This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

MAIN BEARINGS FAILURE, MUST CHANGE

popup said:
ausberg ,just had a look on the *** bearing bearing site,on the 6308 bearing for the xr650,it has less load rating than the standard nj206.
nj206 45kn dyn.compared to 6308 42.5kn dyn.
size really doesn't mean anything when you compare a roller to a ball.

Hang on guys, have you both completely lost the plot here :? :wink:

You've got two completely different bearing types - one ball and one cylindrical roller - PLUS they are both completely different dimensions (i.e. the 6308 has a 40mm i/d).

They are therefore not like for like comparisons and even though I know you are making a comparison with the XR650, it is a totally different engine with totally different design parameters and intended function.

And the 6308 isn't relevant to us because it would never fit.

I really thought we might have moved on slightly, a step away from just looking at the mains to considering some of the issues that haven't yet surface, but have only been vary briefly circumscribed.

One subject I would certainly be interested to learn more about torsional vibration, for example.

All the best,
Simon
 
Ausberg,
I don't race SM, but was particularly intersted in this main issue because my bike was noisy around the left main. Never had any issues with the bike til a couple of desert races and the noise started. Turned out to be a loose primary gear. Really thought the main could be the problem, cause in the past other fellows were having main failures in the dirt, some from down under I think. Anyway, it wasn't the main causing all the gear slap, bike is now fixed and set up for dual sport and the soft suspension is perfect for that. The old 610 Husky will be retired from the trails. This fall am going to buy a new strictly desert racing bike so now my only worry is which one. Thanks for the info on the KTM.
dan
 
AUSBERG said:
I would gladly pick up say 750grms of main bearing mass if it meant they had a lot longer service life. :)

sometimes the bearing is dictated by the OD. a larger OD bearing would bring the machined hole far too close to another machined hole such as the timing shaft hole. the resonance of revs would see these two crack towards each other.

this is something never thrust upon the owner because husaberg R & D may have found this out? so 62mm is it then!

we also realise that there is nowt wrong with the bearing. so that leaves about 10 other things that it could be?

we've talked about:

poor oil (doubtful)
too tight bearing housing - probable
ball race too loose at C4 - doubtful because dale recommends a C3.
c/b sits on crankshaft = design fault - possible
crankshaft too weak - possible
crankcases can't cope - possible
axail tolerance problems - probable as each crank gives up at any time and any miles. seems very possible! however DL says too loose whereas i think it's too tight?
revs - no
bearings - no

ausberg again:
But I wonder how much more axial loading the 6308 could take- being a ball my guess is a lot more- especially as how a lot of the guys attribute the NJ206s failures to the thrust flanges cracking

i know that orangeberg said that the inner flange crap? is that it? and why cracking? ideas? the SKF tech bloke siad that you would find spalling on the inner race especially if the bearing started to pick up?

regards

Taffy
 
we've talked about:

poor oil (doubtful)
too tight bearing housing - probable
ball race too loose at C4 - doubtful because dale recommends a C3.
c/b sits on crankshaft = design fault - possible
crankshaft too weak - possible
crankcases can't cope - possible
axail tolerance problems - probable as each crank gives up at any time and any miles. seems very possible! however DL says too loose whereas i think it's too tight?
revs - no
bearings - no

Taffy[/quote]

Hallo
I can agree on all of above mentioned points, specially since SKF factory with R&D is located in Gothenburg 30km from our workshop, and they have tested our bearings after they brake and they have no cause why it breaks. Its not the bearing it self, but internal construction fault.
Well its a race engine, live with it and pre-maintnence will keep the engine running, on some engines we change the bearings after 4hrs just in case( supermoto). The Enduro factory team changes often i belive.
Regards
Thomas
 

Attachments

  • CadlinkMIMEa03528.pdf
    18.5 KB
Taffy said:
bearings - no

Yeah and lets get this one nipped in the bud, FFS :roll: :wink:

Just to back it up a bit and doing an on-the-back-of-a-***-packet (not mine coz I stopped years ago) (and that's a cigarette packet to most of you), the force on the crank at max piston acceleration I estimate to be around 24kN (KiloNewtons) at 9000 rpm, 20 kN at peak power on the 644 engine. Given that the ball bearings have a dynamic load of just above 24kN they are essentially taking half their max dynamic load at extreme engine rpm - 12kN each. The rollers cope with double that.

