Torsional frame flex reduction

[vette427-sbc]

I posted this on another forum and Im looking for some additional input...


Gathering parts and ideas for a T56 swap this winter... I have a welded in factory crossmember now (4speed). I was thinking of ditching the whole crossmember in favor of a large tubular X-brace to stiffen up the frame a bit and also serve as the trans crossmember. I am yet to actually get under the car and look at obstructions, but I dont remember anything being too low or in the way. It also has to bolt in... I have sidepipes, so exhaust clearance is not an issue.
Will it stiffen up the frame enough to be worth it?

Green would be the new X brace (rough drawing)
The car has a good amount of frame flex and Im just looking for relatively simple ways to reduce the flex without adding a full cage or alot of weight.
https://static1.pt-content.com/images/pt/2012/10/xbrace-1.jpg

[exwestracer]

I posted this on another forum and Im looking for some additional input...



Green would be the new X brace (rough drawing)
The car has a good amount of frame flex and Im just looking for relatively simple ways to reduce the flex without adding a full cage or alot of weight.
https://static1.pt-content.com/images/pt/2012/10/xbrace-1.jpg

It will help, but to really improve torsional stiffness, you need height. Best case scenario would be if you could build that X member out of 3 or 4" tall rect. tubing without cutting any notches in it. Probably not possible...66579
Sorry for the quickie line work... Raising the sills and tying the front kickup to the rear would probably do the most good short of putting a cage in, but that's pretty major fiberglass (Vette, right?) work.

[astroracer]

Hey Chris,
This seems to be a popular subject. What you've added in green will help longitudinal stiffness but will do nothing for torsional stiffness (okay "maybe" a little). The only way to get good torsional stiffness is to add a full cage with good triangulation between the outer rails and upper bars of the cage. You have to "create" an element for the outer rails to push against to reduce torsional roll. That element is the upper bars of the roll cage. There is really no other way to do it. You could possibly add firewall and rear seat bars that will tie into the A & B pillars. This would gain a little torsional stiffness but to tie the chassis into those bars and ,especially the rear suspension forward pivot points, would cause a lot of grief to your interior.
Don't know how far you want to take it.
Mark

[dontlifttoshift]

If I could rephrase the op's question....would the x frame be stronger torsionally than a simple bolt in crossmember?

Ray, how would raising the sills (making them taller) help with torsional twist?

[exwestracer]

If I could rephrase the op's question....would the x frame be stronger torsionally than a simple bolt in crossmember?

Ray, how would raising the sills (making them taller) help with torsional twist?

Donny, if we look at the frame as a whole, torsional flex is really bending one of the rails up in relation to the other. This load is balanced out at the spring pickup point at the opposite end of the rail.

Now, I DID make an assumption that the front and center crossmembers are holding the rails stable at those points. If we ran tubes from the front spring pockets to the top of the kickups (basically the bottom half of a roll cage) we would be removing the hinge point between the front spring load and the opposing spring load in the rear. Basically, the higher sill is acting like a big "gusset" along the same lines. Not as effective, but there isn't much else you can do without going right through the center of the car. That's actually why the original 300Sl Mercedes had to have a gullwing door. The sills were raised radically to help torsional stiffness.

And, yes the X member would be better than a bolt in straight crossmember. How much better depends on how tall the material is and how much notch has to be taken out of the center for trans clearance and a mount.

[dontlifttoshift]

So not just making the sills taller, but making the sills taller and tying them in front and rear to suspension pickup points. Like you said though, only as strong as as the crossmembers tying them together side to side.

300 Merc had a pretty good tunnel through the middle, right? So maybe some strong bulkheads (firewall, behind the seat) and a spine down the middle and you would end up with a pretty strong tub.......lots of work.

Chris, if you could make that X out of some rectangular tubing, taller, not wider, I think you could make some nice gains in rigidity. Wall thickness won't count for much, .083" would probably do it and help minimize the weight penalty, assuming you can get some height on your tubing.

