[natala1987]

Hi,
I've got big problem with NACA0012 meshing. I do not have experience with CFD and I really need your help I've done 2D mesh and it works in Fluent. Now I should create mesh in 3D with the same parameters (something like extrude for 0.2). I attache my 2D ICEM CFD file.
Thank you for your help

[diamondx]

here is a link of a video tutorial on how to mesh a 3D wing made by Mr Far:
http://www.youtube.com/watch?v=mgjRT4WY_iA
hope it can help

[natala1987]

http://sendfile.pl/171206/NACA0012.zip
http://sendfile.pl/171207/NACA0012_2.zip

[Far]

Now I should create mesh in 3D with the same parameters (something like extrude for 0.2)

This option works fine, if you want to simulate the 3d effects for 2d case (quasi 3d with symetry condition on both sides of wing). But if you really want the 3d wing mesh, that is meshing is also going after the wing and also inside the airfoil block exteded in the 3d then you should start with the 3d blocking and or you can make the 2d blocking and revolve it (as pointed by Simon and I am quoting exact words below)

There are several ways to handle a 3D wing. This one's key features are a sharp trailing edge (which suggests a C Grid) and a flat wing tip, which is most easily captured with (Blocking => Create => 2D to 3D).

Start with a 2D blocking... But because you will need to keep the flow volume past your wing tip, you also need to mesh inside the airfoil. The topology Far showed works well, or if you can handle wedges in your model, you could just keep the very simple collapsed block you got from collapsing the blocks behind the trailing edge, then add the Ogrid to the front inner block of the airfoil.

Then go to Create Blocks => 2D to 3D and choose the "by translation" option. Put in the width of your box and apply.

Then go and "delete" the blocks that should end up inside the airfoil (or put then in a solid blocking material).

Associate all the edges with the 3D curves, set up edge parameters, etc.

Done.

Detailed procedure is discussced here and diamondx already proivded you the link for YouTube video.
http://www.cfd-online.com/Forums/ans...mesh-wing.html

Blocking is also attached with above post for four different procedures. If you still find it difficult, let us know and we shall help you out.

[Far]

I am also going to add the voice in these videos soon.

[natala1987]

Thanks I am going to do it tomorrow and I will let you know ab.results

[PSYMN]

Just to be clear, it is possible to extrude the blocking or the mesh (2.5D), but Far is assuming that you are actually trying to mesh a real 3D wing with a wing tip...

If you really want to mesh the full wall to wall airfoil, extruding what you already have is the easiest way.

Best regards,

Simon

[natala1987]

My wing is symmetry so "extrude" is easier way than create geometry, blocking...I had problem with "extrude" because my boundary conditions did not work in Fluent, but I solved a problem. Before "extrude mesh" necessary is tick "lines" in mesh tree. Mesh and boundary conditions are working and results of lift and drag coefficient are the same as experimental results

[Far]

Actually, if you are trying to include the 3d effects then extrude is the best option. However if you want to solve the 3d wing then you need to consider the 3d blocking.

Can you post the few pics of goemetry, mesh and results.

[natala1987]

No problem but tomorrow, becuse I've back from university

[natala1987]

I can send you few print screens of my mesh. In 3D case I have 4 layer (every one 0.05). It is not too much but it is enough to compare with experimental results. In the next part I will simulate accretion of icing on the airfoil.

I am not sure what mean Area and Length in references value in Fluent? could you help me?

[Far]

Are you taking 30% of chord to extrude airfoil in 3rd direction? Four nodes may be sufficient I am not sure. Length is air-foil chord and area is the chord * unit because you are simulating the 2d airfoil

Model top and bottom wall with slip condition (zero shear) and dont apply pressure far field.

[natala1987]

I am confused ;/ I should talk with my supervisor. So in your opinion which depth will be the best, 1?

In 3D case to compute area I should multiply chord *depth?

[Far]

depth should be 0.3*chord to 0.4*chord for the 2d cases with 3d effects. I have learned it from the research paper.

Since you are simulating the 2d cases, therefore you should take the 2d properties to normalise the data.