Electromagnet-Powered Pendulum

We’re always happy to see hackers inspired to try something different by what they see on Hackaday. To [SimpleTronic] has a project that will let you stretch your analog electronics skills in a really fun way. It’s an electromagnet pendulum analog circuit. Whether you’re building it, or just studying the schematics, this is a fun way to brush up on the non-digital side of the craft.

The pendulum is a neodymium magnet on the head of a bolt, dangling on a one foot aluminium chain. Below, a Hall Effect sensor rests atop an electromagnet — 1″ in diameter, with 6/8″ wire coiled around another bolt. As the pendulum’s magnet accelerates towards the electromagnet’s core, the Hall effect sensor registers an increase in voltage. The voltage peaks as the pendulum passes overhead, and as soon as the Hall Effect sensor detects the drop in voltage, the electromagnet flicks on for a moment to propel the pendulum away. This circuit has a very low power consumption, as the electromagnet is only on for about 20ms!

The other major components are a LM358N op-amp, a CD4001B quad CMOS NOR gate, and IRFD-120 MOSFET. [SimpleTronic] even took the time to highlight each part of the schematic in order to work through a complete explanation.

In the end, this analog circuit should help newcomers get familiar with electromagnets so they can move on to the next logical steps: coil guns and web shooters.

[Via Hackaday.io]

44 thoughts on “Electromagnet-Powered Pendulum

  1. I think there’s room for reducing the component count here. One of the spare gates could be biased into linear operation and the input(s) AC coupled to the hall sensor. Use the remaining gate to square it up and use that to trigger the monostable.

    1. One of my uncles was in advertising, and in the 70s I believe, a point-of-sale displays he helped create consisted of a cardboard image on a wire armature that rotated or twisted back and forth. The motor powering this was a solenoid, and a steel rod on the rocking armature that passed through that solenoid. Triggering the solenoid was a simple contact arrangement that connected up the coil to a D battery. I cannot recall the specifics of the contact arrangement, but the net result is that the solenoid was powered briefly at the right time to impart a little energy to the armature at the bottom of the swing. So the total electrical parts count was: D cell, contact arrangement, solenoid. I recall it would run for months on that D cell.

      Anyway the featured project is a nice little challenge, and would make a great school project.

      1. I was gifted one of those, in the later~mid 1960’s.
        Coil of at the bottom of frame along with the “D” cell holder.
        Chunk of (rough cast appearance) material on the pendulum.
        A leaf or spring type switch on one side of the frame.
        Can’t recall much more construction detail.
        As the pendulum swung, it caused a make-break action with the switch that energized the coil to repel the magnet as it passed over the coil.
        Was probably one of My very first introductions (wonderfully visible action/reaction!)to understanding what what happening with electricity being transferred into motion.
        You could play with the “tuning” by wrapping tape or something onto the armature and see what affect it had on the function, etc.
        In other words why an electric motor turned or stalled etc.
        This simple “do-dad” was pretty darned instrumental in my appreciation of and getting handle on things electrical.
        I think it eventually was squashed (one time too many) in a box of my youths souvenirs, thus finally tossed out. :(
        IF he’s still alive, Please pass a thinks to your Uncle!

        P.S. Speaking of running for months.
        This was one the first things to cross my mind when the coin cell challenge was posted.

  2. In my childhood (some 12,000 years ago now) there was a fun toy called something like “Bing Bang Boing”, which consisted of a metal ball bearing bouncing over a path of stretched ‘balloon trampolines. What I have often wondered is if an electromagnet could be positioned to compensate for frictional loss so that a set of these trampolines would bounce the bearing forever in a ring. (here’s a slightly more recent effort to replicate the ‘game’: http://trampoline-game.blogspot.com/2005/10/years-ago-i-really-liked-game-called.html )

    1. It would be difficult getting the magnetic fields lined up properly to give the (I assume) magnetic bearing the proper kick when bouncing off the balloon. The ball will be tumbling, and an electromagnet would be as likely to pull as to push.

    1. that sure wouldn’t hurt.

      What I like about his website is the detail in the explanation of the circuit, don’t see that much these days. Well done.
      I like the approach of the hall sensor, this makes the design a little easier then when using the coil itself as the sensing device.

      Hopefully [SimpleTronic] will add the diode soon before the FET breaks. But perhaps many FET’s have already died…

      1. That doesn’t help. Current in an inductor continues to flow in the same direction. The voltage inverts, driving the drain more positive so that reverse biased integral diode will never be forward biased. The damage occurs because the voltage from drain to source increases until the MOSFET conducts, usually damaging it.

