antenna-theory.com

[aidan]

Hi,
I have been scratching my head over this and I hope an expert can alleviate the frustration.

I use centre fed grid parabolic reflectors as p2p antennas. They are not the full circular reflector but rather a rectangular segment from a fully circular parabolic dish, I'm sure you know what I mean. They have a single polarity feed that can be mounted to give either vertical or horizontal polarisation. Does the fact that the reflector is not fully circular actually affect the performance when vertical or horizontally polarised? The manufacturer does not suggest this is so. They indicate that all that is required to alter the polarisation is to turn the feed through 90deg, rather then rotate the entire antenna. They do however quote a beamwidth of 6deg in the V plane and 4deg in the H plane (relative to what, do they assume you will mount the antenna so that it is wider than it is tall?). My understanding of the theory is clearly very lacking. I can only think about a yagi where the relector is always in the same plane as the driven element and polarisation is clear from the physical position of the elements. I have a feeling that this is not the same with a parabolic reflector? Can someone help clear this up for me.

So basic question. If the parabolic reflector is wider than it is tall will it give more gain in the horizontal plane assuming you orient the feed in the horizontal also? Now, if you still maintain the reflector to be wider than it is tall, but instead mount the feed in the vertical plane, will the gain drop? If so is it possible to use such a none circular parabolic relector to create a dual polarised dish for MIMO purposes?

Thanks
Aidan

[Schubert]

Hi aidan,

If the parabolic reflector is wider than it is tall, then the horizontal plane will have a sharper beamwidth than the vertical plane. The horizontal plane won't have a higher gain, because the peak gain of each plane will be at the same point.

The manufacturer is right to say that rotating the feed will alter the polarization. The source antenna will radiate a field, lets say it's horizontally polarized, and then the reflected field (what is radiated by the reflector) will have the exact same polarization.

As to whether the gain drops or not, it's possible, but hard to say, depending on what type of antenna is the source antenna. If it is a horn, then probably the gain doesn't change much.

Finally - on MIMO: you can't get polarization diversity with a single antenna. Circular polarization does contain two polarizations, but you cannot send independent information with each polarization, so using separate receive antennas to pick up the two polarizations won't help.

[aidan]

Thanks, OK sounds it sounds as if my feelings are correct. If I can just clarify: If the beamwidth in the H & V planes are different does this not mean that the energy if some widely distributed in the V rather than the H and hence at the recieving end of the link the overall system gain will be lower.

Regarding MIMO, actually I realise that sepearte physically diverse antennas are expected, but the systems I use achieve MIMO through different polarisations. One MIMO chain on H plane and the second MIMO chain on the V plane. Yes they have discrete coaxial feeds into the antenna, so they are in fact different driver units, so in effect 2 antennas. However they are not seperate.

So I wish to use this grid we speak about but replace the single feed unit, with a dual polarised feed. Giving a seperate feed for each MIMO chain.

What I am trying to work out if how to create a center fed horn that will give spearte H & V radiation patterns simultaneously.

The ones I have seen for fully circular dish reflectors use a dual feed waveguide mounted on what appears to be a 'champagne glass' center mounted feed.

Any ideas how this is achieved physcially?
Thank again
Aidan

[Schubert]

In regards to your first paragraph:
Thanks, OK sounds it sounds as if my feelings are correct. If I can just clarify: If the beamwidth in the H & V planes are different does this not mean that the energy if some widely distributed in the V rather than the H and hence at the recieving end of the link the overall system gain will be lower.

Let's say your V plane is plotted from -90 to +90 degrees (+90 is straight up, -90 is down), and the H-plane goes from -90 to 90 (-90 is to the right, +90 is to the left). The V- and H- plane intersect each other when both are 0 degrees. Hence both have the same peak gain. So I'm hesitant to agree that the overall system gain will be lower.

For horns with dual feeds, you can use small probes as the source, each 90 degrees from each other. The hard part is getting it radiate well across a large frequency band. The dual feed horns I've seen for anechoic chambers feed across a small band that flares out inside the horn.