[krikre]

Hello everyone!

Why is in compressibleInterFoam T first calculated from the temperature equation TEqn and then again from internal energy he() in heRhoThermo.C with calculate()?

        TCells[celli] = mixture_.THE
        (
            hCells[celli],
            pCells[celli],
            TCells[celli]
        );

I tried to comment this out and it seemed to impair the calculation of pressure. I'm guessing this piece of code enforces the equation of state? My idea was to eliminate internal energy calculation since temperature is governed by the TEqn. So if T needs to be corrected to enforce the equation of state, could I make this correction without calculating the internal energy and just use the equation of state to correct the temperature? Could I correct pressure instead?

[aunola]

It isn’t.

I am going by memory and it’s been a long time since I’ve been working with OF (2.2.x), so I may we wrong about the details. But as far as I remember the twoPhaseMixtureThermo that compressibleInterFoam uses inherits from basicThermo and this is where the solver gets its p, T fields from. The twoPhaseMixtureThermo has two thermo objects, one for each phase. Each of those also inherit from basicThermo so they both have their own p, T fields as well.

The solver uses the p, T fields of the twoPhaseMixtureThermo. Those fields are not known to the phasic thermo objects. The twoPhaseMixtureThermo in turn doesn’t know how to calculate transport properties of the constituent phases. Only the phasic thermo objects know that.

In the correct() function of twoPhaseMixtureThermo there should for each phase ‘x’ be a line something like this: thermox->he() = thermox->he(p,T). The right hand side creates a new volScalarField with enthalpy initialized from the current p and T fields of the solver (see heThermo). It then assigns that field to the same thermo object. In turn, the correct() function can be called on each of the phasic termo objects to update the transport properties.

You don’t state exactly what you commented out, but if it is the calls to the thermo object correct() functions inside the twoPhaseMixtureThermo then it will not work. The missing update of transport properties – e.g. compressibility used in the pressure equation - leads to field inconsistency. Eventually something will blow up, quite likely pressure.

[krikre]

You pointed me in the right direction, thank you!

I then checked the phase temperatures T.air and T.poly and they weren't changing during the calculation, because of the commented code. I wasn't aware of the different temperature objects.

#include "mojHeRhoThermo.H"

// * * * * * * * * * * * * * Private Member Functions  * * * * * * * * * * * //

template<class BasicPsiThermo, class MixtureType>
void Foam::mojHeRhoThermo<BasicPsiThermo, MixtureType>::calculate()
{
    //const scalarField& hCells = this->he().internalField(); //commented out
    const scalarField& pCells = this->p_.internalField();
    const scalarField& strigCells = this->strig_.internalField();

    scalarField& TCells = this->T_.internalField();
    scalarField& psiCells = this->psi_.internalField();
    scalarField& rhoCells = this->rho_.internalField();
    scalarField& muCells = this->mu_.internalField();
    scalarField& alphaCells = this->alpha_.internalField();

    forAll(TCells, celli)
    {
        const typename MixtureType::thermoType& mixture_ =
            this->cellMixture(celli);

        //TCells[celli] = mixture_.THE //commented out
        //(
        //    hCells[celli],
        //    pCells[celli],
        //    TCells[celli]
        //);

I didn't realize it, but I guess the individual phase temperatures are a feature of the original solver. In my case, only the polymer is to be modeled and air is there just to fill the empty space.

This now allows the phase temperatures to be overridden.

void Foam::mojTwoPhaseMixtureThermo::correct()
{
    //thermo1_->he() = thermo1_->he(p_, T_);
    thermo1_->T() = T_;
    thermo1_->correct();

    //thermo2_->he() = thermo2_->he(p_, T_);
    thermo2_->T() = T_;
    thermo2_->correct();

[mikulo]

Originally Posted by krikre

You pointed me in the right direction, thank you!

I then checked the phase temperatures T.air and T.poly and they weren't changing during the calculation, because of the commented code. I wasn't aware of the different temperature objects.

#include "mojHeRhoThermo.H"

// * * * * * * * * * * * * * Private Member Functions  * * * * * * * * * * * //

template<class BasicPsiThermo, class MixtureType>
void Foam::mojHeRhoThermo<BasicPsiThermo, MixtureType>::calculate()
{
    //const scalarField& hCells = this->he().internalField(); //commented out
    const scalarField& pCells = this->p_.internalField();
    const scalarField& strigCells = this->strig_.internalField();

    scalarField& TCells = this->T_.internalField();
    scalarField& psiCells = this->psi_.internalField();
    scalarField& rhoCells = this->rho_.internalField();
    scalarField& muCells = this->mu_.internalField();
    scalarField& alphaCells = this->alpha_.internalField();

    forAll(TCells, celli)
    {
        const typename MixtureType::thermoType& mixture_ =
            this->cellMixture(celli);

        //TCells[celli] = mixture_.THE //commented out
        //(
        //    hCells[celli],
        //    pCells[celli],
        //    TCells[celli]
        //);

I didn't realize it, but I guess the individual phase temperatures are a feature of the original solver. In my case, only the polymer is to be modeled and air is there just to fill the empty space.

This now allows the phase temperatures to be overridden.

void Foam::mojTwoPhaseMixtureThermo::correct()
{
    //thermo1_->he() = thermo1_->he(p_, T_);
    thermo1_->T() = T_;
    thermo1_->correct();

    //thermo2_->he() = thermo2_->he(p_, T_);
    thermo2_->T() = T_;
    thermo2_->correct();

Hello,

Can you please explain what are your goals here and why you commented on some lines? I can barely follow what you did. I need to understand more since I want to modify the TEqn also to my purpose.

Regards,
Mike