Hi @simscale8,
I’ve post-processed your results in ParaView. As you can see in the screenshot below for CL and CD. Your results are very off by almost 95%. This is probably due to a number of reasons.
The first thing I must clarify is that the calculation of CL and CD is based off the lift and drag equation. While the variable of density is most certainly correct, the speed and effective area might be incorrect. I have based off the speed as your inlet velocity of 14 m/s to obtain this CL CD data that you see now. It may be incorrect on the other two variables so I’ve gone ahead and explain what I used.
However, if we think about the rotating zone, your inlet velocity may not represent the correct velocity, as such I’ve went ahead and measured the geometry to obtain the velocity which is slightly less than 14 m/s, so assuming this is correct, the results don’t deviate as much so no worries there.
Effective area however might be an issue. I’ve merely extracted the surface area from Paraview as you can see in the figure under “area” and from what I can see it may not be accurate. So do let me know what the surface area is of the wheel.
On to the other possible sources of error.
Referring to the figure below. Your inflation layers for achieving your Y+ is not present at all. This is likely due to the mesher erasing your Y+ due to the quality control. Refer to this post here to possibly fix the issue. Also ensure that your “Minimum overall layer thickness” is significantly less than the actual overall layer thickness. It will just erase the layers if the inflation does not exceed this limit.
Next, same issue with your bounding box inflation layers, they are not showing up. This one you can refer directly to your project and observe that at the bounding box, there are no layers at Y min. The likely fix is similar to that of the previous paragraph.
Also for you boundary condition “Bottom-MW” is it supposed to be assigned to the inlet face as well? I don’t think so, so do remove that unless I’m mistaken.
You will lastly need to ensure convergence is met. Your current residual levels are quite high and unless somehow this is alright, you want to get them below 1E-4 at least. This means adjusting the numerics for your relaxation values for p or U by decreasing them. Do ensure the end time is longer if you do this as it takes longer to converge. Ensure that your mesh and boundary conditions are alright first.
Hope this helps.
Cheers.
Regards,
Barry