Adaptive Mesh Refinement for 3D

[abeer.vaishnav]

Hi everyone,
I'm trying to uae ElmerFEM for EM simulations to enable simulation and analysis of superconducting microwave circuits as a part of my work at IBM Quantum for superconducting qubit design. Mostly my work includes integrating Elmer with our quantum-EDA software -- Qiskit Metal (https://qiskit.org/metal). I'm making use of the Electrostatic solver for capacitance extraction, and Vector Helmholtz solvers for E-field and H-field eigenmode simulation for our use-case.

Right now I'm using Gmsh for geometry generation and post processing. I am facing an issue with the Adaptive Mesh Refinement (AMR) feature in Elmer as it turns out that it's only available for 2D right now. I tried searching the whole forum and documentation and couldn't find any solutions for using Gmsh or Netgen with Elmer to do AMR using the first order field calculation results. AMR is the only thing that is stopping me to use Elmer instead any other proprietary tools and I'd appreciate if someone can help me find any examples of using the generated 'bgmesh' when we set `Adaptive Remesh = True` in Elmer, and specify it a background mesh density field in Gmsh or Netgen to achieve AMR.

Thanks!

[Rich_B]

Hello,

You may want to look at this wiki post:

http://elmerice.elmerfem.org/wiki/doku. ... ms:calving

The test case uses MMG3D.

Rich.

[raback]

Hi

There is no AMR for this equation in 3D. Rich is right that one avenue for this would be to use MMG3D. This feature is on the long list of things to do but unfortunately there are also many other things on the list and this one is not completely trivial...

-Peter