PDA

View Full Version : wear on suspension plates and bearings



theknurl
11-18-2007, 10:54 PM
i just removed suspension plates;
the length of pins through dogbone and width of shock eye 36mm
width of dogbone 32mm
4mm of float! so why did the plates move to the left and grind 0.16mm off the right plate in the narrow section at the swingarm mount?
because something is crooked!

the pin in the rear of the dogbone shows signs of lack of lube, pin is turning brown, with flat spots from the needles, like in unlubable u-joints.
needs new needle bearings and pin and a damn grease nipple [Zerk fitting]
pin in swingarm mount is perfect.

revised dimentions of failing area; 8.9 x 5.9mm with surface roughness
so its actually smaller.
isn't bead blasted, its sand blasted VERY BAD FOR FATIGUE LIFE, creates stress risers everywhere.
weight of part 53 grams, identicle part in steel 153 grams

ENGINEERING??? my ass, this is like leaving the conrod thrust washers out, it always trashes something....usually the whole motor.
our problem is that link plate failure can be FATAL.....
Noel-theknurl

MDRSV
11-18-2007, 11:10 PM
the pin in the rear of the dogbone shows signs of lack of lube

ANY bike (japan or Italian) you take apart will have very little lube in the dogbones or steering head as it comes from the factory. Thats just a fact of life.

Italijet
11-19-2007, 04:01 AM
our problem is that link plate failure can be FATAL.....
Noel-theknurl

I aint seen anyone come in here and say that it has actualy killed them yet. Those that have reported it said it "could have" or "nearly did", but no one has said it actually did....

Quite frankly, you don't want too much grease in the dog bone bearings. It's in the direct line of the rear tyre and a prime place to colect all of the grit and carbon that the tyre pulls off the road. Mixing grit, carbon and grease in one place just makes a quite effective cutting compound, and causes more damage than if the part is regularly cleaned.

dimdoc
11-19-2007, 05:57 AM
can we see pictures of that Noel?

Thanks

Roadless
11-19-2007, 06:07 AM
I aint seen anyone come in here and say that it has actualy killed them yet. Those that have reported it said it "could have" or "nearly did", but no one has said it actually did....

Quite frankly, you don't want too much grease in the dog bone bearings. It's in the direct line of the rear tyre and a prime place to colect all of the grit and carbon that the tyre pulls off the road. Mixing grit, carbon and grease in one place just makes a quite effective cutting compound, and causes more damage than if the part is regularly cleaned.

From where should they post, those that "got killed" by the failing links?
Just asking, or should we wait patiently untill someone posts " the link broke, and now I have a harp "

spoonz
11-19-2007, 06:20 AM
My conrod bearings were very short on grease and were also full of water. I don't jetwash the bike either. I repacked it with water resistant lithium grease.
I don't think you can actually over grease them as any excess just oozes out when you install the pin.

I also had the wear on the plates always on the left side. The dogbone always migrated to that point after i centered it. There were 4 pins missing from the bearings in total which didn't help. Not sure if that was the dealer during the swingarm change or from the factory.

spoonz
11-19-2007, 06:33 AM
Out of interest does anyone know why they made the dogbone float as they did.
It can't be by accident so there must be an engineering logic there somewhere. Stiction or something ?

Italijet
11-19-2007, 07:38 AM
My conrod bearings were very short on grease and were also full of water.

Spoonz, you will having the tech savvy Americans asking if you have a fully blown engine that needs FI water cooling :eek:

They know it as the dogbone, connectong rods are for their pistons...

:happy:

Im thinking of using a rubber washer to prevent ingress of road grime and allow movement of the dog bone. I beleieve that there is play there to allow movement when the swing arm flexes during cornering.

millemille
11-19-2007, 07:46 AM
I'm pretty sure the end float is designed to overcome the relative misalignment between the swingarm pivot points in the frame, the suspension linkage mounts on the frame and the suspension linkage mounts on the swingarm.

If you visualise the suspension linkage in motion when the suspension compresses and factor in one or more of the - ideally perpendicular to the axis of suspension compression - pivot points being even a few seconds of a degree out of alignment (which would be within perfectly acceptable manufacturing tolerances of the frame or swingarm) then the linkage side plates need to float laterally to acomodate this.

IMO, understanding this is key to understanding the side plate failures. None of the normally applied ISO statisticall analysis (which lest we forget is what is keeping cars on the road, trains on the tracks, planes in the sky and ships at sea) criteria support this being a design issue of the side plates in normal - or expected - operation. The failures were not on highest mileage vehicles, heaviest owners, roughest roads etc and the failure population (even if factored at the extreme end of pessimism regarding number of actual failures over number of failures there "may" have been and take the lowest estimate of population size and duty cycle) as a %age of the available total population is too low to be considered a design or manufacturing issue relating to the side plates when exposed to "normal" loading within the design envelope parameters.

I'd like to see what results you'd get if you undertook Faroe measuring (or any similar x/y/z accurate measurments) of the frame and swingarm pivot points in relation to each other as the suspension compresses.

spoonz
11-19-2007, 08:10 AM
Spoonz, you will having the tech savvy Americans asking if you have a fully blown engine that needs FI water cooling :eek:

They know it as the dogbone, connectong rods are for their pistons...

:happy:

Im thinking of using a rubber washer to prevent ingress of road grime and allow movement of the dog bone. I beleieve that there is play there to allow movement when the swing arm flexes during cornering.

Actually i know it as a dogbone as well but i had just read a post on another forum abount con rods and it just came out.

MilleMille did i interpret you correctly that you think the linkage failure might be as a result of a combination of factors over stressing an otherwise sound linkage plate rather than a design oversight with the link itself.

I know Scott ran a stiffer rear spring as do I. I wonder if the other failures did ?

millemille
11-19-2007, 08:23 AM
It's my WAG that you could (and Aprilia more likely than not did this as part of the R&D process) put any of the models with the suspect link side plates onto a road impact replecating shock dyno - the ones I've seen are just a driven revolving drum with an ellipitical profile under the rear wheel - with any stiffness of rear spring and rider weight and subject the linkage to a duty cycle equivalent to tens of years of riding the roughtest roads known to man and the linkage components wouldn't fail....

