Umbrella And Tin Cans Turned Into WiFi Dish Antenna

There’s something iconic about dish antennas. Chances are it’s the antenna that non-antenna people think about when they picture an antenna. And for many applications, the directionality and gain of a dish can really help reach out and touch someone. So if you’re looking to tap into a distant WiFi network, this umbrella-turned-dish antenna might be just the thing to build.

Stretching the limits of WiFi connections seems to be a focus of [andrew mcneil]’s builds, at least to judge by his YouTube channel. This portable, foldable dish is intended to increase the performance of one of his cantennas, a simple home-brew WiFi antenna that uses food cans as directional waveguides. The dish is built from the skeleton of an umbrella-style photographer’s flash reflector; he chose this over a discount-store rain umbrella because the reflector has an actual parabolic shape. The reflective material was stripped off and used as a template to cut new gores of metal window screen material. It’s considerably stiffer than the reflector fabric, but it stretches taut between the ribs and can still fold up, at least sort of. An arm was fashioned from dowels to position the cantenna feed-horn at the focus of the reflector; not much detail is given on the cantenna itself, but we assume it’s similar in design to cantennas we’ve featured before.

[andrew] hasn’t done rigorous testing yet, but a quick 360° scan from inside his shop showed dozens of WiFi signals, most with really good signals. We’ll be interested to see just how much this reflector increases the cantenna’s performance.

24 thoughts on “Umbrella And Tin Cans Turned Into WiFi Dish Antenna

    1. You mean biaxially-oriented polyethylene terephthalate (mostly known as Mylar), I do not think that it has been used to reflect RF. But 2.4GHz WiFi has a wavelength of roughly 125 mm and 5.5GHz WiFi has a wavelength of around 55 mm and if you assume that as long as any holes or imprecations on the flatness of the parabolic dish are one twentieth of a wavelength or less there should be a strong focusing effect. It should work fine, but from a wind perspective, if mounted outside, you are far better with large 3mm holes that allow air to flow through the antenna, while still reflecting and focusing nearly the same amount of RF energy.

  1. With greater gain comes greater responsibility to stay within your countries radiation power rules. There is a limit, and the operator needs to adjust the power down, as the gain goes higher. Alas, irresponsibility is an art form, and the wifi bands are now completely polluted as enforcement takes a knee.

    1. True, but with a parabolic antenna like this, with increased gain also comes narrowing of the beam width. That’s where the gain comes from. So, if ones uses a high gain parabolic reflector on both ends of their link and makes sure they are aimed properly, the chances of causing unwanted interference are probably actually lessened rather than increased.

      1. IIRC, the rules say that your radiation in any direction must be assumed to be radiated all around and the rules limit THAT power. So if your wifi dongle is at the limit of a omnidirectional radiation limit of 0.1W, and you’re concentrating that power across only 1/100th of the sphere, you are now effectively broadcasting at 10W. And people in the beam will indeed experience interference from you as IF you were actually transmitting at 10W.

    2. WiFi bands aren’t polluted because of directional antennas. Remember that the power focused in one
      direction is taken away from others, so a directional antenna although “polluting” more in one direction
      reduces that pollution in all other directions, therefore the sum of pollution is the same, but the user
      can reach the destination without using an amplifier (which especially if paired with a non directional
      antenna would pollute a lot more).
      Pollution aside, the WiFi spectrum is saturated because of lack of channels compared to the huge
      growth in the number of devices being used (most desktops, all laptops, all phones, all tablets, most
      TVs, and do I have to mention IoT devices?).
      If the nice folks who get money to produce standards would allocate a number of channels an order
      of magnitude higher than it is now, it still would be vastly inferior to the minimum necessary.

  2. If he really wants to “stretch the limits” of wifi using a parabolic dish, just get a C band satellite dish. You can get them up to 13 foot and perhaps even larger if you get a commercial dish.

      1. Sure, can go bigger. andrew mcneil noted this build being “fold able” and “portable”.

        Also, like other thoughts… we want to be careful so to not cause interference issues with ISM bands unless you have a license to do so. I’m thinking with all the downsizing at the FCC, unless you’re doing something for profit… they’re not going to warn, or fine especially, unless there is interference.

        andrew mcneil has some great videos for sure and will be neat to see what he does with the RTL-SDR’s also. Basically, you can put an RTL-SDR like he has with the ALFA Wifi adapter, maybe with a conical spiral antenna design he has. I have some extra cream horn moulds if you’re interested that I ordered from the UK.

        1. A large satellite dish is probably less likely to cause interference. With the increased gain also comes a narrowing of the beam width. The larger dish will focus the signal more tightly on exactly whatever it’s aimed at. Also a satellite dish is likely to have a higher surface accuracy, and be a more true parabola, which will likewise result in a more tightly focused beam.

          1. I’m guessing that is correct Stefan; makes sense if mounted higher than all those in the beam path so to point directly at the target. On the ground however… the larger dish will most likely cause issues I’m thinking. My interest was more from a like radio astronomy use in radio directional finding with two or more large satellite dishes that can be rotated and tilted. Even better would be a Z-axis effect where the dish/trough can change shape to change the focus point of the beam to be more of a spherical concave reflector and not just parabolic beam for more accurate distance sensing.

            Interesting is the parabolic trough designs that are even more effective than a dish design due to increased surface area, though the beam pattern is larger also when transmitting and I can’t even think of what would be COTS to make that portable other than using wires like some designs use as ground planes or reflectors. Then having some sort of reflector cross sectional pieces that can attach to tubes possibly telescoping or folding into shape.

    1. Might be true, because the dish seems to be rather deep with a f/D ratio of ca 0,3 if I were to guess, and if so its not about distance, the horn beam may be to narrow for this deep dish.
      I guess the horn is at focus, so he should not move it further away, rather he should shorten the horn, or dispose of it, and use a dipole with reflector, so as to illuminate the dish more effectively.

  3. Something that must be taken into account when making parabolic reflectors; is not to reflect the counter-phase of the wave.
    That is, if the reflector is too long, it may be reflecting parts of the signal that are canceling out in focus.

    A simple solution is to draw a plane normal to the vertex and focus of the paraboloid, and then eliminate any part of the paraboloid that is more than 1/2 wave away from the normal plane.
    For 2.4GHz frequency; 1/2 wavelength is 62mm.

    A more complex solution involves considering the distance from the focus to the reflector; eliminating only the parts of the reflector that are reflecting the counterphase; that is, after a full wavelength (125mm), the reflection returns to being in phase.

Leave a Reply to Robert ZachrissonCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

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