The mass assumed in this calculation is the "reciprocating" mass of 515grams which is half the conrod plus the piston and pin not the rotating mass of everything else (essentially the same estimate done for balancing the crank) and a conrod length of 134mm with a stroke of 82mm.

And just in case I haven't made this point earlier :? - minus the c/b and the problem on most accounts goes away. What is the job of the c/b - to work with vibration.........

All the best,
Simon
 
I think it would be interesting to see a real world measurement of the lateral crank case expansion at full engine temp. versus the real world expansion of the crank shaft under the same conditions. At least then we would see by how much the end float grows/ or reduces in running conditions.

Anyone done something like this?
 
what's your take on the cracking of the inner race/flange on these/

anybody?

regards

Taffy
 
Lets assume, for the sake of argument that the crank is absolutely fine, Husaberg have done their calculations perfectly and the crank is therefore perfectly stiff enough to do the job.

What else would remain?

Too much axial thrust which is rapidly wearing the bearings out?

In part the possibility of this can be reduced by ensuring satisfactory end float. (it would still be interesting to know the expansion ratios between crankshaft and crankcase).

The other can be to "lock" the crank in place to one side of the crankcase as I've already mentioned to prevent sideways movement of the crankshaft in operation. This is done with the use of a carefully made collar that sits under the primary gear.

Another reason could be something that we've all heard and we're familiar with in that the inner race has too much of an interference fit with the crankshaft. In these circumstances it is perfectly possible that the inner race changes shape according to the shaft it sits on causing problems. It may also change somewhat the internal clearance.

On engines where main failure is frequent has anyone measured the crankshaft runout and ensured trueness throughout? Ensuring it is within tolerances when it has been put together?

I haven't seen much discussed on this other than by DL....

These are further reasons why I still fail to accept the hypotheses put forwards.

There are still too many other variables that still have to be eliminated...

All the best,
Simon
 
Put an old crank in a press with it resting where the left and right mains sit. Lube where it is sitting with a bit of motor oil and with the big end up, press on the big end. Will spread on the opposite side of the big end without a terrible lot of pressure. The further you move from the crank throws, the easier it gets. Spread it too much and endfloat would dissapear and bearing shoulders would take the heat. Sorry didn't make it clearer.
dan
 
dsducati said:
Put an old crank in a press with it resting where the left and right mains sit. Lube where it is sitting with a bit of motor oil and with the big end up, press on the big end. Will spread on the opposite side of the big end without a terrible lot of pressure. The further you move from the crank throws, the easier it gets. Spread it too much and endfloat would dissapear and bearing shoulders would take the heat. Sorry didn't make it clearer.
dan

Its been perfectly clear for a while now dan :). You dont need to convince me :wink:

The fact that the Husaberg engineers increase the emd float spec as the stroke increases in length pretty much supports exactly what you have just said.

Well, in my eyes anyway 8O .
 
So, come on then guys and if it is that obvious:

How much pressure is required and how much does it spread by? What are your measurements? What is the resulting crankshaft runout - statically and dynamically?

And with all due respect, the test mentioned is only a static test and wouldn't necessarily bear resemblance to a dynamic test....

Unless you can actually back up what you are saying with some figures, it is simply conjecture and if you have done the test you should have the figures readily to hand.

I personally would be more than happy to accept that I am totally wrong and that everything is down to the crankshaft BUT not one single person who blames the crank can back it up with concrete evidence. That says a lot.

