# COMSOL - Implementing Perfectly Matched Layers (3D)

If there are any COMSOL gurus here, I want to simulate a 3D RF system infinite in 1 direction. I realize I need to use PML's, but am unsure how to configure them. it seems that simply scaling them by lambda isn't enough - anyone know if I need to configure the material itself in a special way? Thanks!

EDIT: I want to solve a problem containing a infinite (in +z direction) cylindrical waveguide filled with vacuum with PEC walls. How can I replace (some of) the infinite waveguide with a PML?

-
It doesn't sound like your question is specific to COMSOL. I recommend posing your question in the most general possible way, e.g., referring to the equation you're solving rather than simply "3D RF". For example, are you interested in solving 3D time-harmonic Maxwell's equations? – Jack Poulson May 19 '12 at 20:50
I am not sure I understand your comment- I am specifically interested in the COMSOL configuration of PML's. I labeled it with 3D because the 2D setup is quite different (and simpler), while it's important to note that I'm using the RF module, not acoustics or such. – Hopeful May 19 '12 at 21:21
My point is that this is just mathematics. A reasonable number of people on the SciComp SE are familiar with PML (myself included), but you are unnecessarily limiting your audience by only asking about COMSOL parameters, as opposed to the underlying equations. – Jack Poulson May 19 '12 at 21:24
Thanks, but I don't think I wanted to "mess" with the equations too much, hoping I could configure COMSOL to take care of this on its own (as I thought it could). However, I edited my question with a more detailed description. – Hopeful May 19 '12 at 21:29
There is someone in my group that uses COMSOL for E&M simulation, and PMLs are built in, I believe. I myself don't use it, so I can't say anything more. – Victor Liu Jun 17 '12 at 10:18