Point Seven Five, thought I'd post a couple of pictures to show where I'm coming from regarding the shim.

The first one clearly shows wear on the shim from the inner bearing race resting directly on it.



This photo shows a little wear on the gear facing from the shim. Note that it is unequal and goes only about 65% of the way around.



This photo shows unequal wear on the side of the shim facing the gear. It pretty well matches the wear on the gear face.



Nothing like developing a good neurosis!  The bottom line is that I'm going to just re-install the shim as it was and be done with it, unless of course someone can feed my neurosis with something like 'Have you been adding an appropriate amount of blinker oil?' 

Another couple of weeks to go before parts arrive.

Mr Rhodes Rash - I like your photos....

I spent hours on an extensive post for this thread recently, then promptly lost it before I got to send - guess it must have been bullshit!

I'm not following what Mr .75 is saying about the shim affecting the position of the bearing. My understanding is that the bearing inner race goes up against the locating shoulder of the shaft and that's where it ought to stay (interference fit - as long as it hasn't been on and off too many times without creating the necessary temperature differential to allow the inner race to move easily on the shaft). I put the stuff in brackets because BMW changed the standard bearing inner race retaining circlip for a more fancy one that applies pressure to the inner race - i.e. presses it up against the shaft retaining shoulder. They did this some time between '86 and '88 - Inge will tell us more exactly - presumably they figured that an inadequately retained bearing inner race was a possible contributor to the issue of knocking noises originating from the output shaft on the K75

The shim goes between the  anti backlash gear - the spring loaded one which Motorbike Mike calls the idler gear that he has removed from his K100 track bike - and the bearing inner race. The face with the 60% wear and the corresponding pattern on the idler gear - I figure this is due to the open ends of the C spring exerting a separating force on the driven and idler gears - you can feel this when you reassemble (at least I did). This is why I think it is easier to measure the clearance accurately on a trial assembly with the C spring removed (left out for the clearance checking exercise). I used two sets of feeler gauges 180 degrees apart to ensure that the clearance is equal all the way around when determining the 'go / no go' scenario. In the case of the '88 K75 we rebuilt, the shim was too big even though it was a high mileage machine which looked like it hadn't been opened before. The good thing was that this meant all the wear steps were polished out of the shim in the process of us getting it down to the right thickness using a piece of glass and valve lapping paste.
Have a look for my post on a different thread, with photos of the notched wear pattern that we found on the contact surfaces between driven and idler gears. The more I consider this issue, the more sure I am that this wear pattern would have adversely affected the ability of the anti-backlash arrangement to do its job. Mike doesn't hear any noise on his track bike because he never accelerates SLOWLY up through the rev range!!!!

The minimum required clearance (end float) for the idler gear which is determined by the shim, is definitely in the Clymer manual that my mate had when we were doing that job.

I will confess that on my engine I did not look all that closely at the shim in question.  When I disassembled those parts I did not notice any untoward looseness or wear patterns on the parts.  When I saw that the backlash spring was still perfect and tensioning the gears I just put everything back together and moved on to other areas. 

As I have stated above, my feeling now is that the backlash is not the source of the noise in my engine.  As I have done nothing to the backlash spring or gear, if the noise is gone after this, I will assume that in my case the noise was related to the balancer shaft being loose and banging around.

Hello all,

Very interesting thread, same problem with my K75 basic, its in bits now awaiting parts.

The photos shown could be of my parts!, same wear marks, same area's, which is interesting as it shows the issue to be pretty consistent, at least in a few engines in this thread.

I will post some photos soon, but one area that is causing me a lot of concern is the actual mating faces where the output/balance shaft contacts the drive gear, there is a pronounced witness mark, polished area, on the balance shaft, I assume as its a softer metal than the actual drive gear, which shows the very faintest of marks.

With out photos its very hard to explain!, but the witness marks show definite rotary movement, allowed by the slack in the fit of the driven gear two protruding dogs to the two slots in the balance shaft.

Sorry if I am making this as clear as mud!, its late and with out photos.....

I think this is proof positive that in my engine at least, there has been definite movement, but strangely, the two locating dogs? tabs?, show no sign of wear/impact, same in the balance shaft slots/cut outs.

I shall post some photos soon, will make more sense, just some thoughts before I go to the land of nod.

All the best,

John

Engine is back in the frame and most of the peripheral parts have been installed.  Injectors are being painted right now.  Should have it running this weekend and will report on if I have been successful in stopping the noises.

Hi John
I was using the term drive gear for the gear that is integral with the crank shaft and 'driven for the one on the output shaft but this of course ignores the engine breaking mode...
I understand you to be saying you have some evidence of relative movement between the anti backlash gear (the narrow one which is spring loaded) and the narrow journal on the drive/driven gear boss which it runs. This is normal - its how the anti backlash assembly works.  For a start, the backlash between crank and output shaft gears varies with engine temperature plus the C spring between the wide and narrow parts of the output shaft gear is unlikely to be stiff enough to have the narrow gear carry the load in engine breaking mode. I envisage the design to be more of a shock absorber - it reduces the impact forces and noise that would be generated by unhindered backlash where the gear teeth will bounce backwards and forwards (rotationally and relatively) when gears are cold / worn / oil is thin or full flow is not yet established. 
I bet Mike uses full synthetic oil and warms his bike up fully before he hits the track and wangs that throttle wide open (excuse the Zimbo slang)

OK, put the engine back in Ilsa yesterday and after a bit of troubleshooting to find a connector that I forgot I got her fired up this afternoon.

The nasty clank at idle is totally gone, and she now sounds as good as my K75S.  Smooth as silk with a bit of gear whine and a little valve noise.  I am now sure that the heavy metallic clank at idle up to about 2000 rpm in the k75 is due to slippage between the clutch basket and the balancer shaft.  The only thing I did that actually related to the noise was to clean the mating faces of the two parts and torque the clutch nut to the top of it's specification. 

I'm equally sure that if it ever comes back I can fix it without pulling the engine.  Just remove the clutch basket and spray some brake cleaner onto the face of the balancer.  Clean the face of the basket and put it back together with 100 ftlb of torque on the clutch nut.   In fact, I plan to tweak the clutch nut every time I go in for a spline lube. 

The only other thing I did was replace the alternator drive dog bearing, and as a result the alternator growl is now reduced.

Top result .75.... You must be well chuffed.
Cheers

Well chuffed, Kaptain, and then some.  The old girl makes some really sweet sounds at idle now.

Good work Point Seven Five

Enjoy the music!

If the music stops - tighten the clutch nut again... but remember there are three parts in that assembly excluding the parts bolted to the clutch basket. If it happens again it would be interesting to have you describe the process you have to go through when you first loosen the nut. When we were stripping a 75 with a higher rpm rattle, we had to use heat on the nut and I still wished I had a 3/4" drive socket instead of my 1/2" drive STAHLWILLE.

High rpm noise is probably something different.  My problem turned out to be at low rpm and seems to have been cause by slippage between the clutch basket hub and the balance shaft.  As near as I can tell it is a problem unique to the way the output shaft is designed in the K75.  The design of the K100 output shaft precludes the problem I saw with the K75.

With the k75 output shaft being protected as it is in the intermediate housing, it is hard to imagine a clutch nut being that hard to loosen as long as it is torqued to 100 ftlb.  With a 6 point socket and a 20" breaker bar I had no trouble loosening the nut.  Based on the evidence of the clutch having been disassembled I suspect that the problem was the result of a film of oil getting in between the faces of the clutch basket hub and the balancer shaft while things were apart to replace the seals.

My advice for anyone going in there on a K75 is to clean the face of the clutch hub and spray some brake cleaner onto the face of the balancer shaft immediately before assembly.  You gotta get the highest possible coefficient of friction between those parts.  Once they're tightened up the fit between them should prevent any oil getting between them.

