Skip to main content
English French German Italian Spanish

NJ/U306E ('superblend') profile

Forums

This should keep quiet those who still insist Commando's have barrelled roller 'superblends'.

We seem to have lost the cut and paste facility? or is it just my compooter?

Permalink

Just one word of caution.

The test roller bearing is clearly well used and worn, but even so, still displays a definite difference in surface finish (blending??) for a millimetre or so before the end radius, as is clearly evident on new bearings. Perhaps any minuscule dimensional difference has disappeared through wear??

I would be very interested in the same exercise carried out on a new unused roller.

Just to add to the conspiracy theorists the attached link from the end of the parallel Access Norton thread totally contradicts the video evidence.

Keep Smiling

Permalink

Previously Jim Brierley wrote:

Just to add to the conspiracy theorists the attached link from the end of the parallel Access Norton thread totally contradicts the video evidence.

Keep Smiling

I'd be keen to know which part of that paper you feel contradicts the video evidence? If it's Figure 12 (on page 12), then Fig. 12(b) - 'cylindrical roller, partially-crowned' shows exactly the situation presented by Jim Comstock's demonstration.

Permalink

I'd be keen to know which part of that paper you feel contradicts the video evidence? If it's Figure 12 (on page 12), then Fig. 12(b) - 'cylindrical roller, partially-crowned' shows exactly the situation presented by Jim Comstock's demonstration.

I think you are confusing the crown with the radius at the end of the roller

Permalink

Hello The original Made in England Ransom & Marles number -- 6MRJA-6 copied by FAG NJ-U306E Made In China and Marketed By FAG Germany , Is Not A Crown nor a Spherical roller Bearing , But Has Sharp Rounded ends toopreventthe roller digging in on this side track , And you guy's ever take and notice of what I have Put witch Is the Correct answers, Witch Engineering institute did you lot go too , yours Anna J

Permalink

Previously anna jeannette Dixon wrote:

Hello The original Made in England Ransom & Marles number -- 6MRJA-6 copied by FAG NJ-U306E Made In China and Marketed By FAG Germany , Is Not A Crown nor a Spherical roller Bearing , But Has Sharp Rounded ends toopreventthe roller digging in on this side track , And you guy's ever take and notice of what I have Put witch Is the Correct answers, Witch Engineering institute did you lot go too , yours Anna J

Hobviosley knot te sam whone haz u hana

hic

Permalink

Previously Jim Brierley wrote:

Just one word of caution.

The test roller bearing is clearly well used and worn, but even so, still displays a definite difference in surface finish (blending??) for a millimetre or so before the end radius, as is clearly evident on new bearings. Perhaps any minuscule dimensional difference has disappeared through wear??

I would be very interested in the same exercise carried out on a new unused roller.

Just to add to the conspiracy theorists the attached link from the end of the parallel Access Norton thread totally contradicts the video evidence.

Keep Smiling

How can you tell the roller is worn? The graph shows a uniform horizontal line to within 0.0001" deviation.

Please point out the total contradiction.

Permalink

Anna, can you explain " spherical roller bearing" and "sharp rounded ends" I went to Bracknell Technical college and Oxford Polytechnic so where did you go? Tharg?

Permalink

It was me that posted the link to the PDF on access norton forum, its supports the video and does not contradict it.

If you read it carefully it states the drawings of the crowning effect in the pdf are greatly exaggerated and just there so the effect can be seen, it should have a bold NOT TO SCALE stamped on it. It then goes on to explain that crowning (chamfers on the edge) is applied to all parallel rollers, the planned loading determines the amount of crowning as with heavier loads the edges of the roller can dig in to the bearing race. With Nortons whippy crank this digging in was worse than the loading calculation suggested, so by increasing the crowning to match a greater than planned load they solved the problem. The example of a roller that is fully crowned is only for a spherical roller bearing which have never been near a Norton engine and best ignored.

