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.

Strange pattern within modes

Please login with a confirmed email address before reporting spam

Hi,

I am simulatimg a fundamental WG with some curves. Since WG is operated in its fundamental band, no mode significant conversion or reflection is present ( coefficients below -50 dB). So I expect to see a very clean TE10 ( input mode) propagating in the WG, expect for the attenuation given by Ohmic losses ( copper walls ). The problem is that I see some strange structures inside the maximum and the minimum of the propagating TE10 ( you find a picture in attachment). Do you think these structures are significant or could they be given by numerical errors? Do you think I need to refine the mesh ?



3 Replies Last Post Feb 1, 2024, 9:51 a.m. EST
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 9 months ago Feb 1, 2024, 8:40 a.m. EST
Updated: 9 months ago Feb 1, 2024, 8:29 a.m. EST

It does look like your mesh is too coarse.

Jeff

-------------------
Jeff Hiller
It does look like your mesh is too coarse. Jeff

Please login with a confirmed email address before reporting spam

Posted: 9 months ago Feb 1, 2024, 9:20 a.m. EST

Hi, I tried to use the "extremely fine" mesh in the physics-controlled mesh option but the pattern is still identical, why?

thanks in advance

Hi, I tried to use the "extremely fine" mesh in the physics-controlled mesh option but the pattern is still identical, why? thanks in advance

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 9 months ago Feb 1, 2024, 9:51 a.m. EST
Updated: 9 months ago Feb 1, 2024, 3:05 p.m. EST

My guess is that the "extremely fine" setting results in a mesh that, its name notwithstanding, is too coarse to capture the solution. I figure from your screenshot that there may only be ~3 elements across that waveguide. If I am right about that, you will need to use more subtle ways of specifying your mesh than a global setting. The chapter on meshing in the Reference Manual contains valuable advice and examples. For this particular analysis, you may want to consider using a swept mesh for the waveguide, see this blog post on swept meshes.

Jeff

-------------------
Jeff Hiller
My guess is that the "extremely fine" setting results in a mesh that, its name notwithstanding, is too coarse to capture the solution. I figure from your screenshot that there may only be ~3 elements across that waveguide. If I am right about that, you will need to use more subtle ways of specifying your mesh than a global setting. The chapter on meshing in the Reference Manual contains valuable advice and examples. For this particular analysis, you may want to consider using a swept mesh for the waveguide, see [this blog post on swept meshes](https://www.comsol.com/blogs/improving-your-meshing-with-swept-meshes/). Jeff

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.