after doing all the work to the rear of the frame (new cross, notch, etc....), and seeing how much the frame flexes on the rotisserie, my new project is going to be frame stiffening. the plan is to box the C channel and use 1.75" tubing to strengthen the rest, kind of like the roadster shop chassis below.

im thinking that the points of connection for the 1.75 will be as low on the frame rails as possible. this should leave about 1.25" above the tubing to run the fuel (1/2" x 2) and brake lines on the boxed frame. it will also leave as little floor pan to tube contact as possible.

the rear cross will incorporate a ds loop and 4" exhaust holes for the 3" exhaust. the plan is to run this cross as far back as possible, and possibly tie in the front lca mounts.

anybody have any tips to look out for? line running? exhaust? trans cross? any input would be appreciated.

thanks, Tim

http://www.roadstershop.com/images/Picture164.jpg

http://www.roadstershop.com/images/Paintedchassiscomplete.jpg

http://www.roadstershop.com/images/Exhaust3.jpg

I can't see why it wouldn't work especially if you keep everything as far away from the floor as possible as you mentioned. You will have to do a custom trann crossmember but thats not a big deal. I think if you did that in conjuction with a cage it would make for a very stiff frame.

Dont forget about the e-brake cables and brackets. I plan on running a Lokar lever style e-brake off the trans tunnel. Make sure you have all your brackets figured out before you paint or powdercoad your frame. Use hard line for as much of the fuel line as possible. Over on Pro-Touring.com there were people having isses running braided line the length of the car. Where are you mounting the battery? Engine Bay or Trunk?

I plan on boxing the rear frame by the gas tank and running a Rick's Hot Rod tank but mount it to the frame, rather then the body. I may raise the trunk floor an inch or two for extra clearance as well.

If you leave a small boxed section along one of the main rails you can use it as a conduit for the electrical stuff. Keeps it very clean especially if you have the battery in the trunk.

Ron

the battery will be in the trunk. im also going to be running a rick's tank, but it will have to be a custom tank due to the watts. since the trunk is already cut up, i was going to run it taller as well to gain some extra capacity since im losing it on the front side.

i plan on running the ISIS system, so im not sure whether or not i even need to run a heavy gauge wire back to the front. is inside the cab a big "no no" for the main wire?

thanks for the suggestions

Tim

also, the plan was all stainless hardline for the fuel and brakes. hard to bend, but they look sick.

i just called my metal yard, and they only want a little over $5/ft for the 1.75" .120 DOM. i thought that was a pretty good price. the regular tubing was a little cheaper, but i think ill stick with the 1020 unless somebody has a reason not to.

Tim

Looks like an awesome project! I wish I had the knowledge to do something like that.

Any reason why your mounting the battery in the trunk?

Have you seen my Braille Auto battery? It's 11.5Lbs and 5.8” x 3.4” x 5.8”.

http://www.bionicengineering.net/media/DSC03402.JPG

http://www.bionicengineering.net/media/DSC03419.JPG

These are work in progress pics. Don't have any of the bracket painted and correct mounting bolts. So far it has been working great. I don't even leave the tender on it.

First Guess:

All that bracing will make a H-U-G-E difference if someone T-bones you at an intersection. It will make a fair difference if you put a pair of jackstands in the middle of the car and play "teeter-totter".

It will not make much difference at all in terms of stiffening the frame to resist TWIST. At least, not unless you tie all that extra tubing into a roll cage--something to provide the third dimension to the existing 2-dimensional frame. All you have there is width and length; resisting twist will require height. And since resisting twist is going to be important in a high-torque, heavy vehicle...you need to think about triangulating UPWARD as well as sideways and lengthways.

But I'm not a mechanical engineer; so don't take my word for it.

Any reason why your mounting the battery in the trunk?

the main reason is to clean up the engine compartment. 11.5 lbs (more with my optima battery) moved from the extreme front to the rear cant hurt either.


im torn btwn the 1.75" DOM and the 2" DOM. my thoughts are that while im at it, it might as well go with the larger tubing. i was even tossing around the idea of the main rails in 2" and the braces in 1.75". looking at the roadster shop site, it says that "Tubular, Triangulated Center Crossmember For Greater Strength" is 1 5/8", but i have a feeling thats just the trans cross and not the entire webbing. it looks much larger in the pics above.

weight really isnt that big of a deal. 1.75" is 2.1 lbs/ft and the 2" is 2.45 lbs/ft. it would be a difference of about 6-9 lbs depending on the amount used.

any suggestions?

