back to article How to build your own DIY makeshift levitation machine at home

Engineers at the University of Bristol in the UK have published a rough guide to building a simple levitation chamber that uses sound waves to suspend objects. Performing levitation experiments requires careful laboratory equipment and conditions. But a paper published in the Review of Scientific Instruments this month shows …

  1. Justin Case

    Like the idea, but...

    ...using an Arduino just to generate square waves? That's a fully fledged computational device doing the job of a simple astable multivibrator.

    Everything these days seems to have an extra layer of unnecessary complexity. I must be getting old.

    1. imanidiot Silver badge

      Re: Like the idea, but...

      blah, blah, I could do that with a 555...

      Yeah, you could. You'd have to do more calculations, breadboard or solder up more components and then tune the vibrator to the frequency you need.

      Or just grab the Arduino off the shelf, plug it in and get going, doing the tuning in the programming.

      Is it overkill? Probably. But it's convenient and Arduino's are cheap as chips compared to spending even an hour extra time building an discrete square wave generator.

      Look at it this way. In the olden days a square wave would be generated with discrete components, transistors, resistors, capacitors, diodes, etc (talk about unnecessary complexity...). Then we got the 555 timer chip to combine all those components in silicon and require less discrete components. Now we have MCU's in various guises to fulfill the job even easier and with more flexibility.

      1. Jason Bloomberg Silver badge

        Re: Like the idea, but...

        A Raspberry Pi would be better; giving it Bluetooth and WiFi, allowing it to be controlled from a mobile phone and reporting its status over the internet.

        1. m0rt

          Re: Like the idea, but...

          Typical of Microsoft....

          Wait...maybe I should read the article...

      2. Justin Case

        Re: Like the idea, but...

        Look at it this way. In the olden days a square wave would be generated with discrete components, transistors, resistors, capacitors, diodes, etc (talk about unnecessary complexity...)

        Those "olden days" of which you speak, they're my days, they are...

        </wipes rheumy eye>

        I fully get the point about ease, convenience and all the rest. It just strikes me as ironic that as a result of progress, it now takes millions of transistors to do the work of two.

        1. m0rt

          Re: Like the idea, but...

          There is beauty in simplicity...

        2. Version 1.0 Silver badge

          Re: Like the idea, but...

          Transistors? You are using Transistors? Paradise .... back in my days we had to do the whole thing with a Colpitts oscillator using one Triode and a handful of cold gravel.

          1. Anonymous Coward
            Anonymous Coward

            Re: Transistors? You are using Transistors?

            All we had were rocks and twigs.


            1. Anonymous Coward
              Anonymous Coward

              Re: Rocks and twigs ? You are using rocks and twigs ?

              All we hade were bananas.


      3. Kevin McMurtrie Silver badge

        Re: Like the idea, but...

        I'd use simple feedback so that it automatically finds resonant frequencies for any conditions. You could still put a 555 in there as a bandpass filter to avoid non-functional ranges of frequencies. That chip is essentially the basic building block of the universe.

        1. Anonymous Coward
          Anonymous Coward

          Re: Like the idea, but...

          There's another vector at work here too.

          If you quickly need to get to a proof of concept stage I'd use the Arduino. However, if you have to push this out in volume it becomes interesting to get it replicated in discrete design because that will reduce your unit cost quite considerably.

          That said, there will probably be a way that you can write the whole show into one chip so you cut down on component count. But I wouldn't boter with discrete circuitry anymore for a PoC, those days have indeed gone and I'm old enough to have played with tuning indicators..

          1. annodomini2

            Re: Like the idea, but...

            Micro < Discrete generally other than for really, really simple stuff and then only in massive volumes.

            Fundamentally more components = more cost, it's not the individual parts, it's the PCB and assembly.

            Which requires more R&D because it's more complex, plus there is a high probably of production issues due to that complexity.

            Also as the bulk of the behaviour of the system is SW, if there is a problem you can replace it with a simple firmware upgrade. A design issue in your analogue circuit generally results in replacing/repairing the whole circuit.

      4. John Smith 19 Gold badge
        Thumb Up

        "But it's convenient and Arduino's are cheap as chips "

        That's the real story of this.

        Acoustic levitators have been around for decades and have been used to do things like measure the properties of balls of molten Silicon and high temperature metal alloys.

        But they've always been ruinously expensive and a PITA to make and keep working.

