Hello, Since the previous posting about LES in PHOENICS, I have been trying to use LES in BFC. However, there is some odd thing going on. Using similar geometry and setting for each case, I find that the results from BFC is not identical with Cartesian case. Actually, I do not see any eddies develping using BFC, althogh I tried many variations of time step sizes and grid dimesions. Is there any library case using LES and BFC together or any publication using PHOENICS as such? Or is LES compatible with BFC in PHOENICS? Thank you in advance.

I am not aware of any LES restriction for BFCs. The default grid filter for the Smagorinsky SGS model is one-third of the cell volume, as may be seen in the file /phoenics/d_earth/d_core/gxmxlen.htm. For wall-bounded flows, wall damping is advisable and so the LTLS variable must be solved to provide computation of the WDIS variable.

It might be a good idea to check whether the ENUT and LEN1 values are reasonable for BFC meshes. Check the values by hand for a few cells. It should be possible to restart from the Cartesian case, and perform a single-sweep run with the BFC case. The velocity fields will then be similar, but check whether the ENUT and LEN1 values returned for the BFC mesh are similar to those for the Cartesian mesh.