[Chris T]

Dear OpenFoam users I have the following issue.


Setup:
I am trying to build a case in OpenFoam v9 using the solver chtMultiregionFoam. The case consist of 5 regions. 3 of air, 1 of solid material, and 1 of liquid. I broke the air region into 3 because the way I make my mesh solid_to_air and air_to_solid are defined automatically but I want to segment that interface to be able address different part of it separately.


Problem:
For solid_to_air1 interface I would like to apply heating to the surface of the solid as a means to simulate heating from solar radiation. In the past my model didn’t include air region(s), which meant that that patch was an external boundary, therefore I could simply use

boundaryField
  {
    in_radiation
    {
      type            externalWallHeatFluxTemperature;
      mode            power;
      Q               40000;
      kappaMethod     solidThermo;
      kappa           none;
      value           $internalField;
    }
 }

in the 0/solid/T file
but now since that wall is also an interface I have to use a different type boundary condition for the thermal coupling:

boundaryField
  {
   "solid_to_air*"
   {
     type compressible::turbulentTemperatureRadCoupledMixed;
     value           $internalField;
     Tnbr            T;
     kappaMethod     solidThermo;
    kappa           none;
     }
 }

My question is. How do I heat a region at its interface with another thermally coupled region?

What I looked into so far:

1. The use of heatSource in the fvModel file. The problem with that is that I don’t see, from the various examples, how to confine the heating just on the interface. Also that method seems to be limited only to volumetric heating.

2. I’m currently looking into using semiImplicitSource in the fvModel file. I’m still in the process of understanding how I can use a semiImplicitSource for h field to simulate surface heating. It seems to me that it suffers from the same issues as the heatSource method.

Extra considerations:
1. The possible solutions should be compatible with the application of a radiation model (View factor to be more specific).
2. Ideally the solution should be generalizable such that I would be able, in the future, to apply non-uniform heating. This is not a must though, for now at least.


If anyone has a suggested solution or some relevant resources to recommend please let me know.



Best regards,
Chris