Hallo Ardavan, than means that in the new meep 0.2 I can directly specify what you call round trip reflection? That would be nice!
Andreas Ardavan F. Oskooi wrote: > Alexander Sprafke wrote: >> Lets say I have a pml layer which is too thin. Do these reflections >> come from propagating waves going through the pml layer and being >> reflected at the end of the computational cell, going through the pml >> layer again and entering the region of interest. Or is it the >> too-thin-pml layer itself? > There are two sources of reflection from the PML: 1) the non absorbed > component of the wave entering, reflecting at the hard wall boundary, > and returning back through the PML and into the computational cell > (which you have already described), what I will call the 'round-trip' > reflection and 2) the 'transition' reflection occurring at the interface > of the non-PML & PML regions. The round-trip reflection can be made > arbitrarily small (in uniform media) by making the PML absorption large, > however this is not the case in the current release of meep (version > 0.10.1) which implements a split field PML so you will have to make sure > that either the PML is sufficiently long or to increase the PML strength > parameter (for those interested, see the structure_chunk::use_pml > function in structure.cpp of the source files). You will need to do > these tests to make sure that PML reflections are sufficiently small so > as not to affect your computations. The transition reflection, on the > other hand, is a consequence of the discretization of space (a PML is > perfectly reflectionless in the exact wave equation) and is determined > by the degree of smoothness of the PML conductivity absorber (we have an > upcoming paper on this) which is typically a quadratic polynomial. In > your case of PML overlapping uniform media, you should be more way of > the 'round-trip' reflection. > > We will be releasing meep 0.20 very shortly which has a much improved > uniaxial PML (uPML) implementation with the ability to fix the round > trip reflection along with user-defined PML conductivity. > > Ardavan _______________________________________________ meep-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss
