Amateur Radio Transmits 1000 Miles On Voice Power

Many of us tried the old “Two tin cans connected by a string” experiment as kids. [Michael Rainey, AA1TJ] never quite forgot it.  Back in 2009, he built “El Silbo”, a ham radio transmitter powered entirely by his voice. El Silbo is a Double Side Band (DSB) transmitter for 75 meters. While voice is used to excite the transmitter, it doesn’t actually transmit voice. El Silbo is a CW affair, so you should bone up on your Morse Code a bit before building one. Like many QRP transmitters El Silbo’s circuit is rather simple. A junk box loudspeaker is installed at the bottom of the can to convert voice power to electrical power. The signal is passed through a step up transformer, and used to excite a 75m crystal. Two NPN transistors (in this case MPS6521) pass the signal on through a second transformer. The signal is then routed through an LC network to the antenna.

Back in 2009, [Michael] brought El Silbo to the Maine coast in an attempt to make a transatlantic contact. This isn’t as far-fetched as it sounds – [Michael] has “crossed the pond” on less power. While the attempt wasn’t successful, [Michael] has made connections as far as 1486km, or 923 miles. That’s quite a distance for simply yelling into a tin can! One of [Michael’s] favorite El Silbo stories is a 109KM conversation (QSO) he had with W1PID. [Michael] found that the signal was so good, he didn’t have to yell at all. He reduced power by dropping to his normal speaking voice for the “dits and dahs”. The two were able to converse for 17 minutes with [Michael] only using his speaking voice for power. We think this is an amazing achievement, and once more proof that you don’t need a multi-thousand dollar shack to make contacts as a ham.

[Thanks Bill!]

50 thoughts on “Amateur Radio Transmits 1000 Miles On Voice Power

    1. This is not transmitting his sound. The sound of his voice drives a loud speaker which generates an AC voltage or some arbitrary waveform. This gets rectified and turned into DC which powers the CW transmitter. It would sound just like a CW signal. It makes no difference what his voice “sounds” like, as long as it is loud enough for the speaker to generate enough power to drive the transmitter. He could play lound music into the speaker and on the air you would hear a CW tone.

        1. I built a little morse do-dad using an Atmega1284p. It had a digital recording of two dog barks, one long and one short. The thing would bark out morse that was ridiculous and perfectly copyable. I used two capacitive touch paddles to key it and had a mode that just keyed out a bunch of text files I stored in the flash. I gave it to an old friend who is now a silent key.

    1. That would actually be great! But a complementary receiver would, I guess, need a power source. Dunno how much antenna you’d need for a crystal set on the relevant bands, and if there’d be enough power, since crystal sets are generally for the multi-kilowatt power AM broadcast band. Still, if you knew Morse I suppose beeping out your co-ordinates is better than nothing.

      1. I don’t see why someone who is adrift at sea would have to transmit and receive on the same frequency. Also, it’s quite likely that whoever would come to your rescue has sufficient power to transmit a signal strong enough to be received with a batteryless crystal receiver. It would prevent you from chatting with other survivors though.

        While this is an amazing achievement, for actual emergency situations, a transmitter and receiver powered by a hand crank would probably be more practical, although it would mechanically be more complicated and as such could be more likely to break.

    2. For maritime emergency a smaller version of the old “Gibson Girl” crank powered CW transmitter would be better. It would be very easy to build one using a small DC cordless drill as crank generator, a uC to auto xmit SOS and a button to use as a key. A simple and compact HF transceiver design and you would be all set. 5 watts out would be easily obtained.

      1. I’d rather have a CO2 cartridge driving a linear generator. Think shake flashlight meets steam engine. He just needs a small CO2 cartridge, valve, and a whistle for that transmitter to be really cool. Toot toot!

  1. i guess the fcc is too busy making sure we dont say 1 or more of the 7 words on tv or keeping the internet neutral or making sure hobby transmitters dont knock out the paramedic’s radios or stop someone’s pacemaker to be dealing with unlicensed radio transmitters.

    1. ham radio transmitters are not licensed, the operator is. I can build and use any transmitter (within power limits) and transmit on the ham bands as long as I have an amateur license that permits transmission on those frequencies.

      there is a long DIY tradition in amateur radio.

    2. Hams are fairly self-policing as well. We’re more likely to report gross/harmful offence among our own to the FCC if needed.
      We stay within our own bands, and take all measures possible to avoid interference with other services.

    3. Nice Try, Troll. Ham operators have a legacy of building their own stuff and advancing telecommunications by way of experimentation, it’s one of the main purposes of the service.

        1. People used to make their own Galena crystals? Never knew that! I’ll go look it up, sound interesting. I’ve heard of people using coke (the type from coal) or rusty razor blades (the type for the other type of coke) with a cat’s whisker.

