Roll Cage Failures

[Stephen Wilson]

B-pillar bracing would have done quite a bit to prevent that, correct?

Yup, I can't believe that's a race legal roll cage. It wouldn't pass SCCA road racing specs without a rear diagonal support for the main hoop (B pillar) .That rear support and an FIA bar of appropriate size would have completely prevented the collapse.

 

[gt1guy]

Forgot I had these, friend of ours from so-cal. Heavy rig due to class rules. High speed roll during qualifying at Ridgecrest a couple years ago. Albert got a bad concussion and his neck seriously hurt because the cage folded in and hit him in the helmet.

He's lucky he's not dead right now.

Maybe somebody can help me understand what I can't seem to understand. How the fuck does a cage like that, pass tech and be allowed to race? Do these offroad sanctioning bodies even have a rule book? Are there specs for how a roll cage should be designed, determined by class/vehicle weight? NASCAR, NASA, NHRA, SCCA, you name it, all have very specific rule books that lay it all out in black and white. There is literally no thinking involved, it tells you exactly how the cage SHALL be built.

I can fully understand the goober with the mud truck building a stupid cage at home that doesn't work. But for a professional sanctioning body running races, there is zero acceptable reasons why a cage with so many design flaws is allowed on track.

 

[JR4X]

I think it’s an issue of grandfathering older chassis’s while all the time things getting faster. That cage wouldn’t pass tech today but it did in 2012. On page 2 my failed eco caged samurai was an 1 1/2” cage. I probably should have decided on my own to take it out of service before I did. But I honestly didn’t even think about the cage when I did a bunch of stuff to make it faster. I had rolled it before and the cage held up fine. Then for the last year it had enough horsepower and shock tuning that all of a sudden the car is going faster than the cage was safe for.

 

[Pyleit]

Poison spyder bolt in cage.

 

[Donk]

Not sure if this is helpful but it's the current version of the UK MSA Bluebook Section K which covers basic cages, seat mounts etc. https://www.motorsportuk.org/wp-content/uploads/2019/12/155-195-Safety-K.pdf Certain series (U4 is one I believe) do go beyond the minimum requirements as laid out here.

 

[CarterKraft]

Poison spyder bolt in cage.

Still a success no?

 

[Pyleit]

Still a success no?

I would say yes. Driver was unharmed and cage will be replaced.

 

[YJ-6 Unlimited]

Any pictures of the overall cage design? Also, what size and material tubing? Any idea what the rig weighs?

Looks like it held up great, I only ask all these questions to provide more tech for those in the process of designing their own cages.

1.75x.120DOM. I would say around 5k lbs. here is some pics.

 

[DannyK]

NOT mine. Downloaded MANY years ago off the web.

 

['84 Bronco II]

Great example of a nearly complete lack of lateral triangulation!

 

[Brad]

NOT mine. Downloaded MANY years ago off the web.

I remember that picture

 

[dnsfailure]

I'm seeing several cages that cracked the tube right below where it was reinforced/gusseted, the heat affected tube, not the weld itself.

Is there any way to avoid this type of welding heat related failure? Or any way to tell if you've over heated areas to the point where it might become prone to cracking?



I want more cage related stuff on here, I miss some of the old PBB cage threads.

 

['84 Bronco II]

I'm seeing several cages that cracked the tube right below where it was reinforced/gusseted, the heat affected tube, not the weld itself.

Is there any way to avoid this type of welding heat related failure? Or any way to tell if you've over heated areas to the point where it might become prone to cracking?



I want more cage related stuff on here, I miss some of the old PBB cage threads.

I don't think what you are describing is due to the heat affected zones around the weld, but rather stress risers created by terminating tubes/gussets in the mid span of another tube. It seems like a lot of the gusset tubes people add to their cages may be doing more harm than good.

 

[bdkw1]

I don't think what you are describing is due to the heat affected zones around the weld, but rather stress risers created by terminating tubes/gussets in the mid span of another tube. It seems like a lot of the gusset tubes people add to their cages may be doing more harm than good.

Long ago they would do the bent piece of 1" that had large notch's on the ends. This gave some flex and large attachment area to spread out the load. Never saw one of those crack the main tube.

 

[gt1guy]

Taco gussets help spread out the load on a joint and reduce the amount of stress that the tube weld sees.

The most important thing you can do starts in the design phase, by having tubes only see compression or tension.


Taco gussets

 

[bdkw1]

Those create the same issues as the large straight tube gussets. The sudden reduction in cross section with a stress riser is just a crack waiting to happen.

 

[gt1guy]

Long ago they would do the bent piece of 1" that had large notch's on the ends. This gave some flex and large attachment area to spread out the load. Never saw one of those crack the main tube.

Those create the same issues as the large straight tube gussets. The sudden reduction in cross section with a stress riser is just a crack waiting to happen.

Is this the type of gusset you were talking about?

 

['84 Bronco II]

Is this the type of gusset you were talking about?

That is a much better gusset design from a stress concentration perspective, but also more work to fabricate than a plate, taco, or straight tube gusset.

 

[arse_sidewards]

Those create the same issues as the large straight tube gussets. The sudden reduction in cross section with a stress riser is just a crack waiting to happen.

That's why we design shit to only be in compression and tension. The tube should never see enough cyclical bending force to have issues with cracking.

Every cage ever will have issues if you start applying bending forces to things. They just aren't meant to take those forces. Heck, the maximum diameter is typically in the 2" ballpark. Even Toyota doesn't feel comfortable going below 3.x" web for the tail end of a truck frame and lord knows that will never see even a fraction of the load of a off road vehicle.

If you want a cage member to take bending forces you need to double the tube up and do a truss structure (as seen in the B-pillars on some race cars, though even those have proper triangulation to help them out) in order to get the effective height of the truss to something that can resist the forces you will apply without moving too much.

 

['84 Bronco II]

The way the main tube behind the driver ripped open and buckled in the pictures JR4X posted on page 2 was largely caused by the gusset design creating a stress concentration at that point.

EDIT: Not sure why the pictures I quoted aren't showing up correctly, but if you click on the attachment hyperlinks you can see what I am talking about

 

[arse_sidewards]

The way the main tube behind the driver ripped open and buckled in the pictures JR4X posted on page 2 was largely caused by the gusset design creating a stress concentration at that point.

EDIT: Not sure why the pictures I quoted aren't showing up correctly, but if you click on the attachment hyperlinks you can see what I am talking about

But what would have happened hat those gussets not been there? Unsupported tube is unsupported tube. A gusset just makes the effective span shorter.

 

[gt1guy]

But what would have happened hat those gussets not been there? Unsupported tube is unsupported tube. A gusset just makes the effective span shorter.

Which adds to it's resistance to bending.


If you look at the buckle at the bottom of the B-pillar, there's a single flat plate gusset on the rear side of the tube. I'd bet money that's what went first. With the direction the pillar was being pushed, that gusset would have aided in the buckling.

 

[bdkw1]

Is this the type of gusset you were talking about?

Similar, but with 1" tube instead of same size as the cage.

Saying that the cage is built so that the tubes are only in compression or tension is naïve at best. Everything flex's, chassis twist, a lot. Accident's don't give a shit if something was designed in tension or compression.

 

[gt1guy]

Similar, but with 1" tube instead of same size as the cage.

Saying that the cage is built so that the tubes are only in compression or tension is naïve at best. Everything flex's, chassis twist, a lot. Accident's don't give a shit if something was designed in tension or compression.

Accidents don't give a shit if someone puts a bend in their B-pillar x-brace. But when the whole thing folds down on the driver..............they sure a fuck will.

 

[bdkw1]

Accidents don't give a shit if someone puts a bend in their B-pillar x-brace. But when the whole thing folds down on the driver..............they sure a fuck will.

But it will keep the tube from shearing like the one pictured above. Even rally cars with their fancy form sheetmetal gussets suffer from design issues. Notice the large gusset and roof bar meeting in the same spot on the outer roof bar. Those tubes should have been run up to the A pillar windshield bar intersection.

 

[gt1guy]

But it will keep the tube from shearing like the one pictured above. Even rally cars with their fancy form sheetmetal gussets suffer from design issues. Notice the large gusset and roof bar meeting in the same spot on the outer roof bar. Those tubes should have been run up to the A pillar windshield bar intersection.

I'd like to see a side view of that.

 

[SLOWPOKE693]

I'd like to see a side view of that.

Me too. I'm trying to wrap my brain around what's going on in the roof area.

 

[bdkw1]

The top tubes were raised like that to clear the duct work going to the radiators/engine. This was one of the 2nd gen Tuarags.

 

[SLOWPOKE693]

The top tubes were raised like that to clear the duct work going to the radiators/engine. This was one of the 2nd gen Tuarags.

So that's a finished chassis? Car has zero overhead triangulation in the drivers compartment? Nothing preventing the top of the A pillar from being pushed inward?

That's kinda fucked, especially when you factor in the amount of protection built around the scoop area.

 

[dnsfailure]

I like those big gussets up top!

what sort of racing is that chassis for?