Index: femsolvers.sch
===================================================================
--- femsolvers.sch	(nonexistent)
+++ femsolvers.sch	(revision 6816)
@@ -0,0 +1,190 @@
+v 20130925 2
+C 40000 40000 0 0 0 title-C.sym
+V 34200 50100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 67300 44100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 69200 46700 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 34200 46200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 58700 51400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 34200 45000 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 34200 48600 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 34200 47400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 58700 52600 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 69200 47900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 48000 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 46900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 45800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 44700 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 43600 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 49100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 50200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 42500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 40300 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 39200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 41400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 38100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 67300 43000 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 37000 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 67300 41900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 68300 38700 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 35900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49700 34800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 67300 40800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 67400 36800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 72000 35500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 72000 34400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+T 34200 50100 9 10 1 0 0 4 1
+librecad
+T 67300 44100 9 10 1 0 0 4 1
+openfoam
+T 69200 46700 9 10 1 0 0 4 1
+paraview
+T 34200 46200 9 10 1 0 0 4 1
+gmsh
+T 58700 51400 9 10 1 0 0 4 1
+salome
+T 34200 45000 9 10 1 0 0 4 1
+freecad
+T 34200 48600 9 10 1 0 0 4 1
+brlcad
+T 33800 49200 9 12 1 0 0 0 1
+3D CAD
+T 34200 47400 9 10 1 0 0 4 2
+open
+scad
+T 33800 50700 9 12 1 0 0 0 1
+2D CAD
+T 58700 52600 9 10 1 0 0 4 1
+onelab
+T 69200 47900 9 10 1 0 0 4 1
+opendx
+T 49700 48000 9 10 1 0 0 4 1
+maxfem
+T 49700 46900 9 10 1 0 0 4 1
+openems
+T 49700 45800 9 10 1 0 0 4 1
+meep
+T 49700 44700 9 10 1 0 0 4 1
+angora
+T 49700 43600 9 10 1 0 0 4 1
+EMTL
+T 49700 49100 9 10 1 0 0 4 1
+fastmaxwell
+T 49700 50200 9 10 1 0 0 4 1
+nec2
+T 49700 42500 9 10 1 0 0 4 1
+getdp
+T 68500 48500 9 12 1 0 0 0 1
+3D Graphing
+T 66900 44700 9 12 1 0 0 0 1
+3D CFD
+T 49300 50800 9 12 1 0 0 0 2
+3D FEM
+w/E&M Ability
+T 49700 40300 9 10 1 0 0 4 1
+FEniCS
+T 49700 39200 9 10 1 0 0 4 1
+dealii
+T 49700 41400 9 10 1 0 0 4 1
+freedrake
+T 50400 46800 9 12 1 0 0 0 2
+openems uses it's own format in matlab/ocatve script to process everything including geometry
+it's output includes but is not limited to paraview and octave's native grapher + ?gnuplot?
+T 50400 49100 9 12 1 0 0 0 1
+fastmaxwell might be abandonware. the project looks stalled post 2008. no info on mesh or geometry input is obivous on the site.
+T 50400 48000 9 12 1 0 0 0 1
+maxfem is more for FEM mechanical issues like mechancial stresses from magnetics than electrical/rf work
+T 50400 50200 9 12 1 0 0 0 1
+nec2 has frequency limitations and is meant mostly for the design of antennas. sadly versions after 2 are proprietary. this is the only solver i have seen that uses method of moments.
+T 50500 41400 9 12 1 0 0 0 1
+firedrake is cool because it has gpu processing. i question their meshing. as yet i see no em work done in it.
+T 50500 39200 9 12 1 0 0 0 1
+dealII is cool but i have trouble understanding it's documentation (only lgpl)
+T 50500 40200 9 12 1 0 0 0 2
+FEniCS is another tool that seems to be mostly for mechanical people. However I like their example work on electromagnetics. 
+The one thing I dislike about their work is that the current front end is python.
+T 50400 45700 9 12 1 0 0 0 1
+meep uses scheme. I feel that says everything. it is actively developed. one positive for it is that it supports hdf5 for output. 
+T 50400 43300 9 12 1 0 0 0 2
+EMTL is hard to google for some reason http://fdtd.kintechlab.com/en/start is the url
+it would be good for modeling semiconductors but they don't really talk about anyting rf or pcb related in the documentation.
+T 50400 44500 9 12 1 0 0 0 2
+angora is also hard to google to get results look for angoraftdt which should yeald http://www.angorafdtd.org/
+i don't like their documentation. it would appear to have been developed over the course of 1 year 2012-2013 and then abandoned.
+T 50400 42400 9 12 1 0 0 0 1
+getdp is run via onelab using gmsh to do the meshing. i am not wild about onelab which is way too gui for me. we want to be able to drive everything from make.
+T 49700 38100 9 10 1 0 0 4 1
+calculix
+T 50500 38100 9 12 1 0 0 0 1
+calculix looks like a cfd/fem that someone wrote to do aeronautical work and then later an example of electromagnetics was run on it for show.
+T 69900 47900 9 12 1 0 0 0 1
+opendx has it's fans and is still in use but it has been with out a new release in like 10 years.
+T 69900 46700 9 12 1 0 0 0 1
+paraview has an excessive number of dependencies but just about everyone uses it. it is maintaned and well documented.
+T 67300 43000 9 10 1 0 0 4 1
+code saturne
+T 57900 53300 9 12 1 0 0 0 1
+FEM front ends
+T 49700 37000 9 10 1 0 0 4 1
+feel++
+T 50500 37000 9 10 1 0 0 0 1
+feel++ (http://www.feelpp.org/) has a lot of current development and has some examples in the E&M domain but they are not really over lapping with what I think our users need.
+T 67300 41900 9 10 1 0 0 4 1
+gerris
+T 68000 41800 9 10 1 0 0 0 1
+Gerris Flow Solver (http://gfs.sourceforge.net/wiki/index.php/Main_Page) 
+T 68000 43000 9 10 1 0 0 0 1
+Code Saturne (http://code-saturne.org/cms/)
+T 68300 38700 9 10 1 0 0 4 1
+code aster
+T 50500 35800 9 10 1 0 0 0 1
+S4 (https://web.stanford.edu/group/fan/S4/index.html) 
+T 50500 34800 9 10 1 0 0 0 1
+Elmer (https://csc.fi/web/elmer/elmer) (http://www.elmerfem.org/) this is able to connect to salome which is nice and lists E&M as a feature but I could not find a single example of it in actual use on E&M via their wiki or examples page.
+T 68900 38600 9 10 1 0 0 0 1
+Code Aster (http://code-aster.org/spip.php?rubrique2) This hooks into salome as well. For the mechanical folks it handles mechanical and thermal stuff. I know people who use it with Salome and love it. Honestly it is all over my head.
+T 68300 39400 9 10 1 0 0 4 2
+3D FEM
+Mechanical
+T 59400 52500 9 10 1 0 0 0 1
+One Lab (http://onelab.info/wiki/Main_Page)
+T 34800 46200 9 10 1 0 0 0 1
+gmsh (http://gmsh.info/) is a very well respected mesh generation program for FEM that has a 3D CAD built into it. For some formats it falls back on opencascade.
+T 49700 35900 9 10 1 0 0 4 1
+s4
+T 49700 34800 9 10 1 0 0 4 1
+elmer
+T 68000 44100 9 10 1 0 0 0 1
+openfoam is impressive and is aiming to be like comsol interms of it's universality. it is very well documented but none of the people i know in the modeling field have ever used it or even know anyone who has used it.
+T 59400 51400 9 12 1 0 0 0 1
+i know a ton of people who use this and swear by it. the trouble is that all of the more appealing stuff in E&M is ftdt which this has yet to connect to as far as i can tell.
+T 34800 50000 9 12 1 0 0 0 1
+librecad was originally qcad from QT. i liked qcad which had a very short learning curve. i still haven't gotten the hang of librecad for some reason.
+T 34800 44800 9 12 1 0 0 0 2
+freecad is reported by some friends to be best in terms of ease of use compared to openscad and freecad.
+I resist it because it uses opencascade which is only lgpl.
+T 34800 47400 9 12 1 0 0 0 1
+openscad is favored by Igor2. It's 3D kernel is opencsg which is distributed seperately. 
+T 34800 48300 9 12 1 0 0 0 3
+brlcad is the old and well respected. It is also painful to use a problem they are trying to address with a new gui called archer. 
+brlcad supports more cad formats than any other open source program i could find. It has it's own 3D kernel.
+they are so well respected openscad, linuxcnc and a few others have their google summer of code projects over seen by brlcad.
