CFD Online Discussion Forums - [swak4Foam] Implementation of Robin BC by groovyBC : difference wrt. theoretic solution

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)

- OpenFOAM Community Contributions (https://www.cfd-online.com/Forums/openfoam-community-contributions/)

- - [swak4Foam] Implementation of Robin BC by groovyBC : difference wrt. theoretic solution (https://www.cfd-online.com/Forums/openfoam-community-contributions/99669-implementation-robin-bc-groovybc-difference-wrt-theoretic-solution.html)

Implementation of Robin BC by groovyBC : difference wrt. theoretic solution

Hello everyone, we consider the following stationary heat equation:

Laplacian(T)=0

Code:

-----------T(x)---------- a bar with length L=0.4 m, mesh Delta x=1 mm

x=0        x           x=L (patch outlet)

BC :
At x=0 : T(0)=T0=273 K
At x=L : kT'(x)=h*(T(L)-Tinf) with k=237 W/mK, h=-100 W/m^2K, Tinf=873 K

We have implemented this BC using swak4Foam, in 0/T for the patch outlet:

Code:

            type                groovyBC;

            variables           "k=237;h=-100;Tinf=873;";

            valueExpression     "Tinf";

            fractionExpression  "1/(1-k/(mag(delta())*h))";

            value                    $internalField;

We are using the "steadylaplacianFoam" where we solve only the stationary heat equation. After about 5 iterations the solution OF has converged.

However, the OF solution differs a little from the theoretic solution :

Attachment 12409

Quote:

Originally Posted by tianyikillua (Post 353794)

Hello everyone, we consider the following stationary heat equation:

Laplacian(T)=0

Code:

-----------T(x)---------- a bar with length L=0.4 m, mesh Delta x=1 mm

x=0        x           x=L (patch outlet)

BC :
At x=0 : T(0)=T0=273 K
At x=L : kT'(x)=h*(T(L)-Tinf) with k=237 W/mK, h=-100 W/m^2K, Tinf=873 K

We have implemented this BC using swak4Foam, in 0/T for the patch outlet:

Code:

            type                groovyBC;

            variables           "k=237;h=-100;Tinf=873;";

            valueExpression     "Tinf";

            fractionExpression  "1/(1-k/(mag(delta())*h))";

            value                    $internalField;

Convergence is relative. Have you tried a stricter convergence criterion? Also try varying the mesh resolution to see if this is not just a discretization problem

After we changed the gradient scheme to leastSquares, everything is fine !! However we still don't know why...