[FLASH-USERS] Collisional Ionization in FLASH 4

[Klaus Weide]

On Wed, 2 Dec 2015, Christopher Arran wrote:

> I was wondering if anyone had successfully managed to use the Ionize
> unit 
> recently (the documentation and code doesn't look to have been updated 
> since v2). If so, it would be great if you had any suggestions.

Hello Chris,

The Ionize unit has been updates enough to still work with FLASH 4.3
(we are running the NeiTest regular as a regression test), but only
with the split Hydro implementation ("split/PPM") and in a 1T (not 3T)
configuration.

Your setup commmand,

 "./setup -auto Ionization -2d +cylindrical +hdf5typeio -nxb=16 -nyb=16 +mtmmmt +uhd3t",

enters uncharted territory in at least two ways:

 * by using the split Hydro solver (the "uhd" part of "+uhd3t"),
    - actually, a multiTemp version thereof (that's the "3t" part), 
 * by using a multiTemp, i.e., 3T, Eos (implicit in "+mtmmmt").

(I think that using Ionize with just "+uhd" should work, but I have not 
tested this either.)
 
The Ionize unit on the one hand, and the multiTemp code on the other hand,
have different ideas of how to represent ionization levels: mass fraction 
of ionization level "species" for the former, separate ion/ele(/rad) 
components of internal energy (and of temperatures, and pressures) for the 
latter.

It may or may not be possible to combine these in a straightforward 
manner, but we have not tried this, and I assume at least some code 
changes will be needed (in the Ionize unit, and probably elsewhere).

HTH,
Klaus