Tuning Fork Clock

2,244

4

18

Introduction: Tuning Fork Clock

Some time ago I succeeded in making an oscillator with a (cheap) 440 Hz tuning fork. See https://www.instructables.com/id/Tuning-Fork-Osci...
I guess this can be seen as an inevitable follow up on that project. What else would one use a stable frequency for, than for a clock. So that's what I did.

EDIT: The code is now available on GitLab:

https://gitlab.com/WilkoL/tuning-fork-clock

Step 1: Mounting the Tuning Fork

As I had another tuning fork available (more on that later) I didn't have to take it out of the oscillator. So filed off the metal lump at the end of the new tuning fork and cut thread on it, just as I had done for the first one. Then I made a mounting block out of some scrap metal so that I could mount it on a back plane.

Step 2: The Dome

Now clocks are sometimes placed in a glass dome, probable to prevent dust from coming in, or just because it looks nice. I wanted to have that too. I found one with a lamp in it in a shop called "Action" (that is a kind of pound-shop or dollar-store) After removing the lamp-socket I had a perfect place for my clock. I wanted to have the fork clearly visible, that meant that the clock display itself had to be small but still visible from a distance. I settled for a TM1637 module with 1.4 cm large seven segment displays. The ones I had available were white. That meant that later I could give them any colour I liked by adding a piece of coloured plexiglass.

Step 3: Electronics

The electronics is for the most part the same as for the oscillator I made before. Added to it is a ATTINY13 that takes care of dividing the 440 Hz to one pulse per second, keeping track of time, checking the adjustment buttons and controlling the TM1637. The software is made with AVR Studio 7. I don't think it is possible to use the Arduino software for this micro controller as it have very little flash memory (1 kByte), you could use an ATTINY85, that one has plenty flash memory available (8 kByte).

Step 4: The Right Frequency

Just as I saw with the oscillator, the tuning fork does not produce 440 Hz exactly. I think that for "normal" use such as tuning a guitar it doesn't really matter but for a clock it does. Luckily both my forks were tuned too low. So I could file a little bit off the tines to raise the frequency.
While doing that I accidentally bumped against the handle and broke it off the fork. Now I had to disassemble my oscillator to continue the clock. Unfortunately it had a smaller thread cut on to it, so I also had to make another mounting block. Lesson learned, those cheap tuning forks are not very sturdy.

In the end I managed to get it to produce almost perfect 440 Hz, at 25 Celsius. With rising and falling temperatures the fork lowers en rises in frequency noticably, but not so much as to make the clock unusable.

Step 5: Noise

What does make this particular clock unusable though, is the noise it makes. I mounted the fork on a big piece of plexiglass, that works as a very efficient sound-board. Also the plexiglass is so big that is hard to prevent it from touching the glass dome, again producing more noise. I tried to reduce noise with pieces of cotton but that isn't very effective.

Step 6: Conclusion and Advice

So in the unlikely event that you want to make a tuning fork clock, use a small but sturdy place to mount the fork, make sure it doesn't touch anything and try to isolate vibrations from entering into and through the base. I might do those things to my clock later, but for now I'll leave it like it is and only switch it on when I want to show it to someone or when I just feel like looking at it.

Be the First to Share

    Recommendations

    • Make it Glow Contest

      Make it Glow Contest
    • First Time Author Contest

      First Time Author Contest
    • PCB Challenge

      PCB Challenge

    18 Discussions

    0
    zanod
    zanod

    6 months ago

    It's beautiful, Wilko. How are your mechanical skills? Have you thought of making a ratchet wheel, so you could drive an analog clock movement, like the Bulova Accutron used? With a 440Hz fork, you would need 440 ratchet teeth per second, and a primary wheel with that many teeth is a non-starter. You would have to have fewer teeth on the primary wheel, and allow it to turn much faster than once per second, then gear it down.
    One of the endearing things about the Bulova was that the second hand swept the face at constant speed, rather than advancing one second at a time.