...as long as the axes of the major pivot/anchor points were all within a certain tolerance range of each other...

..if the pivot/anchor points were misaligned by a few seconds of a degree over the acceptable tolerance range - either through manufacture of through damage post build - then the side plates are subject to a whole load of new forces that start loading the side plates in a lateral and/or torsional manner..

..and that's when plates could well be failing.

That's why I asked scott in the main thread on the subject whether his bike had ever been crashed in the past which might have caused a misalignment.

I still believe that placing anything in the shock linkage to limit the lateral freeplay - as some members have done - is potentially very risky.

Italijet
11-19-2007, 08:46 AM
I'd like to see what results you'd get if you undertook Faroe measuring (or any similar x/y/z accurate measurments) of the frame and swingarm pivot points in relation to each other as the suspension compresses.

Faroe Measuring, as in Egyptian Ruler of People or the islands ? :happy:

Triangulation of reference points in relation to each other will be an endless task as the direction that the forces through the wheel and subframe take, will be in relation to the angle of the vehicle in relation to vertical pull of gravity and the forces inparted through the weight.

Each and every possible angle will need to be tried and tested. Computational modeling would make it possible, although phyical measurements should also be taken to prove the computer model.

spoonz
11-19-2007, 09:00 AM
Faroe Measuring, as in Egyptian Ruler of People or the islands ? :happy:

Triangulation of reference points in relation to each other will be an endless task as the direction that the forces through the wheel and subframe take, will be in relation to the angle of the vehicle in relation to vertical pull of gravity and the forces inparted through the weight.

Each and every possible angle will need to be tried and tested. Computational modeling would make it possible, although phyical measurements should also be taken to prove the computer model.


errr all very technical then.

So takeing everything into account that we know is a stronger linkage plate likely to solve the problem ?
If not i might start praying to that Egyptian god like ruler before each ride.:worship:

Italijet
11-19-2007, 09:58 AM
errr all very technical then.

So takeing everything into account that we know is a stronger linkage plate likely to solve the problem ?
If not i might start praying to that Egyptian god like ruler before each ride.:worship:

That raelly depends on the nature of the failure. From what I see in the pictures, the failure looks to be a result of metal shearing under stress. However the pictures are not clear enough to show whether the metal was torn (leaving crisp shiney alu graining) or if the shear point is dull indicating a fissue or fracture was present in the original ingot.

From the look of the plates I had suspected that they were cast, however it's been said here that they are punched from rolled plate and then shot peened. If this is correct, with all of the pictures showing the failure happens towards to the eyelet for the shock, it is unlikely to be due to a problem with the original material (except possible torsional strength).

As millemile indicated, it would be good to see the Poisson measurement characteristics (lattitudinal and longitudinal stress calculations) of these components. Without actually having physical access to the broken parts, there is little that can be understood by the nature of the failure.

However, it hasn't stopped me riding (the weather this week may) I just keep them clean and apply a coat of light oil.

theknurl
11-19-2007, 05:45 PM
the 0.16mm deep wear mark was caused by the swingarm rasping on the plate the groove is 24mm from the center of the bolt through the swingarm and plate.
the plates have failed in tension... from the pictures right about where the groove is on my right plate
the dealer had just replaced the recalled swingarm so can't see the wear mark on the swing arm
given the size of the swingarm section i doubt there is any flexing there
as millemille said a few seconds of misalignment will cause things to move around, the only thing stopping it is the seal friction laterally.
the problem is the wear on the plate creates a stress riser and weak spot in the plate. when its strength is exceeded it lets go...followed immediately by the undamaged plate.
replacing the plates with stronger ones is a must but the swingarm shouldn't be grinding on them either
Noel-theknurl

duc slayer
11-19-2007, 05:51 PM
did you post pictures or pictures from the previous thread ??

ChicagoFactory
11-19-2007, 06:21 PM
Just asking, or should we wait patiently untill someone posts " the link broke, and now I have a harp "


That's f*ckin funny Roadless. :funnypost

theknurl
11-19-2007, 11:06 PM
duc slayer;
sorry, HP isn't talking to Motorola....or Epson either
i'm still in the world of Hasselblads and Leicas
i'm an old fart sticking to things that i trust.
pictures wouldn't show you the depth of the abrasions, hopefully the dimensions i gave will show you where to look
Noel-theknurl

ckruzel
11-20-2007, 12:45 AM
anyone read up on the 636 front forks that break apart under hard breaking and stoppies

ckruzel
11-20-2007, 12:46 AM
my race cr's used to have wear on the dogbone and links, i used to replace them regularly

millemille
11-20-2007, 04:12 AM
Italjet - FARO measuring... http://www.faro.com/content.aspx?ct=uk&content=ind&item=20&tab=0 ... As part of a project team from the engineering consultancy that I work for we measured the dimensional accuracy of 160 rail vehicle bogies and the FEA modelled the results - including all service loadings to capture extremes of dynamic distortion - to understand the lateral loadings on axle bearings - all on a laptop and in 3 weeks.

You are going one step too far as an intial investigation, there's no need to consider flex of the frame and swingarm. The pivot/anchor points are close enough together and there are more flexible elements of the structures to absorb loadings first. Consider if the left side swing arm mounting point in the frame is at the upper rear most range of it's tolerance and the right hand is at lower forward position.. your pivot point induces the swingarm to not just have a vertical component to it's movement but also a horizontal one. Repeat the same alignment scenario with the suspension linkage anchor points on the frame and the swingarm and also the relationship between the swingarm pivot on the swing arm itself and the suspension linkage anchor point on the underside of the swingarm. If all of those have slightly conflicting lateral arcs of movement then the side plates flexing and the linkage end float are all that can accomodate this...

This might explain why the linkage failures appear not to fit a statistical norm for duty cycle variables vs. failure rate.. it's not the use they have but the "condition" of the suspension geometry..