Tim

the battery will be in the trunk. im also going to be running a rick's tank, but it will have to be a custom tank due to the watts. since the trunk is already cut up, i was going to run it taller as well to gain some extra capacity since im losing it on the front side.

i plan on running the ISIS system, so im not sure whether or not i even need to run a heavy gauge wire back to the front. is inside the cab a big "no no" for the main wire?

thanks for the suggestions

Tim

You will need to run a large, 2 AWG min, wire for the starter and a 10AWG for the wire that charges the battery and supplies power to the starting circuit on the car. I would also run a return (ground wire) from the battery to the engine block. For the ISIS system you can just some off of the battery, with a fuse of course. Don't forget the fusible link on the charging wire coming off the battery either. You might be able to use the ISIS system as the starting signal for the solinoid mounted with the battery; I don't know for sure because I havn't done it but in theory it should work.

ISIS will save the rear light wires from the front, fuel pump wire and car audio turn on wires for sure.

I would still do a box section for a conduit. It is still a good number of wires.

Yes, big wires on the inside are a bad idea. If they happen to short to something and cause a fire there is no firewall to protect you. These high power wires should be run outside the passenger compartment.

Ron

Any reason why your mounting the battery in the trunk?

Have you seen my Braille Auto battery? It's 11.5Lbs and 5.8” x 3.4” x 5.8”.

These are work in progress pics. Don't have any of the bracket painted and correct mounting bolts. So far it has been working great. I don't even leave the tender on it.

That is a great battery! I think you went too small but the 15lb battery might just make it, even if it was cold!

Think I will put one in my trunk!

Thanks!

Ron

All that bracing will make a H-U-G-E difference if someone T-bones you at an intersection. It will make a fair difference if you put a pair of jackstands in the middle of the car and play "teeter-totter".

It will not make much difference at all in terms of stiffening the frame to resist TWIST. At least, not unless you tie all that extra tubing into a roll cage--something to provide the third dimension to the existing 2-dimensional frame. All you have there is width and length; resisting twist will require height. And since resisting twist is going to be important in a high-torque, heavy vehicle...you need to think about triangulating UPWARD as well as sideways and lengthways.

im not an engineer either, but you are correct that a cage would add the ULTIMATE rigidity. but, i actually cut my cage out before i started the frame off, and i dont have the desire to put a new one back in.

although it would seem that the bracing is not adding the third dimension (height), it is to a small extent. not only is the tubing dia contributing to height, but any drop from level with the bottom of the frame is as well. again, although small, the tubing must drop towards the center of the car a few inches to clear the floor pans. its only about 5" total (dia + drop), but any little bit helps.

in addition to that, theres the torsional rigidity of the tubing itself. im really not sure how much to expect, but i know it cant be any worse.

Tim

You will need to run a large, 2 AWG min, wire for the starter and a 10AWG for the wire that charges the battery and supplies power to the starting circuit on the car. I would also run a return (ground wire) from the battery to the engine block. For the ISIS system you can just some off of the battery, with a fuse of course. Don't forget the fusible link on the charging wire coming off the battery either. You might be able to use the ISIS system as the starting signal for the solinoid mounted with the battery; I don't know for sure because I havn't done it but in theory it should work

thanks for all that info. i need to look into a remote solenoid and see how it needs to be wired up. also, what is the main source of power for the stock system's fuse block? was it the other wires that tied into the solenoid, or am i missing something?

Yes, big wires on the inside are a bad idea. If they happen to short to something and cause a fire there is no firewall to protect you. These high power wires should be run outside the passenger compartment.

you are right, safety first. i only asked because of the stereo installations ive had done. all have had big dia wires run to the amps inside the cab.

also, your point about the firewall may be valid for some cars, but ive NEVER seen a veh fire stop at the engine compartment, unless somebody put it out before we got there with big red. it will buy you some time, and in a tc with fire, it may just save your life.

thanks again guys,

Tim

one last thing to add; my neighbor who will be bending the tube doesnt have a 2" die, so 1.75" is the max i can do, unless there is a reason for 2".

