Reocurring gearbox bearing failure


XS650 Member
Reaction score
Bristol, England
Hi all, I would really appreciate any of your thoughts on why my XS650 has twice broken the double row ball bearing that is on the gearbox input shaft, behind the clutch basket.

First some history: I bought this bike in 1993 with about 20 000 miles on the clock, (it was registered in 1981 and had 6 former owners).

In 1995 I discovered some broken clutch shock absorber springs. I had Halco Tuning fit their conversion (3 new springs and 3 plastic spacers)

In 1997, now with 34 000 miles the engine was vibrating so badly that I had the engine overhauled and this including the crank being overhauled and was re-bored to 2nd oversize. A year later I found that one of pillars of the clutch boss was cracked and replaced just this part of the clutch.

In 1998 I fitted a Boyer Branson electronic ignition system, the type that also does away with the auto advance mechanism.

In 2003 (48 500 miles) while crossing France the engine nipped up. Once the bike was recovered back to England, my local XS650 specialist (LPS in Wiltshire) stripped the engine and found that the gearbox input shaft bearing’s outer race had broken up and some fragments has circulated around the engine and damaged the bores. At this time they were unable to obtain a new replacement for this double ball bearing and so they used a carefully examined second hand one when they rebuilt the engine again.

This time it was over bored to accept modified J.E pistons to take it to about 710cc, 10:1 CR and kawasaki GPZ500S carburettors (EX500 in USA) were fitted. I had the carb jetting set up on a rolling road dyno and showed 53bHp at the rear wheel but would only run on one cylinder at tick over, it was very tricky to start and would kick back. (The electric start and its gears had been removed during this last rebuild). I spent a lot of time trying to sort these second hand carbs out and looking for air leaks and dealing with some electrical issues.

Eventually I bought another second hand set of GPZ500S carbs and swapped the jets and needle shims. Now it started easily and ticked over well but only made 23 bHp.

I was riding the bike, trying to sort out the carb issues until, with about 55 000 miles the clutch behaviour felt wrong (I can’t remember the detail now) so I took the clutch off and found the bearing behind it had cracks in its outer race. Fortunately all the bits were still held in place but I was very concerned that this had happened again!

Other things probably not connected to this problem but I will mention it it case it:

Sometime after the second engine rebuild I noticed that with the engine running and in gear and the clutch pulled in, I could only paddle the bike backwards a few inches before something locked up and stopped it going back further. If I then turned the ignition off it rolled back easily.

Also, I was revving the engine to the red line while riding, and later was told by the operator of the rolling road dyno that the bike’s tacho was very slow to rise above 6 000 rpm. I wonder if I was actually going way past the red line and this could have caused harm.

At this time I put the bike to one side, wondering what to do until I was told about someone who modified XS650 engines for sidecar motor-cross and I took the engine to him for examination. He told me that none of his many 880cc (I think) conversion race engines had ever had a problem with this bearing. He spent some hours looking for an underlying problem but didn’t find one, so fitted all new bearings to both gearbox shafts (the double race bearing had now become available again) and bolted the crank cases back together.

I have recently put the engine back in the frame but haven’t fitted carbs yet. I would have liked an explanation for this repeated failure. On one hand I feel that if it fails a third time I will be kicking myself and asking ‘What did I realistically expect!’, but I would also like to have the bike back on the road this summer.

On line searches have not found this to be a common problem, has anyone else experienced it?

Does anyone have any suggestions?

Yes sir, the first time I ever heard of that bearing failing was in your introductory thread. I've been around XS's, in some capacity, since 1972. I do have some thoughts about the clutch basket and also the main shaft sideplay; I'd like to defer comment until someone who perhaps has experienced the issue posts findings.
Taking this out of memory and have not read through.
But some of the bearings was not available for a long time ( I notice now the same here )
If i recall right some of the crank bearings . There was a hole in them for lubricating i Believe.
The same bearing dimension but minus the lubricating hole was available used for some agricultural equipment.

