[Zheng.Zhi]

the UEqn.H at interPhaseChangeFoam

fvVectorMatrix UEqn
(
fvm::ddt(rho, U)
+ fvm::div(rhoPhi, U)
- fvm::Sp(fvc::ddt(rho) + fvc::div(rhoPhi), U)
- fvm::laplacian(muEff, U)
- (fvc::grad(U) & fvc::grad(muEff))
//- fvc::div(muEff*(fvc::interpolate(dev2(fvc::grad(U) )) & mesh.Sf()))
);

but the UEqn.H at interFoam

fvVectorMatrix UEqn
(
fvm::ddt(rho, U)
+ fvm::div(rhoPhi, U)
- fvm::laplacian(muEff, U)
- (fvc::grad(U) & fvc::grad(muEff))
//- fvc::div(muEff*(fvc::interpolate(dev(fvc::grad(U)) ) & mesh.Sf()))
);

I don't understand why need - fvm::Sp(fvc::ddt(rho) + fvc::div(rhoPhi), U) at interPhaseChangeFoam , Could anybody please answer me ? thanks very much.

[nimasam]

fvm::Sp(fvc::ddt(rho) + fvc::div(rhoPhi), U)
this term would be zero mathematically because
fvc::ddt(rho) + fvc::div(rhoPhi)
is global continuity equation, and i guess it is added to consider
any numerical error in global continuity equation

[abe]

I am agree with Nimasam. It returns the continuity equation value. I have checked the results in two conditions with and without this term. Results are fairly the same. However, I should say that by using this term it seems that the convergence speed is much better.
PS: in the OF22, this term has been removed!

[abe]

Originally Posted by nimasam

fvm::Sp(fvc::ddt(rho) + fvc::div(rhoPhi), U)
this term would be zero mathematically because
fvc::ddt(rho) + fvc::div(rhoPhi)
is global continuity equation, and i guess it is added to consider
any numerical error in global continuity equation

Hi, have you been able to use interPhaseChange in OF21 or 22?
I can get results by previous versions (i.e. OF1.5). But by these new versions results are not good at all.


ABE