RPiTX Turns Rasberry Pi into Versatile Radio Transmitter

Since the discovery that some USB TV tuner dongles could be used to monitor radio waves across a huge amount of spectrum, the software-defined radio world has exploded with interest. The one limiting factor, though, has been that the dongles can only receive signals; they can’t transmit them. [Evariste Okcestbon, F5OEO] (if that is his real name! Ok c’est bon = Ok this is good) has written some software that will get you transmitting using SDR with only a Raspberry Pi and a wire.

There have been projects in the past that use a Pi to broadcast radio (PiFM), but this new software (RPiTX) takes it a couple steps further. Using just an appropriately-sized wire connected to one of the GPIO pins, the Raspberry Pi is capable of broadcasting using FM, AM, SSB, SSTV, or FSQ signals. This greatly increases the potential of this simple computer-turned-transmitter and anyone should be able to get a lot of use out of it. In the video demo below the break, [Evariste] records a wireless doorbell signal and then re-transmits it using just the Rasbperry Pi.

The RPiTX code is available on GitHub if you want to try it out. And it should go without saying that you will most likely need an amateur radio license of some sort to use most of these features, depending on your locale. If you don’t have a ham radio license yet, you don’t need one to listen if you want to get started in the world of SDR. But a ham license isn’t hard to get and at this point it shouldn’t take much convincing for you to get transmitting.

58 thoughts on “RPiTX Turns Rasberry Pi into Versatile Radio Transmitter

  1. Very interesting. The article claims this will make it a “ridiculously powerful” transmitter. Define “ridiculous”. I’m guessing it’s in the order of a few mW and if that’s the case, you don’t need a ham license to transmit so long as it adheres to FCC part 15 rules (think of the kits Radio Shack sold back in the day). If you want to run more power then yes, you’ll need a license. [Source: myself, n0xmz]

    1. Regardless of the power output, it 100% fails the FCC rules because it throws harmonic interference all over the spectrum. Without an appropriate low-pass filter in place, this is completely illegal for anyone (licensed or not) to use.

      1. ^^^ THIS ^^^
        If you’re going to use the RasPi as a transmitter, be aware of the purity (or lack thereof) of the signal you’re transmitting, and don’t run it through an amplifier unless you know it’s clean.

      2. Lots of complaints about this. But if the only solution that is offered is to buy an HackRF or USRP for $300+, then you will get lots of people speweing “harmonic interference” anyways beacause it’s actually available and at the minimum 10cm range it doesn’t really matter right? I’ve seen the RasPI used for things like this before this post, it was just not advertised because the HAMs in question didn’t want idiots like me to use it.

          1. A simple filter like that would be a good thing to sell as a cheap kit to people interested in using this software. Even just plans would be good, often software types aren’t brilliant at hardware, and RF stuff is a whole black art to outsiders (ie me).

            As it is, the range is going to be very limited, low output power, I’m sure hams have worse things to worry about. The FCC aren’t going to even receive any of this interference.

          2. Also… do they make RF DACS? And cheaply? That would do a better job. Actually, wonder if an R-2R DAC would be good enough, could the Pi send data quickly enough? GPIOs tend to be set one pin at a time, can you send a byte to 8 pins at once?

          3. Greenbaum is totally wrong. The FCC does not monitor ham radio transmissions. No, OTHER HAMS do! And the only way we can keep the radio spectra that we are allowed to have is by SELF POLICING. And when a couple of hams zero in on your crappy transmission they are going to report you and you will have a nice “Pink Ticket” to hang on the wall alongside the photocopy of the $5000 check that you wrote to the FCC Violations Division (sic). This is not a trivial matter.

            And as for that few milliwatts, Google JT65 and “WSPR” and see what fun hams are having seeing how far (around the world basically) their couple of milliwatt transmissions are reaching.

            This is not a trivial matter

        1. @Greenaum: The kit idea is very good. Problem with SDR is that you want a tuneable output filter. Or you could make a couple / many plug-in modules with the right filter and antenna on each.

          An R-2R DAC would work fine, but I don’t know if the Rasbpi can handle the speed on an (eg) 8-bit bus. That’s also an interesting idea.

          @Everyone: This is a quick and dirty hack and, as everyone’s complained, it’s spewing square waves. But it’s doing so quietly enough that you’re not going to bother anyone unless one of your neighbors is a ham. And if he/she is, then he/she would doubtless help you with some filters if you asked.

          You’d make a new friend, and the world would be a better place. More square waves!

      3. Anyone with a ham license that permits building your own radio should be capable to verify harmonic suppression. Anyone that isn’t licensed should not be using this software without making sure they aren’t breaking local law. FCC rules only apply to Americans on American soil.

    2. Sometimes you just don’t care about legality. Once I was transmitting nasheeds on an active military frequency standing near army base, that could end in some serious legal troubles but it was fun as hell.

        1. It’s not that awesome, just using the onboard (presumably) IF oscillator, which leaks a bit. At a fixed frequency, and presumably with lower power than you’d get from the RPITX.

          1. low power? amplify it.
            leaking? filter it.
            There is no similar solution in the same price range(or even close).
            This is not just on board oscillator, you can’t do real SDR with (say) si5351 and micro controller.

      1. No, many key fobs use the 300-400mhz range. There might be a few manufacturers using 2.4ghz but I don’t know why they would. Lots of interference, requires more power to transmit and no need for a faster bit rate. RC uses 2.4ghz because they can utilize spread spectrum to avoid interference with other transmitters, where keyloq (and equivelant) to my understanding doesn’t do frequency hopping, only rolling codes.

    1. FM is permissible on 10 M and he was using a repeater. Although I haven’t read all the info the designer builder has this project I’m going to assume this build is incapable of SSB.

