BMW Luxury Touring Community banner

1 - 20 of 25 Posts

·
Registered
Joined
·
2,176 Posts
Discussion Starter #1
Hey guys,
I feel like I've been monopolizing this forum with my tear down, and I do apologize for that, but I just found something that opened my eyes. Here's a few pictures of my crown bearing in the final drive I just pulled. The PO (Goldendragon) told me that the FD was rebuilt at just under 36K miles as a precautionary measure. The bike now has 68K.
I've been doing the 6 & 9 rock and rolling the wheel while feeling and listening, and would have bet I was OK. Then I opened it up. I PM'd Curtis and told him that I would check the shim that is in there now to see if it was done correctly, but it'll take me a while to get a rig built to handle that.
What's scary to me is that, had I not found the main seal leak, I would likely have taken off on my trip scheduled next month, to the tune of about 3,000 miles. Wonder how far I'd have made it.
I am now totally convinced that I'll be doing a precautionary rebuild of the FD at least every 30,000, if not sooner.
Hope this helps someone else that's on the fence about rebuilding.
Frank
 

Attachments

·
Premium Member
Joined
·
1,345 Posts
Is the bearing OK? And if not, what do you believe is wrong with it? A crown bearing will never show any wheel wobble. By the time it gets bad enough to wobble, your bike will be making so much noise a deaf man would be afraid to ride it................ ;)
 

·
Premium Member
Joined
·
1,345 Posts
Doh! I just enlarged the picture. The only way you might have known anything at this point is if you spun the wheel with both plugs out of it. It would have made some noise if the races were scarred............. :D
 

·
Registered
Joined
·
2,176 Posts
Discussion Starter #4
James, that's what was odd. When I pulled it, it was covered in oil, and was super smooth and quiet. I just dunked the whole thing in my parts cleaner, and after blowing it off, I spun it again. Still smooth, and still quiet. It was just something that caught my eye that made me look closer, and what you now see is what I saw. I don't believe that you could ever detect this without opening it up and actually seeing it.
 

·
Wrencher Extraordinaire
2005 K1200LT
Joined
·
14,171 Posts
fpmlt said:
I am now totally convinced that I'll be doing a precautionary rebuild of the FD at least every 30,000, if not sooner.
Hope this helps someone else that's on the fence about rebuilding.
Frank
By all means rebuild it but if you shim it correctly you won't have to touch it again. I think every one with and ealy bike that has NOT failed yet do a preemptive preload check. Curtis has found that several were shimmed in excess by 0.80mm. I am convinced (until substantitave other evidence surfaces) that excessive preload is the cause of early failure and that the design is sound.
 

·
Registered
Joined
·
2,176 Posts
Discussion Starter #6
Makes sense John, but let me ask this: has anyone ever rebuilt one, know to be correctly shimmed, just to look inside, then reassemble without changing bearings?
 

·
Registered
Joined
·
989 Posts
fpmlt said:
Makes sense John, but let me ask this: has anyone ever rebuilt one, know to be correctly shimmed, just to look inside, then reassemble without changing bearings?
That is true - I doubt that anyone would open the drive just to check it, without having a replacement bearing at hand.

Still, I agree with John and with Curtis: I am convinced that the "preventative" rebuilds only delayed the inevitable. That is, if the shimming was wrong and it was killing the bearing, just swapping the bearing started the destruction cycle again.

On the other hand, a properly shimmed bearing should last practically forever - after all, there are the 96% of drives that happily go about their business...

From reading the K12LT posts for many, many years, I came to conclusion that most of the "preventative" rebuilds as well as "repairs" involved only replacement of the bearing, without correcting the underlying shimming problem. Needless to say, these bearings are due to fail again.
 

·
Registered
Joined
·
4,404 Posts
Hi Frank,

I sent my PM to you before seeing the pics posted here. You clearly have a broken bearing retainer.

The retainer sits very close to the seal. Since you caught it early the retainer hasn't torn the seal the shreds. But... I'll bet if you closely inspect the inside of the seal, you'll see where the broken retainer has been tearing at the seal. If not, I'd still wager your oil leak is caused by the broken retainer distorting the oil seal causing the leak.

