Drag coefficient of square cylinder
I'm testing a new CFD code using the turbulent flow around a square cylinder as a validation case. Pulblished drag coefficient Cd (measured and CFD) are typically between 2.02.1 for Re of 20000. My initial 2D runs gave Cd of 2.8 which is significantly higher than established number. At this point I can either try finer mesh size or move to 3D which takes a lot longer. My question is, can one get the 2.0  2.1 drag coeff. using 2D modeling?

yes ofcourse . i had done the simulation with Re=22000 and i obtained the value of cd=1.98
(kw sst model ) 
Hi All,
Thanks for posting this topic. I also got about the same value, CDmean = 2.13 and CLrms = 1.36, using 2d kwsst model and a wall function. The number of cell along the edge is 30 x 4edges. But when I refine the grid to 80 x 4edges and without wall function (y+ < 3), the CDmean drop to 1.57 and CLrms drop to 0.45. Any comment?? Thanks, 
All times are GMT 4. The time now is 17:15. 