Rear Toe Link Failure Analysis and Definitive Fix!

[marcw]

You may recall the now quasi-famous forklifted Elise at the WCLM this year.
It was forklifted off the track because it suffered a broken rear toe link.



If you missed the latest "all about the rear toe link" thread, here it is:

http://www.elisetalk.com/forums/showthread.php?t=23291&highlight=toe+link

This problem is potentially serious and dangerous enough for many to demand a recall:

http://www.elisetalk.com/forums/showthread.php?t=23472&highlight=toe+link

Stan did an analysis of the OEM part that sheared off at the track:



Lotus sells a "track pack" brace kit that seems to help with this problem, although its primary purpose is to reduce bump steer during extreme track driving. The root of the problem seems to be the inadequate size (10mm) of the OEM ball joint bolt that carries the rear cornering load.

Well, that's not the entire story as it turns out! Take a look at these pictures of the OEM ball joint bolt from the non-failed-side of this same car:




Notice the red arrows pointing to the shiny areas. This is evidence of the stock 10mm bolt moving back and forth against the tranny mount. It was discovered that the inside diameter of both the hole in the tranny mount and the factory rear A-arm bushing are a bit too large for the stock 10mm bolt, and there is therefore a noticeable bit of slop inherent in the factory inner ball joint bolt to chassis interface. In other words, it does not so much matter if your inner toe link bolts are properly torqued, they will still wiggle inside their mounts under stress of hard track driving because of the sloppy tolerances! This will lead to the gradual weakening and failure of this bolt as outlined in Stan's analysis. Therefore, the factory track pack brace really fails to address the root of the problem, since it still uses a little 10mm bolt through the tranny mount and the factory rear A-arm bushing.

So, how to properly remedy this situation??

Well, to make a long story short, the car has been up at Buttonwillow since the WCLM '06 track event getting this issue expertly resolved (it still as of today has all the masking tape on it!)


It was decided obviously that we needed to go with a bigger diameter, higher-grade inner toe link bolt than what is offered by Lotus. The heart of the new system comprises of an aircraft grade 7/16" AN bolt (suitable for safety wiring) and stover nut, together with a new custom machined spherical ball joint spacer featuring a machined insert channel that fits into the factory keyed hardened insert on the tranny mount's inside toe link location:



Before they re-installed the diffuser, I took a shot of the new system. It uses all new race-grade hardware including spherical ball joints throughout, and a center brace like the Lotus track pack brace, to bring the system into double-shear:



While this new redesigned system may be overkill for many Elise owners, I feel very strongly that the OEM setup is not adequate for track duty, and is even more of a potential hazard given what I know now about its design.

I have asked the race shop that did this system for me to make it available to other Elise owners. So far, my car is the first to have this new setup. I looks like there is a fellow running and Elise in SCCA T-2 who will be getting the second one installed shortly. This "kit" will require minor machining of the A-arm busings and the inner toe link mounts, plus it will require a rear-end re-alignment, so I dont think it's really a "bolt-in" solution for the shade-tree mechanic or do-it-yourself'er.

Pricing has not been announced, nor do I think it ever will, because this will likely remain a relatively low-volume product.

If you want this for your Elise, call Tom at 7's only racing at the Buttonwillow track: 661-764-5456 http://www.sevensonly.com/


Good luck and see y'all next time at the track!

Marc

[ZTEC]

Thank you for doing this and I am sorry you were forced to make this correction on your own, I hope Lotus Takes this seriously.

[achiu]

Nice Swift DB1. Is that yours?

Alex

[MitchT]

marcw,

It looks like you are running ra-1s on smaller wheels, any suspension upgrades too. Will lotus stick to the position that your car is getting close to race specs and the car is only designed for occasional track use not competition driving. Could the failure be due to loads beyound normal spec? Not siding with Lotus but I think we need to be unbaised, here. I want the problem solved too as it is disappointing that the car has a flaw of this nature.

[fishguy]

I was thinkning about this allot, as I use my car for non competivetive HPDE day level 3-4. how much are the toe end links (stock)? couldnt this be something that track guys just change out on a basis of preventitve maintenance. if we are going to be in there checking the torque before every HPDE day(which I am going to check every time out now). can the end link be marked say with tape or something to show where the old link was set up at for rear alignment purposes, and a new one just put in when inspecting this part warrants change out? I am a DYI guy, but by now means a pro mechanic. maybe my line of thinking is WAY off, and there is something i dont understand, but wouldnt this be an OK solution as well, keep 2 of these around, just like extra brake pads or something, and watch them just as closely as brake pads, or checking oil. it might be a pain, but so would hurting the car, or my body as a result of this part letting go on the track. I am putting this out there for other minds to ponder, and would like Stan to think about this solution(for me and other HPDE guys) as an alternative to this new design, which looks really good, but my car is not going to be able to go to this shop thats doing the install, as its hard enough for me to get away with my car outside AZ as it is. Just putting this out there for my own piece of mind, and as a way of continuing dialgoue on this subject, as it has been on my mind allot since i have become aware of the issue.
thanks for the valuable input in adavnce, hopefully no flames from this idea, or at least not from people who are using the cars for track days anyway.
Josh

