Clutch and input shaft spline - need advice please

nrpetersen said:

Barrel shaped wear on spline teeth will only happen with simultaneous inverse barrel shaped wear on the clutch hub.

Click to expand...

Not true. Look, I'm really not trying to pick on you but while I can theorize with the best of them, theory is only valid when it's applied in a relevant way. The facts don't support the way you've been theorizing.

Looking at heavily worn 1150 clutch splines we can see the angled wear pattern. Let's say it's 5??. Does anyone - ANYONE - honestly think there is a 5?? misalignment between the engine and transmission, or a 5?? warp in the clutch carrier? Believe me, the former would have the bike going in circles and the latter wouldn't have room to turn between the engine and transmission.

Understanding that the 5?? comes from something else is critical.

Hi Anton, as you have said you have never seen a repeat failure after repaired once so maybe I missed something but what do you think accounts for the problem if not mis-alignment? Could it be they all will eventually fail due to normal wear and the design just not beefy enough ( spline shaft not long enough, hard enough, etc.) and Paul is right when he says lube every 40K or so to hold off the problem? It has always seemed a stretch to me to believe BMW can't hold manufacturing tolerances close enough to avoid bolting on a bunch of misaligned trannies.

i have nowhere near the experience that Anton or Paul have- but my assessment of the situation, based on failure circumstances of either these things blowing apart at painfully low miles (30K range, and then again and again) or at a much higher mileage (70K+) is that both misalignment and inadequate/incorrect lubrication are at work here.
low mileage with repeated failure = misalignment. will continue unabated, unless teh alignment between engine and trans is addressed.
high mileage = lubrication. only needs to be lubed to prevent/avoid future failures.

any part can/will eventually fail, on any machine- so yes, whatever you have will eventually fail, given enough usage.

just my $.02, worth what you paid for it.

AntonLargiader said:

Not true. ...The facts don't support the way you've been theorizing.

Looking at heavily worn 1150 clutch splines we can see the angled wear pattern. Let's say it's 5??. Does anyone - ANYONE - honestly think there is a 5?? misalignment between the engine and transmission, or a 5?? warp in the clutch carrier? Believe me, the former would have the bike going in circles and the latter wouldn't have room to turn between the engine and transmission.

Click to expand...

Have you ever seen that sort of angular tooth wear on one spline - but NOT on the mating one?

This is the type of rig you need to accurately measure engine-transmission alignment. First the transmission must be disassembled and the case remounted onto the back of the engine. The run out measuring rig attaches to the flywheel using a couple of rare earth magnets salvaged from a junk hard drive. The magnets are epoxied to a standoff and extension crudely made from a piece of 1/2 inch conduit and a metal pad.

The end of the conduit is sawed at ~45 degrees and notched to hold a standard dial test indicator. A screw clamps the dovetail on the back of the indicator.

The test rig (painted blue now) is then magnetically mounted onto three of the flywheel bolts. Those rare earth magnets hold like all fury.

The dial indicator is mounted in the notch and with a lot of fussing is set to read the bore of the transmission input shaft bearing. This type of indicator has a measurement range of only +/- .015 inch so it takes a little while to get everything into proper position.

The dial indicator can be read thru the transmission housing as the crankshaft/flywheel is rotated:

It can also be used to measure the crankshaft rear main bearing clearance in any axis by simply prying on the flywheel and noting the backlash.

OMG!!!

Thank heavens my splines are pristine at 38K miles.

To tell the truth I don't understand the info in your pictures. I mean I can't figure out how to set up or perform the measurements you are describing. You must dis-assemble the transmission??

Red100RT said:

Thank heavens my splines are pristine at 38K miles. .............You must dis-assemble the transmission??

Click to expand...

You very likely don't have an alignment problem. Just renew the spline lube on a normal schedule per Paul Glaves.

Yes, this check requires the transmission to be fully disassembled. It is only for those that have to put in a new input shaft.

Red100RT said:

Hi Anton, as you have said you have never seen a repeat failure after repaired once so maybe I missed something but what do you think accounts for the problem if not mis-alignment? ... and Paul is right when he says lube every 40K or so to hold off the problem?

Click to expand...

I personally don't know of any repeat failures after I have done a complete clutch replacement. Several bikes I've fixed were two-time losers when I got them. I don't know of any other such bikes either, but it's by no means impossible. If the root cause isn't fixed, another failure is possible. It's quite possible that the root cause is not always what is addressed.

Misalignment is a broad term. I think just about everything can be called misalignment in some way. Some people need to specify more clearly which sort they are suggesting.

I really don't think lube will save a bad bike. I'm not the only one that's seen the remnants of stripped splines covered in fairly fresh grease. And there are plenty of dry and rusty splines discovered that look like new once they're cleaned up.

nrpetersen said:

Have you ever seen that sort of angular tooth wear on one spline - but NOT on the mating one?

Click to expand...

Of course; why else would I refute what you said? I've seen that, logically enough, on the splines we're discussing. R1150-generation six-speed input shafts. Not /5 splines, not early 5-speed splines, not '81-on splines, not Oilhead 5-speed splines like Paul's 350,000 mile bike. Let me stress that the spline arrangement under discussion here is geometrically unlike the others. This is significant. Anyone seeking to understand this problem needs to understand the implications of the difference.

nrpetersen said:

This is the type of rig you need to accurately measure engine-transmission alignment....

Click to expand...

BTDT! Mine was bolted to the crank, but the magnets should work too. I made a bunch of spiderweb charts and then concluded there was nothing in it. If you want to establish causality, you'll need to measure a bunch of good ones also because otherwise you won't know that your 0.0whatever millimeter misalignment is or isn't also present on every 'good' bike out there. By the way, the bike you show is NOT one of the models under discussion.

And BTW2, that rig alone doesn't detect angular misalignment (which I doubt exists at all). It only detects translational misalignment of the front bearing bore.

My 2002 R1150RS went thru similar ordeal at 28k mi. on odometer.

Since the bike had less than 30k mi., the clutch/tranny shaft replacement were performed under warranty late 2003.









OEM vs Touratech ceramic clutch plate:

Interesting in that your second photo showing the bad splines also shows what looks like metal particles which might have been visible had you removed the starter motor and taken a peek?

AntonLargiader said:

I personally don't know of any repeat failures after I have done a complete clutch replacement. Several bikes I've fixed were two-time losers when I got them. I don't know of any other such bikes either, but it's by no means impossible. If the root cause isn't fixed, another failure is possible. It's quite possible that the root cause is not always what is addressed.

Click to expand...

Do you always use a clutch pack alignment tool - BMW or of equal precision? (I presume so) That could be why you see no repeated failures.

AntonLargiader said:

Misalignment is a broad term. I think just about everything can be called misalignment in some way. Some people need to specify more clearly which sort they are suggesting.

I really don't think lube will save a bad bike. (agree)I'm not the only one that's seen the remnants of stripped splines covered in fairly fresh grease. And there are plenty of dry and rusty splines discovered that look like new once they're cleaned up.

Click to expand...

I don't see how the internal and external spline surfaces could go non-conjugal with either type (radial or angular) of misalignment. Non-conjugal would imply that wear is magicly occurring without contact.

AntonLargiader said:

Of course; why else would I refute what you said? I've seen that, logically enough, on the splines we're discussing. R1150-generation six-speed input shafts. Not /5 splines, not early 5-speed splines, not '81-on splines, not Oilhead 5-speed splines like Paul's 350,000 mile bike. Let me stress that the spline arrangement under discussion here is geometrically unlike the others. This is significant. Anyone seeking to understand this problem needs to understand the implications of the difference.

Click to expand...

How are they functionally different?

AntonLargiader said:

BTDT! Mine was bolted to the crank, but the magnets should work too. I made a bunch of spiderweb charts and then concluded there was nothing in it. If you want to establish causality, you'll need to measure a bunch of good ones also because otherwise you won't know that your 0.0whatever millimeter misalignment is or isn't also present on every 'good' bike out there. By the way, the bike you show is NOT one of the models under discussion.

Click to expand...

But the problem seems to be across all oilhead (and a few airhead) R bikes. I recognize only radial and angular misalignment between the engine and transmission axes. Are there others?

AntonLargiader said:

And BTW2, that rig alone doesn't detect angular misalignment (which I doubt exists at all). It only detects translational misalignment of the front bearing bore.

Click to expand...

Yes it specifically looks for radial misalignment of the only transmission bearing that can radially fatigue load and fret the spline system. Angular misalignment would be much easier to check by sweeping the mating surface from each element with a dial indicator and a spanning flat bar to handle the changing effective radii of each mating surface.

Anton - have you ever tried to have a clutch housing (and pressure plates, etc) refurbished and rebalanced? Would the issues with a bad clutch unit be something that a good machinist with proper equipment could straighten out?

Since you have not seen a repeat problem with a new clutch (presumably new, not a used one from another bike), does this imply that there was a manufacturing problem with the original clutch that has now been remedied with later spare part production?

My 49K mile '00 1150GS is sitting tail up with this exact problem and the transmission input shaft is being replaced right now, so I am interested why you think that a new (or a different used?) clutch would not have the same likelihood of re-causing this problem as the original clutch?

Thanks!

FatChance said:

Anton - have you ever tried to have a clutch housing (and pressure plates, etc) refurbished and rebalanced?

Click to expand...

Yes. It was a total failure; I had to go back in and replace it because the clutch chattered so badly. But it could be a successful approach if it's done right; I can only conclude that this one wasn't. And by right, I mean that EVERYTHING has to be true. The radial and axial alignment of both the mounting points and the spring seat, correct balance, maybe something else that doesn't occur to me.

I think there are just some crappy clutch housings out there and half a percent (or whatever) of the bikes got them. In that case, get a new one and you have a 99.x% chance of getting a good one.

Several years ago I moved out of the research phase and into the solution phase. I don't have anything new to add here.

AntonLargiader said:

I think there are just some crappy clutch housings out there and half a percent (or whatever) of the bikes got them. In that case, get a new one and you have a 99.x% chance of getting a good one.

Click to expand...

Do you also replace the cover, pressure plate and diaphragm spring along with the housing? Thanks!

This has been one of the most fascinating, and educational threads I think I have ever read on this site. Extremely illuminating. Great pictures, too. I just wanted to thank everyone involved for taking the time to share their hard won expertise and experiences. It's got to take many years of hands on work to get to the level of understanding some people exhibit about these machines. I believe all of us in the BMW community have learned a great deal on this topic. At least I know I certainly have.

jconway607 said:

This has been one of the most fascinating, and educational threads I think I have ever read on this site. Extremely illuminating. Great pictures, too. I just wanted to thank everyone involved for taking the time to share their hard won expertise and experiences. It's got to take many years of hands on work to get to the level of understanding some people exhibit about these machines. I believe all of us in the BMW community have learned a great deal on this topic. At least I know I certainly have.

Click to expand...

I'll second ALL that! I feel ready to pull the transmission and lube the spline.

I remember riding my 02 R1150RS pretty hard the first 26 months of ownership..
I took a pretty big spill on it at around 11k mi., no major damamge, but the subframe was tweeked pretty good.
I remember getting that bike airborn good number of times when crossing raised railroad crossings. There is one particular ones that was on my commute, I can see that landing a heavy bike like the R1150RS can put some stress on teh clutch spline if throttle was applied while the bike is airborn.
After my clutch and tranny was replaced under warranty, I never quite push that bike that hard again, once in a while rode it spiritly, but took it easy on the railroad crossings. Sold that bike about 6000 mi. after the clutch/tranny shaft was replaced.. got a y2k VFR800 in its palce... seemed to have better power to weight ratio than the RS.

I'm by no means a mechanic. About how much would a dealer charge to do this?

I'm the original owner of a '99 RT with 67,000 miles, and I've never had this done. I was aware this was an issue with airheads, but it's only recently (within the past year) that I've heard that this is also necessary on oilheads.

I have heard anything from ~$750-$1400 quoted for a lube, ~$1000-$2400 quoted for an input shaft replacement.

bikerfish1100 said:

I have heard anything from ~$750-$1400 quoted for a lube, ~$1000-$2400 quoted for an input shaft replacement.

Click to expand...

Well, the two parts that we are talking about wearing are the clutch disk (hub) and the transmission input shaft. MSRP for the clutch disk is right at $200 and the input shaft is just over $700. So totally aside from labor, or other clutch parts ($400 or so), or any other transmission parts, (seals and bearings) they are expensive parts to wear out.

So I try to prolong their life by making sure they are lubricated.

If you think the clutch housing (flywheel) is bent, wobbles, has axial run-out (however you want to express it), that is a $300 part. Add for new bolts.