Good point. What is the least invasive way to accomplish this? Hammer the two surfaces together and spot (plug) weld?

Edited to add: If I end up going as fast as I'm hoping, I will need a six-point roll bar anyway. If that's tied into the subframe connectors, are the floor pan welds still important?

Spot welds are factory. I would run a full bed along the edges on both sides of the crossmember.
I did (had to do) so on my '73 Dart because I partly removed the upper part of the crossmember in the tunnel for a 518-transmission. Over time the factory spotwelds were tearing loose from the floorboard.
Remember the (dynamic) weight of the transmission/engine is leaning down on that crossmember all the time, as well as the torsion bars are putting a varying rotation torque/force onto the anchors.

I drilled small holes from under thru the crossmember's web as it folds out to the floor pan. Then on the top side I used sheetmetal screws to suck the floor pans down tight and then back under I stitch welded the pans to the x-member. Then I removed the sheetmetal screws and just welded the holes up.
But drilled holes and Clico's should also work.

RE the snipped out section above, I thought at one point over all these years we had talked about how Mopar's Hustle Parts engineers had developed the sub-frame connectors to resist the beaming motion of drag launches way back in the early 70s.



IMO, unless rules require it or you are pursing a more full frame approach, there really isn't the need for .125 wall structural reinforcements pieces in a uni-body construction. However, it is a common approach taken by many that if large is good, bigger is even better. If one is willing to add weight for structural improvements, much greater gains can be made with stitch welding factory structure over adding a bunch of eight inch wall structural components. Since your car is complete, proceed with the thick wall connectors and enjoy the increase in rigidity they will provide.

they provide a good place to land the kick in's from the main hoop.

And for the road racing group, a place to tie the outter unibody to inside.

I agree, but in drag race beaming improvement is the single objective often sought, for the OP, laying the 2x3 over is rather inefficient from a design standpoint in that goal, and the main basis for my comment. And I wonder, if the prospective SFC installer is not nearly lifting the front tires on a launch, not sure "beaming" issues should be their main concern. I'm more focused on torsional improvement, and once that direction is decided, it seems there is a plethora of ideas of the various solutions that SFC's actually contribute to the problem, albeit including some contribution with "beaming" issues.

TC, you are seldom one I ever seek to technically argue with, your continuous contributions here are well regraded, and I hope this is not one of them.

Nah, not looking for a debate on this one. Just recalling when we did really dig into the sub-frame connector piece and couldn't find much supporting evidence for torsional rigidity increases. Probably best not to blur any lines for the more casual tech readers.

Laying the tube flat may not be as good as standing it up but its still better than no SFC, IMHO.

I'd just spend the 2 hours to remove the seats and carpeting to fit it standing up, weld it in good and then 2 hours putting the carpet and seats back in....

How about - any that cause the entire body/chassis/suspension mounting points to deflect from it's static shape and dimensions. SFCs are only a part of the picture. A well designed cage structure that incorporates the original stamped sheet metal to further strengthen and stiffen the entire assembly, is the other part. There is substantial strength in those body panels; use them. Don't just cut a hole in the floor, pass a bar through it and patch it with a piece of aluminum. Fit it correctly to the sheet metal and weld it up. “The whole is greater than the sum of its parts.”

Why hang expensive suspension components and try to tune them with any predictable results when they're attached to a flimsy structure? I've seen high end Super Stock/Comp cars that were cleaned and had all the individual spot welded body parts stitch welded together before anything else is done. Time and money? Yes. But a substantial and verifiable improvement. Stop thinking of it as a cage stuck inside the body and look at it as a complete structure.

2x2 maybe if we insist on not sectioning it into the floor. 2x3 stood on end is much stiffer and allows for more of the floor to be sectioned into it and the cage structure to be attached directly to it. .083 wall is plenty substantial. We're still concerned about weight.

i think the common 2x3 tubing is usually 11ga, which is .120 wall.

Yep. You have to look for or order the .083. UPS will ship 8' lengths. Saves almost 10lbs for 8'. It's also available in .065.

I'm not that concerned about weight, except my own... being embarrassingly close to 300 lb with a helmet on
So the extra 20 lb of steel isn't that big a deal, since it's not a class racer
Once I get the aluminum heads and the bolt-on glass hood installed, that'll knock off a lot more. The best thing I can do is lose the 50+ lb of excess body weight that I already should have...

If we are spending money for fiberglass and aluminum, I would think that saving 10-12 pounds and probably saving a little money as well would be an easy decision.

Well, it's primarily a street car with possible occasional strip use. "We" spent the money for aluminum in order to increase performance (via porting). Iron heads are not only limiting but ridiculously expensive to port to relatively modest aluminum head levels. The weight loss is a bonus. Similarly, the glass hood has a scoop built-in to allow more carb/manifold clearance and admit more cool air, without having to spend money and time hacking up the original flat steel hood or paying a body man to do it. Again, the weight loss is a bonus.

You pointed out in your earlier discourse that structures work as a system. Same with every decision to buy, modify or replace parts - none of that happens in isolation.

I know dealing with 8 foot sections of 2x3 boxed tubing building motorcycle trike frames we found little difference in flex between 0.120 and 0.095 and 0.083 when tied in with the other frame parts all diagonally braced. And the diag's were all 0.063" DOM round.
But we are talking a 800 pound trike and not a 3200 pound car. But then again the SFC's are not real structural in most use. I bet you could get about the same effect with a stick of 1 1/2" 0.095 DOM buired half way into the floor pan and stitch welded to the pan.

These days, most serious builds and those that need to have an SFI tag are all round tube. And it's 4130. 1.500x.065 frame rails but there is also plenty of diagonals and bracing as well.

Using the totally unscientific method of jacking up one corner to change the tire methodology, I got the same observed stiffening results from .120 2x3 square tube installed logitudinal as from an Xed structure of 1.5 x.090 DOM round tubing. Or course the X arrangement did not allow much room for exhaust and may not be practical in a street car.

The guy I was working with made ladder frame sections 8 foot long. At the rear it was 18 inches tall and the top tubes were curved down to 12 inches up front. As said all the diag's were 5/8x0.063 DOM. He then placed the single ladder on blocks and set a dial indicator under and loaded 300 pounds in the center to measure deflection. From memory, the difference between 0.120" and 0.085 was about 0.25 inch. In a bounce, he lifted up on the center and dropped it 4 times, I think it was like 7/16 inch difference.
But in these tests he did not do a 100% full weld on the diag's so there may have been some loss of rigidness.
He ended up using 2 of the 1.5x0.085" tubes on the bottom and 2 of the 1.5x0.063" tubes on top.

"This thread is worthless without Pic's" Yes I lost all those pictures when a hard drive failed. I'll see if he has pictures on the Web of the build.

The problem that sticks out to me in your ladder bar example, the thicker wall thickness main chord might mainly just strengthen the welded node connection between the larger tubing and and the smaller 5/8" D diagonal tubing, and that accounted for a lot less of the deflection, would be my guess.