Measure your existing shim and compare it to what your new bearing measures for preload requirement. I'll bet you a beer and a burger that the existing shim is thicker than what a careful measurement for the new bearing indicates.

Also, I bet if you cut that failed bearing open and inspect the balls and races closely, you'll see degradation of the races somewhere, and maybe on the balls as well.

Your pics are of a classic crownwheel bearing failure, caught early. I'd speculate that you weren't gonna get more that 100 miles from home with that, more like less that 50.

PS I will say again here for the benefit of the group, checking for wheel play will not reveal a bearing problem. It has been posted in the past as something to check for, and rear wheel play will indicate a problem like loose swing arm or FD pivot bearings, BUT will NOT indicate an impending crownwheel bearing failure; the evidence is pretty good that the bearing will remain quite stable even after retainer fracture. You have to ride a leaking, grinding FD a while before the wheel starts to wobble. You will get oil leakage, rough ride from a failed crownwheel bearing long before any detectable play occurs at the rear wheel. Roughness during wheel rotation (get the brake pads away from the rotor), and metal shards on the FD drain plug are the only two indicators of impending failure that I know of, and these don't come early. Roughness during wheel rotation and/or metal shards on the drain plug indicate things are gonna be bad pretty soon.


fpmlt said:
Hey guys,
I feel like I've been monopolizing this forum with my tear down, and I do apologize for that, but I just found something that opened my eyes. Here's a few pictures of my crown bearing in the final drive I just pulled. The PO (Goldendragon) told me that the FD was rebuilt at just under 36K miles as a precautionary measure. The bike now has 68K.
I've been doing the 6 & 9 rock and rolling the wheel while feeling and listening, and would have bet I was OK. Then I opened it up. I PM'd Curtis and told him that I would check the shim that is in there now to see if it was done correctly, but it'll take me a while to get a rig built to handle that.
What's scary to me is that, had I not found the main seal leak, I would likely have taken off on my trip scheduled next month, to the tune of about 3,000 miles. Wonder how far I'd have made it.
I am now totally convinced that I'll be doing a precautionary rebuild of the FD at least every 30,000, if not sooner.
Hope this helps someone else that's on the fence about rebuilding.
Frank
 

·
Registered
Joined
·
4,404 Posts
fpmlt said:
Makes sense John, but let me ask this: has anyone ever rebuilt one, know to be correctly shimmed, just to look inside, then reassemble without changing bearings?
Well, it is impossible to know if it is correctly shimmed without opening and checking it. I think what you may have meant to ask, is has anyone opened and inspected a drive that had nothing known to be wrong with it, no history of failure, and just put it back together?

The answer to your question is yes. Before I learned of DMAN's dial micrometer method for calculating shim thickness, I opened my final drive (several years ago) just to look inside. I just bolted it back together.

Later, after DMAN posted his technique, I did a preemptive rebuild on my 2000 and replaced the bearing, about 40K miles ago.

A few weeks ago I re-opened the drive, rechecked the measurements, inspected the bearing, and reassembled without changing anything.

I had a couple of motives for doing this. At the time I rebuilt the drive 40K miles ago, I had only DMAN's dial indicator technique for measuring for shim thickness. Since then, I have fabricated a "BMW special tool" to allow for using the BMW service manual technique using a depth micrometer to measure for shim thickness.
And my other motive, I just wanted to inspect the bearing.

When I compared the BMW Service Manual technique to my previous calculations using DMAN's dial indictor method 40Kmiles ago, I got exactly the same results. I felt no reason to change anything. I'm gonna ride that FD to Alaska.

I have also done a couple of "open and inspects" of other people's FD where a "good" FD of unknown history was checked and reassembled. IIRC these were FD that folks bought as "spares" and wanted to know if they were set up well.

I am sure there are others out there who have done the same thing.
 