All the best,
Simon
 
dan

i think what you said was: press down on the middle of the crank and it squidges out axially left and right? right? yes i can grasp that!

simon
the axial float is suggested as .65mm isn't it? by the factory which is .026" (thou). on the other hand dale recommends 0.21mm (or .008") so we should read something into that. don't ignore it!

i also totally disagree with holding the crank hard left or right as the rollers will graunch hard against this lip everyone is on about. you want to hold the crank still: what have i advocated? angular contact ball race bearings!

the next one si, is that on the ball race dale doesn't recommend a C4 (the loosest fit) nor the C2 which is almost an interference fit but the C3 which one could describe as "taking up the slack under the normal thermal expansion found in yer average motor!". this i suspect cuts out the question i had put up previously and you do now: is the inner race crushing the balls and therefore squeezing the oil out?

apparently not!

anon.

i will now add a question as you are on about the crank: my ducati had excellent primary balance coz it was a V-twin and each cylinder was only a baby 250cc. yet the bearings were 72mm od and 80mm od. as well as that, the crank journalks were 35mm and not the 62mm and 30mm of the 650 so come on folks.....they have upped and upped the CC over the years, they have shortened the stroke and made every one of the three engines well over square. i think the crank is part of the problem.

i think it was Drc who said that the big end was the wrong diameter versus it's width but i suggest that as long as you don't make the big end wider but make it fatter you have helped stiffen the crank surely?!?! you gain more contact area ayt the press fit and surely dan's scenario is not going to happen? so why the move to 35mm big end when - and correct me if i'm wrong? - the 650 big end was having no trouble?

as ausberg has had his mains go, i've asked him if he would measure the bearing housing in the cases, to measure the run out on his crank, and if the housings are out to have an arbor made and check the alignment!

BTW, popup/weed - if you have made one and could pass it on i'm sure he'd appreciate the loan of such a fine piece of equipment!

it's interesting that dale doesn't 'give up' on the roller bottom end, he maintains that if the housings are correct, if the alignment is right then rollers aren't a problem!

that's why everyone here should be getting their internal dial gauges-a-twiddling and measure the crank run out and housing sizes down to almost the micron!

regards

Taffy
 
Taffy said:
don't ignore it!

I'm not :wink:

I'm probably one of the people who is most aware of it as it seems I have been the only one to be able to post the data :roll:

And the .65mm is really the max figure..

Taffy said:
i also totally disagree with holding the crank hard left or right as the rollers will graunch hard against this lip everyone is on about. you want to hold the crank still: what have i advocated? angular contact ball race bearings!

I was not talking about holding it hard left, it is simply a spacer of the correct dimension to hold it in place. That does not mean there is a preload on the bearing.

I'm also not the only one. There are plenty of engine builders who use this, whether right or wrong. One of them I believe is JBS of the 750.


Taffy said:
the next one si, is that on the ball race dale doesn't recommend a C4 (the loosest fit) nor the C2 which is almost an interference fit but the C3 which one could describe as "taking up the slack under the normal thermal expansion found in yer average motor!". this i suspect cuts out the question i had put up previously and you do now: is the inner race crushing the balls and therefore squeezing the oil out?

You haven't understood my statement taffy.

It is possible to use a C4 even if Dale recommends a C3. Personally, I would also select a C3 bearing. C4 is not the largest internal clearance, or loosest as you put it, a C5 is.

I have taken C4 bearings out of these engines that came from the factory even.

In my previous statement I am talking about the interference fit on the crank shaft of the inner bearing race. If the fit is too tight it can change the shape of that inner race and cause problems from there. Including unnecessary vibration, increase stress on the rollers and so forth, including reducing the internal clearance from spec.

A sign of such an impending problem can be a slightly interference fit when installing the crank.

There are also plenty of engine builders out there who will machine the bearings or internal engine components the make sure the tolerances are good.

Taffy said:
i will now add a question as you are on about the crank: my ducati had excellent primary balance coz it was a V-twin and each cylinder was only a baby 250cc. yet the bearings were 72mm od and 80mm od. as well as that, the crank journalks were 35mm and not the 62mm and 30mm of the 650 so come on folks.....they have upped and upped the CC over the years, they have shortened the stroke and made every one of the three engines well over square. i think the crank is part of the problem.