I bought on e .bay and payed a lot for them Yamaha parts

So the quetion is .As always what bearings have been used twice here If it is not Stock

This is a tuned engine over bored . This is something one should not do in my Opinion
The life span has a finite number of load changes " Fatigue " And it can take fewer of these load changes if those loads are higher.
So over the lifetime there is a accumulated load history.
And loading heavily at the end .Makes it break sooner in time.

So assuming the shop really installed a Factory used bearing It had seen use. Already used up some of its life
Into service and now for Higher loads and maybe ( In some cases back in the day ) loads the bearing was not designed for.

Remember one
Big job on a Honda four Bought cheap --1000 miles --- Kabong Kaputt

Can of course be other things .But My view is if you load heavily and more than the Japanese designer had reason to design for . If you do that on a brand new vehicle the warranty can be devoid.

Not read through but 35 --55 000 miles one cannot be surprised if things break.

To do that used a carefully examined second hand one when they rebuilt the engine again.

That is something a Motorcycle repair shop rarely have the Knowledge to do. Not even sure if they can split it to do it.
Brief recap of pertinent history:

'93 – 20,000

’95 – ? clutch damper spring failure, repaired with solid “plastic” (vesconite?)/spring combo

’97 – 34,000 crankshaft vibration, rebuilt & top-end rebuild, .50mm over

’98 - ? clutch boss cracked

’03 – 48,500 mainshaft bearing outer race cracked & fragmented and replaced (used) and BBK 53bhp

? 23bhp and strange rolling backward behavior

? – 55.000 mainshaft bearing outer race cracked and replaced (new)

As no one has piped-up with first hand experience, I’ll offer some commentary and speculation.

First, I suspect the last rebuild may have fixed most, if not all, the issues; as he would have had the transmission out (for new bearings) and should have checked and set side-play for the gears and shafts to spec. One of the possible causes of the previous issues may have been incorrect side-play, particularly on the main shaft (~.003”). Also, as he knew the issue has repeated and is unusual, he may have checked runout on the main shaft (<.001”). He probably checked for other unusual stuff – beyond normal inspections – like the journal to bearing fit in the case and perhaps lash between primary driven and crank gears.

It is noteworthy – to me – that you did not mention any shifting issues during the previous clutch and bearing issues. Also noteworthy is the number of miles accumulated between various failures. Typically, if a part is defective or assembled incorrectly it shows-up in a few hundred miles (or less), not several thousand. So, some of the concerns that I may otherwise have had are remote.

One concern that I would still have is the damper assembly on the clutch basket. I’m not personally familiar with the thermoplastic inserts installed in ’95, but would note that they are not widely used nowadays and being 28yrs old now, I’d suspect they may need replaced. Excessive driveline shock loads won’t do the transmission bearings any favor. I know the UK market is tight, but used clutch baskets are available in the US for ~$100. Heiden Tuning also sells a new aftermarket basket for ~$300.

My wild-ass guess is that the original bearing failed as a result of debris (perhaps previous clutch damper bits). The second failure perhaps a combination of factors – it sounds as there were other issues with that rebuild: rolling backward issue. The “lock-up while rolling backward” is an uncommon, but heard of, issue that has been attributed to the kickstart gear misassembled or perhaps the countershaft (1st). The only thing that changes - running vs not running – regarding the transmission would be oil pressure at the shafts’ needle bearings. Anyway, one first good test of the new rebuild would be to see if the rolling backward issue remains (doubtful).
I believe it's possible to install the clutch shaft bushing incorrectly where it partially or completely blocks oil flow through the countershaft. Not sure what other indications you'd see, but it would definitely starve the roller on the clutch side.

but it would definitely starve the roller on the clutch side.
Ah, yep; that's another condition not mentioned: galling or blueing from lack of lube or excess heat. Just noted "broken" (1st occurrence) and then cracked (2nd occurrence).
Thanks for all your replies, they are appreciated.
I have replaced the bush that the clutch push rod passes through near the gearbox output sprocket. Is that what you were referring to Jim?
I will do some reading up on clutch basket shock absorber mechanisms. I thought I had read that just using 6 more springs allowed them to become coil bound and so to damaging themselves.

It’s reassuringly to hear that this isn’t a common problem,

Thanks again
the bush that the clutch push rod passes through near the gearbox output sprocket. Is that what you were referring to Jim?