BMW R1200R R1150R/T F800 K1200R R1100R/T R1250R RnineT Message Board

We have gone full circle on the clutch spline failure and the following documents the research, pictures, measurements, related incidents and conclusions.

The topic is essentially a root cause failure analysis and I’ll try to be careful with the explanation and language so as not to confuse the readers. So off we go:

Background
The BMWR259 models (4 valve heads) are no strangers to clutch failures as they have been documented on the various forums representing the brand time and time again. The failures are not model specific as they affect the R, RS, RT and GS, albeit I haven’t found many GS’s the have been affected.

On the R1150R forum the topic has come up several times dating back to December 24th, 2005 describing specific instances and September 12, 2007 among others, documenting specific failures across model lines.

One of the most interesting aspects, to date, is that the quantity of failures is significant enough and potentially dangerous to have warranted acknowledgement and action by the OEM but this has yet to happen. Some owners have been lucky and their repairs were warranted, some others had to endure partial or full cost for the repair along with the lay-up time.

Data
I was provided with actual samples of failed parts kindly donated by RiftonRoadster, along with pictures from tor1150r parts, for inspection, evaluation and measurement.

Rifton parts – 02 R1150R @ 43K miles
Input shaft had no visible/measurable wear other than the splines, the bearing was tight and free. Clutch disc measured 0.248’ (thickness) between the friction faces, as new thickness is ~0.350’ or ~ 9 mm, complete wear, to the rivets, is 0.175’ or 4.5 mm. Based on these measurements and the specific operating habits of the owner. I would have predicted a clutch life of 60K miles with a safety factor of 0.5 mm before rivet to flywheel contact.

The input shaft displays significant spline wear starting at the transmission end and tapering off towards the front of the input shaft. This wear, in the vernacular, is the pre-cursor to the failure. The wear is uniform around the OD of the input shaft indicating the presence of angular misalignment between these two elements. Normal lubrication can’t cope with this phenomena and is therefore “squeezed” out by the heavy contact and accelerating the failure.

Another issue is the length of the spline engagement between the clutch disc and the input shaft, the result is the “overhang” of the clutch disc on the input shaft. I measured .583” or 14.82 mm for the “worn” input shaft spline and .720” or 18.29 mm for the length of the clutch disc spline, that computes to 80% contact between the two. Do remember that as you decrease the contact area, the force over unit area increases by the same amount. See pictures below:









Tor1150r parts – Failures at 39.9K and 66.7K
The pics below show one of the potentially dangerous failures as once the disc splines are lost the bike will loose forward momentum. See pics below





Other Data
BMWsporttouring - This site relates the trials and tribulations of an owner that endured 7 failures on a 98 R1100R. The highlighted link RickPon his post takes you to another user that experienced the failure along with pictures on a 04 RT with 12K – 13K miles

http://bmwsporttouring.com/ubbthreads/s ... Post596643

Pelican BMW forum - Relates a double failure in Australia of a 03 R1100S, the first at 13K kms and the second at 17K kms.

http://forums.pelicanparts.com/showthread.php?t=381041

Pelican BMW forum - Relates a failure in the UK as early as 2000, the owner was lucky as he received a free tranny.

http://forums.pelicanparts.com/showthre ... adid=24006

On December 24, 2005 on the R1150R forum, Boxer started a thread to which all of the pundits contributed; it will be worthwhile for you to visit as the causes were pretty much out on the open.

Likewise on September 12, 2007, on the R1150R forum I started a thread collecting info on those that had failures and the following was reported then:

EVmachina, 04 Roadster, failure at 17K miles
Tor1150r, 04’ R1150R, 2 failures, 39.9K and 66.7 K miles
Marty Wadd, 35K miles
Paul Mihalka 02 R1150R, failure at 29 K miles

Most of the gearboxes are Getrag, but some have no visible ID.

Google “BMW Motorcycle clutch failures” and you will be exposed to a vast array of sites and experiences.

Mention has been made of the guide pins or lack thereof between the gearbox mounting flange and the engine flange as a potential source for the failure. As I understand there are two pins that establish the centerline between the crankshaft and the transmission input shaft, so as to prevent lateral misalignment. Being that I never had the need to drop the tranny, I could not attest to the quantity and location.

Conclusions
It is my opinion that there are many contributory elements to this failure and I will try to dissect each one of them. However the type wear exhibited on both the input shaft and clutch disc splines is characteristic of angular misalignment between the gearbox and the engine (this could also be experienced at any of the bolt up vertical surfaces, i.e. the crank flange, flywheel flange, flywheel surface or square ness with the crank centerline and the same on the tranny side.

This failure is not maintenance or operator induced, it is merely a machining/alignment error that the OEM has not seen fit to accept responsibility for.

In addition to the above there are other contributing factors that will eventually affect the longevity of the input shaft/clutch disc, as some of the early models are coming up on some serious miles by now.

Transmission input shaft – The design allows having the throw out bearing as an integral part of the clutch/flywheel, unfortunately to accomplish this the tranny input shaft doesn’t pilot onto the crankshaft as it is the norm in all automotive applications to insure that the centerline of the two is not compromised. Design issue.

Clutch disc - The clutch disc splines do not engage fully with the input shaft counterparts, this increases the transferred load. In addition the disc portion that overhangs the input shaft allows the “rocking” of the disc as the clutch begins to engage. The disc should have full engagement to the input shaft splines. Design issue.

Overall the basic engine design remains, the improvements have been evolutionary and within the confines of the original engine/tranny case, this limits any type of improvement onto the size or design of the clutch or its components, meanwhile the displacement and output increased over the years. Basically the OEM has been living of the safety factor built onto the original design. Design issue

Last but not least, this little problem will eventually come to haunt the BMW community as the bikes in question are coming of age and the spline problems, if they are to be had, will be imminent. I would want to have piece of mind if I acquire an R259 second hand, at least to the point of knowing if it has had the spline failure.

Anyone is welcome to the parts for their evaluation…..I volunteer to work with the next unfortunate user that befalls upon this problem and will compile a list of things to look for and measurements to take to get down to the point source for the problem.

44

I wonder if there is a simple way to align the clutch/spline interface. I know that on slash 2 BMW's (technically 61'-69', but the same as pre slash 2's from 55' to 60'), there is a way to align the transmission to the crank, a similar situation I think, by loosening the 4 bolts between the the transmission and engine so that the two parts can move just a little bit, start the engine, run up the RPM's a to maybe 3K or so, pull in the clutch and then while running and clutch engaged, reach down and tighten the mounting bolts.

Why does this procedure on the /2 align the transmission in a way that a static procedure cannot?

Al, there are two aspects of the wear pattern that still bug me:

1) I would think that a small amount of wear would accommodate some misalignment early on by distributing the misalignment loads fairly evenly along the tooth faces. However, badly worn splines have a huge gouges that dig deeper than a reasonable misalignment angle.
2) I can't explain why there is no wear at the end of the axle splines. I would have expected equal wear at both extreme contact points on opposite faces of a spline tooth.

Do you have an explanation?

Carlos

In answer to your questions:

1) You are absolutely correct on your assumption, however, the misalignment here is much greater as depicted by the extreme wear on the back of the input shaft spline. Let me point you to the post on this forum, titled Clutch Splice Lube (Crosspost) where the very first post by R4R&R posts a link to a friend that lubed the splines on a O2 GS with about 40K miles on it. I have included the link for simplicity, I would have used the pictures in my post, however I was unable to copy/open.

http://www.bmwbmw.org/bmwforums/viewtopic.php?t=10801

Please look closely at the tranny input shaft splines, both uncleaned and cleaned and if you look really closely, really closely, you can see the very same wear pattern on the back of the splines, albeit, significantly less. The reason for this is, as I have explained, is that the clutch disc is cantilevered on the tranny input shaft and when the engine transfers rotational force via the clutch to the tranny the inut shaft is able to move ever so slightly to preload the rear portion of the tooth. Nevertheless, I would consider this bike one with no spline problem.....

2) There is no wear at a portion of the clutch disc splines (the ones on the clutch disc away from the tranny) because the relative axial position of the elements does not allow the clutch disc to fully engage the input shaft splines, they are there merely for the ride.....about 20% of the full engagement.

Al, thank you for the work. A few questions, please

did you get any idea of the incidence of the failure? This is an exceedingly rare occurrence, isn't it? Unless it happens to you, then it is 100%

Is the failure more common in certain years of production. I believe that word around the campfire suggests that it is even more rare 2004 and later; did you see this?

Is there a photo or a drawing showing where these splines are located, if viewed on an intact bike?
Thanks again
JOhn

Al -

Thanks for the info. Semi-troubling issue, since Mrs. Sweatmark's '02 Roadster is of the more problematic vintage. My '04 Rockster might be less suspect, but we'll see.

Is there any chance that the clutch slave's applied force causes the cantilevered clutch disk to wobble, or rotate off-axis, thereby inducing the wear pattern shown in your pics above?

Moderators-

Would it be possible to place links to this and other problem-solving technical threads somewhere, perhaps within the Tech section? We have some good info on the potential problems for our R1150Rs, e.g. clutch/transmission splines, final drives, clutch slave, fuel line quick connects... and it would be helpful to at least index these threads for future reference and possible additions of info & fixes as time marches along. Thanks!

John,

I can't asign the failure a % of total population, as I don't have the figures for either. However, in the write up I made reference to at least 6 bikes plus 2 or 3 more on the links.

Initially, I thought that it favored the 02 models, however, with the passage of time I find that the 04 models are also affected. Of course the 02's will theoretically have more mileage than the 04's so by design they will show up first.

Model wise there isn't much different, they are all pretty much affected, I think it shows off earlier on the RT's as they are a heavier bike.

Regarding the location of these parts, the tranny input shaft and the clutch disc should be at the elevation of the starter motor and pretty much half way accross it's length.

Hope this helps..........

Great & revealing photos....

To me the clutch spline failure appears to result from under-engineering that is possibly...

1. Too much engine torque for the designed splines
2. Clutch splines which are not physically sized robust enough
3. Clutch splines that have the incorrect temper hardness
4. Inadequate tolerances for initial spline machining alignments (ie too loose)
E.G. Too tight tolerances has the effect of the splines binding together and staying stuck.

This unfortunately appears to be a very big chink in the armour to the invincible BMW knight
and has shown that BMW is now mortal and their legendary status is now able to be wounded.

Arbee

As you and I have brought forth, there are several design aspects that can precipitate failures of this nature, unfortunately neither you nor I are in the motorcycle bulding business, however, there are two fundamental issues pivotal to this issue.

1) The problem, IMO, is derived from a combination of an old design that has endured several different upgrades in both displacement and HP over the years, while remaining within the basic packaging envelope.

2) The arrogance by the OEM to acknowledge and assume responsibility for their product in the face of those that promote their brand......PRICELESS....

Good info, great analysis. Many thanks.

Any BMWRA members here? I would love to read the complete article from OTL: http://www.bmwra.org/otl/splines/

Great post, boxermania. Thanks for taking the time to gather the data. Placing accountability on the OEM may be a daunting task, if that's what the intentions are.

I'm thinking that the next chapter in this saga would be an attempt to figure out which of our bikes are susceptible to the spline failure without removing the transmission, if that's even possible. I know that I'm concerned, and have been for a while now. I guess that's why roadside assistance is a good thing.

Carlos......

I second your desire to seek the article as I would also be interested in reading the full account. Fundamentally we have reached the same conclusion; he seems to have a disc or discs that show heat checking, as a sign of angular misalignment as shown on his second drawing. The parts I have show none of that, in fact, the disc is uniformly worm across both faces.......but as I also mentioned, there are other places where the angular misalignment can be located.....

I measured the backlash, between the disc and the input shaft (had to turn the disc on the input shaft to engage the “good” splines and found .004" to .006" backlash or .10mm to .15mm. He mentions 0.10mm as an acceptable tolerance......the point is that you don't need much misalignment to take up the gear backlash to "dingle" the clutch disc onto the input shaft splines. The clutch assembly is a complex arrangement since it contains the flywheel, the pressure plate, the clutch disc and the TO bearing all in one package, due to its complexity and various components this would b the logical place where a missed dimension, ever so slightly, could cape the QA/QC process.

There is no question in my mind as to what is happening, but not having a bike to take down and effect all the run out and angular misalignment checks makes it difficult to pinpoint the actual source. I have an old design CAD program and if time permits I will draw all of the potential locations where the problem can be located.....time is the key element here.

Anthony....thanks for the encouragement as I must confess that FI is both a passion and a strong suit.

In regards to the post, the intent is not to rally the users into "critical mass" against the OEM, but to share the nature of the problem, albeit not what was the one event that precipitated such.

There are many other little quirks in the character of the BMW boxer that had already made up my mind as to the product stewardship of the OEM......but the potential danger and cost to repair this one, takes the cake.

In answer to your question as to how to check if the problem is there, I'm afraid that the compactness and location of the elements simply preclude any type of inspection, barring removal of the tranny, which albeit not difficult is a royal pain in the arse just to gain access.

Long live the Volkswagen Beetle, in my heyday, on a good day with everything ready I could bring the engine and tranny down in 15 minutes.....I remember my mother making comments to my brother and I as to why we had to take the engine out so often......we were bored and just wanted something to do.....I guess that's why I'm the eternal tinkerer...

I remember my mother making comments to my brother and I as to why we had to take the engine out so often......we were bored and just wanted something to do

I here ya!! I used to change the engine in my '73 Nova on the weekends "just because."

Since my brother owns an '03 RTP, I may practice on his tranny before I yank mine. Maybe a good winter project for next year.

ASQTec

There, I knew I wasn't alone.....practice makes perfect....go for it......

boxermania -

Thanks for the informative post. I appreciate your time.

You offered a conclusion, but really no recommendation. Are these bikes considered (by you) to be non-trustworthy? If we are worried about expensive failure, should we sell now before anything happens? Does this happen enough that I should be worried about taking my bike very far from home? Honestly, I can't stand the thought of having to call the local red-neck tow truck driver when I'm 2000 miles from home. I didn't think I would have to worry about this type of thing when I bought a BMW.

I just bought an old Ducati fixxer-upper and now I'm wondering if the Duc would be a more dependable long-trip bike

My personal best solo time for VW engine removal is 18 minutes... with a lift and a floorjack all staged up. Good times.

I wish that I had read this post because my clutch spline went out at around 26,000-miles and just got fixed. The teeth simply wore out on both ends, so that also suggested to me misalignment due to either bad assembly (highly doubtful) or bad manufacturing. It had nothing to do with lube because mine still had lube on them. I should have taken some pics and measurements too. Too late, the damn thing is in the trash and gone.