Thanks for the info .75

What's your take on using loctite on that clutch nut?

You should not use loctite on that nut it has its own locking mechanism which I reset by squeezing the side of the nut and reuse them, never had a problem.

+1 on the Loctite.  100ftlb is hard enough to break loose without the added resistance of Loctite. 

I like the idea of reusing the nut.  Same with the clutch assembly bolts.  I would use a drop of medium Loctite on them and reduce the torque a couple ftlb, per Chris Harris, to put a hair less stress on them.

BTW, you can get the clutch o-ring at any bearing supply.  The rear main seal and bearings too.  Local prices are a little less than BMW but not that much.  It does save postage and gets them in your hands faster if you have to mail order from an out of town dealer.

Well done Point-Seven-Five! I was called away for a week and getting back home I’ve just read about your success. Given the similarities between our bikes’ problem and what we’ve found (or not found) upon digging into them I’m feeling re-enthused about getting back to work on mine. My enthusiasm had waned badly. I expect my parts to arrive this week and I can start the rebuild. 
You, as well as others, have put lots of good thought and advice into this thread. Thanks everyone for your input. Having said that, I’m sure you haven’t heard the last of me as I start to put things together.

Hello all, back again, sorry for silly late reply.....

Well, I have spent some time examining the parts from my engine, all the wear marks are the same as the photos in this thread, very consistent.

Hi Rawdonball, sorry my explanation was as clear as mud!, I can't seem to post photos yet, do I need a certain number of posts first?, The parts I was referring to, if you look at the exploded drawing in the Haynes manual, output/balancer shaft K75;

Part number 5, the balance/output shaft, the faces where it mates against part number 6, the drive gear, my parts show a polished witness mark on both faces, more pronounced on the balance/output shaft, due I guess to the drive gear being so much harder.

I spent about an hour lapping the two surfaces together with some valve grinding paste, so darn slow due to the minimal movement allowed by the gear dogs to balance/output shaft grooves.

I ended up with about twice the contact area, so am hoping it can only help to keep the two parts from moving, semi rotary movement, once the clutch housing retaining nut is fully tightened.

Point seven five, I think you have got it right, I could not find any thing wrong in my engine, apart from the polished witness marks mentioned above, which I think is proof positive that movement was occurring in that area, allowing the gear dogs to clatter in the shaft grooves.

I did find a bit of roughness in the alternator bearing, the drive bearing that fits in the bell housing, which I will replace, other than that, all seems very good, I took the parts to show the boss of our local engineering shop, he could see no issue with wear, just a few polishing marks he said, but did agree on the witness marks on the balance/output shaft to drive gear faces.

He did comment on not being surprised as to the design, the loose fit of the dogs and grooves, he said its a modern way of design and manufacture, he said so many engine parts these days rely simply on the tightness/torque of the retaining nut/bolt, things such as camshaft sprockets being held in tapers with no key-ways! etc.

So, now to re assemble, keep those faces spotlessly clean!, I have a new anti backlash spring, and am re using the thrust washer, which has clearance to the large bearing, numbers 10 and 14 respectively, in the above mentioned Haynes drawing.

The old and new anti backlash springs look identical!, have not tried to measure the tension though.

Thanks for the update point seven five, very heart warming to hear you now have a quiet running engine!, I shall hopefully be able to report in due course the same outcome with my engine!.

I wish I could of posted some photos, but there does not seen to be an option to attach them, I am guessing I do not have enough posts yet?.

