norvalwilhelm

As most of you know I am installing a 8 possiblly a 10 point roll cage, . I also want to bolt it in because of insurance, some don't like it and I have to get the car appraised for insurance, the cops are giving guys hassels , I don't want to totally gut the car for welding so I can tac and finish welding outside the car and I want the whole thing double powder coated.
The main reason for the cage is for rigidity. Sure in the even of an accident I want all the protection I can get but the cage is going in for rigidity. I have one in the mustang and it made the car much more rigid and driving in a driveway really shows the difference.
Anyway some scoff at the idea of bolt in and rigidity in the same breath.
There is a right way and a wrong way to bolt one in.
Take a 2 x 4 x 1/4 inch piece of steel, lay it on the shop floor flat and lay a 5 foot long bar on top of it and weld the two solid together.
Now bolt that to the frame. The question is how?? YOu can not get at the top of the frame to drill holes and what good would it do anyway, nothing thick enough to bolt into. You can try putting it on the side of the frame but how agian do you bolt it?? Through bolts? The frame would collapse if you really tighten the bolt down. How about welding another 1/4 inch plate with 4 tapped holes on the side of the frame then bolt our test bar to that?? The problem is you would be testing the friction between the two plates created by the clamping force of the 4 bolts. Tighten the plate down and if you could start rocking the 5 foot bar back and forth in no time you would have play. Why because the bolt holes have play and the 4 bolts only created a friction joint between the 2 plates. A few miles down the road this would happen to your cage, it would loosen up the tight friction joint, the bolt holes would begin to get sloppy and the rigidity that you were after is gone.
Now take another plate like before and lay it flat on the floor but stand the 5 foot pipe on end and weld it solid to this plate.
Now the only way to mount this setup is flat on the frame. since we have no bolt holes we need to create them. We can not weld nuts to the inside, we can not pass bolts all the way through because of collapsing the frame so we need to weld a threaded mounting plate solid to the frame and bolt the plate to this welded plate.
Now we are not dealing with a friction joint but the torsional strength of the bolts pulling straight down. Rocking the bar back and forth cycles the bolts through a tensile load and hopefully within their elastic limit. The friction is no longer a factor and neither is the bolt slope in the holes.
So the load applied to the bar must be perpendicular to the bolts. No 90 degree loading to the bolt. It must be in a staight line comming form the bolts to apply tensile loading.
Since we would have problems bolting to some parts of the frame I am welding standoffs the leave plates parellel to the frame but on the interior of the car.
So it is critical how a roll bar/cage is bolted down. Holes through the frame will not do.
If looks are the only concern go for it but if rigidity is important bolt everything down in a straight line from the bar.

lostpatrolman

interesting discussion. Thanks for the info.

70_Shark

Interesting idea. I would like to see some pictures if you get the chance.

bobs77vet

wow thks for the insight, i think i followed it and now i know why i'm not an engineer. Does that same logic re the stresses apply to welded joints? so that welding it flat on top is ok? or is there a whole another set of stresses involved?

mrvette

NO, good thinking Norval, nice to hear of someone with another approach to the problems....tempted to follow along some day, but first projects first....allways something comes b-4, and so the new style operating headlights are first...lo profile opening only 3-4 inches high, instead of 6-8 inches whatever....

as usual, sounds like a good accomplishment....



GENE

LiveandLetDrive

Correct me if I'm not understanding the example, but in a situation where the whole cage tries to twist, isn't there still shear load?

-Chris

Red73Vert

Norval,
Some very good perspective, well described - thanks. Even with the plates flat, the frame twist/flex is going to put those bolts/plates into a tension/shear condition - as would a side impact during an accident.
Your thinking is right regarding the bolted attachment in shear/cantilever. Bolting into a threaded steel plate is only going to be as strong as the steel in the plate's threaded area, and as the plate flexes there will be some distortion of the threaded holes as well. Threading the bolts into grade 5 (or better) nuts on the backside of the plate would provide for a much stronger attachment.
To further strengthen the connection between the flat plates you could machine a hole(s) into one plate and have a matching protrusion(s) on the other plate so that they would interlock when in shear. Or you could do something similar to how your standoff interlocks with the frame.
The other thought I had was to make the bolted connection in the tubes versus the plates. You could use bar material with an OD close to the ID of the tube. You could weld a length of the bar (2x-3x the OD) perpendicular to the plate. Then slip the tube over the bar and cross drill and bolt the tube and bar - twice such that the bolts would be at least 1/2 an OD apart on the tube and at 90 degree angles.
Of course all these bolted connections should be adequately washered to spread the loads.
Best of luck with your project, and thanks for sharing.

NHvette

Good thoughts - well expressed, Norval.
I do like the idea of being able to remove ... and assemble with body on.

I tend to agree that the twist of the frame will have
force vectors that apply to the friction joint in your situation.
I know you already started, but ...
How about a mounting plate with two perpendicular mounting
surfaces (three would be ideal). That way, there are bolts
holding the tensile joint in both vertical and lateral directions.

This is why I love being an engineer.
John Lennon said it: "There are no problems ... only solutions."

632C2

What a great project. The finished product should be outstanding. Keep up the good work.

Steve

The Dude

The only right way to bolt in a roll bar or cage is if you're doing it only for style. They have to be welded to be legal in the eyes of most sanctioning bodies.

