[francesco_capuano]

Hi everyone,

I have some problems in setting up the proper boundary conditions for an internal compressible flow (e.g. flow through a convergent-divergent nozzle), using a generic compressible solver like rhoSimpleFoam. I have found out that many OF users have encountered the same issues, so I think it would be nice if, once and for all, we could find a solution.

Tipically, such cases comprise, in addition to a wall (on which a no-slip condition is assumed), an INLET face and an OUTLET face. So the question is: what are the right boundary conditions to be set for p and U on such faces in order to reach a convergent solution?

Thanks in advance.

[francesco_capuano]

Hi,

someone who could send a converged test case for a similar flow would be greatly appreciated.

Thanks

[crixman]

Hi Francesco,
did you make any progress? I am trying to modify rhoSimplefoam as well as I am having problems with floating points due to thermal properties, maybe we can work on that together, let me know!

[stosse]

Hi, guys!
I have been working on a similar problem for the last couple of weeks (using a modified version of the fireFoam solver), and have had severe difficulties getting a correct flow field with a module with inflow/outflow. I have tried many different configurations, but what has worked for me is to use totalPressure BCs for p_rgh with a pressure gradient between the inflow and outflow patch, and pressureInletOutletVelocity for U.

p_rgh:

outlet
{
type totalPressure;
rho rho;
psi none;
gamma 1.4;
p0 uniform 101275;
value uniform 101275;
}
inlets
{
type totalPressure;
rho rho;
psi none;
gamma 1.4;
p0 uniform 101325;
value uniform 101325;
}

I am not very experienced with OpenFOAM, so this is a result of trial and error. It would be nice if some more experienced users know why this setup (compressible flow, inlet and outlet) is so unstable.

Regards,
Sindre

[crixman]

Hi Sindre, thank you for sharing your findings, would try it out when I have some free time!