I have started building roll cages. I have all the proper tools, I am a professional Tig welder by trade and have been a car guy since I can remember. I am, however, not an engineer. My problem is that I don't want to use people as guinea pigs and test my skills. I need some definitive answers in regards to cage building. I have read so many articles on this site, and many others in regards to cage design, that I'm even more confused. I have about 5 cages lined up to build, but if I can't come up with some hard design facts, I'm going to pass on building the cages until I do. I have read the info of many sanctioning bodies, but they tell you what you can do, not neccesarily whats best to do. How do I find this info, I want to build the best, safest, and strongest cages possible, but I need some hard facts, where do I find it???
This was my first try: http://www.redpepperracing.com/gallery/cage_build

Thanks

 

You are definitely heading in the right direction. Take a very close look at what KIRK Racing has for a cage. Their weld-in type is very strong. We need to see what you are doing for the Halo and forward downtubes too. The mounting points for the main hoop should be close to the rocker panels and the kickup (CRX anyway) behind the seats. You need to consider the total surface area of the mounting plate along with possibly providing a vertical element on said plate. There are several excellent designs floating around, including the ones in the GCR. One of your greatest challenges will be using a mandrel bender and a notcher. As to locating the rear down tubes-there are several ideas on that one too. We happen to like the way the KIRK cage is installed in the CRX, others prefer the Autopower arrangement. Which one is stronger-I wonder if someone out there does. Final thought-take a look around the SAE website; may just find some additional ideas.

 

you're best bet is probably just try to look at as many other designs as possible. Either in person, or in pictures.. If you can determine how well each functions, you're on your way..

i doubt a lot of cage builders are going to want to just put some of their proprietary R&D up for all to read though..

 

Good point about guys not giving up their info, I guess I never looked at it that way. So no books on general chassis design then? Alright, I guess I'll have to start fooling around with my own cars and see what happens

Thanks

D.

 

for whatever it's worth here are some pictures of the cage that was designed for my Civic. http://switch.uwtc.washington....1.htm

 

That cage is awesome!!!!! Why does it look so shiny? Welds are gorgeous and I really like the rear box on the shock tower. My next question involves attaching gussets to the A and B pillars. Is it necessary? What about tubular gussets at certain parts of the cage? Who built that BTW?

D.

 

My cage, which I'm thrilled with, can be seen here:

http://www.nogumracing.com/karl/integra/rollcage/

Sorry the picures aren't great, I don't have a great digicam.

 

One of the best chassis design books that I ever read was an old "How to Build Stock Cars" thing. I don't even know if I still have it. Some standard rules, however, are...

** The centerlines of tubes should come together at one point, where multiple tubes are joined. You appear to have done that in your example.

** Where possible - and assuming that weight is not an issue - triangulate any two- or three-dimensional bay (planar rectangle or rectangular prism) but running a tube to opposing corners.

** Picture your structures as having NO rigidity at the corners, as if they were held together with swiveling joints. This reinforces the understanding that elements must be loaded in compression or tension - which is what triangulation does.

** No element should be loaded in bending, by the way other elements attach to it.

** Visualize load paths in any structure, considering what will get pushed where if a force is applied on a particular point of a structure. These might be chassis dynamics loading forces or those resulting from crash accelerations but they should - again - be taken in compression or tension rather than by bending elements or "lozenging" (distorting) a bay.

** Devise ways to get welds completed all around tubes while getting the elements themselves as close to the chassis as possible (e.g., A-pillars, roofline, etc.)

** Try modeling some structures out of 3/16" wood dowel and hot glue. You can learn a LOT with 1/10th scale models out of these materials.

Have fun and keep asking the good questions.

Kirk

 

My barZ..... http://members7.clubphoto.com/...phtml

Built by Russ at http://www.racetechfab.com

 

Would the stock car book happen to be written by Steve Smith? Thanks for all the tips, I think I understand most of it. Am I assume when you talk about an "element" you mean tubing? I don't quite understand what you mean here.."** No element should be loaded in bending, by the way other elements attach to it." Everything else makes sense. Now what about plate gussets to the A and B pillars? Are they needed? Should I make "boxes" at the shock towers and the areas where tubing is welded at a right angle? Thanks for all the info guys, tis is what I am looking for.

D.

 

for the love of god avoid "x" designs and negative bends on Nascar bars.

triangles are strong , just remember that.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Doctor CorteZ &raquo;</TD></TR><TR><TD CLASS="quote">for the love of god avoid "x" designs and negative bends on Nascar bars.</TD></TR></TABLE>

isn't that a kirk cage in the one set of pics with an x design?

 

Another question we had was about the diagonals. On some cars (like johng's), the diagonal brace on the main hoop to the lower main hoop goes from by the drivers head down to the passenger side floor of the main hoop, while the diagonal for the rear section goes from passenger side top of main hoop down to drivers side shock tower. Is there any advantage/disadvantage to that as opposed to how we did my car where all the diagonals come to one point on the main hoop?

Thanks
Tom

 

think of it from the standpoint of solely protecting the driver.

the stiffness the cage adds is merely an added bonus.

 

In terms of bending loads, if two elements (yup, tubes) intersect in a T and there is a load upward from the base of that T, the load on the cross piece at the top is in bending.

I am perhaps in the minority here but I am not so convinced that X-style door bars - those with all elements in a single plane - are inherently less safe than the typical ones that extend into the door. I won't go into the entire treatise here but it IS true that if an "S" or double bend is put in a tube, it is Mr. Newton's way of telling a force to "bend here." If any bar has a SINGLE bend in it, the message becomes, "bend here but I'll only go one way." A bent element is never as strong - in any loading situation - as a straight one.

