User Boundary conditions for vector potential. More...
#include "decs.h"
Go to the source code of this file.
Macros | |
#define | EXTRAP 0 |
Variables | |
int | inboundloop [NDIM] |
int | outboundloop [NDIM] |
int | innormalloop [NDIM] |
int | outnormalloop [NDIM] |
int | inoutlohi [NUMUPDOWN][NUMUPDOWN][NDIM] |
int | riin |
int | riout |
int | rjin |
int | rjout |
int | rkin |
int | rkout |
int | dosetbc [COMPDIM *2] |
User Boundary conditions for vector potential.
For fluxes, e.g. F1, assume fluxes exist everywhere – including j/k boundary zones. Only i-boundary zones need to be bounded. This assumesCOMPZSLOOP(is,ie,js,je,ks,ke) is over boundary zones in flux.c, which in general to be compatible with any flux method (including finite volume) this is how it should be.
With fluxes, only need to bound each dir-flux along that direction (as presently used by ENO-type schemes)
Assume flux at 0 through N are computed correctly. So only need fluxes in other boundary zones. Self-assigns for 0 or N for simplicity of coding
OUTFLOW leaves true edge of boundary unchanged Therefore, if FIXEDOUTFLOW, then extrapolation is always ok. if OUTFLOW, then extrapolation is ok as long as flux is from active zones out of boundary
Definition in file boundsvpot.c.
#define EXTRAP 0 |
Definition at line 24 of file boundsvpot.c.
int dosetbc[COMPDIM *2] |
Definition at line 34 of file boundsvpot.c.
int inboundloop[NDIM] |
Definition at line 28 of file boundsvpot.c.
int innormalloop[NDIM] |
Definition at line 30 of file boundsvpot.c.
Definition at line 32 of file boundsvpot.c.
int outboundloop[NDIM] |
Definition at line 29 of file boundsvpot.c.
int outnormalloop[NDIM] |
Definition at line 31 of file boundsvpot.c.
int riin |
Definition at line 33 of file boundsvpot.c.
int riout |
Definition at line 33 of file boundsvpot.c.
int rjin |
Definition at line 33 of file boundsvpot.c.
int rjout |
Definition at line 33 of file boundsvpot.c.
int rkin |
Definition at line 33 of file boundsvpot.c.
int rkout |
Definition at line 33 of file boundsvpot.c.