Started this same thread, earlier today, but someone destroyed it w/in 4 posts. This was a HUGE success, elsewhere, so I'd like to try again...

I started a thread similar to this on a 'vette forum, about C4 Corvettes...because I had little luck finding technical information on that car's frame. The thread turned into a windfall of information, so I thought I might try the same thing here.

Likewise, there is the ongoing argument regarding the effectiveness of various frame stiffening products. But as some people have commented, there is no objective data to prove that these parts actually do anything meaningful. The first place to start is w/the original specifications for the frame. Torsional stiffness, bending, etc. IS this type of data out there? Can we post it up here? How about drawings of the frame? What data exists for this platform? Post up!

 

Good luck, I've posed the same question many times.

From what I have seen, the standard test to measure torsional rigidity is to rigidly fix the suspension (front and rear) at ride height position (I figured i could do it with some old shocks welded into this height) then rigidly mount the ends of the rear axles to the floor via a beam. The front you would do the same except at the center of the beam you would place a fulcrum, you then add weight to the end of this beam and figure on how many ft lbs of force is required to deflect the chassis 1 degree.
Of course that would only tell you torsional rigidity and not flexural rigidity along the longitudinal axis of the car.

Now even though I'm sure you can stiffen up our cars significantly, I've often postulated that the result isn't necessarily going to be an improvement in handling. So far no one has done a scientific test to show otherwise. I think the fact that thirdgen F bodies have crushed the competition for years in E street prepared WITHOUT things like sfcs and STBs is at least some empirical evidence demonstrating that it's certainly not a significant detriment to not have them.

Now as far as stress cracks and whatnot are concerned, there may be some legitimate rationale there as a preventative measure. I have seen my fair share of stress cracks on thirdgens. But then again I have seen them all over multimillion dollar aircraft too so that puts things into perspective for me.

 

I should also point out that in order to truly conduct the torsional rigidity test I mentioned, you would have to have solid bushings. You would also have to consider that loads are shared between the shock mount and the spring pocket. So ideally I would think you would want some sort of solid connection on both the shock and the spring pocket so that the loads are transferred at least evenly. Even then, that wouldn't be 100% accurate because the shock and the spring transmit energy to the chassis at different times to and to a different degree.
But, I think that just doing the solid shock plus bushings isn't a bad baseline. From what I've read you want to make your shock mounts the most solid anyway since its the only point capable of dampening susp movement. I.e. if your spring pocket were to flex upward in a non linear fashion to your shock/strut mount, you could at least dampen it like it was part of the spring. Hopefully I'm making sense.

 

I'm reading your "similar thread" and I have to ask what the purpose of this exercise is? I'm going to play devils advocate, because this really is a can of worms being opened. Having the factory claimed torsional strength numbers would be nice (I've never seen any myself for these cars), but I'm not sure how much merit they would have on a mass produced car let alone a thirdgen. That is not to mention the aforementioned wear and tear, and accident/road damage as well.

So in order to quantify changes and their effectiveness, and do so in a manner that is a little less destructive and allows for other ideas to be tried, the welding that is so common in practice in attaching structural items such as SFC's needs to be replaced with properly bolted connections for testing purposes. You would need to test the body itself before (making the factory strength numbers kind of useless) and after, and then try something different and test again. That is the only way you will have a good, quantifiable piece of information. As a bad example (that was in yet another locked post):
http://www.musclemustangfastfords.co...ang/index.html
The more I thought about this, the more I realized how flawed their method was. One of the things they did entirely wrong was measure the change across the door jamb with the doors shut. How much do you think the door could deflect differently with the striker in place, all the rubber bumpers in place, glass closed, etc? So having a good/proper method is a must. Another thing is this is simply static testing, what happens dynamically is entirely different. Frequency plays a part in this, but there are a lot of things going on.

On that note, given the complexity of the body of the car, the panel thicknesses, the doubled up panels (even triple in some places), the spot welds (often somewhat "random") and other mass production anomalies, it would be... really difficult to model the chassis accurately. This applies to the C4 as well, BTW. A solid works model that may not (will not I'll bet) 100% accurately represent the body panels, thicknesses, and welds (including strength of properly joining panels) could easily misdirect someone in correcting problems that don't exist in the real world, and missing problems that do exist in the real world. Its nice for general analysis, but really the only thing that might be of use from that would be finding serious flaws in the General's design. There are problem areas, but I'm not sure there are any real serious failure points.

