At THz frequencies, the microstrip-line width W becomes of the same order that its thickness t. In these conditions, are CST or HFSS tools able to take into account this restrictions (W very close to t) ?
Simulation codes will have no problem with this, as long as you model the thickness and edge profile correctly, and don't use thin layer approximate models. Doing this will mean it takes longer and uses more memory.
You will also need to model the edge roughness correctly or take account of it in some way, because this will be more significant at wavelengths smaller than 1 mm.
Thank you for your response Mr. M. White,
So, we have to create our own simulation code ?. But do you know if simulation software tools (such as CST or HFSS) can take into account both thichness t and roughness edges at THz band ?
No you don't have to create your own simulation code. I don't know where you got that idea from.
The frequency makes no difference to electromagnetic codes, unless there is something other than dielectric and magnetic material effects. You will get the same result at 10 GHz as you will at 1000 GHz, if you make the model 100 times bigger, and use the same dielectric constant and loss tangent, and magnetic properties too. But when you make it 100 times bigger, surface roughness and the thickness of the track becomes 100 times larger, so you need to remember to model these. An edge that looks smooth at 10 GHz may look zig-zag at 1000 GHz. It is up to you to tell the code what the edge and thickness and taper of the sides look like.
I don't know if the codes have built-in edge roughness routines, but it may be accurate enough if you increase the track edge resistivity to take account of it - there will be an increase is the length of the current path along the edge, if the skin depth is less than the RMS roughness.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Waves