Hacker News new | past | comments | ask | show | jobs | submit login
DIY: Cheap, ultra low-power radios that communicate over thousands of miles (nycresistor.com)
87 points by elimisteve on July 25, 2010 | hide | past | favorite | 29 comments



I wonder if this is similar to what Paul Lutus did to communicate from the open seas back to his Oregon home during his sail around the world?

Here is an excerpt from his book:

My ham radio link is working better than expected. Before I started this sail, I spent some time installing and testing ham radios and computers in both the boat and my house in Oregon. I wanted to be able to write a message here, transmit it by radio, and print it on paper in Oregon. At the Oregon end, because a normal person (not a radio nerd) has to be able to use the system, there's a simple "message screen" on display. A person just sits down at the computer and types a message, then presses a key that saves it. The next time I make contact I collect the messages.

http://www.arachnoid.com/sailbook/Chapter_2_--_Oregon_to_Haw...


It should be noted that he is on the 30m Amateur Radio band. I would not suggest doing this unless you are a licensed ham of general or better.

That begs a good question. How many hackers here are also hams? I'm W1OFZ.


Used to be. Honestly don't even remember my callsigns.

I became a Technician when I was in high school because I lived in NYC (went down to Varick street to take the test) but didn't know anyone who could administer the Novice test.

Built a small QRP rig but never made contact with anyone and lost interest.

A few years later I went back to get either my Extra or Advanced...who remembers. I did that more to see if I could copy code at the required rate than for any other reason. Actually I think my last license might be in the safe, I should take a look someday.

I was never that interested in being a ham and talking to other people; SWL was much more fun. I had QSL cards from all over the world.


I used to be. I got my technician class license at 14, just before morse code became optional (which I think they never should have done). In fact, I passed the 20wpm code test for general, but never took the written. My license expired while I was in the middle of graduating college, buying a house and having a kid. I had other priorities on my mind at the time, obviously, and didn't even think of it until the grace period had ended. I want to get back into it, but there are no longer any VE's in the area and it's just a lot of hassle to travel a long distance to take the test. The last time I checked, though, my call sign was still available (KE4BZY), and if I did test again, I'd probably pay the vanity license fee to get it back, if possible.


Low-code Extra here. I've got a couple Altoids tin kits, but I've never put enough time into code to really work CW QRP. I've worked some far contacts with 2-5 watt PSK though.


"I would not suggest doing this unless you are a licensed ham of general or better." Why? Is being licensed required for this?


Yes, the 30m band is allocated to amateur radio operators by the ITU, and this is enforced by the authorities in each member country (e.g. Industry Canada in Canada, or the FCC in the US).

In practice, especially on bands this low, it's mostly unenforceable. Just don't be a jerk: keep your power low, your bandwidth small, and make sure no one is using the channel before you smear your transmission all over it.


a) The FCC love to exert their dominance over everyone and it's illegal to use a radio channel without permission / licensing. b) You would be being an ass to hams who are trying to use the channel to communicate amongst themselves but now it's filled with ununderstandable garbage.


I am, though I haven't done any ham radio stuff for a few years now. My callsign is VE3MTM.


No code tech. Don't really operate, though.


w8lvn, ex-k7qvf, ex-yj0vn.

30m rocks for mobile as well.


K1GF


N7LQX


KF8QE


n2bmw


Reading this thread is giving me the bug again :-)

I might just go into the basement and put together a simple 40 meter receiver tonight just to see if there's any activity there and maybe build a small transmitter later.

ISTR that the FCC went to non-expiring licenses. My last one is from the late 80's/early 90's. Anyone know what the likelihood is that I still have a valid license?


3600 miles, 100 milliwatts, 5Hz of spectrum near 10MHz, rooftop mounted dipole, $25 of parts to make radio (including obligatory Altoids tin) => 0.05 bits per second.

[Edit: to lose that extra zero. Thanks Joe.]


Make than near 10MHz.


hey there i'm kd8mek, i wrote a blog piece about software radios and such the other day http://verily.posterous.com there are a billion of these qrp small radio blogs out there, all kind of people dreaming up all kinds of things


While it is interesting I can't really get excited because to me a QSO is a two way conversation. Far more interesting to be running a few watts and be having two way conversations with people instead of being merely captured on their computer screens.


Why do they transmit a square wave, instead of just short and long dashes of a single frequency? Only the top of the square wave is the signal, the bottom is the 'negative' of it.


