Okay, flame if you must. I am completing my first gearbox overhaul. Didn't send it to a guru. Hey, a guys got to do his first one sometime, right? Things seem good. Shims in, cover on. Everything rotates with just the right feel of not-too-tight, not-too-loose. But in running through the gears, while I can get them all, I am also getting false neutrals, and here and there it takes an extra click-clunk to get into gear. Is this just a fact of life when hand rotating and shifting? Or, what do I need to go back and adjust? Gurus! I need your help after all! Side note: I used two methods for shim measurement, the typical machined plate and calipers route, and the mold-a-gauge from Cycle Works. I believe solder is a flawed method. Details to follow...
They can be tight when you first build them and turning by hand is never going to replicate the same action with oil in the box and engine running. It'll be fine! get on the road and give it some beans
Aw, how'd you know that "it'll be fine" is my favorite phrase just now. Thanks for that. I wouldn't mind some darker thoughts though. The yin/yang thing maybe. What's the worst that could be wrong? If everything's meshing nicely, I'm not sure what I would go in to tweak.
You don't say what year box it is, 4 speed or 5 speed, any up dated parts or how many miles on box currently. Since the bike in your sig line is an R75/5 I suppose the box is a 4 speed? No mater. The only test that counts is when you put it in your bike and take it for a spin. Are you ready now for the second one?
Oh, yes, right: 1977 R100S with a gearbox from Waterworld. I replaced the input gear and shock-spring which were the worst pitted parts. And all of the bearings. I also switched out the plastic shift bearing for a ball bearing. I will have to take that ride before committing to another, but it all seemed to go well. When I get a few minutes, I'm going to defend mold-a-gauge as the way to go for shimming.
You can verify the input and output shaft shimming without disassembly. For the output, pull the flange and see if you can move the shims with a pick. You can even bend a feeler gauge and see how you did. For the input, warm the cover slightly and see if you can push the shaft up and down. If it moves, you'll have to go back in. It is difficult to get the box to shift well on the bench; you can't spin the shafts fast enough to align the shift dogs.
Congrats on getting through it. No flames from me, I'm going to do my first one (a 76 R90/6 five speed) this winter. And I've got some of that mold a gauge stuff, too. On my trannie, the dogs for first gear are shot (and the clutch is slipping a bit at higher rpms, another story.) I have a complete output shaft assembly which I may just install instead of pulling off the offending gears/dogs. I've been looking for another sick transmission to practice on, but haven't found one in my price range (about $100 plus shipping.) I suspect knowledge is good, knowledge and practice is even better. I'll be interested in how your trannie works once you get it in. That's going to be the test and the proof of the pudding moment.
The Airhead 5 speed went thru many changes over the years. Some parts do not swap with out without swapping something else. Since you have a transmission apart be sure the output shaft you use has the groove in front for a snap ring and use the snap ring. Be aware that there is a change in the angle of the helical gears used (March/April 1982) There are three helical gears and the ones on the input and lay shaft are in gear all the time. The one on the output shaft is only in gear in 5th gear. The updated shift mechanism is considered an improvement by most. It is not cheap in the USA but can be about half our cost if you get it from Moto Bins in England. If the dogs on any gear are bad it is likely the holes they fit in are distorted also. Many Web sites to use, Snowbum, Anton Largiader, Duane Ausherman and a few others. (I'm on my way now to Turkey dinner.)
Hope your dinner meets expectations. When I bought the shaft, we did comparos on build date for the trannies. Turns out they were made within a month of each other in late '75 for '76 model 5 speeds. Hoping I've got it covered in the due diligence department. I've been collecting articles about fixing trannies. I've read Sno bums stuff, and have other writeups. I wish I could find a mentor in Tucson, but so far I'm not close enough to the to do time to make the trip across town to the local BMW dealer for a q & a session and to order parts.
The leeway in shimming is huge. The leeway in getting pre or post pressure face angle change helical gears is zero. I would read up on the helical gear pressure face angel change disston mentioned and make sure that you not only got the right one but that the other two already in there are the earlier PFA gears. To confuse the matter I swear I have seen newer late PFA gears without a X on them anywhere. You can barely see the difference IF you look closely. PFA is NOT the helical angle. They never changed that. PFA is about the shape of each individual tooth. I have seen the PFA's mismatched many times and they work just fine for a while. A couple of thousand miles is usually enough for mismatched gearsets to eat themsleves. They do make more whurring noise while doing it.
