Birth of a kit

[jonashford]

Old shit breaks, Always amazes me when someones lambretta part fails and they say "Dunno why that happened, It was an original part!?. yes its nearly 50 years old!!! Throw that crap out and buy a new Tino, Mb or other such new part, with 4 8 or 12 studs on it!!! but dont put old crap on ur scooter if u want to ride it fast and far.

[Andy B.L.C.]

innit...

[dickie]

I agree with Warkton. The failures in the pictures posted by strummer and nudger are in a section which is not subjected to drive torque. Drive torque is transmitted via the splines, not the nut.

My conjecture is that the shaft has failed due to fatigue induced by the ability of the hub to rock slightly on the shaft and therefore 'bending' the nut up/down and forward/aft on the shaft.

I'd really like to get a look at a variety of close up shots of the failed shafts to try to determine the failure mode.

[HxPaul]

I had a 1948 Harley-Davidson panhead and I was riding it one day when it locked up and threw me off.When I stripped it down I found that one of the teeth on the second gear had sheared and locked the gearbox.I was told by an engineer that it was because it was so old that the gear had become hard/brittle through use and that had caused it to break.I'm no metalurgist so I took this as being correct.The bike at that time was 30+ years old,we are riding around on vehicles that 50 years old,some are even older,so if the engineer was right about age and metal getting harder through use and age,its no wonder that some of the original layshafts break through the constant torquing and unfastening of the rear hub.

[dickie]

I had a 1948 Harley-Davidson panhead and I was riding it one day when it locked up and threw me off.When I stripped it down I found that one of the teeth on the second gear had sheared and locked the gearbox.I was told by an engineer that it was because it was so old that the gear had become hard/brittle through use and that had caused it to break.I'm no metalurgist so I took this as being correct.The bike at that time was 30+ years old,we are riding around on vehicles that 50 years old,some are even older,so if the engineer was right about age and metal getting harder through use and age,its no wonder that some of the original layshafts break through the constant torquing and unfastening of the rear hub.

He's sort of right but steel doesn't harden simply due to the passage of time. Rather what happens during that time adds up to cause failure usually either through fatigue or crack propagation. I'm assuming here that it's not due to corrosion.

Innocenti's advice that a small amount of play in the hub is acceptable indicates to me that the hub/shaft design isn't exactly great. And I suspect is a significant contributing factor.

[Strummer10]

I agree with Warkton. The failures in the pictures posted by strummer and nudger are in a section which is not subjected to drive torque. Drive torque is transmitted via the splines, not the nut.

My conjecture is that the shaft has failed due to fatigue induced by the ability of the hub to rock slightly on the shaft and therefore 'bending' the nut up/down and forward/aft on the shaft.

I'd really like to get a look at a variety of close up shots of the failed shafts to try to determine the failure mode.

Thanks Dickie for your thoughts, if useful I will try and source the second layshaft I mention and post pictures.

Just going to 'float this one', but is it conceivable that a new rear wheel bearing could be a potential cause, IF it was perhaps not locking tight against the aluminum engine case housing from install.........thus moves within the aluminium engine casing as the layshaft turns ?

[dickie]

Just going to 'float this one', but is it conceivable that a new rear wheel bearing could be a potential cause, IF it was perhaps not locking tight against the aluminum engine case housing from install.........thus moves within the aluminium engine casing as the layshaft turns ?

That would induce a bending moment at the other end of the shaft in the gearbox end plate with the outcome being either end plate failure or the other end of the layshaft failing, so no, I don't think that's a factor.

Don't take that as a put-down by the way. I'm a big fan of thinking out loud. And often talk utter bollocks to stimulate thought at work.

[Strummer10]

Just going to 'float this one', but is it conceivable that a new rear wheel bearing could be a potential cause, IF it was perhaps not locking tight against the aluminum engine case housing from install.........thus moves within the aluminium engine casing as the layshaft turns ?

That would induce a bending moment at the other end of the shaft in the gearbox end plate with the outcome being either end plate failure or the other end of the layshaft failing, so no, I don't think that's a factor.

Don't take that as a put-down by the way. I'm a big fan of thinking out loud. And often talk utter bollocks to stimulate thought at work.

Cheers Dickie and don't worry about the 'put-down' thing .......honest open and intellectual debate is always useful to 'flush out' things and i'm always open to learning from this forum and through practical experience.......

The reason I threw in the loose bearing, was that this was felt to be contributory in the issue with the photo in my post above, but as you say this might have more of an effect gearbox side. Good point also on endplates coming loose...btw.

I have a keen interest in scooters and have been re-building Lambretta engines since 1979 so have a reasonable degree of experience and have built 5 engines this year, but still don't know everything of course and surprisingly am still finding new issues coming up. Thanks to 'Stickys' book I have learnt the benefits of using a 'torque wrench' 'loctite' and now have the proper issued tools for the job......a far cry away from the 1980 rebuilds .