+T 67300 40800 9 10 1 0 0 4 1
+su2
+T 68000 40800 9 12 1 0 0 0 1
+su2 http://su2.stanford.edu/
+T 67400 36800 9 10 1 0 0 4 1
+openfcst
+T 68100 36800 9 12 1 0 0 0 2
+OpenFCST (http://www.openfcst.mece.ualberta.ca/index.html) open source fuel cell simulation tool box
+Uses Deal.II, salome, and dakota
+T 67400 37600 9 10 1 0 0 4 2
+3D FEM
+Left Field
+T 72700 35500 9 12 1 0 0 0 1
+dakota (https://dakota.sandia.gov/) ???
+T 72000 35500 9 10 1 0 0 4 1
+dakota
+T 72000 36200 9 10 1 0 0 4 1
+Associated Libraries of Interest
+T 72000 34400 9 10 1 0 0 4 1
+opencascade
Index: femworkflows.sch
===================================================================
--- femworkflows.sch	(nonexistent)
+++ femworkflows.sch	(revision 6816)
@@ -0,0 +1,382 @@
+v 20140308 2
+C 40000 40000 0 0 0 title-C.sym
+V 42700 49400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 43100 49700 43800 50400 3 0 0 0 -1 -1
+L 42700 48900 42700 47900 3 0 0 0 -1 -1
+V 42700 49400 1500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 47700 50400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 47700 47300 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49900 53400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49900 51000 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 47700 54000 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 50000 45500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 49900 52200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 47700 51600 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 42700 46900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 42300 49700 41600 50400 3 0 0 0 -1 -1
+L 43800 48400 43100 49100 3 0 0 0 -1 -1
+L 42300 49100 41600 48400 3 0 0 0 -1 -1
+V 41400 47400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 44000 47400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 46000 45500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 41200 49400 42200 49400 3 0 0 0 -1 -1
+V 47700 48500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 47700 52800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 46300 48500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 44500 47400 47200 47400 3 0 1 0 -1 -1
+L 44200 49400 45800 48500 3 0 1 0 -1 -1
+L 46800 48500 47200 48500 3 0 1 0 -1 -1
+L 44200 49400 44200 47900 3 0 1 0 -1 -1
+V 42700 52400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 42700 53900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 43200 52400 45800 48500 5 0 1 0 -1 -1
+L 43200 53900 45800 48500 5 0 1 0 -1 -1
+V 50000 44200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 44200 49400 47200 50400 3 0 1 0 -1 -1
+L 44200 49400 47200 51600 3 0 1 0 -1 -1
+L 44200 49400 47200 52800 3 0 1 0 -1 -1
+L 44200 49400 47200 54000 3 0 1 0 -1 -1
+L 43200 53900 47200 50400 3 0 1 0 -1 -1
+L 43200 53900 47200 51600 3 0 1 0 -1 -1
+L 43200 53900 47200 52800 3 0 1 0 -1 -1
+L 43200 53900 47200 54000 3 0 1 0 -1 -1
+L 48200 52800 49400 52200 3 0 1 0 -1 -1
+L 48200 52800 49400 53400 3 0 1 0 -1 -1
+L 48200 51600 49400 52200 3 0 1 0 -1 -1
+L 48200 51600 49400 53400 3 0 1 0 -1 -1
+L 48200 50400 49400 51000 3 0 1 0 -1 -1
+V 49900 54600 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 48200 54000 49400 53400 3 0 1 0 -1 -1
+L 48200 54000 49400 54600 3 0 1 0 -1 -1
+V 49900 49500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 48200 51600 49400 49500 3 0 1 0 -1 -1
+L 48200 48500 49400 49500 3 0 1 0 -1 -1
+L 48200 47300 49400 49500 3 0 1 0 -1 -1
+L 48200 48500 49500 45500 3 0 1 0 -1 -1
+L 49500 45500 48200 47300 3 0 1 0 -1 -1
+V 52200 46100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 54400 49900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 56000 49500 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 56000 47300 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 52200 50200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 56000 45100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 56000 42700 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 56000 46200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 56000 48400 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 54900 49900 55500 49500 3 0 1 0 -1 -1
+L 54900 49900 55500 48400 3 0 1 0 -1 -1
+L 54900 49900 55500 47300 3 0 1 0 -1 -1
+L 52700 46100 55500 48400 3 0 1 0 -1 -1
+L 52700 46100 55500 47300 3 0 1 0 -1 -1
+L 52700 46100 55500 46200 3 0 1 0 -1 -1
+L 52700 46100 55500 45100 3 0 1 0 -1 -1
+V 58400 47200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 54900 49900 55400 50300 3 0 1 0 -1 -1
+L 55400 50300 56600 50300 3 0 1 0 -1 -1
+L 56600 50300 57900 47200 3 0 1 0 -1 -1
+L 52700 46100 55500 44500 3 0 1 0 -1 -1
+L 55500 44500 56500 44500 3 0 1 0 -1 -1
+L 56500 44500 57900 47200 3 0 1 0 -1 -1
+L 56500 45100 57900 47200 3 0 1 0 -1 -1
+L 56500 46200 57900 47200 3 0 1 0 -1 -1
+L 56500 47300 57900 47200 3 0 1 0 -1 -1
+L 56500 48400 57900 47200 3 0 1 0 -1 -1
+L 56500 49500 57900 47200 3 0 1 0 -1 -1
+L 56500 43800 57900 47200 3 0 1 0 -1 -1
+V 52200 48800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 52200 47200 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 50000 42900 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 48200 48500 49500 44200 4 0 1 0 -1 -1
+L 48200 47300 49500 44200 4 0 1 0 -1 -1
+L 48200 47300 49500 42900 4 0 1 0 -1 -1
+L 48200 48400 49500 42900 4 0 1 0 -1 -1
+L 52700 48800 53900 49900 4 0 1 0 -1 -1
+L 52700 47200 53900 49900 4 0 1 0 -1 -1
+L 48200 48500 51700 48800 5 0 1 0 -1 -1
+L 48200 48500 51700 47200 5 0 1 0 -1 -1
+L 48200 48500 55500 42700 5 0 1 0 -1 -1
+V 52200 44100 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 52700 44100 55500 42700 3 0 1 0 -1 -1
+V 56000 43800 500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 52700 47200 55500 43800 4 0 1 0 -1 -1
+L 48200 48500 51700 46100 8 0 1 0 -1 -1
+L 56500 42700 57900 47200 3 0 0 0 -1 -1
+T 42700 49400 9 10 1 0 0 4 1
+libgeda
+T 42700 50400 9 10 1 0 0 4 1
+gschem
+T 41700 49100 9 10 1 0 0 4 1
+gattrib
+T 43700 49400 9 10 1 0 0 4 1
+gnetlist
+T 50000 45500 9 10 1 0 0 4 1
+gerbv
+T 47700 47300 9 10 1 0 0 4 1
+pcb
+T 47700 50400 9 10 1 0 0 4 1
+iverilog
+T 47700 51600 9 10 1 0 0 4 1
+ng-spice
+T 49900 52200 9 10 1 0 0 4 1
+gtkwave
+T 42700 46900 9 10 1 0 0 4 1
+tragesym
+T 42200 48400 9 10 1 0 0 4 1
+gsymcheck
+T 49900 53400 9 10 1 0 0 4 1
+gwave
+T 49900 51000 9 10 1 0 0 4 1
+dinotrace
+T 47700 54000 9 10 1 0 0 4 1
+xyce
+T 42700 51100 9 10 1 0 0 4 1
+geda
+T 44000 47400 9 10 1 0 0 4 1
+gsch2pcb
+T 41700 49700 9 10 1 0 0 4 1
+gschlas
+T 41400 47400 9 10 1 0 0 4 1
+sarlacc
+T 41200 45900 9 10 1 0 0 2 3
+* sarlacc and tragesym use libgeda 
+but only for colors?! They do not 
+apear to parse symbols via libgeda.
