Jens
The Robin boundary conditions are most useful for solving elliptic equations, i.e. for setting up initial conditions. For time evolution, most prefer radiative boundary conditions that let (approximately) outgoing waves leave the domain, while (approximately) no ingoing waves are generated.
One issue with the built-in boundary conditions (both Robin and radiative) is that they are only second order accurate. Most time evolution codes these days support 4th or even 8th order accuracy. We thus generally don't use the built-in boundary conditions any more, but rather implement them ourselves, together with the evolution system, using a compatible discretization method.
I do not have such an example that you ask for. However, you can call the Robin boundary condition routine yourself. There should be respective examples in the CactusWave/WaveToy* thorns. It might be that one of the ID* thorns (which set up initial conditions) even use the Robin boundary conditions.
I'll be happy to answer further questions if the pointers above do not help.
-erik
On Wed, May 23, 2018 at 6:23 AM, Jens Mahlmann jensmahl@alumni.uv.es wrote:
Dear all,
I am currently experimenting with the setup of Robin boundary conditions (I need a decay to the power of 3 in my physical field variables). However, I am finding myself a bit troubled with the correct understanding and implementation of the corresponding table handles.
Does anybody have a minimal example for the implementation of Robin boundary conditions with a non default decay power? Or a hint where I could improve my understanding of the implementation?
Thank you very much and my best from Valencia
Jens
Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users