Why is my radar return NOT symmetric ?

JR_22148

I was trying to study the amplitude and phase response of  a dielectric sphere to L band ( ~ 2 GHz) radiation.

The center of the sphere is at the origin. The radar  positions were  at 45 deg elevation  and were distributed in azimuth 5 degrees apart.

I was expecting to see a symmetric response in azimuth but it is not the case. Please see attached polar plot. I am also attaching the .cfx file and the  .out file for your inspection. Any suggestion , observation ? Thanks.

wooden_ball.cfx

wooden_ball.out

wooden_sphere_2GHz.png

Torben Voigt

Hi JR

1. You have chosen RL-GO as solver, which is a high frequency solver. This means that it should be used for electrically very large models. For asymptotic solvers like RL-GO, PO, LE-PO or UTD, the size of the model should be > 20 lambda. In principle, the larger the model, the more accurate the results. Your sphere is only 2/3 lambda with 1 m radius and thus much too small for asymptotic solvers. You should just use MoM, the simulation time for such a small model is much shorter than with RL-GO.

2. you could force a symmetric mesh. With full wave methods like MoM, FEM etc. you get 100% symmetric results, with RL-GO at least approximately.

I attached two models:

wooden_ball_alt.cfx (RL-GO, increased size to 20 lambda and symmetric mesh)

wooden_ball_alt2.cfx (MoM and symmetric mesh)

Hope this helps!

 

Best regards,

Torben

wooden_ball_alt.cfx

wooden_ball_alt2.cfx