Re: [Libmesh-users] Poisson w/ Penalty Dirichlet Example

[Caleb M P. <cal...@ut...>]

John,

I wasn't too sure how to accomplish that, but here is the main file. This
is a module in a larger code, but the function I'm looking at is
"assemble_pressure".
https://github.com/CalebPhillips5/ABM_Ang/blob/master/pressure_ABM . In
this instance I'm just trying to set the top boundary to 1 and the bottom
boundary to 0 with 2nd order elements.

Cheers,
Caleb

On Wed, Jul 8, 2020 at 5:10 PM John Peterson <jwp...@gm...> wrote:

>
>
> On Wed, Jul 8, 2020 at 4:48 PM Caleb M Phillips <cal...@ut...>
> wrote:
>
>> Hey all,
>>
>> Quick question that is completely throwing me for a loop. I'm working on a
>> poisson equation for pressure and wanting to specify dirichlet boundary
>> conditions. I used the penalty method (per libmesh documentation), and it
>> works beautifully, aka matches the real solution. However, when I change
>> to
>> 2nd order polynomials, everything falls apart. The penalty method does
>> implement the boundary conditions, but the solution is complete garbage.
>>
>> Any explanation for changing (.., FIRST) to (.., SECOND) and the solution
>> (the code runs fine) but the solution being garbage?
>>
>
>
> I just ran introduction_ex3, which uses penalty BCs and second-order
> elements, and wrote the result to an Exodus file, and it looks OK to me.
>
> So my guess is it's a bug of some sort in your code. Is it possible for
> you to share it somehow?
>
> --
> John
>
>