You can import S-para,eters model in HFSS simulation. Then, without any extra linear network analysis you can see the frequency domain result. Build your design, click on "Circuit element", create multi/single terminal, browse your .S2P file. You can also see S21 and etc. Connect the port to the model. Sometimes the ports that you assign are hidden. Do not worry, just rerun the simulation and see the results.
Watch this video for better understanding: https://youtu.be/0wZ8W_Dxows
Adam Weidling I don’t know their design. Maybe to define two ports just to explain the concept.. Just define ports and assign them to terminal. It should work.
Milad Eyvazi Hesar Thank you! I may have gotten it to work just by defining two lumped ports and creating s-parameter object for them! The model is running now! Fingers crossed! :)
I think you need the connection to ground as the reference for the S-parameters. I would suggest you try it both ways. You can also simulate just the S-parameters block in HFSS and confirm the results match what you expect.
Adam Weidling there is another way you can use HFSS with S-parameters which I personally prefer to importing the S-parameters in HFSS. This involves creating an HFSS with circuit ports where the S-parameter model will be placed, simulating the HFSS design, then importing the model in Circuit (also in Ansys). With Circuit you can place the S-parameter file between the circuit ports from the HFSS file and simulate to get the combined result from HFSS + S-parameter file.
I tried this with a microstrip line (see the first picture, with a gap and circuit ports on either side of the gap, and two wave ports deembedded to edges of the gap). I simulated and used this model with an s2p file of a bandpass filter in Circuit, as well as the s2p file without HFSS (see second picture). The results seem to match for both cases (see third picture) which suggests the method works quite well.
I also attached the HFSS/Circuit project and the s2p file of the filter for reference.
Adam Weidling I just noticed that my integration line directions are opposite for the circuit ports, it seems these must point in the same direction to get the correct transmission phase.