So on that note:
High strength steels have more carbon. They bend less, take more load before breaking, but break catastrophically. They do not weigh less, common misconception. Steel is steel, it all weighs the same. The difference is, stronger means you can use less if you dont need the strength.
The vette2vette rods I believe I was asked about around the same time they were posted over there. After thinking about it a while (did not have the answer offhand when it was asked) they're tension rods. The basic idea is to put the piece/s they're attached to in compression. This is similar to the concept used in post-tensioned bridges, although the reason is slightly different. Concrete is fine in compression, and is basically useless in tension. They run steel cables through the bridge, and often tension them after the concrete is poured and cures. This transforms the bending stresses from tension and compression (depending on where in the concrete it is) to compression everywhere. Probably what they do in the C4 is reduce any potential resonance or change the resonant frequency of the chassis, and its also preloading that piece which, from what I could see, would counteract the natural bending tendency of that portion of the chassis.

Pretty much any vehicle, torsional strength/stiffness is the primary concern. This absolutely should NOT be confused with the torsional strength of any one element in the frame. For example if the body is twisting along its centerline F-R, any element directly centered on that centerline will be in torsion, and other elements not on that centerline will see some torsion from 100% along that centerline to zero perpendicular to that centerline. This little detail is very often confused and is why there are so many locked posts on this subject. Put simply, just because one element is in torsion does not mean they all are, nor does it mean any of the rest of them are.

There are a few problem spots in these cars, from design. I would not really raise them to the level of a flaw on the part of GM, but surely things can be improved. Everyone knows about the cracks on the roof. This is just the result of the L shape where the glass is, and the fact lots of force is being transferred through there. I heard a theory, I believe from 1 Mean Z, that poor spot welds on the inside break and allow this to happen. Its almost a chicken and egg argument, which came first? Is it poor welds, and then damage from too much load, or is it too much load, that tears apart ok (but not enough) welds? Anyway, an L shape results in a very high concentration of stress at the angle point in the L. Fact of life. Given that these cars also like to crack the A pillar at the base, I'm going to say there is just too much force trying to be transferred through the roof and it needs to be carried elsewhere. SFC's.

I have seen a cutaway of the front and back sub"frames" (I use the term "frame" on these cars very loosely, its mostly built up sheetmetal sections), a picture of the underside that is pretty nice, and recently someone posted up dimensional drawings of the chassis for checking to see if their car was bent. I dont have any of this saved, but I'll make an attempt to find them.

As far as reinforcements, I've been trying to come up with a couple ideas up front but I'm not too far into researching it so far. Basically tying in the strut towers, the front "frame" to the body, the rear "frame" to the body, and the rear "frame" rails to themselves.

I've been contemplating trying to provide kind of a guideline on structures recently, but still not sure I want to post anything. Been working on something for years, far from done.

 

It would be more about making the motions of the body predictable and consistent. For example, the jacking effect of the rear panhard bar. You know its there, you compensate accordingly and deal with it. Some choose to eliminate that "feature", stiffening the chassis would fall along the same lines.

Exactly.

On that note, last time Pablo posted up about this and put up that MM+FF garbage article I was looking around and came across a guy with I believe a 1stgen Rustang who was testing torsion of the body and the significant differences of deflection depending on where he stressed the chassis from. I think there was also a website where a guy with a 'Cuda did the same thing, probably with similar issues and results. Places like Hendrick have more money than brains and accurately load the chassis, even dynamically. None of us have even remotely close to the funds needed nor the expertise to accomplish this kind of thing. Its more of a guessing game for us.

 

Hey Monster stock C6 guy, Whats the tortional load in lbs per degree of a poly urethane control arm bushing for these cars...molre specific, the rear A-arm ear- not the front?

As madmax summed it up in 8 paragraghs, this whole subject is just plain stupid.

 

I don't know the number but when I built my chassis for my vette I simply bolted a big beam to the front and rear suspension with the frame up on stands under the susp mounting points (3 of them, the corner with the weight had the stand removed). Then I placed a known weight at a known distance on the beam and measured the deflection. I repeated this and checked it again after adding beams/bracing to get an idea of the result so I could estimate whether i needed to add a little more structural stiffness or if it would be a waste of material and dead weight.

I'm not am professional builder, it was my 1st attempt at building a structure like that, i also built the front and rear double a arm coil over suspension and the front end take off r&p steering system. I'm by no means a guru but I have tinkered a bit on these cars (I'm an autodidact, have no formal training or books)

 

C4 vette isn't classified as a unibody is it? Either way I've driven both C4 vette and 84 ws-6 trans am and even a later IROC ... sorry but a f-body can't compare now was it ever intended to do so. I would think even with a f-body heavily mod'ed it still would have it's work cutout staying with a C4 vette that is stock.

I've driven a base suspension FE1 vette and so a Z51 C4 vette would even put the hurt on a f-body even more so.

 

Please stay on topic.

JamesC

 

Maybe the real question is not whether stiffening the chassis improves "handling" in and of itself, but how any such added stiffening makes suspension tuning efforts more responsive to changes in springs/bars/shocks.

