Ferrofluid Clock Is A Work Of Art


It is not usually too difficult to separate functionality from art. Consider a clock. It’s a machine that has a clear and distinct function. It provides information. Nothing could be more different from a clock on a wall than a piece of artwork.  A painting, for instance has no clear function and provides no information. It’s just…art. It’s nice to look at. If we were to ask you to build a functioning, information providing clock that is also a piece of artwork, you would surely have your hands full. Where would you even start? If your name was [Zelf Koelma], you’d grab a bottle of ferrofluid and build us a beautiful, almost mesmerizing clock.

clock_01There’s little to no information on the details of how the clock works other than the use of ferrofluid. But it’s not hard to guess that it uses dozens of electromagnets and an Arduino. You can even pick one up for a cool $8,300 if you’re lucky enough to get a spot on the list, as he’s only making 24 of them.

Want to make one of your own? Pick up some ferrofluid and keep us updated. We’d love to hear from you in the comments on how you’d implement a build like this one. We had a fun time hearing your ideas when we covered the clock made of clocks.

Thanks to [Itay] for the tip!

58 thoughts on “Ferrofluid Clock Is A Work Of Art

      1. The ferrite particles start to separate from the surfactant, and no longer stay suspended in the carrier fluid.
        In some application where there is a lot of movement (like cooling speaker voice coils) this is not an issue, as long as it is not exposed to the atmosphere, where a high vapor pressure carrier fluid can evaporate over time,

        1. Also, some of the surfactants used are acidic. Oxalic acid is a common one. If the acid reacts with something it looses it’s attraction to the ferrite particle, and once “wetted” by the carrier fluid, the particles start to agglomerate to form clumps. The clumps settle out and you end up with a magnetic solid mass.

      1. There’s probably not a lot of CPU behind this unless they’re calculating the transitions on the fly. More likely it’s whatever controller they like, a few display drivers and a large chunk of memory for the animations.

        Each digit on the clock looks like it’s 7×4, the entire area for a digit maybe 10×6, giving 10×24 all up. Easy enough these days to drive a few hundred outputs. Wiring them all up, now that’s fun.

        Controlling the power of each output looks like it’s needed. To move the blob to about spot, it looks like it turns the neighbour on then slowly reducing the power of the current dot.

        Mapping out all of the transitions would be fun. Doing the digits individually (that is the fluid doesn’t transfer between them) would make life easier as the number of ‘frames’ would be way less (well 10 really).

  1. It’s a really cool piece of work. Too bad this article is so terrible.
    People have been creating artwork clocks for thousands of years, and the author cannot think of a single example?
    Also, art isn’t just something that is nice to look at. A beautiful sky is nice to look at, but it isn’t art.

    1. I agree there could have been a bit more depth to the exposition – Function vs. Art is are two ends of the spectrum and I think Will set them up in a false dichotomy. Clocks and lamps are two very practical technologies that I would argue are *normally* presented as art, at least on some level – and there are a plethora of examples of that just on HaD. (Never mind that Art could also be considered functional, in “providing enjoyment”.)

      Having said that, it’s a minor nitpick and I’m glad for the article – this clock definitely one-ups the awesome for aesthetics and cool technology.

    2. Hackaday covered a lot of artwork clocks, and the article gave you an example: “clock made of clocks” project.
      After the video, there’s a link on phrase “Posted in clock hacks”, and clicking there you will find a lot of artwork clocks projects.

  2. My jaw dropped at this. I’ve built several clocks, I’ve tinkered with ferrofluid, I’ve been itching to make something like a CNC zen garden or a drawbot or magnetic levitation or something, just waiting for a great original idea to come to me.. and this fits it all just too perfectly. My god is it beautiful. But it’s not mine.

    Maybe I’ll try to make one anyway.

    1. Make a panel for a coffee table or something. You could have ripples like a lake and a touch panel to run lines in it with your fingers. Maybe something like those pin cushion toys that extrude shapes when you push something into the back, but horizontally.

  3. I reckon a microcontroller and some EPMs and this would be a fairly simple project. It’s a shame nobody’s selling cheap Electro Permanent Magnets yet, we’ll have to make our own

  4. If the wowgasm from that video ever subsides enough that I can think straight, I’m going to order some ferrofluid and get started! I wonder if there’s a way to simulate the slow motion effect in real time…

  5. This is very cool, and would be an order of magnitude cooler if the designer had bothered to share with us how it was done.

    I think that it would be fun to make a similar clock out of magnetic Silly Putty. It doesn’t have the drawbacks of ferrofluid (see comments above), but would work a lot more slowly due to its high and variable viscosity. Maybe it could be made to change only every 5 minutes. That would give enough time, providing the magnetic fields were strong enough, to change from one digit to the next. Perhaps an analog face would work even better.
    I might just have to try that…

    1. I like the idea. But it would take a good bit more than 5 minutes to fully change from one “state” to another most likely. You could modify the viscosity range (and other factors) though by custom compounding your own PMDS “silly putty” though.

      1. The stuff I have is fairly soft and moves readily when using strong magnets (Crazy Aaron’s brand, available on Amazon.com). Also, the distance between elements (the size of the clock) would be a factor as well. A smaller clock could be made to work, I think.

        I like your idea of custom compounding the putty, but I wonder if sourcing the ingredients would be a problem.

  6. Dang. My wife and I worked on a ferrofluid clock for weeks before giving up because the electromagnets would cost too much. We tried all sorts of stuff like linear actuators to move permanent magnets up to and away from the fluid, etc. I guess the only thing we didn’t try was spending all that money for tons of electromagnets!

    1. did you happen to price up what it would cost to make the electro magnets yourself? The cheapest ones I could find on aliexpress were about $3/ea and I figured it would end up costing $800-1000 just in magnets at that price. I’m curious exactly how strong the magnets need to be in order to move the ferrofluid around because weaker EMs are cheaper and/or easier to build.

  7. I thought of doing this exact thing a couple years ago. It’s on my list of projects that I never get to. :( Now there’s not much point since I don’t really think I could do any better. Guess I should take that as a hint to get going on the other projects on my list before someone else thinks of the same thing, like has happened so many times before…

  8. ​Hi,

    I have made ferrofluid from laser toner and motor oil to find it too heavy for nice experiments. So I am looking for white water or white oil (silicon oil) ferrofluid solution alike ferrofluid clock


    I can do ferrofluid clock now, it’s easy trick ;)))

    Digit segments are made of electromagnets

    and there is one another solid magnet moving from down to top, moving ferrofluid to digits.

    If you set electrognets off, electrofluid goes down as heavier than water/ silicon oil.

    My experiments with water and laser toner failed since water cannot moisture toner particles, so they float above.

    Ok, I can use mixer.

    global innovator
    manta103g at gmail com

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.