What's new
By:
Filters

Roll Cage Failures

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

A shorter span is stiffer, but it also won't be able to deform as much (i.e. absorb as much energy). The lower plate gusset on the B-pillar initiated a premature buckling failure which pretty much defeats the benefit of having a short tube span anyway. The gussets at the top of the B-pillar concentrated the stresses at that point and caused the tube to completely rip out since it could not deform at all. Once the tube rips, it is doing nothing to help support the structure. Had the gussets had smooth transitions (a nice radius) into the tubes they tied into, they would have had the desired effect of stiffening the structure and removing the possibility of a single weld joint failure without causing stress concentrations that would promote tearing and premature buckling of the primary tube.

However, the primary issue at hand here is the poor cage design. There is no triangulation at all in the cage to resist bending and shifting fore and aft on the vehicle, and I don't think changes in gusset design would have substantially changed the outcome in this case. A secondary A-pillar tube and a kicker from the top of the B-pillar to the C-pillar mounting point at the tub would have made a world of difference. Here's a bad paint sketch to illustrate what I am talking about:

photo41184.jpeg


The black line is to represent the existing door bar that is hard to see in the picture and the red lines are where I am suggesting tubes should have been.
 
'84 Bronco II said:
I don't think changes in gusset design would have substantially changed the outcome in this case.
Click to expand...
Probably, the cage would still have collapsed without the gussets due to lack of triangulation. The tube would not have ripped in half but it would still have bent.
 
photo41181.jpg

fish mouth gusset.jpg

With the direction that the tube was forced in the rollover, Im not sure this gusset would have done any better. The force applied found the weak point and snapped it. Am I wrong in thinking that? Would love to add some more strength to my chassis with this type of gusset, I just don't see how you can plan for every scenario. Can anyone explain?
 
YeeP said:
photo41181.jpg

fish mouth gusset.jpg

With the direction that the tube was forced in the rollover, Im not sure this gusset would have done any better. The force applied found the weak point and snapped it. Am I wrong in thinking that? Would love to add some more strength to my chassis with this type of gusset, I just don't see how you can plan for every scenario. Can anyone explain?
Click to expand...


The idea with a curved gusset is to gradually change the section which reduces the stress concentrator at the end of the weld. I'm not sure the gusset you show is large enough to gradually change the section but it's better than the straight tubes in the cage that failed. The stress concentration factor reduction can be as high as 5X with the right design. The gusset also reduces the effective length of the tube which can reduce the likelyhood of buckling (L/D) and allow the straight section of the tubes to bend without failure.

In many cases I would use different radii gussets front and rear so they don't land at the same point on the tube. This would give you the lowest section change and the highest stress concentration factor reduction. I'd want the end of the tighter radius to land about midpoint in the straight section of the larger radius if possible.

I would probably make the gussets a thinner material than the base tube to allow some flexibility of the corner.

The idea with a roll cage is not to make it completely stiff but to have it absorb energy and not collapse on the driver. Energy absorbed in the cage is not trasferred to the driver which can lead to less driver injuries. Less of an issue with most rock crawling, but more of an issue with Ultra 4 and rock bouncers.
 
lt1yj said:
The idea with a roll cage is not to make it completely stiff but to have it absorb energy and not collapse on the driver. Energy absorbed in the cage is not trasferred to the driver which can lead to less driver injuries.
Click to expand...
This is pure bullshit for the kind of use any vehicle with a roll cage sees, especially as speeds get higher and the potential range of impact speeds gets higher.

A cage that deforms enough to keep your dumb ass from breaking ribs when you lawn dart off a sand dune at 50mph will crumple like it's not there if you lawn dart at 65mph or barrel roll off some rock ledges in a way that is less favorable to the cage.

You're basically advocating for using the cage as a single-use spring to absorb impact. And we all know how good springs are at absorbing impacts except in the most narrow of ranges.

I can understand that people who get their engineering knowledge exclusively from the soyboys on Reddit or some other cesspit of degeneracy may feel temptation to mimic automotive crumple zones but you have to understand that those primary exist to buy time for airbags to deploy and don't really do fuck all for the occupants except in a very narrow window of speeds and impact types.

If you're worried your cage is too strong buy a better seat and harness.
 
arse_sidewards said:
This is pure bullshit for the kind of use any vehicle with a roll cage sees, especially as speeds get higher and the potential range of impact speeds gets higher.

A cage that deforms enough to keep your dumb ass from breaking ribs when you lawn dart off a sand dune at 50mph will crumple like it's not there if you lawn dart at 65mph or barrel roll off some rock ledges in a way that is less favorable to the cage.

You're basically advocating for using the cage as a single-use spring to absorb impact. And we all know how good springs are at absorbing impacts except in the most narrow of ranges.