At the ridiculous extreme, a chassis that has a torsional "hinge" somewhere in the middle of it cannot be tuned for understeer/oversteer handling balance no matter how much roll stiffness you add to either suspension (or to both, for that matter). You simply can't drag roll moment that wants to be taken at one end through the chassis over to the other end (which is what you're doing when you play with roll stiffnesses). A unibody convertible with rusted-out sill structures (and enough tunnel strength remaining to keep it from collapsing under its own weight) probably comes closest to this.

I am assuming that there is enough chassis stiffness in the transverse planes of any independent suspensions to keep the cambers from going all wonky.

Locally, spotwelded panel construction is not quite as rigid as the calculations for identical one-piece shapes predict. There are small deformations at the seams that cannot be avoided, and I'm pretty sure that roof cracks happen due to fatigue effects from this sort of local flexing. It'll concentrate right at the edges of the spotwelds since those are points of sudden change in local stiffness and stress flow.

The only paper I've seen that goes into the effects of added stiffening is the Clemson paper that was done a few years ago for a NASCAR chassis.

http://www.ces.clemson.edu/~lonny/pu.../sae983053.pdf

This might also be of interest (just found it while searching for the above link, haven't read it yet)

http://www.ces.clemson.edu/~lonny/pu.../sae983054.pdf


Norm

 

What is so hard to understand the simple addition of a cage to box 3 dimentionally the suspension points and the fore and aft chassis? This is done in any car. not just 3rd gen related so a model of the entire body flex is pointless.

Things all goes to deflection of the weakest link in the ENTIRE chassis and suspension combined as one. Who cares if upper weight and structure of a vehicle is bending and flexing, as long as the suspension points remain in tact on a dimesional scale.

I have said and will conmtinue to say that this thread is pointless beyond the other thread already closed when the anwers were givin in the first 3 or 4 posts about a good aftermarket support already exists and has been proven to work.

Are we going to birdcage the entire cockpit so maybe event he driver will not fit just so we reduce some roof panel cracks - just cage the car for optimum performance of the supension points----Your done.(With ANY car)

 

You would loose alot of money if you ran against me. Your assumption is far wrong. there is nothing common on the street that I can not walk up and around on any freeway cloverleaf or sweeper at about 25-100 mph lateral range.

They are all just cars with tires. Wreight to grip ratio is what counts- In other words, its how much weight on each tire and how that tires footprint statyys in contact with the ground. It does not care what kind of cat the tire is attached to- the tire doesn't care, its how the chassis and suspeion work in conjuction with that tire to keep it and its 3 cousins all in optimum lateral grip........
.......its all about the person that adjusts the suspension.

Last night I was doing a chassis setup for this weekends race. We race spec Supertrucks in NASCAR. Rules provide that all trucks be basically equal in parts aditions and restrictions. So..Why are some trucks as much as 2 seconds faster than others with all the same parts? its the people working on them for the most part. Their was a figure published that analyzed the work of the crew vs the work of the driver in importance of a winning vehicle. That percentage was posted as 70% car/ 30% driver.

Last night I did a different setup with my new driver (new team I am on- 1 race history so far) and basically took about 7 hours adjusting the same parts he has had on his car all year. To take one particular example of what I worked on for about 1 hour was bumpinmg the car through the suspension travel with toe settings. Sounds simple right? Most just set it somewhere between 1/8 to 1/4 toe out. Not me.

The shock indicators are showing 1 7/8" travel on the LF and 2 5/16" on RF. So I bump it to those ride height travel ranges. I am running .102 gain of toe on the RF and a .014 loss on the RF. When the car tracks downt he stright I am at aporx .150 toe out(just over 1/8" out down the straight, and when it compresses under braking and corner entrance it toes out on the drivers side more increasing overall toe to .238 or just shy of 1/4" toe out for turn in.

How many people take the time to check bump never the less set it accordingly for less drag yet still have corner turn in adj.?

You ever check stuff like this on your C4 Vette? Gm engineers have gotten better at this kind of stuff over the decades so this is why chassis feel much better stock, and why when people change stuff themselves they have no clue as to why they are hurting their factory specs in light of the more performance upgrades they made. Keeping the suspension factors in check along witht he upgrades and the car wil be incredible, most poepls cars are not due to lacking foundations like this example.

Do what you want to your car, whether its a Vette or a Camaro, or a Transam, but please do not go around stereo typing a car because of its emblem- that don;t mean crap to someone like me that knows how things really work. Your vette is just anoher 3000+ lb car with tires that needs adjusting to make things work to perfection just like any Camaro or transam. Tires do not dicriminate with car brands or emblems- just wheelbases, and footprints combined with a marraige of chassis and suspension parts & adjustments. I can make a Yugo beat a Vette dollar for dollar against the average car enthusiets.

