Hi, I'm very new to Fluent and Gambit and CFD but I have worked the examples in the Cornell tutorials. My question is, what type of turbulence model I should use to model the flow separation of an airfoil and how fine should the C-grid be to accurately obtain the section lift coefficient at that particular angle of attack (AOA) or higher AOA?

Previously I've used the SST k-w model with the grid being the same as the one in the Fluent tutorial by Cornell for the NACA 0012 airfoil. It has 12150 elements. The stall angle was off by about 2 degrees and the Cl obtained above it does not even approximate the experimental data. Any help in resolving this is greatly appreciated. Comments and extra readings please tell me about it. Really like to know more.

Thanks.

i would try using the realizable k-eps model. grid adaption might be a good idea as well

Hello Kevin, just a few pointers over here. 12150 elements might be too little. There would be a lot of "action" in the nearfields, hence you'll need sufficient elements to capture it. As such, i would reccomend that you:- 1) Try using a boundary layer grid around the airfoil 2) Try grid adaption ( but this might be time consuming) 3) Increase the number of elements, perhaps somewhere around 80,000 to 100,000. Check with a Grid Convergence Index (GCI).

As for the solver, SST k-w apparently works well for scenarios such as this, but mypreference is always RSM.

Thanks, Matt

Hi all, thanks very much for the replies. As for the number of elements, 80000 is pretty high for my comp which has only about 2 gigs of ram. Don't know how long it'll take, but will try it somehow. Thanks again. At least I got the model ok. LOL.

hi sir,

I would like to ask you a question about modeling a mesh around NACA 4412 at angle of attack 12o. I will use the many block type to divide the one block in many block as it necessary. What is the best way and which type of mesh should I use?

Thank and appreciate your support

Khalid al anazi

Check Fluent Tutorial of Cornell University