Clockmaking: 1

This entry is part 1 of 2 in the series Clockmaking

As most of you know, I’m working on (another) book about time, The Clockless Clock, which I’m serializing on the Breaking Smart email list. In the spirit of getting a hands-on understanding of the subject, a while back I decided to build an actual clock as a semi-homemade project. Maybe more than one, but let’s start with one. This one, the ROKR 3D wooden mechanical pendulum clock:

I bought the kit ($45.99 on Amazon in case any of you wants to join me in the build) several months ago, but only just started building it. The thing is almost entirely laser-cut parts on several sheets of wood, so the first order of business was over an hour of painstakingly popping out the parts. It’s like a masochistic version of popping bubble wrap.

After over an hour of part popping, and fortunately no mistakes or damage, I was left with this surprisingly large number of parts (the pile of interesting looking crap on the lower right is the leftover negative space wood).

I’ve looked inside clocks before of course, but a fully assembled clock doesn’t give you a sense of the sheer number of parts involved. I think there’s at least a couple of hundred here.

I’m also going to be reading David Landes’ seminal Revolution in Time alongside. I’ve owned the book for nearly a decade now, and browsed it enough to be able to pretend to have read it, and cite it while writing Tempo, but it’s time (heh!) to actually read it now. After all, I’m trying to write a book based on the premise that the clock ruled us for 400 years and is now being eaten by software, so I should know the story in a lot more detail than I do.

As of now, I’ve assembled 2 of the pieces of the clock. The first part is some sort of ratchet assembly, and the other is a stepped gear. Putting this stuff together is a good deal more delicate than putting together IKEA furniture. It’s not hard exactly, but it takes patience. I can easily see myself damaging a part irreversibly out of impatience, so I’m going to approach this as a meditation as much as a construction project.

We’ll see if I make it through with something that looks like the picture on top, and hopefully actually runs. I could probably binge-build this in a day, but I’m going to force myself to do it over multiple short sessions, so I can meditate better on it, and guess at the logic of the mechanism as I go. I’ll be able to see if I guessed right at the end. It’ll be like solving a murder mystery, except in reverse, since the clock should wind and come alive at the end.

Ratchet #1

Stepped gear #1

For starters, stepped gear #1 has 30 and 10 teeth, so it’s a ratio of 3. The ratchet also has 30 teeth and is on a separate axis, so I’m going to guess it meshes with the other gear 1:1. I know the ratchet axis attaches to the winding key, so it will probably attach to the mainspring.

We’ll see if I’m right.

It’s kinda fascinating to do this kind of detailed, patient, precision assembly work with no clear sense of how a clock is supposed to emerge out of this. I don’t think I’ve ever built a model this complex. While I have a basic conceptual understanding of how mechanical clocks work, with gears, escapements, ratchets, and so forth, I’ve never actually taken a detailed look at the innards of one, or even taken one apart. So putting one together should be fun.

I was going to live tweet this project, but that would interrupt the assembly sessions too much, so I’ll be live-blogging it instead.

Series NavigationClockmaking: 2 >>

Get Ribbonfarm in your inbox

Get new post updates by email

New post updates are sent out once a week

About Venkatesh Rao

Venkat is the founder and editor-in-chief of ribbonfarm. Follow him on Twitter


  1. I have recently discovered the joy of watching YouTube videos of watchmakers repairing and restoring wristwatches and pocket watches. I’m finding it deeply soothing and endlessly fascinating. All the while understanding almost nothing about how/why the timepieces work.

  2. As a comparison to clock-based time, you maybe should read about the prior cyclical experience of time. Clocks were built on a cyclical sense of time in being built with a circular clock face. But it’s interesting that, combined with the dominance of text with the printing press, that time became increasingly a linear experience. Anyway, I’d suggest checking out Circles and Lines by John Demos.

  3. says

    The nefarious ninnies are able, even at screen res, to get a partial finger and palm print from you holding the ratchet.

    In time, this may be a problem. Just sayin’.

    And of course, thanks for your time – which is really a proxy for ‘life’. Contact me if you want a sideways flashbiulb view of lifetime.

  4. I’m happy that you are putting your hands on the guts of a clock.
    My understanding is that pendulum style clocks are mostly gearing to slow the fraction-of- a second pendulum cycle to register as minutes and hours.
    The ingenious part is the mechanism that takes a bit of the spring (or weight) energy and gives it to the pendulum so the pendulum doesn’t slow down.

    Which is all fine in a conceptual way, but has not much to do with making a functioning clock.

    My first job was making panel meters. Volt & ammeters, which shared the point & jewel pivots of modern clocks, for example. They also needed to respond to very small impulses of electric current (magnetic field).
    Consequently, much of what we did was roll masking tape stickum onto a needle, and using an eye loupe, pick up dust and hairs from inside the meter that would cause it to hang up.
    It is all about the friction.

  5. Wait are you building this without the instructions?