All the best,

John

Good luck with the reassemble!  Here's hoping it comes out as well as mine did.

cheers,

Hey Pond Snail,

Use the search function on the 'portal' page to find threads on posting pictures. Also find other related threads and if you have the energy, read my pet theory on impaired operation of the anti-backlash arrangement as a source of noise from drive gears on crank and output shaft gears

The nasty clank at idle is totally gone, and she now sounds as good as my K75S.  Smooth as silk with a bit of gear whine and a little valve noise.  I am now sure that the heavy metallic clank at idle up to about 2000 rpm in the k75 is due to slippage between the clutch basket and the balancer shaft.  The only thing I did that actually related to the noise was to clean the mating faces of the two parts and torque the clutch nut to the top of it's specification. 

Thanks. I know someone with a K100 which has a very loud metallic knock which sounds a lot like this description.

It sounds to be coming from just under the starter motor.

If it is a K100 the noise source will be different.  In the K75 the output shaft is different and the noise is the result of the unique way the balancer shaft is assembled to the output gear and clutch basket.

In the K100, I believe the source of the noise is damaged cushions in the shaft's coupling to the clutch assembly.  These output shafts are one of the main differences between the K75 and K100(besides the number of cylinders).  If you can find an exploded diagram of them for comparison you will see how different they are.

Thanks, the guy has put his bike into a garage now anyway. I'll see what he makes of it.

Just an update to my post above, have now started reassembly, balance/out put shaft is back in, clutch carrier/housing torqued up, getting there!.

Here are a couple of photos, pretty self explanatory,

First shows the witness marks on the shaft and gear, which I think are proof positive of semi rotary movement between the parts,



The second, shows parts after about an hours worth of lapping with valve grinding paste, hopefully the greater contact area will help,



So, hopefully I will be rewarded with a quiet K75!, will let you know how it all turns out!.

Hope I am not hijacking the thread with this!,

John

I have been reading this thread with a mix of amusement, horror and interest.

So can we assume that the moral to the story is (if you do go after the noise)

output shaft spring clip 
bearings
shim
nut to 100ft/pounds and no oil

The pleasure awaits me

Congrats to all of you who have been through the dark days and succeeded.

I guess treatment under previous owners is both an unknown and significant. 

Regards

Guy

I started this thread several months ago. Sorry I’ve been silent for such a long time. I fell into the trap of not fully anticipating what bits and pieces I would need to replace, and felt more comfortable about replacing just because they were in front of me (seals, o-rings, a couple of bearings, etc, etc) and as a result was ordering parts as I went along. As a result, while waiting for orders to arrive (one got lost in the mail) all the spare time that I had during the winter evaporated with the arrival of spring.
 
The good news is that everything is back together, the bike’s on the road, and the engine clatter I was chasing is gone! As Point Seven Five and others had discovered with their K75’s everything inside appeared to be in factory-new condition, and I have to agree that the problem arises from torque relaxation on the output shaft/clutch housing retaining nut. I had never re-torqued it since new.
 
Here are a couple of observations I made during the project:
 
1. I replaced the output shaft rear bearing (the large one) as I (and others) felt it was showing a little wear. On re-assembly I put the anti-backlash gear assembly in the freezer for several hours and the bearing in the oven for one hour @ 90 C. The bearing didn’t go on particularly easy and when tapping the bearing onto the shaft the gear assembly sprang apart (one of the ends of the anti-backlash spring became disengaged from its mounting pin). I had to pull the bearing off and start again. I put the gear assembly back in the freezer for several hours and put the bearing in the oven, this time @ 100 C. Prior to mounting the bearing I clamped the gear assembly together. Because of the higher oven temperature, this time the bearing literally fell into place on the shaft. No need to tap the bearing on, so ... hotter is better! I still think it was a good idea to clamp the gear assembly ... just in case.
 
2. I replaced the output shaft oil seal, not because it had been leaking, but because it had been disturbed and was 25 years old. The new seal is slightly different from the old one and is brownish in colour. Moto-bins says to pre-soak the seal in oil for two hours. I assume this is to soften the sealing compound that coats the outside edge of the seal. There has been talk about not mounting the seal flush with the crankcase face, but to leave it extended by 1 mm. (as per BMW service bulletin from days of yore). I noted that on the old seal (original from 1991) the distance from the seal face to the inside sealing lip is ~7.3 mm. On the new seal that distance is only ~5.3 mm. This negates the need to extend the face of the new seal by 1mm. It can go on flush and achieve roughly that same result (a 2 mm repositioning of the seal lip).
 
