On Joe Saward's blog, USF1's Ken Anderson says:

"“The major cost savings will be that the engineering and production of the cars will be done in the United States. Our technical partners located within a 30-mile radius of our shop contribute to this savings, as there are some departments we don’t have to have in-house, such as a wind tunnel, shaker rig, K&C machine, additional CFD support and a center of gravity machine. What most people see – the transporters, motorcoaches and the “lifestyle” side of Formula One – are a much smaller part of the overall budget and will be located at our European facility, which we’ll tell you about soon!”"

(http://joesaward.wordpress.com/2009/10/07/usf1-interview-with-ken-anderson-and-pictures/)


I'm not familiar with some of these devices... what's a K&C machine? What's a center of gravity machine?


Meanwhile, in an Autosport article today, Nick Wirth says the Manor car "won't do even one day in the wind tunnel, not even for a check." That sounds like a weird mix of being really frugal, cutting edge, and stupid... all at once. Hmm.

K&C=kinematics and compliance. That measures the way that the wheel moves in response to forces, and gives you bump steer curves and RCH and so on. Spiffy ones use hexapod tables to drive each wheel. "Mine" does.

A CG machine measures where the CG of the car is. I'd love to have one locally. You can also get the polar moment of inertia measured, on the one I use.

I've always wanted to see a pure CFD car in F1!
However, we'll never know if the design is truly Manor's or a slight modification of a current F1 car provided to them by some other sources.

I guess we'll see in winter testing! Wirth's no-wind-tunnel-i-trust-my-cfd-thank-you-very-much acura prototypes seem nice....but they haven't really had much competition in the ALMS this year..

Is there any evidence to suggest that a pure-CFD car would look any different to an existing car?

And how would you differentiate it from the general innovation of Mr Wirth and Co?

Didn't Stewart Grand Prix build a CFD car on it's debut?

If you skip the tunnel validation couldn't one theoretically take a more or less generic F1 design, plug it into some sort of CFD/iterative GA tell it to run and come back in 6 months with a design?

I was lectured by SGI around 1995 on how they helped developing New Zealand's Americas Cup champion at the Naval Architecture school here in Sao Paulo University.

The 12 meter rule for sail boats was very interesting on how loose it was and how completely different designs would fit into it. That's pretty the opposite of F1 regulations today.

So they designed 10 different approaches that would fulfill the rule, did a simulation run and picked the best. Then create another 10 variations around the wining one and re run, until they found what would be the optimal design. It worked back then.