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 The information provided may be out of date.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.

RF port settings to take advantage of model symmetry

Please login with a confirmed email address before reporting spam

I have a model with two ports. I am simulating the transmission of a TE(10) wave. I running a "frequency domain" study to generate S-parameters. The model runs fine, but takes a while to simulate so I am trying to take advantage of symmetry in the model.

The wave propagates in the z direction and the model is symmetric about the xz and yz planes. I've set up a quarter model and assigned the surfaces on the planes of symmetry perfect magnetic conductor boundary conditions.

I can't seem to figure out how to set up the ports to duplicate the results I get when I simulate the entire geometry. (When I simulate the entire geometry I set both ports as rectangular, turn the excitation of the input port on, and select TE with the 10 mode number).

4 Replies Last Post Apr 7, 2011, 10:43 a.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 15, 2011, 9:23 a.m. EST
Hi,


As far as I know, in COMSOL 3.5, once you have "cut" your model and defined symmetry (typically E-plane and H-plane for a rectangular waveguide), then you have to define manually your ports (because the automatic port definition does not work any more once you have cut the model...).

Also, if you are interested in the power, take care that once you have cut your model, you should divide the input power accordingly (1/2 or 1/4).
Hi, As far as I know, in COMSOL 3.5, once you have "cut" your model and defined symmetry (typically E-plane and H-plane for a rectangular waveguide), then you have to define manually your ports (because the automatic port definition does not work any more once you have cut the model...). Also, if you are interested in the power, take care that once you have cut your model, you should divide the input power accordingly (1/2 or 1/4).

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 6, 2011, 12:00 a.m. EDT
Since you are more advanced regarding the problem of simmetry, I would like to know what are the boundary conditions to simulate a fourth of a nanowire where the electromagnetic wave is incident in the axial direction of the nanowire (small cylinder. I have tried to use with this fourth of cylinder the boundary conditions of floquet periodicity and continuity (see figure). How should I apply these conditions?
With respect to what you are discussing, Why should the port be modified manually? and what should be specified?

PS: I am attaching a picture of the fourth of cylinder representing my nanowire. This is not my original simulation. It is only meant for more clarity.
Thanks in advance
Since you are more advanced regarding the problem of simmetry, I would like to know what are the boundary conditions to simulate a fourth of a nanowire where the electromagnetic wave is incident in the axial direction of the nanowire (small cylinder. I have tried to use with this fourth of cylinder the boundary conditions of floquet periodicity and continuity (see figure). How should I apply these conditions? With respect to what you are discussing, Why should the port be modified manually? and what should be specified? PS: I am attaching a picture of the fourth of cylinder representing my nanowire. This is not my original simulation. It is only meant for more clarity. Thanks in advance


Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 6, 2011, 9:35 a.m. EDT
I'm not sure to fully understand your model: is your wire a sort of hollow cylinder?

If yes, I would say that you would need an E-plane and an H-plane, but not knowing the problem...
I'm not sure to fully understand your model: is your wire a sort of hollow cylinder? If yes, I would say that you would need an E-plane and an H-plane, but not knowing the problem...

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Apr 7, 2011, 10:43 a.m. EDT
the problem consists in measuring the absorption of radiation in a silicon nanowire. The incident electromagnetic wave reaches the nanowire (a cylinder in the figure) with "k" parallel to the axis of the cylinder. The wave is TM.

The figure shows the modeling of the problem: a solid silicon cylinder surrounded by air where the "air cubes" at the ends of the figure are Cartesian PMLS. My question is: taking advantage of the symmetrical geometry, How can I get the same results using only one fourth of the actual geometry? How do I do with the boundary conditions? Should I maintain the same boundary conditions? How should these boundary conditions be applied?

thanks!
the problem consists in measuring the absorption of radiation in a silicon nanowire. The incident electromagnetic wave reaches the nanowire (a cylinder in the figure) with "k" parallel to the axis of the cylinder. The wave is TM. The figure shows the modeling of the problem: a solid silicon cylinder surrounded by air where the "air cubes" at the ends of the figure are Cartesian PMLS. My question is: taking advantage of the symmetrical geometry, How can I get the same results using only one fourth of the actual geometry? How do I do with the boundary conditions? Should I maintain the same boundary conditions? How should these boundary conditions be applied? thanks!

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.