Cross-pol component in offset reflector geometry

[FMeyer]

Hi all,

I would like to understand in more detail the physical reason, why an offset in a reflector antenna leads to an increased cross-pol component. I look at linear polarized antennas. For sure the asymmetric illumination of the reflector is the cause for the cross-pol component. My question is whether the asymmetry in amplitude or in phase is the main reason for the cross-pol component? (I'm not even sure whether it this distinction is possible.)

The background for my question is, that if I have a high control over the feed pattern, shall I mainly try to keep the amplitude or the phase constant over the reflector in order to reduce the cross-pol component.

Thanks a lot in advance

 

[VEBill]

Are you referring to an offset feed?

Offset feed parabolic dishes are absolutely ubiquitous in satellite TV reception applications. Most are CP (RHCP/LHCP used in DSS), but there are some that are linear (e.g. Shaw Direct, formerly StarChoice in Canada, FSS).

Conceptually, an offset parabolic dish is (typically) simply an oval cutout offset subsection of a larger parabola. In short, it's just part of a full parabola. So I'm not sure why the offset dish itself would affect cross-polarization ratios any differently than would a full reflector. Conceptually, one is a subset of the other.

The feed itself must be optimized to illuminate the reflector of course. There could be some complications in that technology; those details are beyond my experience.

Dish illumination is typically about -10dB at the edge. Years ago I read about the derivation of this rule of thumb. It's related to the overall efficiency and the balancing of full fill (using the entire dish surface) versus spilllover (due to feed beamwidth roll off).

If the feed phase varies over the surface, then (conceptually) one might bend the dish surface to bring the far field waves back into phase. The point of this is to highlight the geometry of the parabola. Typically, dishes are accurate parabolas to lambda/10 or better. This implies that phase compensation isn't often a consideration.

Hopefully all this is related to your actual question.

 

[VEBill]

In reading what I've just posted above, the following thought occurred to me.

The feed for an offset dish will presumably have higher spillover on the edge close to the feed. Higher than the usual -10dB.

This spillover would then 'bounce around' and some small fraction of it would thus impact cross-polarization ratio.

I wonder if that secondary effect of offset feeds might have something to do with it?

Some absorption in the right area might claw back some of the rejection ratio.