Radio For Everyone New Posts: Building an ADS-B Station, Easy Homemade Beginner ADS-B Antennas

Akos the author of the radioforeveryone.com blog has recently added two new articles to his blog. The first post is a comprehensive guide to setting up your own ADS-B station. The guide focuses on creating a system that is easy to use, has good performance and is value for money. In the post he shows what type of computing hardware is required, what software can be used and what RTL-SDR dongles work best. He also shows what choices are available when it comes to amplification and filtering to improve signal reception and goes on to talk a bit about adapters and the antennas that work best for him.

BuildingADS-Bstation
Building a ADS-B station

 In the second post Akos shows more on how to build your own beginners antennas for ADS-B reception. The post focuses on showing how to modify the stock magnetic mount antenna that comes with most RTL-SDR dongles, and how to build a half-wave ‘spider’ antenna entirely out of coax cable. The post is full of easy to follow images which make it great for beginners.

EDIT: It’s been pointed out in the comments by antenna experts/enthusiasts that the 1/2 wave ground plane antenna described by Akos in his tutorial may not be technically correct. A 1/2 wave antenna has a huge impedance which requires some sort of matching. Without matching there is going to be about 10 dB of loss due to the mismatch, and so the antenna will perform poorly. We recommend sticking with a 1/4 wave design, which is essentially the same as Akos’ 1/2 wave ground plane antenna, just with the element lengths halved.

26 comments

    • snn47

      The antenna you refer to is in principle 3 stacked ½ antennas. But as far as matching to the coax cable is concerned still ½ wave antenna.

      This design was intended for terrestrial use. Due to the identical dimensions/phasing shown, the stacked ½ dipoles this design will provide a beam-width with a horizontally increase in gain. Most the aircraft will fly and be received at higher elevation angles, which is why the performance of this design should be worse, compared to a ¼ antenna on 3 or more ¼ wave wires or a larger ground plane /counter poise.

      The elevation of the main beam of antennas purposely built for aviation have somewhere between 3° to at max 6° of elevation, and allow together with higher lobes operation to about 40° to 50° in elevation. Above that most antennas have a so called cone of silence where it is unlike that you get very often contact, because of antenna lobes with minimal gain. Design of such anennas will require seperate phasing of the antenna elements to achive the desired elevatedd antenna bean and gain.

      Due to the high frequency, the width of the wire and if a pcb is used width length and the substrate characteristics have to taken into account to adjust the w/h of the will require adjustment of the dimensions, this summary gives you an idea that it is nut just the legth that counts http://www.jpier.org/PIERB/pierb35/12.11091102.pdf .

      ½ wave dipoles have a impedance of a bit above 200 ohm, which is why you cannot connect a wave dipole directly to a low impendence coax cable (50 to 75 ohm), without matching. The matching to the ½ wave J dipole is done with the ¼ stub at the bottom.

      Without adjustment of measurements, accounting for the wire dimension, matching the antenna impedance to the coax cable impedance,most people will, unless they are extremely lucky, only have placebo-gain/-performance nothing else.

      • Akos

        You said horizontal gain, and airplanes fly out from antenna location, eventually being on your horizon. Kerteszradio said best coverage, and he said he builds antennas.
        “Should be worse”, “cannot connect” etc is just hocus-pocus focus on regurgitating books, and really no offence meant, but when was the last time you’ve connected a Pro Stick Plus to an antenna made from a dog food tin with common TV coax? Or when was the last time you’ve actually built an ADS-B antenna, measured real-life performance? I don’t doubt you’re right in theory, but practice is different.

        • Adam

          Well Akos, the antenna field we are talking about is the field where the antennas are built after the theory and not vice versa. We do not use cut and try approach and then upon the results create the theory. There is a “fat” electromagnetic theory behind any design. What snn47 said is based on theory but also on big experience from his side. I can say that based on how he write about the problem. I can immediately recognize that he has a lot of practical experience on that field as he addressed properly the problems with stacked vertical antenna and the gain related.

          I do not say that cut and try is not fun, but a minimum of theory should be accepted not to make some basic mistakes. The fact that you can use the cut and try approach with the antenna design we can thanks to the fact that most of the todays frontends are unconditionally stable and they support the variety of impedance mismatch. Let say 10 to 15 years back this was not possible so often. This cut and try approach usually end up with burning the frontend GaAs FET due to oscillations and mismatch.

