Chassis Torsional Stiffness

[BruceA]

Now, what gets to me is why would you need to stiffen a chassis well beyond the maximum twisting force applied?? What I mean is there is a theoretical maximum amount of force that would 'appear' at a wheel, say a bump on the road. This force is transferred from a horizontal to vertical force being put into the shock absorber mount. Would it be fair to say perhaps that envisage a 1 ton force hitting the wheel as the car drives along. Lets say that 1 ton is presented to the shock absorber and mount as 1 ton. The chassis then resists this through the rear shock mounts. If the chassis deflects 1 degree it would have a torsional stiffness of 1d per 2200 lbs (or there abouts).
Now if a force hits any wheel of your car at a weight of 1 ton, do you think your wheel would still be attached to the car? We know the wheel will ride up over a bump so it would take a severe amount of 'bump' to project 1 ton into the shock mounts.
Does anyone have any numerical information on forces against wheels vs shock input? What values does anyone know of that are actually applied to shock mounts in extreme conditions??
My thoughts are why over develop a chassis with weight etc to achieve a stiffness that is really not required? Why develop for say 10,000 lb/degree when the wheel could be ripped off at 5?

 

[Scott Calabro]

RamboLambo,
I think because the distance from one suspention pivot to the other acts like a lever arm taking that 2000 lbs you mentioned and multiplying it several times according to the distance between points. Like using a breaker bar instead of a socket wrench. There is more to it than that, and one of our chassis guru's here will chime in for the complete answer.
I don't wan't to drift off the subject though.

Regards,
Scott

 

[Randy V]

Do you think that all chassis need to be so stiff that they are capable of absorbing none of the forces applied to it?

Let's think smart about this..

Not only racecars but Trucks, Street-Autos, aircraft and boats (et al) need to have a "certain" amount of compliance in their frames lest they be hammered to the point of failure.

Now your job just became much more of a challenge as you struggle to define the "Certain" above.

 

[Ron Earp]

Interesting discussion on stiffness:

View topic - Torsional Stiffness - F1Technical.net

F1 cars at ~20-30,000 N/m per degree from various sources.

Keep kicking around on the net and you will definitely convince yourself that the measurement method has a high impact on the resulting figure. Unless each chassis manufactuer takes a chassis to the same place and has it tested the same way I imagine the comparison figures will be fairly useless.

R

 

[Mike Pass]

As a bit of an aside whilst still in the general area - the addition of a welded in rollover cage will make a big difference as it "closes the box" as seen from the side. Martin Gough's SouthernGT chassis from Mick Sollis has a rollcage which is accepted by the MSA for hillclimbs and sprints and whilst only subjective at this stage seems to have added a great deal of stiffness to the structure. See pics on Builders forum - SouthernGT No 8.
One of the issues with GT40 chassis is the narrow f-r link across the top of the cabin caused by those doors cutting into the roof. I have seen pics of two crashed originals and both had buckled very severely at this point. it would seem to me that a bit of emphasis in this area would be very beneficial in both rigidy and safety. Having seen a car land upside down at over 100mph and from a height of 20 feet in a sprint at Aintree some years ago I know what I would like over my thick skull!
I also remember from the past an original monocoque being tested at about 25,00 lb.ft/deg.
Roll on - not over
Mike

 

[Scott Calabro]

Randy,
I think you will find if you do the research, that generally speaking in car design the stiffer the better. The chassis is not meant to take the place of the suspension.


Ron, none the less the way its measured, the stiffer the better.

But we are starting to drift...
Regards,
Scott

 

[Fran Hall RCR]

Cobra......read Shelbys thoughts...he wanted exactly that....Flexy flier to keep all the wheels on the ground with crappy suspension.

 

[Russ Noble]

Ultima - 3300 ftlb/Deg (coupe) and 2500 ftlb/Deg (spyder)mass 135 kg,
Lotus 23 - 1500 ftlbs/Deg, 45 kg
Lotus 7 replica, "locost book" spec - 1200 ftlbs/Deg, 82 kg
Lotus 7 replica, uprated "cymtriks" spec - 2540 ftlbs/Deg, 78 kg
Lotus Elise - 7350 ftlbs/Deg,
Lamborghini Countach - 1900 fp/degree.
Ferrari 360 spider - 6250 fp/degree.
Lotus Elan: 5,000 - lb-ft/deg
Porsche 959 - 9,500 lbs-ft/deg
Lotus Esprit SE Turbo - 4,300 lbs-ft/deg
GTD Lola T70 replica - 3300 ft/lbs per degree.


Fred W B

So..... You guys that think extreme torsional rigidity is such an important issue..... What is/was so terrible about a Lotus 23?