+T 46000 45500 9 10 1 0 0 4 1
+make
+T 47700 48500 9 10 1 0 0 4 1
+pcb-rnd
+T 43100 48400 9 10 1 0 0 4 1
+gaf
+T 47700 52800 9 10 1 0 0 4 1
+gnucap
+T 46300 48600 9 10 1 0 0 4 1
+gsch2-
+T 46300 48400 9 10 1 0 0 4 1
+pcb-rnd
+T 42700 52400 9 10 1 0 0 4 1
+ltspice
+T 42700 53900 9 10 1 0 0 4 1
+qucs
+T 47100 49200 9 12 1 0 0 0 1
+PCB Layout
+T 41800 54600 9 12 1 0 0 0 1
+Schematic Capture
+T 46800 54700 9 12 1 0 0 0 1
+Circuit Simulation
+T 50000 44200 9 10 1 0 0 4 1
+linuxcnc
+T 49900 54600 9 10 1 0 0 4 1
+gnuplot
+T 49900 49500 9 10 1 0 0 4 1
+cray lc
+T 49300 55200 9 12 1 0 0 0 1
+2D Graphing
+T 49500 49300 9 12 1 0 0 0 1
+(closed)
+T 42300 52200 9 12 1 0 0 0 1
+(closed)
+T 49500 50000 9 12 1 0 0 0 2
+3D FEM
+parasitics
+T 52200 46100 9 10 1 0 0 4 1
+salome
+T 54400 49900 9 10 1 0 0 4 1
+onelab
+T 53700 50600 9 12 1 0 0 0 1
+FEM front end
+T 56000 49500 9 10 1 0 0 4 1
+getdp
+T 56000 47300 9 10 1 0 0 4 1
+elmer
+T 52200 50200 9 10 1 0 0 4 1
+gmsh
+T 56000 45100 9 10 1 0 0 4 1
+code aster
+T 56000 42700 9 10 1 0 0 4 1
+openems
+T 56000 46200 9 10 1 0 0 4 1
+code saturne
+T 56000 48400 9 10 1 0 0 4 1
+openfoam
+T 51600 50800 9 12 1 0 0 0 1
+Mesh + CAD
+T 55500 50100 9 12 1 0 0 0 1
+3D solvers
+T 58400 47200 9 10 1 0 0 4 1
+paraview
+T 49500 46100 9 12 1 0 0 0 1
+gerber view
+T 49500 44800 9 12 1 0 0 0 1
+CNC Mill
+T 52200 48800 9 10 1 0 0 4 1
+brlcad
+T 51800 49400 9 12 1 0 0 0 1
+3D CAD
+T 52200 47200 9 10 1 0 0 4 2
+open
+scad
+T 50000 42900 9 10 1 0 0 4 1
+openpnp
+T 49500 43500 9 12 1 0 0 0 1
+CNC Stuff
+T 41400 42000 9 12 1 0 0 0 8
+Key
+Green - Workflows known to operate
+Blue - Workflows believe to work based on format compatibility
+Yellow - Workflows we would are considering in the near term.
+Red - Workflows someone should do in the extreme far future. 
+Probably by/as other projects.
+(state) - status of a project
+[parent] - who is this a part of
+T 52200 44100 9 10 1 0 0 4 2
+csxcad
+[openems]
+T 57700 47900 9 12 1 0 0 0 1
+3D Graphing
+T 56000 43800 9 10 1 0 0 4 1
+FEniCS
+T 50800 54900 9 10 1 0 0 0 8
+Note: For E&M solving you really need direct route from 
+pcb-rnd to the solver or atleast a front end for it. Meta-
+data is lost when you transition through a 3D CAD about 
+where connections are and etc that is critical to E&M. 
+The flows via 3D CAD are documented here so that they 
+can be visualized for people who might want to look at 
+heat/air flow over populated boards and/or vibrational 
+modes of complete assemblies.
+T 49300 40500 9 10 1 0 0 0 6
+Note: I am not proposing we actually build all or any of this 
+[block of 3D CAD&FEM stuff] past the part I have already 
+gotten involved with. I just wanted everyone to understand 
+how the workflows could be if one of these tools jumped 
+out as a "must have" to them or if one of their users 
+jumped at the chance to bridge the gap.
+L 52700 50200 53900 49900 3 0 0 0 -1 -1
+L 52700 50200 55500 48400 3 0 0 0 -1 -1
+L 52700 50200 55500 43800 3 0 0 0 -1 -1
+T 45400 46100 9 12 1 0 0 0 1
+Automation
+T 44000 47200 9 10 1 0 0 4 1
+[geda-gaf]
+T 42700 46700 9 10 1 0 0 4 1
+[geda-gaf]
+T 41400 47200 9 10 1 0 0 4 1
+[geda-gaf]
+T 46300 48200 9 10 1 0 0 4 1
+[pcb-rnd]
+L 52700 50200 55500 47300 3 0 0 0 -1 -1
+L 52700 50200 55500 46200 3 0 0 0 -1 -1
+L 52700 50200 55600 45400 3 0 0 0 -1 -1
+L 44200 47900 44200 48000 3 0 0 0 -1 -1
+L 51500 50000 51700 50200 4 0 1 0 -1 -1
+L 51500 48600 51700 48800 4 0 1 0 -1 -1
+L 51500 49000 51500 50000 4 0 1 0 -1 -1
+L 51500 49000 51700 48800 4 0 1 0 -1 -1
+L 51500 47400 51500 48600 4 0 1 0 -1 -1
+L 51500 47400 51700 47200 4 0 1 0 -1 -1
+L 51000 49800 51700 50200 4 0 1 0 -1 -1
+L 51000 44500 51000 49800 4 0 1 0 -1 -1
+L 51000 44500 51700 44100 4 0 1 0 -1 -1
+L 51700 47200 51300 46900 4 0 1 0 -1 -1
+L 51300 44600 51300 46900 4 0 1 0 -1 -1
+L 51300 44600 51700 44100 4 0 1 0 -1 -1
+L 51300 49900 51700 50200 4 0 1 0 -1 -1
+L 51300 47500 51700 47200 4 0 1 0 -1 -1
+L 51300 47500 51300 49900 4 0 1 0 -1 -1
+L 51200 48500 51700 48800 4 0 1 0 -1 -1
+L 51200 44600 51200 48500 4 0 1 0 -1 -1
+L 51700 44100 51200 44600 4 0 1 0 -1 -1
+T 56000 53200 9 10 1 0 0 0 16
+Note: Looking at rats nest between CAD programs some 
+might ask why not just use gmsh because it has the most 
+interconnections and there for looks the most capable. 
+The answer is simple, it isn't just the model you have to 
+move or for that matter the components that have to be 
+composited into it. You have to consider the solder fillets,
+fasteners and chassis. Those are probably going to be 
+hard to add manually in the post pcb-rnd process. Really 
+someone will have to add a tool later to generate that 
+stuff in the correct locations. Doing that ether way in gmsh 
+is a large question but more importantly drafting the 
+chassis and etc in gmsh would be brutal for all but the 
+simplest geometries. Really you want gmsh in the flow for 
+most solvers but for non-E&M problems it has 
+to come after a 3D CAD where everything around the 
+board is developed.
+T 56000 51700 9 10 1 0 0 0 7
+Note: Another question that might be commonly asked is 
+why do we care about paths to solvers not involving a 
+front end. The front ends make life easier but a lot of 
+solvers particularly multiphysics ones like openfoam fenics 
+more limited in functionality when channeled through a 
+front end. Those front ends are like help wizards they look 
+pretty but they narrow your functionality a lot.
+L 48200 48500 50800 49000 8 0 1 0 -1 -1
+L 50800 49000 50800 49800 8 0 1 0 -1 -1
+L 50800 49800 51700 50200 8 0 1 0 -1 -1
+T 50800 51500 9 10 1 0 0 0 16
+Note: The blue vertical lines between CAD programs have 
+their limits. Moving anything from one 3D format to another 
+causes degeneration. Most 3D CAD formats are just bags 
+of triangles. More modern ones have splines and more 
+complex geometry but the point is that changing how these 
+are processed adds rounding error which is a problem. The 
+bottom line is don't hope 3D CAD programs just to add a 
+feature here or there to a model. It will also cost you most 
+if not all meta data about what features are. Most of these 
+programs only understand a very little about what the 
+geometry is in the first place. 
+Making matters worse each CAD engines native format can 
+store a whole model but to move that model to a new CAD 
+with all meta data require splitting the model up into parts 
+each in their own file and importing them into the other CAD.
+This is a brutally unpleasant process.
+L 51500 46900 51700 47200 4 0 1 0 -1 -1
+L 51500 46400 51500 46900 4 0 1 0 -1 -1
+L 51500 46400 51700 46100 4 0 1 0 -1 -1
+L 51500 44500 51700 44100 4 0 1 0 -1 -1
+L 51500 44500 51500 46000 4 0 1 0 -1 -1
+L 51500 46000 51700 46100 4 0 1 0 -1 -1
+L 51100 48600 51700 48800 4 0 1 0 -1 -1
+L 51100 46700 51100 48600 4 0 1 0 -1 -1
+L 51100 46700 51700 46100 4 0 1 0 -1 -1
+L 52700 50200 54500 49200 3 0 0 0 -1 -1
+L 54500 49200 55300 49200 3 0 0 0 -1 -1
+L 55300 49200 55500 49500 3 0 0 0 -1 -1
+L 52700 48800 55500 48400 4 0 1 0 -1 -1
+L 52700 47200 55500 48400 4 0 1 0 -1 -1
+L 52700 48800 55500 43800 4 0 1 0 -1 -1
+T 52200 45800 9 10 1 0 0 3 1
+(LGPL2.1)
+T 57200 41700 9 10 1 0 0 0 15
+*brlcad
+example 
+https://brlcad.org/wiki/SGI_Cube
+
+*openscad
+The whole thing was built to be scripted.
+
+*salome
+???
