[leroyv]

Dear OpenFOAM users,

I am new to OpenFOAM and I am trying to perform heat transfer DNS on a geometry consisting of both solid and fluid regions. The flow in the fluid region is laminar and compressible. My investigations led me to the conclusion that the best solution for me would be to use the chtMultiRegionFoam solver or its steady-state version chtMultiRegionSimpleFoam.

I looked into this solver, and I followed the little tutorial on the chtMultiRegionSimpleFoam solver available on the OpenFOAM wiki:
http://openfoamwiki.net/index.php/Ge..._-_planeWall2D

When looking at the inner boundary conditions, it happens that these are pretty elaborate, defined by the turbulentTemperatureCoupledBaffleMixed patch. My model is very simple, and the inner heat boundary conditions are merely temperature and conductive flux continuity at the fluid-solid interface, with a potentially non-zero interface source term.

My question is: Is it possible to reduce the turbulentTemperatureCoupledBaffleMixed patch so that there is no temperature or heat flux jump at the interface? If not, is there a less elaborate patch that I could use?

[NewtoFOAM]

Hi,

I am in no way an expert in this. But if you look at the src code for turbulentTemperatureCoupledBaffleMixed ($FOAM_SRC/turbulenceModels/compressible/turbulenceModel/derivedFvPatchFields/turbulentTemperatureCoupledBaffleMixed) then you can see that this boundary condition is simply trying to make both regions agree on temperature value and gradient at the boundary.

Good luck
Dave

[leroyv]

Originally Posted by NewtoFOAM

Hi,

I am in no way an expert in this. But if you look at the src code for turbulentTemperatureCoupledBaffleMixed ($FOAM_SRC/turbulenceModels/compressible/turbulenceModel/derivedFvPatchFields/turbulentTemperatureCoupledBaffleMixed) then you can see that this boundary condition is simply trying to make both regions agree on temperature value and gradient at the boundary.

Good luck
Dave

Hi Dave, thank you for your answer. We are probably reading the same comment in the source code:

// Both sides agree on
// - temperature : (myKDelta*fld + nbrKDelta*nbrFld)/(myKDelta+nbrKDelta)
// - gradient    : (temperature-fld)*delta

What this actually means is unclear to me. What temperature should both sides agree on? The wall temperature? It's called 'mixed', so it must do sometime with gradients too, probably invisibly using class inheritage (see in the associated header).

To my understanding, this patch does more than just apply a temperature and flux continuity BC. From what I could learn from various sources I can't find back, turbulentTemperatureCoupledBaffleMixed includes (among others) models for near-wall radiative transfer, turbulent and laminar conducto-convective transfer and for conductive transfer in "baffles" (thin solid layers). All of these examples involve a temperature jump BC.

And what intrigues me the most is this kappa setting, which I don't know anything about.