Simple NOAA/Meteor Weather Satellite Antenna: A 137 MHz V-Dipole

Over on his blog Adam 9A4QV (seller of various RTL-SDR related goods including the LNA4ALL) has just made a post detailing a build of a high performance super simple NOAA/Meteor M2 weather satellite antenna. Most antenna designs for polar orbiting weather spacecraft are based on circularly polarized turnstile or QFH designs. However, Adams antenna is based on a very simple linearly polarized dipole, which makes construction almost trivial.

The idea is that by arranging a dipole into a horizontal ‘V’ shape, the radiation pattern will be directed skywards in a figure 0 (zero) pattern. This will be optimal for satellites travelling in front, above and behind the antenna. Since polar orbiting satellites always travel North to South or vice versa, we can take advantage of this fact simply by orienting the antenna North/South. 

There is also another advantage to Adams design. Since the antenna is horizontally polarized, all vertically polarized terrestrial signals will be reduced by 20 dB. Most terrestrial signals are broadcast in vertical polarization, so this can help significantly reduce interference and overloading on your RTL-SDR. Overloading is a big problem for many trying to receive weather satellites as they transmit at 137 MHz, which is close to the very powerful FM broadcast band, air band, pagers and business radio. In contrast a circularly polarized antenna like a QFH or turnstile only reduces vertically polarized terrestrial signals by 3 dB.

As the satellites broadcast in circular polarization there will be a 3 dB loss in Adams design from using a linear polarized antenna. But this can be considered as almost negligible. Adam also argues that the home construction of a QFH can never be perfect, so there will always be at least a ~1dB loss from inaccurate construction of these antennas anyway.

The final advantage to Adams design is that construction is extremely simple. Just connect one element to the center coax conductor, and the other to the shield, and spread apart by 120 degrees.

Adam 9A4QV's V-Dipole for 137 MHz Weather Satellites.
Adam 9A4QV’s V-Dipole for 137 MHz Weather Satellites.

Adam has tested the antenna and has gotten excellent results. If you want more information about the antenna design, Adam has also uploaded a pdf with a more indepth description of the design and his thoughts.

45 comments

  1. Red Dragon

    I was going to use some thin aluminum tubing from an antenna. Would I need to change the length of the tubing or something else because of the bigger diameter?

    • Antofa

      Not for receiving, no. For transmitting though, there would be an end-correction. Usually a nice large diameter makes the antenna more broadbanded.

  2. Yves Reiser

    I got my RTL USB yesterday.
    This afternoon, I built your antenna, and no later then this evening, I was able to get 2 perfect IR-Channel Pictures. Many thanks for this excellent tutorial!!!

  3. Baris

    Hello Adam,

    So this is a really nice piece of work. We are also trying to get signal from NOAA satellites. After recieving signal, we need to calculate doppler shift and deduct orbital parameters from that.

    My question is, can this be achieved with this V-Dipole antenna design? Also we were able to find 4mm dia rods, would that have big impact on this design?

    Thanks for help in advance.

    • peter chamberlain

      Hi i Built this antenna with 4mm rod and its working pretty well , i have 22 metres of mini 8 coax and at the moment the V dipole is 8 feet off the ground and i have just realised its pointing the wrong way .. its total build cost was £4ukp .. .aprox $5.50

  4. Mark

    Built and tested this antenna. It works really well while sticking out of my window, so I can’t even imagine what would be the performance if I get it high in open area.

    • Adam

      The simplest way is to use the 1 Mega-ohm resistor across the dipole on the choc block. At some point (before reaching the receiving equipment) the braid/coax should be grounded.

      More advanced is using the quarter wavelength shorted coaxial stub, also across the dipole on the choc block.
      This stub can notch some frequencies and if we chose smartly the length we can kill the strong blockers. At some point (before reaching the receiving equipment) the braid/coax should be grounded.

