Why S-parameters are passive in fdtd simulation but not passive in Interconnect?

MGURMGUR Member Posts: 1


I did a 4-port S-sweep in FDTD and exported the S-parameters to INTERCONNECT format. The S-parameters are passive according to the diagnose from S sweep in FDTD. However, when using it in 'Optical N-port S-parameters' element in INTERCONNECT and enable the passivity test, it shows the warning 's-parameters failed the passivity test', and the ONA gives different spectrums with and without 'enforce' enabled.

The fdtd S passivity plot:

INTERCONNECT |S41|^2 Results:

What's the reason causing this mismatch? and which one is more reliable? The exported S-parameter file is attached.

Any help would be appreciated.

Best Answer

  • GWANGGWANG Posts: 112Ansys Employee
    Accepted Answer

    Hi MGUR ,

    Thank you for attaching the figures. As long as the s-parameter measured in INTC matches the result in FDTD, I wounldn't worry about forcing passivity. The results are passive to me, you may just use in in INTC as it is.


  • GWANGGWANG Posts: 112Ansys Employee
    edited November 10

    Hi MGUR,

    Unfortunately we cannot open any attached files here. What is the unit for the y axis for the second plot? I believe for the first plot, the y axis is the power transmission? FDTD usually reminds you when the s-parameter is not passive, then there is an option to force passivity. It is also this option in INTC, and that's when you use the "force" passivity option, the power transmission will be forced to be lower than 1. If you see any warnings like the one you attached, you may want to force passivity in INTC.

  • MGURMGUR Posts: 4Member

    Hi GWANG,

    Thank you for replying. Actually I was simulating a Silicon directional coupler with 4 ports:

    So the first plot in the last post is the S parameters passivity plot from 'S diagnostic' of 'S-parameter sweep', the fdtd file is just too big to be uploaded. The whole plot window is here:

    For the second plot, it's the power transmissions at 'port 3' of the simulated coupler using exported S-parameters in INTERCONNECT simulation like this:

    And, yes, I was using 'force' passivity option in INTERCONNECT. But I'm just confused that why S-sweep in fdtd shows that S-parameters are passive but it failed the passivity test in INTERCONNECT. I attached the S-parameters file again in txt, you may be able to open it this time.

  • GWANGGWANG Posts: 112Ansys Employee

    Hi MGUR ,

    Based on ITAR, we cannot open any attached files here. Could you please have a check on the S-paeameter generated in FDTD itself then show the plot here? If the power tranmission (S^2) is not showing any gain, it won't trigger the passivity warning. It will usually remind you when you export the file that the s-parameter is not passive.

  • MGURMGUR Posts: 4Member

    Hi GWANG,

    The s-parameters plots (both using Abs^2) from FDTD are here: S11 and S21 are close to zero (<1e-8)

    Zoomed plot for S31 and S41:

    |S|^2 should be smaller than 1 according to the plots from FDTD. And actually, in INTERCONNECT, the ONA gives the same S31 and S41 results when 'force' is disabled as the following figure:

    Thank you.

  • MGURMGUR Posts: 4Member

    Hi GWANG,

    I got it, thank you so much for helping on this.

    I'll use it with 'force' disabled in INTERCONNECT in this case.


Sign In or Register to comment.