Front straight lower suspension arm failure - can it be prevented?

[Davezj]

well i have been doing some thinking and it has all been about safety which is not like me. I have come up with a possibility.

it will not stop the ball joint failure but engineered right it could stop the joint falling apart. I have been thinking we as club are so lucky that no one has been killed by this style of suspension failure yet, well not that i know of.

in basic terms you weld a nut/bolt to the top of the top wishbone and then bolt the top of a length (correct length) of box section steel to it.
fit a longer bolt to the lower suspension arm where the shock fork attach's and bolt the bottom of the box section length to it.

this should stop the joint breaking apart when the ball joint fails, potentially at high speed.

there will probably be a need for a bit of compliant bushing in the top and bottom fixing holes in the box section, so it does not stop the correct suspension movement. the length of the box section is the critical thing. it should not stop the suspension from moving up and down as the hub carrier fixes the distance from top wishbone to the lower arm anyway.

my theory might fall down if the box section sits on an angle rather than going straight up and down which would change the movement, i think. i will have to have a proper look at the suspension over the weekend to see if it will work or just create triangles effectively and lock the suspension in place and render the suspension useless.

please feel free to comment, especially if you have a "don't be stupid that will never work due to ........." comment.

i have been worrying about suspension failure as you can probably tell.

if you think the link between top wishbone and lower arm can not be solid then a cable of the correct breaking strain could be used instead but this would cost a lot more to impliment.

[The Vee]

I too have been trying to come up with alternatives including using USDM set up which unfortunately is a non starter on our cars as the turret is completely different on the inner wing.
With your idea I can see what you're trying to do and you may even stop the lower arm dropping a little. Unfortunately that's not the biggest issue. When then ball joint shears or pulls out the whole hub and wheel breaks free at the bottom and splays out. This would be unaffected by your mod.

[The Vee]

Another idea I had was if we could turn the lower arm upside down so the ball joint went in from the top. But again , probs with the ball and socket being reversed would require redesign of socket and not enough room between the cv joint and hub. Then redesign hub mounting point by lowering it, but costs, strength etc etc again prohibitive. And then all alignments would alter. For now I think just pulling it apart every year and checking is what, unfortunately we're left with.

[Davezj]

i was hoping that if you stop the bottom arm moving away from the ball of the joint, the ball would go really slack in the hole but it would not pull out of the hole because the box secsion would hold the top of the hub carrier the same distance from the lower arm. stopping it coming apart.
does this not work then?

but it would not help the senario when the actual threaded section of the ball joint shears off the ball bit.

[The Vee]

I can see that part of it but then your basically transferring the weight of the car to the upper arm for which it's not designed. Itwould have to carry weight to stop the ball dropping enough. Not also too sure what would then happen to the "dynamics" of the car, although there are one or two cars that are designed with the shock/spring sitting on the top arm.
Not in any way trying to dishearten you, in fact just the opposite - I would love to find a solution to this that wouldn't cost the earth. It is one of my biggest concerns with the car. Pity it never came to light when the cars were newer as Mitsi would probably have to have had modded it rather than the rather lame recall. It's by no means impossible to cure, just the costs and if it were taken up would there also be design and construction issues.

[Davezj]

Another idea I had was if we could turn the lower arm upside down so the ball joint went in from the top. But again , probs with the ball and socket being reversed would require redesign of socket and not enough room between the cv joint and hub. Then redesign hub mounting point by lowering it, but costs, strength etc etc again prohibitive. And then all alignments would alter. For now I think just pulling it apart every year and checking is what, unfortunately we're left with.

the trouble with inverting a ball joint is the ball gets forced in to the socket and creates extra resistance.

think of the way you undo a ball joint if it spins in the socket when it is the normal way up. you can put a jack under the ball end and apply a bit of pressure which increases the Resistance which in turn allows you to apply more turning force to the nut allowing you to under it.
So when you invert the ball joint are you going to be making the steering heavier, by how much i don't know.

but i get what you are saying, as i spoke to Paul (PSBarham) at donington, we spoke about this very thing.