Tim

Very true that to resist twist ya need the third dimension. Very true that what you are doing will help a little with torsion.

And its the job of a civil engineer, not a mechanical.

And its the job of a civil engineer, not a mechanical.
So I don't even know what kind of an engineer I'm not? Good Grief.

Stuff like this is the core of mechanical engineering...those civil guys only build bridges and stuff. :)

There is a fairly interesting, easy to understand section that describes torsional stiffness in Herb Adams' Chassis Engineering book. Finite element is the best way to determine how effective your bracing will be, but the Adams book has good directional advice on what to do. As has already been said in this thread, you won't do any better than a cage....

actually both civil and mechanical deal with structural analysis .....how do I know I'm and a mechanical eng. and a civial sits next to me at work and we both do this type of work on aircraft

so besides a cage, what is my best option for stiffening it? also, with what it is being a "1" and a 12 pt cage being a "10", where does the roadster shops form of bracing put it?

my frame had a pretty decent twist to it. from front to rear, it was almost 1.5 degrees of gradual twist. it was also proportional to the distance from the front pt of reference, meaning half way read half the degrees in twist. when on the rotisserie and the front rotation locked, i could twist the frame back to 0* with very little effort, and i could continue to twist it 2* in the opposite direction with just a little more additional effort.

yesterday i added the watts crossmember and i made sure all reference pts were at 0* when i welded it in. once done and the frame was unlocked, i only had a .7* twist. and it stopped at the uca crossmember, meaning that this one piece is providing some torsional resistance. im guessing that just boxing the frame will get rid of the rest.

the more i think about it, it really seems as if the chevelle is just a unibody car that just happens to have the sub frames connected from the factory in the form of a full frame. this thing really isnt strong AT ALL.

Tim

the more i think about it, it really seems as if the chevelle is just a unibody car that just happens to have the sub frames connected from the factory in the form of a full frame. this thing really isnt strong AT ALL.

Tim
Add the body stiffness to the frame stiffness via fresh body mounts and hardware...you might be surprised.

Yes, the frame by itself is a weak-ass piece. The "unibody with a frame under it" was EXACTLY what I was alluding to here:
http://www.chevelles.com/techref/body_bushings.html

What you're doing will work fine for how you will likely drive it. Adding fresh bushings and hardware is a given on a project like this and will help over the old stuff but it does mean that loads are transferred to the body which causes it to flex along with the frame. Only real option to be totally happy with frame stiffness is the cage and that includes tying the cage into the front part of the frame up near the engine.

Not to argue but what I meant was that a civil (at least me) takes structures, reinforced concrete, prestressed, steel design, indeterminate structures, timber, etc while I dont know of a class the MEs take beyond mechanics of materials. Also remember no one is doing calcs on this stuff. I doubt any cage out there has anymore design into it than what makes sense.

P.s. nice link from the above poster.

i made up some half height poly bushings, and although they wouldnt work on a stock vehicle without mods, ive already modded everything that needs to be done.

check out this thread that i found.
http://forums.corvetteforum.com/c3-tech-performance/2143610-c3-frame-torsion-testing.html
hes using a "torque tube" style middle member and running round tube inside the c-channel before boxing.

im trying to figure out the amount of torsional stiffness that is being added. all i can find is calculations for sway bars, but it should be along the same line. i come up with 5690 ft/lbs per tube, but i dont know how to use this info. i guess what i really dont know is what it takes to get 5690 ft/lbs. is it any attempt to rotate? 1/4 rotation..... etc.....

Tim

5690 ft/lbs is a torque. What do you want to know? When you put a torque on a tube it will twist a bit. To calulate the amount a tube will twist is a function of the inner and outer diameter of the tube, the material property in shear, the torque on the tube, and the length of the tube. the function is that the angle of twist measured in radians is equal to the torque(T) times the length (L), divided by the quantity of the polar moment of inertia (J) (function of diameters) times the shear modulus of rigidity (g)

angle=(T)(L)/((J)(G))

In order to find J, for a hollow tube the formula is J=(pi/2)(OD^4 - ID^4).