Well, you are confusing two separate things here. However, the two bikes have totally different design parameters and Husaberg will have designed the mains to be the minimum size required for racing.

I go completely with Enginehardware's comment to that effect. In these engines they are not designed to last a lifetime. The Duc. will have had a much longer lifetime in the design parameters.

However, what I want to know is that I can have mains that last a specific number of hours and know that they will last that time. The problem lies here - in many cases they don't last even the short time required.

Taffy said:
i think it was Drc who said that the big end was the wrong diameter versus it's width but i suggest that as long as you don't make the big end wider but make it fatter you have helped stiffen the crank surely?!?! you gain more contact area ayt the press fit and surely dan's scenario is not going to happen? so why the move to 35mm big end when - and correct me if i'm wrong? - the 650 big end was having no trouble?

In supermoto, as soon as you increased the compression, big end problems did arise. They increased the 650 compression in several stages, the first of which was in 2003.

It is worth keeping in mind that the force onto the big end pin is distributed almost across its entire width, it's not like it's a 2 foot cantilever. The force acting upon it at peak acceleration is therefore more a shearing force than a bending force, but I suppose that resultant forces may cause a bending force.

If you are considering loads on the crank, you also have to consider conrod length as this will also have an affect on peak acceleration and therefore crankshaft load.

Because it's a larger diameter the big end will also add stiffness and it will also change the resonant frequency of the whole crankshaft assembly.

The other thing that should be mentioned is that it is generally thought that peak force on the crank is at tdc on the exhaust stroke, not from combustion pressure.

Taffy said:
it's interesting that dale doesn't 'give up' on the roller bottom end, he maintains that if the housings are correct, if the alignment is right then rollers aren't a problem!

And that should be a clue on the crank and he added mass quite happily to the things.

But in every occasion he removed the counterbalancer, particularly in his racing engines.

This should be an even bigger clue.

Let me explain a little more about my harmonic theory.

I think that the crank shaft may be flexing naturally but under most conditions it is not enough to cause any problems. in most things we expect some flexion anyway. But by and large the crank on its own is stiff enough.

Add to the crank a significant weight spinning away contrary to the crank shaft which, while the engine is running adds noise and vibration to the whole assembly. On many occasions this works absolutely fine. Then on top of that you have vibrations added by the running gear etc. through the primary gear and so on.

However, there are certain times where the vibrations end up matching the resonant frequency of, lets for arguments sake say the crank shaft and suddenly the vibrations amplify causing that dangerous significant movement in the crankshaft assembly.

It is a bit like those first generation suspension bridges that under normal circumstances happily took the load of the traffic but when the wind blew in a certain way, the bridge shook itself to bits - without any load on it

This doesn't mean I think that the crank is woefully bending but I think it end up vibrating in such a way as to cause damage.

It can be a very distructive force, like adding all the amplifiers in a nighclub to the output of your ipod headphones.

So to quote your words of wisdom "i think the crank is part of the problem"

But I don't think it is the problem per se. If you know what I mean.:wink:

This is also why I don't think it is quite a simple as the factor making the big end fatter.

If it were just crank flexion, I think they'd have solved it ages ago, becuase it would be simple and cheap. KTM and Husaberg are pretty good mechanics.....

All the best,
Simon
 
As i said it's only my Guess, same as everone else is doing. I would also guess that KTM engineers know the answer but because of small sales of Husabergs and expensive retooling aren't going to change anything major. If there was a simple answer such as installing spherical rollers it would already be done by the factory, so I'm probably wrong about that too.
 
si

i don't care how you dress it we are on about an expanded inner race causing problems which i would take to start with pressure onto the balls and crushing them between the races. anything else i reckon they cann cope with. because dale recommends C3 and not C4 or yes C5, then i think it's fair to say that this inner race theory hsn't been found to be there by dale.

i don't believe you can hold the crank anywhere axially, i think it must be left to float left to right and should be shimmed at the crank. i believe that it wasn't necassary to shim the ball races and didn't particularly matter. however, when you introduce a .21mm tolerance it must be controlled and i take that tto be shimmed. this the owner hasn't been in control of and the factory don't appear to utilise.

