[chenhu]

Being an old OpenFOAM user (almost 15 years), I still strugle with thermophysical library of OpenFOAM. The temperature calculation in OpenFOAM is quite messy and problematic. A very common error is temperature out of range when using Janaf polynonial equation for getting Temperatur from enthalpy.
Anyway, in order to figure out /debugg, where is the problem, I need to implement a new thermo type: hePsiThermo->myHePsiThermo. But my case cannot recognize this new thermo type:

Note that I changed psiThermo -> myPsiThermo (class reference, abstract)
and hePsiThermo->myHePsiThermo (class tempalate reference)

Selecting thermodynamics package
{
type myHePsiThermo;
mixture singleStepReactingMixture;
transport sutherland;
thermo janaf;
energy sensibleEnthalpy;
equationOfState perfectGas;
specie specie;
}



--> FOAM FATAL IO ERROR: (openfoam-2406)
Unknown psiReactionThermo type

Valid psiReactionThermo types :

23
(
hePsiThermo<homogeneousMixture<const<hConst<perfec tGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<homogeneousMixture<sutherland<hConst<p erfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<homogeneousMixture<sutherland<janaf<pe rfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<inhomogeneousMixture<const<hConst<perf ectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<inhomogeneousMixture<sutherland<hConst <perfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<inhomogeneousMixture<sutherland<janaf< perfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<multiComponentMixture<const<eConst<per fectGas<specie>>,sensibleInternalEnergy>>>
hePsiThermo<multiComponentMixture<const<hConst<per fectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<multiComponentMixture<sutherland<janaf <perfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<multiComponentMixture<sutherland<janaf <perfectGas<specie>>,sensibleInternalEnergy>>>
hePsiThermo<pureMixture<const<eConst<perfectGas<sp ecie>>,sensibleInternalEnergy>>>
hePsiThermo<pureMixture<const<hConst<perfectGas<sp ecie>>,sensibleEnthalpy>>>
hePsiThermo<pureMixture<sutherland<janaf<perfectGa s<specie>>,sensibleEnthalpy>>>
hePsiThermo<pureMixture<sutherland<janaf<perfectGa s<specie>>,sensibleInternalEnergy>>>
hePsiThermo<reactingMixture<const<eConst<perfectGa s<specie>>,sensibleInternalEnergy>>>
hePsiThermo<reactingMixture<const<hConst<perfectGa s<specie>>,sensibleEnthalpy>>>
hePsiThermo<reactingMixture<sutherland<janaf<perfe ctGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<reactingMixture<sutherland<janaf<perfe ctGas<specie>>,sensibleInternalEnergy>>>
hePsiThermo<singleStepReactingMixture<sutherland<j anaf<perfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<singleStepReactingMixture<sutherland<j anaf<perfectGas<specie>>,sensibleInternalEnergy>>>
hePsiThermo<veryInhomogeneousMixture<const<hConst< perfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<veryInhomogeneousMixture<sutherland<hC onst<perfectGas<specie>>,sensibleEnthalpy>>>
hePsiThermo<veryInhomogeneousMixture<sutherland<ja naf<perfectGas<specie>>,sensibleEnthalpy>>>
)

type mixture transport thermo equationOfState specie energy

[chenhu]

The thermophysical library is probably one of the most complex or messy parts in OpenFOAM with 6 levels objects and spider net connections, I think.

The problem is solved by modifying the following classes, including

src/combustionModels/CombustionModel/CombustionModel/CombustionModels.C
src/combustionModels/infinitelyFastChemistry/infinitelyFastChemistrys.C (I use this combustion model in my case)
src/thermophysicalModels/basic/psiThermo/*
src/thermophysicalModels/reactionThermo/psiReactionThermo/*

Then I could call a new thermoType as follows:

18 thermoType
19 {
20 type myHePsiThermo;
21 mixture singleStepReactingMixture;
22 transport sutherland;
23 thermo janaf;
24 energy sensibleEnthalpy;
25 equationOfState perfectGas;
26 specie specie;
27 }

Then I can start debugging the temperature calculation in the hePsiThermo.C file.

[Tobermory]

Haha - yes - agreed! It is the messiest part of OpenFOAM, and what's worse is that it keeps changing (in the Foundation version at least), so when you finally get your head around it, it's all out of date! Well done in figuring it out.

Incidentally, on the Janaf issue (usually T undershoot in my experience), you can use limitTemperature to keep it within bounds ... ofc that can still cause your code to blow up, but at least you don't end up with gigabytes of Janaf warning messages in your log file.

[dlahaye]

Do collaboration diagrams as shown at e.g. https://www.openfoam.com/documentati...calModels.html help here?

[chenhu]

Originally Posted by dlahaye

Do collaboration diagrams as shown at e.g. https://www.openfoam.com/documentati...calModels.html help here?

Yes, the diagram helps.

[Mid_Night]

Hi. I had a need to create a new viscosity calculation model for chemical components in OF. I have a little experience working with the source code OF 12, but I can't figure out how to add a new transport component. I found a couple of guides where this was discussed, but it was for very old versions OF. Could you help me, where should I start? As far as I understand, I have to copy a number of libraries and rename them, but which libraries should I copy? For example, if I want to change the simplest library where viscosity is constant, which parts OF IT should I copy?
Thanks!