+
+*freecad
+Look at "The parametric way" here.
+http://www.freecadweb.org/wiki/index.php?title=Drawing_Module
+http://www.freecadweb.org/wiki/index.php?title=Topological_data_scripting
+https://www.youtube.com/watch?v=Cann15ajYlY
+T 58200 48700 9 10 1 0 0 0 11
+Converting geometry between cad programs...
+
+openscad can be read perfectly in freecad but the reverse requires reconstruction because of freecads fundamental architecture being BREP and not CSG.
+http://www.freecadweb.org/wiki/index.php?title=OpenSCAD_Module
+
+to understand the difference between BREP and CSG look at how the brlcad developers explain it in this document page 4.
+http://brlcad.org/w/images/6/66/Converting_Geometry.pdf
+(note this document is rather out of date but the concepts it describes are still valid)
+
+https://sourceforge.net/p/brlcad/mailman/message/25933181/
+from the cocreator himself. don't change formats
+T 52200 43800 9 10 1 0 0 3 1
+(LGPL3)
+T 56000 42500 9 10 1 0 0 3 1
+(GPL3)
Index: openems-layout-outline.sch
===================================================================
--- openems-layout-outline.sch	(nonexistent)
+++ openems-layout-outline.sch	(revision 6816)
@@ -0,0 +1,110 @@
+v 20130925 2
+C 40000 40000 0 0 0 title-C.sym
+B 42000 51000 4500 4500 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+A 42000 55500 500 90 180 3 0 0 0 -1 -1
+A 46500 55500 500 270 180 3 0 0 0 -1 -1
+L 42000 56000 46500 56000 3 0 0 0 -1 -1
+L 42000 55000 46500 55000 3 0 0 0 -1 -1
+T 44800 55700 9 10 1 0 0 0 1
+trace
+T 42300 56300 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+T 47800 56300 9 20 1 0 0 0 1
+What we need in OpenEMS.
+A 42000 51000 500 180 180 3 0 0 0 -1 -1
+A 42000 55500 500 0 180 3 0 0 0 -1 -1
+L 41500 51000 41500 55500 3 0 0 0 -1 -1
+L 42500 51000 42500 55500 3 0 0 0 -1 -1
+T 41900 54000 9 10 1 0 90 0 1
+trace
+A 46500 51000 500 270 180 3 0 0 0 -1 -1
+A 42000 51000 500 90 180 3 0 0 0 -1 -1
+L 46500 50500 42000 50500 3 0 0 0 -1 -1
+L 46500 51500 42000 51500 3 0 0 0 -1 -1
+A 46500 55500 500 0 180 3 0 0 0 -1 -1
+A 46500 51000 500 180 180 3 0 0 0 -1 -1
+L 47000 55500 47000 51000 3 0 0 0 -1 -1
+L 46000 55500 46000 51000 3 0 0 0 -1 -1
+T 43500 50900 9 10 1 0 180 0 1
+trace
+T 46700 52600 9 10 1 0 270 0 1
+trace
+B 48000 51000 4500 4500 3 0 0 1 -1 100 1 -1 -1 -1 -1 -1
+B 42000 43500 4500 4500 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+A 42000 48000 500 90 180 3 0 0 0 -1 -1
+A 46500 48000 500 270 180 3 0 0 0 -1 -1
+L 42000 48500 46500 48500 3 0 0 0 -1 -1
+L 42000 47500 46500 47500 3 0 0 0 -1 -1
+A 42000 43500 500 180 180 3 0 0 0 -1 -1
+A 42000 48000 500 0 180 3 0 0 0 -1 -1
+L 41500 43500 41500 48000 3 0 0 0 -1 -1
+L 42500 43500 42500 48000 3 0 0 0 -1 -1
+A 46500 43500 500 270 180 3 0 0 0 -1 -1
+A 42000 43500 500 90 180 3 0 0 0 -1 -1
+L 46500 43000 42000 43000 3 0 0 0 -1 -1
+L 46500 44000 42000 44000 3 0 0 0 -1 -1
+A 46500 48000 500 0 180 3 0 0 0 -1 -1
+A 46500 43500 500 180 180 3 0 0 0 -1 -1
+L 47000 48000 47000 43500 3 0 0 0 -1 -1
+L 46000 48000 46000 43500 3 0 0 0 -1 -1
+T 44800 48200 9 10 1 0 0 0 1
+trace
+T 42300 48800 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+T 41900 46500 9 10 1 0 90 0 1
+trace
+T 43500 43400 9 10 1 0 180 0 1
+trace
+T 46700 45100 9 10 1 0 270 0 1
+trace
+A 43500 46500 200 0 180 3 0 0 0 -1 -1
+L 43300 46500 43300 45300 3 0 0 0 -1 -1
+L 43700 46500 43700 45300 3 0 0 0 -1 -1
+A 43500 45300 200 180 180 3 0 0 0 -1 -1
+A 44100 46500 200 0 180 3 0 0 0 -1 -1
+L 43900 46500 43900 45300 3 0 0 0 -1 -1
+L 44300 46500 44300 45300 3 0 0 0 -1 -1
+A 44100 45300 200 180 180 3 0 0 0 -1 -1
+A 43500 46500 200 90 180 3 0 0 0 -1 -1
+L 43500 46300 44100 46300 3 0 0 0 -1 -1
+L 43500 46700 44100 46700 3 0 0 0 -1 -1
+A 44100 46500 200 270 180 3 0 0 0 -1 -1
+A 44100 45300 200 270 180 3 0 0 0 -1 -1
+L 43500 45100 44100 45100 3 0 0 0 -1 -1
+A 43500 45300 200 90 180 3 0 0 0 -1 -1
+L 43500 45500 44100 45500 3 0 0 0 -1 -1
+B 48000 43500 4500 4500 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+A 50100 45300 200 270 360 3 0 0 0 -1 -1
+A 49500 45300 200 90 360 3 0 0 0 -1 -1
+T 49000 43100 9 10 1 0 0 0 1
+Board outline shown not filled in for readability
+A 50100 46500 200 270 360 3 0 0 0 -1 -1
+A 49500 46500 200 270 360 3 0 0 0 -1 -1
+T 55000 40700 9 10 1 0 0 0 1
+Board Outline and Cutouts
+L 50100 46300 53100 46000 3 0 0 0 -1 -1
+L 50300 45800 53100 45500 3 0 0 0 -1 -1
+L 49300 46500 49500 46700 3 0 0 0 -1 -1
+L 49500 46700 50100 46700 3 0 0 0 -1 -1
+L 50100 46700 50300 46500 3 0 0 0 -1 -1
+L 50300 46500 50300 45300 3 0 0 0 -1 -1
+L 50300 45300 50100 45100 3 0 0 0 -1 -1
+L 50100 45100 49500 45100 3 0 0 0 -1 -1
+L 49500 45100 49300 45300 3 0 0 0 -1 -1
+L 49300 45300 49300 46500 3 0 0 0 -1 -1
+L 55000 44300 55200 44500 3 0 0 0 -1 -1
+L 55200 44500 55800 44500 3 0 0 0 -1 -1
+L 55800 44500 56000 44300 3 0 0 0 -1 -1
+L 56000 44300 56000 43100 3 0 0 0 -1 -1
+L 56000 43100 55800 42900 3 0 0 0 -1 -1
+L 55800 42900 55200 42900 3 0 0 0 -1 -1
+L 55200 42900 55000 43100 3 0 0 0 -1 -1
+L 55000 43100 55000 44300 3 0 0 0 -1 -1
+T 53200 46000 9 10 1 0 0 0 1
+The rounded corners are done with cylinders.
+T 53200 45000 9 10 1 0 0 0 3
+The polygon is used to take up the body of the cutout is 
+made using tangent lines between the outside edges of 
+the cylinders and extruding the Z axis
+T 54400 44600 9 10 1 0 0 0 1
+The polygon on it's own.
Index: openems-layout-simple.sch
===================================================================
--- openems-layout-simple.sch	(nonexistent)
+++ openems-layout-simple.sch	(revision 6816)
@@ -0,0 +1,183 @@
+v 20130925 2
+C 40000 40000 0 0 0 title-C.sym
+B 49000 53000 6500 2000 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+A 50000 54000 500 90 180 3 0 0 0 -1 -1
+A 54500 54000 500 270 180 3 0 0 0 -1 -1
+L 55800 54400 55000 54100 3 0 0 0 -1 -1
+L 55800 55300 55500 55000 3 0 0 0 -1 -1
+B 41500 53000 6500 2000 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+A 50000 54000 500 270 180 3 0 0 2 100 100
+L 50400 54300 51300 55900 3 0 0 0 -1 -1
+A 42500 54000 500 90 180 3 0 0 0 -1 -1
+A 47000 54000 500 270 180 3 0 0 0 -1 -1
+L 42500 54500 47000 54500 3 0 0 0 -1 -1
+L 42500 53500 47000 53500 3 0 0 0 -1 -1
+L 42200 53600 42600 52700 3 0 0 0 -1 -1
+L 46500 52600 46200 53000 3 0 0 0 -1 -1
+T 52100 52700 9 10 1 0 0 0 1
+Body of the trace replaced with a rectangle
+T 55800 54400 9 10 1 0 0 0 1
+Ends of the trace replaced with a smaller polygon extruded vertically
+T 55800 55300 9 10 1 0 0 0 1
+board outline is now a polygon extruded vertically
+T 51300 55900 9 10 1 0 0 0 1
+annular ring is here but occupies the same space as the body of the trace and it's round end.