[APOGEE]

I've never seen link failure on a street car that had not slid into a curb.
If you change your wheels, tires, spring rate, or shock rate and race prep your car, you may want to think about a Lotus rear link brace.We have them
in stock for around $ 700.00 bucks.( Newport European Motorcar L.T.D. )
Street car guy's with oem parts only need to worry about keeping their ride
serviced by the book.
All the best, Don

[Dro]

Originally Posted by fishguyAZ

I was thinkning about this allot, as I use my car for non competivetive HPDE day level 3-4. how much are the toe end links (stock)? couldnt this be something that track guys just change out on a basis of preventitve maintenance. if we are going to be in there checking the torque before every HPDE day(which I am going to check every time out now). can the end link be marked say with tape or something to show where the old link was set up at for rear alignment purposes, and a new one just put in when inspecting this part warrants change out? I am a DYI guy, but by now means a pro mechanic. maybe my line of thinking is WAY off, and there is something i dont understand, but wouldnt this be an OK solution as well, keep 2 of these around, just like extra brake pads or something, and watch them just as closely as brake pads, or checking oil. it might be a pain, but so would hurting the car, or my body as a result of this part letting go on the track. I am putting this out there for other minds to ponder, and would like Stan to think about this solution(for me and other HPDE guys) as an alternative to this new design, which looks really good, but my car is not going to be able to go to this shop thats doing the install, as its hard enough for me to get away with my car outside AZ as it is. Just putting this out there for my own piece of mind, and as a way of continuing dialgoue on this subject, as it has been on my mind allot since i have become aware of the issue.
thanks for the valuable input in adavnce, hopefully no flames from this idea, or at least not from people who are using the cars for track days anyway.
Josh

A set of the ball joints cost around $100. I recently picked up a pair.

[pixlpush]

Originally Posted by marcw

If you want this for your Elise, call Tom at 7's only racing at the Buttonwillow track: 661-764-5456 http://www.sevensonly.com/

Just a quick note on the Sevensonly shop at buttonwillow. They're great

They've helped me out more than a couple of times when I broke my miata at the track. Once My clutch slave cylinder went so I had a clutch that wouldn't disengage. they got it functional enough that I could make the drive back to LA over the grapevine and replace it at home. Another time the plastic top of my stock radiator split in half and I had to leave it with them while I ordered a new radiator. Both times they helped me out it was late in the afternoon on a Sunday. They pretty much droped everything to help me out. I was fully expecting to pay through the wazoo for race shop labor on a Sunday at the track, but they didn't take advantage of the situation at all, very fair prices both times. Anyway I've got nothing but good things to say about them.

[srwhitman]

Marc - did you try contacting Lotus Sport about this? Maybe it would be worth talking to Malcom? I recall seeing your car fork-lifted away and felt very sad. To read your story is worrisome. A few questions:

- others have pointed out MODS you may have done - how long have you had these modifications on the car?
- how many track days have you done with your car?

I am not a DIYer - I'm not sure what to tell a shop to look for on my car to see if its "going" to fail. It sounds like there is potentially slop but without taking it apart, how does one know? Even the "good" side on your car looks ok.

Scott

[marcw]

Originally Posted by Stan

…..What is the shop's opinion of what would happen in a wreck? Right now it seems that the stock toe link would fail before the items on either end of it……

Hey Stan: Give Tom a call at 661-764-5456. He is the guy that did this project and took the stock setup apart. Im not the technical person. You and he can probably discuss this issue in much greater detail than I could. My write-up is based on my conversation with him when I picked the car back up. Like pixlpush says, these guys do this for a living. They are a real honest to goodness race shop, they do not cater to the posuer street scene, or wannabes.

Originally Posted by achiu

Nice Swift DB1. Is that yours?

Alex

No, but the #50 yellow car up on the rack used to be mine. I sold it a few months ago. Those GT cars are bloody fast and fun not to mention pretty economical as far as running costs!! The Swift DB1 is faster of course! Speed costs $$$$. How fast do you want to go?

Originally Posted by MitchT

marcw,

…..Will lotus stick to the position that your car is getting close to race specs and the car is only designed for occasional track use not competition driving……..

More to point out this problem to them rather than to pursue a warranty claim (although the car did suffer minor rear clam damage from the wheel hitting the rear of the wheel arch, which I would not mind getting fixed) I have left 2 messages at the Lotus USA “customer service” line, plus an email to Lotus. No response from them I am sorry to say, and basically not worth the hassle at this point IMHO. As far as I am aware, there is nothing that the factory offers that would adequately address this problem based on what I have learned about the mode of failure, so a warranty repair on the toe link system really does not help me out.

Originally Posted by APOGEE

I've never seen link failure on a street car that had not slid into a curb…….

Agreed, let’s not start a panic here, this only seems to be a problem for the track guys amongst us.