The Dr.Jon
11-20-2007, 04:15 AM
Italjet - FARO measuring... http://www.faro.com/content.aspx?ct=uk&content=ind&item=20&tab=0 ... As part of a project team from the engineering consultancy that I work for we measured the dimensional accuracy of 160 rail vehicle bogies and the FEA modelled the results - including all service loadings to capture extremes of dynamic distortion - to understand the lateral loadings on axle bearings - all on a laptop and in 3 weeks.

You are going one step too far as an intial investigation, there's no need to consider flex of the frame and swingarm. The pivot/anchor points are close enough together and there are more flexible elements of the structures to absorb loadings first. Consider if the left side swing arm mounting point in the frame is at the upper rear most range of it's tolerance and the right hand is at lower forward position.. your pivot point induces the swingarm to not just have a vertical component to it's movement but also a horizontal one. Repeat the same alignment scenario with the suspension linkage anchor points on the frame and the swingarm and also the relationship between the swingarm pivot on the swing arm itself and the suspension linkage anchor point on the underside of the swingarm. If all of those have slightly conflicting arcs of travel in a laterall dimension then the side plates flexing and the linkage end float are all that can accomodate this...

This might explain why the linkage failures appear not to fit a statistical norm for duty cycle variables vs. failure rate.. it's not the use they have but the "condition" of the suspension geometry..

Millemille: If someone can get you some link plates are you able to test them to breaking point on a rig at you place of work? :confused: All for free of course.....pro (dog) bono.:)

ARAIHEAD
11-20-2007, 04:23 AM
DJ,sounds like you're concerned about these, dont worry, yours will be getting tested to the extreme with your git lardy arse bouncing on them!:eek::happy:

millemille
11-20-2007, 05:10 AM
Millemille: If someone can get you some link plates are you able to test them to breaking point on a rig at you place of work? :confused: All for free of course.....pro (dog) bono.:)

Dr.Jon.... testing the side plates in isolation is - IMO - not going to be that constructive. Whatever results you got would be meaningless unless you can relate them to the forces they are actually exposed to - the link plates might fail at (plucking a number out of the air) 3kN which may seem low, but if the system is only capable of exerting 1.5kN then there's no problem - and the only way you can understand these is by looking at the bike as a whole.

Noel's theory might be correct - the surface damage to the side plates from the sliding faces fretting might be the propagator of cracking.

The point I was -and am still - making is that there is no evidence to support the assertation that the side plates are not strong enough by design, material or manufacture... something external to the side plates opearting within an acceptable set of parameters (a combination of tolerances in suspension alignment or surface damage reaching a critical level, for example) is the catalyst for failure.

kzmille
11-20-2007, 08:16 AM
The point I was -and am still - making is that there is no evidence to support the assertation that the side plates are not strong enough by design, material or manufacture... something external to the side plates opearting within an acceptable set of parameters (a combination of tolerances in suspension alignment or surface damage reaching a critical level, for example) is the catalyst for failure.

There is also no evidence to support your theory that something other than a lack of an adequate amount of metal in the vertical element of the side plate is ultimately responsible for the known failures.

Italijet
11-20-2007, 09:34 AM
There is also no evidence to support your theory that something other than a lack of an adequate amount of metal in the vertical element of the side plate is ultimately responsible for the known failures.

IMHO, the number of failures would be considerably higher if it were.



You are going one step too far as an intial investigation, there's no need to consider flex of the frame and swingarm. The pivot/anchor points are close enough together and there are more flexible elements of the structures to absorb loadings first. Consider if the left side swing arm mounting point in the frame is at the upper rear most range of it's tolerance and the right hand is at lower forward position.. your pivot point induces the swingarm to not just have a vertical component to it's movement but also a horizontal one. Repeat the same alignment scenario with the suspension linkage anchor points on the frame and the swingarm and also the relationship between the swingarm pivot on the swing arm itself and the suspension linkage anchor point on the underside of the swingarm. If all of those have slightly conflicting lateral arcs of movement then the side plates flexing and the linkage end float are all that can accomodate this...

I hear what you are saying, but:
During cornering the forces exerted on the tyre, wheel and axial are lateral to the axel. This can cause the un braced swingarm to twist (hence previous cracking issue). This twisting results in force being placed on the pivot, which in turn causes the twin spar frame to distort. This minor distortion of both the Swingarm pivot and the frame results in the rear shock to cause torque on the linkages (offset by the movement in the mounts.

Too little play in the linkage or suspension movment and this torque exerts is's force on places that are not equipped to carry it.

kzmille
11-20-2007, 10:00 AM
I hear what you are saying, but:
During cornering the forces exerted on the tyre, wheel and axial are lateral to the axel. This can cause the un braced swingarm to twist (hence previous cracking issue). This twisting results in force being placed on the pivot, which in turn causes the twin spar frame to distort. This minor distortion of both the Swingarm pivot and the frame results in the rear shock to cause torque on the linkages (offset by the movement in the mounts.

Too little play in the linkage or suspension movment and this torque exerts is's force on places that are not equipped to carry it.

With all these cornering and axial and twisting and distortion and torquing forces it seems like it might have been prudent to make the plates a little thicker, wider, stronger, heavier beefier so they would be equipped to carry it. Yes indeed, sounds like a design flaw to me.

Roadless
11-20-2007, 10:11 AM
Being a tech myself, we sometimes make things overly complicated.
And then we try to explain things in techno to the normals...
And the normals get upset because they dont understand, and get upset if you simplify it for them.

In laymens terms, one link braking is a fluke.
Two links braking is cause for concern.
Three links braking is cause for new links for those that donīt want to get "scotted".

I personally belive in statistics and math, and those two have a way of catching up on you.
( if something happens one time in a thousand, how many times do you have to do it before it becomes certainty )

I work in places like steep roofs with heavy objects and lots of ice and snow.
I also go to strange foreign places, for work or pleasure, or both, rollerblading in baghdad was for instance fun.
BUT statistics catch up, and murphyīs law is an actual law of nature, so I never leave anything to chance, specially when the odd chance is disastrous.

