PML in HFSS, Proper port to get S21

[monir]

Hi

I have a simple rectangular wave guide filled by air. I want to apply PML for boundaries in x and y direction and electric wall for z direction boundaries. and then take the S21 of of the structure which should be 1 since wave guide filled by air.
There are some tutorials on how assign PML in HFSS but they are applying plane wave incident wave and calculation the RCS. However I need to have 2 ports to calculate S21 and I can not use waveport since I have PML.
I also have tried lumped port, it is working but it has wring results. (S21=0.3 instead of 1)
could anyone guide me what should I do to build this simple structure and get S21?

Best
Monir

 

[PlanarMetamaterials]

Hi Monir,

I'm not sure what you are describing for your setup - is the waveguide along the z axis? So you're terminating the waveguide on either end in PMLs?

If this is the case, you don't need/want to use PMLs to get the scattering parameters. Simply apply a waveport to a rectangle on either end of the waveguide and ensure you are exciting the desired mode.

Good Luck

 

[monir]

Hi

This is a test set up to see if the PML works or not. I have an air wavegide along x axis. If I consider electeric wall (perfect E) in top and bottom (boundaries in z) and magnetic wall (perfect H) in y boundaries and 2 waveport at both end I could get S21 =1 becouse all energy will transfer. Now I consider PML at y and x boundaries and remove perfect H, it should still have S21=1. The problem is that in the location of ports I have PML. So I can not use wave port. So how can I get S21 of the structure.

Best

Hi Monir,

I'm not sure what you are describing for your setup - is the waveguide along the z axis? So you're terminating the waveguide on either end in PMLs?

If this is the case, you don't need/want to use PMLs to get the scattering parameters. Simply apply a waveport to a rectangle on either end of the waveguide and ensure you are exciting the desired mode.

Good Luck

 

[PlanarMetamaterials]

Now I consider PML at y and x boundaries and remove perfect H, it should still have S21=1

Why? With the PML, the structure now has 2 "open" surfaces. Saying that this structure should not leak power does not seem correct to me. I assume you're exciting a TEM PPW mode?