All car design is a compromise. Do you want to trade off a nice light nimble good handling car for an overweight pig that goes slower and doesn't do anything as well just so you can have higher torsional rigidity?

Just a thought.....

Cheers

 

[Scott Calabro]

I don't care what the numbers are I just want to know why its such a F@#$ING secret !
I bet you can find bump steer info and chamber curves all day long...
Surely no offense Russ !!!

Regards,
Scott

 

[Jac Mac]

I don't care what the numbers are I just want to know why its such a F@#$ING secret !
and chamber curves all day long...
Surely no offense Russ !!!

Regards,
Scott

Wots a ''Chamber curve" ,a secret formula for calculating the 'S' bend on a toilet? No apologies for the drift guys!!

Jac Mac

 

[Joe T]

Yeah Scott its a secret, Sorry mate...........rockonsmile

Build a chassis and start swinging on it, all will become clear.

 

[Ron Earp]

Scott,

I think if you use the search engine you'll find figures for CAV, ERA, Roaring Forties, and GT40 Australia and maybe one or two GTDs and KVAs - all on this website. I know the first three are here. Once you've got them I'm not sure how useful they'll be since they were not all done with the same method thus they are completely incompariable.

I don't think the numbers are a secret. I think most folks realize the numbers are useless.

What is the discerning customer who is not a chassis engineer going to do with the following data:

ABC = 15,000 N/m per degree using Method 1

XYZ = 21,000 N/m per degree using Method 2

DEF = 31,000 N/m per degree using Method 3

A Formula 1 car is at 25,000 N/m per degree using Method 4

NET = Not Equal To

Method 1 NET Method 2 NET Method 3 NET Method 4

Questions:

1) Which car has the stiffest chassis, ABC, XYZ, DEF, or the Forumla 1 car?

2) How do you know which one in question 1) above is the absolute stiffest chassis? Explain.

I think this is one of the reasons why you rarely see torsional rigidity numbers touted by car manufacters (Ford, GM, Audi, etc.). But you DO see them write press releases of the following form "The new E98 chassis is 35% stiffer than last years' chassis" all the time. You see lots of % changes based on the manufactuer's testing method which they use repeatedly. But I don't think you'll find chassis comparisons across brands too often, and if you do, then what to make of them?

Maybe the SAE has a well described testing method. I don't know.

I wonder what my 260Z race car is? I wonder if I knew the number if it'd make me like it less? Nah, I'll just shut up and race that thing!

Nothing gets the forum going like a good transaxle discussion, or a torsional rigidity discussion!

Ron

 

[Scott]

Is a stiffer chassis better for drifting?

 

[Scott Calabro]

Thanks Ron,

I bet if I knew what each different testing method was and how it was used, I could give you a definative answer ie. question 1.

For the record I dont think I'm asking anything too far off the wall, just trying to gather as much info as possible.

JacMac.. did I spell chamber wrong ? Maybe camber then.....

Joe T.... Thats too bad, its a shame, its probably because we would all be shocked at the low numbers of some cars.

Thanks anyway guy's

Regards,
Scott

 

[Fran Hall RCR]

Scott,
its no secret...we just dont want to tell you...:flameon:

 

[Scott Calabro]

Touche' my friend!

Very Best Regards,

Scott

 

[Ian K]

I wonder what my 260Z race car is? I wonder if I knew the number if it'd make me like it less? Nah, I'll just shut up and race that thing!

Nothing gets the forum going like a good transaxle discussion, or a torsional rigidity discussion!

Ron

With roll bar - probably OK. Without, probably not very stiff at all. At least
that's what I "felt" was true with my 240

Ian

 

[Jim S.]

I'm with Ron and those others that have said it. If all of the tests aren't performed the same they're basically of no use. The only true reason to know this info is for comparison of chassis design. Your trying to figureout the pros and cons list of all the mfg's your considering. Problem is this could only be part of the list if they were all performed the same by the same testing facility or if the testing facilities are all accurately calibrated back to NIST standards.

One of "possibly" the major reasons why some of the Kit builders do not post the numbers is because of this reason. They end up being judge by non-common standards.

:dead:

 

[Scott Calabro]

Good point Jim, and well taken.

Thats enough for me for one day.

Time to go get a cold one !

Goodnight all,
S

 

[Russ Noble]

I don't care what the numbers are I just want to know why its such a F@#$ING secret !
I bet you can find bump steer info and chamber curves all day long...
Surely no offense Russ !!!

Regards,
Scott

I don't know that it's a secret. But unless the methodology is the same, it's meaningless.

Chamber curves and bump steer are easy to measure, but I bet not many GT40 manufacturers publish camber curves, not because they are a secret, just because people are more interested in whether to run Gulf livery or Essex Wire!

Cheers