As you can see, its a bit of work and this is just to see how much a tube will twist with a torque on it. Trying to find out how much a chassis will flex is a whole 'nother story.


Edit: this may be easier to study

http://en.wikipedia.org/wiki/Torsion_(mechanics)

Also 5690 ft/lbs must be the amount of torque this type of tube can be subjected to before breaking if it hasnt been cut down in length

the 5690 ft/lbs was obtained using a torsion bar formula (rate = 1.139E6*D^4/(A^2*L) ). i guess im just trying to see how much twist these bars will resist. my intention is to find how much torque is required to twist the frame "X" amount, and then build in resistance to keep that "X" as close to 0 as possible. i know that x=0 is impossible for my application, but id like to see how much twist i can remove.

i know that with the front of the frame stationary, that i can rotate the rear of the frame 3* with 500 ft/lbs. my intentions are to cut this at least in half. i think just boxing the c-section would accomplish that, but im not sure how much more with the webbing.

Tim

i enjoy math, especially trig, but some is over my head. i started college as a mechanical engineer major, but ended up getting my degrees in political science and criminal justice, only to become a firefighter. go figure.

Tim

so besides a cage, what is my best option for stiffening it?A "backbone" built around the trans-tunnel would be the next-best thing.

hmm, dont know if ive seen that formula. to clarify, is it ftlbs ft/lb you're thinkin of?

I dont think the tranny hump backbone would help at all as it can't provide its own torque on the frame to resist the flex, it can only apply normal forces.

How about one of these up front?
https://secure.mysuperpageshosting.com/kirbanperformance.com/store/photos/6756.jpg

and these run from the center of the front cross member to the front frame legs
https://secure.mysuperpageshosting.com/kirbanperformance.com/store/photos/6564.jpg

I don't know if they ever finshed the 68-72 kits but this may be what your looking for.

http://www.chevelles.com/forums/showthread.php?t=159800&highlight=frame+gussets&page=2

hmm, dont know if ive seen that formula. to clarify, is it ftlbs ft/lb you're thinkin of?


according to what i think im reading, that number is in ft/lbs.

can't provide its own torque on the frame to resist the flex, it can only apply normal forces.

could you dummy that down a little. how would an object apply its own torque? do you mean that if an object was twisted with 1000 ft/lbs, that the equal and opposite reaction would be the twisted object exerting 1000 ft/lbs on the object thats twisting it?

don't know if they ever finshed the 68-72 kits but this may be what your looking for.

http://www.chevelles.com/forums/show...gussets&page=2

thats a decent kit for a completely stock chassis, but i no longer have the stock uca crossmember or shock/spring bucket.


by the way, i boxed the frame today and the twist is GONE. i leveled it out prior to welding on the rotisserie, and when i release the rear rotation pt, the frame didnt re-twist. also, the 3* of rotation i was getting with 500 ft/lbs on the rear is now very minimal. ill have to get some numbers tomorrow. it looks as though the c-channel is a huge enemy of these cars.

Tim

Pardon me while I get all tutorial for a bit here... there is a point but sometimes I like to establish a common baseline of concept :)

I think the point has already been made that material at the centerline of flex is pretty much ballast. I wanted to put some numbers on just how much difference in strength and thus flex there is between the exact same material at different distances from centerline. There are some tables running around out there on steel suppliers web sites but I do not have any handy right here. Suffice to say that the change to the positive is very significant.

