Benetton Formula 1 - The Race Is On With STAR-CD!

During the last two years, leading Formula One teams have been racing to get aerodynamic design improvement through whole-car CFD (Computational Fluid Dynamics) simulation using CFD codes such as STAR-CD.

Benetton Formula 1 are using CFD techniques to gain more insight into the flowfield than can be achieved by wind tunnel measurements. In collaboration with CD (Computational Dynamics Ltd, a member of the CD adapco Group) Benetton Formula 1's design engineers are using STAR-CD to "fine tune" the aerodynamics of their race cars.

The method starts by importing geometry of a CAD model of the car and then building a CFD mesh automatically using STAR-CD's engineering-based expert software tool "EZ-Aero". The mesh exploits symmetry for half-car modelling with about 10 million cells. Thanks to scalable parallel computing, these large scale simulations can typically be completed overnight. By building up fine mesh layers from the CAD surface, an incredibly high flow resolution can be achieved. For example the flow for the leading edge of a Benetton front wing can be accurately resolved to within 1 millimetre!

This approach to high resolution CFD simulation provides results that capture the car's aerodynamic flowfield in extreme detail, including the effects of wheel motion and the thermal air-density affects of the cooling air leaving the engine bay.

The race is truly on at Benetton Formula 1 to use STAR-CD to push aerodynamic design of their F1 cars to new heights.

For more information about STAR-CD and its abilities in the automotive sector please contact us.

Apparently Ferrari is using Fluent for their analysis. The race is on!!

(1). Let us know how the code is being applied to the study of Formula-1. (2). I mean, were they able to find anything new in the flow field?

Hi!

I thought Benetton were using Fluent. There was a big campaign saying that a few years ago.

Does anyone know why they changed?

It is possible that they use both codes.

They changed because they realized that "being easy to use" wasn't real helpful when you get the wrong answers. They also noted that one of their competitors who tore up the track the last couple of years was a STAR user.

(1). Formula-1 is sure a good proving ground for CFD codes. (2). Since a CFD code does not produce the 3-D solution by itself, I must say that the key factor will be the experience of the code developer and the Formula-1 designer. (3). Still, there are effect coming from the wake of other cars all the time.

Benetton did use Fluent a couple o years ago. At this Time they became World Champions. Last Year Renault bought the Benetton Racing Team, so I think as Renault also uses Star-CD, Bennetton has to use it.

Does everyone really think the CFD makes that much difference ? One team changing code because the oposition streaked passed, their car developed using Star.

I hope i understand that correctly and that it was a joke ??

Lets be realistic, how much of the Aerodynmics is developed by the CFD guys and how much by the wind tunnel guys !! 20% CFD 80% WT ??? Dave

(1). Good point. (2). I would say that if the wind tunnel test data is used in the turbulence modeling of the CFD code, then there is a chance that the CFD solution will match the data. (3). Otherwise, given two teams using the same code, it is likely that they will get different CFD solutions. (4). I don't think it is a joke. It is just a promotion to use CFD codes. In the end, the CFD approach will fail. But for now, there is still a chance especially for those who have the money and time. (5). Actually, related to the US government report on the DNS and LES in a recent article, it says:"The continuing development of this technological effort offers the possibility of vastly improved performance prediction for practical design and analysis applications." Direct Numerical and Large-Eddy Simulation of Aerodynamic flows (VA-00-04),Sept.2001, :http://www.afrlhorizons.com/

Cheers for the reference John, i guess we'll all have to wait and see what the future brings - only a matter of time !! oh and resources. Dave

Folks, As a long time adapco employee and STAR-CD user I have my prejudices. But will admit that there is some marketing going on. That said I have a technical point you should consider.

Formula 1 cars (and most race cars) have very low ground clearance. Now you can suck and you can blow and you can spin a treadmill but you can never really get completely rid of the boundary layer on the floor of the wind tunnel. You also have to stick these really large (relative to a F1 car anyway) support structures into the flow. So that front wing that sits less than an inch above the wind tunnel floor is sitting in air that is probably behaving as a more viscous fluid than what it sees on the track.

The CFD solution within its own limitations of turbulence modeling and boundary layer seperation eliminates that particular "difference" from the "real world". Additionally the support structure can be removed, the tires and brake rotors can spin. The radiators can reject heat. The intake can suck in air and the exhaust can blow smoke. If one has the CPU you can place several cars in a row. You can have cars pass each other, though I doubt anyone has actually tried that. All very difficult in the wind tunnel.

Both the wind tunnel and the CFD analysis are simulations. Each can tell you things about how the real car will behave in the real world. Obviously the trick to it all is knowing how to weight the two methods, and how to extrapolate them to that real world. Both simulations cost money and F1 guys may have big budgets, but they are not unlimited. They wouldn't do it if they didn't think that it would help their designers.

Anyway I think that CFD whatever tool you use can give a lot of insight into flow past, around, and through automobiles. And the automotive industry seems to agree.

Sorry for the long post, but external aero has always been one of the CFD topics that most interests me.

-Andy Robertson

(1). I like your positive thinking. And I think, there is a bright future of CFD in F1 racing car design. (2). To get there, we will have to identify each limitation first. So that improvements can be done on the long term basis.

i also agree with you whole heartedly and I like to think i too am very posative about CFD. it was just the original hype that drew my attention. Don't get me wrong I'm not a skeptic. What was interesting from the last but one response was the wind tunnel problem of modelling the boundary layer (or removing it). How much effect does this have upon the overall performance of the car ? I have no actual experience of vehicle aerodynamics, but would be interested to find out. I guess the sucktion improves cornering, where as the drag is more imortant on straight sections and full on speed ???