[ace_xp2]

I would think that the sill height would bring advantage in a lot of designs as they're typically the longest untrussed member of a unibody chassis. No cowling or wheel tubs/b pillars to help them out. I'm assuming the vette isn't using that frame as its only stiffening member though.

[dontlifttoshift]

Yeah, look at the height of the inner rocker on a mid 60s mustang.....very tall.

[vette427-sbc]

Just took a good look underneath the car...There isnt any room to build a "taller" x-member. Looks like a roll bar has far more benefit/lb in this case.

Now after looking the car over with structural height in mind, does anyone run any kind of mid plate from the bellhousing to the frame rails? Ive got pretty much a straight shot from the back of the block to the frame rails (again, sidepipes, so no header issues)
I already have solid motor mounts and a "spreader bar" connecting the two shock towers. Im thinking it might be overkill since there are already two "tall" crossmembers in the front of the car. Opinions? Benefit per lb is the major deciding factor here.

[exwestracer]

Just took a good look underneath the car...There isnt any room to build a "taller" x-member. Looks like a roll bar has far more benefit/lb in this case.

Now after looking the car over with structural height in mind, does anyone run any kind of mid plate from the bellhousing to the frame rails? Ive got pretty much a straight shot from the back of the block to the frame rails (again, sidepipes, so no header issues)
I already have solid motor mounts and a "spreader bar" connecting the two shock towers. Im thinking it might be overkill since there are already two "tall" crossmembers in the front of the car. Opinions? Benefit per lb is the major deciding factor here.

That can help, but unfortunately it works best when tied into a more vertical structure. Just bolting the midplate to each rail with one or two bolts on each side won't do a whole lot to "lock" the rails together.


Let me ask, WHY do you think the frame is flexing too much? I've been through this with a C4 Vette in the past. Just curious what the "symptoms" are...

[Supervette]

I think I know why he asking this question. Just backing out of my driveway I use to hear a slight creek or two. I think its due to it being a T top car. I had a 4 point chromoly cage put in to try to fix this, and handle the Prochargers power without tearing up the car. The 4 point helped a ton but like any racer I think we all look for ways to improve our chassis as that is how we get power to the ground and handle better. I'd recommend the 4 point as it has minimal interior impact on these cars. I have contemplated going to the 6 point but that is where Corvettes begin to run out space for comfortable seating or gaining entry to the seat. You can get swing outs put in though to take care of this. I have seen pictures of it being done that have looked good but you need a good reputable shop. Anyways, good luck with your decision and if you go with the cage post some picks up!:twothumbs

I'd personally like to hear more about the mid plate and its function. I have had people in the past recommend them to me once before. Is it more for drive line rigidity??? I also see front plates put on some older stock cars. Even with my solid motor mounts my engine still torques the procharger pulley to scrap the bottom of my hood. Thats when I went to a smaller pulley:cheers:

Even though the car has a little flex on the Corvette forums I have heard of guys running into the 9's on these stock frames. This could be myth but I have heard this on the Corvette forum so don't chew me up to badly guys. From what I have heard the frames are very strong and solid its the IRS parts that will kill ya though, although there are expensive options to make it work.

I know I am getting long winded here but I agree with the other guys. You have to decide how far you want to go. How much fun do you really want to have? How fast do you dream your car should be?

[silvermonte]

Ive been wondering about this topic for a bit, for those who dont want to put a roll cage in their car how much is there to gain by putting in a spine down the center of the car around the drivetrain, and would it have to be made out of round tubing or would square tubing work? I could keep the spine level with the bottom of the frame but could then raise it up a few inches above the top of the frame and into the body, i lowered my car and i have to cut the trans tunnel out anyways and this would be an option for me

[exwestracer]

Miles,
It depends a lot on how you tie that spine back into the "main" chassis structure. Here's a classic example (Lotus Elan) of using the spine for the whole works.
71829

Tying that back into the rocker sills with some transverse boxes should make for a pretty stiff structure. Don't build it like you would a tube chassis...because in the end you'll have a tube chassis.