          At the end, anything can be used as antenna, even the wet salty rope 🙂 but the question is how efficient will this antenna be.

          • Akos

            Thanks, comment gives me power to go and test antennas 🙂
            “We do not use cut and try approach and then upon the results create the theory.”
            Yes and no. We test antennas for reception for ADS-B, and we don’t create theory, we use what works, because we can’t rely on assumptions and what has been done before. Harold H. Beverage had no clue what he was doing, just like you, me and snn47 have no clue when it comes to antennas, or ask an university professor and s/he will say it’s black magic. Sidescan sonar was discovered by a failure, when the head went sideways. I built an antenna from a plastic bottle and coax which received signals from Inmarsat. I relied on theory, and used ingenuity, just like Mr Beverage did stringing up copper on a tree (as an example, I’m nowhere near his genius).
            In the intervening 96 years radio seemed to forget the wonder of trying and failing, then try and fail. Radio antenna theory has been written years ago, not now. Read any PhD level paper and see references to papers written in the 60’s. Physics don’t change, but the operating environment does. Majority of readers live in a city, with lots of interference. I’m building antennas at the moment, we’ll see results. I can be wrong or not, I can’t know.
            If you have anything to contribute I’d love to hear from you, but please don’t mention “assumption” and “theory” to a man who spent the afternoon on a roof screwing antenna mounts for four antennas.
            Thanks,

            Akos

            • Adam

              The antenna theory is not a black magic, the only connection to black is a black body radiation 🙂
              ADS-B test that I have seen on various forums on the other hand can be called black magic 🙂 The 1090 MHz frequency is the one where everything can be calculated and the results are known. No strange propagation effects included so theory and software predictions are in accordance with the results on the field. The professionals get the best performance and all you need to do is just copy their design. No need to invent the hot watter as this days….
              Operating environment changing because mainly we cause the changes and according to that the proper theory should be applied (already known) end use the appropriate antenna design, already known. There is no black magic in that, simply, if it is raining – use the umbrella 🙂 do not try to invent something that will have the same or worst effect of the umbrella.

            • snn47

              „just like you, me and snn47 have no clue when it comes to antennas,“

              At least Adam and me have provided information as to why we say something, and seem to have a similar to identical understanding of RF application.

              “Physics don’t change, but the operating environment does“

              The understanding of physics, for RF applications, incl. antenna design and impact of the operational environment has significantly increased, e.g plasma or dielectric antennas. Supported by more processing power than available in the 60’s and later decades, it is now possible to calculate and model antennas.

              “If you have anything to contribute I’d love to hear from you, but please don’t mention “assumption” and “theory” to a man who spent the afternoon on a roof screwing antenna mounts for four antennas.“

              Point taken I won’t argue with you any more, since it is obvious that it is not knowledge, nor facts you are looking for.

              When I used “assume” I tried to give you the benefit of a doubt, that due to some lucky compensation you had achieved to have matched the ½ wave ground plane correctly, so that it performed well for you. But since there is no reference to the dimensions and material used, no one will be able to reproduce the design, even if it really worked well.

              PS.: Adam I remember not just GaAs FET being instable. Some MMIC, despite being in 50 ohm impedance environment, oscillated at 23 cm, and I could not find a way to stop it when the case was closed. I still prefer a salty wet noodle, you always can enjoy the cut and try leftovers with sauce as pasta.

              • Adam

                Agree, even the perfect 50 ohms match design can be spoiled 🙂 The box resonance, coupling the output to the input through the cover, high gain devices may result the oscillator from the amplifier design. We use to say: when you want to design an oscillator go for the amplifier and vice versa 🙂
                Such design may be saved using the RF absorber on the box cover but not always. I heard from some RF gurus that when you see such design using the RF absorbers to keep the device “calm” this is wrong design. On the other hand, I saw a lot professional devices where such approach with RF absorber was used. The RF absorbers are made for that purpose at the end 🙂

                To end this “blag magic” antenna story I advice the guys that are starting with radio hobby to watch the numerous youtube video where the transmision line and a dipole theory/operation is explained, the current and voltage distribution on a dipole where this can be a simple way to learn some basics upon the rest of the antenna they are using or constructing can be characterized.

