In early 2002 I learned that it was possible to change the ratios of both first and fifth gears in the GS transmission by swapping out gears. Surprisingly, only one of the gears of the respective pairs need be changed. I called Cal BMW and Ozzie's BMW for pricing, and found that each wanted almost $300 per gear for the parts, plus labor to do the swap. That was prohibitively expensive,and the idea went onto the back burner. Several months later I was leafing through the back pages of a Classic Bike magazine Gene had loaned me (it was the issue that profiled Norton's Production Road Racer from the '70s), and I noticed an advertisement from S. Meyer in Hillesheim, Germany advertising BMW parts. Having previously dealt with many parts suppliers in Great Britain, I composed an email to S. Meyer inquiring about the gear swap.

6 % lower 1st gear Euro 90
18 % lower 1st gear (Set 2 gears) Euro 250
6 % quicker 5st gear Euro 120
10 % quicker 5st gear (Set 3 gears) Euro 620

At the time, a Euro was going for about $.95, so these prices represented a vast departure from what was being quoted locally. The gears were made by a company called Kaiser. I FAXed a copy of my Visa charge number along with a note explaining what I required (6% lower 1st gear and 6% quicker 5th gear) and in about 3 weeks the parts were sitting on my front porch.

In the mean time, I started local inquiries for someone to do the installation, along with a general reconditioning of the transmission: replace all bearings and bushings, replace shifter spring (prone to breakage), etc. One of my buddies suggested I call Bob Grauer, who seemed quite knowledgeable and willing to undertake the work. I told Bob the parts were on the way, and that I would contact him again when they were in hand. When the gears arrived, I emailed Bob asking when I could bring the transmission by, and he indicated, by reply email, that he was no longer interested in doing the work. I called him, asking for some elaboration, but all he would say is that he wouldn't do the job, and that I shouldn't have any trouble finding someone else. I still have trouble adequately expressing my shock, dismay, and anger at this behavior. Absolutely unprofessional to put it mildly. But enough said.

Gene suggested I try Dave Gardner at Recommended Service, an independent mechanic he'd been using for years for tires and, more importantly, transmission work. Dave was interested, but was concerned about the quality of the parts until I mentioned they were sourced from S. Meyer. Dave was familiar with the company, explaining that they had partnered with BMW in the '40s, '50s, and '60s to assemble some of the most successful race bikes in the company's history. We set a date for the next Friday, and I got busy pulling the transmission from the GS.

I won't go into the removal process here, except to say that it was very nice that BMW uses the frame to support the center stand on the GS, and unlike the K series, it isn't necessary to rig up some sort of auxiliary support for the rear of the bike when pulling the tranny. The job went very smoothly, and after three or so hours I had the transmission on the bench. I hate working on dirty parts, and I figured Dave would appreciate getting a clean transmission case. I Gunked it, first sealing off the speedo drive hole with duct tape. After drying, I went after the corrosion on the top (where the air box sits) with a brass brush and contact cleaner. It never did get to silver white, but the crusty white deposits were removed.

Recommended Service is located in San Francisco, occupying one bay in a fairly large industrial park. When I visited, the shop had a half dozen or so BMWs in varying states of repair, and most were of airhead (R) design. I was impressed by the shop's neat, clean appearance, with everything in it's place and no stacks of junk lying around. Even the floor had a shine. Dave was out running an errand, but his wife Gaby wrote up a work order and fixed me up with a comfortable chair and appropriate reading material while I waited. It didn't take long for Dave to come along, and after introductions I showed him the gears and transmission. He seemed pleased with the look of the gears, but did express one concern: he wasn't sure he'd be able to press first gear off the shaft, as his press only went to 20 tons. He also indicated that, since I was supplying the gears, he would not be able to guarantee how well they would work. That seemed reasonable enough, and if we had trouble with the press, at least I'd have a reconditioned transmission and a taller fifth gear at the end of the day.

Dave washed down the outside of the transmission in his parts washer, then began disassembly by heating the end cover with a pair of heat guns. Out came the screws, and after a few taps with a soft faced hammer, off came the cover. The inside of the tranny was spotless and shiny, good news to me as I'd seen plenty of rusty parts inside the various Norton transmissions I'd worked on. Moments later the first gear shaft was in Dave's press, and we knew we were in trouble. Even with a three foot cheater bar on the lever, the gear wouldn't budge. Ugh! Undeterred, Dave got on the phone, and after a couple of calls to his neighbors in the industrial park, got ahold of "Cowboy". Cowboy didn't like strangers, but a passion for all things mechanical had kindled a friendship between he and Dave. More importantly, Cowboy had a 120 ton press, and if that wouldn't do the job, nothing would. Off we went, down about a block to a plain steel door, a couple of thumps on which were answered by the Cowboy. He turned out to be a middle aged, well muscled, long haired, bearded character that would have been as much at home wrangling cattle as he was at rebuilding classic cars. Very soft spoken, bordering on shy, he led us through the dim recesses of his shop, which contained a jumble of cars and associated parts, most of them from the '50s and '60s, and all of them dusty from the body work from which he derives his income.

