Dear Richard,
You are absolutely right in saying that if you have the Stokes equation
your problem should be linear, even with the PSPG stabilization term.
Laplacian of u is only going to be zero for Q1-Q1 elements if your mesh is
perfectly aligned with the axis of your domain.
However, this is
Dear deal.ii community,
I solved it, sorry for bugging you with it but simple mistakes can bug you
for long...
I simply forgot to re-distribute the dofs for the finite element after
refining the mesh. Ooofff :-/
Best,
Konrad
--
The deal.II project is located at http://www.dealii.org/
For ma
Dear Wolfgang,
Thanks also to you for taking the time, I really appreciate it!
I already tested the integration, increasing up by order+1, +2, +3 ... gave
identical results.
Now it seems, that I should introduce some switches to a direct solver!
This might give me a hint, if I need some rescalin
Hi Bruno,
Thank you very much for taking the time to read my post & giving me some
ideas!
I am already familiar with lethe, thats where i compared my implementation
of the residual to, but for Stokes,
we actually do not introduce a nonlinearity, since we do not have the
(nonlinear) convective t
