[yongxiang]

Hello everyone, I am quite new to OpenFoam, especially to the programming part. Now I would like to solve the following equation:

ddt(U)+div(phi,U)-laplacian(nu,U) == -grad(p) - chi * (U-Ubar)

ddt(Ubar) == omega * (U - Ubar)

in which Ubar is another field vector. For the last red terms in both equation, they are direct multiplication of scalar and vector field. How should I discrete it? Like for other terms: fvm::div(phi,U), what is the operator for such a direct calculation?

Many thanks
Yong

[mkraposhin]

first equation

fvm::ddt(U)
+
fvm::div(phi,U)
-
fvm::laplacian(nu,U) 
+
fvm::SuSp(chi, U)
-
fvc::Sp(chi, Ubar)
==
-grad(p)

second equation

fvm::ddt(Ubar) 
+
fvm::SuSp(omega,Ubar)
-
fvc::Sp(omega,U)

[yongxiang]

Hi Matvej, thanks for the quick reply. I've tried your suggestion and find that the following term can not pass compilation.
fvm::SuSp(chi, U)

with the error:

error: no matching function for call to ‘SuSp(Foam::dimensionedScalar&, Foam::volVectorField&)’
+ fvm::SuSp(chi, U)

it seems the match problem of dimensionedScalar and volVectorField.
But if I use
fvm::Sp(chi, U)
the compilation passed.
Would you mind to explained a bit for me the difference of Sp and SuSp.
Regards
Yong

[mkraposhin]

fvm::Sp - implicit source term
fvc::Sp - explicit source term

fvm::SuSp(chi,U) - dynamically switched term, for cells where chi >= 0 it is treated as implicit , where chi < 0 it is treated as explicit to preserve diagonal elements untouched

if you want to use fvm::SuSp, then variable chi must be of type volScalarField

[yongxiang]

Hi Matvej,
this time I understand, thank you very much
regards
yong