Great video. Btw if your results are mesh dependent then that means you have to do grid verification. In CFD, the mesh change shouldnt influence your results, especially in area of interest near the body. Drag, Wall shear stress should not be dependent on the mesh.
Interesting, I have not heard of this before. I will look into grid verification. However, with my current CFD, the mesh size does seem to affect both drag and downforce by 1ish percent with minor modifications to the mesh near aerodynamic elements. Does this variation mean that my CFD is not percise?
@@Idontknow-ec5bj If it is that small a variation then your CFD is precise and solution can be considered grid independent. Most literature and industry accept errors less than 5%. For example, Ma < 0.3 is incompressible with density variation being less than 5% . This means your solution is converged. Btw, when you refine the grid, are you reducing the wall y+ as well? I have noticed that y+ < 1 and y+ approx 0.1 lead to around 2-3% variation in wall shear stress.
I'd say - there is a long way in here till thinking about any kind of mesh independency: at 08:49 transition from hex cells to prism layers (of uniform thickness!) is .. beyond good and evil %)
@@sergeianpilov5222 Indeed. that abrupt change would manifest itself in errors when doing grid sensitivity. It may be okay if that region is outside the BL. That depends on the flow conditions.
Very bad video, or at least title. "Mastering" implies understanding the settings of the software, that is not what you do. You just show your own settings which you think they are the best. I would title this video "How to START with meshing"
Awesome, thank you
✈
Great video. Btw if your results are mesh dependent then that means you have to do grid verification. In CFD, the mesh change shouldnt influence your results, especially in area of interest near the body. Drag, Wall shear stress should not be dependent on the mesh.
Interesting, I have not heard of this before. I will look into grid verification. However, with my current CFD, the mesh size does seem to affect both drag and downforce by 1ish percent with minor modifications to the mesh near aerodynamic elements. Does this variation mean that my CFD is not percise?
@@Idontknow-ec5bj If it is that small a variation then your CFD is precise and solution can be considered grid independent. Most literature and industry accept errors less than 5%. For example, Ma < 0.3 is incompressible with density variation being less than 5% . This means your solution is converged. Btw, when you refine the grid, are you reducing the wall y+ as well? I have noticed that y+ < 1 and y+ approx 0.1 lead to around 2-3% variation in wall shear stress.
I'd say - there is a long way in here till thinking about any kind of mesh independency: at 08:49 transition from hex cells to prism layers (of uniform thickness!) is .. beyond good and evil %)
@@sergeianpilov5222 Indeed. that abrupt change would manifest itself in errors when doing grid sensitivity. It may be okay if that region is outside the BL. That depends on the flow conditions.
Very bad video, or at least title. "Mastering" implies understanding the settings of the software, that is not what you do. You just show your own settings which you think they are the best. I would title this video "How to START with meshing"