R1150 trans offset to the engine did you fix it?

[Mrmerlin]

I am working on my 2001 R1150GS and the input shaft splines are worn,
the disc has just about sheared the teeth off ,
and am working on recentering the transmission to the crank centerline.

With the trans split I put a dowel into the crank and found apprx .0095in. difference in the trans centerline to the engine.
looking from the rear of the bike the transmission input shaft is high and to the right in the 2 O clock position.
I was wondering what has anyone else done to fix this
would simply adding one alignment collar with an offset to it work or should both collars be replaced. just trying to figure out what to do before machining any new parts .

Next is trying to figure out how to get the clutch disc splines to have full contact with the input shaft ,
it has been suggested by a machinist to simply reform the disc center and push it to the rear so the rivets holding the splined hub are not removed.
This idea sounds like it would damage the disc.
Note I am going with the Siebenrock clutch disc

[Dr. Curve]

Hello Merlin. I mentioned this problem that BMW has with some bikes one time in a post here but others did not seem to think that there is much of a problem. Well, even with just nine thousands of a inch.......this means that while the clutch is engaged..........the driven disc must slide across the face plates eighteen thousands of a inch..........every time there is one revolution.....and that will, as you have observed, reek wear and eventually havoc on the splines of both the input shaft and the driven disc. This will also wear out the input shaft bearing over time which is what did happen of various bikes.

You will have to reposition the transmission to fit the center line of the crank by removing metal from the centering hole in the transmission. The position dowel in the block should be left alone.......but the hole it normally fits into in the trans case must be filed down in the correct direction to allow the fixed dowel to go over the 9 thousands needed. I know this sounds hard but if you measure carefully you can do it with a dremel tool using a tiny cutter. Scribe the ellipse needed very carefully and take out the metal.

Before you start, get hold of a pin the exact size of the one in the block. As you remove metal, tiny by tiny, insert the pin and monitor the progress until you have it right.

Clean hole completely and with pin inserted note that a area like a one day moon is now empty where the old hole has been enlarged. Re-clean the hole again with acetone. Take the pin and clean it in acetone so that it is also totally clean. Now oil it very lightly and wipe clean leaving a thin oily surface. Now, after mixing your JB weld completely until it is flowing and consistent............insert the pin into the original hole but over into the ellipse you have cut. Now, with transmission sitting of its rear, pour and let ooze the JB weld into the vacant area where the pin is not. Let it set up. Remove the pin which will slide out because of the oil. Dress the face and get ready to reinstall. Note that YOU MIGHT need to relieve the original bolt holes that mate the trans to the engine.......... but they most likely will have a 9 thousand free clear already. You get the picture. Good Luck!

If your bike has the fixed dowel in the tran then work on the engine block hole instead..........but it is easier and safer to work on the trans hole. Thoughts........?

[Jerry Feather]

Dr Curve:

I have been consulting with Merlin in regard to the issue he has started this thread about. I have done some search of this topic on this Forum and have read much of the thread you mentioned posting in before about this. I have also read the other major thread about this, the one by Bill Pierce, and his thread has given us pretty good insight to the problem of the misalignment of the transmission input shaft from the crankshaft, and about how he corrected it.

I think there are some other threads that relate to this issue, but in which it appears to me that there may be a failure to recognize that this misalgnment is the actual root of other problems such as relating to spline lube and one about a vibration deep in the innards of the middle of the motor/trans area that occured at idle with the trans in neutral. In that latter case there was a suggestion that in neutral the clutch was not engaged, so the problem could not be related to the clutch. Of course, that is not correct since I think the problem was occurring with the clutch engaged but with the trans in neutral, so I suspect that a misalignment was/is in fact that guy's problem.

I found one thing interesting in your previous posts on this topic in the earlier thread and that is that you suggest that the kind of misalignment Merlin has found has, in other cases, amounted to as much as .015 in. to .045 in. off. One thing that needs to be cleared up, and I think it was a source of confusion in the Pierce thread, and that is that when the misalgnment is reported as a certain amount, what exactly is the amount a measure of?

In Merlin's case I too took his report of misalignment of .0095 in. as center-to- center misalignment, but when he and I met about this recently I learned that the amount of .0095 in. is how far one side of the trans input shaft bearing bore is off relating to the other side measured from the centerline of the crank. That means that the shaft misalignment is only about .005 in. off center-to-center as relating to the centerline of the input shaft to the crank centerline.

Now, I am not sure what the .014 in. misalignment was in the Pierce case. However, given his/their description of how is was finally corrected I think it was center-to-center, although oddly I think they thought it was the other, because their first correction was only about half that, about .007 in. in each pin offset which they then found only corrected about half of the problem.