I can understand that people who get their engineering knowledge exclusively from the soyboys on Reddit or some other cesspit of degeneracy may feel temptation to mimic automotive crumple zones but you have to understand that those primary exist to buy time for airbags to deploy and don't really do fuck all for the occupants except in a very narrow window of speeds and impact types.

If you're worried your cage is too strong buy a better seat and harness.
Click to expand...


Yup, I agree 100%
 
I rolled my SxS 1.5 times last weekend and was hit by another car in mid air and my chassis/cage is still as straight as an arrow.... :smokin:

The aluminum bumpers act as my "crumple zones" to absorb impact but not transfer it to the chassis.


IMG_1224.jpg


20211024_102038.jpg


20211024_102046.jpg


20211024_102136.jpg


20211023_133543.jpg
 
Last edited:
I'll add that there's nothing wrong with designing the cage so that when it bends it doesn't instantly start tearing but sacrificing ultimate strength so that it can deflect nicely on the way there is beyond stupid. That's basically the old "grade 5 vs grade 8 for recovery points" argument. The rigid cage might not deflect as far before failing but the force to get to complete failure is going to be greater.

Think about it like this. You're building a table with four pipe legs. Which is going to take more side load before the table winds up on the floor, saddle gussets that go 3" down each leg or mitered tube that goes 6"? Of course the 6" mitered tube is stronger, but if you are really worried about bending the tube proper triangulation is the only answer.
 
SLOWPOKE693 said:
I




20211023_133543.jpg
Click to expand...

That vertical that ties into the A-pillar and roof node is key. If the bent cage pictured above had that (or triangulation going to the B pillar and roof node) the B pillar never would have failed because it would not have been subject to the fore/aft forces. Those forces would have been compression/tension forces in the A pillar and other tubes going into that node.
 
cage failure.png
I have 0 expertise in this matter except what has come from magazines and Irate. I thought the bigger problem was the lack of gussets and elsewhere and where I circled in red. Doesn't this spot focus all the force that cage recieves into a point causing it to pinch thus removing all its resistance to bending. I think we are focusing on the break but to me it looks like it only broke where it did because it was the stiff spot. I think it shows more that you can't just put triangles in one spot and expect it to be enough for the whole cage.
 
arse_sidewards said:
This is pure bullshit for the kind of use any vehicle with a roll cage sees, especially as speeds get higher and the potential range of impact speeds gets higher.

A cage that deforms enough to keep your dumb ass from breaking ribs when you lawn dart off a sand dune at 50mph will crumple like it's not there if you lawn dart at 65mph or barrel roll off some rock ledges in a way that is less favorable to the cage.

You're basically advocating for using the cage as a single-use spring to absorb impact. And we all know how good springs are at absorbing impacts except in the most narrow of ranges.

I can understand that people who get their engineering knowledge exclusively from the soyboys on Reddit or some other cesspit of degeneracy may feel temptation to mimic automotive crumple zones but you have to understand that those primary exist to buy time for airbags to deploy and don't really do fuck all for the occupants except in a very narrow window of speeds and impact types.

If you're worried your cage is too strong buy a better seat and harness.
Click to expand...

You're 100% right and wrong at the same time.

I just finished a year long project to develop ROPS (roll over protective system) subcomponents for a large offhighway truck to exceed SAE requirements for force-deflection (energy absorption). I've done several ROPS/FOPS development projects including 2 post, 4 post and 6 post ROPS for the earthmoving equipment industry.

These projects usually take 6-12 months and have 4 or 5 engineers and 100's of thousands in analysis software. Something the average offroader does not have access to.

In every case the energy absoption and the operator space claim are the key metrics we achieve. There are requirement for maximum plastic deformation so hinges that cause tearing don't form among many other less important requirements. Operator safety is the ultimate goal.

So in the sense that offroaders don't have access to these tools and that the average Joe adds additional stiffness to ensure the cage doesn't collapse and drop on their head you're 100% correct. For most of us that recreational wheel we just need something that doesn't fold like the one shown above. We don't need crumple zones as you called it.

And you're 100% correct that the system would need to be designed for the worst case scenario at top speed and with the highest impact load. That's exactly where energy absorption is needed. There is only so much a seat and harness system can do to protect the driver at high speeds. In the slower rolls there would be little or no damage.

25 years ago we were mostly slow speed rock crawlers with a few into sand and mud on the old board. When guys started competing for titles they really pushed the sport to another level. KOH was the real driver for significant change. Real money started to flow into the sport to make cars faster and more reliable. Safety has gotten better as well. With how fast KOH and Ultra 4 cars are going now I expect changes to protect drivers.

For the normal offroader just add more triangulation.

