I used UpwindConvenctionTerm, which seems to have solved the problem of the tail.
On Mon, Nov 25, 2013 at 2:32 PM, yuan wang <[email protected]> wrote: > Hi, > > > http://www.ctcms.nist.gov/fipy/examples/convection/generated/examples.convection.exponential1D.mesh1D.html > > I'm working around this simple example but found that there are some cases > that the results look weird. > > I made three changes to the original code: > > 1. convCoeff = (-10.) > > 2. change Right hand side boundary condition into: > var.faceGrad.constrain(0, mesh.facesRight) > > 3. change the Left Hand side boundary condition into: > valueLeft = 1. > > So, the physical meaning of the changes would be a zero gradient boundary > at RHS and fixed concentration boundary at LHS. Velocity direction is > positive towards right. From these changes, I am expecting a straight line > at 1 throughout the domain. However, the results have a very large tail > that extends way beyond 1. It looks like the zero gradient boundary is not > working for the right hand side. > > Is it a bug or is there any condition I missed to input for this kind of > boundary condition? I've attached the code. Thanks for your help. > > Thanks, > Rose >
_______________________________________________ fipy mailing list [email protected] http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