·
Registered
Joined
·
2,176 Posts
Discussion Starter #10
Thanks Curtis,
I got your PM, but for the masses, I think you answered my next question. I would have asked if you thought there were any dynamic forces taking place that would change the preload measurement.
In other words, if I go in and check my old bearing for shim thickness now, and compare the required amount to what was in there, would that measurement likely be the same as when rebuilt 30+K ago?
After watching your video several times, I'm going to build a stand for the FD and the mic., and measure what I have before I pull the bearing. I'll let you know what I find. I'll also have to drink several more cups of coffee before venturing out to see if my seal took a hit :).
Thanks again to all
Frank
 

·
Registered
Joined
·
4,404 Posts
Most of the failed bearings I have seen were way beyond any level of integrity that would allow for remeasurement. Your's is pretty well intact, so you might measure it before you pull it off the crownwheel hub. That will also give you a "practice run" before you do measurements for your new bearing. Given that your bearing has failed, I would expect there are some changes that have occured and measurements taken today won't likely be exactly the same as they were when the bearing was new. As to whether a functioning bearing, without any signs of failure, would measure differently after 30-40K miles of running, I would expect not; but that is a question best answered by a mech. engineer/bearing expert.

I will say that the bearing I put in 40K miles ago, and remeasured just a few weeks ago was "spot on" in terms of preload. That is to say, I calculated preload shim thickness when I rebuilt it, ran it for 40K, and checked it again. No change. And the lube has always been clear, without ANY mud, metal, gray fuzz, during lube changes. After 40K it was as clean in there as it was the day I put it together 40K ago.

From what mech. engineers have told me, the tolerances of bearings from the factory are very tight, so the amout of difference in dimensions and axial play will be small compared to the shim thickness and preload range we are looking at when we set things up. So the assumption is, and I emphasize "assumption", that comparing the old shim, even when we can't measure the failed bearing, to the shim thickness measured for the new bearing will give an idea if the original shimming was incorrect.

Have fun,



fpmlt said:
Thanks Curtis,
I got your PM, but for the masses, I think you answered my next question. I would have asked if you thought there were any dynamic forces taking place that would change the preload measurement.
In other words, if I go in and check my old bearing for shim thickness now, and compare the required amount to what was in there, would that measurement likely be the same as when rebuilt 30+K ago?
After watching your video several times, I'm going to build a stand for the FD and the mic., and measure what I have before I pull the bearing. I'll let you know what I find. I'll also have to drink several more cups of coffee before venturing out to see if my seal took a hit :).
Thanks again to all
Frank
 

·
Registered
Joined
·
3 Posts
Hi, I'm new here but not to other forums, mostly K-Bikes and ADVRider. I have been following a lot of threads on RD failures, but yours related so much to something else I just saw today I had to sign on here also.

There is a thread going on ADVrider that somehow degenerated......but this came out of it and finally put that thread back on track.

I'll steal a few words out of it:

bmw k1200rs 1998 19700mi. oil leak, final drive. metal chips, bearing cage broken. bought 1999 bmw k1200rs FD unit with 92997mi. with 19 ball bearing still intact. cut top and sides; replaced with plexiglass. with highspeed flim 300 frames per sec. I placed red dye #3 in cage ball area. spun drive and struck wheel to simulate bump, all load was absorbed by cage twice once at impact and at 90 degrees later in spin direction. after 144 impacts the 92,997mi. cage failed in the same way as the 19,700mi. cage did. same test new 17 ball unit failed after 661 impacts in the same way as the 19 ball bearings. stay away from potholes.

http://www.advrider.com/forums/showthread.php?t=573835&page=16

There is more, starts at post 230, most of the previous posts are arguments over % of failures...hence my interest in the title of your thread, and the pic of the failed cage. :thumb:

Interesting to see that someone tested a rear drive and made it fail...at the bearing cage.Hopefully it won't happen to me until I can do some preemptive maintenance, until then I will carry a bearing and seal on long trips....just in case but thanks to you guys, I know now how to rebuild my rear drive. :thumb: :bmw:
 

·
Registered
Joined
·
4,404 Posts
Hmmm.... interesting, I want to know more but his posts raise more questions than I'm prepared to address here.
Here's a guy who reports putting a window in a final drive, making a test device out of a cement mixer and various other components, films it at 300 frames per second, but doesn't post a pic of the setup when asked for it?
I try to be non-judgmental on posts like this; I'd like to see more as I am curious if there is good info to be had. I try to glean good information where I can find it, but I also have a healthy sense of scepticism, and I can't help but wonder if this guy isn't blowing smoke and barking up the wrong tree. Until I see a more methodical description by the guy, and pics of the setup, I'm suspecting smoke and mirrors as likely as good data.
And if he actually did the tests he describes, I'd want to know how the FD was set up before the tests with respect to preload. I'd also want to know what his window cut into the top and side of the FD housing did to the integrity of the drive.
And I'd want to see the bearing races and balls. I have never seen a crown wheel bearing with a broken retainer that didn't have spalling of the balls and races. And having cut open quite a few failed bearings, I can say that the evidence is pretty good that race degradation precedes retainer failure.
So if his theory that impacts (potholes) on the FD are the cause of stresses on the retainer is correct, then I believe that damage to the races is also involved in failure process.
And since a great percentage of these drives have not failed, you'd think there'd be more failures if bearing design and potholes were the primary cause of failure.
And, since all of the failed FDs that I have rebuilt have been measured to require less preload than what was originally in there, I still strongly suspect that excess preload is playing a significant role in failure.

If this guy has done the stress tests that he describes, I believe that he is seeing part of the picture, but not the whole picture with respect to the failure process; that adding "pothole shock" to a drive with excess bearing preload would accelerate the rate of failure makes sense to me.




H96669 said:
Hi, I'm new here but not to other forums, mostly K-Bikes and ADVRider. I have been following a lot of threads on RD failures, but yours related so much to something else I just saw today I had to sign on here also.

There is a thread going on ADVrider that somehow degenerated......but this came out of it and finally put that thread back on track.

I'll steal a few words out of it:

bmw k1200rs 1998 19700mi. oil leak, final drive. metal chips, bearing cage broken. bought 1999 bmw k1200rs FD unit with 92997mi. with 19 ball bearing still intact. cut top and sides; replaced with plexiglass. with highspeed flim 300 frames per sec. I placed red dye #3 in cage ball area. spun drive and struck wheel to simulate bump, all load was absorbed by cage twice once at impact and at 90 degrees later in spin direction. after 144 impacts the 92,997mi. cage failed in the same way as the 19,700mi. cage did. same test new 17 ball unit failed after 661 impacts in the same way as the 19 ball bearings. stay away from potholes.

http://www.advrider.com/forums/showthread.php?t=573835&page=16

There is more, starts at post 230, most of the previous posts are arguments over % of failures...hence my interest in the title of your thread, and the pic of the failed cage. :thumb:

Interesting to see that someone tested a rear drive and made it fail...at the bearing cage.Hopefully it won't happen to me until I can do some preemptive maintenance, until then I will carry a bearing and seal on long trips....just in case but thanks to you guys, I know now how to rebuild my rear drive. :thumb: :bmw:
 

·
Registered
Joined
·
2,176 Posts
Discussion Starter #14
Would one then also need to consider: tire pressure, load, shock type and setting, and even speed?
 

·
Premium Member
Joined
·
3,108 Posts
Curtis,
Have you ever seen a failed bearing where the retainer cage wasn't damaged? I'm thinking we could be looking at photos of a "smoking gun". Too much shim, too little shim..... neither one matters if pieces of the retainer cage randomly come loose and get into the system.
 

·
Premium Member
Joined
·
3,121 Posts
With respect to the question raised have you seen a failed drive without the retainer damaged I too am curious.

I suspect the only way that could be is if it spun?

Or you inspected it with a jewelers loop (microscope) and saw pitting on the balls or race themselves? I would think that once the balls start pitting the retainer is not too far behind?

Either way I think the shimming is the secret. After reading your recent post on how much difference you are finding from OEM shim to your measurement, and your shim.

I mean there is much ado about the FD, but if 96% is close or even 90% then it is a sound design, just that one Eastern Block sonnabitch that can not read the shim sizes correctly, or was truly getting even :D
 

·
Registered
Joined
·
4,404 Posts
deanwoolsey said:
Curtis,
Have you ever seen a failed bearing where the retainer cage wasn't damaged? I'm thinking we could be looking at photos of a "smoking gun". Too much shim, too little shim..... neither one matters if pieces of the retainer cage randomly come loose and get into the system.
Answer: Yes.
Here's one.
History: rider noticed rough ride. Stopped riding.
I removed the bearing, cut it open to inspect.
Pitting on both inner and outer races.
Retainer ring intact, I had to cut the retainer to get it off the inner race after having removed the outer race.

Clearly in this case races were degrading before the retainer gave up.
I have never seen a broken retainer where the races weren't damaged.
I have seen damaged races where the retainer was intact.
Conclusion: It ain't primarily a retainer problem.
(This conclusion has been arrived at some time ago. )
 

Attachments

·
Registered
Joined
·
4,404 Posts
fpmlt said:
Would one then also need to consider: tire pressure, load, shock type and setting, and even speed?
In my opinion, no.
I tried to be kind in my evaluation of the posts on the ADVrider site by the guy who reports having done a cement mixer stress test of a K1200rs final drive; this will be a little more direct:
Here's the salient part of this guy's post: "....cut top and sides; replaced with plexiglass. with highspeed flim 300 frames per sec. I placed red dye #3 in cage ball area. spun drive and struck wheel to simulate bump, all load was absorbed by cage twice once at impact and at 90 degrees later in spin direction. after 144 impacts the 92,997mi. cage failed in the same way as the 19,700mi. cage did. same test new 17 ball unit failed after 661 impacts in the same way as the 19 ball bearings. stay away from potholes"

Let me put it this way, I think it is either a very inadequately controlled test, or he lacks the ability or motivation to report it well, or the entire thing is a fabrication. In any case, his report gives me no valuable information, just questions. I mean seriously, he installed a window in the drive housing, filmed with "highspeed film 300 frames per sec.", and hasn't posted a picture, and can't write a better description than that? I'm waiting for pictures of his test rig, and a well written report of his findings. Until then, it is malarky as far as I am concerned.

I have 40K miles on a FD rebuilt with a 17 ball bearing. I have pounded over potholes, two up, overloaded with camping gear, ridden the "washboard" of the hardpacked dirt roads of rural Vermont at speed. My drive has been hammered. I opened it recently to inspect the bearing. After 40K miles it looks like new. That is just a case report of one, but frankly a better indication of things than is his reported simulation of potholes.

If this guy actually simulated 661 potholes and made a bearing fail, I'd like to know exactly what his experimental setup is, and I'd like to see good quality pics of the failed bearings after they've been cut open. Until he reports with good information it is meaningless. Let's face it, potholes aren't good, they can bend rims, and breakstuff, and maybe enough potholes hits will break a final drive. The mech. engineers who have kindly posted on this subject in the past have indicated that micro-stresses to the balls and races during installation or due to excessive preload may be what sets the stage for development of the pitting of balls and races that I am seeing in failed bearings. It stands to reason that if you hit the bearing hard enough you might initiate such micro-stresses eventually leading to failure. But I think that this is unlikely in a properly lubricated, properly assembled final drive.

So, an emphatic "NO", I do not think we need to consider: "tire pressure, load, shock type and setting, and even speed".
That's a considered opinion, but granted, just an opinion.
 

·
Registered
Joined
·
4,404 Posts
fpmlt said:
I'm sorry Curtis, but I was being a smartass when I asked that. :rolleyes:

PM sent
LOL, sent you a PM.
No apology needed
Dry wit humor 'eh? Fair enough. :histerica

But seriously, your question about speed, tire pressure, and shocks isn't the most off point question that's been asked about FDs. I mean, I've been reading, thinking about, and exploring the final drive issue for years. Some questions become inane for repetition, or just off the wall lack of thought; I try to not sound too much like an intolerant smart ass myself when answering 'em. :)

I have posted in jest too. Without the wink ;) it sometimes looks like a serious question, and in the context of so many other questions, I did take you seriously.

But I'm a much worse troll than you are, having queried here some years ago what the best snow tire for the K1200LT might be and other such nonsense. Without the ";)" many well intentioned folks will take the post seriously and try to give good advice.
So, having made such tongue in cheek posts myself on more than one occasion, I'm fair game.
Well played. :thumb:
 
1 - 20 of 25 Posts
Top