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 The information provided may be out of date.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.
Discrepancies using periodic ports at off-normal incidence
Posted Nov 5, 2013, 2:16 p.m. EST RF & Microwave Engineering, Modeling Tools & Definitions, Parameters, Variables, & Functions, Results & Visualization Version 4.3b 3 Replies
Please login with a confirmed email address before reporting spam
When performing 2-D full-field simulations of an opaque periodic structure at off-normal incidence, my results (near- and far-field profiles) vary as a function of the height of the upper (air) region. At normal incidence, however, there is good agreement.I know this topic has been addressed, but I have accounted for the solutions presented previously, such as meshing, diffraction orders, and reflections.
Model included. My BC's are Floquet PBC on the sides, impedance BC on the bottom (aluminum substrate) and a port with diffraction orders on the upper boundary.
At an angle, the phase distribution would shift as I change the height of the air port--is this the source of my disrepancy? I have tried adding a phase shift such as ky*delta_h, where delta_h is the change in height from a baseline model (height of 4*period), but the two results still differ significantly.
Any thoughts?
Thanks for your attention!
-ado
Attachments:
Please login with a confirmed email address before reporting spam
The problem is that you have defined the "Port phase" on the periodic port as:
phi_shift = {see variable definition} = exp(-j*kaz*delta_h)
This should be the phase angle as seen from the desired unit [rad].
Change the variable definition to;
phi_shift = -kaz*delta_h
and it will work.
--
Magnus
Please login with a confirmed email address before reporting spam
Please login with a confirmed email address before reporting spam
I thought I would post the final solution to this issue, which was presented to me by the online support team. My default angle of incidence was set to theta=0, and so my diffraction order calculation only accounted for the +1 & -1 orders. I needed to include the orders at the full angular span of the model, which, in my case, added a very significant -2 order. Now, incorporating the phase advance into the port and diffraction order phases properly gives identical solutions at any port length.
Another problem that occured was that a strong field appeared near the ports once the diffraction orders were added. Apparently this was solved in the most recent patch (v.4.3.2.189), while I am at v.4.3.2.164.
Thanks, support team!