Guyer, Jonathan E. Dr. <jonathan.guyer@...> writes:
> > * Replacing the T.faceGrad constraint with a flux divergence source helps: > > eq = DiffusionTerm(coeff=D) + (alph * (Tinf-T.faceValue) * mesh.faceNormals * mesh.facesTop).divergence > > Max,Min 500.0 450.943343505 49.0566564949 > 450.0 expected surface value 50.0 150.0 0.333333333333 > Thank you! After playing around I also found that putting the T.faceGrad constraint into the sweep also resoved the problem. My best regards Burak > * Increasing the mesh resolution to dx = dy = 0.000025 helps further: > > Max,Min 500.0 450.093747663 49.9062523366 > 450.0 expected surface value 50.0 150.0 0.333333333333 > > Surprisingly, replacing T.faceValue.value with T.faceValue doesn't make any difference, but you > should do it anyway. > > Wheeler may be able to speculate why the T.faceGrad constraint doesn't do a very good job in this case. > _______________________________________________ fipy mailing list [email protected] http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
