[Aeroengineer]

I am currently working on a CFD analysis of 2D and 3D airfoils. I am using Pointwise to mesh the geometry and using ICEM to convert the .cas file to .msh (to be used on an in house code). The code is a RANS solver and I am using k-omega SST with the y+ value being less than 1.

I have been running unidimensional and compressible simulations but for some reason when I plot the Cp curve it is very unsteady (choppy or erratic). I have included the screenshot of the issue. I am not sure why this is happening.

I have carried out a grid independence study and as you make the grid fine, the unsteady behaviour is even more profound. I am using an o-grid for the mesh and it is normally extruded from the surface. My supervisor said it has something to do with 'spatial resolution' and I do not know what that means.

I have tried different boundary conditions but nothing seems to be working.

Any help would be greatly appreciated.

https://imgur.com/a/coIxnUi

[pdp.aero]

Originally Posted by Aeroengineer

I am currently working on a CFD analysis of 2D and 3D airfoils. I am using Pointwise to mesh the geometry and using ICEM to convert the .cas file to .msh (to be used on an in house code). The code is a RANS solver and I am using k-omega SST with the y+ value being less than 1.

I have been running unidimensional and compressible simulations but for some reason when I plot the Cp curve it is very unsteady (choppy or erratic). I have included the screenshot of the issue. I am not sure why this is happening.

I have carried out a grid independence study and as you make the grid fine, the unsteady behaviour is even more profound. I am using an o-grid for the mesh and it is normally extruded from the surface. My supervisor said it has something to do with 'spatial resolution' and I do not know what that means.

I have tried different boundary conditions but nothing seems to be working.

Any help would be greatly appreciated.

https://imgur.com/a/coIxnUi

Hi there,

Here is my general recommendation:

1. Regardless of what kinda CFD solver you would use for your simulation, do a simple verification case like NACA 0012 to make sure for a standard test case everything is working fine and there is no problem. When you do this, you’ll understand that numerical methods and models implemented in your solver is fine and there should be no problem with your specific case. If problem exists in your verification case, then something is probably wrong with the numerical methods you used for the simulation.

2. Check your airfoil and make sure, your CAD model is smooth and there is no problem with upper and lower surface of your airfoil.

3. Make sure you converged!

4. Use a very course mesh to do the simulation and compare the cp obtained from RANS with SST and k-epsilon or SA.

5. Switch between gradient options in your solver and try green-gauss cell based or node based in comparison with least square. (For structured mesh go for cell based for unstructured mesh go for node based.)