961 gearbox

When you get bearing out, take it to a good bearing supplier, given the number he can replace exact size with a heavy duty bearing, usually having a extra ball or two
 
I had heard some time ago, and recently again that the bearings are available as gene says. I don't know whether it's a worn shaft or bearings or both on mine. A fellow member of this parish and all round 961 nut has offered me his old assembly to try to put together a working unit.
 
Well folks, here's a conundrum. Early bike, no evidence that it's been apart before. I wanted to put my mind at rest, so stripped the primary to inspect and look what I found....

961 gearbox


961 gearbox


961 gearbox


961 gearbox


961 gearbox



Yep, the "rivets" are already welded, plus the shock springs have top hats in them.

I found both of the 6205 bearings loose in the housing and there are signs on the inner that the track has been spinning in the basket. That bearing is a bit grumbly but not too bad. They'll both be replaced. There is a large circlip between the bearing to locate the clutch basket and a spacer shim which is supposed to stop the bearing centres pinching up when tightened. That spacer was thicker that the circlip so I've skimmed a bit off it to match, so the outers should be held securely in place and the bearings can control the end float.

Ring gear nice and tight, no signs of any movement, but I will pin it anyway.

A question - In the manual it says to use Threebond 1327 on the shaft nuts etc., but I cannot find it anywhere (even on the Threebond catalogue) does anybody know what it been superceded with? Or would Loctite 243 do the job as well?
 
Well folks, here's a conundrum. Early bike, no evidence that it's been apart before. I wanted to put my mind at rest, so stripped the primary to inspect and look what I found....

View attachment 82055

View attachment 82056

View attachment 82057

View attachment 82058

View attachment 82059


Yep, the "rivets" are already welded, plus the shock springs have top hats in them.

I found both of the 6205 bearings loose in the housing and there are signs on the inner that the track has been spinning in the basket. That bearing is a bit grumbly but not too bad. They'll both be replaced. There is a large circlip between the bearing to locate the clutch basket and a spacer shim which is supposed to stop the bearing centres pinching up when tightened. That spacer was thicker that the circlip so I've skimmed a bit off it to match, so the outers should be held securely in place and the bearings can control the end float.

Ring gear nice and tight, no signs of any movement, but I will pin it anyway.

A question - In the manual it says to use Threebond 1327 on the shaft nuts etc., but I cannot find it anywhere (even on the Threebond catalogue) does anybody know what it been superceded with? Or would Loctite 243 do the job as well?
https://threebond-europe.com/products/tb1327/ we prefer 277
 
Thanks for the link & tip !

Well folks, here's a conundrum. Early bike, no evidence that it's been apart before. I wanted to put my mind at rest, so stripped the primary to inspect and look what I found....

View attachment 82055

View attachment 82056

View attachment 82057

View attachment 82058

View attachment 82059


Yep, the "rivets" are already welded, plus the shock springs have top hats in them.

I found both of the 6205 bearings loose in the housing and there are signs on the inner that the track has been spinning in the basket. That bearing is a bit grumbly but not too bad. They'll both be replaced. There is a large circlip between the bearing to locate the clutch basket and a spacer shim which is supposed to stop the bearing centres pinching up when tightened. That spacer was thicker that the circlip so I've skimmed a bit off it to match, so the outers should be held securely in place and the bearings can control the end float.

Ring gear nice and tight, no signs of any movement, but I will pin it anyway.

A question - In the manual it says to use Threebond 1327 on the shaft nuts etc., but I cannot find it anywhere (even on the Threebond catalogue) does anybody know what it been superceded with? Or would Loctite 243 do the job as well?
The spacer is supposed to be thicker than the circlip. This is so that the clamping pressure of the shaft nut is applied to the inner race of the bearing , not the outer !! Being thinner will now put severe load into the spheres of the bearing and it will fail.
Also, the fact that the rivets are welded suggests that there is nothing clamping the inner parts of the assembly together, as the function of the 'peened' rivet is to swell the shank and grip the inner components.
 
