[ayush1989]

Hi,

I am new to CFD online and I would like to discuss about a problem I am having with my multigrid solution for supersonic MHD duct flow. I am using simple parabolic equations to solve axial velocity along the channel and a multigrid solver to solve elliptic equations for cross-stream velocities.

I am quite confused because of the way my solver is behaving. Whenever I am using 2 grids, I am having convergence but less accuracy. But when I am trying to use 3 grids, the solution converges on the coarsest grid but the solution gets stuck at a particular residual value on the interemediate fine grid and then continuously switches between the coarsest grid and the interemediate fine grid as the residual on the fine grid are not small enough.

Can anyone give me any suggestions or if anyone has faced such a situation?

Thanks