Problem in HFSS as axial ratio changes not symmetrically to the change of the airbox

Hi

I have problems regarding the axial ratio for a circularly polarized waveguide antenna in hfss.
As i have always done for waveguide slot antennas , i choose an airbox of λg/4 of height and use PML boundaries.
with this boundaries, the structure was optimized for an axial ratio of 0.1 dB.
but now, as i change the dimesnsions of the box, the results change and if it's not complitely symmetrical, the axial ratio riser up to 13 dB

Could anyone help me ?

thanks
saras

hi,

Adding to this question

is there any general formula or design for optimization of axial ratio in a circularly polarized microstrip patch antenna?