Electromagnetics: Discontinuous Galerkin and Multiple Bodies

Numerical methods and mathematical models of Elmer
Post Reply
tylerm
Posts: 5
Joined: 09 Jan 2023, 19:00
Antispam: Yes

Electromagnetics: Discontinuous Galerkin and Multiple Bodies

Post by tylerm »

Hi,

I'm running into an issue in trying to calculate accurate inductance values for a coil in the presence of various materials. As a first step, I've set up a simple case with the geometry in a bounding box and have found a combination of boundary conditions and solver configurations that converge well and appear to provide passable results (within 10% of benchtop at 10E-8 convergence) using L = 2*E/(I^2) with GCR and Discontinuous Galerkin (based on this post: https://www.elmerfem.org/forum/viewtopi ... 1&start=10).

When I try to increase the complexity of the model to 3 bodies (coil, variable material set to air for testing, and boundary) this approach seems to fall apart. While the calculated magnetic field strength is similar, the current density through the ends plummets and results in wildly inaccurate results. I've tried playing around with the mesh fineness/quality and the general trend remains the same. Is there a limit for the number of bodies that the discontinuous galerkin method works with? In a related manner, is there documentation describing the internal variables available and what the difference is between the vector and scalar current densities (as well as AV potentials for that matter)?

.SIF files and the simple mesh are attached.

Thanks!
Attachments
coil_sims.zip
(182.32 KiB) Downloaded 24 times
tylerm
Posts: 5
Joined: 09 Jan 2023, 19:00
Antispam: Yes

Re: Electromagnetics: Discontinuous Galerkin and Multiple Bodies

Post by tylerm »

Here's a coarse version of the "complex" mesh to fit the file size requirements:
Attachments
complex_mesh.zip
(345.71 KiB) Downloaded 25 times
Post Reply