Permalink

Simon - thanks for posting the NASA report on a damp November day. I'd searched in vain for something as comprehensive as that.Top right of page 7 says rollers are usually barrel shaped. But Page 12 says they have parallel sides and radiused corners. As shown by the measurement report.Maybe this is about beer. They don't have wooden cask shaped barrels for beer in the USA: they are what we call kegs - cylinders with rounded ends. As shown in detail on page 12. So a cylinder with radiused ends is barrel shaped over there.We really want to see the measurements on a new old stock Dommie bearing, plus a new old stock Commando bearing, don't we? As John suggests - the difference is likely to be in the bigger radius on the Commando bearing.I don't think I have ever read a complicated technical report that cannot be misunderstood - and if advertising managers then get involved - no wonder the argument has lasted 40 years! I bet the man who devised it all had a good laugh!
Permalink

Previously David Cooper wrote:
Simon - thanks for posting the NASA report on a damp November day. I'd searched in vain for something as comprehensive as that.>You've lost the plot - I didn't post it and it wasn't November.
Permalink

Previously david_evans wrote:

Anna, can you explain " spherical roller bearing" and "sharp rounded ends" I went to Bracknell Technical college and Oxford Polytechnic so where did you go? Tharg?

Dave,

You should have payed attention, no use just turning up! Spherical roller bearing incorporates two rows of symmetrical barrel shaped rollers. The outer race bearing track has a radius so the bearing can cope with mis-alignment and shaft deflection.

Permalink

Dear SimonBut of course you did not post anything useful. Silly me! Jim Brierly did that.Thank you Jim for shedding the light.And I apologise unreservedly for trying to be polite to Simon.Regards David
Permalink

Previously Jim Brierley wrote:

I think you are confusing the crown with the radius at the end of the roller

Two things:

Firstly, I did not confuse the crown with the end of the roller - the paper showed quite clearly the situation in Figure 12.

And

Secondly, you stated that the roller was obviously worn. It wasn't - Jim Comstock makes it clear in another post that the bearing was in excellent condition.

Permalink

So the Superblend is a myth. But how do we stop people saying "barrel-shaped"? Electric shocks maybe? Can the police help?

Permalink

Previously patrick_hook wrote:

Two things:

Firstly, I did not confuse the crown with the end of the roller - the paper showed quite clearly the situation in Figure 12.

And

Secondly, you stated that the roller was obviously worn. It wasn't - Jim Comstock makes it clear in another post that the bearing was in excellent condition.

Hi Patrick

I don't know if you're looking at a different fig 12 to me so I've attached a copy just in case.

The author of the video clearly states that the bearing is used so by definition must exhibit some degree of wear. The only thing that I suggested was that it would be interesting to measure a totally new bearing. We are probably only looking at one half of one tenth of a thousandth of an inch or even less, so measuring it would be a bit beyond a well used micrometer. This has also been suggested in the Access Norton Thread, but Mr Comnoz has asked for a donation of a new bearing so that he can undertake the examination. ANY OFFERS??

There's also some interesting information contributed by J M Leadbeater on the same thread which is worth a read.

Personally I'm in total agreement with she who must be obeyed AJD. Forget the FAG, go for the original Ramsom and Marles 6MRJ-6 if you can get one.

With all the chaos happening in the world, life's a lot easier being an angle dancing on the head of a pin!

Attachments figure-12-jpg
Permalink

Did anyone ever really think that the bearings have the same form as something that would pass muster at a barrel-rolling competition ?

It was made clear in non-specialist publications as long ago as the 1970s that the radiusing of the rollers was carried out in order to prevent line-contact with the inner and outer race as the crank flexed...no-one who has ever taken a pair apart could imagine that the whole bearing face was elliptical...the loaded area there would be hardly wider than with balls.

The term 'barrel-shaped' was surely invented at the time in order to convince the average non-technical punter who couldn't chose between a 750-4 and a Commando that the Combat problems were behind them...they had to keep the terminology simple, it was being aimed at potential Honda owners !

Attachments fireshot-screen-capture-164-2-facebook
Permalink

Jim - we are talking about the same diagram - as Richard says in the post below yours, if the roller was as per Fig. 12 'a', it would provide very little support area. The video Jim Comstock supplied on the other hand clearly showed that the bearing's profile was as per 12 'b'.

