drive-side bearings

timetraveller

Well Known and Active Forum User
VOC Member
Recent communications from Robert Watson on this Forum and from John Renwick in the May 2010 MPH, page 15, concern fitting two back to back taper roller bearings to the drive side main shaft. Robert was good enough to send me a copy of his hand drawn original note from which Dan Smith did the modifications and fitting. As this is likely to be of interest to others, as well as ‘Roy the Mechanic’, and with Robert’s permission I have made a drawing which will explain to those who have not seen Robert’s sketch just what is involved. Note that the Robert Watson/Dan Smith modification has the flange with the countersunk holes on the outside of the bearing housing, that is inside the primary drive chamber, while John Renwick has chosen to put the flange inside the crank case near to the flywheel. Both have fitted a housing to take an oil seal on the outer end of the main shaft and in Robert’s case this is housed in a screw in oil seal holder with a left hand thread. The outer spacer, which sits between the outer parts of the two taper roller bearings, is an integral part of the bearing housing and therefore the position of this is dictated by what bearings are used and their dimensions. The inner spacer has to be accurately machine/ground to give a four thou interference fit in Robert’s case or a two thou interference fit in the case of ‘passenger0-0’. As both have been successful modifications it seems that the exact figure is not critical.

Please note that I have deliberately not put dimensions on the sketch as these will depend upon what size main shaft you are using and which bearings. There is quite a choice when one looks through the bearing catalogues. For either beefing up an engine or for recovering a worn set of castings this seems to be a useful piece of information which might help any of us in the future.
VinTaperRollers.jpg
 

roy the mechanic

Well Known and Active Forum User
VOC Member
Nice one Timetraveller, this is about as envisioned, except for the screwed sleeve. As i have new cnc machined cases i reckon to try without, bearings available to suit standard od+id. Regards Roy.
 

timetraveller

Well Known and Active Forum User
VOC Member
Hi Roy. Do you intend to rely on the bearing being a tight fit in the new crank cases or do have some scheme to fit the outer spacer firmly to the outer wall of the bearing housing? The reason I ask is that if I have understood correctly what is going on here then there has to be some sort of really positive location (circlips in the original design) to stop the bearing assembly creeping inwards or outwards.
 

roy the mechanic

Well Known and Active Forum User
VOC Member
Hi, Timetraveller, circlip grooves are in there as per standard, reckon to use them. The most important factor must be that the rods sit central in the bores. Roy.
 

timetraveller

Well Known and Active Forum User
VOC Member
I agree totally with that. The circlips and the ball race are the reference point with the original system and then shims used to get the centrality correct. It will be interesting to see how you get on with the taper rollers. I imagine that you are going to have to do some careful measuring of the overall thickness of the taper roller bearings and then fiddle the outer spacer thickness to get the inner part of the inner taper roller very nearly right up against the flywheel boss. Centrality is not going to be critical to within a thou but the use of a shim to get the final position correct seems to be likely to me. Good luck, sounds like an interesting project.
 

Howard

Well Known and Active Forum User
VOC Member
Sorry to butt in to this discussion. I can see what you're doing and how you're doing it, but my question is why?

It obviously works, but what shortcoming are you you designing out?

H
 

Robert Watson

Well Known and Active Forum User
VOC Member
Two things.

The outer spacer is integral in the housing, so the cups are not going to creep. No circlips either

The use of a steel sleeve reclaims a set of otherwise trashed cases. The Timkens are also very much less expensive.
 
Last edited:

roy the mechanic

Well Known and Active Forum User
VOC Member
Hi, Timetraveller, thanks for the positive waves. As I'm building the bike first, it will be a while before I can pull the motor to build it! Will report according. Howard, I have heard horror stories where standard main bearings go "walk-about" the big-end nut hits the bearing with undesirable results! Roy.
 

Howard

Well Known and Active Forum User
VOC Member
OK - I had it explained to me at the section meet last night. It's not a problem I've had, but I understand why there's concern.

What I'd got mixed in my mind, because I haven't had a bottom end apart this century, I thought the timing side was located axially, and fixing the drive side with back to back tapers sounded wrong in lots of ways.

H
 

timetraveller

Well Known and Active Forum User
VOC Member
I started riding Vincent twins in either 1957 or 8 so at that time the bikes were only a few years old and yet even at that stage it was known that the drive side main bearing could walk its way inwards until it fouled the crank pin nut. Once I started racing and sprinting a Vin, about 1960, I always tried to prevent this by grinding two small hollows on each side of the outer race of the drive side inner roller bearing and then fitting two large headed screws each side which fitted into holes sunk into the crank case. All this was done by hand as in those days I had no access (or the skills needed) to machine tools. The screws were punch locked to stop them rotating and the screw heads stopped the outer race rotating and working its way inwards. I never had a bearing come loose after doing this and all my engines both for others to race and for road use have had this mod done if the crank case was split. More recently my advice to those with access to the required machinery and skills is to grind two small flats into the edge of the outer race of the drive side inner main bearing and then place two flat pieces of steel into milled shallow slots which take the small metal plates and are held in place by punch locked countersunk screws. Even more recently it has become the fashion to use a full width outer race for the inner roller bearing, rather than the narrow special item originally used, and to thin down the outer edge of the spacer between the drive side bearings to allow this outer race to fit further in to the bearing housing. This gives a larger area to create the required friction between the outer race and the crank case casting.

At about the same time, late fifties and early sixties, it became clear that the drive side main shaft was likely to come loose in the flywheel with high rev use and sheer if it was used in sidecar racing.. Stopping the rotation of the main shaft in the flywheel was undertaken by drilling a hole, half in the main shaft and half in the flywheel, and forcing in a quarter inch diameter roller bearing and then peening over the surface to keep the roller in place. Several of the sidecar racers went for oversized main shafts following Picador engine development. I believe that the Picador main shaft was 1.125” diameter, as opposed to the one inch standard and more recently several of the racers have even taken this up to 30 mm diameter and I think that Patrick Godet might use that size as standard.

It seems to me that the Robert Watson/Dan Smith modification is a particularly elegant way of reducing potential problems in this area. Given the fact that the large flange is on the outside of the bearing housing and that there is an outer locator machined as an integral part of the new flanges tube there can be no way that the bearings or their housing can work their way inwards towards the flywheels. Additionally the sandwiching of the inner taper roller bearing between the flywheel boss and the outer spacer means that the outer race of the inner bearing also cannot move inwards. Given the fact that this mod can either be used in new cases, as Roy the Mechanic is going to do, or to rescue already damaged cases means that it might at some stage be of use to any of us which is why I am happy to now see it in the public domain.
 
Top