No wonder the clutch slave cyl's don't last

[macx]

During reassembly, I was shocked when I realized the release
pushrod always has compression inside that little throwout bearing
in the slave cyl, exerting constant tension on it against the
spring inside the cyl and making it turn constantly while the
engine is running.

No wonder they don't last! And at $100+ a pop, it's another guaranteed
recurring parts revenue source for BMW and repetitive revenue for dealers
that would be a totally unnecessary expense for owners if it instead was engineered to avoid deliberately wearing out.

That's totally counter intuitive and the opposite of what automotive
setups are - they retract the throwout bearing away from the spring
tension of the clutch when the clutch is engaged so the only time
it turns and is under compression is when the clutch is being disengaged.

It should be set up so the spring in the slave cyl tensions in the
opposite direction - is positioned in front of the throwout bearing
to hold it away from contact except when hydraulic pressure is
applied to disengage the clutch.

And then have at least an oilite bushing in the flywheel to locate the
forward end of the clutch pushrod, or as in better automotive setups
a small bearing which could be similar to the one in the slave cylinder.
It would still only take direct significant compression when the clutch is
being disengaged.

Enlightened engineering?

[sweatmark]

macx-

Great observations and comments.

The combination of clutch slave piston/bearing design PLUS the partially-engaged clutch hub splines makes one wonder who was responsible for the R1150** transmission design. Addition of 6th gear from earlier R1100** transmission was welcomed, along with better/smoother function, but the problems associated with the R1150** gearbox & clutch should simply never occur with a single plate dry clutch.

[macx]

Agreed most wholeheartedly.

All kinds of other vehicles, from low to high powered cars and up to huge big rigs with well over 1,000 ft lbs of torque have dry plate clutches and none of those problems.

I took my old slave cyl apart and saw the rather strong spring behind the throwout bearing and the double lip seal.

"Thinking out loud" here -

IF the clutch master cyl has enuf capacity for just a couple mm more stroke - which it does feel like it would judging by how much lever travel there's left when the clutch disengages- it wouldn't maybe be that difficult to shorten/weaken the spring behind the slave piston or replace it with a softer spring, or why is a spring needed behind there at all really?, and add a small / short one in front of it just strong enough to push the piston back against the residual hydraulic pressure remaining in the system after the clutch lever is released - just enough to relieve the tension on that pushrod and on the little throwout bearing inside the slave cylinder piston and pull the pushrod back just a couple mm so it is not in contact with the clutch diaphram spring.

There would probly need to be some means of pulling the pushrod back just enuf so it wouldn't remain in contact with the diaphram spring or it would wear out rather quickly at the taper on the pushrod where it seats in the center of the spring. That shouldn't be that difficult to accomplish with a little study. Maybe a small groove in the pushrod at the right point with a rubber or teflon ring - or one of those metal circles that you have to compress just a little to get it inside of a groove - if there's a way to i.e. attach it just inside the opening in the slave cyl piston such as a small groove. Then when the piston retracts it would pull the pushrod with it. Only would have to move back a couple mm.

Maybe also need a positive stop of some sort (thin sleeve? - that would actually reduce the volume behind the piston to make up for the required extra couple mm of travel to disengage the clutch too!) inside the cylinder so the piston only retracts a certain amount from the spring between the clip and the outer end of the piston pushing on the piston. Really, no need for
the spring behind the cylinder piston at all then.

The rest of the slave cyl could remain the same - same piston, seal, inner bearing. The one other thing that would need to be solved would be the pushrod end inside the end of the flywheel - with the flywheel rotating but not the pushrod. Some type of oilite bushing or something. Will have to wait to take a look at that until next time I check and lube splines. Hmmmm, gonna have to study my old one some day when I feel creative Sounds like a little retrofit kit might just be possible.

Without constant tension on that little throwout bearing that slave cylinder
should last a very long time.

EDIT: Gosh, if I'm so darn smart, though, why did it take me 1/2 hour to find the plug the gear selector wire hooks to!

Ah, well, all the more reason for a "refreshio" - after all, the clouds ARE gathering (beer clouds) and it IS beer-thirty. Tomorrow sounds like a good day to finish putting it together.

[BigJohn]

Most modern cars have a clutch release mechanism similar to BMWs. The type you're describing where a spring pulls the slave cyl away from the clutch is the old style that requires frequent adjustment to keep the bearing from riding on the pressure plate and wearing the bearing out and causing clutch slip. More recent designs have the release bearing in constant spring-loaded contact with the pressure plate, the bearing spinning the entire time the motor is running. This results in an auto adjusting clutch, which is the goal. I think it works similar to disk brakes, in that the piston floats in and out as required. Modern bearings and lubricants have made this system viable and it does work on most bikes most of the time. A lot of cars use a concentric slave cyl which is located around the first motion shaft and requires gearbox removal to replace. Ours is, at least, relatively easy to get to.

[boxermania]

All excellent points, however, they don't account for the basic issues with the design. The basic transmision envelope has undergone several engine size and hp increases.

