Description: I am trying to model an antenna above an infinite soil plane, with small objects embedded inside the infinite soil plane, to see the impact of inhomogeneous soil on the behavior of the antenna. The Manual and other forum questions I've read suggests doing this with an infinite half plane (exact Sommerfeld integrals) with cuboids of dielectric embedded below.
At first my tests went well and I did not notice anything wrong with large embedded blocks of 1m+, but now I am coming up against the case of a small 10cm x 10cm x 10cm cube embedded in the infinite plane. To verify my results I ran three tests:
1) A cube of identical material to the infinite soil plane
2) A cube of slightly varying conductivity to the infinite soil plane
3) A cube of dielectric material with the same properties as air
I expected to see essentially no change from the infinite homogeneous case for case 1, and a minimal change for case 2, and possibly a larger change for case 3, however instead I saw almost identical deviations from the infinite homogeneous case for all three cases. My first thought was that some kind of numerical convergence issue was behind this, so I increased the mesh density of the cube, but I'm getting the same results. I have tried positioning the cubes at different locations and distances from the antenna, but keep getting similar results.
Does anyone have any suggestions on what I could do to improve my simulations to make the identical material case closer to the homogeneous case and separate out the impact of varying soil types? Is this a numerical limitation of FEKO, or a setting issue of some kind?
Product/Topic Name : FEKO v2026.1
