[maximus23]

Hello,

I am new to CFD and I am starting off with some 2D test cases. The simulation is of a 1 meter NACA 2412 airfoil at 0 degress angle of attack with a velocity of 30 m/s in a viscous flow. I have set the flow to incompressible and ran four simulations with Roe and JST solvers and SA and SST turbulence models. The results where fairly consistent but all had lift lower than experimental data and drag higher than experimental data.

I refined the mesh by increasing the number of elements along the airfoil curve as well as reducing the first boundary layer cell height and increasing the total inflation layer height. I repeated the four runs above but JST diverged almost instantly and only one of the turbulence models converged when using Roe, which yielded almost identical results to the coarser mesh.

I created another mesh, this one using a blocking strategy, and reran the simulations again. Now, for all of the solvers it says that SU2 has diverged before a single iteration.

I have modified the CFL number with no results and have the configuration file set up almost identical to the airfoil test cases provided with SU2. I have attached my configuration file. Any feedback would be highly appreciated.

Thank you.

[fpalacios]

Hi Max,

Looking at your config file, I would change the following:
CFL_NUMBER= 4.0 -> CFL_NUMBER= 1.0 (at least for ~100 iterations, then you can increase the number or use the adaptive CFL number)
SPATIAL_ORDER_TURB= 2ND_ORDER -> SPATIAL_ORDER_TURB= 1ST_ORDER (more robust and the solutions are acceptable - if you are using a fine mesh)
KIND_TURB_MODEL=SST -> KIND_TURB_MODEL= SA (I think that SA is the only that has been validated with the incompressible mode).

Anyway, SU2 has been developed and validated in a community focus on compressible applications. For that reason, my main recommendation is to run the software with
REGIME_TYPE= COMPRESSIBLE and you need to follow a good tutorial to set the problem as
https://github.com/su2code/SU2/wiki/Turbulent-ONERA-M6


Thanks for using SU2,
Best Regards,
Francisco

[maximus23]

Francisco,

Thank you for your reply, I will try your recommendations this week.

Regards,
Max