i will make my own code to investigate turbulent flow around a cube using RANS. which method of medelling is best for flow around a cube. as i know there is various medelling methods(k-epsilon or k-omega...). thanks for all

the RSM is best RANS model but lots of extra equations to solve. k-epsilon is the best basic model.

If you want to program it into a code I have no experience of this but I expect k-epsilon is easiest but have you considered DNS? For reynolds numbers not too high it can be done on a decent pc. As far as I know it's the most basic to program as simply no turbulence model required.

phil

PS - make sure you use a very dense grid close to the cube to try and resolve the boundary layer. You need to read up on this if you are serious about developing a code.

i am coding rans because it is my turbulance modelling course project. i would choose les or rans and i chose RANS. thanks.