lio007 wrote: ↑
Tue Dec 31, 2019 9:57 am
Capharol wrote: ↑
Tue Dec 31, 2019 9:26 am
lio007 wrote: ↑
Mon Dec 30, 2019 9:02 pm
A limitation I never understood...why shouldn't teams be allowed to use more CFD for their development?
Or do the big teams have so much better models, representing real world behaviour more precisely?
I think the effort from the teams is now a lot bigger to deal with the limitation and trying to get out as much as possible.
first of all they limited CFD because of $$$$$$$
and 2nd yes the bigger teams have better tools then others so they get more and better data
But is CFD not better (in terms of $$$$$) compared to the wind tunnel?
BTW, can CFD be flawed as well as the wind tunnel? We often hear correlation issues between track and tunnel (e.g. in 2017 RBR had problems), but at the moment I can't remember reports of CFD-correlation issues. Or is it always the kind of "one without the other"?
Correlation issues means we buggered up and went the wrong direction. The wind tunnel and CFD are tools, lab tools, they can help, but they are approximations at best. The real world behaves in a particular way, we try to model that behavior, and our models are built with underlying assumptions about that behavior.
We don't need to have the correct assumptions to accurately model behavior at a certain scale. Our models simply need to be close or match experimental results. Here's the issue, our assumptions may model 90% of behavior accurately at a certain scale but my only be 40% at another scale. How do you reconcile your assumptions so that you can model behavior equally at both scales?
It's figuring out and getting all these different models to work together that's challenging. Mistakes are euphemistically called correlation issues.