My original post in response to Stickys thread was just to see whether the new engine cases had considered making the new layshaft any larger across its section (whether this was fruitful or not, I don't know) and if there was a way to keep the back wheel / hub harnessed against the engine when it did fail to provide a secondary safety mechanism. Also pointing out many are using 50 year old layshafts and not replacing them and with higher BHP this might be a problem.

This was mainly geared by hearing the news two scooterists we knew well had layshafts fail recently and from the responses above the issue seems to perhaps be larger than first felt (not sure).

Ultimately I was after someone (SLUK / Scootering) doing some form of technical test and producing results, which would interest me as I like the 'techy stuff' to demonstrate 'pros and cons' of a 50 year old layshaft versus the new ones on the market just to provide us scooterists with the facts and let us make an informed decision whether to run off and replace our 50 year old layshafts if we had concern.

[Warkton Tornado No.1]

Post by dickie » Sat Nov 05, 2016 11:13 am

Innocenti's advice that a small amount of play in the hub is acceptable indicates to me that the hub/shaft design isn't exactly great. And I suspect is a significant contributing factor.

I’m not contradicting you, but wonder if I am mistaken as my interpretation of Innocenti’s advice would be the inevitable clearance in the rear hub bearing & the end plate bearing would amount to a little perceptible rock, which is acceptable.

I don’t think that there should be any play whatsoever within the taper fit of the hub to the cone as it should result in a transition fit if the cone surfaces match & the tightening torque are to the correct values.

You’ve reminded me, though, that I’ve voiced concerns about this area before on this Forum....

In that scenario, of the OEM hub/cone/OEM lay-shaft interface, the part that worries me is the remade cone itself that doesn’t seem to be hard enough in the bore.

I have only purchased the Serveta type from Scooter Restorations as the company has a good reputation. Yet, if the hub is fitted to the (lapped in) cone & removed after negligible miles, the bore of the cone will have strong witness marks of the lays-haft spline. This hasn’t been a one off, either, as I’ve noticed it three or four times...

Inevitably, the result is the loss of the actual taper fit & some of the tightening torque from where it should be, if you understand?

[dickie]

OK Warkton. I see what you mean. I think you're interpretation is correct, not mine.

But I think we need to separate the issues. One being security of the hub against the cone and the other being the bending moment in the threaded section. I'm cautious about being blinkered in my view here as I've only seen 2 pictures of failures. But from what I've seen, they have failed due to fatigue (repeated bending) or sheared (torque). Again, like Warkton I'm very sceptical about the latter as there should be diddly-squat torque going through that portion of the layshaft.

[Warkton Tornado No.1]

Post by dickie » Sat Nov 05, 2016 1:19 pm

OK Warkton. I see what you mean. I think you're interpretation is correct, not mine.

But I think we need to separate the issues. One being security of the hub against the cone and the other being the bending moment in the threaded section. I'm cautious about being blinkered in my view here as I've only seen 2 pictures of failures. But from what I've seen, they have failed due to fatigue (repeated bending) or sheared (torque). Again, like Warkton I'm very sceptical about the latter as there should be diddly-squat torque going through that portion of the layshaft.

I agree with you.

To back up my belief that the OEM lay-shafts are more than ‘fit for purpose’ I have earlier tried to justify my feeling with some engineering that might be understood.

Although it’s indisputable that these things will ‘age’, it will be hardly @ all! Certainly not to the extent that some suggest if treated appropriately.

There’s the rub! There appears to be a great deal of dismissing OEM lay-shafts simply because of an eagerness to buy into a replacement that, unless I’m mistaken, has not gone through anything like the rigorous standards of manufacture that the OEM products will have! Where’s the certification, accreditation, call it what you will, for these ‘improved’ lay-shafts? Are they TUV approved for road use, I wonder.

For the moment, it’s gone deafeningly quiet as far as backing up the assumption that the ‘re-made’ lay-shafts are better with some form of proof.

Where there would appear to be some common ground within this debate is in that we can’t allow for the human element of abused, badly fitted, & even corroded lay-shafts if we don’t know the heritage.

But that is hardly the fault of the OEM lay-shaft....

[ToBoldlyGo]

We must assume that SIL layshafts, at least, benefit directly from Innocenti development.

[Warkton Tornado No.1]

Post by ToBoldlyGo » Sat Nov 05, 2016 3:04 pm

We must assume that SIL layshafts, at least, benefit directly from Innocenti development.

I would say definitely, if they are genuine SIL, they are likely to be as good as Innocenti.

