MAIN BEARINGS FAILURE, MUST CHANGE

i think that all the other manufacturers are 8 and the NTN is 9 standard?

i was also told about a cage that disintegrates without damaging the balls. ausberg?orangeberg?

regards

Taffy

Well, you've done an amazing amount of work on this Taffy, good on you.

My comments are as follows:

I think is is misleading to have photos comparing SKF and OEM bearings that are completely different types - i.e. ball bearing versus roller bearing. It implies one is inferior to the other when this is patewntly not the case and it is not a true comparison without a clear explanation of the illustration. Also, given that OEM bearings are SKF it's kind of weird.

You've used the data sheets from *** when the OEM bearings are SKF and they have different data - e.g. the SKF bearings have higher load limits and limiting speeds than ***.

The suffix ETN9 is not a reference to 9 balls. The E suffix stands for reinforced ball set and the TN9 stands for a type of reinforced polyamide cage. It does have a higher load rating than the 6206

You say:

"The counter-balancer should be removed and a wider (20mm roller instead of 16mm) roller bearing be fitted with the appropriate spacer made to suit."

I'm not in agreement with this and I don't see any evidence supporting this statement in terms of reliability. JBS is using some sort of convoluted spacer and wider bearing and we've heard someone else say they are doing it. However, I believe in Sweden where this is allegedly happening, there is still a problem with main bearing failures so it certainly doesn't resolve any problems. The JBS solution is also pretty much a different engine...

Before making this type of statement we actually need some evidence - at the moment it is conjecture.

OR we need to be very clear that is is just one solution a couple of people have used.

AND

Removing the balancer and keeping the standard 16mm bearing is abolutely fine and is a simple fix. The wider bearing is not necessarily a better solution than this (I don't believe Dale used a wider bearing either).

You state:

"The one on the right is against the flywheel cheek while the one on the left sits some 16mm away".

The actual figure is just over 12mm, not 16mm

You state:

"The rollers can catch a ringlip and start chaffing the edge, the chippings cause spawling and the rest - as they say - is history!"

In terms of the left hand bearing failures, we still haven't seen conclusive evidence that this is what is happening.

Historically older single cylinder engines showed this type of problem but is was actually the point load of the cylindrical rollers digging into the bearing that caused the spalling. However, the left hand bearings don't show this kind of damage - or at least none of the photos we have so far been shown.

One of the causes of this digging in was the shape of the cylindrical rollers but nowadays I believe the rollers actaully have a chamfer to prevent this from happening.

On the rarer right hand failures, we also don't know exaclty what is happening because things go quiet when we ask specific questions.

We should be very careful not to get carried away with out of date notions from the days of Brit thumpers

A final minor comment:

"Bearing notes: The NJ6206 series bearings"

It is of course 206 rather than 6206.

If my memory serves me correctly the stander ball bearings are 9 balls, but then I've lost count on this topic.......

I hope this helps,
Simon

i've made many of the changes you have suggested but at the end of the day i have taken some of the information in good faith from tuners and without studying every little deytail taken their info as correct.

like all these things we lay down an outline and then 'embellish' it.

i think that maddel will move some photos/charts sometime today to the relevant section and then it'll be pretty much there.

regards

Taffy

Simon said:

Well, you've done an amazing amount of work on this Taffy, good on you.

My comments are as follows:

I think is is misleading to have photos comparing SKF and OEM bearings that are completely different types - i.e. ball bearing versus roller bearing. It implies one is inferior to the other when this is patewntly not the case and it is not a true comparison without a clear explanation of the illustration. Also, given that OEM bearings are SKF it's kind of weird.

You've used the data sheets from *** when the OEM bearings are SKF and they have different data - e.g. the SKF bearings have higher load limits and limiting speeds than ***.

Simon

Click to expand...

Hi Simon,
I dont believe that the bearings have always been SKF. I think you will find the first of the Husaberg NJ roller equipped motors used NTNs?? :?