I will have a new link in there before spring, I simply donīt want to become the 4th victim to somthing that "really canīt happen"

mike p
11-20-2007, 11:42 AM
did you post pictures or pictures from the previous thread ?? Duc, I think this it what the Nurl is talking about. When I replaced mine these marks were by the holes. This one is from the kick stand side the other plate showed no marks. The marks you see not by the holes were caused by the machinist when he was making me a new set. The one shot shows a little dent in the material just above the scrape mark which just barely catches my finger nail. In the other picture my finger nail will get caught in the scratch.Just a thought for us that have there swing replaced. When I went to take off one of the nuts it felt like it was torqued way more then the suggested 36 foot pounds so you might want to check yours for proper torque setting.04 factory 9500mi

http://i238.photobucket.com/albums/ff171/piemk/IMGP0619.jpg
http://i238.photobucket.com/albums/ff171/piemk/IMGP0618.jpg

spoonz
11-20-2007, 12:15 PM
Something Uli from MPL found was that the holes in the plates are bigger than the bolts by quite a bit. The link plates he made up had smaller holes because they felt there was potential for the bolt to move up and down on the plate with a hard impact.

Could this be part of the problem

theknurl
11-20-2007, 01:28 PM
the wear marks [gouges] in my right plate is 24 mm from the center of the bolt through the rear[swingarm] mount, amost in the middle of the narrow verticle piece.

the load on the part;
with a 130N/mm spring with 0.5 in preload
over a bump you bottom the suspension [40mm travel to the bumper on the Sachs] lets say 2 in of travel
thats a 742#/in spring compressed 2.5 in. Hooks Law 2.5x742= 1855 lbs
section at breaking point 5.9x8.9mmx2= 105 sq mm= 16% of a sq in
6061 T4 =14 ksi yield x 0.16 = 2278 #
6061 T6 =35 ksi yield x 0.16 = 5697 #
since Aprilia has shown in the past not to have used the best materials,
lets assume they used an equal to 6061 T4
22% safety factor..... NICE, not good enough for me
thats for a part with NO stress risers, unlike my grooved one
hole size 10.2mm, no complaint here
Noel-theknurl

spoonz
11-20-2007, 05:45 PM
the wear marks [gouges] in my right plate is 24
hole size 10.2mm, no complaint here
Noel-theknurl

It seems i misunderstood the description because of the German > english thing but when he said smaller holes for less play he actually meant the cut outs in the plates centre was smaller for less flex. Or at least i think that's what he meant.

I wish i listened more at school when I did langauges.

William the Third
11-20-2007, 10:16 PM
the wear marks [gouges] in my right plate is 24 mm from the center of the bolt through the rear[swingarm] mount, amost in the middle of the narrow verticle piece.

the load on the part;
with a 130N/mm spring with 0.5 in preload
over a bump you bottom the suspension [40mm travel to the bumper on the Sachs] lets say 2 in of travel
thats a 742#/in spring compressed 2.5 in. Hooks Law 2.5x742= 1855 lbs
section at breaking point 5.9x8.9mmx2= 105 sq mm= 16% of a sq in
6061 T4 =14 ksi yield x 0.16 = 2278 #
6061 T6 =35 ksi yield x 0.16 = 5697 #
since Aprilia has shown in the past not to have used the best materials,
lets assume they used an equal to 6061 T4
22% safety factor..... NICE, not good enough for me
thats for a part with NO stress risers, unlike my grooved one
hole size 10.2mm, no complaint here
Noel-theknurl
They are 2024-t6 according to this post.
http://www.apriliaforum.com/forums/showpost.php?p=1638755&postcount=501

ckruzel
11-20-2007, 10:35 PM
given the size of the swingarm section i doubt there is any flexing there


you'd be surprised to see how much a swingarm can flex, it has to.

millemille, so your saying leave the plates the oem thickness?
while i don't understand everything you are saying, your posts seem to be making the most sence

theknurl
11-20-2007, 11:12 PM
my 'standards for aluminium mill products' book shows;
sheet and plate;
2024-0
2024-T3
2024-T36
2024-T4
2024-T351
2024-T42
2024-T62
2024-T81
2024-T851
2024-T86
no T-6 though
it failed in tension ......its not strong enough, period
Aprilia screwed up when they made the legs of the plate narrower, to save weight
gee, folks, it can only maim or kill you .....don't whine.....fix it!

Noel-theknurl

Micah / AF1 Racing
11-20-2007, 11:43 PM
I have personally only seen two of these failures since the design was released in 2004. Both were on bikes that crashed into something causing extreme forces to be transmitted through the rear. Both failures were post crash, moment of impact type damage. One was a bike in 2004 that skidded of of a banked (positive camber) corner on the race track and came crashing to earth on the rear wheel which was bent along with the swingarm. The other was a bike that lowsided and slid into a rock (very large rock) striking the rear wheel first. Both bikes had damaged swingarms in addition to damaged wheels at the rear. The link plates broke as a result of the impact in both cases I have personally seen.

That said, since there is so much concern over this issue we at AF1/Acculign have designed and are in the construction phase with Mark at Acculign for replacement link plates that are at a bare minimum 125% stronger than stock while weighing no more and yes, are still made from a strong aluminum alloy. They will be 100% CNC machined and anodized and made from a strong alloy. They are fully radiused to eliminate the "stress risers" in the stock parts. They will use the stock bolts. They are identical in ratio dimensions to stock.

Price will be beyond reasonable. DO NOT PRE ORDER! DO NOT CALL AND ASK FOR DETAILS. THEY WILL BE ONLINE AS SOON AS YOU CAN ORDER THEM. The production run will be final in the next two to three weeks. I am on vacation till next Tuesday so please do not PM me about these either.