They are not transmitting a square wave. The square wave is the drive waveform that is modulating the transmitter. If you look at the square wave shot, you will see longer "high" periods (dash) and shorter "high" periods (dots).

In classic "CW" (continuous wave, aka. Morse Code), it would be keying the transmitter on and off. When the transmitter is on, it is transmitting a pure sine wave at the Tx frequency (10.140015 MHz). When the transmitter is off, it is transmitting nothing (duh).

Looking at the schematic, the kit in question is transmitting continuously and modulating the carrier to be slightly higher or lower in frequency based on the square wave modulation drive.

For a direct conversion receiver[1], (heterodyne[2] works the same way, but uses intermediate frequencies to improve selectivity) the receiver has an internal oscillator that is tuned to approximately 1000 Hz difference from the transmitter. By mixing the two signals in the receiver, it demodulates the transmitted signal into a 1000 Hz tone when the transmitter is transmitting and nothing when it is not transmitting (CW) or higher and lower pitches with the kit from the article.

The CW on/off modulation is what you hear (simulated) in the WW2 movies with the beeping ("dit" and "daaah") sounds.

[1] http://en.wikipedia.org/wiki/Direct-conversion_receiver

[2] http://en.wikipedia.org/wiki/Heterodyne see also http://en.wikipedia.org/wiki/Superheterodyne_receiver


Because it would be very difficult to decide if your box was not transmitting or just unable to be heard at any given time. The square, spread over a large time allows you to say with confidence, "yes I have received and yes its the bottom/top of a dot/dash".


Are you involved with this as an amateur, or taking a stab at coming up with a reasonable answer? I would like to know the actual reason, and so far Google hasn't been helpful.

I am starting to think the square wave is an artefact from the circuit design.


Hi I am the designer of this circuit ( http://www.hanssummers.com/qrsskit ) and can answer that question. I've just seen the thread on this.

There are various possible "modes" in QRSS and certainly plain on/off keying as you suggest, which is in fact just very slow morse code, is one of them.

The mode I am transmitting here can be known as FSK/CW. CW (Morse Code) in a Frequency Shift Keyed (FSK) way. As you say, the top level is the "key down" ordinary morse code.

The reason for this is NOT just an artefact of the circuit design! The circuit is designed this way for a reason! I could just have easily designed it such that it keyed the signal on and off, like real morse code but much slower.

There are several reasons why FSK/CW is more suitable than plain on/off keying. 1) and 2) below are the most important.

1) Readability: Many kinds of interfering signals on the band are just a plain carrier. It might be some mixing product coming out of your computer monitor, or a harmonic signal of some oscillator in your neighbour's TV, or plain carrier transmissions from great distances. A weak carrier, drifting in and out of visibility, can look VERY much like a weak on/off keyed CW signal in QRSS. Experience of QRSS experimenters has shown that it is much easier to read the "squarewave" FSK/CW style transmission. It appears to be more resilient to interference and weak signal propagation conditions.

2) Chirp. This is a temporary shift in frequency which occurs on key down, in an on/off keyed CW transmitter. That makes the end result much harder to read, less resilient to interference, and just plain ugly. See some examples at http://www.hanssummers.com/qrss/qrssqrv.html . This "chirp" can be very hard to get rid of in such simple circuits. It may only be a few Hz of frequency shift but it is very ugly and visible on QRSS transmissions. Having the transmission continuously ON and just shifting its frequency slightly, avoids the chirp problem.

3) Drift: as the oscillator components and crystal heat up, the frequency can drift a few 10's of Hz. On key down, components will be transmitting some hundred or so milliwatts of power and can slightly warm up, changing the frequency ever so slightly but visibly. Again, a problem that is not so easy to resolve in such a simple circuit. Keeping the transmission permanently ON solves that problem too as once the steady state thermal equilibrium is reached, everything stays there.

I hope this helps to explain it - you can read more about my QRSS experiments at my web page http://www.hanssummers.com and let me know if you have any more questions

Hans http://www.hanssummers.com


Why does this nonsensical reply have 2 points ?


Hmm. I seem to have answered the wrong question. I answered "why does it look like a square wave on a spectrum analyzer over a very long time instead of just appearing and disappearing?". Even that I did badly.

Sorry. Hip-shot from the old iphone....


I wonder if this could be combined with UWB-like principles: you encrypt the clock at both ends and only look at the signal at certain timestamps.


Hmm, interesting.

Makes me think about an alternet for when they switch the internet down.




Consider applying for YC's Summer 2025 batch! Applications are open till May 13

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: