Bar Stiffness in lb/ins per degree, Q.

Q = [10^4*T^2*K^2*(D^4-d^4)]/[R^2*(L/2)]

Where,
T = Track (ins)
K = Lever arm ratio (Movement at ARB pickup/Movement at Wheel)
D = Outside diameter of bar
d = inside diameter of bar
R = Effective Arm Length
L = Length of Bar

Taken from Allan Staniforth's Race & Rally Car Source Book

Quote from tristancliffe :

Bar Stiffness in lb/ins per degree, Q.

Q = [10^4*T^2*K^2*(D^4-d^4)]/[R^2*(L/2)]

Where,
T = Track (ins)
K = Lever arm ratio (Movement at ARB pickup/Movement at Wheel)
D = Outside diameter of bar
d = inside diameter of bar
R = Effective Arm Length
L = Length of Bar

Taken from Allan Staniforth's Race & Rally Car Source Book

That equation looks about right, but it makes no mention of the Young's Modulus of the material used to make the ARB. The (D^4-d^4) term is the second moment of area of a hollow circular bar (minus some coefficients which have probably been absorbed somewhere). To get flexural stiffness you must mupltiply this by the elasticity (Young's Modulus) of the material. Since it's not explicit in the formula the author must have assumed a value.

If you're using the same material then you don't have to worry, but there might be a small correction to be made if you're using a different metal.

Yes, I realise that. There are various constants (a 4, a pi, etc) that are amalgamated into the equation, and then rounded to give an easy 10^4 at the beginning.

I think it's reasonably safe to assume that the material differences will be pretty negligable compared to the rounding errors above and the measurement approximations used. Oh, and the motion ratio is variable, because the ARBs are driven on the pushrob bellcranks, so I'll have to approximate that to a constant (at least to start with - I'd love to make a numerical model of the suspension at some point, so that ride height changes in motion, or setup changes in preload/spring rate can all be taken into account).

Just been sent this - it's taken a lot of chasing, and they could only supply actual data for radials.

Average seems to be about 1200lb/in or 21kg/mm or 210N/mm

No data about crossplies as yet, but I'm going to try sitting on them whilst my Dad measures the change in distance to the rim from the floor. Only a static test, of dubious scientific value, but it's a real numerical answer, and shouldn't be a MILLION miles away from sensible.