I found it doable, but I would ask first-timers if they have ever pulled and pressed bearings before. And more than once. And, do they know just how stuck a taper can be (man, my output flange did not want to come off, destroyed my homemade tool). And do they know their way around the measuring bit. Otherwise, crack into it. The worst that happens is you feel overwhelmed, and then you ship it off to whoever. Give yourself plenty of time, of course. The nuanced ability to interpret various things that a first-timer lacks are certainly real. Just how much wear here and there can be ignored, and how much can't, etc. And then the shimming. Experienced rebuilders will have proven practices, and preferred targets for the shimming that I just don't. Proof in that pudding awaits. As I said my opinion on the measuring is that mold-a-gauge is the way to go. I milled up a plate, checked everything for flatness with a good straightedge (and I know that's not just a ruler). I have good, maybe even very good measuring tools. I was pretty confident in the results, but a little squeamish about the variation point to point around a single bearing. I wanted to contrast that with mold a gauge just out of scientific curiosity. The results were different. The "measured" gap was a bit smaller all around, the mold a gauge gap larger. Not way off, but significant, and not perfectly consistent. So, I would have shimmed them looser if going by the "measurement" technique. My result might still have been fine. As SS said, plenty of leeway I guess. The compelling thing was just how perfectly the mold a gauge worked. Here we're these perfect hard little plastic rings that had been filling the very gap I was measuring. They obviously flowed all over as they were squeezed, and this is why I think solder is flawed. If your gap is 1.5mm, and you have 2mm solder, well, sure, you are going to get a good result. But if your gap is slightly smaller, that solder will reach a point where it will start distorting the cover, and I believe it would be easy to miss that. You would have to have very good touch when tightening the cover to feel that the solder had reached its plastic limit. In my case, one shaft had much less a gap than the other two. Solder would have been fine for two shafts, and the third would have distorted the cover, then distorting all three measurements. Only mold a gauge will allow measurements from virtually zero all the way up with no distortion. Now, it's possible that this stuff (and as soon as I saw it I thought, i know what this is, of course, why didn't I think of that! But you should order from Cycle Works for the valuable instructions) this stuff might have some wildly large thermal expansion coefficient. That would be a problem, but I bet it doesn't, and the error would not be large anyway. The only downside is the oven. I have an electric kiln, so it was easy. But think of all the tools, plates and whatnot that you don't need. YMMV
I guess I don't quite grasp what your point is--I'm not trying to be a smart ass or snarky here. I'm not singling you out, but there does seem to be a trend towards making this all a bit more complicated than it needs to be. I know that 74/75 were years of change in the gearboxes--and that parts from a 75 probably might be problematic in my 76 five speed. If, as Disston says, 82/83 was a change year for the helical gears, I should be golden for helicals if both my replacement shaft and the offending shaft are from the same displacement and model year, no? At the very least as the offending gears are the 1 & 2 gears on the shaft (last ones before it outputs to the flange) I may be able to just pull those bad gears off and put the good ones in?
685, You're doing fine mate. What SS is getting at is theres a thin gear and a fat gear and sometimes people mix them up. Later 17.5° tooth on the left. Earlier 15° on the right. I've seen a few mixed up and as SS says it's ok for a while but wears the intermediate and the fifth gear eventually. Theres supposed to be an X on each of the gears when they're all 17.5° but often there isn't. They also have a habit of putting the X on the layshaft on the inside face where you can't see it. When they're mixed up they do make things tight. To be honest the best thing you can do is whack the end casing with a rubber mallet. This will settle things down.
I think it is a hard concept to voice is the problem. A really big change in the middle of 1982 production was the helical gears changed from 15 to 17.5 degrees and it is hard to see this with the naked eye. It is not the angle you first notice when you see these are helical gears but it is the shape of the teeth. It is hard to impossible to find the 15* gears if you need them. I think only the 5th gear is readily available. That's the one on the output shaft that riders are changing all the time to a higher ratio (after market part). So people have installed a 17.5* gear not knowing better or not caring. I have seen it said that extra noise was the consequence implying this was the only consequence, and it should be noted this is not the only consequence. The 17.5* gears are supposed to be marked on one side with an X. Many reports from many sources of 17.5* gears that were not marked. So it behoves you to pay attention and if you find a gear with out an X know where it came from. Sometimes the only solution is to replace all three helical gears and since the one on the middle shaft is only available with a new complete lay shaft the price really starts to add up. I do wish there was a readily available gauge to measure these teeth with. Maybe there is. I've never heard of it's being used. There are gauges for measuring screw teeth but I think that is something a little different. BTW, the last gear on the output shaft is first gear but the gear next to it with dogs on both sides is 4th gear.
I think I knew that. I was referring to gear position as it comes off the output shaft. I'm kinda thinking that as long as I don't try to "make things better" by mixing parts from different years and models, I should be in the ball park. I don't have any sort of press yet. I've seen the pic of the press tool for pulling off the first (last one before the flange on the shaft) bearing. That bearing looks like it might be an impediment. On my replacement shaft assembly, it's already been pulled and wasn't included when I bought it. I'm thinking rather than trying to source the tool, I'll just spring for a replacement bearing (which is where all the info and worry about shims and end play come in.) My understanding is that it's not too tough getting the bearing back on--it's getting it off that's a problem. Plan B is that I have a friend at an automotive trannie shop that could probably safely press it off. BTW, thanks to Rob & Disston, et. al. for the input. I'm not squeamish about trying this. And there is a very good BMW dealership in town, Ironhorse BMW, if I end up out of my comfort zone.
The output shaft bearing is easy to pull using a bearing splitter. You can pretty much do all the bearings with a cheapo bearing kit off ebay. Whatever force your using to pull the bearings is a fraction of what they take when you're using the bike. Saying that I hate reusing bearings. It was never meant as a definitive guide or to be techy but a few of us threw a thread together on UKGser after an airhead weekend to show the guys who went what was inside the gearboxes. It's just a whistle stop tour through a gearbox and a few of the issues http://www.ukgser.com/forums/showthread.php?271570-Airhead-Gearboxes
thanks for the pic of the bearing splitter. I don't know why the pic didn't show up in the thread. I can get one of those at Harbor Freight for about $30--that's what? Like 20 British Pounds?
I own some HF tools. They are good enough but I haven't checked the pullers they have much. I do own a Chinese blind hole puller that I have used some. I mostly own Snap On or OTC pullers. The OTC Stinger line is also mostly made in China but is top quality. As good as most made in the uSA tools. This is the OTC 4518 puller kit. It is only rated at 5 tons but I think is big enough for all the bearings in a 5 speed. It has a medium and a small bearing splitter. This is the OTC 4517 puller kit. This puller is rated at 7 tons. The 4518 sells for about $74 on Amazon dot com and the 4517 for about $95 on Amazon. Shipping is usually extra. I post these so you can compare to what you see at Harbor Freight. In pullers as in most other tools the name brand does mean something. A press, even a borrowed press, is the easier way.
