[kyfrankie]

The case is ran with SimpleFOAM using kOmegaSST turbulence model at RE=50000, free stream velocity of 4.8m/s. The solution reached a steady "converged" stage but the residual of p fail to drops below 1e-3, others residuals are okay though. However, when viewing the U in paraView, there's two weird low velocity region at the leading edge. I am puzzled about what is going on. It would be great I can find some tips and advices here. Thanks in advance.

[tas38]

The "lines" appear to coincide to the transitions in surface mesh density. I suspect the low velocities are of numerical origin.

[kyfrankie]

Thanks a lot for your reply. I have made the refinement level to the same (6,6) and the "line" seems to be less aggressive but still appear. In addition, if the force coefficient converged to a stable value but the residuals fail to drop below 1e-4. Is it still considered converged? Any advice to improve convergence? Thanks in advance!

[tas38]

Originally Posted by kyfrankie

Thanks a lot for your reply. I have made the refinement level to the same (6,6) and the "line" seems to be less aggressive but still appear. In addition, if the force coefficient converged to a stable value but the residuals fail to drop below 1e-4. Is it still considered converged? Any advice to improve convergence? Thanks in advance!

Yes, I would consider the case converged considering the residuals and lift/drag coefficients you presented. If you desire to reduce the continuity residuals further, it may be possible to do so by adjusting the solver settings (e.g. discretization schemes) and/or mesh density/quality. You can think of the solver as "having converged to within the constraints imposed by the current mesh and discretization schemes". Note that by "mesh" I also include the proximity boundary surfaces to the wing/airfoil surface.

[mAlletto]

There may be some reciculation regions on the right edge of the airfoil which explains the high pressure residuals.

Regarding the low velocity close to the leading edge did you check if you can expect a laminar recirculation bubble.

Michael

[kyfrankie]

I believe the "line" may be caused by an issue in the STL file. Also, I found that by setting the refinement of the Wing to be uniform (6 6) instead of (6 7) helps the convergence after adding a absolute layer near the wing wall.

[mAlletto]

The stl seams to be not smooth. So probably you have to fix this first. Do you have wall layers? What kind of boundary conditions are you using