I do agree though - 'angles' dancing on the head of a pin would be something to behold!

Permalink

I suppose next we'll be told Superman wasn't all that 'Super' either.... and as for 'Super' Unleaded...

Oh well, at least we've all got Father Xmas to look forward to :)

Permalink

All very interesting. Just a note I ran just a pair of ball bearings in my 650 Dommi which I raced in the 1980's in VMCC events. It was as smoooooth as silk I changes at 7200rpm most of the time never stripped the cases in 5 years and won 3 or 4 unlimited championships (brain fade). I sold the bike and it won another one still on the same bearings and the cases not parted.

My road 1962 650ss had the same set up, two ball bearings no rollers and when I sold it after many years and many thousands of miles it was sounding and running fine. This bike was not potted around at 40mph I assure you.

Maybe the 750/850's needed these but you will never convince me. Or was it the fact I thrashed the balls off it not just rode under 3k revs

If you think about it the crank is so long and cases so thin the crank has to bend. Balls allow for this rollers do not.

I raced early pre unit Triumphs before this, again winning many races and a few championships. I always used the old 3 piece cranks and ball bearings because of the same flexing problems and had no problems. I know of a few others that converted to later one piece cranks and one guy broke 3 in one season racing.

You cannot deny this evidence chaps.

Permalink

The profiles shown in the Hamrock report are exagerations for the sake of illustration. The profilimetry results shown in the first video are consistent with his Figure 12(b) roller.

Norton experienced huge problems with short Main Bearing life, 3000 miles according to the late John Hudson, with Wolverhampton built Commando engines.

Apparently, according to Bob Manns who tested early Combat Commandos, the cranks came in as a casting , the big end journals were machined and then the whole thing split, the machines was worn so that the splitting was inconsistent. The halves were put in separate bins and latter assembled with no attempt at matching'.

John Hudson said that the result of this haphazard production plus the use of a more ridgid crankcase lead to multiple mains failures due to the edges of the rollers digging into their track.

Roller bearings are goood at supporting loads relative to ball races but are intollerent of misallignment.

The solution was to use rollers with radiused edges to avoid dig in , the crankshaft production remained rubbish. Warranty claims returned to accptable levels.

Earier versions of our beloved twins did not suffer unduely from this problem, but if you are changing your mains, you might as well use a partially crowned roller.

Permalink

Previously charles_bovington wrote:

The profiles shown in the Hamrock report are exagerations for the sake of illustration. The profilimetry results shown in the first video are consistent with his Figure 12(b) roller.

Norton experienced huge problems with short Main Bearing life, 3000 miles according to the late John Hudson, with Wolverhampton built Commando engines.

Apparently, according to Bob Manns who tested early Combat Commandos, the cranks came in as a casting , the big end journals were machined and then the whole thing split, the machines was worn so that the splitting was inconsistent. The halves were put in separate bins and latter assembled with no attempt at matching'.

John Hudson said that the result of this haphazard production plus the use of a more ridgid crankcase lead to multiple mains failures due to the edges of the rollers digging into their track.

Roller bearings are goood at supporting loads relative to ball races but are intollerent of misallignment.

The solution was to use rollers with radiused edges to avoid dig in , the crankshaft production remained rubbish. Warranty claims returned to accptable levels.

Earier versions of our beloved twins did not suffer unduely from this problem, but if you are changing your mains, you might as well use a partially crowned roller.

Hello,

Without wishing to be antagonistic is it likely that someone could explain what all this is about please?

Also, what exactly are superblend bearings? as having spent 60 odd years in various forms of engineering I have never actually heard this description before.

Regards, Ian.

Permalink

Ian

It's worth reading the linked thread on the Access Norton Forum (also referenced 'this' at the top of this NOC message thread).

