Re: Wheel bearing replacement

Hi,

Thanks for the note! I saw the same site, and noted that warning NOT to grease the bearings... I think I'm one of those "guilty" of giving flack about greasing bearings, but this is the first time I've seen this warning. Guess I learn something new every day! ;) Apologies for the flack!

Anyway, I kinda wonder if Subaru's insistence on using ball bearings has anything to do with premature failures. It seems roller bearings have pretty much taken over the automotive world, at least for hubs, and it also seems Subaru's answer to some of the problems has been to install tapered rollers. Years ago in my VW Bug days, some of my cars had ball bearing hubs, others roller bearing hubs, and it seems life expectancy was greatly improved with the roller units. Maybe Subaru has some interest in a ball bearing manufacturer?

The bulletin mentions the bearing housing (hub?) roundness being compromised by a press as being a cause of problems, then gives a tolerance figure of 0.2mm IIRC. This seems to give virtually no room for error, and I wonder if the bearings being pressed into at the factory are actually causing some of the warpage problem before the car is ever sold? This could cause a preload, which I understand is necessary with rollers but deadly to balls. That's kinda the impression I got from the bulletin. I doubt any kind of "super" grease will offset that, but we'll see?

Anyway, hope this new procedure helps alleviate the problems!

Rick

recommendation.

Yeah - I think Subaru has favored ball bearings over the years - I know my '86 had ball bearings on all four wheels.

Probably ball bearings have less rolling resistance, but the pressure (pounds per square inch) on the contact surface of the balls vs. that of the rollers is much greater, and so it is probably much harder to design in the same safety factor in a ball bearing vs. roller bearing of the same physical size. Perhaps Subaru compromised the design a little too much (gave up some durability) in order to gain a precious couple of tenths in mpg.

Bill Putney (to reply by e-mail, replace the last letter of the alphabet in my address with "x")

Bill,

Does it seem a little haphazard as to where they choose to use balls over rollers? My '90 has rollers in the rear, balls in the front. I agree with Eddie H that balls should be loaded perpendicular to the axle and rollers will take some side loads better, so the arrangement in my car seems 180 degrees bassackward to me (IF you're not going to simply use rollers at all four corners.) Now it seems they're going back to balls in the rear from the factory but replacing them with rollers when they fail... what are they using in the front? It's got me confused!

Rick

As far as Subarus, I've never worked on anything later than the '86 I used to own, so I can't say.

Balls of course can take some side loading (the bearing design manuals have charts of such stuff, and the race design can be tweaked a little to improve the side loading capability), and you must admit that the overwhelming majority of the time, the loading is perp. to the axle, and that's where the failure occurs due to brinelling (a fatigue type of phenomenon) in spite of periodic side loading. The bearing design books have charts about time-and-speed-at-load to failure for different directions of loading, and, again, I would think the failure even for a ball bearing on a car is from the radial (vs. side/thrust) loading simply because of the 99% time duration in that mode.

I think, in a car, it goes back to the pounds per square inch on a microscopic scale on the ball and race contact points rather than the radial vs. side loading, even for ball bearings.

Bill Putney (to reply by e-mail, replace the last letter of the alphabet in my address with "x")