[Nek5000-users] Sponge layer in Nek5000

[nek5000-users at lists.mcs.anl.gov]

Hi,
we have also implemented various sponges and other boundary conditions, 
including the Dong condition and convective conditions. The sponge we 
typically base on a "direct" forcing approach as e.g. described in the 
paper by Appelquist (JFM 2015); the implementation of that type of 
forcing is in userf.

The Dong condition is essentially a modified "o" condition (where one 
has to change a few lines here and there in the code, but I could send 
you our code), and the convective condition, which behaved the best (see 
e.g. the Master's thesis by Erik Boström) can be implemented in 
usercheck. The latter condition did create the least upstream disturbances.

Philipp

On 2018-03-01 20:24, nek5000-users at lists.mcs.anl.gov wrote:
> Hi Daniel,
> 
> We did implement the Sponge layer a couple of years back in nek5000 at 
> the outflow boundary condition. We used a Smagorinsky type of 
> eddy-viscosity closure where the Smagorinsky coefficient quadratically 
> varies in the x direction (where is the direction of dominant flow), 
> close to a few elements near the outflow. You can set it up with uservp 
> in nek5000. However, the stabilized natural outflow boundary conditions 
> by Dong et al. (2014) behaves much better than the sponge layer in my 
> opinion. For my type of problem the sponge layer actually creates 
> artificial diffusive structures close to the boundary, even when they 
> are applied very smoothly, and you need to chop off the near-outflow 
> region for data analysis.
> 
> On Thu, Mar 1, 2018 at 9:56 AM, <nek5000-users at lists.mcs.anl.gov 
> <mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
> 
>     Hi all,
> 
> 
>     I'm trying to implement outflow boundary conditions for a turbulent
>     plume using a sponge layer to absorb the energy of the flow and
>     prevent reflections back into the domain.  I was wondering whether
>     there is anything set up in Nek already to construct this sponge
>     layer or if it's something I need to do myself, and if so whether
>     anybody could help with something they've done before - I've read a
>     couple of other threads here that mention a sponge layer but not its
>     implementation.
> 
> 
>     I look forward to hearing from you.
> 
> 
>     Best,
> 
> 
>     Daniel Ward
> 
> 
>     _______________________________________________
>     Nek5000-users mailing list
>     Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov>
>     https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>     <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>
> 
> 
> 
> 
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>