Checking my bearings

[Les is more]

Working on my bike we felt appreciable play in the rear wheel. That combined with too much shiny stuff on the final drive drain plug, made us decide that after 110,000 miles, it might be a good time to replace the rear wheel bearing. Every thing looked spiffy when Jamie pulled the crown gear/wheel bearing, but a little preventive maintenance never hurt.

My question, for those who have actually held this bearing assembly in their hands is, how much movement should I expect to feel in the tapered roller bearing at the end of the crown gear. The outer race moves axially over the inner race by at least a couple of mm. Is this normal?

[RFW]

Working on my bike we felt appreciable play in the rear wheel. That combined with too much shiny stuff on the final drive drain plug, made us decide that after 110,000 miles, it might be a good time to replace the rear wheel bearing. Every thing looked spiffy when Jamie pulled the crown gear/wheel bearing, but a little preventive maintenance never hurt.

My question, for those who have actually held this bearing assembly in their hands is, how much movement should I expect to feel in the tapered roller bearing at the end of the crown gear. The outer race moves axially over the inner race by at least a couple of mm. Is this normal?

This is probably normal. Tapered roller bearings are hopelessly sloppy until installed. As you hold them in your hand, there is no pressure pushing the inner race against the outer race's taper. Depending on the design of the specific bearing, some just fall apart (the inner race together with the rollers just fall out of the outer race). Other types are just loosely held together. The need to be shimmed correctly and installed, at which point, the inner race is adjusted (by shimming) to JUST touch the outer race, with a specified preload. At this point, all slop is taken up.

 

Hope I understood your question correctly.

 

Bob.

[Stan Walker]

1st question - how sure are you that the play you felt at the rear wheel was caused by the final drive crown wheel bearing? Any chance it was the pivot bearings?

 

2nd question - is there any sign of damage to the crown wheel bearing? Clean it and look for race damage, roughness when spinning, etc. If no damage is found then it probably is/was good.

 

Bob is correct that the tapered roller bearing needs to be preloaded to work or measure correctly. So take the same approach as with the ball bearing and check for race damage and smooth rotation.

 

That roller bearing is pretty cheap as I recall, if you suspect it, replace it.

 

The crown wheel bearing isn't, at least not from BMW.

 

It will be interesting to see how the two of you set the bearing preload after this. I avoided doing that when I decided that the crown wheel bearing in my '02 was good to go and therefore didn't replace it after all. The maint book shows a special fixture that bolts everything together and allows you to access internal parts in order to measure and thereby determine the required shim size. Maybe your local dealer will lend it to you?

 

My thinking was that the replacement crown bearing is identical in size to the old bearing and that replacing it shouldn't change the spacer size needed. A good theory? But is it as true for the tapered roller bearing? How about when you replace both at the same time?

 

Once upon a time we talked about having parts available on an emergency basis. Tools are often even more important!!!

 

Good luck, let us know how it all turns out.

 

Stan

[ShovelStrokeEd]

As others have stated, you can't tell about a taper bearing until or unless it is already preloaded. The spacer on the crown wheel bearing side is tied to the housing and shaft dimensions and bearing specifications being what the are, very little variation in dimension from bearing to bearing, you are probably good to go with the spacer that came in there. I would venture it is pretty much the same with the taper bearing. The spacer seems to be used to adjust preload based on housing and shaft dimensions. Again, bearings, from one to another, maintain pretty tight tolerances. You should be good to go replacing either bearing using the spacers that you have.

 

Inspection should be done after a careful washing of the bearing in solvent and then use a good magnifier on the races and rollers. Anything other than smooth and shiny is cause for replacement.

 

Some precautions about removing/installing the races for those taper bearings. Fit in the housings or on the shaft is usually pretty tight. Judicious use of heat and cooling is necessary for this process and access to a press might be necessary as well. A hint for the inner race, a small electric deep fryer and some high flash point oil like peanut oil will quickly heat that inner race. For the outer, I use a freezer bag that I squirt full of Dust Off to get the humidity out and then stick it in the freezer. If you work fast, not too much frost will form on the race as you stick it into the housing which should have been in the oven at 200 or so degrees.

[CTCarl]

I don't know about the application but with regard to the bearings... A ball bearing has very tight ID and OD tolerances of tenths of thousands of an inch depending on diameters but the width tolerance can normally be up to 0.005".

Tapers are in the same range so it is important to check clearances and preload for a new bearing.

 

A word about heating a bearing. You only need to heat a bearing to 250F MAX to achieve ANY normal fitup. If you find you need more heat to get the expansion you need double check the fits. If the bearing becomes discolored during this process you have annealed, softened it, so trash the bearing and start with a new one.

 

Hope this helps.

[EricJRW]

There is article about this very subject in the April '06 MOA Owners News... Complete with pictures...

 

I don't have the magazine, but saw it at a club meeting... The article was discussing "Final Drive Failure"

 

Might be worth a look if you have access to a copy.

[RFW]

I don't know about the application but with regard to the bearings... A ball bearing has very tight ID and OD tolerances of tenths of thousands of an inch depending on diameters but the width tolerance can normally be up to 0.005".

Tapers are in the same range so it is important to check clearances and preload for a new bearing.

 

A word about heating a bearing. You only need to heat a bearing to 250F MAX to achieve ANY normal fitup. If you find you need more heat to get the expansion you need double check the fits. If the bearing becomes discolored during this process you have annealed, softened it, so trash the bearing and start with a new one.

I had a look at the NTN bearing site, and their info on tapered Roller bearings seems to confirm that tolerances on the width of a tapered roller bearing are in the vicinity of +0.1mm/-0.0mm for class 0 bearings, and about half that for class 6 bearings. This approximately fits with your comment of "up to 0.005".

 

My suspicion is that the width tolerance on the big ball bearing in there, is even more.

 

If I have read this correctly, this means that when replacing rear wheel bearings, you cannot just reuse the same shims. It will be necessary to check the crown clearance (to the pinion) as well as bearing preload, and adjsut as required.

 

Now I remember why I hated rebuilding differentials in cars

 

Bob.

[KMG_365]

1st question - how sure are you that the play you felt at the rear wheel was caused by the final drive crown wheel bearing? Any chance it was the pivot bearings?

We both felt the play at the rear wheel while doing the twist/wiggle--I thought the push-pull wiggle was solid. I've got a pair of bronze paralever bushings on the way from Rubber Chicken Racing to make sure, and I suppose I could just re-install the crown gear with the same shims and bearings to be sure. Frankly, I was almost hoping to find more obvious damage when I opened it up (not enough to crater the FD of course), at least enough to know I'd found the source of the "wiggle" and be confident I was solving the potential problem.

 

2nd question - is there any sign of damage to the crown wheel bearing? Clean it and look for race damage, roughness when spinning, etc. If no damage is found then it probably is/was good.

On casual inspection, the crown wheel bearing looks fine. If I use my imagination I think I can feel some friction when spinning the outer race by hand, but that small amount of play felt with the whole wheel/tire installed would be virtually undetectable on a 5" diameter bearing.

 

Bob is correct that the tapered roller bearing needs to be preloaded to work or measure correctly. So take the same approach as with the ball bearing and check for race damage and smooth rotation.

The outer race of the tapered roller bearing however wiggles freely twisting side to side and in and out. I'd have to invest in some more sensitive measuring devices to verify required shim thickness, and without removing/replacing any of the bearings or shims the replacement is pretty simple. Maybe I'll wait for the paralever bushings and reinstall it to verify that I've got bearing problems before breaking into it further. I saw tiny shiny flakes on the magnetic plug this time and that with the play and her long trips coming up led me to want to make it right.

 

That roller bearing is pretty cheap as I recall, if you suspect it, replace it.

That entails a whole 'nuther bit of anal retentive measuring I'd have to "gear" up for!

 

The crown wheel bearing isn't, at least not from BMW.

The crown bearing is about $125.00 from most dealers and as much as $365.00 from private suppliers. I guess BMW bought 30,000 of these in one go (and it's an odd size) to get them so cheap. To make matters worse, they've replaced the 19-ball bearing with a 17-ball bearing after looking into the numerous LT FD failures, so unless the tolerances on the two different bearings are identical it'll mean re-shimming.

 

It will be interesting to see how the two of you set the bearing preload after this. I avoided doing that when I decided that the crown wheel bearing in my '02 was good to go and therefore didn't replace it after all. The maint book shows a special fixture that bolts everything together and allows you to access internal parts in order to measure and thereby determine the required shim size. Maybe your local dealer will lend it to you?

Amazingly, after 110,000 miles both the crown and pinion gear have no visual signs of fretting--they look brand new! We have been pretty anal about keeping the FD oil fresh and clean. But yeah, reading up on all the FD stuff in the shop manual makes the EVO brake thing look like a stroll on the beach by comparison!

 

The manual states you need to re-shim when you replace the housing or the crown/pinion gear(s), but then later it states you also need to re-shim if you replace the crown (pinion too?) bearing as well. My only hope is to order the new C-B, pull the old C-B and measure them both to the nearest 0.005mm and swap if they measure out to the same. Otherwise, I suppose I could carry the guts in to SD BMW M/C's and have them confirm measurements for shims and just pay the bench time. Honestly, I don't see any obvious wear on the C-B and if the P-B is also just loose when it's not installed, I'm more inclined to replace the paralever bearings with bushings and call it good until it fails (it's covered under the extended warranty).

 

Of course it's Leslie's bike and the final call is hers. She's so pleased with how well the bike's performed over these many, many miles she's loathe to mess with it, too!

 

Thanks for all the info everyone!

 

(Now I've got to go dig that magazine out of my bag and finally read the MOA article!)

[Boffin]

Just thinking out loud here,

 

If you get new bearings and measure them old vs. new, you would know how much to vary the shims by, rounded to the nearest shim thickness - does anybody know what the increments are for the shims?

But then, you don't know how close to one end of the tollerance or the other the old him/bearing combination was....

 

 

 

Andy

[RFW]

The crown bearing is about $125.00 from most dealers and as much as $365.00 from private suppliers. I guess BMW bought 30,000 of these in one go (and it's an odd size) to get them so cheap.To make matters worse, they've replaced the 19-ball bearing with a 17-ball bearing after looking into the numerous LT FD failures, so unless the tolerances on the two different bearings are identical it'll mean re-shimming.

 

Looking at the standard bearing tolerances on the bearing's width, they are not very tight. As a result, I believe you will have to re-shim, no matter what.

 

As matter of interest, I was looking at the bearings used on my old K100, and both the large ball bearing, and the smaller taper roller, are standard bearings available from various manufacturers (for a hell of a lot cheaper than from BMW!). It is highly unlikely that BMW has designed their own bearings when there are so many standard ones available, so you might try to cross reference the number on the bearing.

 

It will be interesting to see how the two of you set the bearing preload after this.

It is not JUST preload!!

 

There are two simultaneous things to adjust by shimming (just to make things miserable!). Actually, three things if you include the pinion shimming, which luckily you don't have to!

 

The first is the bearing preload, but at the same time the shimming affects the play in the gears themselves.

 

Therefore, you not only need to know what the preload is supposed to be (and how to measure it), you also need to know how much gear tooth clearance is required.

 

This is why there are 2 sets of shims, not just one. The shim that goes behind the taper roller bearing, is what sets the gear tooth clearance. The second shim (the one behind the large ball bearing) is what sets the bearing preload. Once the first shim is selected to give the correct gear tooth clearance, then you don't touch that one any more, and then make adjustments only to the second shim to get the bearing preload right.

 

If it is anything like setting up an automotive differential, it is a tedious pain in the butt!

