Thanks a lot Johnathan for looking at it ... I just runned it, and it failed after 169 steps.
seems I am using the 2.2 version >>> print fipy.__version__ 2.2-dev I hope it is accurate that _CUSTOMMetalIonSourceVariable doesn't use the deposition rate. The boundary condition I am trying to implement is that the concentration is always c=0 at the interface. so the speed of the interface, appears only for the advection equation of phi, does that make sense ? On Mon, May 2, 2011 at 5:39 PM, Jonathan Guyer <[email protected]> wrote: > > > On May 2, 2011, at 9:03 AM, Julien Derr wrote: > > > Here is the script, if you have time to take a look, > > I ran it for over 130 steps and it didn't fail. It slows my machine down > too much to let it run longer. > > Does it really need to be such a big problem? Does it fail in a 100x100 > mesh? > > I realize it seems random, but how many steps does it typically take before > you get the recursion error? > > I don't know that it matters, but what version of FiPy are you running? > `python -c "import fipy; print fipy.__version__"` > > > > > note that I don't have this error if I remove the noise line 106 in the > deposition rate. > > That's interesting, although I don't know why it would matter. > > > > > #depositionrate > > #no noise: > > #depositionRateVariable = (abs(v0*c.getGrad().dot(phi.getGrad())) + > v0*c.getGrad().dot(phi.getGrad()) ) /2.0 + 1E-18 > > > > #with noise > > depositionRateVariable = abs(GaussianNoiseVariable(mesh=mesh, mean=1., > variance=0.02))*( (abs(v0*c.getGrad().dot(phi.getGrad())) + > v0*c.getGrad().dot(phi.getGrad()) ) /2.0 ) + 1E-18 > > Is it accurate that depositionRateVariable is not used in > CUSTOMbuildMetalIonDiffusionEquation, but is only used to set the value of > extensionVelocityVariable? _CUSTOMMetalIonSourceVariable doesn't seem to use > it. > > > > > >