            • University Professor

              Is quite clear that you don’t have a background in science or engineering. A University Professor is a hundred percent NOT going to say that antennas are black magic. The physics behind antennas are certainly well understood already. I think what the others are saying is that you can’t just build a bunch of antennas and say that one works the best. You need the theory to understand why. Without the theory it really is black magic. No one knows if your best antenna will be the best for someone else. It could be totally different. Even the simplest antenna theory about radiation patterns and impedance matching helps us understand why some particular designs are better than others and in what scenarios they’ll work best.

              It seems to be a disservice to beginers to says that one antenna is best from only some random practical tests, especially as you picked out a half wave ground plane which is not even an antenna that can work well for most people or even anyone. If you can explain why it is working better than the standard, then at least some people in the same situation as you could use it.

              There is already a lot of black magic designers out there in the ham world. Which is fine, but if your posting to an audience claiming that you know whats what, when you don’t then thats not good.

        • snn47

          cite=“ You said horizontal gain, and airplanes fly out from antenna location, eventually being on your horizon. Kerteszradio said best coverage, and he said he builds antennas.“

          I accept that Kerteszradio builds a structure consisting of electrically conductive elements that have the shape of the shape of a collinear antenna, but I cannot conclude from this that his structure really performs in reality well.

          Reception at 0° main beam elevation is not desirable, because your receiver will be subjected unnecessarily to (man made) noise and signal sources, nor is reception due to e.g. gracing effects be that desirable, especially since concentrating the main beam gain at 0° gain, will limit performance at higher elevation angles were most aircraft within 200 NM slant to your antenna could be otherwise receivable.

          Just scaling down antenna designs to the calculated wavelength is not sufficient if you are at ~1 GHz or higher, won’t provide the gain nor performance, unless other factors, e.g. the dimensions of the conducting elements in relation to the wavelength, are taken into account and are compensated for.

          BTW then all MLAT-sensors used for ATC that I know of, which all have >=3° main beam elevation, are not the optimal choice for ATC use, correct?

          cite=“when was the last time you’ve connected a Pro Stick Plus to an antenna made from a dog food tin with common TV coax?”

          I seldomly use a SDR stick, unless I am interested in modulations not supported by my equipment (TRx and Rx). Since I have no dog and therefore no dogfood cans. I have to make to with whatever is sensible to provide me a mechanical and electrical stable antenna design. This can be brass, aluminum or cooper or even steel welding rods if they have a solid cooper coating, solid wires or tubing, tubing (rectangular round), PCB, 240 ohm symmetrical- Semirigid-, Flexwell- Cellflex-cable I have a sufficient length for the wavelength desired to work with, just to name a few a few materials.

          cite=“Or when was the last time you’ve actually built an ADS-B antenna, measured real-life performance? I don’t doubt you’re right in theory, but practice is different.“

          I never claimed that I did ever measure an antenna gain or pattern, at best compared it with an reference antenna of known measurements that could be reliably reproduced. Unless you have access to a commercial antenna test site, test chamber, or have built your own antenna test field, which some HAM’s with sound engineering background have shown is possible, you just have a placebo antenna gain and pattern you can believe in, but not an antenna that performs as you hope it does.

          Yes, theory is not enough unless you align theory with the many effects that apply under operational antenna applications, but so far you did not provide any facts to prove Adam or me wrong, nor did you answer my question where did the ½ ground plane without the need for matching network came from.

          • Akos

            Many thanks for your and Adam’s comments above, especially for the inspiration to test different materials as listed above.
            The idea for the half-wave ground plane came up after testing antennas during Christmas and recently in a remote location, where they performed better than a 1/4 ground plane.
            I’m running real-life ADS-B antenna testing at the moment at home (city center location) against a Flightaware antenna as reference to get hard data.
            Cheers,

            Akos

              • Akos

                Many thanks, will do!
                I truly appreciate your input above, because for the last week since we started this conversation / argument / discussion I’m thinking why that antenna worked so well. I’m testing at a different location at the moment, will make details public.
                Again, if either you, Adam or snn47 would like to get access to the site to write your comments on results, or provide suggestions for the best antenna design, you’re more than welcome, but I’ll migrate to the forum link provided above.