 

Bob.

[KMG_365]

If you get new bearings and measure them old vs. new, you would know how much to vary the shims by, rounded to the nearest shim thickness - does anybody know what the increments are for the shims?

Andy

That's a damn fine idea--I like it! Since I'm not replacing either the housing or the gears, the only variable(s) is(are) the bearing(s). Assuming the tapered roller bearing is not damaged to the point that the width has changed it should be a much easier task to just measure and account for any differential! If I can get the tolerances exact, it shouldn't matter too much on which end of the original tolerances under which the FD was originally set up. Obviously, the original settings did fine for 110,000 miles so they could not have been too far out of spec.

 

The question of accurate measuring though, is the bugger. According to the fiche, there are three different shims for shimming the various parts of the FD. There are dozens of part numbers covering the various shims which are sized in 0.05-0.1mm increments. I think I've got three different ones behind the crown bearing for example. The odds are good I could measure those and get an idea how accurate my equipment is, then measure the two bearings and extrapolate from there.

 

I'll do some more research and get back to you all!

[smiller]

Also thinking out loud... when you are done getting things as close as you can could you use some indicating dye to look at the tooth contact pattern? If the contact pattern looks good can one make a reasonable assumption that things are set up OK?

[Eckhard Grohe]

Measuring a bearing on the table or surface plate and installed will probably give you different results. It all depends on the internal geometry of the bearing. It would be best to check the clearance as BMW suggests.

 

The good news is that the shimming you have is with in +/- 0.005 inches of what you need because you already have the shims to compensate for the manufacturing tolerances of the other parts.

 

Probably easiest to bring the parts to a shop and let them do it or get a shop to rent the tool to you.

[smiller]

Probably easiest to bring the parts to a shop and let them do it or get a shop to rent the tool to you.

That is really at the heart of my question. Since your shim packs will already be close all you will likely be doing is making a small adjustment due to the bearing replacement. If this is the case, rather than worrying about special tools for measurements or having to involve outside help, couldn't you just go by the tooth contact pattern and use that as a guide to make any small adjustments?

[Stan Walker]

I wonder just how much that fixture costs?

 

Stan

[RFW]

If you get new bearings and measure them old vs. new, you would know how much to vary the shims by, rounded to the nearest shim thickness - does anybody know what the increments are for the shims?

Andy

That's a damn fine idea--I like it!

Unfortunately, it may not work.

 

The reason for replacing the bearings is that they are worn. Thus, measuring their width will result in an erroneous number.

 

There are 2 different shims if your bike is anything like my K100. The shim that is behind the taper roller bearing is used to adjust the gear backlash only.

 

The second shim, that is next to the large ball bearing, is used to adjust bearing preload. This one is adjusted LAST, once the smaller shim has been finalized.

 

The small (roller bearing) shim is available in steps of 0.05mm thickness (at least it is on the K100). The larger (ball bearing) shim is available in coarser steps of 0.1mm.

 

I have not found out yet what the required gear backlash is supposed to be, but the bearing preload is supposed to be anywhere from 0.00mm (no preload and no play) to 0.10mm preload. This is consistent with the 0.10mm available steps of thickness that the larger shims are available in.

 

It is also clear from the smaller 0.05mm steps of thickness that the smaller shims are availave in, that gear backlash is relatively more critical.

 

Once the shims are selected, it is a good idea to put some mechanic's blueing on the hear teeth, assemble the thing, spin the gears, then disassemble it to check that the tooth contact patches are centralized on the gear teeth.

 

Lots of fun!

 

Bob.

[RFW]

does anybody know what the increments are for the shims?

I thought I posted this earlier, but it seems to have disappeared.

 

I checked the MAX BMW parts fische, and the available shim thicknesses for the R1150RT are the same as for my K100 I referred to in an earlier post.

 

The shim that is behind the taper roller bearing (that adjusts gear backlash) is available in 0.05mm thickness increments.

