[ChRiZZ]

Hello! I am investigating the aerodynamic profile of the Naca 0018 at various angles of attack. The experimental data does not match the ones I received at Ansys and it freezes me out wildly. Can you tell me what I'm doing wrong? I used the Spalart, k-e, k-w, SST model and others models. Methods everywhere second order. Lift calculated as (-y;x), drag as (x;y), where x and y are the angle of attack. If the angle of attack is 0 degrees then x=1 and y=0. What could be wrong and where could I have made a mistake? Thanks for the help

[LoGaL]

With a mesh like the one you are showing, I am suprised fluent converges... Do you think the white shape you show correctly represents the naca0018?

[ChRiZZ]

Originally Posted by LoGaL

With a mesh like the one you are showing, I am suprised fluent converges... Do you think the white shape you show correctly represents the naca0018?

Unfortunately, it is impossible to make a more accurate grid in the student version, I tried to make the most of what I have. And the aerodynamic profile itself is built correctly, I took it from an open source. Thanks for the answer

[LoGaL]

Its not the number of elements, Look at what is happening at the leading and trailing edge! It's horrible to see, what are those giant triangles???


Check the screenshot I made

[ChRiZZ]

Originally Posted by LoGaL

Its not the number of elements, Look at what is happening at the leading and trailing edge! It's horrible to see, what are those giant triangles???


Check the screenshot I made

I hope you are right and this will help me solve the problem, I will try to fix it somehow. Thanks

[Peppinfante]

I have done a similar simulation in the past so maybe I can help you. As already said, the mesh doesn't look good. Those triangles on the LE seems to come out of a 3d setup. Before starting meshing have you checked on the right side of the worbench that the analysis is 2d? If, instead you want to do a 3d simulation, be sure to insert a sizing or a face meshing on that part. Said that, the inflation also looks too strong, try keeping it closer to the profile since you don't expect such a big boundary layer. Finally, a good rule is to make the transition from inflation to free mesh as smooth as possible and yours doesn't look like that. I am attaching some pictures of a mesh I used some years ago to do 2d incompressible viscous calculation (the profile is not the same).

P.S. Be sure to check the exact value of the first layer height using some Y+ calculator online (giving the Re as input) and be sure that the skewness value is not higher than 0.9 and that the orthogonal quality is higher than 0.1

[arnie333]

1. You need to add surface sizing to your entire wing surface and add inflation layers.
2. To achieve reasonable correlation with literature in Cl and Cd, ensure that your Reference Parameters are correct I.e. frontal surface area for drag
3. I doubt a student license will allow you enough cells to get correlation.