[vette427-sbc]

I admit I kind of forgot about this thread... as to why I think I need to reduce frame flex? Pretty much the same reason I think I need a big set of Brembo brakes. Its not a race car, but I would like it to perform as close as I can to one while still remaining a street car. (that makes alot of sense :screwy:)
But like supervette said, I can feel the car flex pulling out of the driveway and after adding a few other frame stiffening pieces, I really liked the improvement in ride quality. The car is having a cage put in it now... 1.75x.095 chromo. I posted up a few progress pics in my build thread. Should be done in a few weeks. :smoke:

When the body comes off the frame I think Im going to look into some structural foam to stiffen it some more. Thats quite a ways from now though

[uxojerry]

You may want to look at aftermarket frames/chassis. 95% of updated vintage corvettes go on a new chassis/frame. An SRIII tube chassis is $7k. The tube chassis also makes it easier to run C5/6 suspension/brake components as they bolt on directly. The extra $$ you spend on the chassis is recouped by using off the shelf C5/6 components. Another advantage to the custom chassis, is you can run OEM ZR1 wheels and tires. OEM C6 ZR1 wheel and tire pkg is -$4k. Custom offset after-market pkg is +6k and you must run fender flares. The ZR1 wheels will fit in the wheel wells as the chassis is narrowed to do so. You then cut out the floor over the wheel wells and graft in a custom floor. The custom floor can be ordered from the vendor.

I learned this stuff the hard way. Tuition at the School of Hard Knocks is very expensive! Good luck with your build.

[silvermonte]

Miles,
It depends a lot on how you tie that spine back into the "main" chassis structure. Here's a classic example (Lotus Elan) of using the spine for the whole works.
71829

Tying that back into the rocker sills with some transverse boxes should make for a pretty stiff structure. Don't build it like you would a tube chassis...because in the end you'll have a tube chassis.

I really dont know how I would tie the spine back into the chassis, thats why im looking into this. Granted im building a s10 so my frame is a bit different but the same theory should apply. My plan was to start right behind where the frame bows out right behind the front suspension then the two sides meet up at the tranny, and then go out behind the cab and then split back apart under the bed where the frame starts to angle back up, should keep my spine level and not get to insane. My frame is 5" tall, there is 1" between the top of the frame and the cab and i wouldnt mind going up into the cab about 2 or 3 inches. So the runners that reach from the center spine are 5" tall and then the main spine could be up to 8" if i wanted and ill just make room for everything else. I dont want to put a roll cage in my truck and this is why im looking into this. My truck is 2900# and ive pulled out alot of weight so i can afford to add a few points in tubing, im not worried about that a whole lot if it will help alot, im going to be boxing the frame also. Im wanting to do something like this with my trans tunnel, just not as high since my tranny is that far into my cab. http://www.hotrodstohell.net/chevelle_from_hell/index.htm

[vette427-sbc]

You may want to look at aftermarket frames/chassis. 95% of updated vintage corvettes go on a new chassis/frame. An SRIII tube chassis is $7k. The tube chassis also makes it easier to run C5/6 suspension/brake components as they bolt on directly. The extra $$ you spend on the chassis is recouped by using off the shelf C5/6 components. Another advantage to the custom chassis, is you can run OEM ZR1 wheels and tires. OEM C6 ZR1 wheel and tire pkg is -$4k. Custom offset after-market pkg is +6k and you must run fender flares. The ZR1 wheels will fit in the wheel wells as the chassis is narrowed to do so. You then cut out the floor over the wheel wells and graft in a custom floor. The custom floor can be ordered from the vendor.

I learned this stuff the hard way. Tuition at the School of Hard Knocks is very expensive! Good luck with your build.

Thats too easy though! More like its just too much $$ to swallow at once...

[exwestracer]

71864
This one?
You could get by with a lot less triangulation. That center section is MUCH more rigid than the attachment points... But, yeah, it's the right idea. Personally, I like the "semi-monocoque" approach Lotus used with less tubes and the sheet metal structure running full length. That also gives you the option of just painting it and making that the interior tunnel surface. When you tie back into the frame, try to get as close to the front and rear suspension as possible (at least get ahead of where the frame widens out). Try to get as much beam height ( 5" rect tubing stood upright?) as possible with the pieces that tie back to the frame. You can tie the rect tubing to an upright piece of round or square at the ends of the center "spine".