BTW I got my degrees from a cracker jack box. None of this Reddit for me:flipoff2:
 
dentedsub said:
cage failure.png
I have 0 expertise in this matter except what has come from magazines and Irate. I thought the bigger problem was the lack of gussets and elsewhere and where I circled in red. Doesn't this spot focus all the force that cage recieves into a point causing it to pinch thus removing all its resistance to bending. I think we are focusing on the break but to me it looks like it only broke where it did because it was the stiff spot. I think it shows more that you can't just put triangles in one spot and expect it to be enough for the whole cage.
Click to expand...
You are 100% correct.

The area you circled caused a plastic hinge that caused the tube wall to buckle. The whole cage folded back and the top of the tube tore at the stiffness transition. If curved gussets were added in front of and behind the area you circled as well as behind the A pillar the cage probably would not have failed as spectacularly.
 
There's a reason that NASCAR requires mild steel and not cromo for the cages in cup cars. It's written in the rule book.

The extra rigidity that comes with cromo will transfer the force of an impact into the driver moreso than the exact same designed cage in mild steel. They want the cage to absorb as much of the impact as it can without failure.
Nobody can say that CUP cars don't take big hits. It's amazing the hits they can take and the driver just walks away.

Point is that cages do absorb impacts. It doesn't matter if everything around the driver compartment is destroyed. What you do want, is that the driver compartment remains intact.
 
lt1yj said:
You are 100% correct.

The area you circled caused a plastic hinge that caused the tube wall to buckle. The whole cage folded back and the top of the tube tore at the stiffness transition. If curved gussets were added in front of and behind the area you circled as well as behind the A pillar the cage probably would not have failed as spectacularly.
Click to expand...

It's less about the type of gussets used and more about the lack of triangulation that allowed the roof to move in the first place. If you chuck a piece of tube up in your bender, it doesn't matter if you weld gussets to the ends or not - when you hit the switch it will bend just the same as if there were no gussets at all. No amount of gussets or curved vs. straight vs. plate is going to fix a poor design.

Try to picture every joint or node as a hinge - and how to eliminate range of motion in different directions. An example for the failure above could be a diagonal tube going from top of b pillar to 'midway' point of c pillar. If every joint of your node or cage was a hinge or joint, it should still be rock solid and unable to be moved if designed well. If every joint or node in the example above was a hinge, the roof would be easily able to fold back and down just like it did in the rollover because of the lack of triangulation preventing it from doing so.

It's like setting up a link suspension. Even though there are heims on the ends of your links, you can triangulate to eliminate different planes of motion until the axle travels along the path you want it to. Same with a cage construction:beer:
 
I agree, the whole problem with that particular failure is a lack of triangulation, it wouldn't have mattered what type of gusset you did or didn't have. There's a reason every road racing body requires a diagonal from the top of the main hoop, down to the rear frame.
If you want to know everything there is about gusset stress concentration, look up "bicycle head tube gusset construction". It's a very highly stressed joint in thin wall tubing that will definitely find any design flaws.
 
Putting my flame helmet on..... My buddy and I built our first cage in 2009 and didn't really get it right. I did the back half of my jeep out of 2" DOM and ran out on the back so I bought 1.5" schd 40 pipe to finish the rest. Didn't get enough gusseting and the cage was a 1 and done.

Flash forward 2 years later.
We were at Gilmer and the girl I was dating at the time wanted to know what it was like to know what it was like to be in a roll over and I had a couple soft flops in the jeep so my confidence in the cage was high. So on Sunday cruise (a twisty trail) I crab walked the jeep up the wall until it fell over. Turned out the jeep went higher than I anticipated and fell in a hole that was also deeper than I anticipated, making it a lot harder hit than expected. It rattled my at the time girl friend and I was pretty good but I could tell pretty quick I bent some shit....

After the gilmer trip I had to cut the entire front portion of the cage off and at that point I ordered 2" DOM to finish it like I should have in the beginning and also made the cage much stronger with more cross pieces and lots of extra tubing added. The jeep took good rolls on the new cage and aside from some dented pipe never moved an inch.

You can tell by the second pic that the front of the cage has shifted pretty hard and was at this point toast!

189564_204021149610447_914474_n.jpg


197848_204021162943779_2735670_n.jpg


376818_307743729238188_1444090955_n.jpg


255005_529481207064438_217115120_n.jpg
 
Seems like everyone is on the same page with the Fury jeep. It's funny how easy the red lines in the above Pic would have been to add with very little weight difference.

Gussets only shorten the amount of unsupported tube, sure they help, but it's not game changing.

