[javier.delasierra]

Good morning,

I’m having problems for getting a proper solution from a typical flow around airfoil simulation. To be more precise the profile is ‘whitcomb-il’ with 3m chord at a wind speed of 80m/s. The problem is the following, I’m trying to calculate the lift coefficient for each angle of attack in order to capture the stall angle of attack however it seems impossible to capture it. Geometry is exactly as show in figure (29deg angle of attack which is pretty high, theoretically stall should be found between 16-20deg)
Whitcomb_29deg.jpg
The boundary conditions are:

Inlet: fixed speed, zerogradient pressure, fixed values of k and omega
Outlet: fixed pressure, zerogradient speed, zerogradient k and omega
Lateral walls: empty type
Top and bottom walls: symmetry type
Profile: wall type with no slip

The simulation was run as steady state, incompressible flow using k-w SST turbulence model. The software used for the simulation is OpenFOAM.

The problem is that even having a 29deg angle of attack the simulation seems not to capture the boundary layer separation giving an unrealistic ‘stuck to surface’ flow around the profile, consequently calculating an overestimated lift coefficient. I would like to know any advice or tip or possible solution for this problem. Any ideas would help.


Thanks in advance for your time and attention!

[fra76]

Hi Javier,

You should be able to capture the stall, maybe a bit too late or too early, but you should see something.
How is your mesh, especially near wall? What is your yPlus value on the surface at the end of the simulation? Which wall function are you using?

Cheers,
Francesco

[javier.delasierra]

Good morning,

The y+ value is around 90 using a 'nutUSpaldingWallFunction'. The mesh has a boundary layer shape around the profile (10 layers with a 1.25 coefficient between layers). The rest of the mesh is made of blocks generated from SnappyHexMesh with 4 refinement levels around the profile surface.




Thanks for your help!