KTM were using NTNs in the RFSs and I believe if you check you will also find Husaberg used NTNs in the first roller mained versions…..

It was only in 04-05 that KTM did the swap to SKF as a fitment for the RFS and Husaberg mains I think.

Please correct my belief if I'm wrong.

Simon said:

You say
"The counter-balancer should be removed and a wider (20mm roller instead of 16mm) roller bearing be fitted with the appropriate spacer made to suit."

I'm not in agreement with this and I don't see any evidence supporting this statement in terms of reliability. JBS is using some sort of convoluted spacer and wider bearing and we've heard someone else say they are doing it. However, I believe in Sweden where this is allegedly happening, there is still a problem with main bearing failures so it certainly doesn't resolve any problems. The JBS solution is also pretty much a different engine...

Before making this type of statement we actually need some evidence - at the moment it is conjecture.

Simon

Click to expand...

I dont really see what the issue is for those who choose to go the 20mm wide bearing.
HOWEVER- I am suprised that no one seems to go to the trouble of welding the cases and remachining the bearing bore so the 20mm wide bearing is supported by the cases for its full width.
As it stands the bearing bore in the cases is only 15.3mm deep- putting a 20mm wide bearing in there leaves a lot of unsupported bearing (and maybe a shift of the bearings " load centre" more inboard away from case support).
A pro job would be as above- 20mm wide bearing, cases welded and machined to support it fully.

Simon said:

If my memory serves me correctly the stander ball bearings are 9 balls, but then I've lost count on this topic

Simon

Click to expand...

I think I may have helped with the 8 ball myth……
When my RHS main failed I recovered 8 balls amongst the shrapnel that fell out when the cases were split (see pic in my gallery)……obviously I didn’t look hard enough
Ball No9 is probably still under the workbench somewhere :lol: .

All 6206s I have checked since - SKFs and NTNs have been 9 ball- as nsman has been trying to tell us ever since.. :wink:

And, yes Taffy job well done compiling all this info for the owners doc- it will no doubt save many people much searching.

PS- theres a thread going on at KTMTalk at the moment in the RFS section re main bearing failures on the RFS engines................. :shock: ???

AUSBERG said:

Hi Simon,
I dont believe that the bearings have always been SKF. I think you will find the first of the Husaberg NJ roller equipped motors used NTNs?? :?

KTM were using NTNs in the RFSs and I believe if you check you will also find Husaberg used NTNs in the first roller mained versions…..

It was only in 04-05 that KTM did the swap to SKF as a fitment for the RFS and Husaberg mains I think.

Please correct my belief if I'm wrong.

Click to expand...

Maybe not all the time but definitely for the majority of the time since about 1989 :wink:

If anything the NTN's were simply a blip because we have to remember we also had the ball bearings for years.

AUSBERG said:

I dont really see what the issue is for those who choose to go the 20mm wide bearing.

Click to expand...

Neither do I and I'm sorry if my opinion has come across that way.

I really don't care what people do to their engines. I do care that they're given the relevant information so that they can make properly informed decisions.

In my view where it said people "should" use a wider bearing ignored the fact that using the same width main bearing is perfectly adequate.

In fact given the weight of evidence, using the same width bearing, ball or roller without counterbalancer wins hands down. It is also a solution that can be done without any machining because in desparation you can use the inner races of the balancer bearing as the spacer and use the crank without rebalancing - it is a very relevant option. Cheap cheerful and works.

I am really wondering what the issue is with accepting this one.....

The only evidence we have of the wider width bearing being used is spurious at best and we have no long term data either. The JBS big bore there just aren't enough units out there and the others no real data either - I am therefore extremely cautious about it being a realistic alternative.

I think therefore it is plain wrong to give a recommendation for it but by all means mention that some people are trying it out.

AUSBERG said:

PS- theres a thread going on at KTMTalk at the moment in the RFS section re main bearing failures on the RFS engines................. :shock: ???

Click to expand...