[The Vee]

[QUOTE=Davezj]the trouble with inverting a ball joint is the ball gets forced in to the socket and creates extra resistance.
QUOTE]


Marginally, perhaps. But this effect is also happenining the other way with the weight of the car pulling the pin onto the shoulders of its socket.
When jacking that other way up one to undo the nut, the resistance is not the ball in the socket. It is pushing the taper fit pin back into it's hole and causing more resistance by the interference fit.

[lancerevo3]

what about some form of clamp that goes around the bottom of the arm and comes up around to both sides of the ball joint [like a fork type ball joint splitter].so when it pops it cant fully come out.but at the same time sitting between hub carrier and ball not clamping it if you get me?

[Davezj]

I can see that part of it but then your basically transferring the weight of the car to the upper arm for which it's not designed. Itwould have to carry weight to stop the ball dropping enough. Not also too sure what would then happen to the "dynamics" of the car, although there are one or two cars that are designed with the shock/spring sitting on the top arm.
Not in any way trying to dishearten you, in fact just the opposite - I would love to find a solution to this that wouldn't cost the earth. It is one of my biggest concerns with the car. Pity it never came to light when the cars were newer as Mitsi would probably have to have modded it rather than the rather lame recall. It's by no means impossible to cure, just the costs and if it were taken up would there also be design and construction issues.

i get what you are saying with this, but isn't that how it is configured already through the hub carrier.

normal configuration:-
the bottom arm is being pushed down always by the spring. the force from the spring is being transfered to the hub carrier via the ball joint, then the hub and out to the wheel forcing it downwards.

in failure with the link box section:-
the spring pushes down the force pulls the link box section down which presses down on the hub carrier, then the hub and out through the wheel.

yes i see the difference now. but hopefully the lower arm will not move by any more than a few milimeters, so the ball doesn't come out of the socket. the upper arm does not normally take any load of the spring at all. It just keeps the top of the hub carrier in the correct place.

[The Vee]

Certainly I see what you're trying to do with the box section idea. Another factor you'd need to consider is the mounting point on the upper arm. The arc drawn by the two arms will be different as they move up and down according to where you fix it.
Is there not some way of fabricating some sort of "catch" system locally to the joint where it attaches to the hub? I did cime up with an idea for this involving a bracket and pin should it drop but it appeared a little cunbersome and not easy to allow for when the steering turned.
Maybe just get the USDM inner wings (or a USDM galant complete) /pan
Your thread does give me renewed interest in this, must be able to sort something!

[scott.mohekey]

What about a length of thick braided wire going between the two parts? I.e. a small plate bolted onto the hub carrier with the ball joint's bolt, and somewhere on the arm?

[Davezj]

my thought for this was started by remebering the staps that are used by monster trucks to stop the shocks from over extending. you might have seen them.
Maybe we could fit a strap under the ball joint a bolt it into the hub carrier some how.

Edit:- simplar idea.

[scott.mohekey]

Ok, a loop of thick braided wire, attached at the ball joint bolt?

[Davezj]

what about some form of clamp that goes around the bottom of the arm and comes up around to both sides of the ball joint [like a fork type ball joint splitter].so when it pops it cant fully come out.but at the same time sitting between hub carrier and ball not clamping it if you get me?


not ignoring you, i had considered this but i couldn't see how it could done at a resonable price, but then how much is life worth.

yes you could put some sort of catch fork/clip under the hub carrier to ball joint connection and round the bottom of the ball joint itself. which would stop the ball as it came out of the socket.

but i don't know how, yet!

i will think about it toworrow and try and get something posted up.

[lancerevo3]

much better idea with the wire instead of fitting suspension to suspension .