 

Guys; I'm not necessarily looking to "solve a problem", or "reengineer" an F-body at the moment. At teh MOMENT, I don't even have and F-body. Still, I have plenty of INTEREST in F-bodies...and I also have INTEREST in body structures. Which is the point of this thread; body structure. That is it. It's not about "handling", or what is the best way to stiffen an F-bod, or how to prevent roof cracks....it's simply an attempt to learn something about the stock structure. And to do so, for no other reason than to satify my opwn curiousity, and hopefully (if this thread actually produces data), become a useful reference for others. Is any of that a crime? Didn't think so.

The way I see it, information can compile here in two ways; engineering papers on the design and criteria of the 3rd G F-bod, and/or owners actually measuring the chassis performance (strength) of their own cars.

Pablo and Twin Turbo were two posters who seemed to "get it" the best, and I agree. To measure the torsional stiffness, you'd need to stand the car, rig up a "degree wheel" (of sorts), and use a dynomometer/strain gauge to measure deflection in ft lbs/some measurement (degree). If I had an F-bod at the moment. My hope was that some f car fanatics here would have GM docs that talk about this type of subject matter.

For those who think this topic is "stupid" etc.; don't post or read it. No one is forcing you to. It's not "stupid", b/c it is a topic of INTEREST, to ME -and some others on this forum, it would seem.

 

^This is excellent. Good stuff, and thanks for posting.



Anything would be great. Thanks for looking.

 

The cars will do this if you brace them adequately with standard aftermarket products. The picture is proof, it does not lie.

The cars will teeter totter on a curb diagonally if you put a rear wheel onto a curb and leave the LR off. The car will teeter here from the LR to the RF. Do a measurement with the car in the same exact spot 9same tire pressure) and then add you chassis braces etc and keep checking this refence....Was that so hard? Its been done.

Why go through all the trouble of jigs and gauges and expensive deflection monitoring devices- thats just stupid and a waste of time and money.

Curbs are free to use and everywhere, so are floor jacks and tape measures. Tom, you are over analyzing this crap.
Look at the picture in post #12
https://www.thirdgen.org/forums/susp...uspension.html

 

Here is an excellent point. A car may have percieved "handling deficiencies" that are a result of...whatever. In this case, we are talking about structure. The structure, may (will) in certain conditions, behave in such a way as to provide feed back to the driver, that is undesireable and distracting/upsetting, etc. The PHB Jacking is an example of that w/in the realm of suspension operation.

However, before you can fix a structure "problem", you have to be able to observe that it exists. Hence, this discussion and an attempt to gather data, if it exists. Applying "fixes" in the right places may or may not actually change the performance limits of the car (it's grip to the surface), but it may dramatically improve the driver perception and experience. That's a big deal (to me).

 

Nope. I'm not. Remember that this is of interest to me.

 

All I can find right now: (the frame dimensions are apparently in the factory service manual)
Notice the one with the body cut off how the rear "frame" is just a bunch of sheetmetal, much like the rest of the "frame" on these cars.

 

Im going to say this up front I don't really have much on technical specifications for third gen stuff except what is available in the factory manual and unfortunately even that I don't have a way to post here. I will say the factory manual has a section in the back that shows the main frame structures and dimensions. If your interested in general chassis structures and designs there's tons of info on that. Unfortunately that topic is so vast we would need to be more specific as to what it is your trying to learn or accomplish before could even begin to talk about it. Maybe it was in your original post but I did not see that one.

 

Good point. I guess the most interesting thing to me is the torsional strength. I understand that it likely degrades over time/use, but a factory spec is something. It's a starting point at least, for comparative purposes. People talk about "how flimsy" these cars are. I KNOW that they are, but compared to what? Fox bodies are flexible fliers too, but how much? People say a C4 is "better"...I can drive my C4 down the street with the roof off, and easily watch the whole front of the car (steering wheel forward) moving independantly of the rear (where I'm sitting). I don't recall anything like that in my F-bod days -though I do recall a good amount of flex before bracing was added.

 

Well if your looking for a strait up comparison between particular models I think you'll have your work cut out for you. Not to mention as noted what may have been true 20 years ago is no longer for most cars and the rate of how fast these chassy degrade over time wont be the same. If your comparing different designs thats even more challenging because some are designed better than others, with different materials, different weights, and with different requirements. You can get general ideas about whats best for what application or a list of pros and cons but it will be hard to get much more out of it than that. If your primarily concerned with torsional strength I dont really think the third gen f body is really a good place to start researching unless it was to use as an example to show the short failings of unibody construction. The classic unibody type construction used to produce these cars is less than ideal. The unibody construction was chosen primarily because it was a easy and economical way to mass produce a car. Unibody construction is actually almost the worst design considering weight vs torsional strength with the exception of the classic ladder frame which is worse yet. You probably already know this though since you have a vette that also has a good deal of flex. The problem is factory designs for your every day car involves a lot more than its actual functionality and thats why we've seen such compromised designs. However if you wanted to look into more interesting designs check out the tubular space or backbone frames used in a lot of higher end cars.

 

Tom, people spend a lot of money on tubular this and that. What you are essentially doing is making them uncomfortable by asking for some kind of proof that it helps. No one wants to believe that the emperor might not be wearing clothes. This is why people get upset over this kind of thing.

This is an interesting site where you'll see a company that actually puts R&D into what they make http://www.xvmotorsports.com/enginee...c_rigidity.cfm
They have some neat videos showing chassis flex on a test setup that looks like hydraulic rams actuating the corners of the car up and down. You can also see their torsional rigidity test rack which looks like they use a hydraulic jack to measure the amount of force applied away from the fulcrum. I have also seen setups where there is just a mount to put weight plates.

I've always wanted to build a torsional rigidity test fixture myself but have way too many projects to want to get into that.

One thing i have noticed about thirdgens is that there seems to be some longitudinal flex accounted for in the design of the car. If you look at the transmission tunnel you will see channels along the lateral axis of the car stamped into the sheet metal. They look to me like areas that are designed to allow flex along the longitudinal axis and prevent buckling.
Another thing I have noticed on at least 3 thirdgens are stress cracks and buckling in the trans tunnel right above the torque arm mount (where there are no channels) I think the logical explanation is that the rear axle and consequently the rear of the car wants to rotate about an axis that is the front torque arm mount. Its not hard to see how stress cracks form at the corners of the window openings.
Actually I'm surprised they don't crack sooner being that there is basically no radius to that corner. Any aircraft that I have worked on that has a sharp corner like that is basically guaranteed to crack which is why a lot of the drawings specify corner radius dimensions. There is no way to completely eliminate flex so you have to account for it in design.

One quick thing to do to give you a qualitative idea of whats going on is to drive around with your finger pushed into the top edge of the door. It looks silly but it's kind of enlightening even if unscientific. I find that I feel significantly more movement from road bumps like reflectors or pot holes than by yanking it around corners. Someone at GM thought it was a good idea to put the door shims on T top cars for a reason. I added them to my hardtop (shimmed tightly), I cant help but think they stiffen the car longidudinally by adding some metal further away from the neutral axis of the bend. vs just having the rocker boxes, trans tunnel, and roof (which seems like it would buckle easily in elastic deformation) .

Those are just some of my thoughts on the issue.

 

By the way, about this chassis degradation issue...

Springs do not lose their rate as they fatigue. The only way to lose spring rate is through cracking (Ive read some talk of "micro cracking" being a factor) or some other damage to the metal. So unless the chassis is actually cracked, the "spring rate" of the chassis should remain the same.

 

If unibody is so bad then why do nearly all aircraft incorporate it? Monocoque construction just makes more sense when weight is a concern, why carry around panels that serve only one function (i.e. aerodynamics) when they can be both structural and aerodynamic? It just makes more sense from a design standpoint. I understand that in practical hot rodding applications, it makes modifications more difficult though.

 

Well Ill say right off the bat I am not qualified nor familiar with aircraft construction to speak on that matter however I can speak about cars which is what were talking about. Unibody construction isnt "bad" it has some advantages which is why it is so commonly used today. However its not used because its the best performing design and in terms of strength vs weight especially when you look at the 80s rendition of a unibody frame. First off i want to clear up something dont confuse stressed skins with unibody construction. Unibodies do use stressed skins but even in other types of frames they do use stressed skins for example floor pans and firewalls are commonly tied into the frames in non unibody cars however theres more to it than that. If youve ever held a front fender in your hand its pretty light. This is because its not a stressed member and its made of very light gauge steel making it light. Now if youve ever picked up part of a rear fender youll see its much much heavier this is because its made of a much heavier gauge material because parts of it are tied into the frame. However this is not very efficient to have a heavy rear fender when a structure of comparable weight could be built to offer better strength. On a space frame for example you can make the body out light weight fiberglass or aluminum and use the save weight to add supports to your space frame. Also when you look at parts of the body that are stressed there not exactly ideal shapes to add to the strength of the car as they are designed for appearance not necessarily maximum strength. For example again the rear fender I think we can all agree is not the best shape to give the car strength. Also as we all know these unibodies are many of many stamped parts. The problems with stamped parts is you have use a heavier gauge metal than is actually required because after stamping the pieces the thickness will not be uniform but you have to ensure a minimum thickness. This is not a problem in other types of designs. Not to mention again the whole shape of a unibody frame is going to be designed with the shape of the car in mind not necessarily what is the strongest shape. Another problem with the many stamped pieces is attaching them all together. Typically factories will spot weld them all together which is not really the best for strength and if you did fully weld them together you would add a lot of weight. Today theres lots of new technology to improved the classic unibody design such as glues to help hold pieces together and hydro-formed pieces but fact remains for performance of the frame vs weight ignoring all other factors unibody construction is not the best design out there.

