Build Your Own Radio Clock Transmitter

Deep in the Colorado foothills, there are two radio transmitters that control the time on millions of clocks all across North America. It’s WWVB, the NIST time signal radio station that sends the time from several atomic clocks over the airwaves to radio controlled clocks across the continent. You might think replicating a 70 kW, multi-million dollar radio transmitter to set your own clock might be out of reach, but with a single ATtiny45, just about everything is possible.

Even though WWVB has enough power to set clocks in LA, New York, and the far reaches of Canada, even a pitifully underpowered transmitter – such as a microcontroller with a long wire attached to a pin PWMing at 60kHz – will be more than enough to overpower the official signal and set a custom time on a WWVB-controlled clock. This signal must be modulated, of course, and the most common radio controlled clocks use an extremely simple amplitude modulation that can be easily replicated by changing the duty cycle of the carrier. After that, it’s a simple matter of encoding the time signal.

The end result of this build is an extremely small one-chip device that can change the time of any remote-controlled clock. We can guess this would be useful if your radio controlled clock isn’t receiving a signal for some reason, but the fact that April 1st is just a few days away gives us a much, much better idea.

72 thoughts on “Build Your Own Radio Clock Transmitter

      1. You’re mistaken. Your claim that “all digital on the wall clocks” are synced to a radio transmitter is wrong. Unless the clock actually advertises that as a feature then you’ll find that it is just a basic quartz digital clock chip or glop top and no external synchronisation.

          1. Probably NTP. Most telcos operate Rubidium standards to control the frequencies in mobile phone base stations. But to be honest I don’t know for sure.

          2. I can second what tekkieneet stated about cell cites using GPS receivers. I install and commission the equipment in cell sites. There is usually a GPS receiver for every technology (LTE, GSM, UMTS, etc).

  1. I would like to propose an entire HaD category in the theme of “with a single ATtiny45, just about everything is possible.” The Tiny25/45/85 is an amazing chip and is worthy of its reputation as an eight-pin Swiss army knife. This also sounds like it could be a great new contest in the format of the 555 contest!

    1. This. Most “atomic” clocks only enable and poll the radio once a day. The is usually done late at night when longwave reception is at its best. This is why you need to set an approximate time on many clocks before they will sync with WWVB.

        1. The transmitter broadcasts a DST change schedule in advance, so the clock should be able to change to DST at exactly the right time even if it only checks the radio once per day.

      1. I never had to do this. It is easy why: When you put the batteries in the clock it, listens for the signal until it has a valid time information (or it gives up after 5-10 minutes). Then it resyncs once a day.

  2. As a tool to show how Radio Clock works- wonderful work . For the love of Darwin… be mindful of how far these signals can really go. Nope- the inch of wire in a safe demo is not that liable to go major distances at these frequencies. My concern is some cretin who uses a bit more power or better antennas- or just gets Darwin Lucky in re-radiation coupling to other wires!

    I can sketch out a credible risk of a Ham Shack/SW listener or someone with old twinlead fed roof antennas etc having that parasitic coupling. And yes- that gets their clock whacker signal out into danger space. It’s why we used to confine some work to a REAL shield room.

    Yes, it is Damned illegal to screw with such signals and even if you think you got away with no damages- it’s a when, not if about the epic doom. Clock and Location systems can indeed pose a life safety issue in ways approaching impossible to predict.

    There’s a layer beyond mere legal compliance. It’s called thinking of consequences. There’s a quite serious potential for Very Bad Things. As these signals can and DO propagate over astonishing distances, with a scary real potential of you pranging stuff on the other side of town or farther.

    It’s not “just” Lulz to screw up our delicate ballet of interconnections It’s risking a lot more.

    if someone at minimum is late for a schedule as the worst, they’re quite liable to NOT wish you well… If someone DIES from several scenarios that we are supposed to be mindful of preventing, it’s going to not only be jail for the perp-it’s going to rain hell on *ALL Hacking of RF stuff! *

    A proof-of-concept in a sandbox, go for it. A small area of Maker Faire equally cool. Do this in the wider world and you lose all honor=you’re a vandal/belong in jail for reckless endangerment.

    Laugh at my reality check , but be careful, eh?