I'm glad I'm not a member over there then :wink:

All the best,
Simon

simon

i didn't recommend 20mm over 16? that sentance has been erased for some 24 hours now so i can't say for sure how it was worded but i would write "should"in reference to doing it the right way and didn't write that it was the best or even a prefered option.

this is the first 'problem' i think i've had with the doc in nearly 4 years now. i rely on hard work after others have posted to get to the facts but where i can't i have run with it.

anyway, i think it's about there now and so i say 'let it pass over'. it's too easy to get bogged down in bollocks.

regards

Taffy

Taffy said:

simon

i didn't recommend 20mm over 16? that sentance has been erased for some 24 hours now so i can't say for sure how it was worded but i would write "should"in reference to doing it the right way and didn't write that it was the best or even a prefered option.

this is the first 'problem' i think i've had with the doc in nearly 4 years now. i rely on hard work after others have posted to get to the facts but where i can't i have run with it.

anyway, i think it's about there now and so i say 'let it pass over'. it's too easy to get bogged down in bollocks.

regards

Taffy

Click to expand...

Trust me Taffy, it wasn't a criticism of you or the doc and I am certainly not wanting to get any further bogged down . I was simply clarifying my position following ausberg's post

All the best,
Simon

Actually, I'm sorry Taffy but the sentence is still there - first paragraph:

"The counter-balancer should be removed and a wider (20mm roller instead of 16mm) roller bearing be fitted with the appropriate spacer made to suit."

Maybe I'm wrong but to me that implies a recommended fix.:wink:

Simon

queen's english old boy: one should remove the cb. ' must' is a touch too pushy. sorry old boy!

but i see the way you're thinking.

too-da-loo!

taffy

just another couple of things worth considering in my opinion.

to ausberg
,building the cases up with weld around the main bearing housing sounds good and may also look stronger but in high load areas will crack right next to the weld every time.
i've tried welding up kick starters a couple of times,and is a complete waste of time ,it weakens the metal next to the weld.in the case of the cr500 engine case that cracks from the kick starter load,they have it welded and within no time cracks right next to the weld again.
new engine case is the only option.
i've seen times when welding the cases will work but only in low stress areas,weld around your main bearing housing and you taking a huge risk.
i used 3 bond bearing retainer in mine.i couldn't believe the pressure i had put on spherical roller to get it out of the engine case.when the bearing let go of the case ,i thought it broke the case,believe me this stuff works.another option is to do what bergroadracer does.
to Simon,
consider this,if the 2206 bearing has a 2mm thicker flange on the bearing inner than a 206
it would have to be less prone to getting cracked off in my opinion.

in the other type of bearing failure where the running surface is worn,it could also possibly help,that is, same load over a wider running surface,but this will make no difference if the problem ends up being a lack of oil to the bearings.

if i was doing another rebuild,i would be doing as enginehardware does,making up a washer that butts up against each wheel of the crank,that supports the flange of the main bearing inner.toss the c/b.then make up a spacer to suit.

i feel ,if the flange is going to break off mine,it will be the thinner 206 flange and not the 2206.
well done to taffy for his efforts again in the doc regarding this subject
looks like its all fine tuned for the politically correct now.
looks like i'm officially named as a butcher,that is probably 1/2 correct
..weed..

hey weed, don't you know it! you get repreived in the next chapter when you invent the wheel - again!

the good news is that there's a happy ending!

regards

Taffy

Taffy said:

queen's english old boy: one should remove the cb. ' must' is a touch too pushy. sorry old boy!

Click to expand...

Oh feck, here we go, Taffy telling us the Queen's English.:wink:

In that case, may I correct you, sir.

The meaning of should as you've used it in English is an auxiliary verb indicating that an action is obligatory.

Then in your sentence you use "and a wider bearing".

Since and is a coordinating conjunction it implies that the fitment of a wider bearing is also obligatory to be done as well as removing the counterbalancer.

So, as per my my comments, I think that sentence should be changed and particularly, which is my issue, the reference to the wider bearing needs to be clarified.