  5. Anne

    Hi can anyone give me some extra information on the calculation used?
    Particularly where the constant 147 is derived from. Many thanks!
    L(m) = 147 / F(MHz)
    L(m) = 147 / 137.5
    L = 1.068 m = 106.8cm
    Each leg = 53.4 cm

    • Bertie

      I have no idea where that number came from but if you put 137MHz into a frequency to wavelength calculator ( e.g. http://www.1728.org/freqwave.htm ) you get 218.83 cm. And a quarter of a full wavelength is ~54.7 cm. I’m interested as well where the 53.4cm came from.

      • AD5NL

        It’s very similar to the formula US hams are taught and tested on, which is L = 468/f. Where L is dipole length in feet.

        Granted there is a slight difference here of a couple of percent, probably because 468/f is usually used for thin wire antennas at HF and is a bit conservative (assumes a slowdown of 5 percent rather than 2)

        http://www.kb6nu.com/468-ham-radios-magic-number/

        In either case the basic premise of both is that the speed of light depends upon the medium.

        • Anna

          Thanks everyone. Clear answers. Anyway I’m planning to make this design with thin aluminum (diam 0.8cm) tubing I have lying around. Would there be other values for the correction factor? Can you recommend a resource to look these things up?

          • Adam

            Hello Anna,

            you can try the following link for the simple approximation:
            http://www.radio-electronics.com/info/antennas/dipole/length-calculation-formula.php

            As a rule of the thumb, bigger diameter will give you more usable bandwidth. I did measure the RL and using the 1/8″ aluminum stiff wire from the video I have 20db RL for the 5MHz BW. On the resonant frequency the RL is 40dB+. For the 10db RL the BW is 24 MHz and I use the same antenna without problems on Aeronautical AM band as well as HAM 2m band. The antenna is working nice also on the government/military band 140-143Mhz where the fighters flying 200km away can be heard loud and clear.

  6. Martijn

    With a slight modification, would this be usable for receiving amateur satellites on the 2M band as well as weather satellites?

    • Adam

      Hello Martijn,
      This antenna is not good for the amateur satellites. Usually, HAM sats are not 3 axis stabilized and they are thumbing. Combined with the low power and linearly polarized antennas on the sats you will experience a deep fading (QSB) on your side. Directional antennas with tracking are required for descent operation.
      You can get a fraction of transmission from that satellites using the V-dipole but this will be rather frustrating experience.

        • Martijn

          Hi Adam,
          Thanks for your help. Due to the space needed and complexity of a directional antenna with a tracking rotor, this won’t be a option for me. I had a small hope that this simple setup will do. 😉

          • Adam

            Another one trying to fill up a 1.5L bottle with 2L of water 🙂
            It can be done but this is completely another antenna / topic.

          • JustJoe

            Not needed. Not too plug a manufacturer – but Arrow makes the most cost affordable for purchase/retail dual band hand held yagi antenna for shooting (amateur) starfish – albeit kinda fun to make a DIY version, the cost actually turns out to be almost the same in parts acquired/needed as you actually end up with quite a bit of spare aluminum in just making a single unit. Plus the
            Arrow one is significantly, lets say *well built*, robust and completely tuned. They were the whopping 45$ USD last time I checked.

  7. AD5NL

    I like this approach. I saw it on Adam’s blog yesterday before this was posted.

    Although it may not be perfectly optimal in theory, in practice it may yield better results because it is very easy to build and tune (I haven’t even attempted to tune my QFH because I don’t have a modern VHF VNA (I have an antenna analyzer but it’s an old analog one — basically a fancy grid-dip-meter — that only works for HF) and besides it would be a real pain anyway to change the dimension of loops!!!).

    The bill of materials is very simple — two linear elements (I might try using solid 14ga. copper wire rather than aluminum), something to act as a center insulator (choc block, piece of plasic with holes drilled out, whatever) and a choke balun (couple of snap on ferrites maybe).

    I have been going crazy trying to work with circularly-polarized antennas, but the thing is NOAA signals are actually pretty strong; 3db shouldn’t matter terribly much; and the biggest problems are terrestrial interference and actual mechanical failure of the antenna. I suspect there are a number of ways where a mistake in a circularly-polarized yagi or QFH could easily result in a 3db loss of performance, or worse.

    So anyway, kudos for the “KISS” approach to antenna design.

  8. Alex

    Adam,
    Can you show/draw exact scheme in PAINT,
    how exactly did you connect cable to aluminium rogs or Chock Block?

    Regards,
    Alex

    • Adam

      Hello Alex,

      the choc block has two screws on each terminal. Under one of the screws the aluminum bar is inserted and under the other screw of the same terminal the coaxial cable wire (center or braid). Same goes for the other choc terminal. Once you do that you can spread the wires to create the 120deg angle. Of course, it will look much better if you prepare the wires and bent the ends before inserting into the choc terminals.

      • Alex

        Thanks Adam.
        Very informative.
        But still some misunderstand as new to this .

        1)We have one coax cabel,it goes to first terminal where it then connected to aluminum bar.?Right?

        2)But how another aluminium bar in second terminal is connected to the cable?
        We need to solder one more cable between coaxil cable and put it under screw of second terminal?

        3)Can I connect coaxial cable directly to RTL SDR , via BNC-SMA adapter ,without any dc block between them?

        Thank you in advance

        • KD0CQ

          1) Yup
          2)Center conductor is screwed or soldered to one terminal, ground/outer conductor is screwed/soldered to 2nd terminal.
          3) you can, I think he put that there to help protect against transient currents on the coax, in other words, to prevent a sudden surge in voltage which would fry the front end of the rtlsdr.

            • KD0CQ

              I’m a retard. The blocking capacitor is there because his RTLSDR has the bias-T mod enabled. He’s protecting the dongle from any potential shorts since there was no LNA in line in the initial test.

              • Adam

                Yes, this is the reason why I have a DC-block in line without LNA. You never know what may happen on the antenna so better protect dongle from short if the Bias-T is enabled.

  9. EA4GMZ

    Very interesting. However I think some kind of balun would improve the radiation pattern simmetry and avoid picking noise from the coaxial braid. You could either use a winding of a few turns of coaxial next to the feed point working as current choke or a half wave coax balun for feeding the dipole arms in oposite phase. In the later case, it also transforms the impedance in a 1:2 ratio.

    • Adam

      A balun to improve the symmetry as suggested to prevent the currents on the braid can improve the things even more. This may be added but the design is more complicated for the beginners. Impedance transformation is not required, the impedance is very close to 50 ohms using a V-shape and the match and RL is good, up 30 dB. I do have the the RL plots but did not want to complicate with this details as they are not so important for this purpose.
      The idea was to present a simple antenna that will deliver good results with minimum efforts and using cheap materials.

  10. CB

    Incorrect: A turnstile has no vertical component. Both elements are horizontal with respect to the ground (same as Adam’s V-dipole), so the discrimination is far greater than 3dB for terrestrial vertically polarized signals.

    • Adam

      Every antenna has a vertical component. Infinite cross-polarization exists only in theory. Cross polarization between V-H or RHCP-LHCP should be infinite, but we can see that in practice it is from 20-30 dB. You can sheck that just turning your whip antenna for 90 degrees. Just in theory you can have only one component. How come then the turnstile is receiving the signals from horizon? This signals are coming from a very low elevations. In theory you should not have any signal, but in practice the signal is present.

      • CB

        Received signals from the horizon (on a v-dipole or turnstile) are still primarily using a horizontal component. Read my comment again – I am not disagreeing with you, but saying that the vertical component of a turnstile vs the v-dipole is the same – in theory infinite, but like you say, in practice, more like 20-30dB down (that’s why I say “far greater than”). My point is that the post above is incorrect in saying that a CP turnstile only attenuates vertical signals by 3dB.

        • Adam

          The main concern and the problems are the strong signals coming from the Airplanes just 10km above and they are using the vertical polarization (mentioned in the full PDF article) If you have a circular polarization, the vertical component is not coming just from the horizon, but from direction with higher elevation. The other problem is that some professional services and mainly the WBFM transmitters are also using the circular polarization in the areas with the terrain with hills and valleys where due to multiple reflections the transmitted signal change the original polarization. Such a reflections may result with deep fading in signals due to cross polarization (30dB). To avoid this problems the WBFM is transmitting circular polarization and the theoretical fading should be not more than 3dB.

Post a comment

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>