I am simulating steady flows in convergent-divergent nozzles with a bidimensional finite volume based code solver. I have some doubts about some results that I have obtained.

First, when I simulated flows with a steady normal shock wave in the divergent zone of the nozzle solving the Euler equations, the plots of the streamlines showed an important deflection across the shock. All the streamlines behind the shock are perfectly perpendicular to the shock, so I think that shouldn't have any deflection of the flow. I don't know if the situation that I described is physically possible or some error in the solution is the responsible.

Another question that I have is about viscous flows with a normal shock wave. When I solved the Navier-Stokes equations for the same case that I described (flow in a convergent-divergent nozzle with off-design conditions at the exit section), it never reaches the steady state. In the transient solution, the normal shock in the divergent zone moves and changes it shape periodically. I want to know if this behavior is correct, I can't find any reference to this situation in the textbooks that I have read.

I really appreciate if someone could give me a hand. Thanks.

the shock in a C-D nozzle isn't straight. see john anderson's CFD book on page 527. for the problem with your viscous problems you may wan to refine your mesh in the streamwise direction in the viscinity of the shock to reduce the oscillations