It may have escaped some people, but the SIL crankshafts were considered rubbish by many, years ago. Now, despite the lack of the bling factor, people are realising that the metallurgy & pin fits are good if not better than some aftermarket crankshafts produced by respected tuning houses. Okeh, the bearings are not great quality & are best swapped, but I'd bet there's many thousands that have served a hard life in India.

[Nudger]

I finally had the chance to give my lay-shaft to an engineer who works in our development centre.
Although we're a medical device company, I value his knowledge & was interested to hear his opinion & what he had to say:

Products fail due to many different reasons, sometimes due to a poor design or because they are used outside of their intended purpose.
Types of failures generally include one or more of the following:
Fatigue
Fretting
Erosion
Stress corrosion cracking
Corrosion & corrosion fatigue
Cavitation
'Creep'
Hydrogen embrittlement
Galling

Looking at my lay-shaft he believed fatigue fracture was the cause and not a design fault in itself. A fatigue fracture is not caused by single-load applications, but by the cumulative effect of a large number of load applications at stresses insufficient to cause fracture with one load application. This includes the possibility of tightening the rear hub nut to incorrect torque settings during the previous ownership of the scooter. The repetitive forces of incorrect load and torque, which may have reached into the thousands or millions of cycles prior to fracture, could have exerted a shearing or wracking action on the crystal structure that tends to cause the inevitable structural micro defects in the crystals to join together until a minute crack develops in certain vulnerable crystals.
The repetitive stressing automatically locates the crystals with the weakest orientation in the regions of highest stress. Continued cycling causes the crack, or cracks, to enlarge, gradually becoming deeper at an increasing rate of growth. As the crack depth increases, the strength of the remaining metal decreases.

[dickie]

I reckon your mate is correct. I also believe (unproven) that it's exacerbated by the distance between the centre-line and the nut, thus acting like a torque arm.

[Swat]

I echo & agree with those thought from your development engineer & those of 'Dickie'. I've been involved in Aerospace structural (full airframe fatigue)testing for a long time & bolt head loss due to fatigue was a regular occurrence on older design airframes such as buccaneer & phantom - bear in mind they were designed in the 1950's just like Lammy's & at that time the effects of metal fatigue & influences such as sharp radii, & section changes weren't widely understood.

looks like classic fatigue growth across the section, to a point were it's failed statically = likely exacerbated by slight hub rock as Dickie described.

[holty]

i have read quite a few posts on layshaft failure recently, with lots of theorys and explainations, all may be possible, surely the question is how to prevent layshaft failure, when you look at the engine case from above it points in the same direction as the frame, the layshaft at 90 degrees the the engine, then the wheel paralell to the engine, this is like three sides of a square, if you were to add a forth side this would make it much more stable and put much less stress onto the layshaft in the first place, i m talking about an arm with a bearing in it that fastens to the rear drum, probably using the 3 threaded hub pulling holes, that then fastens to the engine, maybe around the engine mounting(gp type) i have seen this type of arm used on high powered autos to stop them snapping the engine case in half, with the search for more and more hp this is becoming a more serious problem.

holty

[Strummer10]

i have read quite a few posts on layshaft failure recently, with lots of theorys and explainations, all may be possible, surely the question is how to prevent layshaft failure, when you look at the engine case from above it points in the same direction as the frame, the layshaft at 90 degrees the the engine, then the wheel paralell to the engine, this is like three sides of a square, if you were to add a forth side this would make it much more stable and put much less stress onto the layshaft in the first place, i m talking about an arm with a bearing in it that fastens to the rear drum, probably using the 3 threaded hub pulling holes, that then fastens to the engine, maybe around the engine mounting(gp type) i have seen this type of arm used on high powered autos to stop them snapping the engine case in half, with the search for more and more hp this is becoming a more serious problem.

holty

Holty good response, Great idea would certainly assist the issue if the hub nut comes away with maybe something beyond the back wheel additionally tied into the rear frame someway.......acting like a 'lateral restraint system' to harness the back wheel to the engine/frame.

[Warkton Tornado No.1]

The basic design of the rear hub supported on the lay-shaft is the cantilever principle as used on V*spa front & rear hubs as well as virtually all cars.

I have no doubt that adding a supplementary support would help but whilst there are other problems that have been indicated by some really well informed responses given recently; I think adding a supplementary support is only masking the problems.

The reaction given by Nudger is not based on my gut reaction, or anybody else’s, for that matter. He has enabled a logical, methodical approach using known techniques that enable us to make sound judgements.

For now, to achieve ultimate ‘peace of mind’ the best, most practical solution may be a non destructive test of any lay-shaft likely to be used, including any ‘super-duper-bling-looking-yet-not-certified’ replacement lay-shafts that some would have us queuing up to buy!

[Scooterlam]

So is this kit related to the one recently run up on JBs Dyno.
Sorry if this post is ON Topic.