[jexposito]

Hello,

I am trying to run a simulation of a 2D-Axysimmetric problem. I am using a density based solver with air as ideal-gas. The problem that I am trying to solve is a super-sonic ejector, i.e a converging-diverging nozzle where a high pressure fluid is accelerated and due to the Venturi effect a secondary fluid is suctioned.

The problem that I found is that if I run a 2D planar simulation I am able to make it converge. However, if I run it as a 2D axisymmetric (I change the boundary condition from symmetry to axis and the type of problem) the simulation does not converge and I have a lot of problems with reverse flow. I have tried to interpolate, as initial guess, from the 2d-planar simulation but it doesn't help.

Do you have any suggestion?

Thank you in advance and best regards.