  2. correct me if i am wrong but doesnt harmonics get weaker as they go up the scale?

    for example 100 mhz will transmit 100 feet the first harmonic at 200 mhz may only reach 50 feet then the second harmonic of 300 mhz will reach 25 feet and so on.

    so as long as you dont set it to transmit miles you should be ok and only receivers in front of your house would be able to hear the harmonics.

    1. Harmonics do get weaker for each octave. However, you need to consider the efficiency of the receiving antenna at the other end. The unintentional victim receiver may be a VHF receiver with a tuned antenna. The antenna itself is a band pass filter. Although you are transmitting more energy at the fundamental frequency, the VHF antenna and receiver will not see it. However, the 9th harmonic may be screaming at it.

      Just because there is little power involved does not mean the signals will not propagate long distances (http://www.naqcc.info/qrpworks.html):

      “According to Rich Arland, K7YHA (now K7SZ), in World Radio magazine (Feb. 1990, year 19, issue 89, pp. 46-47) the long-distance low power record is held by KL7YU and W7BVV using one micro-watt over a distance of 1,650 mile 10-meter path between Alaska and Oregon in 1970. This is the equivalent of 1.6 billion miles per watt.”

  3. The wider meaning of this is that we now have a generation of hackers who don’t know why a free-space sine wave is cleaner and better than a square wave. They don’t understand that there is no such thing as ‘digital’ electronics. There is only analogue electronics driven to semiconductor saturation with accompanying harmonics and nasty signal artefacts. It’s not because new-generation hackers are dumb, it’s just that ‘digital’ is so exciting and productive that the knowledge and insights of old-school radio hams and audio enthusiasts have been shoved aside. Oh well. It just means a polluted RF environment for us all and a SNR arms race in the future.

    1. [sigh]…Wow! Respectfully that that make such little sense, to the point it could be considered nonsense. I agree there i no such thing as “digital electronics” just like there are no color or digital TV antennas. Sure sine wave transmissions should be better that the transmission of square waves, but I even don’t know who is deliberately transmitting square waves. The digital representation is not a square wave. Having said that the output of many AC converters are best described as modified square wave rather than modified sine wave. Although I’m sure they are called modified sine wave because the out put using the trapezoidal wave form approximates the RMS output of 60 cycle sine wave Digital transmissions wherecharacterized as “analogue electronics driven to semiconductor saturation with accompanying harmonics and nasty signal artifacts”;really? in the event that’s the case satellite, shortwave and other terrestrial Radio/ TV broadcast spectrum along with Land mobile spectrum as been rendered useless given your assessment, but that spectrum clearly hasn’t been rendered unusable. Respectfully if that broad brush comment was composed by an old school ham, perhaps some old school hams should be shoved aside, I say that as a ham, but I may not be sufficiently old school enough :) [shrug]

      1. You hapless troll.

        I never said digital comms systems had rendered any bands unusable, but if you want to refute things people never said and you feel like you’ve won in your own mind, then cool.

        You are not aware of broadcast squarewaves causing interference? Everyone else is. Those crunchy sounds when cell phones break into audio circuits and radio receivers. Those are from the time division multiplexed transmit pulses. Not bad magic like you might have thought.

        To save you another rant, I know there is no such thing as a broadcastabke perfect squarewaves. We are just talking a shorthand for waveforms with gross harmonic distortion because they have been sent into free space by people who have learned enough digital but too little analogue electronics.

        Now go away and improve the social SNR.

  4. While the possibility of transference is real, but all the hand wringing isn’t going to prevent it. How about the hams wringing their hands over interference start promoting how to correctly use these components to activate seldom use amateur radio allocations?

  5. Myth: Because it is low power and I only observe it working over a short distance it will not interfere with anyone.
    Predicting how far and where RF will go is a very difficult thing to do. It may seem to only be receivable at a short range however that signal which gets spread thin as it gets farther away will interact with all sorts of objects. Reflections may re-concentrate it in unexpected places making it strong enough to interfere with someone. Rather than objects even the atmosphere itself can do this often in unpredictable ways resulting from both earth and space weather. It’s rare and unlikely but milliwatts strong signals have crossed oceans.

    Myth: The only person that might get interfered with is a ham and he/she will just help me fix the problem anyway.
    Those harmonics, going up and down the bands may interfere with broadcast frequencies, military, police, emergency workers such as ambulance and fire, etc… Your interference probably will not but could cost someone’s life! A ham very well might chose to be friendly and help you but interfering with these other services could result in a really big fine.

    Myth: Building a filter is some horribly complicated task that one could never accomplish themselves without a whole lot of help.

    It’s just a combination of a few coils (just wrapped up wire) and capacitors. If you are here you have probably done more complicated stuff. http://bfy.tw/2eh5

    1. I agree. i have one small problem, this new solution (and other like si5351) are capable to cover wide spectrum. all the bandpass filters plans I saw were for single band (or huge bandpass filter arrays).
      Do you know and variable band pass filter for TX (DIY or buy)?

      1. You generally use one filter per band or set of bands with some way of switching them out. If you’re generating a reasonably clean sine wave, I think you can get away with sharing a lowpass filter between two or three bands and end up with somewhere around one lowpass filter for every octave (doubling/halving of frequency) that your transmitter covers.

  6. This is a cool concept and does belong on Hackaday, but the Hackaday editors have an ethical obligation to the tech community to make readers aware of how disruptive (hence illegal) this is everytime something like this is posted, with a demo video like https://youtu.be/LnkGHV-tKe0

    Otherwise it increases the number of people who don’t know how bad this is running them for hours, which could eventually lead to the FCC clamping down like crazy on Amateur Radio.

    To not clearly demonstrate the disruption is irresponsible. This lack of warning and education is what everyone in this thread and All The Previous Threads have been angry about.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s