On Jan 18, 2013, at 12:01 PM, John Assael wrote: > I have a couple more questions: > 1) The problem is a quarter symmetry and I am solving the problem in the > first Quartile how can I specify the symmetry of the experiment?
I'd use a CylindricalGrid1D mesh for the problem you've shown us so far. If you really need a quarter round, then you could do that with Gmsh. See http://www.ctcms.nist.gov/fipy/examples/diffusion/generated/examples.diffusion.circle.html for a starting point. > As if I take the dT on (0,r) it is different than (r/math.sqrt(2), > r/math.sqrt(2)) although it should be as both points are on the radius. Because of the square domain your problem doesn't have cylindrical symmetry, although I'd expect that effect to be small. Because of the square cells, the value of center of the cell nearest to (0, r) and the center of the cell nearest to (r/math.sqrt(2), r/math.sqrt(2)) are probably not at the same radius from the center. You could try dt(((0,), (r,)), order=1) to get the value interpolated to the point you requested rather than at the nearest cell center. > 2) Finally, is this the right way to get the dT (cell-variable) value inside > each step? dT(((0,), (r,))) Yes, that should work. _______________________________________________ fipy mailing list fipy@nist.gov http://www.ctcms.nist.gov/fipy [ NIST internal ONLY: https://email.nist.gov/mailman/listinfo/fipy ]
