[tomp1993]

Hi everyone,
I'm just wondering is there a generally accepted 'best practice' way to set up a RANS case in SU2 specifically for capturing separation regions?
I have tried a couple of different grid topologies, Roe and JST schemes, SA and SST turbulence models, and a few other options in the config file. my case goes from M=0.15 to M=0.5 so incompressible isn't suitable. I'm not expecting amazing stall angle prediction but i would like to see roughly where the separation and attachment lines are.



unfortunately, so far either my solution doesn't converge (no force convergence) or converges on a solution which does not feature separation (when i definitely expect separation in reality).


on this note, are there any best practice methods for gridding this scenario too? i'm currently using a pointwise-generated hybrid grid with t-rex BL cells which have y+<2 and growth rate 1.2. with surface t-rex on the leading and trailing edges of the wing.


my geometry is a wing-body-tail half-span.


Unfortunately i cannot upload the case but this should be fine as i am asking for more general best practice advise.

Thanks in advance!
Tom