We all know frame stiffness is golden but what does it mean to be stiff and why is it important? There are two things we are trying to accomplish with frame stiffness, one easy and one quite hard. The first thing we are trying to do is control wheel geometry and this is pretty easy. The rear is already about as good as it can get thanks to the solid axles and housing. The front is probably decent thanks to the large profile of the stamped cross member. It could be made better using solid motor mounts so as to use the engine as a structural member. This may not be desireable for other reasons (like NVH) so we could use some form of side to side bracing ala shock tower brace. It would have to tie the UCAs together around the engine, trivial so specifics of design left to the student ;-)

Now for the hard one: with frame stiffness we are trying to focus all of the lateral weight transfer controls into the actual suspension components. Now we had quite a discussion of weight transfer back in some panhard bar thread, so I won't go into it here again. Suffice to say that tuning the chassis is a matter of adjusting weight transfer and is largely done by adjusting roll stiffness. If the frame is too soft and flexes we will not be able to move weight transfer through roll stiffness control as we would like. Thus this is where we get into twisting of the frame from front to rear. I would expect that boxing in the frame would make a noteable contribution here. If you consider that for the frame to twist front and rear cross members must be out alignment, and this requires a twist of at least one frame side member. By boxing in those side members we greatly increase their twisting stiffness.

With this understanding of the problem and objectives it is not too hard to figure out that the typical hot rod shop frame reinforcements are 90% ballast. Adding extra stuff towards the center of the car, as all of these products do, will do less good unless it can be formed as a tube. As Derek suggested a backbone or tube down the transmission tunnel can add substantial stiffness to the frame. Check out the frames on older Lotus sports cars where there were NO side members, only a large backbone. The secret to getting the benefit from the backbone is tying it into the front and rear sections of the frame. On the Lotus the suspension pickups were all formed extensions off of the center backbone. If you can figure out how to do that for an A body it will work quite well.

So I guess if I wanted to stiffen an A-body frame I would:

A) Insure the body mounts were attached correctly (GREAT writeup Shurkey posted on this one!).
B) Be sure to run full metal inner fenders up front.
C) Add in cowl flex reinforcements over the top of the engine compartment.
(note that A/B/C are all about taking advantage of what the factory engineered and put there already)
D) Box in the side rails.
E) Replace the side rails with taller profile square tubing.
F) Integrate a transmission tunnel torque tube.
G) Integrate a front end sub-cage using triangulated structural members tying into the firewall appropriately.
H) Integrate a full internal cage tying into all suspension points.

So just how much do stiffness do you need and/or want to buy? ;-) :beers:

BTW I just read and noted Tim's comment about his results after boxing in the frame after writing this. Quite a correlation I would say ;-)

Hey All,

Good thread here, been learning a lot! I wanted to chim in that Hotrods from hell http://www.hotrodstohell.net/pro_custom_chassis/index.htm will box, install their
"BOLT-IN CENTERDRIVE KIT" and powercoat for around $2500, with a week or so turn around. This is the route I'm looking at.

For those who don't wish to cut up their car, Air Ride technologies has their TigerCage. It's built out of 304 grade stainless, bolt in with no cutting,
around $2500 as well. Only thing is, it's pending its competition cert. http://www.ridetech.com/catalog/TigerCage-2490-1.html
The chevelle cage, WAS there about a month ago, not showing currently..

vrooom, can you post a link to the write up on body mounts?

Add the body stiffness to the frame stiffness via fresh body mounts and hardware...you might be surprised.

Yes, the frame by itself is a weak-ass piece. The "unibody with a frame under it" was EXACTLY what I was alluding to here:
http://www.chevelles.com/techref/body_bushings.html

vrooom, can you post a link to the write up on body mounts?
There you go... from post #21 in this thread.

And one random thought on stiffening... always, always think "triangles". Those geometric shapes are much stronger than rectangles.

Take that thought and look at those prior pictures of frame reinforcements. Note how many rectangles versus triangles you see. Open wallet accordingly ;)

With this understanding of the problem and objectives it is not too hard to figure out that the typical hot rod shop frame reinforcements are 90% ballast. Adding extra stuff towards the center of the car, as all of these products do, will do less good unless it can be formed as a tube.

i guess i differ slightly on the comment about the inner structure. my thought is this; in order for flex to occur, the front and rear portions of the frame must be out of alignment as stated. if the bar connects to a portion on the front and a portion on the rear, the bar will twist in opposite directions. this will place torsion on the bar which the bar will try to resist, just as a sway bar does for the suspension. how much it does, i still havent been able to figure out. but, if a small dia bar can keep a car from leaning in a turn, it should be able to do the same for the frame twisting.

any opinions?

