Here is a question, I simulated a grounded co-planar wave guide (conductor-backed) on HFSS with different surface roughness, 0.3 um, 0.6 um and 0.7 um (only the top layer of the copper, dielectric side and two other side remains 0 um).
Here is how I did it, I selected the face of the transmission line (copper box), then set the roughness in assigning finite conductivity, didn't set the thickness, chose the material as copper. (as shown in attached file)
But the s21 is overlapped (as shown in the file), Did I set this wrong? or the top surface roughness of the conductor doesn't effect the insertion loss?
Hello
Instead of selecting the face of the transmission line (copper box) and you should make separate geometry(layer) and set small thickness as in real liefe thus the S21 will quite different as you get it.
The top layer roughness only has has a small effect on the losses - look at where the current is concentrated. The simulation should let you do that. Also, what frequency are you using - is the skin depth bigger than 1 micron?
Levon Odabashyan ,,,, Could you plz be more specific on this? did you mean creating more than a lot boxes to create rough face?
Malcolm White ,, Thanks, I plotted J-Surf of both bottom and top side of the copper, then I found out that the current distribution on bottom side is much larger than the top. My simulation is on 77GHz, skin depth is much smaller than 1 micomillimeter, around 0.2-0.3 um.. I think...
Hayrat Akmj if you will set the thickness you will et different result, I suggest you set thickness
- Badges
- Science topic
- Similar topics
- Filing