 

Tried my suggestion anyway huh?

Its possible for there to be degradation, that is not visible or quantifiable without testing a small portion. Maybe some sort of microscope, sonic test, etc might show this but honestly I have no idea. I've seen discoloration form in most materials so with a microscope there may be visible evidence. With painted surfaces, who knows.

The old school image is courtesy oregon.edu I think it was. Once you get past the elastic deformation area and hit the yield point (B on the diagram), depending on the material it may deform, it may work harden, it may show no deformation, and ultimately it may break (F on the diagram, or the end of the line that says Hard Steel). In any case past the B point the material has definitely been compromised to some degree, but it doesnt necessarily break from this alone. So there's a pretty large range with a common mild steel (the sheetmetal will mostly be this material) where it may not visibly to your eyes show damage, but its definitely been damaged.

 

Rolling thunder, great post. You bring up some excellent points. Unibody as its done for cars is indeed a compromise. But then again so is a space frame setup in a race car. If you could choose a style of building a car from scratch though, I would put my money on monocoque construction being lighter and stronger. If the thirdgen was a factory body on frame car I would think it would be a bit heavier. So in the sense that its lighter is at least an advantage even if it may be less stiff because of how its implemented.

 

Wait, maybe I'm not understanding, but I take what you are saying to mean that once you go past elastic deformation you damage the metal. I understand that. But then you say it may look just fine even though its damaged internally. Wouldn't it have to be bent if you went into plastic deformation and thus visible? That is of course unless it somehow was bent back to its original shape. Do you think the chassis is going into plastic deformation and then coming back to its original shape? Seems kind of unlikely to me.

BTW about "your test" I noticed this as far back as 1995 riding in my buddies 87 T/A. I would rest my finger up in the t top gap and could feel lots of flex.

 

Is it possible to build a poor mans chassis flex tester to do this?

For torsion, a chassis rotisserie with one end fixed so it can't rotate and the other have a torque input on it and an angleometer so you can tell how much torque equals one degree of rotation.

For longitudinal flexture, something similar to the rotisserie but fixed on both ends and a tension force applied at the center of the chassis then measure the force required to deform the flex the chassis something like a half inch.

Sure, it wouldn't be as awesome as a 7-post test rig, but we could start to get real data and begin to research how our chassis react under load.

 

Yeah a cheap way would be like what I mentioned in one of my posts above, you really just need two beams, some solid bushings, shocks welded into position. Get some old steel wheels, weld them to the beams. Put a fulcrum on the front beam and then apply a force to the end of the beam. See how much your beam moves from 0 deg. Of course you'd have to be sure you didn't start picking up the corner of the rear beam.

Oh yeah, from what I've read, you want the fulcrum to be around where the roll center of the front suspension is

 

Yes, it should show up as bent metal being a mild steel but the question is, are you going to see it? Probably not. Initially on an exterior panel you might notice a wave in the panel. The underbody, I dont think anyone would notice a change until it was too late.

 

Well everythings a compromise there is no perfect frame and thats why there are so many different designs out there. This is why engineers get payed big bucks to figure out what is the appropriate frame for a given application and design it too meet the requirement. They all have their pros and cons. From the perspective of most automakers the unibody chassy is simply the right way to go its easy and economical to produce and does not have the space restrictions you have with other types of frames. On the other hand it would not have been practical for GM to produce a space framed f body though it would have been lighter and stronger if done. It was a compromise they chose to make and a compromise most automakers continue to make and rightfully so. It also is a step up from a typical ladder frame with a body bolted on. If a third gen had a full frame on it its hard to say if it would have been heavier or if it would have been stronger that depends on how they would have designed it. I can say though that I am very confident it would not have been stronger for a given weight. So the unibody was a step in the right direction over the classic ladder frame. Though I am curious why you say a space frame would be a compromise in a race car? Most scratch built race cars im familiar with do use space frames because it is the best choice to build a light weight frame generally speaking. The only exception I can think of would be say formula one cars which are full carbon fiber now but even those used to use space frames until then. Actually thats where space frames were adopted from originally if I'm not mistaken.

 

I think the fact that carbon fiber F1 cars are monocoque shows that it is the way to go. I think the reason you see space frame construction of metal tube race cars is because its not economically feasible to form or mold a monocoque chassis from metal. To do one off stuff, its easier to weld up tubes, especially when there is a long knowledge base on the construction of these cars. Carbon fiber on the other hand can be tubes or can be formed to various shapes pretty easily. They could have easily made a carbon fiber tube chassis but it wouldn't make sense from a structural standpoint.