You did fooking ask! :wink:

All the best,
Simon

popup said:

if i was doing another rebuild,i would be doing as enginehardware does,making up a washer that butts up against each wheel of the crank,that supports the flange of the main bearing inner.toss the c/b.then make up a spacer to suit.

Click to expand...

i asked if they had taken into account the positioning of the crank in relation to the middle line of the barrel...... and also the radiusing of the corners in the crank.

regards

Taffy

Ok, here is a final suggestion for the doc which I have drafted as I thought it might be helpful to list the various approaches people have taken. It may or may not be appropriate. It reads as follows:

What's being done to fix the problem?

Unfortunately, there is no common consensus about what the most reliable fix is. Instead, we have a number of people around the world trying out different things. Some of these can be done without needing any machining or making of non-standard parts, some of them require specialist equipment and a capable engineer to get them done.

Most main bearing failures seem to have occurred when the roller bearings (NJ206) were introduced and while failures with the ball bearing (6206) are not unheard of they are a rarer occurrence. The counterbalancer unit is also suspected by some to play a role in the main bearing failures.

The various approaches to this problem that we are aware of are listed briefly below but if you know of one we haven't covered, please let us know. They've been roughly listed in order of engineering complexity:

1.This is the most straight forward approach and it is simply to replace the roller bearing with ball bearings;

2.The next is to replace the rollers with ball bearings and in addition, remove the counterbalancer. When reassembling, simply use the old counterbalancer bearing in its entirety or just its inner races as a spacer on the crankshaft. Obviously, removing the balance unit without rebalancing the crank will change the balance factor slightly and therefore there will be a vibration change to the engine;

3.Next in line in terms of complexity is to do the above but also have the crankshaft rebalanced to the figures supplied above and machine up a spacer in replacement of the balancer;

4.The next approach is to continue to use the roller bearings, remove the counterbalancer and re-balance the crankshaft but in order to make this work, it is highly advisable that you check the bearing alignment in the as well as tolerances as per the above mentioned tests;

5.Now we have an approach we don't have a lot of data on but we do know it is being used in the JBS big bore engine kit. This involves removing the counterbalancer, re-balancing the crankshaft as per above but instead of the usual 16mm wide roller bearing on the left hand side of the engine, a 20mm wide roller bearing is used instead, then a thinner spacer is inserted where the counterbalancer used to be.

6.From Sweden (enginehardware) we have been told of an approach they are using and as far as we aware it seems to be fairly successful and it is being used in their 670cc kit too. This approach is fairly straightforward in that it involves making up a 3mm washer that sits between the left hand main bearing and the counterbalancer bearing - the objective behind this is to provide additional support to the main bearing flange. In order to get this extra 3mm the old single row counterbalancer bearing is used and a lighter counterbalancer weight installed.


The choice is up to you all.

All the best,
Simon

Good work. Section 4 should read ".....bearing alignment in the _cases_ as well as....."

Simon said:

Ok, here is a final suggestion for the doc which I have drafted as I thought it might be helpful to list the various approaches people have taken. It may or may not be appropriate. It reads as follows:

What's being done to fix the problem?

Unfortunately, there is no common consensus about what the most reliable fix is. Instead, we have a number of people around the world trying out different things. Some of these can be done without needing any machining or making of non-standard parts, some of them require specialist equipment and a capable engineer to get them done.

Most main bearing failures seem to have occurred when the roller bearings (NJ206) were introduced and while failures with the ball bearing (6206) are not unheard of they are a rarer occurrence. The counterbalancer unit is also suspected by some to play a role in the main bearing failures.