Mark and I and a couple of other metal savvy types here at the shop have performed our own analysis of the stock parts and do indeed believe they are strong enough stock to withstand anything the stock bike can throw at them but since this is a customer/enthusiast confidence issue we are attacking at full force. Can they be a hell of a lot stronger and weigh no more, hell yes. That is what we are getting ready to make available. Just having a fully machined surface with NO SHARP edges and made of a superior strength material will do the job times 2 at least!

farva03R
11-21-2007, 12:07 AM
you guys ROCK. :cheers:



I have personally only seen two of these failures since the design was released in 2004. Both were on bikes that crashed into something causing extreme forces to be transmitted through the rear. Both failures were post crash, moment of impact type damage. One was a bike in 2004 that skidded of of a banked (positive camber) corner on the race track and came crashing to earth on the rear wheel which was bent along with the swingarm. The other was a bike that lowsided and slid into a rock (very large rock) striking the rear wheel first. Both bikes had damaged swingarms in addition to damaged wheels at the rear. The link plates broke as a result of the impact in both cases I have personally seen.

That said, since there is so much concern over this issue we at AF1/Acculign have designed and are in the construction phase with Mark at Acculign for replacement link plates that are at a bare minimum 125% stronger than stock while weighing no more and yes, are still made from a strong aluminum alloy. They will be 100% CNC machined and anodized and made from a strong alloy. They are fully radiused to eliminate the "stress risers" in the stock parts. They will use the stock bolts. They are identical in ratio dimensions to stock.

Price will be beyond reasonable. DO NOT PRE ORDER! DO NOT CALL AND ASK FOR DETAILS. THEY WILL BE ONLINE AS SOON AS YOU CAN ORDER THEM. The production run will be final in the next two to three weeks. I am on vacation till next Tuesday so please do not PM me about these either.

Mark and I and a couple of other metal savvy types here at the shop have performed our own analysis of the stock parts and do indeed believe they are strong enough stock to withstand anything the stock bike can throw at them but since this is a customer/enthusiast confidence issue we are attacking at full force. Can they be a hell of a lot stronger and weigh no more, hell yes. That is what we are getting ready to make available. Just having a fully machined surface with NO SHARP edges and made of a superior strength material will do the job times 2 at least!

mike p
11-21-2007, 12:11 AM
my 'standards for aluminium mill products' book shows;
sheet and plate;
2024-0
2024-T3
2024-T36
2024-T4
2024-T351
2024-T42
2024-T62
2024-T81
2024-T851
2024-T86
no T-6 though
it failed in tension ......its not strong enough, period
Aprilia screwed up when they made the legs of the plate narrower, to save weight
gee, folks, it can only maim or kill you .....don't whine.....fix it!

Noel-theknurlBetter update that book.

http://asm.matweb.com/search/SpecificMaterial.asp?bassnum=MA2024T6

millemille
11-21-2007, 04:30 AM
you'd be surprised to see how much a swingarm can flex, it has to.

millemille, so your saying leave the plates the oem thickness?
while i don't understand everything you are saying, your posts seem to be making the most sence

Chris - most swingarm deflection occurs in the "legs". The chances of flexing the large casting/fabrication that joins the pivot to the the legs is minimal.

I'm not saying leave the plates alone..yet... I'm saying that it is good engineering practise (that is applied every day in the aircraft, rail, maritime, automotive etc industries) to not "solve" a problem until you know the cause. Fitting thicker side plates might only postpone the inevitable for an unknown - hopefully longer - period or it might cause the next weakest component in the system to fail.

KZ - using the accepted statistical analyis principals as given by ISO, and are used in all disciplines of proffesional engineering daily to make safety critical decisions, the liklehood of the side plates failing due to simple lack of "fatigue strength" or tensile strength is improbable/as to be considered improbable.

For the failures to be related to "fatiuge strength" the duty cycle of the side plates, or the factors that make up the duty cycle, of the failed plates would be apparent as convergent. The failed plates that we know of alone should have either been on the highest mileage bikes fitted with the suspect side plates and/or had the heaviest riders and/or had lots of pillion carrying and/or had the heaviest weight springs fitted and/or been used on the poorest quality roads and/or been consistantly ridden faster that any other example. There are enough bikes on this forum alone - let alone within the whole worlwide population - that exceed the above parameters of the failed bikes by enough multiples as to make duty cycle, and hence "fatuige strength", the cause of failure so unlikely as to be impossible -again as defined within the ISO parameters.

For ultimate tensile strength to be the root cause of failure is not quite so clear cut as to be dismissed at this stage - but of the 3 failures described on here not one of them had an "event" immediatly prior to failure that could be considered likely to exposed the side plates to loadings that would have exceeded their ultimate tensile strength. One of them was riding along a road in Holland which isn't known for it's poor quality roads....unless the bike had just been dropped out of a plane the chances of a major impact to the rear suspension being the intitiator of the failure are in the realms of highly unlikley.

Italjet - you could well be right that ultimately the lateral flex induced by dynamic loading of the rear suspension system is a significant factor in the side plate failure. But at this moment to me, considering that puts us back into the realms of duty cycle or catastrophic event... and the same comments as above seem to apply. I believe the key to understanding the cause is to look at the suspension interaction and the loadings on the side plates statically first.

spoonz
11-21-2007, 04:54 AM
good job at AF1/acculign

You seem to realise what Aprilia don't which is that most of us don't have a degree in Engineering and when something like this happens, it hits confidence of existing customers and potential customers.

To stonewall everybody helps nobody (Scott's situation for example) and it's commendable you have the business savvy to recognise an issue even if in your opinion it's not a bad engineering issue.

Italijet
11-21-2007, 07:40 AM
Italjet - you could well be right that ultimately the lateral flex induced by dynamic loading of the rear suspension system is a significant factor in the side plate failure. But at this moment to me, considering that puts us back into the realms of duty cycle or catastrophic event... and the same comments as above seem to apply. I believe the key to understanding the cause is to look at the suspension interaction and the loadings on the side plates statically first.

Millemille, thanks...

Looking at the stress loading (angle, force, direction etc) that is exerted on the rear subframe, through the shock and onto Swing arm. When aligned to the same loading from the ground through the wheel and swing arm, it would suggest that "hanging off" plus increased ride height (extenuating the leverage) could cause sufficient flex between the two ends of the shock to cause a twist in the linkage.

Of the Failures that we have reported here, how many occurred:
post corner (exiting), pre corner (entry), on a straight bit of tarmac ?