 

Flash back. I had forgotten about that, but I used to do the same thing in my TA's and Camaro...and think about what that movement meant. Recalling that now, I am even more convinced that the Fbod my actually be stronger than the C4 Y car. I should shoot a vid of this car driving down the road.

 

Hmm, so basically, in order for the chassis to lose its rigidity it will have to have cracked or undergone plastic deformation. Do you think there is that much "invisible" plastic deformation going on?

 

Well actually if im not mistaken I believe one of the big reason why they went to full carbon fiber was for safety. When you look at the history in how they evolved into full carbon fiber cars it stemmed from safety issues where carbon fiber panels were added to protect the driver. Prior to that to my knowledge there wasn't much use of carbon fiber save the bodies. Carbon fiber is definitely not practical in most cases as it is expensive to work with and space frames would be considerably easier and cheaper to produce not to mention repairable but im not convinced they went to carbon fiber because it was a better way of doing it only a better way given their restrictions.

 

Only in and around the typical areas these cars like to fail in. Generally speaking, they're fairly solid. There are some weak areas, such as the entire roof structure, the strut towers, the front frame where the idler and box mount, the front rail to firewall/pan/rocker transition, the rear LCA area, the rear shock mounts on the body, probably some other things I'm forgetting. All somewhat minor aside of the first two.

 

Well, Im no expert in F1 construction but I can tell you that carbon fiber is not a good choice from a safety standpoint. When you exceed its yield point it just breaks unlike metal which bends. Carbon fiber is also very bad in impact and abrasion resistance. Actually aramid (kevlar) would be a much better choice for protection. This is why bullet proof vests, helmets, and the armor surrounding cockpits in combat aircraft is made from kevlar.
I think the main reason for going to carbon fiber is its PHENOMENAL stiffness to weight ratio and its workability. When you work with carbon you are kind of like creating a metal at a mollecular level. Depending on the weave, orientation, resin, resin content, etc, you can make two panels that are exactly alike in every dimension but completely different in their stiffness characteristics.
A good example is any serious road racing bicycle. You see almost no metal frames at the professional level, furthermore, they could have used simple bonded tubes and lugs (early cf frames were) but now they are all molded monocoque design. For less weight you can target specific areas for stiffening.
I had mentioned that a fairly simple way to increase the stiffness of an f body would be to just overlay some CF on honeycomb along the entire floor pan. I would go with nomex honeycomb bonded directly to the steel with cf vacuum bagged over top. The thicker the honeycomb the stiffer the panel. That would make the car crazy stiff.
Here is an example of a small panel I made for a composites course I took recently. IIRC I laid up three plies of 3k 8 harness? (cant remember) 0-90-0 orientation on each side over aluminum hc core. Your supposed to put fiberglass on the aluminum as an insulator because cf is a conductor and will corrode the al.. anyway. This thing weighs almost nothing. Id say slightly heavier than a piece of white foam about the same size. I hand it to my buddies and i tell them to try and *bend* it "dude im gonna break it" just try i say. You cant get it to even begin feign a bend. I see the same guys turning all red trying to crank on the thing. Its amazing stuff. It actually isnt stronger than steel though when taken to complete failure but if you take steel that far you go way into plastic deformation anyway.

 

Yea im not exactly sure why they used carbon fiber for a safety cell you would have to ask them about that one. But as far as I can tell that where the transformation began. From there it may have been more efficient to build a complete carbon fiber monocoque seeing as how they had to use it for part of the car anyways. Again that would be a question for an engineer that works with F1 cars. However beyond that I dont know of any race car that utilize anything but space frames. I actually had the pleasure of taking a composites course as well. I was actually graced with an instructor who used to work for a ferrari dealership doing the repairs of the composite bodies and also things like making the custom molded seats which by factory spec could not weigh more than 3 lbs. Composites are an amazing materials and one of the most important things i learned was what you demonstrated in your pic. Strength is related to thickness however that dosnt mean it has to be comprised on 1 material all the way through. With a foam or honeycomb core you can significantly increase the thickness without increasing weight too much but dramatically increase strength. That sort of construction is commonly used on high $ boat hulls. Still though I have not seen this sort of construction in race cars again save formula 1 cars.

 

did a quick search and it looks like F1 uses a hybrid carbon zylon fabric for the safety capsule. I was unfamiliar with zylon but looked it up, looks like it has similar characteristics to kevlar. But anyway, I think there are quite a few more cars that incoproate CF than F1. I think some racing organizations probably prohibit its use to keep costs down. Didnt the Z06 use a CF over balsa core floor pan? heck I believe most of the upper level ferraris use a ton of CF.