    1. It would be an extremely far fetched scenario to see someone die as a result of this hack. First of all, 60 kHz signals don’t travel very far unless you have a huge antenna and a lot of transmit power. But most importantly, people at the risk of dying because of reading a bad time should not depend on a single clock that could stop when the battery runs out, or the mains power is interrupted.

    2. I agree with you that messing with systems where legality is a concern is probably not a good idea. But can you give a good example where someone is going to die(or even get hurt for that matter) because their clock suddenly changed time? I’m pretty sure the only clock I trust is my cell phone and even then I’m not going to die if its not correct and also it’s not even affected by this.

      Seriously, not being a smart ass, I’m curious at to a specific scenario that could cause a dangerous situation.

      I would like to think that such devices wouldn’t rely on a system like this to keep people out of harms way.

      1. It’s ye olde butterfly effect. A police officer has two clocks, the radio controlled one being his backup. His usual clock fails, the radio controlled one is off by a few hours. He wakes up late and during his drive to work he isn’t in place to prevent an assault on a street corner which he would’ve caught if he had woken up in time.

        It’s not about how low the risk is, it’s about not being a c_nt and making it a risk in the first place.

        1. Haha, I hope this is a troll comment. Seriously.

          I understand not being a -insert highly offensive word-, but you just proved my point from the fact you had to refer to the “butterfly effect”. An effect that is based around making everyone at blame for anything that has ever happened in the entire known world.

          There are far worse *legal* pranks you could do to someone than to change the time on their clock. Again I’m not denying that you shouldn’t be messing with people(unless its your friends or they have it coming to them), I merely started this discussion because of the way over the top statement made that this could possibly harm someone. Without creating some absolutely ridiculous slew of events, it would never be the case. People don’t rely on time like that, just look at how everyone is late now-a-days. Time means nothing.

          I was hoping that someone could provide a real situation when this would affect something; like maybe some old school timing device that still relies on these signals to, I don’t know, deliver medicine to someone at a specific time – now that would be dangerous. But the truth is, there isn’t anything like that. Prove me wrong though.

      2. The best I can think of is you’re messing with this in a room of which your neighbour has their clock on the joining wall.
        They could be using that clock to tell when to do something like take pills or test blood sugar.
        Some of these radio clocks are sold as ‘science clocks’ or ‘never tells the wrong time’ so people could be misled by the marketing to thinking they can rely on it.

        1. That is a very valid response. There are most definitely some people who would buy something like that, maybe an elderly person who is taking meds, and rely on it. Though I’m obviously hoping that is rarely the case.
          Its funny you mention blood sugar because I am diabetic, and there are so many clocks around me I would know pretty quickly if one was reading incorrect. But I’m not too worried about time accuracy in my case. Again though my cell phone is almost always number one, then computer.

        2. I used a radio controlled (switching clock/timer) for some years as an alarm clock (controlling my stereo). A few times the stereo was blaring away at 1 or 3 in the morning, because of mis-reception of the time signal. It was a few hours off. So this shows nothing is completely reliable and should not be used alone if an error could cause a life threatening situation.

      3. Example… Say you are Sammy Jenkins, and your wife needs her insulin shot at 3pm. And because you keep changing the time to 2:59pm, and sammy can’t remember the last 2 minutes, and his wife thinks he is pretending then bam, you are responsible for her overdose. sinner.

    1. NIST provides their own range maps for WWVB, with a threshold of 100µV/m as a function of time of day:

      You don’t need a particularly strong transmitter to overpower that, nor a particularly well-tuned antenna (wavelength of 60kHz is 5km). Most consumer WWVB receivers I’ve seen use a loop antenna, so an adjacent matching loop antenna (forming a lossy transformer) is probably the best way to override only the device you want overridden.

  3. Can we be mindful of what responsibility our skills need to have!

    I have personally seen a cell phone in a Chicago Basement Garage link to a Benton Harbor Cell System. Miles per watt arguably up there with this chip’s signal deranging some mere clock a few miles away or closer . Arguing with which band is a fail- it’s watts per mile under consideration.

    How could someone die? How many dangerous things do you know of with a clock in them? And thus a system gets built with potential for a SPOOFED clock to cause disaster. The handful of ones I know about with any chance for risks are bad enough to contemplate.

    Envision the number of systems that we may not KNOW of that are now using such time signals.