As far as Rusty giant jeep and lack of driving skill causing a roll on a non obstacle :flipoff2: never mind, I actually read the post :laughing: did it all for the nookie right? :flipoff2:

The roof x only ties the a and b pillars together. It helps, but doesn't do a whole ton of support for the type of incident pictured. What you want is a b pillar x, especially with exos. They inherently weak by being larger and typically mounted to a less sturdy structure like the sliders. I like to do an x down to the frame. It also helps my piece of mind that in a really hard endo that the cage won't just push the rocksliders off the frame and crush you.

I've had 3 exos I built on heavy 1 ton Toyotas take hard endos off man made large culvert pipe obstacles and hold up really well.

Found some good pics of 1 :laughing:


CalROCSSaturdayOroville477.jpg


Bed took most of it on that one and folded like you'd expect. Same truck went off the big pyramid of culverts and went straight to the cab, as well as a 2nd similar truck. Both took damage to the a pillars, but the b and c pillars were intact. They both got complements on the cage holding up well. The other truck went pass rear over driver front off an 8' culvert. It was a single cab, and took no damage.

What's funny is that all those cages were 1 1/4 schd 40 :laughing:

So really, there are better and worse materials obviously, but design trumps all imo.
 
Last edited:
dentedsub said:
cage failure.png
I have 0 expertise in this matter except what has come from magazines and Irate. I thought the bigger problem was the lack of gussets and elsewhere and where I circled in red. Doesn't this spot focus all the force that cage recieves into a point causing it to pinch thus removing all its resistance to bending. I think we are focusing on the break but to me it looks like it only broke where it did because it was the stiff spot. I think it shows more that you can't just put triangles in one spot and expect it to be enough for the whole cage.
Click to expand...
One thing missed is the gusset is perpendicular to the tube when it should be tangential if it’s a plate gusset. You can see where it folded strait in on the end of the gusset. Better cage design would obviously have helped more but plate gussets can do more harm than good if not put in the proper location. That why I am a huge fan of taco gussets.
 
YJ-6 Unlimited said:
One thing missed is the gusset is perpendicular to the tube when it should be tangential if it’s a plate gusset. You can see where it folded strait in on the end of the gusset. Better cage design would obviously have helped more but plate gussets can do more harm than good if not put in the proper location. That why I am a huge fan of taco gussets.
Click to expand...
Either way you're splitting hairs. Better gussets may have let it collapse 1" less, it's still a failure imo. Actual diagonal braces like what 84 bronco posted are the only thing that would have been a major difference.
 
YotaAtieToo said:
Either way you're splitting hairs. Better gussets may have let it collapse 1" less, it's still a failure imo. Actual diagonal braces like what 84 bronco posted are the only thing that would have been a major difference.
Click to expand...
Correct, like I said better design would be best, but the point was plate gussets should never be perpendicular to the centerline of the tube. All that causes them to do is crease and fold.
 
YJ-6 Unlimited said:
Correct, like I said better design would be best, but the point was plate gussets should never be perpendicular to the centerline of the tube. All that causes them to do is crease and fold.
Click to expand...
Definitely.

If I feel like doing a plate gusset I will put a ~1" 90 on the end, or add a 1" tube. Taco is better also.
 
YotaAtieToo said:
The roof x only ties the a and b pillars together. It helps, but doesn't do a whole ton of support for the type of incident pictured. What you want is a b pillar x, especially with exos. They inherently weak by being larger and typically mounted to a less sturdy structure like the sliders. I like to do an x down to the frame. It also helps my piece of mind that in a really hard endo that the cage won't just push the rocksliders off the frame and crush you.

I've had 3 exos I built on heavy 1 ton Toyotas take hard endos off man made large culvert pipe obstacles and hold up really well.

Found some good pics of 1 :laughing:




So really, there are better and worse materials obviously, but design trumps all imo.
Click to expand...


Agreed.... The saving grace of my cage was the original factory hoop behind the seats. It had a really solid x brace in it and we cut holed through the tub down to the frame for the x brace. On the original cage I didn't tie it to the hoop, but on the second revision of the cage I tied it to the hoop in 4 differen't spots on the b-pillar. So then we figured with the b-pillar being rock solid that the x on the roof and all the corner braces would help the A-pillar from moving at all.
 
SLOWPOKE693 said:
I rolled my SxS 1.5 times last weekend and was hit by another car in mid air and my chassis/cage is still as straight as an arrow.... :smokin:

The aluminum bumpers act as my "crumple zones" to absorb impact but not transfer it to the chassis.






2
Click to expand...
I'd not be so quick to post this. I rolled a well built jeep cage one time. It "looked" as straight as an arrow also. ...
But I thought to myself, perhaps it's time for a newer cage with better reinforcements. So I used a sawzall to remove the 'old' cage and it sprung about 3 or 4 inches once the stress were removed with the saw. ...
 
You must log in or register to reply here.
Top Back Refresh