Superblend bearing profile
- Thread starter comnoz
- Start date
- Status
- Joined
- Oct 28, 2014
- Messages
- 1,996
A very convincing demonstration. So...what ,then, is so special about superblends?
- Joined
- Dec 10, 2008
- Messages
- 7,253
seattle##gs said:
The "Superblend" as it was unofficially named was the first of a new generation of bearings with an improved interface between the parallel part and the start of the radius. The flat part and the radius were blended to avoid an edge that could cause damage when the alignment was not perfect. [hence the name "superblend"]
Getting a perfect blending of the flat and radius required new and improved bearing grinders. The blend was thereafter adopted by most [if not all] manufacturers. Jim
Thanks Jim, if only this had been done 40+ years ago.
Would have killed all that "barrel-shaped" BS.
BTW, FAG have no knowledge of that 'superblend' BS either, it seems.
Some of us suspect that Nortons at some point figured out that Randsome and Marles bearings weren't all they were cracked up to be,
and going to something with better steel and 1 more roller was the simple answer (??).
To avoid embarrassment, they didn't come out and say so outright.
This might be reading between the lines a bit, but stranger things have happened at sea....
Would have killed all that "barrel-shaped" BS.
BTW, FAG have no knowledge of that 'superblend' BS either, it seems.
Some of us suspect that Nortons at some point figured out that Randsome and Marles bearings weren't all they were cracked up to be,
and going to something with better steel and 1 more roller was the simple answer (??).
To avoid embarrassment, they didn't come out and say so outright.
This might be reading between the lines a bit, but stranger things have happened at sea....
- Joined
- May 7, 2005
- Messages
- 7,163
Superblend was always a marketing term and not an engineering term, the roller is not barrel shaped and never has been, what changed was the crowning applied to the roller.
For more info read this, info on crowning is on page 12, and all 'plain' rollers has some degree of crowning.
http://ntrs.nasa.gov/archive/nasa/casi. ... 018943.pdf
For more info read this, info on crowning is on page 12, and all 'plain' rollers has some degree of crowning.
http://ntrs.nasa.gov/archive/nasa/casi. ... 018943.pdf
- Joined
- Dec 10, 2008
- Messages
- 7,253
Rohan said:
I was told that "superblend" was an "in house" term that was used to describe the new manufacturing process which "blended" the forged radius into the load bearing area of the roller.
The process was applied to all straight, tapered and barrel rollers.
A different forging slug was used, requiring a different grinding process.
All roller bearings manufactured by them at that time used the improved manufacturing technology.
He also sent an engineering drawing of the bearing with all the dimensions blacked out.
All this is from memory as the correspondence was lost in my shop fire 10 years ago. Jim
- Joined
- Nov 10, 2006
- Messages
- 1,530
Jim,
I didn't think the crown would have had a "big" change in dia though I would expect it to be measurable, and looking at your photo the roller has what appears to be polished unused ends.
Are you using a ball point stylus on your indicator? You haven't got any new rollers knocking about to try ? :wink:
Nice work,
Thanks
Dave
I didn't think the crown would have had a "big" change in dia though I would expect it to be measurable, and looking at your photo the roller has what appears to be polished unused ends.
Are you using a ball point stylus on your indicator? You haven't got any new rollers knocking about to try ? :wink:
Nice work,
Thanks
Dave
- Joined
- Dec 10, 2008
- Messages
- 7,253
cash said:
The roller I picked was the lowest mileage OEM roller I could find. I had written 7000 miles on it.
It looked pretty cherry but the inner race had not been kept with the outer or I would have re-used it.
The stylus was a carbide point with a .005 radius.
The only new rollers I have are in new bearings. [but if you want to donate one I will measure it...] Jim
- Joined
- Apr 21, 2015
- Messages
- 322
I would suggest that the term 'superblend was used initially by Norton ownerswho thought they were the cure for the main bearing failure problem when as anyone capable of reading Mortor Cycle Sport January 1977 pages 27 and 28 knows they were only PART of the cure.
The terms the Chief Bearing Designer used for it in a ltele[hone conversation and letter to me on the subject a great many years ago was LOGARITHMIC PROFILING and CROWNING.