At the press, Dave put together the supports needed for the shaft, and Cowboy manned the press lever. A few slow strokes and the gear was moving on the shaft, emitting a low groan with each movement. The pressure gauge said 30 tons as the shaft and gear parted. We didn't want to impose on Cowboy's time, so there wasn't an opportunity for pictures of the two gears side-by-side. Rearranging the supports, a few more strokes of the lever had the new first gear in place, again accompanied by the same groaning noises as during the removal of the BMW part. It was smiles all around, and as we left, I asked Cowboy what his favorite beer was. He answered "Corona" with a somewhat puzzled expression, and we were out the door with profuse thanks.

Back at Dave's shop, it was about lunch time, and the caterers next door had dropped off a couple of extra meals. Enchaladas and black beans with a wonderful fresh salad! Dave offered one to me, heavily twisting my arm until I said "yes" (not!). We shot the breeze as we dined, and in between mouthfulls I learned that Dave's passion these days is ultra light airplanes, which he pursues from a hanger in the far east bay. The motors are suprisingly BMW like, and his insistance on perfection complement the reliability requirements imposed by flying.

After lunch, Dave removed the various bearings and bushings from the transmission housing and shafts, cleaned out the inside of the housing with solvent, then hot water and soap, drying everything with shop air and soft towels. He reminded me of a surgeon, what with his rubber gloves and penchant for cleanliness. It gave me a vary warm feeling to see how particular he was, as I tend to be that way myself.

With everything clean, Dave began the reassembly process, which included re-shimming the end play of the shafts. Dave uses a special plate to keep alignment of the shafts constant and enable him to measure the overall length of the various stacks of gears and bearings. There's some math required to get the right shim height, and as we learned, it's a lot easier to get the right number when nobody's pestering you with foolish questions. Dave got it right after I grabbed Gene and walked him out the door for a few minutes.

With that challenge surmounted, and a slew of new parts safely tucked away inside, the tranny was buttoned up and ready for re-installation. The only task remaining was a quick trip to the liquor store down the street for Cowboy's Corona, which he received with a wide grin and surprise in his eyes.

As you can see from the photo above, reconditioning involved the replacement of five bearings, two bushings (one bronze, one nylon), four seals, one circlip, and two springs. Including parts, the total came to just under $400, an amazingly low number considering the futzing around that was needed to move first gear onto and off of the shaft.

Initial impressions are that there's just a bit more steam available in top gear, making passing a little easier and riding less buzzy at freeway speeds. And down in first gear, while the change is less noticeable, chugging along over rocks and other rough stuff it is a bit easier to keep the engine from stalling, and parking lot (or MSF) slow speed maneuvers are easier, too. Would I spend $600 to do it again? Given a transmission with 45,000 miles on it, from a previous owner that gave it an unknown level of service, I think my answer would be yes. I'd been contemplating the rebuild anyway, so in reality the added cost was about $200, and again, the difference in performance is noticeable and worth while.

By the way, with almost 10,000 miles on the conversion, everything is still ticking along just fine. But I couldn't figure out how the ratio change could be effected by changing only one gear. The Cycle World magazine has a technical column each month, so on a lark I popped off an email asking how the change was accomplished. About a week later I got a reply from Paul Dean, Cycle World's editorial director, along with a note from Kevin Cameron, CW's technical editor. Kevin's reply wasn't too specific, but Paul threw down the gauntlet when he noted "So, in other words, he doesn't know, either".

A few days went by and a lengthy description from Kevin appeared that demonstrated, that while he may not have known at the time of his original response, he was able to find references that explained it to him. More importantly, he was able to interpret those references and translate them into terms that I was able to understand. Here it is:

Paul Dean forwarded to me your surprise at discovering that a given gear, on a fixed center distance, could somehow mesh with other tooth numbers than its original mating gear had.

I too was similarly surprised in 1969 when I saw that Kawasaki had done the same thing with the conversion from the stock A1 gearbox to the closer ratios of the A1-R. But it turns out that by playing clever games with tooth depth and thickness, just such an outcome can be successful. Normally, in any standard gear pitch system, you would expect that the sum of the tooth numbers of each meshing pair would be constant, straight across the gearbox, because all are at the same centerline distance from each other.

