[ericclaro]

Hello there,

I am trying to simulate the flow over the supercritical airfoil OAT15A. I have defined a 2D c-grid around it and want to use RANS equations to solve it. So far I have used the SA turbulence model since it yielded the best results. Other settings were taken from RANS simulations I did on su2 that yielded good results.

I am having some trouble converging the simulation. The lowest my Rho residual gets is around -4.5, indeed insufficient. It does not go lower than this and at some point it diverges. I am attaching the cfg file with the simulation configuration that gave the best convergence I got.

I attempted to simulate it using a fine and coarse grid, but both did not converge well enough. I am posting some pictures of the grid as well as of the domain (which in my opinion seems large enough). Boundary layer is discretised with first elements with 5e-6 m height (which in my experience yield decent y+ values).

Please let me know if you have any ideas to improve the convergence of the simulation!

[aero_amit]

Few suggestions -
1- Your CFL value is on higher side to start with, use lower CFL value initially and later use higher value.

2- You can also see if switching off multi grid helps.

3- Give a try to other upwind schemes.

4- Use latest version SU2-6.2.0

[ericclaro]

Hello, thanks a lot for your help, I will try to implement these changes.

I also thought about starting with an upwind scheme, butI am quite new to su2 and couldn't find a suitable option. I tried once to use SCALAR_UPWIND, but apparently this method is only defined for turbulence computations, the software did not allow me to use it at CONV_NUM_METHOD_FLOW.

Any recommendations?

[aero_amit]

Actually SCALAR UPWIND you have mentioned is Convective numerical method (CONV_NUM_METHOD_TURB) for turbulent flow.
I was talking about CONV_NUM_METHOD_FLOW.
Try ROE, HLLC, AUSMPLUSUP OR SLAU2 for this Mach number.

[___cony7___]

Could you please share the mesh file for this airfoil?