+T 45500 53900 9 10 1 0 0 0 1
+trace
+T 42600 52700 9 10 1 0 0 0 1
+end of trace
+T 46500 52600 9 10 1 0 0 0 1
+board outline
+T 42300 56300 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+T 50300 56300 9 20 1 0 0 0 1
+What we need in OpenEMS.
+L 54500 54500 54500 53500 3 0 0 2 100 100
+L 50000 54500 54500 54500 3 0 0 0 -1 -1
+L 54500 53500 50000 53500 3 0 0 0 -1 -1
+L 50000 54500 50000 53500 3 0 0 2 100 100
+T 44500 40300 9 10 1 0 0 0 6
+To me the most annoying corner case is the trouble of board outline vs board cutout. 
+PCB fabs by default assume the outline of the board is the middle of the lines drawn 
+around the layout. Board cutouts are however milled to the exterior edge of the lines 
+drawn in the gerber. Since both are done on the outline layer in pcb-rnd we have to 
+figure out which is the outline of the board and which is the outline of the cutout 
+before converting the geometry.
+L 50000 54600 50000 53400 3 0 0 4 100 100
+A 54500 54000 600 180 360 3 0 0 4 100 100
+L 54500 54500 54500 53500 3 0 0 2 100 100
+L 54500 54600 54500 53400 3 0 0 4 100 100
+L 54500 53400 50000 53400 3 0 0 4 100 100
+L 54500 54600 50000 54600 3 0 0 4 100 100
+L 52100 52700 51500 53500 3 0 0 0 -1 -1
+A 42500 54000 600 90 180 3 0 0 4 100 100
+A 47000 54000 600 270 180 3 0 0 4 100 100
+L 47000 53400 42500 53400 3 0 0 4 100 100
+L 47000 54600 42500 54600 3 0 0 4 100 100
+A 50000 54000 600 180 360 3 0 0 4 100 100
+A 54500 54000 500 90 180 3 0 0 2 100 100
+L 55000 53700 55800 53500 3 0 0 0 -1 -1
+T 52800 55600 9 10 1 0 0 0 1
+clearance around the trace's body is just a square sharing 2 of the same edges with the body only wider
+L 52800 55600 52100 54600 3 0 0 0 -1 -1
+T 55800 53100 9 10 1 0 0 0 3
+Clearance around the end of the trace (required if we are in a 
+	polygon) is a cylinder of what ever matterial the copper 
+	was displacing (air, solder mask, substrate/adheasive).
+L 43700 54600 44500 55400 3 0 0 0 -1 -1
+T 44500 55400 9 10 1 0 0 0 1
+Clearance around the trace
+B 47500 47000 2500 2500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 48500 48500 200 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+V 48500 48500 300 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+A 48500 48500 800 270 360 3 0 0 4 100 100
+A 48500 48500 500 90 360 3 0 0 0 -1 -1
+T 50100 52000 9 10 1 0 0 0 3
+Clearance around the end of the trace (required if we are in a 
+	polygon) is a cylinder of what ever matterial the copper 
+	was displacing (air, solder mask, substrate/adheasive).
+L 49700 53500 50100 52400 3 0 0 0 -1 -1
+L 48500 49300 50300 50600 3 0 0 0 -1 -1
+L 48900 48800 50300 49100 3 0 0 0 -1 -1
+L 50300 48300 48700 48500 3 0 0 0 -1 -1
+L 50300 47800 48700 48300 3 0 0 0 -1 -1
+T 50300 50200 9 10 1 0 0 0 3
+Clearance around the via (required if we are in a polygon) is 
+	a cylinder of what ever matterial the copper was 
+	displacing (air, solder mask, substrate/adheasive).
+T 50300 49100 9 10 1 0 0 0 1
+annular ring is here but occupies the same space as the body of the trace and it's round end.
+T 50300 48300 9 10 1 0 0 0 1
+via hole is now a cylinder of air which ends just at the start of the solder mask layer.
+T 50300 47800 9 10 1 0 0 0 1
+the vias plating is a cylinder which is at a priority above the substrate.
+T 50300 49700 9 10 1 0 0 0 1
+board outline is now a polygon extruded vertically
+T 47500 50900 9 20 1 0 0 0 1
+Vias in OpenEMS
+A 42000 48500 500 90 360 3 0 0 0 -1 -1
+V 42000 48500 300 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+V 42000 48500 200 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 42200 48400 43800 47800 3 0 0 0 -1 -1
+L 42100 48200 43800 47500 3 0 0 0 -1 -1
+L 42100 48000 43800 47200 3 0 0 0 -1 -1
+L 43800 49700 43500 49500 3 0 0 0 -1 -1
+A 42000 48500 800 270 360 3 0 0 4 100 100
+L 42000 49300 43800 50100 3 0 0 0 -1 -1
+T 43800 47800 9 10 1 0 0 0 1
+via hole
+T 43800 47500 9 10 1 0 0 0 1
+via hole plating
+T 43800 47200 9 10 1 0 0 0 1
+annular ring (overlapped by end of trace)
+T 43800 49700 9 10 1 0 0 0 1
+board outline
+T 41000 50900 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+T 43800 50100 9 10 1 0 0 0 1
+Clearance around the via
+B 41000 47000 2500 2500 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 50300 49700 50000 49500 3 0 0 0 -1 -1
+L 47700 44600 49200 43800 3 0 0 0 -1 -1
+L 47700 43800 48000 43600 3 0 0 0 -1 -1
+B 53000 41800 4200 3700 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 55300 46000 54300 44100 3 0 0 0 -1 -1
+L 57500 43100 57200 43000 3 0 0 0 -1 -1
+A 54600 42700 1500 0 90 3 0 0 0 -1 -1
+A 54600 42700 500 0 90 3 0 0 0 -1 -1
+A 55600 42700 500 180 180 3 0 0 0 -1 -1
+A 54600 43700 500 90 180 3 0 0 0 -1 -1
+A 54600 43700 500 270 180 3 0 0 2 100 100
+L 54600 44200 54600 43200 3 0 0 2 100 100
+L 56100 42700 55100 42700 3 0 0 2 100 100
+A 55600 42700 500 0 180 3 0 0 2 100 100
+L 57500 42600 56000 42400 3 0 0 0 -1 -1
+B 48000 41800 4200 3700 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 57500 43900 55800 43600 3 0 0 0 -1 -1
+A 49700 42700 500 0 90 3 0 0 0 -1 -1
+A 49700 43700 500 90 180 3 0 0 0 -1 -1
+A 49700 42700 1500 0 90 3 0 0 0 -1 -1
+A 50700 42700 500 180 180 3 0 0 0 -1 -1
+A 49700 43700 600 90 180 3 0 0 4 100 100
+L 47700 45300 49600 44300 3 0 0 0 -1 -1
+A 49700 42700 1600 0 90 3 0 0 4 100 100
+A 50700 42700 600 180 180 3 0 0 4 100 100
+A 49700 42700 400 0 90 3 0 0 4 100 100
+A 54600 42700 1600 0 90 3 0 0 4 100 100
+A 55600 42700 600 180 360 3 0 0 4 100 100
+A 54600 42700 400 0 90 3 0 0 4 100 100
+L 56200 42700 55000 42700 3 0 0 4 100 100
+L 54600 44300 54600 43100 3 0 0 4 100 100
+L 47700 42400 50100 42700 3 0 0 0 -1 -1
+L 55900 42200 57500 42300 3 0 0 0 -1 -1
+L 56100 43300 57500 43500 3 0 0 0 -1 -1
+A 54600 43700 600 180 360 3 0 0 4 100 100
+T 50200 43400 9 10 1 0 0 0 1
+arc
+T 47700 44600 9 10 1 0 0 6 1
+annular ring (overlapped by end of trace)
+T 47700 43800 9 10 1 0 0 6 1
+board outline
+T 48000 45700 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+T 57500 43900 9 10 1 0 0 0 1
+Rotational box to form the body of the arc
+T 57500 43100 9 10 1 0 0 0 1
+board outline
+T 53000 45700 9 20 1 0 0 0 1
+OpenEMS
+T 55300 46000 9 10 1 0 0 0 1
+cylinders for the ends of the arc
+T 57500 42600 9 10 1 0 0 0 1
+cylinder used to add rounded end to the arc
+T 47700 45300 9 10 1 0 0 6 1
+trace clearance
+T 47700 42400 9 10 1 0 0 6 1
+outline of trace clearance
+T 57500 42300 9 10 1 0 0 0 1
+cylinder used to add clearance in plane(s)
+T 57500 43500 9 10 1 0 0 0 1
+rotated box used to create clearance around arc
+T 55200 46300 9 10 1 0 0 0 1
+cylinder used to add clearance
+L 55200 46300 54200 44200 3 0 0 0 -1 -1
Index: openems-notes-layout.sch
===================================================================
--- openems-notes-layout.sch	(nonexistent)
+++ openems-notes-layout.sch	(revision 6816)
@@ -0,0 +1,181 @@
+v 20130925 2
+C 40000 40000 0 0 0 title-C.sym
+B 48500 50500 6500 2000 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+A 49500 51500 500 90 180 3 0 0 0 -1 -1
+A 54000 51500 500 270 180 3 0 0 0 -1 -1
+V 49500 51500 200 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+V 49500 51500 300 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+L 50900 49900 49700 51500 3 0 0 0 -1 -1
+L 50400 49500 49700 51300 3 0 0 0 -1 -1
+L 55300 51900 54500 51600 3 0 0 0 -1 -1
+L 55300 52800 55000 52500 3 0 0 0 -1 -1
+B 41500 50500 6500 2000 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+A 49500 51500 500 270 180 3 0 0 2 100 100
+L 49900 51800 50500 53400 3 0 0 0 -1 -1
+A 42500 51500 500 90 180 3 0 0 0 -1 -1
+A 47000 51500 500 270 180 3 0 0 0 -1 -1
+V 42500 51500 300 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+V 42500 51500 200 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 42500 52000 47000 52000 3 0 0 0 -1 -1
+L 42500 51000 47000 51000 3 0 0 0 -1 -1
+L 42700 51400 43700 50200 3 0 0 0 -1 -1
+L 42500 51200 43400 50000 3 0 0 0 -1 -1
+L 42200 51100 42800 49800 3 0 0 0 -1 -1
+L 46500 50100 46200 50500 3 0 0 0 -1 -1
+T 51600 50200 9 10 1 0 0 0 1
+Body of the trace replaced with a rectangle
+T 55300 51900 9 10 1 0 0 0 1
+Ends of the trace replaced with a smaller polygon extruded vertically
+T 55300 52800 9 10 1 0 0 0 1
+board outline is now a polygon extruded vertically
+T 50900 49900 9 10 1 0 0 0 1
+via hole is now a cylinder of air which ends just at the start of the solder mask layer.
+T 50400 49500 9 10 1 0 0 0 1
+the vias plating is a cylinder which is at a priority above the substrate.
+T 50500 53400 9 10 1 0 0 0 1
+annular ring is here but occupies the same space as the body of the trace and it's round end.
+T 45500 51400 9 10 1 0 0 0 1
+trace
+T 43700 50200 9 10 1 0 0 0 1
+via hole
+T 43400 50000 9 10 1 0 0 0 1
+via hole plating
+T 42800 49800 9 10 1 0 0 0 1
+annular ring (overlapped by end of trace)
+T 46500 50100 9 10 1 0 0 0 1
+board outline
+T 43100 53900 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+T 50000 54600 9 20 1 0 0 0 1
+What we need in OpenEMS.
+V 49700 45700 300 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+V 49700 45700 200 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 49700 45500 47700 44800 3 0 0 0 -1 -1
+L 47700 44500 49700 45400 3 0 0 0 -1 -1
+L 47700 46600 49200 45800 3 0 0 0 -1 -1
+L 47700 45800 48000 45600 3 0 0 0 -1 -1
+T 50500 45100 9 10 1 0 0 0 1
+arc
+T 47700 44800 9 10 1 0 0 6 1
+via hole
+T 47700 44500 9 10 1 0 0 6 1
+via hole plating
+T 47700 46600 9 10 1 0 0 6 1
+annular ring (overlapped by end of trace)
+T 47700 45800 9 10 1 0 0 6 1
+board outline
+T 48100 47800 9 20 1 0 0 0 1
+What we have in pcb-rnd.
+B 53000 43800 4200 3700 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+V 54600 45700 300 3 0 0 1 -1 100 0 -1 -1 -1 -1 -1
+L 54800 45700 57500 46700 3 0 0 0 -1 -1
+L 57500 46300 54900 45600 3 0 0 0 -1 -1
+L 55500 48300 54300 46100 3 0 0 0 -1 -1
+L 57500 45100 57200 45000 3 0 0 0 -1 -1
+A 54600 44700 1500 0 90 3 0 0 0 -1 -1
+A 54600 44700 500 0 90 3 0 0 0 -1 -1
+A 55600 44700 500 180 180 3 0 0 0 -1 -1
+A 54600 45700 500 90 180 3 0 0 0 -1 -1
+T 57500 45900 9 10 1 0 0 0 1
+Rotational Polygon to form the body of the arc
+T 57500 46700 9 10 1 0 0 0 1
+via hole is a cylinder filled with air
+T 57500 46300 9 10 1 0 0 0 1
+via hole plating a cylinder of metal through the core
+T 55500 48300 9 10 1 0 0 0 1
+two cylinders: 1 from via annular ring
+T 57500 45100 9 10 1 0 0 0 1
+board outline
+T 53100 47800 9 20 1 0 0 0 1
+OpenEMS
+A 54600 45700 500 270 180 3 0 0 2 100 100
+L 54600 46200 54600 45200 3 0 0 2 100 100
+L 56100 44700 55100 44700 3 0 0 2 100 100
+T 56600 48100 9 10 1 0 0 0 1
+2. from the end of the arc
+A 55600 44700 500 0 180 3 0 0 2 100 100
+V 54600 45700 200 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+T 57500 44600 9 10 1 0 0 0 1
+cylinder used to add rounded end to the arc
+L 57500 44600 56000 44400 3 0 0 0 -1 -1
+B 48000 43800 4200 3700 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 54000 52000 54000 51000 3 0 0 2 100 100
+L 49500 52000 54000 52000 3 0 0 0 -1 -1
+L 54000 51000 49500 51000 3 0 0 0 -1 -1
+L 49500 52000 49500 51000 3 0 0 2 100 100
+L 57500 45900 55800 45600 3 0 0 0 -1 -1
+A 54600 45700 800 270 360 3 0 0 4 100 100
+A 49700 45700 800 270 360 3 0 0 4 100 100
+T 47700 45200 9 10 1 0 0 6 1
+via clearance
+L 47700 45200 48900 45600 3 0 0 0 -1 -1
+L 56700 47900 55000 46400 3 0 0 0 -1 -1
+T 58700 48000 9 10 1 0 180 0 1
+cylinder for via clearance
+A 49700 44700 500 0 90 3 0 0 0 -1 -1
+A 49700 45700 500 90 180 3 0 0 0 -1 -1
+A 49700 44700 1500 0 90 3 0 0 0 -1 -1
+A 50700 44700 500 180 180 3 0 0 0 -1 -1
+A 49700 45700 600 90 180 3 0 0 4 100 100
+T 47700 47300 9 10 1 0 0 6 1
+trace clearance
+L 47700 47300 49600 46300 3 0 0 0 -1 -1
+A 49700 44700 1600 0 90 3 0 0 4 100 100
+A 50700 44700 600 180 180 3 0 0 4 100 100
+A 49700 44700 400 0 90 3 0 0 4 100 100
+A 54600 44700 1600 0 90 3 0 0 4 100 100
+A 55600 44700 600 180 360 3 0 0 4 100 100
+A 54600 44700 400 0 90 3 0 0 4 100 100
+L 56200 44700 55000 44700 3 0 0 4 100 100
+L 54600 46300 54600 45100 3 0 0 4 100 100
+T 47700 44000 9 10 1 0 0 6 1
+outline of trace clearance
+L 47700 44000 50100 44700 3 0 0 0 -1 -1
+T 57500 44300 9 10 1 0 0 0 1
+cylinder used to add clearance in plane(s)
+L 55900 44200 57500 44300 3 0 0 0 -1 -1
+L 56100 45300 57500 45500 3 0 0 0 -1 -1
+T 57500 45500 9 10 1 0 0 0 1
+rotated box used to create clearance around arc
+A 49500 51500 800 270 360 3 0 0 4 100 100
+L 49500 52100 49500 50900 3 0 0 4 100 100
+A 54000 51500 600 180 360 3 0 0 4 100 100
+L 54000 52000 54000 51000 3 0 0 2 100 100
+L 54000 52100 54000 50900 3 0 0 4 100 100
+L 54000 50900 49500 50900 3 0 0 4 100 100
+L 54000 52100 49500 52100 3 0 0 4 100 100
+L 51600 50200 51000 51000 3 0 0 0 -1 -1
+A 42500 51500 800 270 360 3 0 0 4 100 100
+A 42500 51500 600 90 180 3 0 0 4 100 100
+A 47000 51500 600 270 180 3 0 0 4 100 100
+L 47000 50900 42500 50900 3 0 0 4 100 100
+L 47000 52100 42500 52100 3 0 0 4 100 100
+A 54600 45700 600 180 360 3 0 0 4 100 100
+A 49500 51500 600 180 360 3 0 0 4 100 100
+A 54000 51500 500 90 180 3 0 0 2 100 100
+L 54500 51200 55300 51000 3 0 0 0 -1 -1
+T 52300 53100 9 10 1 0 0 0 1
+clearance around the trace's body is just a square sharing 2 of the same edges with the body only wider
+L 52300 53100 51600 52100 3 0 0 0 -1 -1
+T 55300 50600 9 10 1 0 0 0 3
+Clearance around the end of the trace (required if we are in a 
+	polygon) is a cylinder of what ever matterial the copper 
+	was displacing (air, solder mask, substrate/adheasive).
