Special Stage Forums banner

Roll Cage Gurus - Less is More?

3K views 10 replies 6 participants last post by  John Sundelin 
#1 ·
Got a question for all you cage gurus... Can you actually over-build a cage? Now, I am not talking from a performance standpoint but rather, a safety standpoint. For example, vehciles are designed with crumple zones for a reason - to absorb the forces of impact. If you build a seam welded chasis with an incredibly strong cage, are you in fact putting the occupants at higher risk? I am thinking there is a point where the chasis is so strong, that all of the forces will be transmitted to the driver and that seems equally as bad as having an underdesigned cage. Anyway, looking to generate some discussion on this item as after seeing what my friend Travis has gone through, I want a cage/chasis combo that will maximize safety, even if I have to sacrifice some cost and/or performance.

Cheers, JC
 
#2 ·
Yes. Basically, think of yourself as the egg that you probably had to put in a box in high school physics and drop off the school roof. There is a certain maximum deceleration that a human body can take, and it depends on the direction that it occurs and the position of the body while it is experiencing it. From a safety standpoint, the ideal would be if you knew what that maximum deceleration was (and it's known to be different between different people, at least for the head) and you could put yourself in a cocoon that would keep you from decelerating at greater than that rate when you hit a point load of something approximating a tree at, say, 100 MPH. Obviously, that's not practicable, and even worse, cars are not isotropic, so the ideal cage for one type of collision may in fact be too rigid for another or too soft for another. The best you can do is compromise. Statistically, the things that have caused the most deaths in US rally are trees (by a long way), so it's best to design a cage that can take a point load in a direction that a tree is likely to come. In my experience, this is most likely the side, which is the hardest to deal with because there is so little crush distance, and there is also a pretty large span, especially for a two door car. Given the intrusion levels that even state-of-the-art cars see, it's probably not really possible to build a car that is too stiff from the side (assuming you can keep your head from getting separated from your spine). From the front, I would say it's most definitely possible. I suspect this is the reason that the FIA prohibits cages from extending beyond the suspension mounting points, as the points forward of that are usually used for the crumple zone. If a tree gets past the crumple zone, you might as well start trying harder to slow it down though, and hope that you don't exceed your personal deceleration limit.

--
John
 
#3 · (Edited)
JC,

The main safety function of the cage is to prevent collapse of the shell onto the occupants. Once that happens, you are in big trouble. The idea is to keep a protected space around you, and then use seats, harnesses, H&N, head side support, helmets, cage padding, etc., to take care of the rest. The recent FIA cage upgrades are to reinforce the basic cage structure to help it maintain it's shape in a wreck; e.g., the roof bracing and A-pillar supports are there to reinforce a heretofore weak spot: the upper A-pillar area.

Cage rules evolve, and the side impact work ongoing with FIA is a result of a death 1.5 years back in a side impact. This is certainly the hardest thing to achieve in a rally car since crush space is not there like in the front and the rear.

Build the cage to better than current standards, adding in some of the optional braces allowed. If you want better, then you need to look to an engineered cage; the specs are minimum standards. If you are building a car for which a homologated cage kit is available, you might look at that, but realize that sometimes these are built as much for light weight as for crach performance.

After the cage, you should put in as much of the head side supported seats as you can afford, and make sure the harnesses are properly installed. Beyond that, only some side impact cushioing can be added; this is a new area of research.

Finally, if you don't want to ever get hurt or killed in rally, then you have to abandon rally. No safety system will insure your injury-free participation in rally, just as you cannot be guaranteed to survive the next trip to the grocery store, whether in a car, train, or on foot. Racing has some inherent danger at all times, and rally is no exception. All you can do is minimize your odds of injury or death by implementing the best current knowledge of crash safety. THEN, always keep in mind that the one factor that you always control is how agrressively you drive. I can think of only one US rally competitor death where the speed or car control exercised by the driver was not a significant contributing factor.

