Hi Wolfgang, Sorry for the reply - for some reason i didn't get the notification.
What i mean by correct is that I do have two uncoupled equations, but that it is correct in that I had already verified it using what you suggested (by using exact solutions - MMS). By it fails, I mean that the top pf value output is not equal to pr as it should be, which I set in the weak form as in my first message in my thread: local_rhs(i) += -( fe_face_values_pf[velocities].value (i, q) * fe_face_values_pf.normal_vector(q) * pr_boundary_values[q] * fe_face_values_pf.JxW(q)); this is in the same way as step-20. I have now done further tests by just letting pr_boundary_values[q] to 1.0 in the blurb copied here - I do not get 1.0 as my value for the solution for pf on boundary with id 1, which I believe I should get. (i get something close but not 1.0. eg 0.96......) I am now wondering whether I am doubling up on the local_rhs in the assembling of cells on the boundary. To clarify, in step-20, f = 0 so that in the rhs_values, there is nothing added to the cells on the boundary other than that which is put in the boundary condition. for my problem, f is not equal to zero, so I am wondering whether I have extra contributions to the cells on the boundary and that's where the difference comes from. Thank you for your help. On Friday, February 22, 2019 at 10:51:11 PM UTC, Wolfgang Bangerth wrote: > > On 2/22/19 4:32 AM, [email protected] <javascript:> wrote: > > > > Wolfgang, so that's exactly what I had done with MMS, and that was > > verified, so i assumed the way i was imposing it was correct. > > When you say "correct", you mean that you have two uncoupled equations? > > > > It then > > fails to get the correct value at the top. Initially, it's correct up to > > about 4 decimal places, then to 2, then to the unit value. > > Try to be precise. When you say "It then fails...", what has changed? > Above you say that things are correct, and now they are no longer. What > changes have you made in between? And what is "then to 4,... then to 2" > -- you are suggesting that something changes form one step to the next, > but what is it? > > Best > W. > > -- > ------------------------------------------------------------------------ > Wolfgang Bangerth email: [email protected] > <javascript:> > www: http://www.math.colostate.edu/~bangerth/ > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