[Davezj]

Ok, a loop of thick braided wire, attached at the ball joint bolt?
i see where you are going with this,

a piece of wire rope with and eyelet/loop on each end, one loop goes on ball joint thread, loop wire round the bottom of the lower arm and then second loop goes on to the thread.

looks like it will work but you are not going to be able to make the loop very tight around the lower arm due to the need to put second loop on to the thread as well. if there is just enough wire length to loop the second eyelet on then when it is pushed down the thread to the shaft bit it will start to sag under the bottom of the ball joint.

but it is a good idea and i will work on it.

[scott.mohekey]

Close, but I think it would be better to have a steel washer with a tab under the ball joint bolt, with the wire loop attached to the tab in some way.

[Davezj]

Close, but I think it would be better to have a steel washer with a tab under the ball joint bolt, with the wire loop attached to the tab in some way.

sorry, but i am not getting your description.

[The Vee]

what about some form of clamp that goes around the bottom of the arm and comes up around to both sides of the ball joint [like a fork type ball joint splitter].so when it pops it cant fully come out.but at the same time sitting between hub carrier and ball not clamping it if you get me?


Sorry mate, missed this. Think that is a similar type thing I was thinking with tha catch idea I was on about. Could loop round and locate under the nut of the pin itself. Only thing is it would have to allow the hub to turn (steering) hence the location pin.
Got some old arms here, maybe time for another look

[scott.mohekey]

Excuse my art skills.. but something like this. It doesn't need to be tight, just loose enough to allow all play.

[Davezj]

Sorry mate, missed this. Think that is a similar type thing I was thinking with tha catch idea I was on about. Could loop round and locate under the nut of the pin itself. Only thing is it would have to allow the hub to turn (steering) hence the location pin.
Got some old arms here, maybe time for another look

sorry not sure what the location pin is for or where it is fitted.

[Davezj]

Excuse my art skills.. but something like this. It doesn't need to be tight, just loose enough to allow all play.

right i see what you mean, great picture says everything that need to be said.

[scott.mohekey]

Obviously, the wire would be thicker, but I exhausted the size limits of MS Paint's brushes /pan

[The Vee]

Location pin would be fixed to the arm that would pass through a metal bracket/loop fixed in a similar position to the wire diagram in turn located by the pin nut shown. This would be engaged to hopefully stop the wheel splaying out if the pin snapped. But as I say, the bracket i think would be flawed by the amount of movement need for steering as it would fowl the side of the arm and probably also get in the way of the curved arm fixing. I need to try a photoshop job!

[Davezj]

the wire loop would have to be big enough to allow the full motion of turn for the hub though.so it would be a suck it see type thing.
we could workit out by using a cable tie round there and tighten it up test movement tighten it up a bit more, test movement, tce then cut cable tie off and measure it and that should give a good start point for making a prototype of the right size.

[Davezj]

Location pin would be fixed to the arm that would pass through a metal bracket/loop fixed in a similar position to the wire diagram in turn located by the pin nut shown. This would be engaged to hopefully stop the wheel splaying out if the pin snapped. But as I say, the bracket i think would be flawed by the amount of movement need for steering as it would fowl the side of the arm and probably also get in the way of the curved arm fixing. I need to try a photoshop job!

i think i get it but a pic would confirm it.

[Davezj]

just seen the time 1.40am

work in the morning time for bed.

[The Vee]

i think i get it but a pic would confirm it.


Yeah I'll try tomorrow - off for some ZZZZs soon!

[The Vee]

Does this make any sense?

[lancerevo3]

on the other hand if wire were to foul movement steering etc.the suggestion of the fork type clamp mentioned earlier would not foul movement due to it location.trial and error i suppose!!

[lancerevo3]

sorry just seen picture .whats the blue piece?

[The Vee]

The blue piece is a bracket fitted to the lower arm which extends into to catch loop and has the pim coming down from it locating in a hole in bottom of loop. The hole being big enough to allow movement.

[lancerevo3]

so if ball joint splits the blue and pink pieces hold it together?

