Re: [FLASH-USERS] advection (fwd)

Dear Erik-Jan,

Flash2.3 cannot advect variables other that chemical or nuclear species,
e.g. mass fractions. The attributes in Config files (ADVECT, CONSERVED,
etc.) are not used and it is not clear that we will pursue that mechanism
for describing variables added.

However, in FLASH2.4, we have added support for any number of "mass
scalars", passive scalars which are described by

d(rho phi)/dt + del.(rho v phi) = 0

phi is unlike the mass fractions we now support in that it has no role in
the composition, thermodynamics (equation of state), etc. (If you would
like source terms, you can add modules for them.) We hope to release
Flash2.4 in March.

Natasha

On Tue, 3 Feb 2004, Erik-Jan Rijkhorst wrote:

> Dear Flash users,
>
> We would like to use Flash2.3 for a setup where we need to advect a number
> of physical variables which cannot be simply identified with chemical or
> nuclear species. First, we tried adding an extra line in the Config file
> in a copy of the setups/sod directory:
>
> VARIABLE test ADVECT NORENORM NOCONSERVE
>
> and added some additional lines to the init_block.F90 to initialize this
> variable. But it seems that the ppm module doesn't do anything with the
> new variables of the type ADVECT (the initialized values of 'test' just
> remain constant for all times). So this was probably to simple an
> approach?
>
> Then we used the materials/composition/fuel+ash as a template. However,
> our quantities don't have a mass, charge nor binding energy so we don't
> really need all that.
>
> My questions are:
> - Is there a simple way to add VARIABLEs to Flash that are advected
> besides through the materials/composition module?
> - Where is the code located that does the actual advection
> a) for the variables
> b) for the fuel+ash species?
>
> thanks!
>
> Erik-Jan Rijkhorst
>
Received on Wed Feb 4 18:22:33 2004