The various approaches to this problem that we are aware of are listed briefly below but if you know of one we haven't covered, please let us know. They've been roughly listed in order of engineering complexity:

1.This is the most straight forward approach and it is simply to replace the roller bearing with ball bearings;

2.The next is to replace the rollers with ball bearings and in addition, remove the counterbalancer. When reassembling, simply use the old counterbalancer bearing in its entirety or just its inner races as a spacer on the crankshaft. Obviously, removing the balance unit without rebalancing the crank will change the balance factor slightly and therefore there will be a vibration change to the engine;

3.Next in line in terms of complexity is to do the above but also have the crankshaft rebalanced to the figures supplied above and machine up a spacer in replacement of the balancer;

4.The next approach is to continue to use the roller bearings, remove the counterbalancer and re-balance the crankshaft but in order to make this work, it is highly advisable that you check the bearing alignment in the as well as tolerances as per the above mentioned tests;

5.Now we have an approach we don't have a lot of data on but we do know it is being used in the JBS big bore engine kit. This involves removing the counterbalancer, re-balancing the crankshaft as per above but instead of the usual 16mm wide roller bearing on the left hand side of the engine, a 20mm wide roller bearing is used instead, then a thinner spacer is inserted where the counterbalancer used to be.

6.From Sweden (enginehardware) we have been told of an approach they are using and as far as we aware it seems to be fairly successful and it is being used in their 670cc kit too. This approach is fairly straightforward in that it involves making up a 3mm washer that sits between the left hand main bearing and the counterbalancer bearing - the objective behind this is to provide additional support to the main bearing flange. In order to get this extra 3mm the old single row counterbalancer bearing is used and a lighter counterbalancer weight installed.


The choice is up to you all.

All the best,
Simon

Click to expand...

Hallo.
Regarding pos5, we have some info on this due to the works/factory team have used this for almost 2 years, but some riders do not like the increase of gyro effect.
we used this as well, 20mm roller bearing with a spacer ring 7.8mm instead of the balancer, only one of our riders liked it, the rest fellt the balancer was the best.
We had no failure on the bearing.
Regards
Thomas

had a call today to say that chris has done summink hard to comprehend!

HE HAS SNAPPED THE CRANK IN THE SAME PLACE AT THE SAME TRACK A YEAR TO THE DAY LATER!

just 24 hours on it. he bought it as a complete assembly with con rod / big end etc all ready to go!

i went over at 5pm and picked up the 4 (bits of) cranks. plus his cases. the bike is an '04 450FE and he got 120 hours out of the first crank and just under 24 out of this one! the difference this year is that the ball-race mains that he converted to a year ago are in perfect order and so is the c/b. the old assembly is toast with the mains stuck on the crank and the c/b sooted up from running by the look of it without oil!

how many snapped cranks have we had? none i thought!

anyway, the crank has snapped both tims across an oil drilling in the actual big end. the drilled hole is pressed into the flywheel cheek to sit over a drilled hole in the crank therefore allowing oil into the middle chamber before going into the big end.

as far as swarf on the drain magnet/ virtually none! although the con rod had rubbed hard on one cheek and there was a silver metalflake look inside the b/e? piston and barrel were excellent.

i'm off to my engineers tomorrow and will have an arbor made, measurements taken etc etc.

as usual i'm amazed where all the members go when their input is needed. finnberg-blah blah!!!! (i could go on....).

regards

Taffy

Would love to see a pic of this one. Having a hard time following the description explaining the exact location. Should have stayed in a Holiday inn.
dan

the big end pin disappears into each flywheel yes? well the end that goes in the left flywheel has a drilling in it and it snapped across there. it therefore snapped the big end pin but just a few mm into the left flywheel.

the odd thing about the drilling is that it is about 3mm wide and stops before it actually reaches the hollow chamber of the big end. it's purpose therefore isn't absolutely clear!

the con rod has been rubbing hard against one cheek and you can see the circular inprint of the con rod sideface.

tomas (engine hardwear) have you checked that the con rod aligns with the middle of the bore?

regards

Taffy

Taffy said:

the big end pin disappears into each flywheel yes? well the end that goes in the left flywheel has a drilling in it and it snapped across there. it therefore snapped the big end pin but just a few mm into the left flywheel.

regards

Taffy

Click to expand...

Taffy,
you are saying it snapped through the big end pins lubrication drilling yes??

If so, its not a one of.

Try a search- I know its occured before and has been discussed on this site.