Modeling dielectric lens in FEKO

shan25_22144

Hi, 

 

I am tried simulating an dielectric lens (attached )  using varies solvers in FEKO.  When I increased the diameter of the lens to be greater than 15 lambda I run into memory issues (I have a 256GB RAM computer), out of core computation also failed. I don't want to use RLGO as it is not very accurate. I tried symmetric planes in FEM simulation but no success.  Can you please  suggest some solution for this problem.  I have a FEKO university license, when I run FDTD solver I get this error message 'The voxel mesh is currently limited to less than 2 000 000 000 voxels' is there a possibility to use symmetric planes in FDTD mesh?

 

Thanks!

 

Manu

Unable to find an attachment - read this blog

JIF

Hello shan25,

 

I would try MLFMM (without stabilisation at first)  with two planes of symmetry (even though symmetry won't help much for the MLFMM, since it will fall back to geometrical symmetry). I did notice that your symmetry planes are not defined correctly for your source - you need to fix that (the one should be a magnetic plane).

 

I can't think of any other approximations to use, if you don't want to use RL-GO.

 

The voxel mesh limit is not just for students and symmetry is supported for FDTD.

shan25_22144

Hello Jif

 

Thanks for your reply. Can you let me know how to define symmetry planes in FDTD. When I activate FDTD solver in solver settings, symmetry button in solve/run tab gets disabled.  

 

 

JIF

Hi
It is disabled because symmetry is not currently supported for FDTD.