[lancerevo3]

not putting your design down or anything but i think this design is gona cause major movement issues.the last design with the wire was less restrictive i think

[White Lightning]

Looks like I may have sparked an Engineering idea frenzy here ... great thread chaps :thumbsup:

[The Vee]

not putting your design down or anything but i think this design is gona cause major movement issues.the last design with the wire was less restrictive i think


Correct, if you read further back I have already stated just that. But that was one of my initial thoughts on how to prevent it all falling apart. Trouble is, it is just that, stopping it falling apart after the joint has broken. I would still rather come up with something to stop it happening in the first place.
I'm still convinced there must be a different set up somewhere to swap over. I've looked at USDM galants, evos, differebt hubs etc etc. All have their merits but then fall down in one area or another.
I have felt very strongly about this issue for some time now (lots of earlier threads) and have since discovered that there was even a recall on the 7G galant for similar issues.
Shame we can't get every galant owner to get together and lobby Mitsi for an alternative, but alas I feel the cars are too old now.

[lancerevo3]

ok .i dont know about this one but anyway ,what about redesigning arm so the outer end curves up a bit to follow the shape of the ball just enough to stop it popping but not so much to prevent movement?
just an idea not to up on how much work involved with this one!!

[The Vee]

Good to see you still thinking about this. I bet, like me, you went to bed thinking of it, hoping an answer might be there by morning!
Think I can see your thoughts on the last one, but yes costs of a redesign I think would be prohibitive. My thoughts are still to find existing components or possibly some kind of "catch" device as we've discussed earlier. Want I want to avoid if possible is major surgery, such as inner wings, because I don't really think it'd be worth it

[Davezj]

Guy at work put me on to these people who can anything to do with wire rope and the likes

http://www.mainco.co.uk/wire_rope_info.htm

so stuff is available to do this.

just thought i would ask, there are 2 different failure method for the ball joint, and they are as follows:-

1. the entire ball gets forced out of the socket and the lower arm falls away.

2.the tapered shaft of the ball joint snaps off the top of the ball leaving the ball in the lower arm and the lower arm would all away.

what is the split of failkure types, i don't remember seeing any more than one of the type 2 failure mentioned (but i could be wrong).

this leads me to my next question, do you want to protect against a type 2 failure?

If no then the solution to the catch device is very straight forwand and will not compromise hub/steering movement. see post 16 for description.

i will try and do a drawing of it.

sorry it is not a .jpg
33165

the drawing is terrible but it is only to the basic idea. i can do full 3d cad drawings of it but then you would not be able to open it. unless you had the free right hemesphere software you can down load to view that sort of stuff.

[The Vee]

Although this would never be a device for driving I would for myself defo have the second failure included also. At least I could hopefully bring the vehicle to a stand were it to fail in that manner. Fortunately so far most have failed at very low speed and the simpler version would probably suffice for these.

[Davezj]

i am of the opinion that if you can protect against both then that would be better.

how about a very simple wire rope ring that can be fixed in place by spliting the ball joint off the hub carrier put the ring onto the hub carrier behind where the nut for the ball joint goes then re fit lower arm to hub carrier and slide the ring back over the end of the lower arm.

the ring wil now be round the hub carrier and lower arm and dropped into place behind the nut.
but then there is the movement issue again.

[lancerevo3]

gona put a hub and arm on a bench and see if anything jumps out at me,or better still if each person does it and see what comes up ?have to be a solution without breaking the bank

[The Vee]

Have rethought that one that I put up earlier. Couple of ideas - will post later tonight as I'm currently at work.

[lancerevo3]

right ,how about a longer bolt that fixes bottom of shock to arm,then attaching a bracket that would sit either side of the arm.from there both sides would meet at a single point resulting in a triangle shape.then a loop/ring attached to top .
now if a braided wire as said earlier was fixed to one side of hub and ran along back of hub where ball joint thread goes through and also through loop/ring of triangle bracket and attached to other side of hub.
NOTE:braided wire could be exchanged for round bar or similar to follow shape of ball joint mounting point.
SORRY IF LONG WINDED JUST TRYING TO DESCRIBE AS BEST I CAN.

