[sbro]

Hi all,

I'm currently simulating flow around an airfoil (different test cases with various trailing edge modifications) using ANSYS Fluent and a Coupled scheme (2nd Order) with Higher Term Relaxation and Pseudo Transient enabled.

My solution (for all test cases) converges for pre-stall Angles of Attack below 13 degrees (all residuals reduce to ~<2E-04 and steady force coefficients converge) but DO NOT converge for post-stall angles (14-18 degrees).

I thought this would likely be due to transient flow features (shedding) not properly captured by the steady solver but, curiously, the solution at 20 degrees (highest angle tested) DOES converge which I don't understand(?). I would've thought if the solution converged at 20 degrees and for the lower angle sweep then the intermediate values should also...

I've tried almost every trick to try and resolve; a higher resolution mesh, switching to 1st order schemes, adjusting under-relaxation factors and I can't seem to get anything to work. Switching the far-side boundary conditions (wing-tip side; see below) to either a slip / no-slip wall achieves convergence but a poor result for lift which isn't feasible next to experimental data (even though the outer domain walls are situated far from the 3D aerofoil model). My mesh statistics could also be better but if this was the issue then I wouldn't expect convergence for any of the angles tested(?).

Any ideas for what I can try would be greatly appreciated and if you need any further information please don't hesitate to let me know.

I've posted my boundary conditions and a vague idea of my mesh sizing below.

- Blue (Front and WingTip-Side Face: Velocity Inlet)
- Red (Rear Face: Pressure Outlet)
- Yellow (WingRoot-Side Face: Symmetry Plane)

BCs.jpg

Mesh1.jpg

Mesh2.jpg