Gears have an abstraction associated with them called the "pitch diameter", which is the circle of such radius that the meshing pair, if reduced to friction wheels of their two pitch diameters, would turn at the same speed ratio as their tooth numbers ratio. In standard gear cutting, the teeth are cut in such a way that this pitch diameter is predetermined at a certain dimension - typically something like 2/3 of the way down each tooth face. But by changing the depth at which teeth are cut it's possible to make this abstract circle somewhat larger or smaller, and to compensate by similar cutting tricks on the mating gear. Liberal use of this is made in producing the many alternative ratios - all very close to each other - used in the gearboxes of road racing motorbikes. It's quite common to have 4-5 alternate ratios available for each speed in a 6-speed gearbox - just so the rider is never caught at one particular corner, in that awful place about 250 revs below where the engine really starts to pull well.

If you'd like to pursue this subject I'm sure you can find standard titles on gear design and manufacture in the library of any engineering school. Wear tweed and no one will look at you twice - I know it works at MIT's Aero & Astro libe.

KC

Kevin's note included a return email address (no, you can't have it), and detailed though it was, didn't address my main concern about the swap : reliability. I penned a quick response by way of thanks, and received yet another even more detailed answer:

The Holy Grail of gearing is so-called "conjugate action", which means that at all times a one-degree rotation of the pinion must produce the same in the gear (naturally multiplied times the gear ratio). This is pretty much built into the tooth shape. As you would expect, any deviation from conjugate action would introduce cyclic loading, which would be not so good. On the other hand, power gearing connected to an internal combustion engine is heavily cyclic anyway, and this is taken into consideration in the experience of the gear designer - a 1000-cc twin, for example, would need slightly more robust gears than would a 1000-cc four. In the case of the great Rolls Merlin aircraft engine, a torsionally compliant quill shaft connected the front of the crankshaft to the pinion in the reduction gear, as a means of smoothing out even the already quite smooth power of a V-12.

Another aspect of gear durability is the so-called "contact ratio", which is a measure of how many teeth are carrying significant load at one time. The more, the merrier, and this is easier to achieve with helical gears of the kind now being introduced in the new 1200 transmission. This is a special concern of those who must design helicopter gearboxes, for example.

Gears are made of steels that can both be adequately surface hardened and core-heat-treated to produce overall toughness. Typically the surface hardening has a depth of somewhat less than a millimeter, but the whole gear cannot be made equally hard because it would then be brittle. Correct heat-treatment and surface hardening depth are more variables to worry about - I watched a friend have alternate ratios made for his Yamaha by an outfit that normally made car gears, and as I'd feared, they batched his gears through with a load of normal production. His much smaller teeth were case hardened to a depth excessive for their size, and were therefore lacking much tough, fracture-resistant core. When he operated them, they broke.

Modern gear oils are wonderful stuff - they contain additives that form metal chlorides or phosphides by reaction with metal surfaces - layers that have quite low friction themselves and which, being less strong than the steel underlying them, can be sacrificed in conditions of extreme pressure. The result is a self-polishing action which reduces the heights of asperities and over time improves the tooth surfaces. This is why gear oils have their typical stink. In racing applications in which the gearbox is separately lubricated from the engine, I have always used gear oils in preference to such notional alternatives as automatic transmission fluid, 20-50 motor oil, Castrol R, &c. I have noticed that in Japanese engines designed to use a common oil for both the engine and gearbox, the gears tend to be rather larger than they would be if they could be lubricated by proper gear oil.

KC

While he didn't directly address my question, he did outline the factors involved in reliability: cyclic loading, contact ratio, heat treat, and lubrication. The BMW twin loads the transmission in a cyclic fashion anyway, so presumably the tooth profiles are already designed with this concern in mind. Fifth gear in the BMW tranny is a helical cut gear, which Kevin indicates helps with the contact ratio. I can't speak to the heat treatment that the gears received, but they do seem to come from a reputable, German, source. And I've been using Redline synthetic oil in motorcycle transmissions for years, and have noted only light polishing on the tooth faces I've examined. And in his first note, Kevin points out that factory road racing bikes can have several ratios for any given gear, some provided by just this technique. I conclude that even though the tricks being played to achieve the alternate ratios aren't widely known among garden variety motorcyclists, racers and mechanical engineers have developed techniques that have shown themselves to be reliable. And while I hesitate to tempt fate, my experience so far has shown that there is no increased wear (as shown by a lack of metal whiskers on the magnetic drain plug at last transmission oil change) associated with the new gears. Coupled with numerous reports from other GS owners that have accomplished the same modification, and I'd say this one is a winner!

Sources:

S. Meyer
Uwe Mayer
Bahnofstr. 22
67586 Hillesheim, Germany
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it
FAX: 0114967338134#
PHONE: 0114967338195

Recommended Service
Dave Gardner
1064 Revere Ave.
San Francisco, CA
(415)822-2041