Originally the rollers had parralel working surfaces with a tiny radius at each end. Perfectly satisfactory for the great many roller beariongs I installed in fairly big AC and DC motors and generators in my more youthful days. I bet the majority are still working perfectly 40 odd years later. However by the time of the original Atlas Mk3 (or Commandos as the 2 year stop gap were later named) with the redicu;lous increases in compression ratios and the use of higher lift cams plus an advance retard unit that fitted tro the end of the cam which VERY quickly fell apart staying fully advanced all the time the poor olde 3 piece crank was flexing in all directions especially with younger death defying riders who tended to rev the nuts off of the poor old motors.....the result was that at tuimes the edges of the rollers were the only bit of the rollers making contact with the raceways causing the surfaces to fail and hey presto at around 4,000 miles and even earlier i beliweve in Germany on one road test........a rumble occured......my money is on the average ownerthinking 'Oh thats an interesting rumble' and thought no more about it.....
The original Atlas bearings, a balll race on the timing side and straight parallel rolling elements roller bearing on the drive side as far as I am aware gave little problem on STANDARD motors with their STANDARD cam and STANDARD low compression ratio but on the stop gap Commqando ........(The low compression ratio being employed to reduce the white finger and turning ones nuts to dust vibes....)
The bearing manufacturer was Ransom and Marles. (All measurements given are APPROXIMATE).
The roller bearing had the designation 8MRJA30 It had 11 rollers... 11.12mm OD, 11.12mm Wide and the parallel working portion of the rollers was 10mm. Load capacity was Static 35,200 Newtons - Dynamic 41,900 Newtons. It had a brass cage and cost £1.50 each in boxes of 8 per box shortly after NVT went arse over tit. The ball race 6206 cost £1.00 each....for cash of course..... One UK dealer in the North West sold the roller bearings to customers for £12.00 each claiming the cages were bronze.... as I remember it....
Originally Norton tried another manufacturers bearings and I bought a few ex Norton stock SKF bearings at one time that were reputed to of been tried. There were two types both with steel cages seperating the rollers. One was a SKF NJ306 with the box marked NJ306(1972). It had 11 rollers of 10mm O.D. and width BUt the parrallel portion of the rollers was 7mm with 'crowning' at each end down to 9.97mm at the edges. I was given load values of Static 20,000 Newtons and Dynamic 36,900 Newtons. The other was a higher load capacity version NJ306E with the box marked NJ306E(1972). It had 12 rollers of 11mm OD and 12mm width but with a central parallel length of 8mm with 'crowning' at each end taking it down to 10.97mm O.D. I was given load values of Static 30,000 Newtons and Dynamic 57,200 Newtons. I suspect(I repeat suspect) these were the bearings that were deemed unsatisfactory mentioned in the Motor Cycle Sport article.
The bearing used eventually called, by many, 'superblend' and which formed part of the cure for the main bearing failure problem were once again manufactured by Ransom and Marles but were marked 6MRJA30. Note 6MRJA30 and NOT 8MRJA30. It had a brass cage and 13 rollers with an O.D. of 9.53mm and 9.53mm width with a 6mm wide cental parallel portion and crowning at each end taking it down to 9.46mm.
The FAG bearing is simply a std high load capacity 306 roller bearing marked NJ306E. It employs 12 rollers of 12mm width and O.D. of 11mm . The central parallel portion is 8mm with crowning taking it down to 10.97mm at each end. Load values are given as Static 48,000Newtons abd Dynamic 51,000 Newtons.
At one time Triumph employed the RHP bearing NUP306ETN It employed 13 rollers of 12 mm width and O.D. of 11mm. The central parallel portion was 9mm with crowning taking it down to 10.96mm at each end. Load values being Static 53,000 newtons and Dynamic 57,000 Newtons.
The only reason I was ever told for going to FAG bearings were that they were cheaper and their office was just down the road......Sounds to me like accountants ran NVT....
As far as I am am aware all major bearing manufacturers roller bearings employ 'superblend' / crowned rollers BUT I found it of interest when examining various used Commando main bearings to note that with the original R & M 6MRJA30 rollers the wear markings did NOT show wear occuring to the very edge of the crowning wheras with every FAG bearing it did........ due to the greater crowning of the R & M rollers?? So can the edges of FAG bearings dig into the raceways?? Or were the FAG bearings out of seriously abused ex Kent police bikes ......
