Lumped port implementation for electromagnetics

1

Hi Fenics community!
I’ve been looking into using FEniCS for full-wave electromagnetics simulations for around a year now and am wondering about interest and prior implementations. The main feature that is missing for me to be able to use FEniCS for my job is an implementation of lumped ports like those available in Comsol and Ansys that would enable the easy addition of lumped circuit elements into the simulation. There is literature on how to implement this in FEM systems as shown in The Finite Element Method in Electromagnetics by Jian-Ming Jin (Section 12.9.1 if you have the book).

I have the interest, time and possibly the ability assuming help from friends and people in the FEniCS community to document and implement this within FEniCS in the next year or so. My question for the community is – to what extent has an electromagnetic lumped port implementation been worked on already and is this an area of interest for other people? If so, are there any demos that show this or any other people who are currently working on this?

1 Like
2

Hi, @awarru,
I’m not going to implement lumped ports in Fenics right now, but I am interested in the topic. Could you please advice some materials (except Jin book) where details of lumped port implementation in FEM are discussed? In particular, I don’t quite understand how we move the known E field at port to the right-hand side.

Thank you in advance, I wish you success in your future work )

3

Hello guys,
maybe I will chime in. During my PhD few years ago I did implement something comparable to COMSOL’s Port based on exactly the book @awarru mentions. You might look at GitHub - bchaber/femtem: Demonstration of a process of solving complex-valued Helmholtz equation in FEniCS 2018.1.0 to see WaveguidePortBoundaryCondition defined in emw.py.
My memories are fuzzy at the moment, but I remember using it for exciting RF devices with 50 ohm coaxial ports.

4 Likes