Tim

If you had a bar connected to the front crossmember and to the rear cross member your analogy would be accurate. But you do not. You only have a structure member parallel to another in a plane perpendicular to the bending moment of interest. This puts it basically on the centerline of the bending moment and renders it least effective.

The other factor you have to consider is the length of the bar and amount of movement. As the bar gets longer it's spring effect is reduced. Further as the distance/length increases so does the amount of movement. Put another way if you have a bar that twists .001" per foot for a given torque then 10' of bar will twist .010" or 10X as much. Take that and work the resultant angles across the frame and you have significant twist.

If you had a bar connected to the front crossmember and to the rear cross member your analogy would be accurate. But you do not.

http://www.roadstershop.com/images/Picture164.jpg
i guess i dont see how this isnt connected to the front and rear crosses. i understand it wont strengthen the area fore and aft, but from the front kickup to their new rear cross, it seems to be as you state.

This puts it basically on the centerline of the bending moment and renders it least effective.

could you explain the bending moment a little further?

As the bar gets longer it's spring effect is reduced. Further as the distance/length increases so does the amount of movement. Put another way if you have a bar that twists .001" per foot for a given torque then 10' of bar will twist .010" or 10X as much

i fully understand what you state. if i had a 1' bar and 1" twist it would be more effective than a 10' bar and 1" twist since each foot only realizes .1" twist. but, wouldnt the bar see the same amount of twist as the stock frame rail?

Take that and work the resultant angles across the frame and you have significant twist.

could you explain this a little farther?

also, how would the tunnel brace be any more effective than the previous displayed webbing?

thanks, Tim

Lessee... taken a bit out of order...

The stock frame rails are about a 4" tall section spread out to about a 4' width. This puts them at the ends of the cross members where the movement is greatest per degree of rotational movement.

Now we attach some smaller diameter members which are going to be much more flexible because of their size. And we attach them closer to the middle of the car where the cross member movement is relatively less. So they have less leverage to work against rotation or the rotation has more leverage to work against them. Double jeopardy much like calculating motion ratios of SLA suspensions from spring rates.

For bending moment the extra ballast members are basically in the same plane as the frame. A piece of paper is really not very strong torsionally. For the cross members to rotate both sides of the frame must bend along their length by some amount. This bend will have a centerline and the added structure is pretty much on that centerline. Since it is of smaller cross section than the existing frame it is relatively weaker.

The reason the tunnel works is two fold: first and foremost it moves material further from the load carrying centerline increasing the leverage of the material against the load. How high/wide is your transmission tunnel? A LOT more than the 4" frame height. The downside is the driveshaft tunnel would need to be enlarged a bit but again would still be quite a bit bigger than the 4" frame. Secondly the forces are changed from tension and compression to shear with a much greater area of material working for you in carrying the load.

I may be overstating things when I refer to the typical frame add-ons as "ballast" :D They will probably add some strength but not in proportion to your wallet reductions or the weight added. I guess at least the weight is in a good place though: low and near the center of the car B)

thanks for the reply......

i think we're thinking about things a little differently. im on this page;

take any given member of the fame and keep one side constant while twisting the other. without the member moving from its original plane, the main force is a torsional one. this is the only force im trying to resist.

now take the frame and keep the front stationary while twisting the rear. lay a straight edge across the front crossmember and you get 0*, while one across the rear nets you 3*(just pulled that out of the air) and you have a 3* difference front to rear. the measurement taken inbtwn the front and rear will be proportional to the distance from the original measurements. but, any measurement taken anywhere along the straight edge will give you the same measurement, no matter where on the straight edge the measuring device is set. meaning both rails, one foot in from the rails, two feet in, in the center..... all net you the same * measurement. this means that any member running front to rear, no matter its location, is twisting an equal amount. it becomes different than a SLA motion ratio because the ratio of twist does not change.

what are your thoughts?

Tim

So given a frame fixed across 3 points: either side of crossmember A and one side of crossmember B. We jack up the other side of B and measure the force and displacement. This would be one reasonable (and probably the most critical) measurement of frame stiffness. I think this is the same page you are on.