At one time a certain london dealer bought a rather klarge pile of FAG NJ306E bearings. I opened one up , popped out a roller and applied eye and micrometer to it. It did NOT have any crowning. I phoned FAG who suggested they were very old stock early versions manufactured before the change over to the later roller design. The dealer flogged them all to Commando owners and to my knowledge non even noticed........Wonder how many failures resulted??
As so few owners ri
The terms the Chief Bearing Designer used for it in a ltele[hone conversation and letter to me on the subject a great many years ago was LOGARITHMIC PROFILING and CROWNING.
Originally the rollers had parralel working surfaces with a tiny radius at each end. Perfectly satisfactory for the great many roller beariongs I installed in fairly big AC and DC motors and generators in my more youthful days. I bet the majority are still working perfectly 40 odd years later. However by the time of the original Atlas Mk3 (or Commandos as the 2 year stop gap were later named) with the redicu;lous increases in compression ratios and the use of higher lift cams plus an advance retard unit that fitted tro the end of the cam which VERY quickly fell apart staying fully advanced all the time the poor olde 3 piece crank was flexing in all directions especially with younger death defying riders who tended to rev the nuts off of the poor old motors.....the result was that at tuimes the edges of the rollers were the only bit of the rollers making contact with the raceways causing the surfaces to fail and hey presto at around 4,000 miles and even earlier i beliweve in Germany on one road test........a rumble occured......my money is on the average ownerthinking 'Oh thats an interesting rumble' and thought no more about it.....
The original Atlas bearings, a balll race on the timing side and straight parallel rolling elements roller bearing on the drive side as far as I am aware gave little problem on STANDARD motors with their STANDARD cam and STANDARD low compression ratio but on the stop gap Commqando ........(The low compression ratio being employed to reduce the white finger and turning ones nuts to dust vibes....)
The bearing manufacturer was Ransom and Marles. (All measurements given are APPROXIMATE).
The roller bearing had the designation 8MRJA30 It had 11 rollers... 11.12mm OD, 11.12mm Wide and the parallel working portion of the rollers was 10mm. Load capacity was Static 35,200 Newtons - Dynamic 41,900 Newtons. It had a brass cage and cost £1.50 each in boxes of 8 per box shortly after NVT went arse over tit. The ball race 6206 cost £1.00 each....for cash of course..... One UK dealer in the North West sold the roller bearings to customers for £12.00 each claiming the cages were bronze.... as I remember it....
Originally Norton tried another manufacturers bearings and I bought a few ex Norton stock SKF bearings at one time that were reputed to of been tried. There were two types both with steel cages seperating the rollers. One was a SKF NJ306 with the box marked NJ306(1972). It had 11 rollers of 10mm O.D. and width BUt the parrallel portion of the rollers was 7mm with 'crowning' at each end down to 9.97mm at the edges. I was given load values of Static 20,000 Newtons and Dynamic 36,900 Newtons. The other was a higher load capacity version NJ306E with the box marked NJ306E(1972). It had 12 rollers of 11mm OD and 12mm width but with a central parallel length of 8mm with 'crowning' at each end taking it down to 10.97mm O.D. I was given load values of Static 30,000 Newtons and Dynamic 57,200 Newtons. I suspect(I repeat suspect) these were the bearings that were deemed unsatisfactory mentioned in the Motor Cycle Sport article.
The bearing used eventually called, by many, 'superblend' and which formed part of the cure for the main bearing failure problem were once again manufactured by Ransom and Marles but were marked 6MRJA30. Note 6MRJA30 and NOT 8MRJA30. It had a brass cage and 13 rollers with an O.D. of 9.53mm and 9.53mm width with a 6mm wide cental parallel portion and crowning at each end taking it down to 9.46mm.
The FAG bearing is simply a std high load capacity 306 roller bearing marked NJ306E. It employs 12 rollers of 12mm width and O.D. of 11mm . The central parallel portion is 8mm with crowning taking it down to 10.97mm at each end. Load values are given as Static 48,000Newtons abd Dynamic 51,000 Newtons.
At one time Triumph employed the RHP bearing NUP306ETN It employed 13 rollers of 12 mm width and O.D. of 11mm. The central parallel portion was 9mm with crowning taking it down to 10.96mm at each end. Load values being Static 53,000 newtons and Dynamic 57,000 Newtons.
The only reason I was ever told for going to FAG bearings were that they were cheaper and their office was just down the road......Sounds to me like accountants ran NVT....
As far as I am am aware all major bearing manufacturers roller bearings employ 'superblend' / crowned rollers BUT I found it of interest when examining various used Commando main bearings to note that with the original R & M 6MRJA30 rollers the wear markings did NOT show wear occuring to the very edge of the crowning wheras with every FAG bearing it did........ due to the greater crowning of the R & M rollers?? So can the edges of FAG bearings dig into the raceways?? Or were the FAG bearings out of seriously abused ex Kent police bikes ......
At one time a certain london dealer bought a rather klarge pile of FAG NJ306E bearings. I opened one up , popped out a roller and applied eye and micrometer to it. It did NOT have any crowning. I phoned FAG who suggested they were very old stock early versions manufactured before the change over to the later roller design. The dealer flogged them all to Commando owners and to my knowledge non even noticed........Wonder how many failures resulted??
As so few owners ri
J. M. Leadbeater said:
JM, you seemed to have missed that that is how ALL of the FAG NJ306E bearings appear ? - and measure up as here with this one.
Did you watch the video ?
These are a PARALLEL roller bearing after all - if the rollers aren't parallel, how would they simply slide into the other track ?
And have enough track for the whole roller have something to roll on...
You also seem to have missed that it was the 1972 only COMBAT model Commandos that had main bearing dramas/failures,
the earlier Commando Mk 3 Atlas not so.
A thread on this very same subject on the NOC Forum a while back (year or 2 ) showed a Randsome and Marles bearing,
the stopgap fix for this bearing problem, with machining/grinding clearly visible on quite a lot of the roller (quite a few mm each end).
It was said to be have been quite expensive to do that for each roller too.
Sorry don't know the link, could probably track down the photo.
FAG has none of that visible, its just simply a better bearing, off the shelf, nothing special done to it.
C3 bearing clearance too, note.
BTW, another BIG part of the 'cure' of this was the extra rolling element in the bearing - i.e. an extra roller.
This is a stock drive side main bearing intended for an early Norton Dommie, circa 1950.
Note that it only has 11 rollers. As you note, the 306E has 12 rollers.
That extra roller (considerably) reduces the loading per roller for the 306E bearing...
comnoz said:
This was FAG saying it was THEIR inhouse term for it,
or Nortons inhouse term for it ?
Or Randsome and Marles inhouse term for it ?
As just noted just above, a pic of a Randsome and Marles 'superblend' bearing on the NOC Forum recently showed
quite a few mm each end of the roller with extra grinding or machining - clearly visible to the naked eye.
As LAB just showed, 'superblend' was used by Norton with R&M bearings, before FAG were involved.
I must dig out a quite early dommie engine I have that I know hasn't been opened, and have a close look at the bearings in it.
The drive side roller being the one of interest.
I don't previously recall any sharp edges on the rollers, but I wasn't taking special notice.
P.S. For many years, british bikes used "Hoffman' bearings, what happened to them ?
The drive side roller being the one of interest.
I don't previously recall any sharp edges on the rollers, but I wasn't taking special notice.
P.S. For many years, british bikes used "Hoffman' bearings, what happened to them ?
- Joined
- Nov 20, 2004
- Messages
- 20,423
Thanks LAB.
It doesn't actually say what happened to them though. (apart from being bombed and rocketed !).
Once-upon-a-time, nearly every british bike had Hoffman bearings in them when you opened them up.
And car too, I gather...
The bearings are often still in good condition too.
But we diverge...
It doesn't actually say what happened to them though. (apart from being bombed and rocketed !).
Once-upon-a-time, nearly every british bike had Hoffman bearings in them when you opened them up.
And car too, I gather...
The bearings are often still in good condition too.
But we diverge...
- Status
Similar threads
