openEMS.de

It is all defined here: https://github.com/thliebig/CSXCAD/blob ... Parser.cpp

If I read my already eight year old code correctly it should be "y0" and "y2".
The reason for y is that this is the name in "math.h". Don't ask me why...

I can only recommend to never use the same priority twice. Unless you are 100% sure they do not touch. STL is a bit tricky as it is unclear what happens if the STL surface is not closed or in others ways not healthy. Make sure the STL is closed, manifold and so on. Then add the STL as metal and noth...

Well maybe I have to be more precise. Intersecting e.g. a lumped port somewhere in the middle is not a good idea. Of course you need to connect the ends with metal signal lines which need to intersect or touch the port.
Just look at all the tutorials and examples on how it should be done.

Intersecting the port with metal parts is not a good idea as the port will not be able to excite a "nice" electric field. But what exactly happens depends strongly on what you are doing...

I agree, you should just do this in the "normal" mesh. Both types of ports should work fine. You "only" need to come up with a transition. But the RWG TE01 und the CWG TE11 mode should be similar enough that this should not be too difficult?

This sound indeed a bit surprising, but maybe you can supply your script and we can have a look? In which direction to you need this dense mesh. In all or only in direction of the waveguide or across the cross section? Do you think that the stair-casing of the waveguide boundaries are the problem? W...

I have played with a SMA to rect waveguide transition but RWG to CWG should be interesting too and should be possible with openEMS (if that was the question). But I have not tried it yet.

Hi Hannes, openEMS itself only depends on the CSXCAD library and both have to interface to any GUI elements and ran run nicely on a server without any graphical interface. All you need to transfer simulation data and run openEMS is for example an SSH access. This is what I did plenty in the past. Th...

Well the mesh usually is the main cause of long simulations. A good indication is if the excitation is already requiring a large amount of time-steps and the nyquist rate is very high too. Example for good numbers: FDTD timestep is: 9.63258e-13 s; Nyquist rate: 173 timesteps @3.00041e+09 Hz Excitati...

I never tried, but sure. Just record and analyze SAR at different frequencies (may take a while) and use Matlab/Octave to plot in any way and combination.

regards
Thorsten