Hi, I would like to understand what are the loops that CFX solves per iteration while solving a steady state problem. Is it the multigrid solver that is looping per iteration? I noticed this while plotting max residuals per iteration and it shows max residual at loop # 2 per iteration. Could someone throw some light on this. Thanks

Hi,

This is explained in the documentation. The Navier Stokes equations are non-linear equations so it uses an iterative approach to solve them.

Glenn Horrocks

Yes i know that. And that iterative approach is wat we see the solver solving every timestep. But im talking about the loops within each iteration that the solver solves.

Hi,

Yes, the AMG multigrid solver does internal iterations as part of its solution technique of starting with a coarse grid then refining down to the actual mesh. Again this is discussed in the documentation and more details is given in the references in the documentation.

Glenn Horrocks