variante wrote: ↑26 Oct 2022, 12:50
Here's the car for COTA.
Lovely car and renders! You scare me a bit with your -5Cl story though...
spacehead3 wrote: ↑26 Oct 2022, 14:12
CAEdevice wrote: ↑26 Oct 2022, 13:07
Who is using 2500 or 5000 iterations? I could not use the fast template for race 3, because it is not suitable for "fine tuning", everything has been done with the "standard" MVRC template of MFlow.
I have started using the standard template now, fast template seems to be leading me in the wrong direction with the floor at this point. Unfortunately standard takes almost 24 hours to run and I don't like to tie up my PC for that long, so my progress as slowed quite a bit. My car really didn't like the low DF Monza setup, should at least be a bit more competitive this time.
Same here. I have run the Standard MRVC setup only once, with "only" 2000 iterations, and it took 24h. So I'll never run this again, but at least it has given me a useful correlation point with my setup.
Which is :
- By default, run the whole car with MRVC_Fast with 3 layers, no "high res" surfaces and 1000 iterations.
- For detailed Rear Wing development, run as above but with the rear wing in the high res folder.
- For Front Wing development, run only the front end (see picture below), with same settings as above except with 5 layers and the front wing in the high res folder.
I am only going after the flowfield and/or large steps of Cl (at least 0.1) so I feel this setup gives me good enough results.
But clearly, if you want to extract every last 0.01Cl bit of performance and optimise every little detail, particularly vorticity under the floor, then you have to run the Standard or Long MRVC setup with at least 2500 iterations.