required torsional rigidity given roll stiffness value

[Ashley Denmead]

Hi everyone,

i seem to remember reading a while ago about a rule of thumb for determining the required TR in a chassis if you have the front and rear roll stiffnesses for your suspension setup....i cant seem to find it anymore! any help would be appreciated....i just wanted to figure out ideally what would theoretically be enough rigidity.

cheers

[Ashley Denmead]

Hi everyone,

i seem to remember reading a while ago about a rule of thumb for determining the required TR in a chassis if you have the front and rear roll stiffnesses for your suspension setup....i cant seem to find it anymore! any help would be appreciated....i just wanted to figure out ideally what would theoretically be enough rigidity.

cheers

[Ben Beacock]

I beleive it is roughly 1 order magnitude higher (ie 10x) than the difference between front and rear roll stiffness.

[alfordda]

I have heard anywhere from 5 to 10 times.

William Riley and Albert George from Cornell wrote an SAE paper (2002-01-3300) that might be useful. You should probably be able to get it from your library for free, but it only costs around 5-10 bucks from SAE.

[DJHache]

This is one of those topics where you hear a lot of stuff but noone actually seems to have any real information. I've heard anywhere from three to ten times the roll stiffness. The guys from Cornell gave a value at about 8 to 10 times the roll stiffness

[Denny Trimble]

The ferarri F1 2000 book has numbers on roll stiffness and chassis stiffness. Very interesting...

[Ashley Denmead]

thanks guys,

that gives me a bit of an insight! however is it a factor of the roll-stiffness or is it a factor of the difference between the front and rear roll stiffness???

[winnie]

Ideally you want the most rigid chassis possible, so all deflections/suspension movements are controlled. Nothing is perfectly rigid, and structures tend to get lighter and lighter the less stiff we make them. It is a compromise and there is no hard and fast formula i am aware of, only rule of thumb, which i have heard/seen to be in the 5-10 times region, which is a pretty big window.
I would be inclined to set a weight target for the chassis and try to make it as rigid as possible.

[Frank]

>3 * total (f+r) rolling stiffness is a good idea.

>5 is my recomdation, this is based on a twist test

http://www.uq.edu.au/fsae/statics.xls

if you (simulink) model including a chassis torsional spring, you'll get a plot something like this

http://www.uq.edu.au/fsae/graph.bmp