How Much Can You Cram Into A Wristwatch

Creating wearable electronics that are functional and not overly bulky is very, very hard. [Zak], though, makes it look easy. He started his DIY digital wrist watch to see how much he could cram into a watch-sized device. The finished product is really incredible, and one of the most amazing DIY watches we’ve ever seen.

The electronics for the watch include an ATMega328p, a DS3231M Real Time Clock, a Microchip battery charger, and a few resistors and caps. The display is an OLED, 1.3″ wide and only 1.5 mm thick, contributing to the crazy 10mm overall thickness of the watch.

The software is where this watch really shines. Along with the standard time and date functions, [Zak] included everything and more a wrist watch should have. There is an interface to set up to ten alarms on different days of the week, a Breakout and ‘Car Dodge’ game, a flashlight with integrated ‘rave’ mode, and a stopwatch. On top of this, [Zak] included some great animations very similar to the CRT-like animations found in Android.

It’s a fabulous piece of kit, and if any project were deserving of being made into an actual product, this is it.

You can check out [Zak]’s demo of all the functions of his watch below.

45 thoughts on “How Much Can You Cram Into A Wristwatch

      1. Think of a thick VFD, batteries and components on a PCB without a case.
        Something that looks out of place on my small wrist and more like a
        timer for a bomb. ;)

        The power consumption of VFD is probably closer to 0.5W, but it is not
        kept on. With a couple of AA rechargeable batteries, there is enough
        juice for 8+ hours even if you left the display on.

        I bought a $25 plastic Casio watch the day before my flight into the
        U.S. and a friend told me about terorist watch wiki afterwards. :) BTW The
        security guy at Narita gestured me to walk through the metal detector
        when I was about to empty my handful of change into the bucket. The
        alarm didn’t go off. ;)

  1. bravo!

    I love this line:

    “with the voltage divider connected across the battery there would be a current of around 350uA constantly flowing through it, this is a huge waste of power.”

    Pretty interesting stuff about the relative efficiencies of different voltage regulation techniques at those currents.

    Also interesting: the poor quality control of those little displays. Seems like it would a lot of work to unsolder a borked display from that tiny circuit. I will burn mine in before I bury them in an enclosure.

    1. It should not be too hard to desolder. I usually desolder desolder wide SMT parts
      with a #14 AWG solid copper wire (house wiring) over the entire side for conducting
      heat to keep solder molten. I add plenty solder and flux to make sure that all of
      the solder on the pads melts and lift the side of the part while rolling back the wire.

      I have helped a friend to upgrade memory on his router. So desoldered TSOP memory
      chips that way and replaced it with one desoldered from his SODIMM PCB only to
      find out it has less memory and finally solder back the original chip. It still works after
      all that and I haven’t broken a single pad. I guess I should have checked, but I trusted him.

      Even easier if you don’t intend to keep the parts. Chip Quik.would probably make it
      10X easier.

      It would be nicer if there is a way for the software to enable a switcher when the
      watch needs higher currents e.g. OLED so that you get a better efficiency.
      That unfortunately adds a bit of complexity in a already crammed PCB.

      1. Using lots of flux and heat is great for normal projects, but when everything is on top of each other you are gonna start melting plastic and getting flux into areas that are hard to clean.

        1. I second this idea. Or, even if the watch auto-synchronises whenever it’s connected to a computer and assumes the host computer’s time is NTP-synchronised. Or, how about a WWVB receiver? How much more complexity and power would that require?

          Still, this is already an awesome build. I think it’s already better than the Metawatch. (c;

          1. You can get wristwatches (a few here http://www.radioandtelly.co.uk/radiocontrolledtime.html) that synch to the MSF Rugby time signal in the UK (even though it’s now in Cumbria everyone knows it as Rugby), and there are others in Europe.

            It’s on 60KHz at 17KW so has a long range, the signal is just a few dozen slow bits in binary, on-off modulated. You can get chips and modules to decode it, but the watch here certainly has the CPU power, if you could get the radio parts fitted in.

            This would be a much more wristwatchly solution than requiring a tether to update itself. Although it would take more hardware. It would have the advantage of taking care of summer time by itself, and would never need re-setting if the battery went flat.

          2. I agree about it being an awesome watch too. I’ve been itching for a smartwatch for ages. There was one with a nice colour display, small and good CPU, but it only had one freakin’ button! They showed it running a ray-casting demo, yeah, lot of use for a Doom where you can only go left!

            Other stuff hasn’t hit the target yet either. Aussie Dave’s watch, and the recent Bluetooth ones have been tempting. I gave in and bought a cheap Chinese phone-watch for the time being, to keep me sated. But I’ve noticed smart watches seem to be coming along, in a few years there’ll be a few good cheap ones hopefully.

            Ideally I want something with Java and a keypad for games. I just want to play Hearts and Sea Towers on my wrist, really.

            A vaguely related note, but anyone know how to get video converted to work on the Chinese mobile phone watches? It’s a Q8, uses MTK 6226 I think. It claims to be “MP4”, but not the same MP4 everyone outside China knows of. It’s some sort of M-JPEG but all my twiddling with Virtualdub has so far been for naught.

      1. Sorry i left out an important line from that section
        “If the watch is in active mode for an average of 1 minute per day (with a 5 second sleep timeout that would be checking the time 12 times a day) and all volume channels set to minimum the watch should last for around 30 days on a single charge”

  2. I liked the analysis of active and sleep power consumption (with battery life). I was surprised that leakage in the diode and transistors is so high compared to the sleeping cpu and running RTC.

  3. On one hand, it’s a nice build, but come on, a CENTIMETER thick strapped on top of your wrist – I’ll pass.

    I’ll stick with a wrist free Smartphone. My Galaxy S4 is less then 8mm thick and does a bazillion more things.

  4. This makes me sad that there hasn’t been a second revision of the uWatch from Dave Jones at EEVblog. I missed out on those and have wished I got one. This makes me feel like I should just make my own at this point. I want a true scientific calculator watch!

    This design seems like a good step in my dream

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.