[Zon]

Hello,

I am a Master's Degree student and I am working on a project that uses sonicFoam as a solver.

The cases I am simulating are basically a fluid in supersonic flow within an ejector. I used the type of cell "wedge" to simulate just one cell (1° of the total cylinder from the original ejector design) in the radial direction and refined the mesh in the longitudinal direction (x and y planes).
However, the cases are diverging, i.e., the Mass Flux Error between Inlet and Outlet is high, mainly when I increase the mesh refinement just in the longitudinal direction . I tried to use the PengRobinson equation but the solver restricted me to use just sensibleInternalEnergy as option to energy in the thermophysicalProperties, so I coudn't use polynomial and hPolynomial as option to transport and thermo, respectively.
Thinking on it, I recompiled the solver (based on rhoPimpleFoam) using the sensibleEnthalpy instead of sensibleInternalEnergy and used the PengRobinson equation, which gave me very good results. But I wanted to know:

Which aspects of the calculation those changes affects the simulated physics of the sonicFoam?



Notes: Changes in the sonicFoam are described below.

----------------------------------------------------------------------------------------
1. createFieldRefs.H
Commented the term "volScalarField& e = thermo.he();";

2. createFields.H
In the line "thermo.validate(args.executable(), "e");", the term "h" was added next to "e", resulting in "thermo.validate(args.executable(), "h", "e");";

3. EEqn.H
The terms "e" were replaced for "he";
Added a line after the "{" with the information "volScalarField& he = thermo.he();".
----------------------------------------------------------------------------------------

Thanks for helping me.