The Hamrock Report referred to is this one

IMHO the term 'Superblend' does not have an engineering definition. It may never have done and is a marketing term created to present Norton's engineering solution to their Commando main bearing failures of the early 1970s. The term has entered Norton mythology with various 'believers' and 'non-believers' that the rollers have to be 'blended' at the edges or even 'barrel-shaped'. It also appears that the term has been applied to different types of bearing (some would say that it is heresy to even state this)!

What does seem clear is that the widely used FAG bearing similar to this one is just an ordinary cylindrical roller bearing. But if you are selling these to a Nortonista they have to be branded 'Superblend' to maintain the 'magic'.

Hope this helps.

Andy

Permalink

Previously Andrew Heathwood wrote:

Ian

It's worth reading the linked thread on the Access Norton Forum (also referenced 'this' at the top of this NOC message thread).

The Hamrock Report referred to is this one

IMHO the term 'Superblend' does not have an engineering definition. It may never have done and is a marketing term created to present Norton's engineering solution to their Commando main bearing failures of the early 1970s. The term has entered Norton mythology with various 'believers' and 'non-believers' that the rollers have to be 'blended' at the edges or even 'barrel-shaped'. It also appears that the term has been applied to different types of bearing (some would say that it is heresy to even state this)!

What does seem clear is that the widely used FAG bearing similar to this one is just an ordinary cylindrical roller bearing. But if you are selling these to a Nortonista they have to be branded 'Superblend' to maintain the 'magic'.

Hope this helps.

Andy

I agree with Andy...just ordinary, albeit,high quality bearings. How anyone could imagine that barrel shaped rollers were necessary is laughable, the crankshaft would have to be made of very soft rubber as to deform enough to change the axial centre line enough to warrant barrel shaped rollers specially as there is a bearing on the other end to brace the crankshaft....reading the Norton blurb of 1971 suggestthat the newbearingsallows a slightly closer fit to the crankshaft webs...that's all.

Les

Permalink

Les is imagining people believing extraordinarily foolish things. I don't believe they do. They look at rollers and believe they are shaped exactly the way they seem to be shaped.

From Accessnorton.com: 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.

I believe that. The OD changes from the middle to the sides. That someone looks at their Superblends and believes they are shaped other than what they see is the laughable suggestion.

Permalink

Previously Jonathan Soons wrote:

From Accessnorton.com: 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.

I believe that. The OD changes from the middle to the sides. That someone looks at their Superblends and believes they are shaped other than what they see is the laughable suggestion.

Wow...Youreally do not believe that people believe "extraordinarily foolish things"??? ....Surely that is naivety taken to the extreme?...Now I am laughing.

Have you checked these measurements yourself?

If they are correct, I would say the shape is not barrel like....just a small rounding off at the edges of the rollers....The Bearing outer track isFLAT isn't it?....OTHERWISE the rollers would not slide in......Les

PS. A roller bearing can handle higher loads compared to a ball race bearing simply because there is a larger swept area of the roller against the track. If the outer race is flat, which it has to be to enable the inner race and its rollers to slide in, even the veryslightest deviation of the rollers profileaway from a parallel shape reduces the contactbetweenthe roller surface and its track, reducing its load capacity. By chamfering the very edge of the rollers prevents scuffing of the track but even this is at a cost of some load bearing capacity.

Permalink

I'm not sure if this answers the question for good but I have a true 'Superblend' roller in genuine Norton packaging and it's old stock from the old address. Only thing is it doesn't seem to fit any Norton I know of. Maybe someone can confirm what it is????

Attachments 20160111_105729-jpg
Permalink

Previously K Glassborow wrote:

I'm not sure if this answers the question for good but I have a true 'Superblend' roller in genuine Norton packaging and it's old stock from the old address. Only thing is it doesn't seem to fit any Norton I know of. Maybe someone can confirm what it is????

Thank you for posting this. Photos, diagrams and measurements greatly help in the debate.

Any chance of providing some measurements, e.g. max diameter, min diameter and length?

Andy

Permalink

Previously K Glassborow wrote:

I'm not sure if this answers the question for good but I have a true 'Superblend' roller in genuine Norton packaging and it's old stock from the old address. Only thing is it doesn't seem to fit any Norton I know of. Maybe someone can confirm what it is????