K

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Knestis &raquo;</TD></TR><TR><TD CLASS="quote">
** Devise ways to get welds completed all around tubes while getting the elements themselves as close to the chassis as possible (e.g., A-pillars, roofline, etc.)</TD></TR></TABLE>
I had a thought about this, but I'm not sure if it would work or still follow rules.
I've been looking into building a cage myself (for my brother's car once he gets on track). I've read that some drop the cage through the floor to weld the top, and also that some cages in a rollover punched through the floor. Don't know if there's a correlation there, but got me thinking. Here's my idea: Make the main hoop ~3" short. Weld plate through floor with 3 1/8" tall tube that has an OD = ID of main hoop. Weld all bars at the top, then slip the main hoop onto the tube. Raise the main hoop, then weld on 2 3" half pipes (same OD as main hoop) onto the smaller tube, and on the main hoop. You can rosette(spot) weld these half tubes to the inner tube.
As further reinforcement/ force distribution, you could run some sheet metal triangles from the main hoop (at the bottom) to the plate. The idea being the triangles would help spread any force exerted on the top to the full surface area of the plate. Or at least more surface area than just the pipe to plate. This *seems* like it would alleviate some of the cage punching through the floor problem.
FYI, my background is in sculpture, not engineering, so I may be way off base.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote &raquo;</TD></TR><TR><TD CLASS="quote">** Try modeling some structures out of 3/16" wood dowel and hot glue. You can learn a LOT with 1/10th scale models out of these materials.
</TD></TR></TABLE>
Great idea. Bamboo skewers work too.

 

That is a very good point. In other words, keep bends to an absolute minimum. Would it be safe to assume then that the combination of an x-door brace and Nascar door bars would be the safest route to go? I still haven't seen anything about plate gussets attached to the a and b pillars. I'm starting to think they are not at all needed. Also, should plates/boxes be utilized on the shock towers? Is it OK to weld to just the body metal? Should the bars coming off the top of the hoop ALWAYS attach to the shock towers or would it be Ok to attach them behind the tower to the rear floor/subframe area? Sorry for all the questions, the info so far is great and if you feel it's time to cut me off, go right ahead. Otherwise keep the info coming, there's hope for me yet

Thanks

D.

 

The three best approaches that I've seen used to get the cage tight to the roof and still complete the welds up there are...

1. Holesaw the floor where the tubes are going to sit, lay plates over the holes, fit and tack weld all of the tubing, remove the plates, drop the structure through the floor, do final welding on the tops, raise it back up, and weld plates in place under the legs and over the holes.

2. Fit and weld the cage on tack-welded box structures, knock themd down, weld the top joints, then rebuild the boxes. This is how Competition Cages did mine.



3. Cut the roof off. This isn't as horrid as it might sound.

It is pretty much necessary under typical rules that tubes get welded to plates that are welded to the body sheetmetal. Ideally, you want to understand where the forces are being fed into the chassis - at suspension pick-up points - and mount them so that they resolve these forces in compression, feeding them into the cage structure. This will vary depending on the type of car.

For example, on a car with strut rear suspension, there is clearly a force upward when the suspension goes into bounce (gets pushed upward by a bump or cornering load)...

K

 

Very nice work there. That's a great deal of work just in that little corner. I had an engineer stop by on Fridayand take a look at the cage we had built. He was quite impressed and we had a good discussion about forces, why bars are put where, and where the loads come from. It was very enlightening! You have really helped me Knestis, to understand why cages are built the way they are, I am very greatful. I am quite confident that boxes are they way to go, and to keep most bars as straight as possible wherever possible. I am much more confident now to build somebody a cage, and be able to explain why it was built that way. I'm really like this place you call H-T!

D.

 

bump for a good idea. I am also a TIG welder and i refuse to pay for 600 dollars for something that costs no more than 50 in materials and a day to make.

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Masta &raquo;</TD></TR><TR><TD CLASS="quote">That cage is awesome!!!!! Why does it look so shiny? Welds are gorgeous and I really like the rear box on the shock tower. My next question involves attaching gussets to the A and B pillars. Is it necessary? What about tubular gussets at certain parts of the cage? Who built that BTW?

D.</TD></TR></TABLE>

that's because it's TIG welded stainless steel.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by bense &raquo;</TD></TR><TR><TD CLASS="quote">bump for a good idea. I am also a TIG welder and i refuse to pay for 600 dollars for something that costs no more than 50 in materials and a day to make.</TD></TR></TABLE>

You may need that cage or rollbar to save your life. Do not go into it without a clue..... a good cage is not money wasted, pay someone that will do it right.

RJ - who saw several cars on their lid this weekend and was glad they all had welded in cages and the drivers all walked away.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Doctor CorteZ &raquo;</TD></TR><TR><TD CLASS="quote">for the love of god avoid "x" designs and negative bends on Nascar bars..</TD></TR></TABLE>

They ain't NASCAR bars with those.

I'm not a fan of mounting the tubes on boxes.

Masta - I think your rear tower tubes are at to steep of an angle. For your floor plates also include plates running up on the rocker panels. Use as much of the max plate area the rules will allow. Also in your first design, I'd like to see the rear cross bar (behind the seat) go from side to side.

 

For another example, check out my signature. I have tons of pictures of the cage and welds. I do believe I have one of the best "quality" cages I've personaly have seen.

 

http://www.e30m3performance.co...m.htm

A nice overview of roll cage designs. Many of the designs shown do not meet SCCA rules, and many are a lot more complex than you need to do, but it does give you some good ideas...

 

[QUOTE=Angry Joe]http://www.e30m3performance.co...m.htm

[QUOTE]

Gustave has a wealth of knowledge put on that website