[FLASH-USERS] Multiple Species Hydro+MHD Simulation

[Weston Lowrie]

Hi Klaus and Mordecai Mac,
Thanks again for the help.  I think I understand what will work with FLASH
now.  I will give some of these suggestions a try.

Best,
Wes

On Wed, Jun 22, 2016 at 10:03 AM, Klaus Weide <klaus at flash.uchicago.edu>
wrote:

> On Tue, 21 Jun 2016, Weston Lowrie wrote:
>
> > Hi Klaus,
> >
> > Thanks for the helpful response.  I think I understand the issues better
> > now.
> > For my option (1) it would probably involve creating a new Unit, but I
> > gather it may be pretty complicated to tie all the other existing
> > Units/physics such that they can work together?
>
> Yes.
>
> [This message was written before I saw Mordecai Mac Low's response
> pointing to the implementation of Duffin & Pudritz for ambipolar
> diffusion. I am not familiar with that implementation, the below applies to
> the unmodified MHD implementations of FLASH.]
>
> > Option (2) or the 3T mode seems like it would be the easiest path.  Do
> you
> > know how the MHD unit works with different species states?  For instance
> > would only the ionized species be affected by the magnetic field?  I
> > imagine the "bulk" density of all species is advanced in the momentum
> > equation, and therefore the field impacts the whole fluid?
>
> Yes to the latter. There is only one "fluid", with just one macroscopic
> velocity field, in the MHD approximation. And MHD does not care how the
> fluid is made up of ions of different charge states, all it cares about
> are macroscopic quantities, including the electron number density (ONE
> scalar field for the whole fluid). So in this approximation, there is no
> way to describe different macroscopic behavior of the ionized part of the
> fluid vs the part consisting of neutrals. What matters to MHD is the
> resulting electron density.
>
> All this is pretty much implied by the system of equations that FLASH's
> MHD solver purports to solve (as documented in the U.G. and other places).
>
> You should find it easy to just add some species to an existing test -
> come up with some reasonable initial condition, and implement it in
> Simulation_initBlock; use a Multispecies Eos - and visualize how this
> changes the behavior of the other quantities (or not).
>
> Klaus
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://flash.rochester.edu/pipermail/flash-users/attachments/20160628/92332384/attachment.htm>