[iansoady]

OK, made the collar as described. I made it 0.200" thick as that seemed the best compromise between allowing the kickstart to function and not letting the clutch operating pin bottom out in its slot.

I'm afraid my material selection was even less considered than Tony's and I just used a chunk of stainless I had under the lathe. I'm sure it'll be all right.

I still need to do some final dressing to clean up marks left by parting off.

Initial trial fitting shows it seems to do the job OK.

Pictures:

[meddlesomebilly]

Hi, I've had a similar problem on my 1929 M90.
I made a split collar as you suggest to maintain the primary chain alignment.. In the absence of anything harder being on hand I used EN16 steel.
The bike has done around 800 miles since without problem and has put off the moment when I have to seriously think about a new mainshaft.
Tony.

[iansoady]

I'm slowly moving forward with investigating this and am now getting somewhere.

With very many thanks to Vic Youell who has provided me with a backplate and splined ratchet assembly I can see it coming together. The only problem I have to solve is that the splines on the ratchet are relieved for about 1/2" on the gearbox side (bottom right in the lower photograph). This allows the ratchet and hence backplate to travel too far towards the gearbox. The old, multiply bodged version, had full length splines and rested against the larger diameter plain portion nearest the gearbox (as evidenced by the witness marks).

The only solution I can see, as the splines are interrupted before the plain portion where the kickstart pinion / ratchet goes and the shaft has a smaller diameter outboard of this (see third picture), is to turn a spacer of the appropriate length with inner diameter that of the reduced portion and outer that of the relieved portion of the splined drive, then cut it into two lengthways to fit it. There is no relative movement between the drive and the shaft (other than slightly in a longitudinal direction when the clutch is lifted), and the spacer will be entirely within the drive section so it can't go anywhere and will only have longitudinal forces applied.

Anybody see a problem with this approach?


Front view of clutch components:



Rear view (showing relieved splines):



Shaft showing reduced portion: