Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

S-Parameter of 1-Port Network

Please login with a confirmed email address before reporting spam

Hi at all,
I need a first advice how to define a one-port network in Comsol to simulate the reflection coefficient S11.
Consider a thin film resistor (or micro strip) on a substrate. Both ends are connected via one rf probes to measure S11 and to calculate the impedance of this micro strip. There are no other pads to measure in transmission.
I want to simulate such a setup in Comsol but I’m not sure how to define the ports. In the tutorial section I found only 2-Port networks and they use lumped ports. In the RF User guide I found: “The requirement on a lumped port is that the feed point must be similar to a transmission line feed, so its gap must be much less than the wavelength.” So in my case the lumped port isn’t the right choice.
Is it right to define the left end of the micro strip as port one and the right end as port two and then my reflection coefficient is the transmission coefficient between both ports in the simulation or is there a other way?
Thanks in advance

1 Reply Last Post 14 ago 2014, 19:23 GMT-4
Robert Koslover Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 10 years ago 14 ago 2014, 19:23 GMT-4
If you have a network with two ports (numbered 1 and 2, for example), there is simply nothing that stops you from plotting S11 and ignoring the other s parameters. You can also have just a single active port (which is actually quite common in antenna problems) at which you can compute S11, while letting the power radiate into either scattering boundaries or PMLs. So you really do not have to do anything like trying to extract S11 from S21 indirectly.

If your lumped port is not small compared to a wavelength, then you need to compute the fields there instead of using a lumped port. If it does not fit one of the other options (e.g., a waveguide-type or coaxial-type port) then you can always compute the fields at your port using the "numeric port" feature, which involves solving for the fields at the port before using them as a boundary condition in the larger/overall problem. There's an example in the RF-module library of a circular to rectangular waveguide transition in which this technique is used.

If you have a network with two ports (numbered 1 and 2, for example), there is simply nothing that stops you from plotting S11 and ignoring the other s parameters. You can also have just a single active port (which is actually quite common in antenna problems) at which you can compute S11, while letting the power radiate into either scattering boundaries or PMLs. So you really do not have to do anything like trying to extract S11 from S21 indirectly. If your lumped port is not small compared to a wavelength, then you need to compute the fields there instead of using a lumped port. If it does not fit one of the other options (e.g., a waveguide-type or coaxial-type port) then you can always compute the fields at your port using the "numeric port" feature, which involves solving for the fields at the port before using them as a boundary condition in the larger/overall problem. There's an example in the RF-module library of a circular to rectangular waveguide transition in which this technique is used.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.