Learning Forum

[itb7]

Hello,

I am fairly new to ANSYS and have been trying to get accustomed to finding the drag coefficients of various airfoils. To start with, I attempted to find the drag coefficient of a sphere, which is well known to be 0.47 at a Reynolds number less than 1E5. However, the result which Fluent returned to me after converging was 0.0041759, which does not make any sense to me. If anyone can identify what my mistake was in either meshing or setup, please let me know. The settings are listed below:

Model type: 0.5 m sphere within a 2x1x1 rectangular enclosure

General:

-pressure-based

Models:

-Realizable k-epsilon

Materials:

-Fluid, Air, ideal-gas model

Boundary Conditions:

-velocity inlet, v=3 m/s

-no-slip boundary on the sphere

-shear force of 0 on the walls

Methods:

-second-order on all spatial discretization settings

Iterations:

5000

Any advice is appreciated!

 

[DrAmine]

Ensure deep convergence and that the reference values used to norm the drag coefficient are set appropriately.

[itb7]

I'm not sure whether or not this constitutes deep convergence, but pictured here are my residuals and reference values. I'm beginning to suspect the issue is in the meshing however; please let me know if you know of a better mesh setup than the one I am using.

[DrAmine]

Your mesh is not good along the sphere: The sphere is highly resoled that is fine but the boundary layer and the jump into the core is accompanied with high cell volume change ration (from picture). Increase the mesh check verbosity to 2 and check the quality again. Another point here is that your domain is containing the sphere: try to put the top and bottom walls / boundaries far way from sphere.

 

And the convergence is not ideal also do not judge convergence just by residual: monitor drag and lift and check if they are still varying.