One might like to think that with only .00475 in. (or about .005 in. in round numbers that can be dealt with) of center-to-center misalignment, that is not enough to be an issue. However, Merlin gave me his original clutch disc and I can verify that the splines in it are worn away by about half. That strikes me as an awful lot of wear for two surfaces that are not expected to move almost at all in respect to each other. It seems clear to me that the wear is induced by the constant rubbing of the splines against each other as the clutch disc is forced from one side to the other in each rotation of the crank, and the total movement overall is the full .095 in. each rotation; or at least from the extreme pressure in both directions, that is causing the wear. Too, you have to consider the load that is being placed each time on the trans input shaft bearing and on the rear main bearing of the engine.

With enough misalignment, perhaps much greater that Merlin's, it would be no surprise that someone's bike blew up while on the dyno and with only about 6000miles of time on it. (Another thread).

I have some more to say about this, but this post is already too long, so I will try to let it soak in first, then I'll try to explain what we are looking for in respect to data. Thanks for looking.

Jerry Feather

P.S. I am new here and I am not a biker, but I love to help with these kinds of problems. I am an engineer of some ilk and a machinist, but neither by trade.

[Jerry Feather]

I think I read in one of the earlier threads that someone suggested that the misalignment, if there is any, is more likely to be in an angular sort of way rather than each center line in question being off with each other. That would mean that although the center lines of the crank and the trans input shaft are centered with each other where they come together, they are not both on the same straight line, therefore being at some angle to each other.

I am pretty sure that that is not the case that we are dealing with, but rather that in some or many cases the centerlines of each shaft, although parallel to each other, do not come together at their juncture. They are offset by varying amounts, and in some measurable direction.

My current theory about this mislaignment is that one of the four holes for the locating pins/collars (two in the engine case and two in the trans case) is off. I think it is obvious that the distance between the hole pairs is always the same because the bikes come fully assembled. This theory, however, is really the purpose of this thread, and that is to obtain some additional data, both is respect to the amount of misalgnment, center-to-center, and more particularly in which direction is the trans input shaft off in respect to the center of the crankshaft as looking at the rear of the engine from the rear. Which clock direction is the trans shaft off, if it is, and by how much. The limited data we have now is Merlin's which indicates that when looking at the fly-wheel, the trans shaft if off in the direction of two o'clock. The other data is from Bill Pierce who said his misalgnment was "up and to the right." I suspect that is also about two o'clock.

If only one of the holes is "off" as I suspect then the misalignment is always going to be off along a line from two o'clock to eight o'clock, about. If two holes are off, but still the same distance apart, so they still fit, then the mislaignment can be in any direction around the clock, and the correction will be needed at both locations.

If the misalignment is based on only one hole being off, it can be corrected at only one of the two locations, or both, I think. If only one hole is off, I can't yet figure out just how to determine which one it actually is; but for small corrections, such as Merlin's, I suspect it doesn't matter.

Merlin and I are going to correct his misalignment with offset locating collars, or only one, that I am going to make, rather than the method suggested by Dr. Curve about moving the hole(s) over. That method has a lot of merit, but seems to be less precise than I think we hope to be.

So, Dr. Curve, and others, if you have some prior experience with this issue your information would be of value. You, Dr Curve, suggested data indicating that this problem could include offsets up to .015 to .045, and I hope you may have some further information about that. It would be very helpful to know what direction that much offset was in. It might also help determine if only one hole is off or two and maybe even which one it is if only one.

Jerry Feather

[Slowgo]

Hello Jerry

Thanks for posting your findings, I am very interested because I bought Bill Pierce’s drive train. I wish I had more time to pull the eng – trans from the frame and get some additional measurements for you. Work has been brutal this year and also need to repair the wife’s bike. Bill might have recorded his measurements and I have his number if you have any more questions.

Mike

[Mrmerlin]

Well back at it here,
i got a few offset collars made and it looks like my machinist was correct about his assumption that only one hole was mis drilled and will need to be moved to restore the alignment.

it turns out the TOP collar is the one that needed moving.

So with the stock collar in the block at the 5 O clock position, I installed a few different sized offset collars to the top position.

All of them were clocked with max offset to the 8 O clock position.

I tried a few different sizes and settled on .020 offset this gives me a total run-out of .002 thousands .

So i put the trans back together last night and am now waiting to see what the new clutch disc will look like.


The plan is to reform the the central web of the disc to push the hub back towards the trans to increase the spline engagement.
The clutch disc being used is the Seibenrock oil resistant disc

It is slightly different compared to the factory part where the central web of the Seibenrock is a single piece of metal ,
the factory part is made of 2 thinner sheets of metal.
Jerry thinks that a slight reforming in a press should move the hub back the required 4 mm to insure full spline engagement and still have clearance between the trans and the disc hub.