Of those reported, how many had set the suspension to increase the ride height and stiffness ?

millemille
11-21-2007, 08:19 AM
Millemille, thanks...

Looking at the stress loading (angle, force, direction etc) that is exerted on the rear subframe, through the shock and onto Swing arm. When aligned to the same loading from the ground through the wheel and swing arm, it would suggest that "hanging off" plus increased ride height (extenuating the leverage) could cause sufficient flex between the two ends of the shock to cause a twist in the linkage.

Of the Failures that we have reported here, how many occurred:
post corner (exiting), pre corner (entry), on a straight bit of tarmac ?

Of those reported, how many had set the suspension to increase the ride height and stiffness ?

Not sure how there can be any "force" exerted through the shock as is it not mounted top and bottom on rose joints?

kzmille
11-21-2007, 09:05 AM
KZ - using the accepted statistical analyis principals as given by ISO, and are used in all disciplines of proffesional engineering daily to make safety critical decisions, the liklehood of the side plates failing due to simple lack of "fatigue strength" or tensile strength is improbable/as to be considered improbable.

I'm not saying that they failed due to a lack of fatigue strength or a lack of tensile strength. It may have been a manufacturing defect, a material flaw or countless other causes. What I am saying is that the failure to take into account these possibilities and build in a large enough safety margin and/or to choose a material that is more forgiving (not requiring shot peening) represents a design flaw. All these many aspects are part of the design of the part.

millemille
11-21-2007, 09:20 AM
But you don't design a component to take into account factors such as manufacturing defects or material flaws.

You choose an apropriate production method and validation method that ensures that the component is produced to the required standard without the need for any comfort factor in the design other than what the intended use component requires - the production method and material flaws doesn't drive the design it drives the production methods.

The side plate designers thought process wasn't " These need to be T6 tempered but just in case there's a cock up in delivery/production/luigi in goods in puts them on the wrong shelf etc I'll assume they'll actually be made from T4 and design them accordingly..."

Hypothetically... what if the side plate failures is caused by a production issue in the swing arm causing a misalignment in the shock linkage - should the designer of the side plates have taken this into account when designing the side plates? No...

What if the side plates are failing due to inclusions in the original material - Should the designer have taken this into account? No...

What if the side plate failures are due to a cumulative misalignemnt issue between the frame and swingarm..even though the individaul misalignments are within production specification - should the designer of the side plates have taken this into consideration? Yes...

The fix for all of this might not be thicker side plates - as this might move the problem else where or leave a risk of failure at a greater periodicity - but maybe a check on frame/swingarm alignment and possible replacement or maybe replacement side plates OF THE SAME DESIGN with a better validation process post production to ensure that material defects do not get onto production machines... or whatever it is once the cause of failure is known and understood.

rx7tt95
11-21-2007, 06:08 PM
Now that's what I'm talkin' about! AF1, you have my loyalty for life. Where do I send my deposit for the V4? :)

theknurl
11-21-2007, 08:54 PM
just disassembled my rear suspension;
as previously reported right plate gouged by swingarm
all 4 needle old bearings are bad, the seals let water in.
Al, you may be partly right there is side load on the dog bone [conrod]
i doubt its coming from the shock, as millemille said rose [Heim] joints at both ends
on my bike the only wear marks are on the plates[didn't see swingarm it was just replaced]
the needles in the front end of the dog bone were skewing 10-15 degrees, you can see the wear marks in the outer race.
i'm putting grease nipples [Zerks] in the dogbone and the swing arm there is a gap between the bearings
the misalignment between the front dogbone mount and swingarm mount for the plate is ~0.12mm about 3.5 minutes, swingarm angled to the left
i clamped a straight edge to the left inner side of the front dogbone mount and a second one to the left side of the plate mount on the swing arm
problem is, this isn't the swing arm that gouged the plate....its gone, can't measure it
mike p, yes my books out of date. thanks
the bearings and seals;
6 Torrington HK 1812V [18x24x12mm]
6 seals, no visible # size 18x24x3mm
Noel-theknurl

ckruzel
11-21-2007, 09:01 PM
i'll probably tear down the bike this winter, i ride alot in the rain, so i'm sure my bearings are in need of replacement, i usually get my bearings from bearing and drives unlimited, i've always replaced needle bearings with caged bearings on my racebikes

mike p
11-21-2007, 10:50 PM
just disassembled my rear suspension;
as previously reported right plate gouged by swingarm
all 4 needle old bearings are bad, the seals let water in.
Al, you may be partly right there is side load on the dog bone [conrod]
i doubt its coming from the shock, as millemille said rose [Heim] joints at both ends
on my bike the only wear marks are on the plates[didn't see swingarm it was just replaced]
the needles in the front end of the dog bone were skewing 10-15 degrees, you can see the wear marks in the outer race.
i'm putting grease nipples [Zerks] in the dogbone and the swing arm there is a gap between the bearings
the misalignment between the front dogbone mount and swingarm mount for the plate is ~0.12mm about 3.5 minutes, swingarm angled to the left
i clamped a straight edge to the left inner side of the front dogbone mount and a second one to the left side of the plate mount on the swing arm
problem is, this isn't the swing arm that gouged the plate....its gone, can't measure it
mike p, yes my books out of date. thanks
the bearings and seals;
6 Torrington HK 1812V [18x24x12mm]
6 seals, no visible # size 18x24x3mm
Noel-theknurl When I changed my plates the needle bearings looked ok though I did not remove them from there mounts. I slid the shaft over to one side so I could stick my finger in there and put some grease on the rollers. The seals also looked good and I did not see any signs of water or dirt in the area. I also did not see any grease where I thought there should be some left over from the original assembly.

Italijet
11-22-2007, 03:32 AM
Al, you may be partly right there is side load on the dog bone [conrod]
i doubt its coming from the shock, as millemille said rose [Heim] joints at both ends
Noel-theknurl

Noel,
Take a look at the assembly. There are only 2 places that can subject it a force loading:
The rear tyre, via the wheel, axel and swing arm.
The Riders weight, via the rear subframe and shock.

The bushings at the bottom of the shock will alow the shock to change it's angle in relation to the bolt. The effect of this will be that the force of the shock will be imparted across the retaining bolt and not directly through it. This then becomes a diagonal force across the plate, the weakest part of any triangular design.

millemille
11-22-2007, 04:01 AM
Noel,
Take a look at the assembly. There are only 2 places that can subject it a force loading:
The rear tyre, via the wheel, axel and swing arm.
The Riders weight, via the rear subframe and shock.

The bushings at the bottom of the shock will alow the shock to change it's angle in relation to the bolt. The effect of this will be that the force of the shock will be imparted across the retaining bolt and not directly through it. This then becomes a diagonal force across the plate, the weakest part of any triangular design.

Al - what you are talking about is one and the same... any force that tries to move the unsprung portion of the frame laterally in relation to the sprung portion has potential to place lateral loading on the suspension side plates (and the dog bone). The amount of hang off a rider uses is irrelevant - two identical bike and riders going round the same bend at the same speed, one hanging off and one not. The hanging off rider will have his/her CoG further out and lower but as the bike is more upright.... the combined CoG of bike and rider will be in exactly the same place as the non hanging off bike and rider therefore the lateral forces acting on the bike will be excactly the same.

The rear shock is mounted top and bottom on rose joints (although normally the top arrangement is a metalastic bush)- there is no way any significant lateral loading can be placed on the side plates by the shock (I'll qualify that slightly - if the displacement between sprung and unsprung portion were to be so great that the shock were to make contact with sides of the shock tunnel in the swingarm then it could)

The Dr.Jon
11-22-2007, 04:08 AM
just disassembled my rear suspension;
as previously reported right plate gouged by swingarm
all 4 needle old bearings are bad, the seals let water in.
Al, you may be partly right there is side load on the dog bone [conrod]
i doubt its coming from the shock, as millemille said rose [Heim] joints at both ends
on my bike the only wear marks are on the plates[didn't see swingarm it was just replaced]
the needles in the front end of the dog bone were skewing 10-15 degrees, you can see the wear marks in the outer race.
i'm putting grease nipples [Zerks] in the dogbone and the swing arm there is a gap between the bearings
the misalignment between the front dogbone mount and swingarm mount for the plate is ~0.12mm about 3.5 minutes, swingarm angled to the left
i clamped a straight edge to the left inner side of the front dogbone mount and a second one to the left side of the plate mount on the swing arm
problem is, this isn't the swing arm that gouged the plate....its gone, can't measure it
mike p, yes my books out of date. thanks
the bearings and seals;
6 Torrington HK 1812V [18x24x12mm]
6 seals, no visible # size 18x24x3mm
Noel-theknurl

For the benefit of the hard of understanding....can we have some pictures?:)

Italijet
11-22-2007, 06:27 AM
Al - what you are talking about is one and the same... any force that tries to move the unsprung portion of the frame laterally in relation to the sprung portion has potential to place lateral loading on the suspension side plates (and the dog bone). The amount of hang off a rider uses is irrelevant - two identical bike and riders going round the same bend at the same speed, one hanging off and one not. The hanging off rider will have his/her CoG further out and lower but as the bike is more upright.... the combined CoG of bike and rider will be in exactly the same place as the non hanging off bike and rider therefore the lateral forces acting on the bike will be excactly the same.


I understand what your saying, but I must disagree.

With the Bike vertical the weight and force imparted through the centre. Everything is aligned and forces exerted on the seat by the rider are taken by the spring directly through the bike to the ground. Everything is aligned and straight.

When "hanging off" the weight is distributed differently and the forces are not applied straight down the centre of the components. This can be shown at the extreme with a cardboard tube and a 1KG weight. Placed on it's open end with the weight placed squarely on the top of the tube. The structure should withstand the weight. Balance the weight towards one side and the tube will crease inwards and fail (the force becomes misaligned with the strength in the tube.

Whan hanging off, the force is applied to the side of the sub frame. The rose allows the shock mounting to swivel, but it won't change the direction of the force.

theknurl
11-22-2007, 03:22 PM
only the needles in the front end of the dogbone[conrod] were skewing
although water got in at both ends
another flex point; the dogbone is bolted to the left side of the frame, the head of the bolt has some play in the right side, so there is some bending load on the bolt
Noel-theknurl

dr_justice
11-23-2007, 06:38 PM
Well, all this talking has made me hungry.

Millemille:
You said that replacing the plates with stronger fatigue/tensile strength would only delay or make the next weakest link break. I understand that, but all I know now is that at this point, the link plates seem to be the weakest link. I agree it's best to fix the cause than try to hide the symptom, but as with a lot of medecine these days, perhaps fixing the symptoms now will save some lives and give enough time to Aprilia to acknowledge the issue and issue a replacement part for whatever may be defective, link plates or something else.

In the meantime, I'll go ahead and replace that until there's a better solution out there.

Diablo1
11-23-2007, 10:58 PM
[QUOTE=dr_justice;1648050]Well, all this talking has made me hungry.

Millemille:
You said that replacing the plates with stronger fatigue/tensile strength would only delay or make the next weakest link break. QUOTE]

If he said that....he's 100% wrong. Just because one part is too weak for the service conditions, doesn't mean that other parts on the bike are nearly as weak and are doomed to failure if the links don't break first. Whether the links are weak or strong doesn't change the stress levels on any other parts of the bike.

ckruzel
11-23-2007, 11:16 PM
got the first set of billet links today. after holding those ss plates, you guys can keep them, i spent money shaving weight off the bike, not adding it, titanium ones are coming soon and should be available to the masses, working on titanium axles, magnesium chain adjuster blocks as well, or maybe titanium blocks, but i'd prefer magnesium

some parts are coming off the bike this winter and being replaced with titanium

Ape Factory
11-23-2007, 11:49 PM
I have personally only seen two of these failures since the design was released in 2004. Both were on bikes that crashed into something causing extreme forces to be transmitted through the rear. Both failures were post crash, moment of impact type damage. One was a bike in 2004 that skidded of of a banked (positive camber) corner on the race track and came crashing to earth on the rear wheel which was bent along with the swingarm. The other was a bike that lowsided and slid into a rock (very large rock) striking the rear wheel first. Both bikes had damaged swingarms in addition to damaged wheels at the rear. The link plates broke as a result of the impact in both cases I have personally seen.

