[guillenmugerza]

Hello there,

I am working on my final degree project, which is about a vertical wind turbine and I am currently on the phase of analyzing the different airfoils for the blades.

I did many trials of a Transient Analysis in fluent but I am getting some incorrect values when plotting the values for the Drag and Lift Coefficients.

The inner domain is a circumference around the air profile. The Outer Domain is a rectangle with half a circumference on one side, and one circumference around the Inner Domain, with a diameter 1e-8 m bigger than the Inner Domain.

I used a simple mesh with an All Triangles Method and some Edge Sizings for the boundaries of the Outer and Inner Domain and for the Airfoil.

For the Transient Analysis I set it up with a rotation of the Inner Domain of -0,01174533 rad/s, which is approximately of 1º/s and the calculation settings with 360 timesteps of 1s each so that it calculates every time the domain turns 1º and I get the Cd and Cl for a full revolution of the profile.

The problem is when the profile turns between 160º and 200º, when the trailing edge is towards the wind, the values of Cd and Cl start to take very anormal values and when plotting them the graph has a section where the values go up and down without following the overall shape of the curve.

I guess that the problem might be because of the wind is hitting directly on the sharp edge and maybe there are turbulences or something, or maybe be because of the mesh, but the rest of the values seem to be correct so I don't think that is the problem.

Any idea why could this be and how I can solve it?

I am attaching images of the Geometry, Mesh, and Cd and Cl graphs.

Thank you very much,

Guillermo Nuñez Muguerza

[vinerm]

From the images, it appears that you are using CFX and not Fluent. Or do you have the images taken from CFD-Post?

In any case, with certain angles, it is quite possible that there is flow separation. If the mesh is not fine enough and turbulence model is not appropriate to resolve the separation, then it could lead to strange numbers for the forces, and consequently for coefficients.

[guillenmugerza]

Hi!

Thank you for your answer. First of all, yes, the images were taken form the post processing and I am using fluent. I made an animation of the velocities and everything seems good until it hts that angle where it appears very different velocity zones.

I will try to refine the mesh or change the settings. And regarding the turbulence model, any suggestions of what might be the issue or what should I check?

Thank you very much again,

Guillemro Nuñez Muguerza

[vinerm]

Which model are you currently using to model the effects of turbulence?