Good luck and realize that the probabilities ARE very much in your favor to have a happy and injury free rally carrer. Regards,
Mark B.
(A student of race safety, who realizes that there is very little original thinking ever in race safety......)
 
#4 ·
There you have it - two completely different opinions.
Dare I say it, this is not the place where you will get definitive answers to that question, only anecdotal beliefs. (Mine being that cages are designed to bend when they reach a certain load. Deformation absorbs energy instead of passing it on to the occupants.)
The key is to find the balance between strength, compliance and weight.
 
#6 · (Edited)
Like John said, Keith, he and I seem to be on the same basic wavelength. If you extend the cage/tubing structure out far enough, you can allow some of that deformation. Refer to the NRS cage rules where we allow some added tubing front and rear, but where these added tubes have to be of lighter size or have prescibed bends in them to allow them to bend and deform, and not put a lot of energy into the cage. Or the newer FIA side impact proposals that add a layer of something on the sides to allow for deformation prior to an inner side layer that does not deform, and protects the occupants from intrusion

The basic cage structure as in FIA 253 has very little space for deformation; really only the rear backstay area could deform without intrusion upon the occupants. Plus, deformation has to be designed in to work the right way. Most basic cage rules sets are oriented for shop construction in the field for core cages that will not deform, partly because they are basic formulas for field construction, and partly because a sanctioning body has to do a lot of costly research if they are going to prescibe a desing that can deform but not cause injury, etc. (There was a lot of costly research that went into the Indy car tubs to eliminate lower leg destruction in crashes. ) It's pretty similar to designing a car for crash energy absorption; you are not going to do it without a good number of crash tests to back it up.

The tubing sizes are set to prevent real deformation up to a reasonable degree, and, from what I can tell, have been arrived at mostly via experience of what works, so that it does not bend too much i most wrecks, and bends so that it does not collapse and fail abruptly; the latter is also a function of material selection and tubing construction. In the relatively simple cage structures that we still use, forecasting how things bend is pretty darned impossible to do with all the different car sizes, weights, shapes, angles of impact, radius and surface hardness of objects impacted, etc., etc., to be able to make a general cage design work as a deformable cage....which I think is what John is saying. (Did we go to the same engineering school John? Cool!)

Finally, in examining several recently built cages after wrecks from mild to severe (as a rally scrutineer), the newer cages just don't defrom very much at all. If you want to consider that anecdotal, well, so be it. But, I do 100% agree with you in one way Keith: engineering cages to deform is not going to ba answered in a forum. That is why I mentioned engineered cage designs in my prior post.

Regards,
Mark B.
 
#5 · (Edited)
I don't think Mark and I have different opinions*, just two different ways of saying the same thing. Now, if Mark had said to weld a 6 inch I beam in place of your bumper, and tie it into the cage, then we would be saying something different...

--
John

*If a diversity of opinions is desired, it might help to have more than two guys who have the same Engineering degree from the same school!
 
#11 ·
In terms of ultimate tensile strength, this is correct (though DOM and CDS are really just two ways of making the tube out of a given material, such as 1020 Steel). Practically speaking, if the small difference between the ultimate strength of a DOM tube and a CDS tube makes a difference, you probably have much bigger problems. For instance, in the "Roll Cage Test" thread, you'll notice that while the cage suffered significant plastic deformation, the tube itself didn't break.

Also, T45 is only allowed in an engineered kit, since the alloying elements exceed the allowable level in the RA/NASA/FIA spec. Generally, these kits are designed to be much lighter than the equivalent steel kit, while providing roughly identical protection. In other words, if you picked up a piece of T45, you'd think it was exhaust tube. It shouldn't cause an experienced welder any real difficulties as long as he knows what rod/wire to use. The thing that I've always wondered about with T45 is whether it's easier to crush a tube with a very localized load than it is a regular tube. I actually have a piece of T45 in my garage, so maybe I'll get around to doing an experiment some time.

--
John
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top