[Supervette]

71864
This one?
You could get by with a lot less triangulation. That center section is MUCH more rigid than the attachment points... But, yeah, it's the right idea. Personally, I like the "semi-monocoque" approach Lotus used with less tubes and the sheet metal structure running full length. That also gives you the option of just painting it and making that the interior tunnel surface. When you tie back into the frame, try to get as close to the front and rear suspension as possible (at least get ahead of where the frame widens out). Try to get as much beam height ( 5" rect tubing stood upright?) as possible with the pieces that tie back to the frame. You can tie the rect tubing to an upright piece of round or square at the ends of the center "spine".

That just seems like a lot of work when I think a simple 4 point will do the trick. If that is not enough for you add the two bars to make it a six point. The 4 point only affects my battery storage routine a little bit.

[Supervette]

Plus it gives something for the race harnesses to fasten onto!

[exwestracer]

That just seems like a lot of work when I think a simple 4 point will do the trick. If that is not enough for you add the two bars to make it a six point. The 4 point only affects my battery storage routine a little bit.

You are absolutely right. Silvermonte DID say he didn't want to put a cage in the car, so that puts the thinking in this other direction.

[silvermonte]

My problem is that i dont want the down bars to be going into the bed of my truck, would making loading and unloading larger items a pain in my truck. It seems also that having the main hoop sitting further back in the extended part of my cab with down bars going forward in the cab is not a safe option so im told. Not sure what the difference is between that and some of the other cages in other cars but rules are rules.

My main plan was to have the main hoop as far back in the cab as i could with the down bar at the 45 degree angle and it comes to about right below shoulder on me and then put another main hoop closer so that its within the 6" of my helmet like alot of rules require. The closer main hoop would be tied into the down bar, and then i would tie the 2 main hoops together the forward and aft direction and then the back hoop would have lots of diaginal brace and the front hoop would have left to right bars for the harness and what not. My problem is nobody seems to want to say that would be a safe way to have the cage and im just wanting something safe, im not really worried about the rules alot, i will just be doing mainly auto-x and a few tracks but in the rare case i put my truck on its lid i know it would crush without some sort of bar so thats what im wanting to do, but if its not safe i dont want to make things worse, so im just sticking with making the frame as stiff as possible till i can talk to someone in person that knows a thing or two that can tell me what i need. I dont do a very good job of explaining how i want my cage over the phone or on the net.

[exwestracer]

Ive been wondering about this topic for a bit, for those who dont want to put a roll cage in their car how much is there to gain by putting in a spine down the center of the car around the drivetrain, and would it have to be made out of round tubing or would square tubing work? I could keep the spine level with the bottom of the frame but could then raise it up a few inches above the top of the frame and into the body, i lowered my car and i have to cut the trans tunnel out anyways and this would be an option for me

Sorry, I'm confused now...

Regarding your cage design, you can have an extra hoop, but the one right behind the seat must go all the way to the floor, and your down bars would have to start there. The one at the back of the cab would be seen as the "extra". Either way, I don't think it will get you out of having the "kickers" going back into the bed area if you need to get it teched for racing.

[silvermonte]

i can understand the confusion, i do a horrible job of explaining. i have been told that the kickers have to go towards the rear of the vehicle,they will not provide the support they need if they go forward, even tho the down bar would be the same length going towards the front or the back if the main hoop was in the rear of the cab. I wanted the 2nd main hoop to be closer to the seat to be within the 6" of my helmet for that reg but as you just said it has to be the one that the down bar starts with.I then wanted the 2 hoops to be tied together however the cage builder said they would be needed to provide strength for a roll over. So since i cant get any roll bar shops to at least consider if my design is safe im just going to go without a cage. One of these days im just going to go to the store and buy a bunch of the round foam that you put in the pool and mock up my idea so people can see it better then me doing a horrible job or explaining it.

