Learning Forum

[Simranjeet Singh]

I am attempting a steady-state simulation on a nozzle using inlet, outlet, far wall, and symmetry boundary conditions. Both the inlet and outlet are specified as pressure boundary conditions. While I can successfully run the simulation using a planar 2D space, I encounter an issue when switching to an axisymmetric setup. In this case, the gas seems to be unable to exit the thruster nozzle. The simulation involves nitrogen as the medium. I appreciate your assistance and insights.

 

[Rob]

For pressure in & out the mass flow becomes part of the solution, so you need additional care with the initial conditions. That doesn't alter the converged solution but does alter the convergence: you may not have a converged solution. That's also the Mach number, what do the velocity & pressure fields look like? 

[Simranjeet Singh]

 

Thank you, Rob, for your response.

You’re correct, I haven’t achieved a converged solution yet. In addition to that, I’m observing reversed flows after just 230 iterations out of the total 3000 iterations. Interestingly, the planar 2D simulation doesn’t converge either even when run for 3000 iterations. Nonetheless, it does provide me with some results that I can compare.

 

 

 

 

[Rob]

2d planar is a slot, axi-symmetric will be a hole so the areas are very different. How much mesh is there in the throat? Did you patch the domain core region with pressure & velocity to give the solver some help? 

[Simranjeet Singh]

 

I’ve implemented edge sizing at the throat of the nozzle, utilizing 10 divisions to refine the mesh in that critical region. For the initialization, I opted for the ‘Standard Initialization’ method and computed the initial conditions based on the ‘Inlet’ boundary condition. As a result of this approach, I’ve confirmed that the pressure and velocity values closely align with the analytical calculations I performed. This gives me confidence in the accuracy of the initial setup.

 

 

 

[Rob]

Have a look at the Patch tools.