 

The larger shim that is next to the big ball bearing (that sets the bearing preload), is available in 0.10mm thickness increments. This nicely corresponds to the requirement that the preload can be between 0.00mm and 0.10mm.

 

It appears clear that the requirement for correct gear backlash (apparently within 0.05mm) is "tighter" than the bearing-to-bearing thickness tolerance, hence it is likely that when changing the taper roller, a different shim will be needed.

 

Bob.

[KMG_365]

Got the brass Paralever bushings from Rubber Chicken Racing Garage in TWO days! They look sweet! I know those paralever needle bearings don't last for crap and remember Leslie's falling to pieces when they were removed for a leaking seal that fouled her clutch at only 38K miles or so. These should be a sweet--and permanent solution! I spoke with Tom Cutter from RCRG and he seemed like a real nice guy and straight shooter.

 

I also had a bitch of a time finding a 13mm Hex Key anywhere locally. I just Googled "13mm Hex Key" YESTERDAY and came up with one for 4.95 from Tool Gopher through Amazon.com and since I've ordered from Amazon before it took all of 30 seconds to order a few--the box came in TODAY'S mail!

 

I got a set of bearing pullers from Harbor Freight (cheap! ) yesterday as well as a set of drifts and seal pushers and a dead-blow plastic hammer. The trick will be measuring the tapered roller bearing with my micrometer (which only reads in inches! ) as the two races are off-set and the measuring faces of the micrometer are not that large--hmmmmm.

 

I'm riding to Phoenix tomorrow to pick up my new (to me) PowerBook laptop and then we're going to Steves1150's BBQ on Saturday, but maybe I'll be able to post about my measurements by Sunday or so. It looks we'll be living in the guest room a bit longer . . . .

[KMG_365]

Okay, I found that the two wheel bearings are only ~ 4-5/10,000ths of an inch different--well within the 0.1 mm tolerance. I'm replacing the paralever needle bearings with bushings and just replacing the crown bearing (leaving the pinion bearing alone as it and the inner race looked fine). I've got the crown gear and the bushings in the freezer now. I'll pop the FD and the wheel bearing in the oven in the morning and see if I can get it all buttoned back up.

 

I also found out that the hex key needed for the paralever bearing stud bolts is a 12mm--NOT a 13mm!

I ended up finding one at the AutoZone, but this is taking me far too long! I finally tracked down some 18" zip ties and I've got to pop in the bearings and properly phase the driveshaft on reassembly and I should be good to test it out!

 

Thanks everyone for your input!

[steveknapp]

couldn't you just go by the tooth contact pattern and use that as a guide to make any small adjustments?

 

That's the procedure I've always seen for automotive diffs. If nothing else, it would be a good way to verify things are OK.

[KMG_365]

couldn't you just go by the tooth contact pattern and use that as a guide to make any small adjustments?

 

That's the procedure I've always seen for automotive diffs. If nothing else, it would be a good way to verify things are OK.

If the pinion bearing had shown any wear, that would've been necessary. I did actually buy another pinion bearing to have on hand just in case (I've still got to do mine ). As it was, the old and new crown bearings measured out almost identical, so I just re-used the old shims and buttoned it back up with the new bronze paralever bushings--SWEET!!!!!

 

We did Gregori's First fRide Day yesterday and Leslie got a chance to put it through its paces. She said it worked like a charm and felt solid as a rock. Previously, she was really noticing the back end moving around and a feeling of having a hinge in the middle of the bike. We went for about a 120 mile mostly twisties ride and that combined with two new tires and she's happy as a clam at high tide!

[DEF]

I'm gonna take a SWAG here and vote that the bushings correct the slight clearance you have observed, felt, heard. I have observed the same on my '01GS at 30,000 miles. In order to insure that it is excess FD bearing lash, you need a magnetic base dial indicator applied to the housing while you attempt to rock the hub to detect any excess play.

 

Please tell us what you finally discover...I'm betting Paralever bearings....and, those Chicken Fat Racing bushings look like the hot setup...