The 30/30 Rule

[MadMatt]

Another advantage of CFD is that if you cache known flow patterns, you can test tons of parts with only a model of that part, in isolation. You can for example record how the air flows off a front wing/splitter/sidepod assembly, and then test 1000 different diffusers in that flow pattern without ever simulating the front of the car again

So in your opinion if I was gonna use a vertical blank plate as a diffuser it would not affect the front of the car?

[beelsebob]

Another advantage of CFD is that if you cache known flow patterns, you can test tons of parts with only a model of that part, in isolation. You can for example record how the air flows off a front wing/splitter/sidepod assembly, and then test 1000 different diffusers in that flow pattern without ever simulating the front of the car again

So in your opinion if I was gonna use a vertical blank plate as a diffuser it would not affect the front of the car?

No, my opinion is that if you want to test lots of subtle variations in the exact same environment, you need to do a lot less work to figure out which ones to test more thoroughly.

[MadMatt]

Another advantage of CFD is that if you cache known flow patterns, you can test tons of parts with only a model of that part, in isolation. You can for example record how the air flows off a front wing/splitter/sidepod assembly, and then test 1000 different diffusers in that flow pattern without ever simulating the front of the car again

So in your opinion if I was gonna use a vertical blank plate as a diffuser it would not affect the front of the car?

No, my opinion is that if you want to test lots of subtle variations in the exact same environment, you need to do a lot less work to figure out which ones to test more thoroughly.

And that is incorrect. If you work in isolation you will for sure find parts that will send you in the ball park (vertical plate VS horizontal plate, the horizontal one will give you better results), but at the level of F1, where you need to find thousands of a second, working in isolation will never give you accurate results.

[marcush.]

as i understand Sauber has relieved itself from Albert some time ago as it was too slow ..there is no hint of a mention of a supercomputer on saubers Web site to be found..One would think they would emphasize they have on site capacity ..
in 2012 MrSauber mentioned they are outsourcing calculations these days:
source :
http://www.motorsport-total.com/f1/spli ... 00202.html



and heres saubers Explanation on the 30/30 rule:
http://www.sauberf1team.com/fileadmin/u ... ons_EN.pdf

[johnsonwax]

but at the level of F1, where you need to find thousands of a second, working in isolation will never give you accurate results.

What's your point? That the teams would get better results without a windtunnel cap? Well no kidding. But the reality is they have a cap and need to work within it, and working in isolation is the best way to test iterations on an idea with limited resources.

[tim|away]

and heres saubers Explanation on the 30/30 rule:
http://www.sauberf1team.com/fileadmin/u ... ons_EN.pdf

Sauber appears to have missed the memo. They are listing the old limitations agreed under the RRA when in fact the limitations have been changed for 2014. See FIA Sporting Regulations 3.6 for more information.

[MadMatt]

but at the level of F1, where you need to find thousands of a second, working in isolation will never give you accurate results.

What's your point? That the teams would get better results without a windtunnel cap? Well no kidding. But the reality is they have a cap and need to work within it, and working in isolation is the best way to test iterations on an idea with limited resources.

You must be working at Marussia then. (no offense to Marussia engineers)

[beelsebob]

And that is incorrect. If you work in isolation you will for sure find parts that will send you in the ball park (vertical plate VS horizontal plate, the horizontal one will give you better results), but at the level of F1, where you need to find thousands of a second, working in isolation will never give you accurate results.

The point is that you are not working in isolation. You are simply caching results (and then working in semi-isolation, and ignoring only the effect that the part will have on other parts in front of it). The result will indeed not be perfect, but making the calculation cost 1/10th the amount means that you can get big gains by testing things quickly this way before testing them on the whole model.

[MadMatt]

And that is incorrect. If you work in isolation you will for sure find parts that will send you in the ball park (vertical plate VS horizontal plate, the horizontal one will give you better results), but at the level of F1, where you need to find thousands of a second, working in isolation will never give you accurate results.

The point is that you are not working in isolation. You are simply caching results (and then working in semi-isolation, and ignoring only the effect that the part will have on other parts in front of it). The result will indeed not be perfect, but making the calculation cost 1/10th the amount means that you can get big gains by testing things quickly this way before testing them on the whole model.

I see your point, but again, this cannot work in F1 where we are looking for thousands. They are not at the stage where they need to narrow down solutions, they already know where to work, they can only do minor gains. Such approach would work for categories with less advanced aerodynamics or for amateur racing. Again just the opinion of an engineer with bit of aero experience.