  2. This is highly common in Amateur radio. Most of what people use today was INVENTED by ham radio operators. It is why a large chunk of the airwaves are devoted to this.

    and it is amazing how far you can transmit on QRP levels of power, a lot of people think you need 10,000 watts to talk around the world, I have done it on 2 watts several times. This guy is why ham radio operators are the innovators in the world of communications.

    1. Innovators? Once upon a time, yeah. Not so much anymore. You wont find a ham operator keying in ipv6 headers or creating algorithms to solve extreme latency communication challenges that are plaguing space exploration. For real innovation they would need some limitations lifted so they can play on a level field.

    1. Double sideband is another way of saying amplitude modulation. CW is continuous wave; he’s transmitting a continuous wave using double sideband (amplitude modulation). Think of CW as the data and DSB as the protocol.

  3. Extremely cool hack.

    I just had a serious “hey, I know that guy!” moment. W1PID is in the same ARC (amateur radio club) as me and we have talked many times, both on the radio and in person.

    Needless to say, I will be asking him about this in the near future.

    1. I’d guess worse. For one thing piezos generate almost no current, but at an impractically high voltage. A few other things too I think would make it impractical even using a step-down transformer.

      The other thing, a piezo doesn’t have a large amount of moving mass when you talk into it. A paper speaker has the cone and the coil all moving about, I’d guess they were more efficient at collecting the energy from speech.

  4. “since crystal sets are generally for the multi-kilowatt power AM broadcast band”

    All that matters is the field strength. I grew up 1/2 mile from an AM station and my crystal set powered a speaker to medium volume.

    The station reduced power at night to only 100 Watts or so – didn’t make any difference..

    I agree though, this is pretty impressive. I’m working on such design right now, solar and hand-crank stepper motor powered electronics. (stepper motors make excellent generators)

    So far I’ve tested a solar powered radio that runs directly off the solar cells (most such radios only charge a battery/supercap that runs the radio for a limited time)

    For rescue you only need to xmit a siren sound on the international distress freqs – 121.5 or 243.MHz (these are depreciated now but still work)

    1. I suppose shouting “EEE-AWW” into this can might count, from the explanation given above it seems like it’d be voice modulated on simple principles. Would … — … on carrier wave do?

      I know ships, even now, have backup “voice-powered” telephones. The replacement for the old speaking tubes. It’s just a small speaker connected directly by 2 wires to another speaker in the other room. The volume’s fine, apparently. I wonder if the impedance of the speakers would matter? Or if using transformers at each end would help? Maybe not, at the low frequencies and power involved.

      I once had a few solar cells I linked up through an ammeter / voltmeter (both, it was a millammeter that I bypassed with a bit of wire for a low resistance, or used open-circuit. Wasn’t calibrated to anything), to an old walkman. It would power the radio all day long, running speakers from the headphone output. If the sun was out, in summer, it would work the tape mechanism. This was 6 cells in series, each one maybe 2×4 inches, cheap amorphous silicon on glass. Just mounted up with sellotape on a bit of old plastic.

      So yeah you can get quite good radio performance, driving a pair of 8 ohm speakers for FM stereo, off a small enough solar panel. I suppose a supercap would be good for all the time you’re not using the radio, store the excess and you’d be able to run it at night. With the ludicrously high kilo-Farad stuff nowadays you could probably run a radio off a full charge for hours.

  5. Re: solar cell powered radios

    I have only 20.ma to work with, the stereo decoder would draw more power and 2 speakers with 2 audio amps would almost double the power needed. This is just a ‘kick start’ project as I begin the transition to solar. Because of the stations here I have signal saturation (the RF input of most tuners is swamped and causes poor reception) so although they have some very low power FM receiver chips (like the TDA7000) they perform poorly. All ‘real tuners’ (tuned 2 or 3 stage RF front end, separate osc and balanced mixer stages) all draw more power than my budget allows (by the time you include I.F., demod and audio amp)

    I’m working to obtain more solar cells then this won’t be a constraint. The distress beacon siren can be made with a CMOS chip, it only needs to gate the RF. With a hex inverter you could have a 2-inverter audio osc, a 2-inverter low frequency osc (to sweep the audio tone) and the final 2 inverters as an RF osc (74HC series would be fast enough for 120.MHz although I would just use a 2 transistor transmitter) Or variations on this general design.

    A lower frequency (to save power – lower speed CMOS) can be multiplied by tuning the input stage of the final to a harmonic of that square wave output, but a simple discrete xal osc transmitter with a second output buffer stage would be about the same amount of work.

Leave a Reply to fartfaceCancel 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.