    0
    WilkoL
    WilkoL

    Reply 6 months ago

    My mechanical skills are bad. We have the tools, lathe, mill, taps, etc. But I am worse than most beginners. My kids went to a school for "mechanical tool makers" https://www.lis.nl/ so they can help me when needed.
    But for the tuning fork clock I didn't need them, making that is (just) within my capabilities. So I won't be making a Bulova Accutron anytime soon. :-)

    Glad you like my clock, impractical as it is!

    0
    zanod
    zanod

    Reply 6 months ago

    Oh- I truly do like it. I like timepieces in general, and have spent a lot of time looking at those beautiful hand-made wooden gear clocks. Google "Wooden clock Clayton Boyer". I'm quite fascinated with Propeller clocks as well - and POV globes (as you know).
    When I was in tech college in 1964-1967, nearly everybody in the class had a "pet project" - usually not an actual project, but something we liked to talk about. The lecturer (Mr. De Klerk) had an idea to make a camera with no mechanical shutter (imagine that!) My pet idea was a tuning fork oscillator - so although I lived my life without making one, I was very interested in yours. I think it was around that time that the Bulova came on the mass market. I remember seeing them in shop windows, though my finances couldn't stretch that far at that time.

    0
    WilkoL
    WilkoL

    Reply 6 months ago

    About those "Wooden Clayton Boyer clocks" , do you know Clickspring on youtube? He made a series of videos of him making a VERY beautiful clock out of brass. He is a master machinist.


    0
    zanod
    zanod

    Reply 6 months ago

    Wilko, you said you have a lathe. Perhaps you could knock together one of these over a weekend?


    I look forward to reading the Instructable ;-)

    0
    WilkoL
    WilkoL

    Reply 6 months ago

    It will have to wait until I finish redecorating the Sistine Chapel....
    Oh dear, that thing is incredible, the maker surely now locked up in an asylum for the insane.
    And I thought this was already amazing:

    0
    zanod
    zanod

    Reply 5 months ago

    Hi Wilko, if you're interested, there are three videos on the total dismantling of that skeleton clock, starting here ...

    I was amazed how modular it all is. Whole sub-mechanisms come out of it very easily.

    My Tektronix 2465 is in pieces at the moment. It had a diagnostic failure on switch-on, but would generally work OK if I simply cleared the error. Then the triggering started to get iffy - though if I switched it off and on again, I could usually get that working as well - then the channel A dies, and after a whole day of trying to find out how to do it, I eventually got the attenuator out. That's the module with the channel input BNC connector on it. I can see the fault, and it's fixable, but there's some pretty inconvenient soldering to do down little holes. I hope to get it working again tomorrow.

    0
    zanod
    zanod

    Reply 5 months ago

    Hi Wilko, if you're interested, there is a series of videos on the dismantlement of that skeleton clock, starting here

    I was amazed to see how modular it all is. By the end of part 3, the basic mechanism is reduced to bare bones.

    I have my Tektronix 2465 in bits at the moment. It has been giving a start-up diagnostic failure for some time. At first, if I hit the "Go away" button, it would go away and the scope would work OK - but now, it has fairly frequent issues with the triggering, If I power it off and on again, I can usually get the triggering working - then channel-A gave up altogether and I had to take the attenuator out - the bit with the BNC connector on it. It turned out to be much more involved than I thought at first, and without any instructions, it took me a whole day to figure out how to get at it. There's a bit of close-quartrs soldering to do down awkward little holes, too. There are a few electrolytics on the triggering board that I'm going to replace, so I hope that will fix it. I replaced all the PSU electrolytics about 18 months ago because one of them blew up - but the scope dates fom1984, I think, and electrolytics have a life span.

    I've been picking out some of your low-level "LL_xxxx" drivers and searching for them in the reference manual for my chip (STM32H743), but they're not all there. In fact, quite a lot aren't. I've also been looking for any kind of instruction on it - free is best, but I can't even find a paid course.

