface provided by knurling if "sized" by pressing in the bearing in is pretty poor too.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources


Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close