[Davezj]

well i have been out today and bought some wire rope and fixing.

i am surprised how low the Safe Working Load is on wire rope. 6mm wire rope 300KG, which does not sound much.
however the breaking load of the same wire rope is about 2 tons.
i do not know how to calculate what the shock load of the ball joint failing, it would have to be some kind of combination of spring rate, and corner weight of the car.

i will try and get some kind of length measurement wire rope will need to be to ensure steering movement is not compromised.

oh i just had a thought, it a ding, light bulb moment.
you can measure the force of a failure if you undo the ball joint nut and move than likely the joint will not split with the use of a ball joint splitter. so if a force gauge is put under the ball joint end when the joint is split from the hub then you will have your answer.
but the cars weight will have to be on the joint as well so the car can not be jacked up while doing this.the gauge would have to be just below the normal resting point of the bottom of the ball joint and to access this the wheel would have to be off. so the gauge would have to on to of some kind of axle stand that can take the weight of the simulated failure.

well the is how i see it anyway. please comment.

oh and if you have any kind of idea as to what could be used as the gauge then that would be good as well. some kind of weighing device with a peek hold on it.

[The Vee]

Don't forget the energy that also has to be stopped should the pin break. Not got round to posting my modified version but will try and get it done later.

[scott.mohekey]

I think that with the wire method, the weakest point will be where the wire is joined/clamped to the fixings.

[Davezj]

if the pin breaks or the ball comes out of the socket, will that not relesae the same amount of energy as the ball joint splitting with the tool. it is just a different method of releasing the energy. I think.

[Davezj]

I think that with the wire method, the weakest point will be where the wire is joined/clamped to the fixings.

yes i agree with that, however there is a methed of joining wire rope into a complet circule without and fixings and the method is as strong as if not stronger than the wire rope on its own. and the wire rope does not actually have to be fixed to anything. unfortuneately it is a time consuming and potentially more expensive mothed. see pic below. you can splice the wire rope together like normal rope.

but how tight the radius can be i don't know, but with this methed the more you pull on it the tight the joint gets.

33177

[The Vee]

if the pin breaks or the ball comes out of the socket, will that not relesae the same amount of energy as the ball joint splitting with the tool. it is just a different method of releasing the energy. I think.

Think of it as holding a hammer in the normal way and have the head on a bench on pushing down on the handle. Then raise the hammer head by just say an inch and then push down with the same force. I Know which one I wouldn't want my finger under! Think it's kinetic energy but it's a long time since I did that /pan

[Davezj]

i am not so sure your analogy is right. but i am stumped as to the measuring device. apart from some very expensive industrial kit i can't think of anything that would do the job.
we could use some kind of corner weight device from a specialist garage but i am not sure they would have a max hold feature on them, plus we need to do this quite cheaply.
does anyone know what the corner weights are on a VR4?

[Davezj]

Think of it as holding a hammer in the normal way and have the head on a bench on pushing down on the handle. Then raise the hammer head by just say an inch and then push down with the same force. I Know which one I wouldn't want my finger under! Think it's kinetic energy but it's a long time since I did that /pan


i have just reread my description of the measuremnet setup and i did not make clear the there would be a gap between the bottom of the ball joint and the measuring gauge. description can be a confusing thing, i have one picture in my mind that i am describing and when you read that description it paints a different picture. I should have done a drawing.
i did not mean,
if a force gauge is put under the ball joint end so it is touching the under side of the ball joint. when the joint is split from the hub then you will have your answer.

So yes i get you hammer analogy now, and agree with you thinking.

[Louis]

(Coming from a rigging background), it is possible to get 4mm diameter wire with a 1ton swl. wires can have crimped loop ends. I have had safety cables made up, when flying/hanging, lighting,PA equipment on trussing.
Some thing similar to F1 cars which would stop the wheel leaving the car, only in our case something that would stop the hub assembly moving out from the car. Poss a steel wire anchored where the straight lower arms inner bush is, and onto the hub?

