There are 3 problems with CFD.
1. Your input model has to be perfect right down to body imperfections in
joints etc.
2. It is still a simulation and as such subject to the rules of the software
correct or not. I would only believe the results if they matched a known set
of conditions on a fully instrumented car. Then I would make a change to the
model that is seemingly for the better and CAREFULLY test again.
3. It takes a tremendous amount of time and money.
There is no substitute for fully instrumenting the car.
Channels
Ride height all 4 corners
Load cell all 4 again
Accelerometer 3 axis these have to be located on the CG in all 3 axis
pitch and yaw sensors
air pressure sensors above and below the car in various places
with this info then go to the CFD software and see if the model and software
agree with what you are seeing.
some basic links
http://www.rhino3d.com/
http://www.fluent.com/solutions/automotive/aero.htm
dave
> -----Original Message-----
> From: owner-land-speed@autox.team.net
> [mailto:owner-land-speed@autox.team.net]On Behalf Of Flowbench@aol.com
> Sent: Wednesday, December 17, 2003 1:15 AM
> To: land-speed@autox.team.net
> Subject: CFD aka Computational Fluid Dynamics
>
>
> Time to ask the list for some input.
> I have recently been involved with a race team that had an
> accident at the
> salt. This car and another I'm involved with were/are running
> very close to what
> we believe to be there aero limits. There are several things that
> could be
> done to help, but what seems to come up the most is CFD. Does
> anyone have actual
> experience using this software? Does anyone know of a program
> where we can
> pay for this evaluation? This software goes for over $35,000 and is still
> undergoing testing in most cases that I'm aware of. Gary Eakers
> wind tunnel is also
> up for discussion but in a limited format because of air speed
> and the limited
> number of changes that can be made to the car vs. computer modeling.
> Thanks for your thoughts
> Mike
|