    If you feel that you have more than enough to fill your time without getting involved with other people, I would understand and respect that viewpoint, but if you have a bit of time to put me on the right road, I'd appreciate if you'd write to me directly at mijewen@googlemail.com

    Are you into any new project?

    All of the Best

    0
    zanod
    zanod

    Reply 6 months ago

    It's lovely. I don't think it's immensely difficult to do something like this, but my word, it takes a lot of time and patience - and if you don't have the jigs (like his filing machine) then you have to make those first. You can't do it without machines.

    I used to enjoy my wood workshop, but I have neighbours now who object to the noise, so I have stopped. My best effort was

    0
    zanod
    zanod

    Reply 6 months ago

    Wilko, Don't look at the whole video above - it's embarrassingly too long. I wanted to edit the entry as well, but apparently the forum doesn't allow for editing after it is committed. Zoom the vid to about 11:00.

    0
    WilkoL
    WilkoL

    Reply 6 months ago

    Sure I watched it all, and It is absolutely beautiful! (I kept thinking the word "pantograph") Do no stop doing that!
    Your neighbours aren't always at home are they?

    Our lathe also isn't very quiet, especially when you switch it on (which I do a lot). And the bench I made for it is bolted to the wall between my garage and the neighbours livingroom so I use it only when they aren't at home, or when I know that it is a small job. And not at 06:00 on a sunday morning ;-)

    Ah, and you aren't a fellow Dutchman... Wales, nice place, I've been there a long time ago. I remember staying in a place called Cwm Craig Farm (in Herefordshire which is or isn't Wales...) and having problems reading Welsh and even more pronouncing it. In Dutch you cannot pronounce words without vowels. So Cwm was strange (I think it was something like "coom")

    0
    WilkoL
    WilkoL

    Reply 6 months ago

    I too wanted to make a tuningfork oscillator for a long time. And I tried several methods, It was only after I thought of using an optical way of detecting the vibrations that I had succes.
    After that my brother challanged me to make a wineglass oscillator, which I did. But I didn't make a clock out of that, it is too cumbersome.

    There is (at least) one timepiece I still want to make, one with a hourglass. I have two of those lying around, one is too big, it takes 20 minutes to complete, but the other one takes about five minutes, someday..... someday....

    Mr. de Klerk? Sounds very Dutch, are you also a "Nederlander" ?
    And in college in '64, I was three at that time, college was still in the far future.

    0
    MatthewD225
    MatthewD225

    10 months ago

    How much power does this clock take to run?

    0
    WilkoL
    WilkoL

    Reply 10 months ago

    As it is, a little less than 300mA at 5V.
    When you put some more effort in it, I guess it is possible to run it with 100mA at 5V. The coil uses most of the power, a better coil will help. The rest of the circuit uses just a few mA.

    0
    nheck
    nheck

    1 year ago

    wow, very nice job! you could make the dome air tight and make a vacuum inside to get rid of the noise.

    0
    WilkoL
    WilkoL

    Reply 1 year ago

    Well it might help just a little, but most noise is contact-noise. The blue perspex plate, on what the tuning fork is mounted shouldn't make contact with the dome or the base. Ideally it should float freely in the air (or in a vacuum).

    0
    nqtronix
    nqtronix

    1 year ago

    Huh, I never though about the acoustic noise a low-frequency oscillator generates, but it's so obvious in hindsight. I guess that's one advantage of the modern 32kHz watch crystals beside their reduced size.

    Nevertheless I appreciate you build as it is. Being able to see what makes a circuit tick - in this case quite litterally - has always a certain charme to it. So thanks posting this :)

    0
    WilkoL
    WilkoL

    Reply 1 year ago

    Ah yes, well apart from being just a tad too high in frequency for humans, a watch crystal (that actually is a tuning fork) is ususally powered by max 1 uW. Also it is shielded, so there won't be a lot of accoustic noise coming from that :-)

    BTW, I know that there are (were?) watches made with tuning forks by a company called Bulova. I think they are called humming watches.