How do you estimate the necessary spatial and temporal resolution for your LES calculations and experimental measurements if you know only the flow configuration?

I just heard that the mesh resolution should be as small as possible, according to your computer's capability. and the time step length should follow the CFL rule. If you receive any further comments, I would like to share it, thanks.

Generally I always find it to be a bit hit and miss. The best estimates are that your viscous stability number should not exceed 0.3, and you CFL limit must not exceed 0.3 either.

Build a mesh, select a time step (~ a few microseconds) and then run the code for about 100 time steps. It'll be obvious quite quickly if the code doesn't like the parameters you have applied.