frequency selective surface simulation in hfss
I have hard time trying to simulate a metalic rectangular ring slot from 0 to 35 GHz individually and afterwards as a skewed array with HFSS version 8. It seems almost impossible, so since yesterday I'm trying it with the new HFSS 9 version.I'm disperately trying to get the reflection
and transmission coefficient while sweeping from dc to 35Ghz. I'm expecting to see stopband around 32 GHz.
It would be great if somebody out there could send me any kind of information, maybe a related type of example(.prj).
In Ansoft's site there are some references(.pdf) concerning FSS simulation but nothing in particular.
I would appreciate any kind of help.
Yours sincerely
vagee
You need to be more specific on the question. e.g. any error message, difficulty in setting up the skew array, etc.
Dear Loucy thanks for answearing.My main problem concerning simulation of a FSS individual or array element is the post processing.Let me describe what I'm doing.
I'm setting up a parametric PEC rectangular ring on FR4, closing it with radiation boundaries(box)(is this a bit tricky?) and setting up an incident plane wave towards the structure.(0,0,1)
Initially when exciting in v.8 with a plane wave you do not have the ability to sweep over the frequency range which for me would be somewhere around 30Ghz for a 8 GHz span. I read somewhere that the only way to get transmission or reflection over frequency is by writting a macro script which is something far away from my knowledge.Is there any anternate way of getting these coefficients?
This is why I'm looking everywhere to find an already running .prj concerning FSS simulation.
I hope I made myself clear.
vagee
Please use the op packages! In it,you can set frequency variable and sweep it!
Your problem sounds like mine. What I did is the following (HFSS9.2):
I Build an Airbox, to get some distance (>lambda) between the FSS an the model border. You can use Radiation while not having large angles. So if your planewave's k=(0,0,1) it will work. I use PML.
The borders to the sides must then be periodic boundaries. Assign Masters an Slaves, see the helpfile!
In the postprocessing switch the plot to "scattered field" (HFSS->Fields->Edit Sources). The scattered field below the FSS, divided by the incident field (I set it to 1V/m) equals the reflection coeff. Use the calculator to get results (I calculate the mean on arbitrary plane).
In this case you can only make a discrete frequency sweep. I takes long, but works well.
it's very helpful, thanks!