You are correct that all points along crossmember B will be twisted the same angle amount.

The rotation of crossmember B will be resisted by 1) twist resistance of one frame beam and 2) bending resistance of the other frame beam. You are focussed on 1 and I have talked about both :D

For 1 you need a closed structure as you discovered when you boxed your frame. Note that it only takes some closure and not a full channel box. This type of structure is used quite commonly on aircraft in the wing leading edge. The metal/wood sheeting is added top and bottom from main spars to leading edge and then shear webs are added between the spars. Try this with a balsa wood model sometime and you will find the structure is locked into place after the first bay of shear web is added.

Once you have a closed structure then the larger the "diameter" the better (this is where the tunnel excells). Again moving the material further away from the centerline of rotation increases it's leverage against the rotation and requires more elastic deformation per degree of rotation. Both work together to greatly increase torsional stiffness.

Now if I have a 4" torsion box in the original frame it will have some torsional stiffness. If I add a 1.5" bar alongside... well that bar is rather small compared to the frame so it is working at a disadvantage. It does have some stiffness which is added though, in particular using an additional rear crossmember as they have in your example pictures.

Note that they probably get more benefit from the fact that those inside tubes connect inline with the rear suspension pickups than anything else. Those angled side tubes are placed more for bling factor than structure. The back angles are not bad to triangulate with the suspension pickups but then the next tubes should go straight out or angle forward. Instead them parallel them with the rear making a cool looking but relatively weak box structure. Those side tubes could be a lot smaller and accomplish the same function too.

Which brings me back to #2 above and bending of the added tubing. Note there was enough need to reinforce the span of the center tubes to add extra side tubes? What about vertical bending of those center tubes? Much less reinforcement in that plane.

thanks again for the reply. like you said, i have been focused on #1, but could see how #2 comes into play.

ill have to see what i end up doing with the rear lca's. the plan was to leave them as is, but with the added crossmember, it wouldnt be hard to just make different lowers that were a little longer. if i end up doing that, the inner structure would serve a purpose, but mainly to reinforce the lca crossmember and keep it from pushing forward due to the weakened nature of it with the ds and exhaust holes cut in it.

Tim

I know you guys are talking about a lot more in depth than what I did, but our 67 Wagon frame was SO WEAK it was incredible. What I did what get a center section out of an El Camino/convertible and welded it into my frame rails to "box them" like the factory did on the weak original cars. Then I welded up all the seams, added 3/16" plate to the rear crossmember (the one in front of the rear diff), then boxed the rear sections of frame rails, and replaced the stock rear crossmember (at the very rear of the frame) with a 2x3 heavy wall box tube. It is VERY stiff now. I know I could have done a lot more, but decided to stop someplace! :) THanks for all the other thoughts and info! :)

What about building a "cage" for the engine bay? It would be 3D. It could come off the front frame rails and bolt to the firewall. (so its not permanent) Then you can add another cross bar simialr to a trut tower bar on something like a late model Camaro. You would probably have to reinforce the firewall where the tube meets it...

The closest thing I could find is the Chris Alson's Nova front clip. It doesn't have the drive sidew to passenger side crossbar I was talking about. The key is thier Gemini Connector on the front of the frame at the dwon tube. They claim its stronger then a regular bolt on.
http://www.cachassisworks.com/images/7700at.jpg http://www.dragracingonline.com/technical/images/upfront_05.jpg

That was pretty much what I was referring to in option "G" several responses back. You would need to study the structure in the firewall area especially over and around the sides. Look for seams where several pieces of metal come together. Things like that. Visualize where the loads are going to transfer and look for loads going into shear not bending. There may well be a fairly reinforced area around the door pilliar and hinge area.

One thing to watch out for here though is building spears into the structure. The front frame area needs to collapse in a collision rather than punch through into the passnger compartment. If it cannot collapse, it must be designed at a minimum needs move outside the bubble of soft stuff. Look at the hooks on the back of modern hoods for an example of forcing parts to crumple rather than intrude. Look at the front structural design on Subarus where the engine is supposed to compress underneath the passenger box instead of into it.