LiveandLetDrive

He's not trying to be legal, he's just doing it for frame stiffness and safety. As far as I know that car never sees the track, although I'm sure it has the potential to kick some ***.

-Chris

DJ Dep

Yep...I wondered when someone would bring that up
They specifiy welding in NHRA FOR safety.
Bolt in would be a show bar...NOT a roll bar.

If you are going to the major effort of installing this for safety and stiffness, why compromise? A roll bar, especially an 8 or 10 point, is gonna make for some VERY awkward climbing in and out. If you are going to hassle with climbing in and out of a very small cage, at least make it worth the effort and have it welded.

Dep

LiveandLetDrive

He mentioned swing outs in another post. I would do the bolt in thing either, but hey we've all got our priorities.

-Chris

norvalwilhelm

Welding is different and you are not worried about friction joints. You can weld to the side of the frame, infact that is the easiest place to get at for welding.
When mig welding vertically start from the top and slowly bring the gun down pulling the bead with you. Do not weld up.
Set the voltage on high, the wire feed fairly fast and weld top to bottom, do not weave and welds can come out great.
Do not weld up, weld down

norvalwilhelm

Sure there are shear loading but you try to keep the bolts under tensile loading if possible. With the frame twisting you get all sorts of loading but a bolt can take repeated tensile loading but using it for friction joint like bolting to the side of a frame is wrong.

norvalwilhelm

Red73vert. You have some good ideas. Some of them I will incorporate in tube passing through the firewall and the swing out bars on the sides. I am considering 2 bars on the sides, one low and fixed and one higher and swingout. The DOT requires for the street that side bars are hinged in the event of an accident.

norvalwilhelm

I am not in the least bit intereste in being legal for anything except the street. I have to please the DOT and the cops. I will not race the car.
As for being just for looks I don't like the looks of a cage, it clutters the interior. I want it for function.
Are you saying that bolts used correctly can't replace a weld? I wouldn't relay on a welded trailer hitch. bolting is far safer. Bolts if used correctly can be stronger then welding and justs as rigid.
I am just pointing out that a bolt should be used in the tensile mode, not used to apply a friction load for 2 plate squeezed together.
If you can not see the difference then I didn't explain it very well.
On my car to my knowledge nothing is for looks other then the paint, everything is functional and if I believed this cage wouldn't add alot of rigidity to the car I wouldn't have it.
This is actually my 4th cage that I installed in a car. The difference before and after is very easy to feel just pulling in the driveway. If you don't own a caged car you wouldn't know.
I could take pictures of the mustang sitting beside the vet but it would be comdemed because it is polished stainless.

DJ Dep

Well I, for one, am confused.
Are the police REQUIRING you to install a roll bar??? Think I will stay out of Canada if that is the case!!!
You say the bar is NOT for looks, but you want to use a method of installation that is NOT approved by any sanctioned racing body. Not drag racing, road racing, or NASCAR. I would tend to think THEY have a better idea of what is the best method for installing a roll bar than ANYONE on this forum. Current class 3 and 4 trailer hitches are a combination of bolts AND welds. The frames used in both funny car and rail classes are WELDED together. Same for NASCAR. I would not trust a backyard weld either. But a skillfully executed weld can actually be STRONGER than the rest of a structure. I still can't understand why you want to have JUST rigidity and not have the other benefits of a welded-in bar. Do you plan on switching the bar out when you sell the car???? A quality, welded installation of a roll bar would be a plus to a potential buyer.

Dep

LiveandLetDrive

Logically a bolted joint can be stronger. It's just alot easier to have a great cross sectional area of connection (ugh, understand what I mean?) It is more difficult to make it rigid though, but correct fastener technique can mostly cure that.

I don't think anyone would argue that an an ideal bolted cage would be as good as an ideal welded cage. However, in a street car where comfort is required, and in this case the ability to take the cage apart for body offs or inspections, I think some compromise is required. Not alot of performance lost if it's done right.

-Chris

vettery

The load on the bolts will hardly ever be 100% tensile (force acting parallel to the bolt or perpendicular on the roll bar). There most always will be a component of force acting perpendicular to the bolts creating a shear stress; at first resisted by the friction created by the clamped joint but ultimately by the bolt(s) once the clearance in the hole(s) has been taken up. Now the shear strength of 4 good sized grade 5 bolts is pretty high but locate 4 (or more) holes so accurately that the clearance in each is the same in all directions. One of those bolts is going to see the shear load first. The shear strength of one single bolt will be high enough to elongate that hole in low carbon plate material; may be elastic deformation, may be permanent yield depending on how far the plate has to move before enough area is accumulated (slop taken out of enough holes) to resist the shear load. I'm not thinking about failure of the joint, only about the amount of flex that could take place before it meets enough resistance to seriously limit additional flex. The welded joint will present it's maximum shear area immediately. I have no idea of what kind of loads we're talking about. Hell, the frictional resistance between the clamped plates maybe be plenty-a-nuff to resist any shear load (which is only a fraction of the tensile or compressive load) created by 10 second ET launches and 1 G turns. Under these conditions the bolted joint may not allow any more frame flex than the welded joint as long as it stays tight. So why don't I shut the f___ up? I want to bolt mine in too!