That said, since there is so much concern over this issue we at AF1/Acculign have designed and are in the construction phase with Mark at Acculign for replacement link plates that are at a bare minimum 125% stronger than stock while weighing no more and yes, are still made from a strong aluminum alloy. They will be 100% CNC machined and anodized and made from a strong alloy. They are fully radiused to eliminate the "stress risers" in the stock parts. They will use the stock bolts. They are identical in ratio dimensions to stock.

Price will be beyond reasonable. DO NOT PRE ORDER! DO NOT CALL AND ASK FOR DETAILS. THEY WILL BE ONLINE AS SOON AS YOU CAN ORDER THEM. The production run will be final in the next two to three weeks. I am on vacation till next Tuesday so please do not PM me about these either.

Mark and I and a couple of other metal savvy types here at the shop have performed our own analysis of the stock parts and do indeed believe they are strong enough stock to withstand anything the stock bike can throw at them but since this is a customer/enthusiast confidence issue we are attacking at full force. Can they be a hell of a lot stronger and weigh no more, hell yes. That is what we are getting ready to make available. Just having a fully machined surface with NO SHARP edges and made of a superior strength material will do the job times 2 at least!

Hello Micah,
While ordering a bunch of stuff from you Guys at AF1, I asked about lowering links for the bike (04's + RSVR/Factorys), not sure who it was. AF1 said they killed the project because it was becoming too expensive/time consuming to test. With all these Linkage concerns, any thoughts on continuing production development for lowering links for the vertically challenged?
Thanks,
Al:cheers:

millemille
11-24-2007, 04:55 AM
The fix for all of this might not be thicker side plates - as this might move the problem else where or leave a risk of failure at a greater periodicity - but maybe a check on frame/swingarm alignment and possible replacement or maybe replacement side plates OF THE SAME DESIGN with a better validation process post production to ensure that material defects do not get onto production machines... or whatever it is once the cause of failure is known and understood.


Don't put words in my mouth...

spoonz
11-24-2007, 06:39 AM
got the first set of billet links today. after holding those ss plates, you guys can keep them, i spent money shaving weight off the bike, not adding it, titanium ones are coming soon and should be available to the masses, working on titanium axles, magnesium chain adjuster blocks as well, or maybe titanium blocks, but i'd prefer magnesium

some parts are coming off the bike this winter and being replaced with titanium

ME ME i like unobtanium. Put me down for some goodies Chris

Diablo1
11-24-2007, 09:01 AM
I believe the key to understanding the cause is to look at the suspension interaction and the loadings on the side plates statically first.

Statistics?? Hell no.
If someone sends me a broken link, I can tell you exactly why it broke from failure analysis. The key is to examine the metallurgical evidence of the fracture. When you want to solve a murder, you don't go to the library and look for statistics on murders, you look at the dead body and crime scene first. There's less than a dozen likely reasons for the link failures, but no one can tell which one(s) are responsible without examining the broken links.

Speedtres
11-24-2007, 10:46 AM
got the first set of billet links today. after holding those ss plates, you guys can keep them, i spent money shaving weight off the bike, not adding it, titanium ones are coming soon and should be available to the masses, working on titanium axles, magnesium chain adjuster blocks as well, or maybe titanium blocks, but i'd prefer magnesium

some parts are coming off the bike this winter and being replaced with titanium

Again i'm in the for Ti. Thanks for putting the work in.

millemille
11-24-2007, 10:51 AM
Statistics?? Hell no.
If someone sends me a broken link, I can tell you exactly why it broke from failure analysis. The key is to examine the metallurgical evidence of the fracture. When you want to solve a murder, you don't go to the library and look for statistics on murders, you look at the dead body and crime scene first. There's less than a dozen likely reasons for the link failures, but no one can tell which one(s) are responsible without examining the broken links.

Hopefully your ability to extrapolate failure root cause from examination of the failed side plates alone is better than you reading comprehension..

Statically... Statistically..not the same word

All you can tell them is the direct cause of failure at best - what you cannot tell them is root cause..

Examining the failed side plates is part of a process to extablish the root cause and solution but on it's own has little merit.

Diablo1
11-24-2007, 01:07 PM
Hopefully your ability to extrapolate failure root cause from examination of the failed side plates alone is better than you reading comprehension..

Statically... Statistically..not the same word

All you can tell them is the direct cause of failure at best - what you cannot tell them is root cause..

Examining the failed side plates is part of a process to extablish the root cause and solution but on it's own has little merit.

You are correct, I didn't read that right. Static loads are important. However, in many cases a proper metallurgical examination of the broken parts can pinpoint the source cause of the failure. For example, I did a failure analysis on a bent drive axle from a Yamaha quad some years ago. The axle failed in a totally ductile manner - bent severely with very little use and gentle driving. The microstructure and hardness showed that the axle had not been quenched and tempered or induction hardened after machining. The alloy they selected was designed to be hardened and not to be used in the fully annealed condition. My conclusion was that this particular axle had missed the heat treatment step during manufacturing. The metallurgical exam will send you in the right direction, but you need to know if the fracture is ductile or fatigue and where the crack was initiated.

So once again...if folks are serious about getting this problem solved...send me a pair of broken links.

ckruzel
11-26-2007, 02:22 PM
ti ones will be around $116.00 shipped to the usa, throwing it out there to see how much interest is out there, if there isn't much it might just be a one off set for my bike

Speedtres
11-26-2007, 03:20 PM
ti ones will be around $116.00 shipped to the usa, throwing it out there to see how much interest is out there, if there isn't much it might just be a one off set for my bike

I'm in.

spoonz
11-26-2007, 05:37 PM
ti ones will be around $116.00 shipped to the usa, throwing it out there to see how much interest is out there, if there isn't much it might just be a one off set for my bike

If you will ship to Uk i'm in.