A Woodworker’s Guide to Gearbox Rebuilding

[Steve in Golden]

My son reckons he is going to put on my headstone (if there is one), "Here lies an atheist, all dressed up and nowhere to go."

If the xtians are right, you (and I), as an unbeliever will be going straight to a rather hot place.

[Airbear]

My son reckons he is going to put on my headstone (if there is one), "Here lies an atheist, all dressed up and nowhere to go."

If the xtians are right, you (and I), as an unbeliever will be going straight to a rather hot place.

Syria?

As a sort of final note to this thread, I have published the whole thing - sans comments - as a PDF with embedded photos. It is 4.5MB, and the shimming spreadsheet in XLS format (for Excel) is 35KB. If you would like a copy, flick me an email or a PM including your email address and I'll send by return message.

[Airbear]

The saga continues with a different box:

As part of tidying up after the above exercise I decided to clean the borrowed gearbox, ready to pack and return it to its owner. I set to with solvent, brush and rags and all was going well until I turned it over and heard a little rattle. I picked it up and shook it and “rattle rattle” it went. I knew what the sound meant.

Short story: I pulled the cover off and there it was, a broken dog …



Not only that – there were three dogs missing from the same sliding gear on the intermediate shaft (as featured in the above story) and only one in the case. This gearbox has unknown provenance. Like the refurbished box that is now back in my bike it is stamped with a “73” which means it is one of the first 5 speed boxes, built for the 1974 model. This one has no kickstarter but the input shaft has the teeth for one and they are in used condition, so it had clearly been opened up previously.

Other issues – the front input shaft roller bearing is quite sloppy, more so than others I’ve felt. Also the big bearing on the front end of the output shaft is badly worn (outer shell can twist quite a lot in relation to the inner shell) and the inner shell slides back and forth on the shaft (about 1mm). The circlip is also broken. There is no snap ring under the circlip but there appears to be room for one.



There are some good things. The input shaft splines and ‘cush’ cams are in excellent condition. The gear teeth on all shafts look good enough. And this is the really good thing:



The spare sliding gear in that pic is the one that Disston (of ADV) kindly donated to the cause. It is the right one for this box but there is no hope of fitting it with the equipment I have. That’s ok, I’ll take it to the local motorcycle-friendly engineer – he has a proper heavy duty press.

Here’s a question: Is there special information I should pass on about disassembling and pressing the gears back on again? Or is it just a matter of applying brute force?

I’ll have to dismantle the output shaft to see what damage might have been done to the shaft by that sliding bearing. I do have a spare shaft from my dead 1975 box that will hopefully do the job if it is damaged.

I’ll be ordering another bearing kit and seals from Motobins and another gearbox will get a new lease of life. This box felt pretty good, by the way. No crazy clunking, no false neutrals, no slipping out of gear and neutral was easy to find. It is likely that the last dog broke while in my care – I did notice a funny rattle that lasted for a few seconds while in neutral a few hundred kms ago. I had ridden that box for nearly 5000kms, most of it keeping up with Jeff (ME109) on some great rides through the mountains.

So how do you tell how good a gearbox is going to be? If it is out of the bike I’ll be picking it up and shaking it like a cocktail shaker, listening for telltale rattles as a first check. Running it through the bench testing would be next. I really don’t know what one should do with an unknown box apart from opening it up to see the condition of gears and to feel the bearings. If all is good, re-shimming and assembly would take half a day if one has the necessary tools.

Good bearings will obviously help every other part to work for longer and early detection of problems is important. I’m now wondering how long a gearbox typically lasts before bearings should be replaced as a matter of course.

[Mal S7]

Gday Charlie,

you should'a just sent that box of junk back to me. It did the intended job, glad it lasted 5000 kms. No you didn't bust that dog, when I was packing it on the back of my bike last April I turned it upside down and heard that same clack and thought it might mean trouble.

Thanks very much though for giving it a going over. Habrá mucho beneficio!! It will allow me to put the old box, which has been magnificent for 100 000 kms + and is still going strong, on the shelf.

I am sure we can come to a happy arrangement ... but don't be telling anyone that it was in exchange for head OK.

cheers
Mal

[qford]

Brilliant rundown on the gearbox rebuild - much great information and photography!

I'm just reassembling a '78 5-speed after replacing a bearing (first time I've ever been inside a tranny), and have run into a problem I'd like to run past you, if I may:

When I installed the shifting mechanism, it won't shift through the whole pattern, hanging up after clicking from 1st through neutral and (sometimes) into 2nd (sometimes stopping before the spring-loaded roller cam drops completely into the round groove for 2nd), sometimes locking up there completely (no movement up OR down).

Any thoughts, or information you could point me to?

Secondarily - maybe a problem, maybe not - the top (largest) gear on the one shaft has about 1/4" of up-or-down play in it that keeps the other shaft gears from hand rotating when gravity lets the large gear drop down to its resting height when the trans is on the workbench. Raising it up by hand that 1/4" allows both shafts to rotate.

Any idea what's up with that? Maybe the gear "dropping down" won't happen when the trans is back horizontal in the bike, but the idea of that largest gear sliding over that little bit and locking the gears up at speed bothers me quite a bit.

I DID NOT disassemble the gear shafts when I replaced the bearing, so everything should be as it came out, and I'm mystified as to why there should be that play in the large gear - as if there should be some spacer to hold it in its correct position.

Any thoughts or help would be MUCH appreciated!

[Airbear]

G'day gford.
I'm glad you found the guide useful.

On the first question, I had similar concerns with the 3 boxes I have now put together. It is tempting to try to test the shifting mechanism before the cover goes on but without the shafts being lined up properly (by the cover) and rotating as they will in use, it is almost impossible to get the forks to slide properly and the gear dogs to mesh. I assume you tested the shifting mechanism when it was out of the box - if that worked well and the forks went back as they should, all should be fine. Once the box is completely assembled it is still tricky to get all gears engaged. You need to spin the input and output shafts while selecting gears. A helpful extra hand is required to do this.

The second matter is more of a concern. I assume you are talking about the large gear at the rear of the output shaft; #15 in this diagram.



This gear spins on a brass (or bronze?) bush (#14) and is stopped from sliding along the shaft by a steel washer (#13). If this is missing you'll need to replace it. Check the fiche at Max BMW for your exact model. Also check the bush for wear or damage.

By the way, would you like the shimming spreadsheet?



If so, send me an email or a PM with your email address and I'll send it by return message.

[Ross]

I had hoped I would never have to open my gearbox... WRONG. So am reading this thread Charlie.

[Duane Ausherman]

Finding a broken off dog is common. It was our standard procedure to test for it anytime the filler plug was removed. We had an articulated arm with a magnet on the end. We would just stuff it in and under the gears and gently swish it around. It wasn't uncommon to come out with a broken off dog. We would just document it and go on. Sometimes we would find two of them.

This is caused by the error in indexing of the dogs or holes in the shift plate. We would always carefully inspect the wear on the dogs and the wear on the holes of the shift plate. In at least 95% of the time one would find concentrated wear on one or two of them and one or two with almost no wear.

I still have a fairly low mileage /2 gear set with almost equal wear showing on all dogs and holes. I could count on one hand with fingers left over of the times that we found this. I kept it some 40 years ago for my own use. Now, I won't ever use it, so it is up for sale. Nobody has inquired about it at all. The price is way under 1/2 that of new parts. This is due to so little understanding of this issue........... I think.

After some 10-12 successful rebuilds, I thought that I had it nailed down. However, by the time that I got to several dozen rebuilds, I realized just how little I knew. That was back before the Internet and now we have a simple way to share information. The learning curve is vastly shorter and we are all richer for it.

The problem that I now see is that we don't bother to use what is readily available. I see many questions asked that are answered on my website, but few bother to look. It is hard to have a BMW motorcycle question that isn't answered on one of several websites.

Google is your friend.

[ME 109]

The problem that I now see is that we don't bother to use what is readily available. I see many questions asked that are answered on my website, but few bother to look. It is hard to have a BMW motorcycle question that isn't answered on one of several websites.

Google is your friend.

Er, that's exactly what Ross has done, Duane. He's telling us about it, not asking questions. Yet.
I could give you an output shaft failure that I doubt you will find answers for on the net.

[Duane Ausherman]

ME, "I could give you an output shaft failure that I doubt you will find answers for on the net."

What is the failure?