Pictures to follow

[Mrmerlin]

Here are pictures of the collars I got made,
you can see the offsets that they have,
the collar that i used is stuck into the trans and that where its going to stay,
it has a press fit into the block as well its just stuck a bit more into the trans.

NOTE I used a factory collar in the lower 5 O clock position,
and the new offset collar is used in the 11 O clock position

[Jerry Feather]

Here is a picture of the tooling I have made to try to form the center of Merlin's clutch disc into better contact with the input shaft so that the two components have a little more surface area to match up with. With this tooling I expect to push the center of the clutch disc by .160 inch and then see how much springback there is. If there is more than about .010 inch I am going to have to work the tooling over so it will push the center a little farther so we can end up with at least .160 or even a little more of offset of the center of the clutch. This method is chosen over the alternate method of taking the clutch disc apart and putting a spacer in with new rivets.
Jerry Feather

[Mrmerlin]

Here are some of the pictures of the metal forming dies to put a bend in the clutch disc

[Mrmerlin]

more pictures

[Jerry Feather]

Thanks MrM for posting these pictures. I don't know why the one I tried to post showes up as a small square with a red x in it. JF

[Mrmerlin]

I was trying to get Sebenrock to make a new clutch plate I sent them a drawing on what they had to make
the response was they didnt think they would be able to sell 150 units per year of the new plate with a bigger spacer. Anyway maybe some others here would be interested in chatting up Siebenrock to make this new disc this would solve the splines not fully engaging issue but not the concentricity of the crank to input shaft

[Dr. Curve]

Thank you very much for this very informative solution and excellent photographs. The offset pins are a much better solution than the shade tree way I suggested and one but wonders why BMW let this go as far as they did before fixing it at the factory. You should have a vibration free running gear and a clutch that last forever, plus spline teeth that stay sharp and clean. Thanks again......for showing us the way to really fix it.

[Mrmerlin]

I had help, I would say that Jerry feather was the driving force to fix this concentricity mismatch ,
his simple idea made it easy to restore the shaft alignment.
Who would have thought that its just one alignment hole thats misdrilled.

The plate pressing idea didnt go so well as after the clutch disc was put into the press the friction material also bent so the disc is trash, a spacer must be used for the parts to work correctly

[adrianw]

Respect.

Just reading this thread made my brain hurt.

I off to play with some plasticine.

[1100s nut]

Originally Posted by Mrmerlin

I had help, I would say that Jerry feather was the driving force to fix this concentricity mismatch ,
his simple idea made it easy to restore the shaft alignment.
Who would have thought that its just one alignment hole thats misdrilled.

The plate pressing idea didnt go so well as after the clutch disc was put into the press the friction material also bent so the disc is trash, a spacer must be used for the parts to work correctly

When you say one alignment hole misdrilled, which are you referring to, the pin on the bell housing? Just curious. And if its just one how did the others pins enter their respective locating holes w/o binding?

Lane

[Mrmerlin]

Lane near as i can tell the first alignment hole is drilled into the trans at the 5 O clock position (looking from the rear of the bike forward),
the second hole is done at 11 O clock this is the hole that when moved will bring the shaft back into alignment with the crank.
So using the lower collar as a pivot point you can move the trans CCW till the shafts align, to hold it there you need an offset collar that will move the trans to the new position
I used a .020 IN. offset collar to do this and the offset was clocked to the 8 O clock position.
This effectively moves the trans shaft into alignment with the crank.
The whole misalignment was caused by faulty tooling as some of the trans housings / input shafts dont exhibit this issue thus they were drilled on other machinery that was in the correct specification

[1100s nut]

MrMerlin, thanks for clearing that up. Its ashame QC did not flag these cases before they were completed and shipped to BMW. I have seen tech. vids where the (can`t remember the correct name) but its automated indicator that checks each hole for the right placement.

Lane

[WillieJ]

Were you able to determine if the machining error was on the block or the transmission? I had the same failure on my 02 1150 GS with about 11800 miles on it. I had no access to the sort of machining and engineering expertise you have. I'm really impressed by the tooling that was developed for your project. I had seen other posts on advrider about using offset pins to correct the misalignment.

I bought a used transmission from a bike that had fewer miles on it than mine and hoped that the machining error was not on the block but on the original transmission. So far so good.

Still have the original gearbox with bad splines sitting on a shelf.

Here's the link to the Advrider thread on my input shaft splines. Ancient history now, 2007. Hope your fix is successful.

http://advrider.com/forums/showthread.php?t=207014

[Mrmerlin]

Willy from the things that i had to do to get the trans lined up,
I would guess and it is only a guess ,
that it was defective tooling used from Getrag,
since there are others that have no spline wear,
it would indicate that the transmission was at fault.
Since this is an outsourced part