On Mon, Apr 2, 2012 at 10:05 PM, Daniel Farrell <[email protected]>wrote:
> Dear list, > > I know that Fipy for coupled simultaneous equations has not yet been > released, but does the trunk contain this code? Yes it does. It is all there along with some examples, which should help explain how to use it. > When it is released I > would like to attempt a two-component mass diffusion problem > (electrons and holes in semiconductors). > It is almost ready, but events keep conspiring against us. Mostly we are trying to document which seems never ending. > The reason I ask is because I noticed that the classes in fipy/terms/ > directory have a 'var' keyword argument which specifies the > CellVariable (a requirement for coupled systems). However, for the > problem that I have in mind I would also like to set different > boundary conditions for the different components of the system. > > That's a good question. We have switched to having constraints on variables rather than passing boundary conditions to equations for the reason you point out (and other reasons too). The old boundary conditions will still work though, but not for coupled equations. If you grep "constraint" recusively in examples, you should find a number of use cases. It should seem a lot less clunky. > For example: > LHS boundary; electrons = FixedValue, holes = FixedFlux. > RHS boundary; electrons = FixedFlux, holes = FixedValue. > > Will this be possible with the next release? I think so. Shouldn't be a problem. > I cannot find and way to > specify which CellVariable the boundary conditions corresponds to in > the fipy/boundaryConditions classes. > That's cos it's not there. Constraints belong to variables. Boundary conditions are deprecated (but are still available, obviously) . Hope that helps. -- Daniel Wheeler
_______________________________________________ fipy mailing list [email protected] http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