                Thanks again,

                Akos

  1. Adam

    WTF? The half wavelength GP antenna? At first I did not pay attention to the published measures but when last night I characterize my GP antenna noticed that there is something odd with this figures. Where this half wavelength antenna came from? This can not work. You better take down that picture. Sorry, but this is all wrong.

    • admin

      Hmm looking into it closer you are right, a 1/2 wave GP makes no sense. Without matching the SWR is over 30, resulting in almost 10dB loss, probably higher in real life. The little bit of extra gain can’t compensate that at all.

      • snn47

        I start to wonder why we are not told the source for this antenna design even though I have asked before.

        I see two possibilities the antenna shown in the picture worked for him well enough

        1.You don’t have to worry about mismatch and high SWR, if the cable is long enough, that it provides enough attenuation to reduce reflection sufficiently to show low SWR, and if you don’t have means to measure performance you will not know how well an antenna works.

        2. that due to the cable-type/-impedance and cable -length chosen for the antenna in the picture, it provided accidentally enough impendence matching to provide some performance. But without any data all on the cable and length this is speculation.

        PS: The reason why I wrote about possibility, because I remember a firm selling an automatic turner in the 70’s, which had no without connections for tuning nor tuning knobs. While most said it can’t work, some said it did for them including the USNY as reference. The operator was only required to choose from the two versions the power rating of his transmitter, attach a cable of min. length of or longer and connect the coax between the antenna and the TRx.
        No one could understand why it was supposed to work for the USNAY. A x-ray didn’t help, because it showed only a complicated pcb and all components were inaccessible, fixed in some hardened glue. Not until someone bought one, picked it apart and made pictures and an article for ARRL showing that the wonder automatic turner was just a few high power resistors in parallel. A surplus pcb which was not connected to the boxes connections was placed on top before pouring the glue in the box. The plausible explanation given at the time was the USNY had transmitters in the kW range, so once the transmitter was tuned, it provided together with the forced matching, cable and resistor, still enough signal to the wired to establish 2 way communication.

  2. Max

    I understand that the antenna post is written for beginners but there is nothing worse for a beginner than to start with poor results and believe that it’s all he can get.
    There are on internet a lot of discussions and test results for homemade ADB-S antennas, expecially on the ADB-S related forums. The best thing that can be done for the beginner is to find these posts, build 2-3 antennas and offer the best solutions, possibly with comparative tests.. At this wavelength any antenna is easy to make.
    I love projects made with zero money or with recycled materials and converted parts but as far as there is good performance and solid contruction. If the beginner want to use a piece of wire and make an antenna that last one day instead of spending three dollars in solid material he is just a time waster and the right place for him is on facebook groups, not in your respectable website 🙂

      • Akos

        Max, thanks for the comment, I’ve included a link to one of the best forum I’ve came across just as you mentioned.
        How well that particular antenna functions will be tested again, I’ve done it already and know the results, but did not wish to overcomplicate the post with more information than absolutely necessary.
        Cheers,

        Akos

  3. snn47

    Antenna are a bit of black magic, while they work for receive you cannot assume that you come even close to the gain that a resonant antenna which is impedance matched to the attached coax cable to the receiver, will have.

    A 1/2 wave radiator with a counterpoise made of 3 or more resonant radials, are to my knowledge not matched correctly to the cable impedance.

    But then the post does not mention impedance, nor which impedance the antenna is supposed to have not that the coax cable is supposed to match it.

    Impedance for coax can be anything between 50 to 75 Ohm, up to 92 Ohm RG-71 or even 125 Ohm RG-63 .

    While the loss due to impedance missmatch between an antenna and the coax coax, and coax to the LNA or SDR-Rx is not so severe for receive only compared when you want to use ist for transmitt, but you still add losses.

    The cable loss at 1090 MHz is also not accounted for. While a antenna with a 1/2 wave radiator has in principle slightly higher more gain compared to a 1/4 wave ground plane, the miss match will imho more than compensate that princiüally higher gain.

    The Length of radiator at such high frequencies to achieve resonance, also varies with diameter of the wire/tubes If not resonant at 1090 MHz it will increase losses .

