Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

How to simulate GVD using HFSS

Status
Not open for further replies.

x529

Junior Member level 3
Joined
Dec 14, 2019
Messages
30
Helped
0
Reputation
0
Reaction score
0
Trophy points
6
Activity points
265
How to simulate GVD using HFSS?
 

Attachments

  • 微信图片_20220408221545.png
    微信图片_20220408221545.png
    139.3 KB · Views: 112
  • 微信图片_20220408221537.png
    微信图片_20220408221537.png
    151.6 KB · Views: 109

Are you obtaining the data from a dispersion simulation, or a driven simulation? The dispersion will directly give you Beta as a function of frequency, but if your structure isn't continuous or periodic this wouldn't be the most appropriate.

You can extract Beta from the scattering parameters of a driven simulation through assuming homogenization (i,e, convert the scattering parameters to something like ABCD parameters, and Beta will be imag{acos(A)/l}, where l is the length of your homogenized line).

You can take the derivative in either of these cases by using the "deriv" function in the output variables (HFSS->Results->Output Variables)
 

    x529

    Points: 2
    Helpful Answer Positive Rating
Are you obtaining the data from a dispersion simulation, or a driven simulation? The dispersion will directly give you Beta as a function of frequency, but if your structure isn't continuous or periodic this wouldn't be the most appropriate.

You can extract Beta from the scattering parameters of a driven simulation through assuming homogenization (i,e, convert the scattering parameters to something like ABCD parameters, and Beta will be imag{acos(A)/l}, where l is the length of your homogenized line).

You can take the derivative in either of these cases by using the "deriv" function in the output variables (HFSS->Results->Output Variables)
Thanks for your apply. I tried to use group delay results to calculate GVD in HFSS.
The group delay dispersion (also sometimes called second-order dispersion) of an optical element is a quantitative measure for chromatic dispersion. It is defined as the derivative of the group delay, or the second derivative of the change in spectral phase, with respect to the angular optical frequency:
(https://www.rp-photonics.com/group_delay_dispersion.html)
group delay dispersion (GDD)


For two-port networks/devices, HFSS has group delay result, group delay(2,1).
GVD=GDD/distance,
(1) To get group delay in HFSS. Group delay is a function of frequency f.

Using Matlab/Python for data processing
(2) GDD is defined as the derivative of the group delay.
To get GDD: Solve for the partial derivative of group delay. w=2*pi*f, then GDD is the partial derivative of group delay with respect to frequency.

(3) To get GDD: Divide by length.

Solving GVD Using Partial Derivatives of Group Delay;
I think this approach is correct in physically, but I'm not sure this is the right approach;
Looking forward to your reply.
 

I wasn't aware that HFSS had a built-in group delay function, shows what I know! Thanks for pointing that out.

Your statements looks fine, expect I think item (3) should read: to get GVD: divide GDD by length. This is because GDD measures the second derivative of phase, whereas phase is Beta x Length, and GVD measures the second derivative of Beta.
 
  • Like
Reactions: x529

    x529

    Points: 2
    Helpful Answer Positive Rating
I wasn't aware that HFSS had a built-in group delay function, shows what I know! Thanks for pointing that out.

Your statements looks fine, expect I think item (3) should read: to get GVD: divide GDD by length. This is because GDD measures the second derivative of phase, whereas phase is Beta x Length, and GVD measures the second derivative of Beta.
Thanks for your apply. I think you are right. I'm glad I learned something new
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top