[silvermonte]

hey Ray i sent you a PM

[j-c-c]

You are absolutely right. Silvermonte DID say he didn't want to put a cage in the car, so that puts the thinking in this other direction.
It doesn't seem to me that the one written in stone simplest solution to reduce torsional flex no one has mentioned, and that solution is a simple tube. It outperfoms all other shapes. Almost all other solutions are simply trying to recreate a simple tube, but always with compromises. Rule number two is the larger the tube the less flex, again all other shapes usually are improved by simply making them larger in cross section. IMO the OP adding "X" bracing does little to increase the size of the holy grail sought after tube shape. Obviously we can't ride around in 5' dia tubes, but at least we can understand the sought after design target to minimize unproductive detours and/or added unnecessary weight. My reply addresses only torsion rigidity the op mentioned.

[exwestracer]

It doesn't seem to me that the one written in stone simplest solution to reduce torsional flex no one has mentioned, and that solution is a simple tube. It outperfoms all other shapes. Almost all other solutions are simply trying to recreate a simple tube, but always with compromises. Rule number two is the larger the tube the less flex, again all other shapes usually are improved by simply making them larger in cross section. IMO the OP adding "X" bracing does little to increase the size of the holy grail sought after tube shape. Obviously we can't ride around in 5' dia tubes, but at least we can understand the sought after design target to minimize unproductive detours and/or added unnecessary weight. My reply addresses only torsion rigidity the op mentioned.

Good point. Now how would you apply that thinking to a factory chassis without cutting the floor pan or adding a 3 dimensional structure like a roll cage?

[T.K.]

there is another option very similar to a x that i like better. some people call it an inner frame but its effectively a larger x and a cross-member. below is a pic borrowed from one of our sponsors art morrison. welded in it is probably the best you can do without an aftermarket frame or cage. http://www.artmorrison.com/images/53-62Vette/final/AM%20vette0232.jpg

[j-c-c]

Not to beat a dead horse, but I visualize a typical ladder frame to a sheet of PW, it twists just like an empty picture frame, and its completely solid, ie adding all the internal bracing you want up and until its completely solid gives little improvement in torsional rigidity. However going from a 3/8" sheet of PW to say 3/4" shows results, and if possible, going to 6" and beyond is the real deal for improvement. I know many want the magic bullet undiscovered brace that gets it done, but its hard to overcome physic, try as we might. Of course if we are talking the new OEM offset crash test requirements, or single wheel rear wheel drive, that big X looks pretty good.

[Norm Peterson]

Torsional stiffness is a 3D problem and all about getting the material resisting it spread as far apart as possible, within any other limits that may apply.

A thinwall round tube is particularly efficient, as all of the material is the same distance away from the center of rotation. Rectangular tubing is slightly less efficient, and the closest that a car frame can come to that is the backbone sort of frame pictured earlier in the thread. A limitation of the backbone and even the rectangular tube is that local distortions tend to concentrate at the corners of the cross section that compromises away some amount of the torsional stiffness that you'd calculate based on the cross section maintaining its 90° (or whatever angle) corner shapes.

An X-brace works essentially by converting the general chassis torsion into bending of the X-brace elements. Spreading the material further apart here means deeper X-bracing beams. This is a reasonably effective means of stiffening a flat plate, but has the obvious disadvantage when the X-bracing becomes several inches deep and running right where most everything else has to live (powertrain, exhaust, seat tracks, feet, etc.).

Another take is to beef up the closed cross-section sills in both width and depth. If you've ever climbed into an XKE Jaguar, you know exactly where this approach can go. You probably have better side intrusion properties than with a backbone, but the transitions to where the frame must narrow to go between the wheels (and permit steering up front) are apt to be more difficult. You still want to tie the sills together so that they act together rather than separately, and you could get into elastic buckling considerations if the ratio of width or depth to metal gauge gets too large. There is always a compromise to unravel.

3-D is why most coupes are much more rigid than their convertible versions - Corvette excepted since they're all based on a separate frame that gets little structural help from the bodywork. Even though there are large openings, there's a lot of 3-D help going on, and when the glass is bonded to the sheetmetal even that stuff adds to the torsional stiffness and has some structural function.


Norm