Appearance
FV reflux on a capped last step
| Cases | Issues | Type | Backends | Figure | Regression |
|---|---|---|---|---|---|
rmf-amr-fv-pulse-timed | #38 | P | CPU, FNL | no | — |
Case and scope
On a time-driven run (it_max <= 0) PRISM caps its last step so that it lands on time_max. The Berger–Colella reflux correction must be scaled by that capped step, the one the update used. Before 4c1dd1a0 it was scaled by the forest's uncapped
With
What the case catches: a reflux correction scaled by any step other than the one the update used, as a jump of
Models and numerics
| Item | Choice | Code |
|---|---|---|
| Equations | Maxwell, SI, no source | models |
| Space | centred finite volumes of order 6 (fv_centered) | numerics |
| Time | SSP Runge–Kutta (5,4), CFL 0.5, time-driven (it_max = -1): time_max | numerics |
| Reflux | correction scaled by the step of the update | apply_reflux_corrections, adam_forest_object.F90; apply_reflux_to_stage_forest, adam_prism_cpu_object.F90 |
Initial and boundary conditions
| Item | Value |
|---|---|
| Domain | |
| Grid | 64 blocks of ngc = 3 |
| Initial condition | Gaussian pulse along sigma = 0.01, B0 = 10 |
| Boundaries | Neumann on all six faces |
| Steps | 4 full steps and one of half a step (time_max = 7.0361171362237545e-12 s); a checkpoint every step |
| Ranks | 2 (--np N) |
The rmf-amr-fd-pulse, 12σ on each side: with the narrower box the Gaussian tails leak through the
Input files
| File | Role |
|---|---|
src/tests/prism/regression/rmf-amr-fv-pulse-timed/input.ini | the case |
src/tests/prism/regression/rmf-amr-fv-pulse-timed/check.sh | runs it and integrates --np N; PRISM_EXE) |
bash
cd src/tests/prism/regression/rmf-amr-fv-pulse-timed && ./check.sh
PRISM_EXE=$PWD/../../../../../exe/adam_prism_fnl ./check.sh # FNLResults
| Leg | Asserts | Bound | CPU | FNL |
|---|---|---|---|---|
| capped step | the last time step shorter than 0.99 of the first | — | pass | pass |
| seams active | the forest: reflux max|F_coarse-F_fine_sum| line is printed | — | pass | pass |
| conservation | at every checkpoint, the drift of each |
| Step | 1 | 2 | 3 | 4 | 5 (capped) |
|---|---|---|---|---|---|
| fixed run | |||||
| fix reverted (negative control) | the same | the same | the same | the same |
Related pages: static AMR and the FV reflux register for the register itself; FLUME's conservation across AMR seams measures the same property on the Euler equations.
Findings
- With the correction scaled by the capped step, the drift grows smoothly to
: the seams' disturbance reaching the boundaries, not a reflux defect. - With the fix reverted the drift jumps to
on the capped step, 6500 times larger. The bound, , sits 45 times above the fixed run and 140 times below the reverted one.