Anyway we are getting off topic It would be neat though to do a cf honeycomb core on the stock floor pan but forget about doing any welding after that. The stiffness would be insane though. if someone wants to give me a roll of CF I'll do it, have the vacuum bag setup ready to go

 

Well there are many cars that use carbon fiber and other composite materials. There are even examples of cars that are made from composite monocoque and this isnt a new thing thats been a round a long time now ive heard of cars going back to the 50s being constructed using fiberglass monocoques. However that didn't necessarily mean they had a better strength to weight ratio than a space frame again just pros vs cons for a given application. For race cars still cant think of any examples of scratch built race cars whos frame is comprised of anything but tubular space frames again barring F1. In these race cars they commonly make use of fiberglass or carbon fiber in various places but not so much for frames. I guess a little more homework needs to be done in this area. There may be rules that prohibit the use of this style of chassy either directly or indirectly I have not heard of any but I cannot say for sure.

Anyways your right this has gotten way off topic (well kinda sorta its still chassy design but still) but if you were to make a comparable floor pan from carbon fiber/honey comb sandwich type construction I would think it would be possable to replace the floor pan all together assuming adequate means to attach it to the rest of the unibody were incorporated. Stock floor pans and attached frame structures are very heavy if they could be replaced would result is a huge drop in weight if nothing else. Course at that rate removing the entire unibody frame and bolting the body to a better designed chassy would be lighter and stronger still. Ive seen that done and its actually not as bad as one might think.

 

Back in the mid '90s when i was working on getting my
third rigid,i used a couple methods to gauge where the
flex was occuring:for the passenger compartment,
a laser was set on a tripod in the cargo well and aimed
at a paper target set atop the dash while a corner of the
car was jacked up and any deviaton of the laser spot
was noticed...
For the front structure ahead of the fire wall(i'll call it
the "doghouse")rods were placed diaganal with one
end secured and the other end hooked to a dial indicator
and the deflection measured...
(BTW that "doghouse"flexes mightily on a stock third)
-just drive one with the hood off over rough ground
and watch that radiator support
i never did try to test for "beaming"of my car because
none was noticed (though it is probably there...)
Once rigid, my third drives oh so nice
(there still is some slight flex noticed under certain
conditions,but far,far less than before-no more feeling
the the steering box moving through the steering
wheel,etc.)

 

Thats a pretty interesting technique. How much flex did you see in your tests? What did you do to stiffen it, and how much did it help?

 

Try to keep the concepts of stiffness and strength separate. They are more or less related but should not be confused.

Since it sounds like you want to throw this chassis at a space frame or finite element program, why not just build your own model from whatever information that you can get hold of, tinker with the element dimensions and metal thicknesses until you end up with something like 4000 ft-lb/deg. If the cross sections still look reasonable (i.e "buildable" and still fit where they actually go), go from there. For finding out the relative effect of adding the various stiffening elements you don't need to be all that precise with your baseline.

It's also quite likely that observations of improvement from adding various chassis stiffening is reflected in improved NVH performance rather than additional general torsional stiffness. Sometimes all it takes is to chase a few particularly disturbing vibration modeshapes off to frequencies where they are much less noticeable. It's easy to interpret a more solid-feeling structure as being generally stiffer overall when it may only be a little stiffer in a few local spots. A static analysis or even a static torsional twist test won't pick this up; for that, you have dynamic analysis and I guess shaker rigs.


Norm

 

I'm taking a surveying course this semester and we are going to have access to a 3D laser scanner for about 2 weeks. I can't promise anything, but if someone in the West Lafayette, IN area has a torn down 3rd gen for me to scan, I might be able to do it. Create a point cloud model of the car and put it into CAD or some other 3-D modeling software for FEA.

 

I said from the start that this post was a waste of time and is pointless. I see nothing came out of it, just speculation.

 

Its not a waste of time if it gets people thinking about the subject. Like I said, I'll have access to equipment this semester that would allow me to make a full 3D model of the chassis if my teacher allows me to do and I can find a 3rd gen chassis around these parts.

The scanning would be the hard part. Once that is done, the converting of the point cloud into a model would be easy as pie relatively speaking.

 

I seriously doubt that one weekend is enough time to go from a rather loosely defined collection of thoughts to meaningful results. Especially for people who are neither getting paid to do whatever might be involved nor have an immediate personal reason to get that deeply involved on their own time (and expense, if actual hardware testing ends up being the direction taken).


Norm

 

Now...


I think this is great. So far, there hasn't been any frame DATA posted...and that would be nice to see, if any exists. But still this thread has produced some great food for thought and analysis of the structure, and how to measure it, going forward. It is of interest to ME, and obviously others here. Therefore, not a waste of time.