On the generation of FDTD mesh using openEMS

How to use openEMS. Discussion on examples, tutorials etc

Moderator: thorsten

Post Reply
Posts: 19
Joined: Wed 22 Sep 2021, 04:27

On the generation of FDTD mesh using openEMS

Post by thinkless » Sat 20 Nov 2021, 15:46

I am reading the instructions on page http://openems.de/index.php/FDTD_Mesh, and have some questions:
(1). At the edge of 2D metal, the mesh line is 1/3 inside and 2/3 outside, so the "metal" constructed with FDTD mesh is smaller than the actual metal, right?
(2). If (1) is the case, can I use the conformal technique to improve the accuracy by modeling the actual metal boundary?
(3). mres is smaller than sres, and the black lines should be set as sparse as possible, but we still have to obey the rule that there should not be an abrupt change of mesh size that's greater than 2 times, am I right?
(4). After the interval between E fields is determined with SmoothMeshLines, we can put the H fields at the midpoints of the E fields, am I right?

Posts: 1520
Joined: Mon 27 Jun 2011, 12:26

Re: On the generation of FDTD mesh using openEMS

Post by thorsten » Fri 26 Nov 2021, 18:04

1) yes
2) I'm not sure what you mean
3) yes, something like 1.5 is even better.
4) the E-fields are located at the center of the edges of the mesh. The magnetic fields at the center of the faces. (see https://en.wikipedia.org/wiki/Finite-di ... ain_method)

Post Reply