+L 43700 52100 44500 52900 3 0 0 0 -1 -1
+L 42500 52300 43400 53300 3 0 0 0 -1 -1
+T 44500 52900 9 10 1 0 0 0 1
+Clearance around the trace
+T 43400 53300 9 10 1 0 0 0 1
+Clearance around the via
+T 50400 53700 9 10 1 0 0 0 3
+Clearance around the via (required if we are in a polygon) is 
+	a cylinder of what ever matterial the copper was 
+	displacing (air, solder mask, substrate/adheasive).
+L 49500 52300 50400 54100 3 0 0 0 -1 -1
+T 49600 48700 9 10 1 0 0 0 3
+Clearance around the end of the trace (required if we are in a 
+	polygon) is a cylinder of what ever matterial the copper 
+	was displacing (air, solder mask, substrate/adheasive).
+L 49300 50900 49600 49100 3 0 0 0 -1 -1
+T 46300 56000 9 10 1 0 0 0 1
+I drew these first and then realized they were too cluttered. There is a second set in the new file that is cleaner.
Index: openems-notes.sch
===================================================================
--- openems-notes.sch	(nonexistent)
+++ openems-notes.sch	(revision 6816)
@@ -0,0 +1,322 @@
+v 20130925 2
+C 40000 40000 0 0 0 title-C.sym
+B 52000 50100 500 4800 3 0 1 1 -1 100 0 -1 -1 -1 -1 -1
+B 52100 50100 300 4800 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+L 53300 55300 53100 54700 3 0 0 0 -1 -1
+B 49500 52100 5500 700 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 49500 53600 5500 600 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 49500 53000 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 49500 53400 49500 53600 3 0 0 0 -1 -1
+L 55000 52800 55000 53600 3 0 0 0 -1 -1
+L 49500 52800 49500 53000 3 0 0 0 -1 -1
+B 51400 51500 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 55000 51900 55000 52100 3 0 0 0 -1 -1
+L 49500 51300 49500 52100 3 0 0 0 -1 -1
+L 55000 51300 55000 51500 3 0 0 0 -1 -1
+B 49500 49900 5500 600 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 51400 49700 51800 50100 3 0 0 0 -1 -1
+B 49500 50700 5500 600 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 49500 54200 5500 200 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+B 49500 50500 5500 200 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+B 51400 50100 1700 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+B 51400 54400 1700 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 52500 51100 54000 49800 3 0 0 0 -1 -1
+B 58500 50100 500 4800 3 0 1 1 -1 100 0 -1 -1 -1 -1 -1
+B 58600 50100 300 4800 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+L 59600 55200 59200 54800 3 0 0 0 -1 -1
+B 56000 51900 5500 1100 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 56000 53400 5500 1000 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 56000 53000 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 61500 53000 61500 53400 3 0 0 0 -1 -1
+B 57900 51500 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 56000 51500 56000 51900 3 0 0 0 -1 -1
+B 56000 49900 5500 600 3 0 0 0 -1 -1 0 -1 -1 -1 -1 -1
+L 57900 49700 58300 50100 3 0 0 0 -1 -1
+B 56000 50500 5500 1000 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 57900 50100 1700 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+B 57900 54400 1700 500 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 59000 51100 60600 49700 3 0 0 0 -1 -1
+B 43000 41000 500 3100 3 0 1 1 -1 100 0 -1 -1 -1 -1 -1
+B 43100 41000 300 3100 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+L 44100 44500 43700 44100 3 0 0 0 -1 -1
+B 40500 41000 5500 3100 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+L 42200 40500 42700 41000 3 0 0 0 -1 -1
+L 43500 41500 45100 40400 3 0 0 0 -1 -1
+B 49500 42400 5500 1100 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 52200 42400 300 1100 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+B 52100 42400 500 1100 3 0 1 1 -1 100 0 -1 -1 -1 -1 -1
+L 51100 54400 51100 55000 3 0 0 0 -1 -1
+L 51100 55000 49500 55000 3 0 0 0 -1 -1
+L 49500 55000 49500 54400 3 0 0 0 -1 -1
+L 53400 54400 53400 55000 3 0 0 0 -1 -1
+L 53400 55000 55000 55000 3 0 0 0 -1 -1
+L 55000 55000 55000 54400 3 0 0 0 -1 -1
+L 51400 55400 51700 55000 3 0 0 0 -1 -1
+B 51400 54800 1700 100 3 0 0 0 -1 -1 3 1 4 200 -1 -1
+L 52800 55100 52900 54900 3 0 0 0 -1 -1
+T 40500 52300 9 10 1 0 0 0 5
+The exporter should ask what the lowest priority it can use is and what the highest 
+is. That way if you run out of room it can warn. I want to be able to set the pcb's 
+base priority so that it can be super imposed in designs. I am not sure how well 
+that will work but it is just a question of what number to use so we are not talking 
+about a minor addition to the complexity we already have.
+T 55400 40800 9 10 1 0 0 0 4
+http://openems.de/index.php/Tutorial:_Importing_with_hyp2mat
+
+3D copper used to model antennas
+http://openems.de/index.php/Tutorial:_Simple_Patch_Antenna
+T 49700 54600 9 10 1 0 0 0 1
+Solder Mask
+T 52100 52300 22 10 1 0 0 0 1
+Air
+T 53300 55300 9 10 1 0 0 0 1
+Annular Ring
+T 50100 53100 9 10 1 0 0 0 1
+trace on inside layer
+T 49700 52400 9 10 1 0 0 0 1
+substrate
+T 54100 52400 9 10 1 0 0 0 1
+substrate
+T 54100 53800 9 10 1 0 0 0 1
+substrate
+T 49700 53800 9 10 1 0 0 0 1
+substrate
+T 53600 53200 9 10 1 0 0 0 1
+adheasive
+T 50800 52800 9 10 1 0 0 0 1
+adheasive
+T 50800 53400 9 10 1 0 0 0 1
+adheasive
+T 40500 53500 9 10 1 0 0 0 15
+we have to assume the board is planar otherwise impedances will be a nightmare.
+yes in real life some vendors squish boards in ways that make the outline of 
+internal copper layers embossed into all the layers over them. This type of 
+manufacturing is impossibly bad for calculating impedances.
+
+To save computational power in the E&M solver we should also ingore the tiny 
+layer of adheasive under and over the copper layer. Making the stackup the same 
+thickness but the substrate slightly thicker. So the stackup above on the left 
+is altered into the one on the right. 
+
+I am not sure if the annular ring actually has adheasive under it or if it is plated 
+directly onto the board. This may vary with process. More research is needed.
+
+If we do mechanical simulation or export later we should add in a feature to 
+support export with adheasive because it adds required boundary conditions.
