Micro Radio Time Station Keeps Watch In Sync

The US National Institute of Standards and Technology (NIST) broadcasts atomic clock time signals from Fort Collins, Colorado on various frequencies. The WWVB signal on 60 kHz blasts out 70,000 watts that theoretically should reach the entire continental US. Unfortunately for [Anish Athalye], the signals do not reach his Massachusetts dorm, so he built this GPS to WWVB converter to keep his Casio G-Shock self-setting watch on track.

Not a repeater but a micro-WWVB transmitter, [Anish]’s build consists of a GPS receiver module and an ultra low-power 60kHz transmitter based on an ATtiny44a microcontroller’s hardware PWM driving a ferrite rod antenna. It’s not much of a transmitter, but it doesn’t need to be since the watch is only a few inches away. That also serves to keep the build in compliance with FCC regulations regarding low-power transmissions. Heavy wizardry is invoked by the software needed to pull time data off the GPS module and convert it to WWVB time code format, with the necessary time zone and Daylight Savings Time corrections. Housed in an attractive case, the watch stand takes about three minutes to sync the watch every night.

[Anish] offers some ideas for improving the accuracy, but we think he did just fine with this build. We covered a WWVB signal spoofer before, but this build is far more polished and practical.

14 thoughts on “Micro Radio Time Station Keeps Watch In Sync

  1. Nifty toy. I like the “software defined radio transmitter” there.
    I prefer to wear a cheap 12 dollar armitron watch that keeps time accurate enough I don’t have to constantly update it. Yeah, it drifts a few seconds a month, oh well.

    1. These watches keep time with similar accuracy, and don’t have to be constantly updated. But some of us like the idea that we can know what time it is to the second, and bonus points for not adding an extra weekly (or however often) ritual of adjusting our watch.

      When I had one of these, I first tried keeping it on my nightstand (where I’d kept my previous watch), but no signal. Then I took to putting it on my windowsill every night; it still failed to sync about 20% of the time, but that was good enough — it displayed time in seconds, and was quite unlikely to drift a whole second between successful updates at that rate. I didn’t need to build something like this, and I’m not sure I’d have actually bothered (I was thinking about ways to couple the watch to a big antenna, but then the windowsill worked), but if that’s what it took to eliminate my prior weekly ritual of resetting my watch from either time.gov or shortwave radio, it would have been time well spent over the years I had that watch.

      Of course now my watch runs android, and updates its time with NTP, so I don’t need such nonsense.

          1. Yes, “network” time sync. I’m pretty sure it works without data transfer enabled, so it cannot be NTP. Also, I use GSM as a word to describe network type, including 3G/4G.

          2. “Yes, “network” time sync. I’m pretty sure it works without data transfer enabled, so it cannot be NTP. Also, I use GSM as a word to describe network type, including 3G/4G.”

            Too bad 3G/4G aren’t GSM though.

  2. This is an awesome idea. It reminds me of a project I have always wanted to do – get one of the old Western Union telegraph-sync clocks and re-create the daily sync pulse with NTP.

    The big problem has been that those clocks are now impossible to get at a price I can afford. So I haz a sad.

  3. Very cool. I use a less-expensive method to keep my non-WWVB Casio wristwatch in sync: Every time I sit down at my ham radio bench, I turn on WWV (2.5, 5, 10, 15, or 20 MHz, depending on propogation), and zero the seconds on the top-of-the-minute beep. Takes less than a minute, and I’m good to within a second for a week (I checked). Plus, I like listening to WWV. I know, weird. But it reminds me of my childhood, when I first got into radio.

    1. Too bad 2.5, 5 MHz. are normally dead, and 10.00, is marginal.
      I rarely hear 20.00 MHz, but dud copy HI. years ago.
      My analyzer hears 10.0 MHz too well, it picks up clock pulses from my multiple timebases.

  4. Interesting since GPS is hard indoors so I imagine just building an amplifier and tuned ferrite antenna combo to boost and retransmit the WWV signal might have be an option too. But maybe GPS is more precise actually, less propagation delay and variation?

    1. GPS with PPS > WWVB; and WWVB > WWV because it travels via ground wave rather than an unknown number of ionosphere hops.

      The thing that WWVB and WWV have is daylight savings time. Not so with GPS. So it’s more “bureaucrat proof”

      It’;s been a bone of contention for me that my GPS receives several satellites at the same time, and all those satellites have on-board atomic clocks and thus I know where I am in the worlds and know Zulu time to a very percise amount, but still need to set local and daylight savings time manually.

  5. If I am getting what his toy does is it takes a gps signal, perhaps the 1pps and uses that to transmit a fake wwv signal. What you are broadcasting may be good for you, but I hope you are keeping it contained to the immediate area around your watch. People use that signal for all kinds of things and your fake rebroadcast may mess some of them up.

  6. Why not combine both? Watch with WWVB, Anthorn, Darmstadt *and* GPS/Galileo/SKYNET references.
    Apparently the latter have atomic clocks which are actually better than the ones on GPS, as newer technology but this is actually classified knowledge (albeit not any more as found it accidentally while searching for a nearby signal on another band, though the signals are encrypted the temporal order is not and can be referenced to!)

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.