[John Stimson]

I agree with Stan, in that a bolt through a hole is held in place by the clamping force between the head or washer and the surface around the hole, not by the walls of the hole. If the bolt has shiny spots from wearing against the sides of the hole, then at some point in time, the bolt wasn't tight enough. That doesn't tell you whether that's due to not being installed at the designed torque spec, or the torque spec in the design being too low.

[uglyduck]

Just a quick question. The 7/16 AN bolt is .0435 larger in diameter than the 10mm OEM unit. Was the hole that much over sized that one can stuff a 7/16 into it?
The other thing is, is the OEM track rod end thread metric, 10mm? If so,what kind of rod end where they using, as most often the hole in the rod end is the same diameter as the threaded shank. Wondering how they mixed metic and imperial threads and dim. Do you know the thread of the track bar,or the spec of the rod end they used?

[Bruce]

Originally Posted by John Stimson

I agree with Stan, in that a bolt through a hole is held in place by the clamping force between the head or washer and the surface around the hole, not by the walls of the hole. If the bolt has shiny spots from wearing against the sides of the hole, then at some point in time, the bolt wasn't tight enough. That doesn't tell you whether that's due to not being installed at the designed torque spec, or the torque spec in the design being too low.

Exactly. I can't agree with the conclusion noted in red that the bolt is going to move and wear regardless of being torqued correctly. The only way I can see that is if the torque spec is not high enough. Those wear marks look like it was loose to me.

[marcw]

Originally Posted by John Stimson

I agree with Stan, in that a bolt through a hole is held in place by the clamping force between the head or washer and the surface around the hole, not by the walls of the hole. If the bolt has shiny spots from wearing against the sides of the hole, then at some point in time, the bolt wasn't tight enough. That doesn't tell you whether that's due to not being installed at the designed torque spec, or the torque spec in the design being too low.

Guys…with all due respect, I am convinced otherwise. This point was made very clear to me by Tom, who is a guy who builds racecars…..from scratch…for a living. He was adamant that if there is play in that hole, the forces are adequate to make the bolt move slightly from side to side regardless of fastener torque, as evidenced by the shiny spots in the above pics. I personally am not an engineer, nor do I feel I have any sort of a background to argue this matter one way or the other, but I am very much more convinced to side with Tom based on both his expertise, and the physical evidence at hand.

Originally Posted by uglyduck

Just a quick question. The 7/16 AN bolt is .0435 larger in diameter than the 10mm OEM unit. Was the hole that much over sized that one can stuff a 7/16 into it?
The other thing is, is the OEM track rod end thread metric, 10mm? If so,what kind of rod end where they using, as most often the hole in the rod end is the same diameter as the threaded shank. Wondering how they mixed metic and imperial threads and dim. Do you know the thread of the track bar,or the spec of the rod end they used?

Call Tom for tech details if you are interested. There is no more metric hardware on my rear toe link assemblies. It’s all SAE now…the inside and the outside attachment points have been totally converted to new hardware. The tranny mount and A-arm pivots had to be re-machined to accept the larger 7/16 diameter hardware.

[Surferjer]

Marc-thanks for taking all the time to post this. Your fellow e-talkers thank you.

[zvezdah1]

Very informative marc, thanks!
Chris

[uglyduck]

Very cool.....that is what I hoped you would say.....swap it all out, not mix and match. Thanks for your updates...really appreciated.

[John Stimson]

Originally Posted by marcw

He was adamant that if there is play in that hole, the forces are adequate to make the bolt move slightly from side to side regardless of fastener torque, as evidenced by the shiny spots in the above pics.

The shiny spots are evidence that the bolt was moving within the hole. However, that shouldn't happen with a properly designed and tightened bolt. If the bolt was sliding around, then that implies one of two things:

1) The bolt was not tightened properly, at some point in time (it only has to happen once to make shiny spots)

2) The forces in reality were greater than Lotus designed for when they chose that fastener and torque specification.

The size of the hole does not determine whether the slippage occurs in the first place, just the distance that the bolt can slip. Reducing the play in the hole will reduce the impact if the bolt does slip, but a more effective solution would be to prevent the slippage from happening. It looks like your fabricator has addressed both issues.

[marcw]

Originally Posted by John Stimson

The shiny spots are evidence that the bolt was moving within the hole.......

The two pictures of the shiny spots are taken from opposite sides of the same bolt. It seems to me that the specific locations of the two discreet shiny spots indicate that the bolt was moving diagonally inside the "hole", which can only mean that the "hole" was too big for the bolt!

[Arno]

The problem is not that the hole is too big, but that the bolt moved in the first place, as it is suposed to be fixated by the clamping force of the bolt+nut.

Interesting to note that on the Toyota powered cars the torque for this connection is so much lower (50Nm) than on the Rover versions (70Nm).

Of course they use different subframes, which may be the root cause of this reduction, but of course this also reduces the clamping force provided by the whole setup significantly.

Not sure if the toyota powered cars use 10.9 grade nut and balljoint shank for this application. The rover ones do, therefore the M10 can handle the 70Nm torque.

Bye, Arno.