    Scared yet? There’s been a move to use over-the air signals in odd places and not all of those uses are sane to let become public knowledge. Sometimes it’s responsibility to not misuse our knowledge. Not censorship- It’s being aware that we can crash systems and kill people too easily to risk it avoidably. Don’t forget- it’s easier to not think first. It’s harder to be a responsible user of our powers but it’s our duty.

    And suspecting we’re kept in the dark can anger us on freedom grounds.

    Sometimes it’s not even deliberate obfuscation so much as no one suspecting signal tampering would ever be a reality. See- clock fail is planned for, not clock spoofing…So maybe the coder never planned for someone bollixing the timedata. Which causes a risk for unimagined havoc that might never happen till some yutz mungs the time stream data. I hope to literal HELL that this stays in theoretical territory and we avoid damages or deaths. That’s my point. SO FAR, we’ve had few reportable incidents of Hackerdom folks doing more than exasperating stupids.

    Many Decades, ago at least one nameless city had reversible traffic lanes controlled in the 11 meter band. More than one whack job tried to exploit the systems. And apocryphal claims of their results ranged from deaths- to mere wasting of the perp’s time. I suspect that a lot was kept deliberately obscure and or disinformation zoned.

    But- yeah- there is a damnably plausible risk to me and as said- don’t risk lives for lulz.

  4. I am surprised that those LF signals work at all in a world full of SWPS’s, specially the cheap ones form China. Longwave and the lower 2/3 of the AM band are just awash in buzzzzzz. Now they won’t replace streetlights on sked, they wait till someone complains to an obscure part of their website. Each light buzzes heavy trying to restart, wiping out AM radio for a 9 block area!

    1. They are not supposed to interfere with radio. That’s why FCC part 15 certification is required for anything that operates above 9kHz.

      Not all Chinese made stuff passes or have FCC/UL/CE etc certification. US Custom people should be checking for shipments from China for those types of certifications instead of superficial things that round corners or yellow edges.

    2. For several years I had no reception of the (European) 77,5kHz time signal in my apartment. Going to the highest point under the roof gave some success. Some years later the interference was gone. Obviously the interfering device was finally sent to the recycling center.

  5. Great hack. I tried to hack the Frankfurt’s DCF77 and it worked too. I used 10 MHz crystal oscillator but it might be OK with the internal too. Only it might be a good idea to calibrate the OSCCAL register because the precision of the internal 8 MHz oscillator is poor (mine got 8.33 MHz at 5 V).

  6. All the exams that I have taken in school and college are moderated with radio controlled clocks. I always wondered if it would be possible to creep the clock back a few 10s of minuets over the several hours that the exam takes place with a spoofed signal for extra time. I may have to experiment with my own clocks out of curiosity.

  7. EVERY clock I have ever encountered is *NOT* “radio controlled” !
    They run off (as someone else pointed out), an INTERNAL quartz
    oscillator – and invariably powered by a AA battery !

    Aside from tech geeks, who would pay extra for a “WWV” capable
    receiver in a clock ? answer – NO ONE!

    On the other extreme, most “mission critical” applications (read,
    public safety or business IT networks), get their timing (again as
    pointed out by others) from GPS satellites that are picked up by
    very expensive 19″ rack mount receivers (think TRAK9100) that
    provide an NTP stratum-2 signal to the network.

    Basically this ‘hack’, is a waste of time, and more fodder for the
    clueless legislators and media to sensationalize and paint another
    “evil hacker” scenario.

    1. Radio controlled clock are quite common, give a second look (I own one). Don’t explain about quartz oscillators to hackaday’s reader, that’s pointelss: they know better. Radio control is about periodical resyncing, not time keeping, you missed the point.

  8. Did they stop transmitting recently? I thought I heard in the news they were going to and during a power outage Wednesday 02/22/2023 when my emerson clock came back on it did not set itself. The cell battery needs replaced but that is only to keep the alarms and the timezone. I may need to build a transmitter and get the time from ntp.

    1. THIS is exactly what I am looking to do, as well. My clock doesn’t receive the signal well. I just want to build a little transmitter which gets NTP time and rebroadcasts it. Did you have any luck? I sadly don’t have the technical prowess of the many other people here to develop this.

Leave a 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.