 |
|
|
|
|
|
|
|
 |
|
||||||
 |
 |
 |
 |
 |
 |
 |
|
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
 |
 |
 |
|
||||
This test demonstrates the successful implementation of a multigrid-scheme solver for elliptic equations. Here, the Rayleigh-Taylor simulation is modified to include the effects of thermal diffusion/heat conduction. The case modeled below is constant-coefficient diffusion Cd. Work on nonlinear coefficient, e.g. dependence on temperature (Cd=Cd,0Tγ) is ongoing.
Initial setup is identical to the ordinary RT simulation, linked above.
The left and right boundaries are periodic, and the bottom boundary is reflecting. The top boundary is extrapolating.
AMR calculation has modified refinement criteria to resolve entire region around instability, to avoid instability seeding. 3 levels of AMR (4 levels total) with base resolution of 32x128, yielding maximum effective resolution of 256x1024.
Given below are movies with and without diffusion to demonstrate its effects.
| Diffusion | No Diffusion |
|---|---|
 |
 |