Linear patch array design

[Parn]

hi all,

i'm new to the antenna design. i have questions concerning a linear patch array antenna.

my design is 5-elements linear patch array connecting by a transmission line with a coaxial feed in the middle patch.

the thing is the width of the patch on the most left and right sides are smaller than other patches. i was wondering why it 's like this. any suggestion?

one more thing , i'd like to know the effect of the width ,the length of the patch, the length of the transmission line and the position of the feed to the return loss and radiation pattern....

thank you very much.

 

[Higgler]

Sounds like a series fed array, whereby each antenna element radiates a small amount of energy.
Each patch can be identical, so the reason for the different dimensions would depend on which orientation the changing dimensions are in. If you describe the physical layout a bit more, maybe it would give more of a hint what the different dimensions are for.

One similar type antenna is a common home TV antenna mounted on your rooftop, fed with a series line similar to your description, and every antenna is slightly different so that you can use it over a wide frequency range.
Series fed arrays form antenna beams that change location as frequency changes if they have a groundplane behind them, as most patch antenna arrays do.
kch

 

[VEBill]

"...the width of the patch on the most left and right sides are smaller than other patches."

I've seen something similar for other arrays (not of patches) where the outside elements are slightly different than all the others. The reason was to provide better side-lobe suppression. In other words, tune the two outside elements to null out the largest sidelobes.

The above is just a guess based on what I've seen previously.

 

[Parn]

Thank you very much for all of answers.

Btw, is there any suggestion how to optimize this linear patch array to have symmetric pattern and less side lobe level?

This case, it is difficult to make it symmetrical, due to its offset feed.

 

[Higgler]

?offset feed? I don't understand that phrase.

To make a 5 element array have less sidelobes, you need to taper the power at each element (strong center element, weaker end elements). If you use 3 standard power splitters (1x2,1x3,1x2), set the center element strongest, the two nearest to the center are 4.7 dB weaker and the edge 2 antennas are 7.7 dB weaker than the center element. Of course phase matching is required.
The amplitude decrease from center to end automatically provides lower sidelobes and symettrical antenna radiation patterns.
You could try to attenuate the signals in the serial path to provide a similar taper, or maybe that change in dimensions for the outer antennas weakens the outer antennas radiating strength, thereby providing amplitude taper.
kch

 

[Parn]

HI Higgler,

what i mean is the feed is not in the center of the middle patch.(i use a coaxial feed.)

 

[Higgler]

Patch antennas have their feeds offset from the center to match them to 50 ohms usually. Thin patch antenna feeds are close to the center for 50 ohms and thicker patch antennas (0.2 wavelengths) are closer to the outer edge.

Do you need a patch antenna for high volume and thin size? There are alternatives to patch antennas. Patch antennas when exposed to dirt, rain, snow or ice change frequency and hence need to be covered by a protective radome so they don't "change frequencies" due to weather, etc. They are typically narrow bandwidth and temperature can sometimes shift their frequency of operation depending on material, bandwidth and antenna thickness.
The one GPS patch we purchased at my company is mistuned in frequency slightly, but still works fairly well.

kch

 

[azezda]

Hi,

I thing that every element has to be fed by a coaxial line I mean by the same signal to improve the antenna gain and bandwidth

 

[Higgler]

If you have options on how many antennas to use, a 4 or 8 element design is better, (most power dividers are multiples of 2). There's no magic in 5 elements, you could have 4 elements spaced slightly larger, or 8 elements at smaller spacings to get similar gain.

To make a single patch, you can trim it empirically and get excellent VSWR. It's a little trickier in an array. The antennas need to be a bit smaller than a single stand alone antenna. Since the antennas are phased together, they all have to be the same size, so manufacturing tolerances are important.

If you really want alot of data on every antenna available, become an IEEE member and for $108 buy the 2 DVD set of "Transactions on Antennas and Propagations" Society compilation from 1956 to year 2000. Great search algorithms to look up patch antennas. Or stick with Pozar's books.

kch