+T 54600 51600 9 10 1 0 0 6 1
+trace on inside layer
+T 50900 51700 9 10 1 0 0 6 1
+adheasive
+T 53700 51300 9 10 1 0 0 6 1
+adheasive
+T 53700 51900 9 10 1 0 0 6 1
+adheasive
+T 54800 50300 9 10 1 0 180 0 1
+Solder Mask
+T 51400 49700 9 10 1 0 180 0 1
+Annular Ring
+T 50400 51100 9 10 1 0 180 0 1
+substrate
+T 54800 51100 9 10 1 0 180 0 1
+substrate
+T 50800 54200 9 10 1 0 0 0 1
+adheasive
+T 53000 54200 9 10 1 0 0 0 1
+adheasive
+T 50800 50500 9 10 1 0 0 0 1
+adheasive
+T 53800 50500 9 10 1 0 0 0 1
+adheasive
+T 53000 49600 9 10 1 0 0 0 1
+copper plating on the hole
+T 58600 52300 22 10 1 0 0 0 1
+Air
+T 59600 55200 9 10 1 0 0 0 1
+Annular Ring
+T 56500 53100 9 10 1 0 0 0 1
+trace on inside layer
+T 56200 52400 9 10 1 0 0 0 1
+substrate
+T 59900 52400 9 10 1 0 0 0 1
+substrate
+T 60000 53800 9 10 1 0 0 0 1
+substrate
+T 56200 53800 9 10 1 0 0 0 1
+substrate
+T 60100 53200 9 10 1 0 0 0 1
+adheasive
+T 61100 51600 9 10 1 0 0 6 1
+trace on inside layer
+T 57400 51700 9 10 1 0 0 6 1
+adheasive
+T 61300 50300 9 10 1 0 180 0 1
+Solder Mask
+T 57900 49700 9 10 1 0 180 0 1
+Annular Ring
+T 56900 51100 9 10 1 0 180 0 1
+substrate
+T 61000 51100 9 10 1 0 180 0 1
+substrate
+T 59600 49500 9 10 1 0 0 0 1
+copper plating on the hole
+T 47900 41300 9 10 1 0 0 0 1
+Note: None of these drawings are to scale.
+T 49400 55800 9 20 1 0 0 0 1
+Worst case for PCB-RND storage.
+T 56500 55800 9 20 1 0 0 0 1
+Worst case for OpenEMS.
+T 40500 50900 9 10 1 0 0 0 6
+I am going to assume for simulation that the tiny amount of adheasive under and 
+over the traces is not important to modeling in the E&M domain. If it was we would 
+have to worry about minor variations in it's thickness and etc which would be brutal.
+
+I am also going to assume that the adheasive is not exactly the same impedance as 
+the substrate.
+T 43100 42400 22 10 1 0 0 0 1
+Air
+T 44100 44500 9 10 1 0 0 0 1
+Annular Ring
+T 40700 42500 9 10 1 0 0 0 1
+substrate
+T 40700 43900 9 10 1 0 0 0 1
+substrate
+T 42200 40500 9 10 1 0 180 0 1
+Annular Ring
+T 45800 41200 9 10 1 0 180 0 1
+substrate
+T 44100 40200 9 10 1 0 0 0 1
+copper plating on the hole
+T 40500 44700 9 20 1 0 0 0 1
+Multilayer accelerated with 2D mesh
+T 45700 43400 9 10 1 0 180 0 1
+substrate
+T 52200 42900 22 10 1 0 0 0 1
+Air
+T 49700 42900 9 10 1 0 0 0 1
+substrate
+T 54100 42900 9 10 1 0 0 0 1
+substrate
+T 49500 44400 9 20 1 0 0 0 1
+3D copper replaced with 2D mesh
+T 51400 55400 9 10 1 0 0 6 1
+Exposed Via
+T 53500 54600 9 10 1 0 0 0 1
+Solder Mask
+T 52800 55100 9 10 1 0 0 6 1
+ENIG Plating
+L 59900 54400 59900 55000 3 0 0 0 -1 -1
+L 59900 55000 61500 55000 3 0 0 0 -1 -1
+L 61500 55000 61500 54400 3 0 0 0 -1 -1
+L 57600 54400 57600 55000 3 0 0 0 -1 -1
+L 57600 55000 56000 55000 3 0 0 0 -1 -1
+L 56000 55000 56000 54400 3 0 0 0 -1 -1
+T 60000 54600 9 10 1 0 0 0 1
+Solder Mask
+T 56200 54600 9 10 1 0 0 0 1
+Solder Mask
+T 51100 55200 9 10 1 0 0 6 1
+Solder Mask Clearance
+L 51100 55200 51300 54900 3 0 0 0 -1 -1
+L 57900 55400 58200 55000 3 0 0 0 -1 -1
+L 57600 55200 57800 54900 3 0 0 0 -1 -1
+T 57900 55400 9 10 1 0 0 6 1
+Exposed Via
+T 57600 55200 9 10 1 0 0 6 1
+Solder Mask Clearance
+T 40500 47100 9 10 1 0 0 0 18
+Every matterial has a priority, matterial and geometry in OpenEMS. We can only 
+add matterials we can not subtract. However we can add non-conductive things 
+at higher priority so that we create voids with pockets of air.
+
+There is an added complexity in that we don't want to have to cut around traces 
+and annular rings that are inside/adjacent to planes.
+
+The following is how the priorities are to be used. The number after the item is the 
+priority relative to the 0 point that users set. I want this feature because there might 
+be a structure people want to layer this over/under after export.
+
+holes, mask clearance & cutouts (air) 6
+copper traces & copper planes with no clearance 5
+substrate (adheasive) clearance 4
+copper planes 3
+adheasive 2
+substrate 1
+environment (air) 0 x and y set at time of export z left as a variable to the user
+L 40500 43100 46000 43100 3 0 0 0 -1 -1
+L 46000 42000 40500 42000 3 0 0 0 -1 -1
+L 42400 42000 46000 42000 7 4 0 2 100 100
+L 40500 43100 44100 43100 7 4 0 2 100 100
+L 42400 41000 44100 41000 7 4 0 2 100 100
+L 42400 44100 44100 44100 7 4 0 2 100 100
+L 51400 42400 55000 42400 7 4 0 2 100 100
+L 49500 43500 53100 43500 7 4 0 2 100 100
+B 58500 43200 500 4800 3 0 1 1 -1 100 0 -1 -1 -1 -1 -1
+B 58600 43200 300 4800 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+L 59600 48400 59200 48000 3 0 0 0 -1 -1
+B 56000 43600 5500 3900 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 56000 46100 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+B 57900 44600 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 57900 42800 58300 43200 3 0 0 0 -1 -1
+B 57900 43200 1700 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+B 57900 47500 1700 500 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+L 59000 44200 60600 43100 3 0 0 0 -1 -1
+B 49500 46400 5500 1100 3 0 1 0 -1 -1 0 -1 -1 -1 -1 -1
+B 49500 47500 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+B 51400 46000 3600 400 3 0 0 0 -1 -1 2 1 -1 100 -1 100
+B 52200 46000 300 1900 3 0 0 0 -1 -1 1 -1 -1 -1 -1 -1
+B 52100 46000 500 1900 3 0 1 1 -1 100 0 -1 -1 -1 -1 -1
+T 58600 45400 22 10 1 0 0 0 1
+Air
+T 59600 48400 9 10 1 0 0 0 1
+Annular Ring
+T 46200 42600 9 10 1 0 0 0 1
+trace on inside layer as 2D mesh
+T 56200 45500 9 10 1 0 0 0 1
+substrate
+T 56200 46900 9 10 1 0 0 0 1
+substrate
+T 61100 44700 9 10 1 0 0 6 1
+trace on inside layer
+T 57900 42800 9 10 1 0 180 0 1
+Annular Ring
+T 61300 44200 9 10 1 0 180 0 1
+substrate
+T 59700 42900 9 10 1 0 0 0 1
+copper plating on the hole
+T 56300 48900 9 20 1 0 0 0 1
+Minimal Multilayer OpenEMS
+T 61200 46400 9 10 1 0 180 0 1
+substrate
+T 52200 46900 22 10 1 0 0 0 1
+Air
+T 50100 47600 9 10 1 0 0 0 1
+trace on inside layer
+T 49700 46900 9 10 1 0 0 0 1
+substrate
+T 54100 46900 9 10 1 0 0 0 1
+substrate
+T 54700 46100 9 10 1 0 0 6 1
+trace on inside layer
+T 48400 48500 9 20 1 0 0 0 1
+What we need in OpenEMS for minimal 2 layer.
+T 40500 45600 9 10 1 0 0 0 6
+In E&M solvers it is not uncommon to cheat a little and make the copper a 2D 
+mesh applied over the substrate to save processing time. The problem is that 
+openems does not currenly have any kind of 2D curved primitive. Emulating curves 
+with a high order polygons would be unpleasant. We should IMHO wait until 
+openems has 2D curve and arc primitives to invest time in acceleration using 2D 
+mesh.
+L 46200 42600 45300 42000 3 0 0 0 -1 -1
+T 46200 43900 9 10 1 0 0 0 1
+trace on inside layer as 2D mesh
+L 46200 43900 43700 43100 3 0 0 0 -1 -1
+T 49200 43800 9 10 1 0 0 0 1
+trace as 2D mesh
+L 50600 43800 51100 43500 3 0 0 0 -1 -1
+T 53800 42000 9 10 1 0 0 0 1
+trace as 2D mesh
+L 53800 42000 53400 42400 3 0 0 0 -1 -1
+T 58000 46200 9 10 1 0 0 6 1
+trace on inside layer
+T 56600 48700 9 10 1 0 0 0 1
+This only works in with 3D representation of copper.