Hi NikolI, if the spacer and the circlip are the same thickness, why would it put any load on the the bearings? surely the only load on them is axial?

I understand What you’re saying about the clamping force of the rivets, but I cannot feel any play. What would you suggest I do?
 
Hi NikolI, if the spacer and the circlip are the same thickness, why would it put any load on the the bearings? surely the only load on them is axial?

I understand What you’re saying about the clamping force of the rivets, but I cannot feel any play. What would you suggest I do?
The 6205 bearings are a standard 'off the shelf' bearing. If I remember correctly they should be FAG6205. I don't recall them being C3 grade though. The C3 has a greater clearance between the sphere and the race groove, and would as a result lead to the clutch 'waggling' around its axis. Considering that the gear teeth are a very basic straight cut form, then this would generate considerable noise. Hence why nearly all road bikes use helical cut gears.

6205 are 'deep groove' ball races, and thier purpose is to control end float with the use of the circlip and mainshaft end nut.
Roller bearings would be an excellant bearing if the 'end float' could be controlled, which it currently can't.
This was looked at, at an early stage of design but not adopted due to other design constraints from the original 'Dreer' design.

As for the centre spacer, I think it should be 0.05/0.10mm thicker than upper tolerance limit of the circlip.
The nut pulls pressure through the shaft via the centre races of the 6205 bearings, thus controlling the float.
If the distance between the bearings is reduced, then this would squeeze the bearings together thus reducing the designed spherical clearance of the bearing races. Its the distance between the inner bearing races that is important.
 
Thanks for the feedback. I can confirm the bearing are indeed FAG 6205, and are C3 spec. Do you think they need to be? Like you say there was a bit of "waggle" and the inner race is a bit rough with evidence that it had been spinning in the basket, hence my thinking of reducing the spacer to make it more like one double row bearing, but if you think that's not a good idea I can easily knock up a thicker one.

I did ask earlier in the thread about using a roller, but obviously end float control is an issue. Do you think a roller on the inner and a ball on the outer would work ?
 
The extra clearance from two C3 bearings next to each other would be negligible and not increase clutch wobble much if at all, completely different for a single bearing where a CN or C2 would be needed. CN to C3 is a change of microns in internal clearance and at 15mm apart centre of ball path to ball path significantly reduced in external effect.
 
The extra clearance from two C3 bearings next to each other would be negligible and not increase clutch wobble much if at all, completely different for a single bearing where a CN or C2 would be needed. CN to C3 is a change of microns in internal clearance and at 15mm apart centre of ball path to ball path significantly reduced in external effect.
Yes it does. I've measured it
 
Thanks for the feedback. I can confirm the bearing are indeed FAG 6205, and are C3 spec. Do you think they need to be? Like you say there was a bit of "waggle" and the inner race is a bit rough with evidence that it had been spinning in the basket, hence my thinking of reducing the spacer to make it more like one double row bearing, but if you think that's not a good idea I can easily knock up a thicker one.

I did ask earlier in the thread about using a roller, but obviously end float control is an issue. Do you think a roller on the inner and a ball on the outer would work ?
A long time ago, one of the Norton engineers called 'Paul' actually looked at the roller option in a bid to try and help reduce the gear mesh noise. It was at this time when we compared the standard bearing to the C3, and measured the deflection. The values I can't remember as it was around 2009 I think. He discovered a roller bearing with a thrust face option, but he left the business before he had a chance to trial it. Some crank shafts use the ball and roller combined method , so It could be worth persuing, but is it really a problem that requires solving to that extent ?
There was a point in the early days were the fit of the 6205 was loose inside the clutch primary gear, and in some cases it just fell into place !
 
It's just the thrashing noise doesn't seem right, but as you say how far is it worth going.

Yep, both mine are loose enough to drop out, so will need some Loctite or similar on the replacements.

Are you still pursuing the idea of making a replacement clutch?
 
As popular as the 6205 is , there may be a option of oversize OD +.002. Little known but their out there
 
You could get the bearing copper plated on the OD to restore the inference fit, more reliable long term than loctite.
 
Back
Top