[FLASH-USERS] Simulation domain (sedov)

[Klaus Weide]

On Mon, 2 Apr 2018, Ian Padilla wrote:

> I’m running the sedov problem where the default computational domain 
> is the unitary box (1,1,1). I’d like to expand the domain and look at 
> physical quantities at longer times. I tried naively to change the size 
> of the domain in the flash.par in 
> FLASH4.5/Source/Simulation/SimulationMain/Sedov, but the simulation 
> reaches the max SimTime really quickly skipping lots of calculations. 
> Has to do with Grid_initDomain?
> 
> The setup was 
> 
> $ ./setup Sedov -auto -2d -maxblocks=2000

Ian,

You don't show what you did to "expand the domain", so it's hard to tell 
what is going on. 

Maybe you have shifted the domain, and your explosion center isn't even in 
the domain any more.

>   lrefine          dx              dy
>       1             0.125           0.125
>       2             0.062           0.062
>       3             0.031           0.031
>       4             0.016           0.016
>       5             0.008           0.008
>       6             0.004           0.004

Are those resolutions at different refinement levels what you want?

>  MaterialProperties initialized
>  Cosmology initialized
>  Source terms initialized
>  sim_rhoAmbient is   1.0000000000000000     
>   iteration, no. not moved =            0           0
>  refined: total leaf blocks =            1
>  refined: total blocks =            1
>  [amr_morton_process]: Initializing surr_blks using standard orrery implementation

That output shows that you are running with only one block. This is the 
root block at the resolution of dx=0.125.

You need to do something so that the root block will be refined and more 
blocks created.

Maybe all it takes is an increase in lrefine_min or lrefine_min_init.

Klaus