[FLASH-USERS] opacity + nuclearBurn

[Klaus Weide]

On Mon, 6 Jan 2014, Amit Kashi wrote:

> I am using FLASH 4.0.1, with
> physics/sourceTerms/Burn/BurnMain/nuclearBurn/Aprox19.
> I would like to use species based opacity as well (together with nuclear
> burning).
> However, using opacity it seems as one must define species using the setup
> variable "species".
> This conflicts with the species defined in
> physics/sourceTerms/Burn/BurnMain/nuclearBurn/Aprox19.
> 
> I tried deleting the species defined in
> /source/physics/sourceTerms/Burn/BurnMain/nuclearBurn/Aprox19/config
> and re-defining them in the setup line
> (species=h1,he3,he4,c12,n14,o16,ne20,mg24,si28,s32,ar36,ca40,ti44,cr48,fe52,fe54,ni56,neut,prot).
> If this is relevant, I am using the Helmholtz EOS.

Hello Amit,

I have not tried this, but in principle, it should work.
It may be inelegant, but FLASH currently does not provide any more 
convenient way to combine the unit implementations you want. AFAIK, 
and given limited knowledge of what you need.

> However, I am not sure that the burning unit works well with these changes.

What makes you think that the Burn unit implementation is not working 
well?

To the Burn unit implementation, it should not matter whether your species
properties are defined in the old way (with SPECIES statements in
BurnMain/nuclearBurn/Aprox19/Config) or with the "species" setup variable.
As you have probably noticed, you just have to set lots of runtime 
parameters in your parfile when you use the "species" variable approach
(see Multispecies/MultispeciesMain/Config for what those parameters are).

If you ALSO have compiled in the Simulation_initSpecies.F90 from 
Simulation/SimulationComposition/Burn/, that would also try to initialize 
some of the properties from "SpeciesList.txt". Not sure which of the data 
sources (runtime parameters/parfile or SpeciesList.txt) would win out if 
there were conflicts, depends on order of initialization. In any case, the 
"Multifluid database" section of the FLASH log file should tell you which 
properties were effective.

> Also, it seems as that I have to define every species opacity with an
> ionmix file (isn't there a more simple way?).

If you want to use "species based opacity", then you do need to define the 
opacities for each species.  They don't all have to come from IONMIX 
files. In particular, you can choose an opacity type of "op_constant" or 
"op_constcm2g" for some, most, or all of your species.

> Does anyone know a better way to use opacity + nuclearBurn together?

Maybe somebody else does. The answer might depend on what you are using 
the opacities for.

Klaus