As I mentioned previously, apart from the clutches that had come with some of the bikes, I had acquired a collection of clutch parts some years previously.
As I had 6 gearboxes, I aimed to get 6 working clutches out of these bits, although I only need 5 for my own projects.
Apart from a couple of 3-plate clutches, these were all 7-plate clutches. The construction of these is:
Backplate – Friction Plate – Plain Plate – Chainwheel – Plain Plate – Friction Plate – Front Plate.
There has probably been as much written about the idiosyncrasies of Velocette clutches as any other aspect of these bikes – and I'm not going to add to that volume of literature. I can understand why that is – clutch operation on other bikes of the era (and subsequently) were generally more intuitive – there was a pushrod passing through the gearbox mainshaft that pushed against the front plate holding the clutch springs and released the pressure on the plates when the clutch was operated. Simple. Velo clutches do not work like that ….but please read about them somewhere else (but not the clutch adjustment procedure in the Haynes manual …it’s incorrect, as I (and many others) found out the hard way many years ago when I first owned a Velo).
However, there is one observation that I will make about Velo clutches: they must have been quite expensive to manufacture. For example, the clutch back plate (KC-1/25), shown below,
really is a work of art but would have required a lot of separate machining operations compared to a more conventional pushrod-operated clutch (see, for example, here and here which describe the clutch that I made for the AJS V-twin).
Among my collection of
bits, I found 2 slightly different chainwheels whereas there is only one (part
number KC-3/25) listed in both of my Spares Parts Lists which cover the years
1925 – 1931 and 1932 – 1934. There are 2 differences: the first is the width of
the slotted rim at the outer edge in which one side is very slightly wider (on
the right side of the sprocket shown below) than the other;
and there is a corresponding change to the bearing housing to move the bearing slightly more inboard. The 2 pictures below show both sides of the bearing housing in one of these chainwheels.
It is a small change but it has the effect of moving the location of the bearing on the backplate to be positioned more on the continuous section rather than the section with the 6 protruding tangs (that enter the front plate), ie a better design. The 2 changes go hand-in-hand to ensure the same position of the sprocket when the clutch is assembled so that chain alignment is unchanged. I can only assume that Velocette made this change without changing the part number ….unless someone knows differently? The other, presumably earlier, chainwheel is symmetrical.
The first step was to make an audit to order bits and pieces from both Grove Classics and Velo Spares Ltd. (the Velo Owners Club spares scheme). Various gearbox bits were ordered at the same time.
Before these arrived, I pressed out the clutch bearings from all the chainwheels so that I could clean them (and the filthy chainwheels) properly to assess what was useable.
Care must be taken to both extract and replace these bearings to avoid damage and, to this end, I machined a piece of round bar, shown below, to ensure that the pressing load is only applied to the bearing outer.
The smallest diameter is a loose fit on the inside of the bearing – just to locate it on-centre, whilst the intermediate diameter is a few thou less than the OD of the bearing to apply the pressing load. The shoulder is also relieved by around 0.005” so that load is not applied to the bearing inner.
And with the housing in the chainwheel supported by a thick steel tube
the bearings can be pushed out easily and without damage to either the chainwheel or the bearing.
New bearings are available from GroveClassics but it is worth preserving serviceable ones if possible.
All of the serviceable parts were degreased with cellulose thinners and then cleaned up in the tumbler overnight (except the bearings) before the new bits and pieces turned up.
There were new inserts (C-25) for chainwheels (22 in each chainwheel – 3 sets were required)
3 sets of new plates (4 for each clutch)
plus thrust bearings, oil-retaining shims, springs, thrust pins etc.
….it’s a bit like Christmas morning when you’re a child.
After a lot of cleaning, measuring, refitting good bearings and inserting new “corks” (they are not really corks but a modern replacement friction material) into the chainwheels I now have 6 good serviceable clutches.
There are still a couple of bits that I need to order but, in the meantime, they will have to occupy my milling machine table as all my other workspace is taken up with disassembled gearboxes.
And these are the bits left over:
I will keep the backplates, chainwheels and front plates for a period of time in case anyone wants one but the rest will go in the scrap bin.
"Wide" assymetric chain-wheels can be 9-plate (heavy wheight RS), i have seen a slim assymetric 7-plate wheel that i preliminarily identified from the parts context as RS-MAC, somebodys old modification is also possible..
ReplyDeleteFor the wider audience: the clutch parts mess is well documented by Dai Gibbison (v-tec.velocette.info)
Sten vW (velonut Sweden)
Hi Sten,
ReplyDeleteThanks for your input. I have 2 of the asymmetric chainwheels which also have the correspondingly offset bearing housing. They are both the same so I don't think these have been modified by someone in the past. If so, it's a hell of a coincidence.
These are not chainwheels for 9- plate clutches - I have those on my Venom and now-sold Thruxton.There is no way that 2 more plates could be squeezed into these.
Maybe they come from a MAC, as you say. I'll research it a bit further when I have time.
Richard