- Joined
- Nov 20, 2004
- Messages
- 20,686
madass140 said:I was informed that the rollers were barrel shaped and that was the idea behind the Superblend. that with crankshaft flex the ends of the rollers would not dig in, my source whoever it was years ago may very well be wrong, I meaured the rollers once and found them to be parallel, Nowhere have I read anything official from FAG that the rollers are other than parallel, ie: normal rollers.
http://nortoncommandoclutch.zxq.net/index_78.htm
34. SO CALLED ‘SUPERBLEND’ MAIN BEARINGS. (All measurements were taken with a micrometer / my even then olde pair of eyes and are approximate). When the Commando main bearing problem started to occur the D.S roller bearing was a Ransome and Marles MRJA30 bearing but with a brass cage and a designation 8MRJA30.The roller elements were the std. shape for that time having 11 rollers 11.12mm in diameter, 11.12mm in length with a parallel working surface of 10mm with a very small std. radius to the corners at each end. It had load values of Static 36,200 Newtons and Dynamic 41,900 Newtons. (The T.S. bearing was a brass-caged 6306 ball race bearing.) Personally I used to use the higher load capacity M306 ball race bearing but we did have a bearing factory in town in those days. In use on the more highly tuned Commando, due to crank flex the tiny area at the edges of the 10mm rollers’ parallel portion would be the only points of contact with the inner and outer raceways which damaged the hard surfaces of the raceways and rollers leading to quick failure.(4000 miles?). ONE of the MANY solutions required to solve the Commando main bearing failure problem was a change in the bearing rolling element design to what became known to Norton owners as ‘super blend’ where the rollers had a narrower central parallel portion with each end slightly radiused for a mm. plus of length which allowed the crank to flex more within the raceways without the edges digging in and increased the contact area within the raceways as the crank flexed, compared to the older bearing that is. I believe there are a couple of technical terms for the shape one of which is logarithmic profiling. The number of rollers increased to 13, the width and diameter of the rollers reduced to 9.53mm with the parallel part of the roller reduced to 6mm with the special slight radius at each end reducing to 9.46mm. This brass caged bearing manufactured by what had by then become RHP (Ransome Hoffman Pollard) and produced at the old Ransome and Marles factory at Newark had an old R&M designation of 6MRJA30 with the ‘6’engraved into the side of the outer raceway and the rest of the designation stamped on. The inner raceway being stamped MRJA30.
The load values for the new ‘super blend’ bearing were lower than that of the original 8MRJA30 bearing it replaced and were Static 31,000N and Dynamic 35,800N. The Nicholson ‘bible’ Modern Motor Cycle Mechanics gives 6MRJA30 as the bearing to be used and tells owners NOT to use the one marked 8MRJA30. I find it interesting that the new bearing which partly helped cure the main bearing failure problem did not require similar or higher load values than the bearing it replaced. A gentleman at the Newark factory kindly calculated the load values for the two bearings for me after digging out the drawings from the cellars. I suspect the drawings along with those for an engine bearing used on a Bristol Bulldog fighter or something similar which the gent had just dug out for another query when I phoned have long since gone. In that Motor Cycle Sport article mentioned earlier (Jan 77 pages 27/28) the writer states that another manufacturers bearings had been tried but that yet another manufacturers bearings ended up being used as they proved better and at one time in a Birmingham warehouse full to the brim with cheap new and part finished BSA Triumph and Norton parts were a small pile of SKF bearings we were told had come from Norton and I SUSPECT these were the original ones tried. There were two types and we bought a few of each. These ‘barrel’ shaped roller bearings were made by SKF in two forms, a std. bearing and a high load capacity version both having a steel cage. One came in boxes marked SKF. NJ306. 1972, which had 11 rollers of 10mm diameter with a 7mm wide parallel section with a radius at each end tapering to 9.97mm. Load values were static 20,000N and Dynamic 36,900N. The other bearing came in boxes marked SKF. NJ306E.1972. (the ‘E’ designates extra load capacity) and employed 12 rollers of 12mm width and 11 mm diameter with a 9mm wide parallel portion reducing to 10.97mm at the end of the slight radius. Load values were Static 53,000N and Dynamic 57,000N. All load values being calculated for me by a Gentleman at SKF and I suspect the NJ306 static one is incorrect but I am not going to try to have SKF check it for me. After Norton etc. had ‘gone’ in one Birmingham emporium were probably several hundred RHP boxes each containing the old Atlas ball or roller mains, 8 bearings to a box and if memory is correct £8 for a box of ball bearings and £12 for a box of roller bearings. One dealer I believe then advertised the roller bearings for sale at £12 each referring to the brass cage as bronze. Not a bad profit margin. Out of curiosity I spent a couple of hours moving boxes and found two marked 6MRJA30 which, after checking the contents were the 6MRJA30 version I bought for £12 per box but have given most away to friends over the years thus ensuring another loss making exercise! At around this time most bearing manufacturers were changing or had changed the shape of their roller bearing rolling elements to include some slight ‘barrel shaping’ to the ends of the rollers, whatever the technical term for it is and this included FAG, RHP Steyr etc… The FAG ‘Super blend’ NJ306E bearing is just a brass caged high load capacity version of their std. NJ306 bearing but the rollers have slightly less radius than the special RHP 6MRJA30 bearings and I found it interesting to note that every old 6MRJA30 bearing I have seen that have been taken from old motors has wear marking on the rollers but not quite to the very edge whereas all the FAG ones (mainly fitted to 750 motors) show wear to the very edge of the rollers, probably because they don’t have as much reduction in diameter toward the outer edges but I suspect that the wear markings will have a bit to do with how the motor was used during its (short?) life. So why did Norton change to using FAG bearings? When I asked a couple of NVT gentlemen the answer given was that they were cheaper and FAG were just down the road.
The FAG NJ306E bearing I looked at had 12 rollers of 12mm width and 11mm diameter with a 8mm parallel section reducing to 10.97mm at the end of the radius. Load values being given in the book as Static 48,000N and Dynamic 51,000N. For a RHP NUP306ETN bearing as used in later Triumph twins the one I looked at had 12 rollers of 12mm width and 11mm diameter with an approximately 9mm parallel section reducing to 10.96mm at the edges. I lost interest in measuring other manufacturers bearing products just to satisfy my curiosity. So really the only special barrel shaped roller ‘Super blend’ main bearings made for Norton motor cycles were the original R&M 6MRJA30 which have not been made or been available for decades. WARNING. Many years ago one dealer bought a pile of very cheap FAG NJ306E brass caged bearings thinking they were SUPERBLEND… I opened one up and could visually see NO sign that the rollers were radiused as one can if you catch the light correctly. ‘Popping’ a roller out, the micrometer showed it not to be ‘super blended’. Phoning FAG they suggested they must have been old stock held by a stockist from the days before the roller shape changed. The dealer still sold them for Commando use and I wonder how long they lasted in service. Bet no one ever just ran a light over the rollers to check they were the correct bearings but why would they have done so? There could still be a few appearing at auto jumbles so BEWARE - but that applies for most things produced for our old Brit Bikes these days and I won’t mention the made a few years ago now………….