        Obviously the sample size in this is tiny, but modern instruments are much more sensitive, so can work with much smaller sample sizes than previously. You can also run property studies one after the other, instead of simultaneously. Slower, but a great deal cheaper.

      5. Simon Harris

        Re: Like the idea, but...

        Turns out it's slightly more tricky than a two transistor (or 28 if you want to use a 555) solution.

        You actually need a square wave* to drive the bottom transducer set, and a second square wave at the same frequency but phase-shifted to drive the top set. By adjusting the phase difference between the two you control the vertical position of the levitated objects.

        So, while it would be possible to make an adjustable phase-shifter with discrete components (you could trigger a second monostable via a comparator on the timing capacitor of the first, for example), it might be easier to get working if you happen to have a microcontroller to hand.

        (* a sinusoid will work too, but a square wave is easier to make).

    2. Steve Foster

      Re: Like the idea, but...

      "an extra layer of unnecessary complexity"

      Ah, so that's what the 3D printer is for.

      (a two-pronged joke!)

      1. Mage Silver badge

        Re: so that's what the 3D printer is for!

        Indeed, I see nothing that needs a 3D printer or a CPU.

        Maybe the square waves are complicated?

    3. asier.marzo

      Re: Like the idea, but...

      Yeah, 555s are great. But you can get an Arduino Nano for 2$ that will provide a fast way of tinkering around. Also, this setup uses two 40kHz that change the phase relatively to each other.

  2. Chris G

    Heavy metal

    I think one or two metal bands I have seen in the past could lift a small car with their output.

    When I was a kid in the '60s I went to see Groundhog, a moving experience that left my ears ringing for a day after, they were the loudest but others came close.

    1. Anonymous Coward
      Anonymous Coward

      Re: Heavy metal

      Saw The Orb at Brixton. Had ringing for about 4 days....

      Granted spent much of the time sat 1m from the speakers for a 8 hour set.

      The best fun was the Dub DJ at one end of the upstairs corridor playing so loud you could barely walk in a straight line, unless you timed each step with the bassline.

  3. TheElder


    I built a magnetic levitator of my own design in high school. We had an electronics shop instead of an auto shop. I also developed a memory system using neon and crossing wires as well as a decimal to binary convertor that I made using diodes encapsulated in a cast block of plastic. I dropped in at Mountain View and a guy gave me a box of about 3000 transistors (floor sweepings) back when a CK722 cost about 70 dollars in today's money.

    The levitator was really cool because I used some LN2 to cool the electromagnet.

    1. werdsmith Silver badge

      Re: Levitator?

      If you can solve the LN2 storage problem (because it's hard to buy small quantities), then magnetic levitation is now very simple with affordable YBCO super conductors and rare earth magnets that can be bought off Amazon. Even the superconductor mag-lev train set kits are a click away but for the decent quality dewars much fun could be had.

      1. TheElder

        Re: Levitator?

        I worked at the Lawrence Rad Lab when I was 14. I was working on fusion power with my father. Plenty of LN2 any time I wanted some.

        As for buying LN2 that is easy. Get a all stainless steel thermos and find out who supplies doctors with LN2 for freezing off warts. The supplier will usually just give you a small fill for free.

      2. handleoclast

        Re: Levitator?

        Wow, that's complicated. All that kit. And LN2 that keeps boiling off. I prefer something a little more passive myself. Like this. Not only passive levitation, but it's a thing of beauty from the iridescent bismuth crystals.

        For more on bismuth crystals (but without levitation) try My World Of Bismuth by Tom Leary, which shows you how to grow your own at home. Note 1: the samples he shows are giant crystals made in a dangerously scaled-up version of the apparatus he demonstrates. Note 2: there are no spoken words, only captions, so when the mood music becomes irritating (after about 10 seconds) you can turn the sound off without missing anything. Note 3: what he doesn't mention is that the tapping allows him to feel when the crystals have grown so much they're about to fuse with crystals growing up from the bottom and it's time to remove them.

        1. TheElder

          Wow, that's complicated.

          My grade was an A+.

      3. John Gamble

        Re: Levitator?

        I had no idea there were such things as "superconductor mag-lev train set kits". So I searched on that phrase, and as a bonus found a demo of a superconductor moebius track.

        (It has a nifty Youtube video of the demonstration; direct link here.)

  4. Gordon Pryra

    None of this is levitation

    What you need is a ton of brown rice, a nice empty cave up a mountain and an old bed sheet.

    And 50 odd years practice......

    1. Rich 11

      Re: None of this is levitation

      What do you need to do during the 50 even years?

  5. Anonymous Coward
    Anonymous Coward

    You call that levitation, you want to see it when I p*ss on the toilet seat, the wife hits the roof.

    1. Sir Runcible Spoon

      +1, but I think there is a difference between levitation and a ballistic trajectory :)

    2. TheElder


      Sit down when you need to piss and a woman is around. It keeps them very happy. It does somewhat depend on tool length.

      1. TheElder

        Re: Pissing?

        Thumb down? Jealousy does not become you... Or maybe it does...

  6. Anonymous Coward
    Anonymous Coward

    "[...] emitting 40kHz waves on a single axis."

    Does that not cause a sample to vibrate and keep changing its orientation on all axes? Would that be a problem for any analysis involving reflections?

  7. Inventor of the Marmite Laser Silver badge

    Thanks for floating this with us

    I'll get my coat. Just plucking it out of the air.

  8. Dave 126

    The laboratory mice aren't going to enjoy this - they can hear up to 75kHz, dogs up to 45.

    Thinking of lab animals, didn't the bloke who got a Nobel prize for graphene once get an Ig Nobel prize for levitating a frog, thus becoming the first person to bag both prestigious awards? Yeah, he did:

  9. Mystic Megabyte


    Excellent! By using an app on my phone I can now raise a small lump of hash into a flame, a fan will blow the smoke my way. What shall I call it? The Vapotron or the, what was I doing?

  10. TRT Silver badge

    Microscope slides interfere with the beam path...

    but that can't be as much of an image killer as, say, ooh... rapidly vibrating the sample? We spend a fortune on pneumatic tables which decouple the instruments from the vibrations of passing tube trains, lorries, people walking a bit heavy footed in the next room.

    1. Simon Harris

      Re: Microscope slides interfere with the beam path...

      I think the idea is that the sound waves are standing waves and you trap the object to levitate in the null points, so you're not really shaking them - by adjusting the phase of one sound generator relative to the other you can move the node points and raise or lower the sample.

      You can do it with transducers at one end and a reflector at the other, but it's harder to compensate for thermal changes to keep the node points stationary.

      1. TRT Silver badge

        Re: Microscope slides interfere with the beam path...

        It still relies on differences in air pressure, though, and that must involve some degree of variability in the position of the pressure boundary.

  11. TRT Silver badge


    does this work by creating standing areas of higher and lower air pressure from the interference of the two sources? Because I thought, long ago about this... you know the heat haze over a road which distorts light waves? Well that's caused by different densities of air, this time caused by heating. So if you create different densities of air through sound waves which are inherently more controllable, then you could create a sort of fresnel lens out of air. A "sonic lens".

  12. Pascal Monett Silver badge

    "new applications of acoustic levitation"

    As long as you are focused on things that weigh only a couple of grams, I guess it works fine.

    I wonder what the settings would be to levitate a car. Actually, I wonder if the car would levitate, or explode.

    1. phuzz Silver badge

      Re: "new applications of acoustic levitation"

      It might melt perhaps? I'm guessing the vibration would make it pretty warm at least.

      1. Sir Runcible Spoon

        Re: "new applications of acoustic levitation"

        Don't bank on having any fillings after being a passenger in such a vehicle

        1. Anonymous Coward
          Anonymous Coward

          Re: "new applications of acoustic levitation"

          Don't bank on having any fillings after being a passenger in such a vehicle

          If it can life a car you'd be really, really clean though. To the point of no-more-flesh-on-your-bones clean.

          I don't know if you have ever experimented with cold vaporisers (typically used in humidifiers and ornaments that need a bit of fog hanging around them - you can buy the transducers in any garden shop, they typically come adorned with LEDs). Just drop one in the water, switch it on and watch the water splash upwards while a fog builds over the water surface. Now stick your finger in the middle of that splashing water so it is submerged above the piezo transducer, and in about 5 seconds you will know why you should not use that in a fish pond. They don't draw over 50W for nothing.

          BTW, the educative effect may linger for 15 minutes or more. Discovering this was not planned, I must add :).

          1. Sir Runcible Spoon

            Re: "new applications of acoustic levitation"

            For some reason, even upon re-reading, my brains keeps substituting 'fog' with 'frog'.

            Not sure what I have against frogs, perhaps I choked to death on one in a previous life or something :)

  13. inmypjs Silver badge

    ultrasonic levitation at home

  14. Stevie



  15. Scroticus Canis

    "careful laboratory equipment" - what the hell is that?

    So this stuff is careless non laboratory equipment?

  16. W4YBO


    Dr. Marzo actually posted an Instructable on this device...

    If you have dogs, they're liable to get a bit irritated if you fire this up in their vicinity. I have a couple of ultrasonic cavitators that I use for a Halloween display, one of which is currently in dry service to prevent Sophie from undermining my porch.

    1. Anonymous Coward
      Anonymous Coward

      Re: Instructable

      "I have a couple of ultrasonic cavitators that I use for a Halloween display, [..]"

      Is that a way to produce a decent sized low blanket of fog?

      I want something like that for this year's Halloween display. Heat based ones are too prone to upsetting people with asthma - and "dry ice" can build up dangerous concentrations of CO2 in enclosed spaces.

      I have a good imitation burning log fire that uses an ultrasonic device to atomise water for the flame effect - "heating" a trident whose glow is randomly animated with A/D controlled red LEDs. A cheap authentic (plastic) skeleton off eBay - plus a realistic mannequin holding a chainsaw - completes the main props.

      1. W4YBO

        Re: Instructable

        "Is that a way to produce a decent sized low blanket of fog?"

        Exactly. Found them on Amazon under "Ultrasonic Mist Maker" for less than ten dollars. Submerged two or so inches (5 cm), they'll churn up a nice thick fog at 24 volts. They create what looks like a drop of water hitting the surface, but it's static, and the fog bubbles up from underneath. I'll run one of them at 12 v to chase off the dog, but it doesn't seem to bother them if it's submerged.

        It sounds like you have a great display! A neighbor takes his chain off his saw and runs it.

        1. Anonymous Coward
          Anonymous Coward

          Re: Instructable

          "A neighbor takes his chain off his saw and runs it."

          Thanks for the pointer to the fog devices. Definitely on my list of acquisitions for this year. Middle of September is the point when I start putting the ideas together.

          Found a kid's toy replica of a proper branded chainsaw. Has a "chain" that rotates and audio to match. The sound is supplemented by a recording of a real chainsaw. Everything is controlled by a laptop and various sensors. All I have to do is appear with the treats at the appropriate point.

          This is the 2016 Halloween display. The skeleton is holding a rat skeleton with added red LED eyes.

        2. Anonymous Coward
          Anonymous Coward

          Re: Instructable

          They create what looks like a drop of water hitting the surface, but it's static, and the fog bubbles up from underneath.

          Umm, just a heads up - splashes above the transducer can go as high as 30cm. This is why the "fog trays" you buy have a little cap or a plate just where those emerge (about 1cm above the water, otherwise the agitated water that gives you the cold vapour never reaches the surface in the first place. You also need to make sure that the fat resistor that tends to stick out from the housing remains submerged as it needs cooling, and NEVER run these things dry because you'll chip the connective coating on the piezo.

          I'll run one of them at 12 v to chase off the dog, but it doesn't seem to bother them if it's submerged.

          If it's underwater, very little sound emerges. If it's above water you may be busy accidentally mistreating your pets.

      2. CrazyOldCatMan Silver badge

        Re: Instructable

        and "dry ice" can build up dangerous concentrations of CO2 in enclosed spaces.

        As well as irritating the hell out of contact lens wearers[1] since it'll dry their lenses out a treat and cause them to adhere nicely to the front of the eye..

        [1] ie - me.

  17. asier.marzo


    Hi, if anyone wants the instructions or parts list.




  18. David 18


    So how long until this is scaled up to make my frikkin' hoverboard I've been waiting for since 1985!

    1. Jonathan 27

      Re: Hoverboards?

      You want a hoverboard that will only work if placed inside a levitation chamber?

  19. Anonymous Coward
    Anonymous Coward

    RE. Re. Hoverboards

    Irony: this actually might work.

    My idea uses three of these, but until the patent application is complete not saying any more.

    Does however not require a special surface as such, will levitate up to 1.3 metres up as the field generator needs proximity of a surface to stay level due to no absolute reference frame.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon