This really states the case so well, I had to break my rule about including entire quotes with a follow-up post.
My step-father-in-law is a retired truck mechanic. We sat at the kitchen table two nights ago discussing the R1150 clutch spline problem, reading some of the online hyperbole, and looking at pictures of damaged parts. He asked some interesting questions and provided opinions.
Questions: (I'm paraphrasing)
Has anyone measured the motor/transmission fit tolerances for a larger number of bikes?
Did the clutch and transmission input shaft designs change over time, from the earliest R1100 oilheads through the R1150s?
Did the clutch operation system (cable versus hydraulic) change over the same period?
What changes in engine torque accompanied the upgrade from R1100 to R1150?
What was the frequency of reported issues with the clutch splines for late airhead, early oilhead, and R1150s?
What other R1100/1150 BMW components have received in-place parts design improvements over the bikes' production runs?
How many different clutch part numbers are there for these applications?
Comments:
The lack of pilot bearing/bushing doesn't matter, since those items rarely last long in automotive applications.
Drivetrain fit tolerances should really be known, and likely require special tooling to make good measurements.
Very similar spline wear & destruction was seen on transmission input shafts and clutch hubs in commercial truck applications, when existing components were re-used with motors producing greater torque.
As a result of the above problems, the spline configuration for truck drivetrains moved to higher spline count to provide increased surface area and resistance to shock-loading.
Damping springs in the clutch disk would surely help.
The partial engagement and cantilever arrangement of the R1150 clutch hub and transmission input shaft, beacuse of the too-short input shaft length, is "CRAP".
The aftermarket transmission input shaft out of Canada is the right idea, but might not be the solution.
Lubricating the splines of a dry clutch is a bad idea: clutch friction material will stick to the grease and create abrasive slurry, accelerating wear.
Most commercial truck splines are fit dry, without any lubrication.
You can machine the transmission input shaft using harder material (and the clutch hub too), but those changes probably won't cure the disease.
Changing from cable- to hydraulic-operated clutch probably hurt the rider's ability to detect spline wear.
There's no question that operator and drivetrain history (specifically the number and severity of clutch engagements) have a direct impact on this type of wear.
The clutch diaphragm spring fingers should be checked for signs of uneven wear.
If the spline pattern, hub/input shaft length, and "bellhousing" fit are designed correctly, this part of the drivetrain should NEVER be a concern.
After learning that the R1150s had been out of production for a few years, he guessed that BMW would officially ignore the spline issue alltogether, and suffer some losses to its loyal customer base. BMW should have quietly revised the input shaft and clutch hub designs as upgrade parts when the problem first arose, likely due to greater power output of the R1150 motor and use of already-on-the-shelf components that worked OK for the R1100s. Finally, he was confident that the aftermarket will find a good solution for the problem, just as it has for similar automotive issues.
He's going to help me dig into the drivetrain of the Roadster next winter. That should be an enlightening experience.