Add the small diameter of the clutch disc (due to dimentional constraints), the clutch disc hub only engages ~ 80% of the tranny input shaft, there are no springs on the clutch disc to lessen the shock load of the torque transfer between the motor and the tranny and the worst blunder of all, the clutch disc is overhung (weight and force) on the tranny input shaft since the shaft is not bossed onto the end of the crankshaft as it is usually the case on 99% of transmission designs. The latter also accounts for some of the missalignmnet issues, between engine and tranny, that have surfaced ocassionally.

That being said, it is amazing of the relatively small amount of failures, but well above what would be considered the standard deviation, of course that will depend if you have experienced a failure or not.

I have a lot of respect for German engineering but in some instances they miss the target by a mile

[macx]

I guess I haven't had any "modern" cars with sticks.

But there are hundreds of thousands of "old" cars running around
with the "old style" clutch setup - and that's still all you see in any
type of performance or race cars.

I've owned and worked on quite a number of them throughout the
years and have never had any problem with frequent adjusting or
wearing anything out.

Also, heavy trucks also have that retracting style of clutch release
bearing, and they run for many 10's of thousands of miles with no
problem. I used to be shop forman for a big paving outfit where we
took care of well over 300 heavy trucks, all with those kinds of clutches.
The only thing we ever had to replace clutch parts for was just plain
wear from heavy use - i.e. starting heavy loads in soft gravel pits, etc.
I have also been parts manager for a heavy truck dealer, and later
owned my own little over the road company.

I'll take that kind any day over something that is extremely difficult to
work on and requires disassembling an awful lot of stuff to access.
Not to mention very expensive.

[BigJohn]

Hey! I'm a guy who would seriously look at a cable clutch actuator kit to go on my 1150 if I could find one! I LIKE to adjust things and personally think that a lot of the things they do to modern bikes is done only because they CAN. I'm only trying to put their engineering into perspective. How many large bikes use hydraulic clutch actuation nowadays? Virtually all, I'd guess. How many modern automatic transmissions have bands in them? Because they require regular adjustment I'd say few or none. This is what drives vehicle design. If you look under the hood of a BMW car almost everything is covered by decorative plastic. It won't be too far in the future that the hood will be sealed (if there is one) and there will be no user servicable areas there. Sensors will measure all fluid levels and you will be directed to a dealer when one is low.

[macx]

I definitely agree there! Don't get me wrong, I have absolutely
nothng against hydraulic or self adjusting clutch setups. Mustangs,
for instance, went to a self adjusting mechanical setup on their
clutches quite some time ago. I don't have a clue what they have now.
And many hot rods use hydraulic clutches simply because most of them
are of unique configuration one way or another and nothng stock would
fit or at least easily or neatly.

I just think BMW could have designed that differently so it still uses
the hydraulic hardware but avoids the constant tension on that tiny
little dry running bearing. It's just simply designed for failure, as
evidenced by places like Beemerboneyard and others selling less
expensive aftermarket slave cylinders. They wouldn't bother if
the failure rate wasn't sufficient to make it worthwhile and nobody
would make them.

I'm still "noodling" on how to retain just about all of the existing hardware
and just make it so the pushrod retracts just a slight amount to break
constant contact with the diaphram spring and take the constant load off
of that little bearing.

It would still be hydraulic and all that.

Stay tuned!

[CycleRob]

I too was stunned by the unnecessarily strong spring pressure continuously applied to the slave piston throwout bearing. It is ridiculous! I considered shortening the spring to decrease the heavy load on that tiny sealed ball bearing, while wondering why such a strong force was required. All I could come up with was that it was needed to prevent a destructive spin wobble harmonic at a certain RPM. That's why I made no spring force modifications. It sure is a head scratcher, when the automotive designs are so very trouble free.

BTW, my new bike has a wet clutch, with the usual cable operated release, that backs away with a no touch no spin freeplay clearance. Duh!

.

[macx]

EDIT: I "what if'd" my way around in a circle (below) and ended
up right back at the status quo.

Took a look at the 1200 setup, of course the diagram doesn't show
much but I did note that the 1200 & 1150 cylinders are listed as
"D=24" maybe meaning cylinder diameter? Also the 1200 has a
similar pushrod setup. Altho the mounting is different, the piston & bearing assembly might work in the 1150 slave as the diameter's appear to be similar. I wonder if they improved the longevity and durability of the throwout bearing for the 1200's. The main question would probly
be pushrod length. Now to see if I can find a closeup picture of a
1200 cylinder to get a look at the throwout bearing.

EDIT 2: Found pictures of both slaves, unfortunately it appears that
BMW may use the same throwout bearing on the 1200 as on the 1150.
I posted a question on the 1200 forum to see how they're holding up.
Maybe BMW upgraded the bearing itself.


I looked some more, did some more pondering.

At the very least, they seem to have room for a more robust bearing
that would last longer under the constant compression. The existing
piston could be machined for a larger OD bearing, and the bearing
cavity be made deeper. As the thrust from piston against the shoulder
on the pushrod is taken at the outer surface of the bearing, it's actually a thrust type bearing altho not of conventional automotive configuration.
That would be one simple way of improving the durability if a more robust bearing could be installed even if the cavity in the piston would need to be machined larger / deeper.

And I most certainly agree that the spring pressure seems excessive.
I, too, thought it must be to make sure the pushrod stays firmly in
contact, and spins with (so it doesn't wear where it contacts) the
diaphram spring.

After studying it awhile longer, the main problem I see so far is how
to prevent wear at either end of the pushrod if a way is figured out
to either withdraw it to get it away a bit from the diaphram spring or
leave it in position and withdraw the throwout bearing from the back
end of the pushrod.

Best I've come up with to this moment is an oilite bushing in the outer / larger cavity in the end of the crank. I think that might be better than a bearing, because if the pushrod is withdrawn to the rear, it would be moving
forward and back inside of whatever locates it. In the inside of a bearing, I'm afraid that would wear the pushrod in short order. An oil impregnated bushing might actually fare better and be easier on the OD of the pushrod. Even a teflon bushing might work, but using either type would be complicated by the dry and probly abrasive dust from clutch disc wear. BUT - that then would cause wear at the pushrod taper to the diaphram every time the clutch is disengaged because the pushrod would not be rotating and the diaphram spring would be - every time the non-rotating pushrod is pushed into the dry diaphram spring seat to release the clutch it would wear there, and pretty rapidly.

So, I also thought about leaving the pushrod in position and withdrawing
the throwout bearing with a short fairly light spring between the outer
circlip and the outer end of the piston to push it to the rear and away
from the pushrod, and replacing the existing spring with a sleeve to
act as a positive stop. The distance would have to be carefully calculated
of course. But then there would be the problem of maintaining sufficiently firm contact between the pushrod taper and the diaphram spring to prevent a difference in rotational speed which would hasten wear at that point.
Maybe a bushing in the end of the crank that is a snug fit on the end of the pushrod to keep it rotating at crank, and thus clutch, speed would overcome the potential for wear at the pushrod to diaphram contact point. Loctite stud locker could be used to secure the bushing inside the rear, larger cavity in the end of the crank, but then the problem of the pushrod having to move back and forth during clutch disengagement arises. So how to keep the pushrod rotating at clutch speed and still be free to move forward and back for clutch release. Hmmm.

For awhile I was thinking it would be easier to withdraw the pushrod a couple mm. The oilite bushing would then be necessary in the end of the crank & around the small forward end of the pushrod to keep it from wobbling. It might still rotate from the inevitable friction, even though relatively light friction, with the bushing, but by releasing the compression against that small dry throwout bearing it should make that bearing last a lot longer even if the pushrod is still rotating inside of it. There, again, however, it probly wouldn't do to allow the pushrod to rotate at a different speed than the bearing which would precipitate rapid wear at that point. That could be lubricated with some moly paste as it wouldn't collect clutch dust in that location, but any constant difference in rotational speed would still cause wear probly on the pushrod and any kind of lube probly not last long with frequent contact and release. Stymied at this end, too.

At first it seems like it should be an easy fix, but preventing wear on either end of the pushrod makes it a head scratcher - no matter what one does there's the possibility of rapid wear on the pushrod which would negate
the advantage of relieving compression pressure on the throwout bearing.

So, given the difficulty of preventing wear on either end of the pushrod from parts rotating at different speeds coming into repetitive unlubricated contact, my thoughts at this stage seem to arrive back to seeing if a sturdier bearing would be commercially available i.e. thru McMaster Carr and have the existing slave cylinder piston machined to accomodate it. That's probly a long shot cuz the ID would have to match the pushrod, or a bushing installed snugly on the pushrod to vary the OD to fit, and would also have to have a cup in the inner portion for the rear end of the pushrod.

Seems I've worked myself full circle and have arrived back at the status quo!

Anyone willing to try a slightly reduced spring pressure to ease stress on that throwout bearing?

Making it better during manufacturing would be relatively easy. Making simple and effective modifications to the existing system apparently is
less so.

Back to the drawing board.

EDIT: After looking at the setup on the parts fiche and considering that
the pushrod needs to rotate at clutch speed AND maintain sufficient pressure against the throwout bearing to keep that moving at the same rpm also,
how about a snug collar loctited onto the pushrod that has 2 or 3 equally spaced fingers just long and thin enuf to go thru the diaphram spring and
maintain contact in both engaged and disengaged position. Something somewhat resembling one of those pronged things that hold the meat on a rotisserie. THEN one could lessen the spring tension in the slave cyl behind the piston and thence on the throwout bearing because the pushrod would not then be dependent on spring pressure to keep it in sufficiently firm contact with the diaphram spring to keep it rotating at the same speed, but retain just enough pressure to keep the throwout bearing rotating at the same speed as the pushrod just thru contact but with reduced tension.

Wouldn't you know I've already got my clutch and trans installed so I can't study that possibility! Will have to look for some closup pics of a diaphram spring.

EDIT: found a good closeup on ebay. Boy, they don't leave much excess room in there for anything! Would pretty near hve to be forward of the diaphram spring with the fingers pointing backwards.