Hello Mihai, > a simple, smooth function u = sin(2pi x) cos (2pi y)
Even though your function u is smooth, the circular vector-field beta is not, it is not differentiable at the origin. The same will be true for your right hand side. So you should not expect optimal orders of convergence, in particular if your u is not constant around the origin. To obtain a smooth case you could shift the origin of the vector-field out of the domain, e.g. to (-a,-a) with some small a. If a is rather large you should quickly see improved orders of convergence. If a is rather small, then beta is still close to singular and the convergence rates will probably improve only asymptotically. Best, Tobias _______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
