Panel antenna for 1090 MHz

Recommended Posts

Hey!

While playing around with ADS-B, I was thinking about boosting the range of detection by building my own panel antenna.

Finally it should look similar to this one: http://www.activefrance.com/Antennas/_wp_generated/wpc12cc178_1b.jpg

The size of the single patches can be calculated from the center frequency:

L = λ/2, so in this case:

f = 1090 MHz --> λ = 0.275 m

and thus resulting in a L of 13,75 cm.

A DIN A2 sized circuit board could carry 24 patches.

As the ADS-B signal already contains information about the location of the signal, it is not necessary to point the antenna always to one direction. Thus finally I want to mount the antenna onto a rotating motor, collecting data from all directions and plot the data with dump1090.

I'm still not 100 % sure how to construct the panel. Maybe I could do it on my own using a patch mask and cutting cooper foil or I'l ask some guys that have access to a PCB milling machine. In the last case the size of the panel might be a problem so I probably should think about a modular design and finally plug everything together.

What do you think?

Share on other sites

First off, don't cut my foil, you bastard!

FWIW, the design looks like something I read about using the term "Fractal Antenna" and if you go to that page your design is very close to what they define as a planar array fractal antenna. I think with intricate designs like this your best bet is by etching the design onto a PCB.

Edited by Cooper
Share on other sites

Haha probably I should change to some gold sputtered silicium wafers, eh?

Yes I guess so that etching would be the best option. However, the more parallel patches you want to run on the panel the bigger the size of the antenna will be. Don't forget that we are not talking about single patches of 1 cm but nearly 14 cm ! I wasn't going to fill my bath tub with HCl/H2O2 although it would be a very cleaning bath.

Thus there was the idea of just making some masks, cut the foil and glue it onto some substrate. Connecting all patches with a wire and there we go.

The only thing to consider might be that all patches are in-phase to eachother and when the signal is mixed together there are no cancelations. This should be avoided by chosing the right lenght of spacers. I'm not sure whether it really matters as the speed of the signal in relation to the frequency could be so much faster that in the range of several cm the amplitude of the signal peaks are not shifted sooooo much.

Share on other sites

Let me just provide you with this pretty compelling article on fractal antennas where the guy goes from a bowtie antenna to a triangular fractal antenna and as he progresses the quality of the measured signal degrades.

My recommendation would be to first try to make your own

. Darren made one of those aswell somewhere in S17 I believe and I think he mostly failed because he just _had_ to put it up on a drone...

Two other options exist.

1. Since your panel antenna is directional, why not just make a biquad? Much easier and probably better at receiving.

?

Since biquads use quarter wavelengths you end up with a diamond shape where each side is just under 7cm at 1090MHz. That sounds to me like something considerably more manageable...

Edited by Cooper
Share on other sites

Thanks for the input Cooper. Those fractal antennas look interesting. When I searched them on youtube I found an interesting video of 'spraying' antennas:

So I just need to buy some graphene (or cooper :P ) spray, print a mask, put in onto my substrate and happy spraying. Awesome!

What I want to achieve is to increase the sensitivits due to the array design. From a theoretical point of view this makes sense, from a practical one ... it depends. As Darren said in one of the episodes, the horizon is the limit. Its more or less just an experiment, playing around.

What I already did is forming kind a colinear antenna by just bending wire. It was easy to build (just bending some wire) and looks like:

http://martybugs.net/wireless/images/collinear_bare.jpg

The antenna works quite well receiving signals from up to 150-200 km (ca 100m above sea level, no mountains around).

The biquad also looks interesting, I wonder if it is more effective than a patch array.

Share on other sites

• 3 weeks later...

Update:

I tried to build a first version but so far without success. I had a wooden plate and fixed paper on it. I sketched a patch mask onto it and cut out the patches with a razor blade.

I bought some cooper spray (500 ml around 7€) and sprayed it over the mask - until here everything worked nicely.

Unfortunatly the cooper spray seems to contain some stuff (maybe silicon oil) that isolates the cooper particles from each other as I cannot confirm conductivity of the surface.

Maybe I will repeat the experiment at some point of time.

Share on other sites

My first thought would be to heat it so you can burn off the oil, but the wood would probably have issues with that idea.

Are you even sure the copper spray sprays actual copper as opposed to paint with a copper-like color?

Where did you find this design anyway? It looks... different.

Share on other sites

I guess heating is not an option as the spray is kind a resistant to heat (as this is its job), but of course there would also be trouble with the wood.

Yes there is cooper inside, the spray is even for increasing thermal conductivity. The main idea for the spray is greasing high temperature car parts like certain screws.

I used this calculator here: http://www.pasternack.com/t-calculator-microstrip-ant.aspx

and to get into the theory I tried to work me through chapter 14 "Microstrip Antennas" of "Antenna Theory Analysis and Design by Balanis". There is also an instruction for an exemplary design of a single patch antenna and of course a lot of optimization stuff, different designs and also about how to design an array from multiple patches.

There are ways to get the ebook as pdf on the internet.

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

×   Pasted as rich text.   Paste as plain text instead

Only 75 emoji are allowed.