Yes this roller is definitely barrel shaped, and if itwere to run in a normal flatinner and outer race it would be worse than a ball typebearing as the only metal that would be in contact is a single line round the centre and it would wear down in no time at all. The only way you could utilise this roller is to have it running on a profiled inner and outer track to exactly match the shape....unfortunately the bearing could not be separatedinto two pieces which is required on at least the drive side of the Dommi engine.

Permalink

Sorry that was tongue firmly in cheek....dimensions of 1.25"dia. centre, tapering to 1.15" dia. just before the radius. It was only my light hearted attempt at showing a true barrel shaped roller. Goodness knows what it came out of....some kind of industrial drive shaft if I recall.

Permalink

The nutty things people believe are "the world is flat because it seems that way" and "rollers are perfect cylinders because they seem that way". Science trumps all that. The rollers are not perfect cylinders. Some of us have no difficulty expressing this shape. Some of us have a lot. Those who have difficulty are antagonistic towards those who don't. Isn't that just what you would expect?

Permalink

Previously K Glassborow wrote:

Sorry that was tongue firmly in cheek....dimensions of 1.25"dia. centre, tapering to 1.15" dia. just before the radius. It was only my light hearted attempt at showing a true barrel shaped roller. Goodness knows what it came out of....some kind of industrial drive shaft if I recall.

Apology accepted! It would have been good for April 1st.

Oh boy not only do we have 'believers' and 'non-believers' we've now got makers of 'graven images'.

It's all got too biblical for me.Smile

Andy

Permalink

Previously Andrew Heathwood wrote:

Ian

It's worth reading the linked thread on the Access Norton Forum (also referenced 'this' at the top of this NOC message thread).

The Hamrock Report referred to is this one

IMHO the term 'Superblend' does not have an engineering definition. It may never have done and is a marketing term created to present Norton's engineering solution to their Commando main bearing failures of the early 1970s. The term has entered Norton mythology with various 'believers' and 'non-believers' that the rollers have to be 'blended' at the edges or even 'barrel-shaped'. It also appears that the term has been applied to different types of bearing (some would say that it is heresy to even state this)!

What does seem clear is that the widely used FAG bearing similar to this one is just an ordinary cylindrical roller bearing. But if you are selling these to a Nortonista they have to be branded 'Superblend' to maintain the 'magic'.

Hope this helps.

Andy

Hi Andy

Thank you for your response to my question posted 10 Jan' concerning S/blend bearings and please accept my apologies for my slow reply. Having waded through the discussion up to the point of my question I was beginning to think that the various writers were in denial of the fact that spherical roller bearings existed. As I'm sure you're aware, they do exist and are manufactured to cater for flex in a shaft, whether it is a crankshaft or any other power transmitting shaft. Having, some time ago stripped and rebuilt the engine of my 1966 BMW R69S I can confirm that this type of bearing was employed as the rear main bearing, the new bearing I fitted just had numbers, no fancy description. It was found that the earlier R69 had crankshaft flex, hence the rear parallel roller was changed to a spherical with barrel shaped rollers for the following R69S.

Regards, Ian

PS:Apologies for mentioning BMWs on a Norton Forum

Permalink

Thanks Ian

Yup barrel roller bearings aka spherical roller bearings or fully crowned roller bearings do indeed exist (see 'tongue in cheek' photo posted earlier in the thread by K Glassborow). IMHO just not in the context of Norton Commando main bearings.

FAG call these barrel roller bearings and they are designed to deal with high radial shock type loads and where radial misalignments have to be compensated. They only have low axial carrying capacity.

An FAG barrel bearing with similar dimensions to a Commando main bearing would also have a limiting speed of 5000 rpm compared with 10000 rpm for an NJ306E, i.e. the type of cylindrical roller bearing marketed as 'Superblend'.

Andy

(Fast becoming an amateur tribologist!)

 


Norton Owners Club Website by White-Hot Design