Check your toe link!

[TimMullen]

Originally Posted by turbophil

This is a metal on metal item that doesn't "settle in" over time like pistons rings do for instance...

Unfortunately, sometimes the OEM toe-link ball joint does "settle in" - there is sometimes too much galvanizing in the "groove" that the ball joint fits into. Even with proper torque, the joint can loosen a bit over time. If you torque your toe-link ball joints and find them loose (at least if more than once), then you should probably remove the toe-link ball joint and clean out the extra galvanizing from the groove so it will fit better and not "settle in".

If your toe-link ball joint remains tight, it probably doesn't have the "too much" galvanizing in on the sub-frame groove.

There is a thread around here somewhere with a good explanation and some photos of the problem..

Note that not all cars have this problem - only some of them...

[turbophil]

Originally Posted by davecarama

FYI, in case you don't want to look, don't have a Nm wrench, whatever:
60 Nm = 44.253728956681634 ft#

My tq wrench only goes to tenths though

[Keeper]

There goes your warranty.

[fpitas]

Originally Posted by turbophil

Ya, pretty sure Lotus doesn't have the best practices for ensuring proper tq at the factory...

Tell me about it. My front lower ball joints were torqued to *maybe* 10 ft-lbs. I doubt they loosened up with use.

[EricH]

Am I missing something? Is using lock-tite not an option here?

[turbophil]

Better to use nyloc nuts, IMO, and check the torque when you do your oil changes (provided you don't track the car). The panel is off and it's easy to do then...It's not a great application for a thread locker, IMO... or just replace the system with more robust hardware and basically forget about it...

TP

[ChrisH]

Originally Posted by EricH

Am I missing something? Is using lock-tite not an option here?

No, loc-tite isn't an option. In the case of the galvanizing compressing (mine had this), it isn't the nut coming loose that is the problem. The material between the head of the stud and the nut is compressing, creating a loose bolt. You have to tighten it each service interval (7500 miles) to keep up with the compression. Eventually, it fully compresses, but I still check the torque every service interval even after over 4 years and 52,000 miles.

[turbophil]

Replacing some toe-links today on a client's car with ours... This car sees a handful of track events a year... has about 40K miles...

I was a bit surprised when I saw these. Passenger side was OK. Drivers sid....um, not so much. The stud is worn down by 18% all the way around

Nicer yet, is the massive indentation it left in the flange surface on the frame. The pictures don't capture the depth of the indention very well, but it's pretty amazing in person. One more track event would have been the end of the line for this lotus. Fortunately, we caught it when we did...

Anyway, thought I'd add a couple more scary pics to the thread...

Best,

Phil

[wdonovan]

The stud doesn't look worn down but rather necked down from going past yield. This is a classic "reduction in area" used in material testing and specifications. It is due to "elongation" when a specimen passes yield point. As the area of this bolt decreases due to this phenomenon, the per-area load goes up and leads to tensile failure. The depression in the washer shows that the stud was bending back and forth. The preload that Lotus recommends (including increasing the specified torque value) is to maintain a clmap load that always overcomes the tensile load cause by side load on the ball joint. If this side load ever produces a force between the stud and the washer that is higher than the preload, the stud moves and the load is concentrated at exactly the area where the breakage occurs. This slight movement causes a condition where the stud will eventually fail from metal fatigue. There was a photo on one of these threads that showed the fracture. It was a classic cyclical fatigue failure. Lotus' claim of soft galvanized coatings that could brinell certainly would have contributed to the situation but only if the preload was not high enough originally to overcome cornering loads. If preload was high enough, there would never been enough movement to allow the stud to bend, which is the actualk cause of the brinelling, not simply the stud sinking into the coating like in quicksand. The photo above is amazing in that it shows the part on the brink of tensile failure. I work in failure analysis a lot and love to see a part that almost failed but didn't. So much easier to analyze failure when you also have unfailed samples.

[shinoo]

Originally Posted by wdonovan

The stud doesn't look worn down but rather necked down from going past yield. This is a classic "reduction in area" used in material testing and specifications. It is due to "elongation" when a specimen passes yield point. As the area of this bolt decreases due to this phenomenon, the per-area load goes up and leads to tensile failure. The depression in the washer shows that the stud was bending back and forth. The preload that Lotus recommends (including increasing the specified torque value) is to maintain a clmap load that always overcomes the tensile load cause by side load on the ball joint. If this side load ever produces a force between the stud and the washer that is higher than the preload, the stud moves and the load is concentrated at exactly the area where the breakage occurs. This slight movement causes a condition where the stud will eventually fail from metal fatigue. There was a photo on one of these threads that showed the fracture. It was a classic cyclical fatigue failure. Lotus' claim of soft galvanized coatings that could brinell certainly would have contributed to the situation but only if the preload was not high enough originally to overcome cornering loads. If preload was high enough, there would never been enough movement to allow the stud to bend, which is the actualk cause of the brinelling, not simply the stud sinking into the coating like in quicksand. The photo above is amazing in that it shows the part on the brink of tensile failure. I work in failure analysis a lot and love to see a part that almost failed but didn't. So much easier to analyze failure when you also have unfailed samples.

Excellent and thorough!

Nitron are releasing a super trick toe link upgrade. Even lighter and with finer adjustment. We'll have them soon.


Our RTVbrace and RTDbrace remain safe, cost effective, & proven solutions.

[apk919]

Originally Posted by shinoo

Excellent and thorough!

Nitron are releasing a super trick toe link upgrade. Even lighter and with finer adjustment. We'll have them soon.


Our RTVbrace and RTDbrace remain safe, cost effective, & proven solutions.

Shinoo, it looks like the inner joint is mounted to a new brace that's bolted to the subframe... it looks like the subframe needs to be drilled to accept the bolts? I do like the idea of replacing the outer joint...

[turbophil]

Good call

The stud is about .04 -.05 longer than it's supposed to be... So you're right, there's a good bit of plastic deformation. Judging on the wear pattern of the hole (before we drilled it out to AN7) and stud currently, it's evident that a good portion of the stud was simply worn down as well... Nevertheless, I overlooked the most obvious.

Thanks for your input, sir!

Phil

Originally Posted by wdonovan

The stud doesn't look worn down but rather necked down from going past yield. This is a classic "reduction in area" used in material testing and specifications. It is due to "elongation" when a specimen passes yield point. As the area of this bolt decreases due to this phenomenon, the per-area load goes up and leads to tensile failure. The depression in the washer shows that the stud was bending back and forth. The preload that Lotus recommends (including increasing the specified torque value) is to maintain a clmap load that always overcomes the tensile load cause by side load on the ball joint. If this side load ever produces a force between the stud and the washer that is higher than the preload, the stud moves and the load is concentrated at exactly the area where the breakage occurs. This slight movement causes a condition where the stud will eventually fail from metal fatigue. There was a photo on one of these threads that showed the fracture. It was a classic cyclical fatigue failure. Lotus' claim of soft galvanized coatings that could brinell certainly would have contributed to the situation but only if the preload was not high enough originally to overcome cornering loads. If preload was high enough, there would never been enough movement to allow the stud to bend, which is the actualk cause of the brinelling, not simply the stud sinking into the coating like in quicksand. The photo above is amazing in that it shows the part on the brink of tensile failure. I work in failure analysis a lot and love to see a part that almost failed but didn't. So much easier to analyze failure when you also have unfailed samples.

[EricH]

Point of reference, at 5500 miles my '08 Exige's toe link bolts needed tightening to bring them to the torque specs.