[poloding-along]

Good evening.

Looking for some assistance in solving an issue I am having when trying to add layers to a mesh around a naca0012 aerofoil. I am currently trying to model the naca0012 foil as per the nasa validation case but I'm finding my Cd values to be way higher than expected and I believe this may be due to a poor mesh.

I am modelling a naca0012 aerofoil initially using the standard blockmesh then the surfaceFeatureExtract tool then snappyhexmesh. The aerofoil is 3D but I am only taking a thin slice of it in the blockmesh and have defined only one division in the z direction.

When in snappyhexmesh, I am trying to add boundary layers around the surface of the aerofoil but I am having particular difficulty in gaining smooth layers particularly around the trailing edge. I appreciate as it is a sharp edge then the snappyhexmesh tool can struggle but so far I have tried adjusting;

the finalLayer size
number of layers
featureAngle
adding an extra refinement region around the trailing edge
refining the initial blockmesh
reducing the sharpness of the edge

With adjusting all these I can occasionally get smooth layers but these tend to be skewed as I then rotate the aerofoil through varying angles of attack.

I have attached two pictures, the first shows the leading edge with consistent layers like I am trying to achieve then the 2nd shows the typical problem on the trailing edge where the layers don't fully generate and create skewed cells.

If anyone has any suggestions for how to improve these then please point me in the right direction.

[Ruli]

Dear Craig,

I am modelling a naca0012 aerofoil initially using the standard blockmesh then the surfaceFeatureExtract tool then snappyhexmesh. The aerofoil is 3D but I am only taking a thin slice of it in the blockmesh and have defined only one division in the z direction.

Why do you use snappy? If it is an infinite span NACA profile, you could create the mesh with blockMesh only.

When in snappyhexmesh, I am trying to add boundary layers around the surface of the aerofoil but I am having particular difficulty in gaining smooth layers particularly around the trailing edge. I appreciate as it is a sharp edge then the snappyhexmesh tool can struggle but so far I have tried adjusting;

the finalLayer size
number of layers
featureAngle
adding an extra refinement region around the trailing edge
refining the initial blockmesh
reducing the sharpness of the edge

What in my experience helped, was indeed refinement of the base mesh. However, I did not get my leading edge meshed perfectly.

Best regards
Julian

[poloding-along]

Julia,

Thanks for the response. Surely I need to use snappy so as to gain a more refined mesh in and around just the aerofoil as opposed to refining the whole domain? Bearing in mind that I have a rather large block mesh so as to deal with farfield boundaries.

When you refined your block mesh, what level of y+ did you aim to work with?

Thanks
Craig

[Ruli]

Dear Craig

Originally Posted by poloding-along

Surely I need to use snappy so as to gain a more refined mesh in and around just the aerofoil as opposed to refining the whole domain?

You can save cells by using the grading option in blockMesh. Check it out. If you are working on a pure 2D or 2D infinite span airfoil, I would recommend using only blockMesh, because it should give much better cell quality.

Originally Posted by poloding-along

When you refined your block mesh, what level of y+ did you aim to work with?

As you probably know, y+ depends on the chosen turbulence model. I tried to get y+ = 30, in the end i have 1 <= y+ <= 500.

Best regards
Julian

[linnemann]

See here for very easy airfoil mesh generation using blockMesh.

http://hvirvel.dk/airfoilmesher/

[poloding-along]

Thank you for the responses. I will try what you have both suggested.