LVDH wrote:realisticly I think that Variante, JJR and TF will push CAEdevice back to P4 next year.
I can not wait, I really can not wait.
I thought for a second to help the staff instead of competing, but now I definitely changed my mind ...
As good as he was, he made his points early in the season by figuring out how crucial cooling airflow was. Now more people got a grip of it.
During the last two races I had issues with the convergence, as you know, and the change of the number of iterations (that I discovered after the 4th race) made me develop the car in the wrong direction for about a month.
I developed a safer car fot the last race (it only lost about 10% efficiency between iteration 1500th and 2500th) considering that I had a good margin of points.
I think that next year we need to have the same number of iterations for every race and geometry (maybe 1500-2000, with a better mesh and some rules to have realistic the diffusers angles (in order to reduce pressure gradients) and lower ride heigth / diffuser heigth ratio (it is not only a matter of floor hieght, on the contrary, higher floor will bring even more extreme diffuser angles).
We should consider that convergence in a virtual challenge should be managed in a different way from an industrial simulation, because our CFD model is not a way to represent reality, but "it is" our reality (yes, Arthur Schopenhauer is my new chief designer).
And I am sure that next year managing cooling air flow will be easier than this year as I will spend some time understanding what made it so difficult.
About the cooling flow simulation: it couldn't be easier as it is now. We just need the possibility to rotate the heat exchangers (or to use two smaller heat exchangers per side).
We should also consider the influence of the position of the heat exchangers on the CoG or we will see something very strange(r) ahead of the front wheels axle on my car.