3. I’ve run synthetic 5w50 oil in my ‘91 since the first oil change, on the advice of the German dealer I bought it from, because of the hot dusty places I often travelled to at the time (Morocco, Turkey, etc.). On engine disassembly my spragg clutch looked like it had just come out of the factory and I attribute that to the synthetic oil and religious 5000 km oil and filter changes. So I was tempted to not bother opening it up to clean it. My curiosity got the better of me and I’m glad I did. A reasonable amount (not a huge amount) of black sludge was built up inside. Probably I could have gone through the rest of my life, and the bike’s, without it being an issue, because I’ve seen some horrid pictures here of gummed up spraggs, but the sludge is there regardless of oil considerations. So ... if you have the spragg clutch out, for whatever reason, take the time to clean it! It’s not a big effort.
 

4. I don’t have a mechanical background, but I’m not afraid of trying things. I’ve done my spline lubes a number of times and had no difficulty with it. With this project though I was into uncharted territory as soon as I went past the intermediate housing and I really felt I was exceeding my capabilities. In short, I was kinda scared about what I was undertaking. What I’m getting at is if you go about a task like this diligently, have a good repair manual, ask questions and follow some good advice I would bet that most people could do this job. This site is a vast source of good information and the people here have no qualms about helping out. Thanks to everyone.

Update: It's now 5 years and 26,000 miles since I tore Ilsa's engine down to chase the clank.

Torquing the clutch nut quieted things down, but about 3 years later(about 20,000 miles) it started to loosen up again. An accident, rebuild of a totaled bike and restoration on another has kept me from going back in tighten up the clutch nut again.

The noise is back, but not as bad as before(maybe I'm going deaf or just getting used to it), so I've had no urgency to do something about it again. If anything, it will be going on Ilsa's list of winter maintenance jobs. Next time I think I will try a little bearing locker on the mating faces when I torque it up.

Hi Point Seven Five, 

Is there any merit in this idea; replacing the circlip and anti rattle spring, even if they do look good? They both provide spring pressure and have been under compression for say thirty years, they might have lost some tension by now. They're both open ended, so might not be as durable as your typical diaphragm spring, and difficult to tell if they're past their use by date. Especially compared with the bearings, which probably also want to be replaced anyway. 

Just a couple of thought bubbles from a carpenter who sometimes asks himself "What ARE you doing offering mechanical opinions on a motorbike forum"

Any chance of some photos if you do go in there?

I purchased a replacement spring before going in, because I thought that was the problem. When I pulled out the original spring I found there was no perceptible difference between old and new. There was no measurable change in that spring in 22 years and 62,000 miles.

That spring has almost no motion in doing it's function, the lash in the gear that it takes up is only a couple thousandths of an inch. It's a low stress application, so I would expect that spring to easily last the life of the bike.

A lot of worn parts could result in a knock, bearings, springs, shims, all sorts of things. Replacing one at a time could reveal the culprit, or do it once and do it all is another option. 

OK, I can think of some reasons this theory might not be sound, but here she goes. 

I've looked at the drawing below of the output shaft with the question in mind "Why would you have a knock under light load or going up a moderate hill?" as the original poster has stated. 

Let's say the circlip has reduced tension, or opened a little. Under light acceleration, pressure from the already weakened circlip is reduced further still, with the result of endfloat and  therefore a noise. While coasting, the spring load from the circlip might be sufficient. 

Here's how the parts interact, although this is a  K100, the gears to circlip arrangement is similar.




They're just some thoughts. Could it be?

Even better, here's a K75 drawing. Not exact or to scale, just to help picture how it works.