[FLASH-USERS] questions about boundary conditions

[Eddie Hansen]

In general, here is how periodic boundaries work...

Just consider the x-direction and say you are using xl_boundary_type =
"periodic" and xr_boundary_type = "periodic". Let's number the cells
starting at n and going up to m, so we have n, n+1, n+2, ... , m-1, m. The
"guard cells" are those cells outside the domain to the left of n and to
the right of m. With periodic conditions, the values of guard cell n-1 are
set to the values of the interior cell m. Likewise, values of guard cell
m+1 are set to values of interior cell n. Depending on the solver order,
more guard cells will be filled accordingly. For example, the values of
guard cell n-2 are set to values of interior cell m-1, and so on.

This effectively makes any flux on the right boundary equal to the flux on
the left boundary. In other words, if material leaves the grid on the
right, it will enter on the left, and vice-versa. You can think of it as
"wrapping around".

Hopefully that answers your question about periodic boundary conditions. To
do something like keep a constant magnetic field value at a boundary, you
have to implement a user-defined boundary like Ryan mentioned.


On Tue, Oct 29, 2019 at 3:13 PM Ryan Farber <rjfarber at umich.edu> wrote:

> Hi Guido,
>
> I'm not sure how periodic boundary conditions work, but to respond to your
> last paragraph on static boundary conditions:
> you can set xl_boundary_type (etc.) in flash.par to "user" and
> you can modify Grid_bcApplyToRegionSpecialized.F90 "case(USER_DEFINED)" to
> set variables to a constant value.
>
> Best,
> --------
> Ryan
>
>
> On Tue, Oct 29, 2019 at 11:59 AM guido <g.granda at irya.unam.mx> wrote:
>
>> Hi Flash users,
>>
>> I've been trying to find out how boundary cells properties are filled
>> when periodic boundary conditions are used. Do you know if these cells'
>> properties like
>>
>> magnetic field, velocity, etc, are filled when periodic boundary
>> conditions are used. Let's say our initial condition has uniform
>> magnetic field or uniform density,
>>
>> does boundary cells have the same value of these properties or these are
>> set to zero or sth else ?
>>
>> Could you please give me a hint of how to set static boundary conditions
>> for a simulation, e.g. a constant magnetic field at the computational
>> volume boundary edges.
>>
>> Cheers,
>>
>>
>>

-- 
Eddie Hansen, PhD
Postdoctoral Scholar
University of Chicago
607-341-6126 | Flash Center
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20191029/62994bb1/attachment.htm>