    Why not use a simple antenna, e.g. what aircraft use. A blade antenna is a 1/4 resonator on top of a suffciently large metallic sheet (>4′ x 4′) will provide about 5 dBi Gain (=relative to an isotropic antenna). Alternatively you can use 1/4 wave wires as counterpoise instead of the metallic sheet which is then called a ground plane. You’ll find all the easy guidance you need in HAM radio literature, but please don’t expect to get miracles by just throwing coax and wires together. Comercially made antennas when provided with measurement of the antenna gain and pattern are not that expensive without reason.

    • Akos

      Thanks for the comment and the ideas, I’ll run a real-life test next week to see how theory holds up.
      I agree on the necessity of impedance matching for maximum performance, but please realize that post has been written for beginners. Cable loss, impedance mismatches, different coax types is too much information for someone wanting an antenna that works, over-complicating an already complicated issue. Getting a 4×4 sheet is more cumbersome than buying coax in a supermarket, but it’s a great idea I haven’t tried, please get in touch either here or via email with more details.
      Miracles do happen, read the post referenced at the end of the post for astonishing images to see what works, or talk to actual ADS-B users on that forum. abcd567 is a safe bet, built and tested lots of antennas, “imho” and “in principle” is not in his vocabulary 🙂
      For the filtering comment, well, dear Anonymous cared enough to vomit some bile instead of using a finger movement and not reading the post, so s/he cares, which fact I treasure deep in my heart as future mental inspiration, therefore I’d love to hear from the same or any other user what s/he doesn’t like about the blog, or my posts in general.
      Feedback is always welcome, previously a kind-hearted soul mentioned that my page is only good for PCB images, so I do more of them – without actually telling me what you like or don’t like, or agree with, or disapprove, I can’t improve or make changes.
      Thanks for commenting now and in advance,

      Akos
      radioforeveryone.com

      • snn47

        “Getting a 4×4 sheet is more cumbersome than buying coax in a supermarket” agreed but it explains the principle. You can also use sufficiently fine mesh or as I wrote built a ground plane using just 3 to 4 wires. But the sheet is what ARINC specifies for L-Band blade antennas to achieve the performance. A blade antenna is basically only a 1/4 wave piece of metal wire above the aircrafts fuselage, hence the sheet.

        Unless you transmitt, and I mean not just a few mW, you will have no problem receiving, but the antennas won’t live up to their full potencial.

        And then there is cable loss. Did you bother to look up how high the attenuation of the thin coax at your magnetic whips are at 1000 MHz? It looks like RG174. Even a good cable from Belden is which has 111.554 dB/100m or 1.1 dB per meter. This is why antennas trying to improve the antenna gain having a few meter of bad cable are such a bad idea, why I said just go for a 1/4 and see that the antenna is resonant and the impedance is matched.

        PS: I would have contacted you, but since I am oldfashioned and don’t use any of the accounts you use, but I’m registered at this website.

        • Akos

          “And then there is cable loss. Did you bother to look up how high the attenuation of the thin coax at your magnetic whips are at 1000 MHz?”
          I appreciate your comments, but please read the site, see for yourself that coax loss has been covered earlier. Just because I don’t use radio terminology doesn’t mean I know nothing. I do my best to write posts digestible for a wider audience, but that effort doesn’t mean I have no clue, I make my living using radio, licensed by two governments from HF to Inmarsat etc.
          1/2 wave is better than 1/4 wave, test has been done before, I’ll redo the test and publish report. Just because everyone is using 1/4 doesn’t mean 1/2 or 5/8 is not as good. Or better. Or worse,will be tested.
          Should I end this conversation with 73, Akos? Or shall I post a video of me copying at 20 wpm visually or by ear to gain your trust?

    • admin

      I believe that a 1/2 wave antenna also has a flatter radiation pattern (with higher gain as you mentioned). Flatter to some extent should work better for receiving distant aircraft, but of course won’t be great in all situations. So the gain+loss may just cancel out depending on the coax length.

      In my experience i’ve seen that most highly successful home made ADS-B builds use a simple 1/4 wave GP antenna.

Leave a Reply to Anonymous Cancel reply

You may use the following HTML:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

This site uses Akismet to reduce spam. Learn how your comment data is processed.