3D Printed Tourbillon Clock

3D printed clocks have been done before, but never something like this. It’s a 3D printed clock with a tourbillon, a creative way to drive an escapement developed around the year 1800. Instead of a pendulum, this type of clock uses a rotating cage powered by a spring. It’s commonly found in some very expensive modern watches, but never before has something like this been 3D printed.

3D Printed Clock[Christoph Lamier] designed this tourbillon clock in Autodesk Fusion 360, with 50 printable parts, and a handful of pins, screws, and washers. The most delicate parts – the hairspring, anchor, escapement wheel, and a few gears were printed at 0.06 layer height. Everything else was printed at a much more normal resolution with 0.1mm layer height.

Because nearly the entire clock is 3D printed, this means the spring is 3D printed as well. This enormous 2 meter-long spiral of printed plastic could not have been printed without altering a few settings on the printer. The setting in question is Cura’s ‘combing’ or the ‘avoid crossing perimeters’ setting. If you don’t disable this setting, the print time increases by 30%, and moving the print head causes the plastic to ooze out over the spring.

There’s a 26-minute long video of the 3D printed tourbillon clock in action that is horrendously boring. It does demonstrate this clock works, though. You can check out the more interesting videos below.


Thanks [LupusMechanicus] for the tip.

27 thoughts on “3D Printed Tourbillon Clock

  1. I’m in love. I’m in the final stage of constructing this guy’s first clock, which has been on Thingiverse for a long time already, but I’ve always wanted a tourbillon. Just didn’t want to pay the $10K+ for a watch that used one. :-)

  2. This is truly impressive. I was certain that it would have needed a steel mainspring, but it has a printed plastic one instead! Remarkable.
    However, he does mention the problems associated with that plastic spring. I wonder if he intends on replacing it with a steel spring, or if this clock will remain a novelty.

    1. He said the power reserve is 30 minutes…plenty for a demonstration piece. Mount it on a wall and power it with weights!

      I was wondering, if it tan continuously, how well the escape wheel and pallet fork would hold up over time.

        1. I know very little about plastics and NOTHING about how 3D printed parts stand up. Well designed gear trains in appropriate applications made with plastic MOLDED gears are very reliable but wonder how the surface finish of a 3-D printed part would stand up. Finish the gears with an Ingold fraise?

          1. This is pretty Rad. Definitely on the make list. The wear on the printed parts wouldn’t be toooo bad. There isn’t much force being exerted on them. The only accelerating factor might be that they’re just quite fine. The printed spring is pretty cool. Most plastics will fatigue after a few thousand cycles but the situation might be helped by using something like an impact modified PLA. Wouldn’t be too hard to make a spring steel spring – or just salvage one from somewhere.

          2. Wow! Someone else actually knows of the Ingold fraise here. That’s some deep knowledge there, nice.
            You’ve been reading Daniel’s Watchmaking or Archie Perkin’s Antique Watch Restoration, haven’t you?
            Actually- the problem that tool was created to correct would not be as present here as you think.

            I should think the layers here would act much as the smoothing lines left by that tool on traditionally
            cut gears do- the lines are there, but as a function physically of the layer height itself. Which should
            mean the printed object would look scale to the end result of a brass gear cut across the blank for
            the tooth, then smoothed in the direction of running using the fraise, with a scale up in roughness
            equivalently. Remember- that tool leaves extremely fine lines in the brass gears, extremely fine-
            but they are there. The effect is to remove the friction as brass gears run by removing the lines
            rubbing 90 degrees from the direction of force transmission in the gears created by traditional
            gear cutting methods (not present on such work as by R. Gauthier, who mills the teeth on the sides).
            These layers would interact in mesh similarly to gears finished with the Ingold fraise topping tool.

            Or so I postulate.

    2. Yea I was wondering that myself, maybe he should build another one with a steel spring and run them in a clock spring death match to see which lasts longer. Also I wonder how small you could 3D print this thing?

        1. In truth, metal hair springs (the ones attached to the balance wheel, or oscillator) are already thinner than the thickness of a human hair.

          Many high end mechanical watch manufacturers (think Patek Phillipe or Rolex) already do laser cutting of immensely small parts. As early as a decade ago, Patek had escape wheels in production watches that were laser cut from sintered plate stock of alloys that were considered impossible to machine. If you want to learn a lot about engineering history, watchmaking is a really good place to start, as even before John Harrison’s Marine Chronometer watchmakers were pushing the limits of human engineering knowledge. Today, completely disregarding electronic watches and smart watches, we can still see improvements to this hundreds of years old technology.

  3. Minor technical point, the tourbillion does not replace the pendulum. The balance wheel performs the pendulum function, as it does in all mechanical watches and many clocks. The tourbillion holds the balance and escapement and rotates in an attempt to average out positional errors. It’s irrelevant for a clock which has no positional variation, and arguably pointless for a watch, but it’s main purpose in both watchmaking, and apparently 3d printing, is to demonstrate the technical skill of the maker.

    1. This is all true. The Tourbillon (whirlwind) was invented for pocket watches. They could sit in a watch pocket at a variety of angles and the rate of running would vary because the moving mass of the balance was not concentric with the balance pinion (axle) as well as other positional effects. Making the whole balance assembly rotate averaged out the errors and resulted in a consistent rate. The hard way. It never caught on because it’s fiendishly difficult to make, and it’s easier to just eliminate the position errors with a better quality balance.

  4. Jeez, something like this sure does make you feel underacomplished as an engineer :\
    I remember seeing a great Glashutte Original video some years back, and they showed a large scale Tourbilion model in it, something like an executive desk toy size, 10x scale from a watch, and I always wanted to try and design and build my own to get some personal branding credit. But, now, it’s been done.

  5. Tourbillon watches can be bought on ebay starting at about $ 15. They are fun to watch and actually work. No really quality but you don’t have to spend thousands of $ to obtain such fine mechanics like the tourbillon.

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.