[geth03]

hi all,

i am basically simulating the mixing time behaviour in a closed stirred tank.
to account for turbulent diffusion i modified the scalartransportfoam code:

            fvScalarMatrix TEqn
            (
                fvm::ddt(T)
              + fvm::div(phi, T)
              + fvm::SuSp(-fvc::div(phi), T)
              - fvm::laplacian(DTt, T)
             ==
                fvModels.source(T)
            );

the third term in the equation is to bound T according to ScalarTransportFoam for RTD calculations. i did test the code without this term, unfortunately T explodes, so this term is helping with that.

so, for mixing time analysis i check the concentration at various locations. i saw that concentration converges to low values everywhere. so i was curious and i intergrated the T value over the whole domain and saw that the total amount of T is reducing during the simulation!

does anybody know the reason for that strange behaviour?

kind regards