as i have said, the inner races need setting far enough apart that the rollers can run clean down the middle of their tracks. we have straight-cut primary drive which should help.

i don't get your point over the size of the mains? you should have mains that last forever in their ability but we change because we have only a small amount of oil, because we only just have a cooling system that works. i don't think you can control the service life of a bearing how the factory and you want it too. for me - a bearing should be up to the job through thick and thin and not just through thick as spec'd on a bit of mattighofen A4 paper!

so to me if the bearings fail they aren't up to it because they should be over engineered.

to you they make the 6206 because it's racing stuff, to me they do it because they can't fit better stuff in. as i said in my first post on this thread - if you make the bearings any bigger there will be a fault line from the MB housing to the timing gear housing. i should know: i got through 4 sets of cases even in my little duke.

so to me they would like to go bigger but can't - that i'm sure of. add to that the classic "we built this bike to be a 501 and now it's a 650SM" and you have a recipe for disaster!

whilst the bearings themselves are doing nothing wrong husaberg may still have been picking the wrong ones off the shelf. we need more data on the housing IDs and journal ODs etc.

i look forward to my winter rebuild.

regards

Taffy
 
Taffy said:
whilst the bearings themselves are doing nothing wrong husaberg may still have been picking the wrong ones off the shelf. we need more data on the housing IDs and journal ODs etc.

i look forward to my winter rebuild.

regards

Taffy

Measured mine up yesterday :)
main bearing bores on my engine cases come in at 61.92mm.
 
well according to the figures i have, the housing should be 61.99mm. as i've said before, 25 microns make one thou, it takes 4 thou to make 0.1mm so that means all-in that 100 microns = .1mm

if there is an allowance of +2 to +11 microns that means (if we take 10 microns) that we're allowed 10 of that 100. that equals 61.9 and 9/10ths. = 61.99. we could have 61.989 to 61.998mm.

it could therefore be said that the housing is out by between 78 and 69 microns.

and what did i (and ausberg) just cure?

nowt!

did you mic the journal or a bearing there ausberg? just a thought?

regards

Taffy
 
Taffy said:
this inner race theory hsn't been found to be there by dale.

Actually Taffy, he did. I think there are still some posts on the site where he talks about it.

Taffy said:
i you should have mains that last forever in their ability but we change because we have only a small amount of oil, because we only just have a cooling system that works. i don't think you can control the service life of a bearing how the factory and you want it too. for me - a bearing should be up to the job through thick and thin and not just through thick as spec'd on a bit of mattighofen A4 paper!

With all due respect Taffy, you're now entering into the realms of fiction. :wink:

Of course you can control, and design, service life to a very good degree. Of every component of the engine. Which means that given a set of operating conditions the relevant parts can be designed with a given service life. That includes each and every bearing. Of course, it may not be 100% guaranteed but the probability can be put fairly and squarely in the domain of knowing by and large what the service life is going to be).

That's one of the ways you get to putting together a service schedule - ultimately it's worked out theoretically and empirically. Of course, there is always going to be some compromise.

Nothing lasts forever, whether it is used or not. Full stop.

Taffy said:
add to that the classic "we built this bike to be a 501 and now it's a 650SM" and you have a recipe for disaster!

Given that this engine was designed from scratch it is inconceivable it was built to be a 501 and ended up as a 650.

If it had been a slow evolution, maybe, but as a newly designed engine line up. Not at all.

Taffy said:
whilst the bearings themselves are doing nothing wrong husaberg may still have been picking the wrong ones off the shelf.

I simply don't think they have picked the wrong bearings off the shelf. Full stop.

The bearings they are using are up to the job. That I thought we've already confirmed.

It also still ain't the crankshaft :)

All the best,
Simon
 

Register CTA

Register on Husaberg Forum! This sidebar will go away, and you will see fewer ads.

Recent Discussions

Recent Discussions