Offset on EM wave in antenna simulation

Numerical methods and mathematical models of Elmer
Post Reply
Posts: 5
Joined: 10 Jun 2023, 21:31
Antispam: Yes
Location: Japan

Offset on EM wave in antenna simulation

Post by TommyA »

Hi, I've been trying to simulate a simple dipole antenna with EMWaveSolver in time.

What I expect to get is outgoing electric fields such as in this video.

See the first picture. I simply have a 0.1-meter metal wire along z-axis embedded in the air (sphere). I specified body force (current density rate 3) to be w*cos(wt) at the middle of the wire as a driving source.

Now I mainly have two questions.
  • I did get outgoing radiation as a result. But there is an offset on Ez, as seen in the right panel of the second picture below. The red curve shows Ez on the x-y plane. This offset oscilates in time. I'm not sure where this offset is coming from. I suspected boundary conditions as there are not much else I can suspect in this simple case. The outer boundary should be open. I didn't specify any values in the outer boundary. Not exactly sure how it affects the calculation. Finer meshes, finer timesteps or lower system tolerances, all resulted in more or less the same result.
  • If I set "electric damping coefficient = √ε0*μ0" for the outer boundary, which I expect to be an absorbing boundary, I get "incoming" wave from the boundary to the antenna. So I'm little confused about it as well. Does it happen in some cases?
I also attached a complete case below. Any advice or suggestion is more than welcome.


description.jpg (22.77 KiB) Viewed 917 times
offset_on_Ez.png (70.63 KiB) Viewed 917 times
(660.24 KiB) Downloaded 118 times
Post Reply