[lancerevo3]

would the force be greater if you were cornering hard and fast when it pops
as opposed to splitting it yourself?

[fluffnik]

Fortunately so far most have failed at very low speed and the simpler version would probably suffice for these.

Low speed and lots of lock seem to be the common feature of all the failures I've read of. I'm not sure keeping the arm and upright attached but not located would help that much.

Some version of the wire loop idea would be helpful in the typical low speed maneuvering situation though.

It's quite possible that the loadings that cause failure only occur with more lock than you'd ever use at speed, Alpine passes notwithstanding...

[Davezj]

yes the failure point do seem to be low speed manoeuvring which has been lucky up to now.

i had another ding, light on moment last night just before i went to sleep.

it is too difficult to explain properly in words but the basic idea is to have 2 curved metal plate with interlocking lips on them that don't touch in general manoeuvring but if the joint fails then they will lock together and stop separation.

i will draw it up later today but i will have to do a 3d cad model of it and create a 2d drawing from that but i am not sure how to get the drawing into a format everyone can see without downloading some viewing software. the Right Hemisphere 3d model software i mentioned above will not work for stuff i model up at how because i don't have the RH file creation software at home only at work but there is a JT file formate that can be exported and there is free software to download to view the 3d model in that.
i don't have a PDF creator at home either so i can't print to PDF for the 2D drawing, unless so one can point me in the right direction of a freebie PDF creator.

[The Vee]

While I think I know what you mean about the plates, the computer speak uh oh! Still havn't drawn up my modified one but will get round to it, Seen another issue unfortunately in how to fit it withoit having a larger gap at the bottom. All will be revealed when I do my drawing.

[The Vee]

OK here's a drawing of later ideas. If your metal plates with lips / hooks on it is what I think it might be then sounds good - Drawing please even if it's just a few lines like mine!!

/ Damn pc. There's supposed to be 2 diagrams - lost one /pan

[The Vee]

Here it Is!

[The Vee]

Or this? Maybe similar to what you're thinking?

[Davezj]

the final one is kind of what i was thinking of,
i would have made the smaller top bit fold down over the edge of the hub carrier then fold back up to form the lip.
the bottom bit would form and arc with a lip on it to inter lock with the top bit.
i was thinking of securing the bottom bit to the suspension fork bolt. with a U shape channel running under the arm that would take the load in a failure. with some kind of loop over the top to stop it falling away when not in failure mode.

it is just the suspension travel up and down we have to think of so the lips do not catch on eachother at the extreme of movement.

[The Vee]

Yep can see what you're thinking. Have got a couple more to put up, then I think have to look at trying one. Will post em later.

[Davezj]

i have had a look at my suspension this morning and the wire rope option will be a quick and easy solution.

i measured the minimum length of wire rope will be 225mm but you will need the right sort of fixing for it. i have wire rope clamps but not sure this will be good enough.

Andy can you give me this info you were talking about, the 4mm wire rope with the high breaking strain and the best sort of clamp to secure it. the rope clamp i was going to try is this one.

i was going to thread the wire rope from either side and then loop the ends back through again, then clamp it tight.

33396

[The Vee]

Think it was Louis with the rope info. Have not really had a chance to do much more on this as I have been busy in any spare time that I have preping the car for Easter. Will get back to it though after the break.

[Davezj]

i did the 3D model of the metal slide past bit but i can't show it to anyone as the file format is not correct for a simple viewer.

[scott.mohekey]

i did the 3D model of the metal slide past bit but i can't show it to anyone as the file format is not correct for a simple viewer.

Just do a few print screens of good angles and post them.

[Davezj]

good point. i will do that!

[Davezj]

here we go,
this is a possible assembly, i am not saying it is an easy thing to make but it would give protection. it is not dimentionall correct, just an example.

first pic is both parts as they would slide past eachother

second pic os the bottom bracket that secures in place with suspension fork bolt going through it

third pic is the top bracket which fixes to the hub carrier with the ball joint thread going through the hole.

33402

33403

33404