From: Kyunghoon Lee <aeronova.mailing@gm...>  20130208 07:29:21

Hello all, I've been testing transient_ex1 with inhomogeneous Dirichlet BC after simplifying it (no convection, no density). If I specify homogeneous Dirichlet BC, the example gives me a solution whose boundary values are exactly what I put. However, if I try inhomogeneous Dirichlet BC, boundary values change after one time step and remains as the changed value throughout time march. For instance, I set inhomogeneous Dirichlet BC with 1 for the left and right side and 1 for the bottom and top sides, the boundary values are no more 1 and 1, respectively, after one time iteration (plz see the figures). To set the boundary values, I modified the exact_solution function as follows: Real exact_solution (const Real x, const Real y, const Real t) { if ( (x == 1.0)  (x == 1.0) ) // left and right sides return 1.0; else if ( (y == 1.0)  (y == 1.0) ) // bottom and top sides return 1.0; else return 0.0; } I'd appreciate if someone could help me figure out the problem. 