        The diode needs to be reverse biased across the coil.

  3. OMSI (Oregon Museum of Science and Industry) in Portland Oregon had a big pendulum that was at least two stories high with some sort of electromagnet to keep it going – at the edge of the swing there was a circle of domino’s and the pendulum would tip over a domino every minute or so – showing that the earth rotated – was a bit hypnotic to watch. Assuming it is still around a good visit

      1. Did they put it back? It was removed about 40 years ago when it became the National Museum of American History in recognition that Ronald Reagan had set the country on a backwards path. Obviously the pendulum had to go, along with French Fries. It was electrically powered because even air has friction.

        1. To be honest, I am not sure if they put it back. I was not aware that it had been taken out. It’s been that long since I was in DC. It’s too bad if they have not, it was a cool piece.

  4. I love these kinds of things. That said, couldn’t one use the induced voltage in the electromagnet coil itself to trigger the magnet circuit on and skip the hall sensor entirely? (answer, yes, and I’ve seen it in the home of an old ham who homebrewd this kind of thing up as part of a super-accurate clock – in a glass pipe to eliminate air currents, invar for the pendulum length, and quartz fiber for the bearing. It worked in pull mode, eg the solenoid pulled the pendulum in as it approached.
    It ran on the power picked up by a big coil – basically wasted radiated 60hz from the rest his house. It really doesn’t take a lot of power to run a good pendulum if you don’t blink a led too.

    I’m not saying this as a critic – I’m pointing out where to go next with it…Like I said, I love this kinda stuff.

    1. We had a Foucault Pendulum at the Science building at Wittenberg (Ohio) in the late ’60s that worked this way.
      Electromagnet pulled until the mass went by, detected by the coil. No Hall sensor.
      More stable than pushing.

  5. DCfusor2015… That pendulum as a clock with paracytic power sounds delish. I your spare eons, maybe see if anyone inherited this or if he is still hamming it up?
    San Francisco Exploratorium has a.2 story pendulum clock. Only reason to have 2 storues, except for the observatory/ widows’ walk.

    Is a free wheeling diode equal to a flyback? Or sisters/cousins?

    1. The freewheeling diode is to absorb the flyback current/voltage so it does not damage the transistor by overvoltage.

      When you turn off the source of current/voltage to a coil, the magnetic field begins collapsing. This generates a current equal to that which created the magnetic field in the first place. The voltage will rise very quickly until this current flows, it can easily be thousands of volts if nothing breaks down by then.

      Power is still equal to voltage times current. So the higher the voltage, the faster the magnetic field collapses and the more quickly the current and voltage decay. That is why sometimes relays have a diode in series with a resistor instead of just a flyback diode.

  6. Just wondering, if you run this thing for a few hours, does the path of the pendulum rotate 15 degrees per hour, as the earth does? Or does the jarring of the pendulum by the electromagnet in this setup (it looks as though it is ‘bumping’ it pretty hard) introduce enough noise to swamp out the rotational effect? Something like this could make a cool clock.

    1. For the pendulum to precess with the earth, it needs a support that rotates. The support for a Foucault pendulum is a bearing that rotates in the horizontal plane. Old WW2 gyroscopes were dynamically balanced on a large pendulum with an electromagnet pulse system. However, the axis of pendulum was fixed relative to the Earth’s axis.

      The most interesting rotating pendulum is the paraconical pendulum. Maurice Allais (physicist who won a Nobel in economics) showed that the gravity of the Sun and Moon affect the precession of the paraconical pendulum. During two solar eclipses, his pendulum radically changed its precession rate. After the eclipse, his it returned to its former position and precession. Pulsing a paraconical pendulum with a magnets is complex because the plane of the pendulum keeps shifting. The paraconical pendulum also has minimum support friction.



      1. Every Foucault pendulum I’ve seen has been a large, symmetric (a ball) mass on the end of a long, thin cable.
        No rotating support – just a simple (symmetric) clamp of the cable up top.
        You can’t have the electromagnet pushing, like this one, and expect to not change the plane of rotation.
        Pulling, gently, as the ball come to the center worked in all these systems.
        Mind you, they were balls weighing 20-50 lbs (yes – lbs) and the cables were typically 30-40 or more feet (yes – feet) long.
        You’re welcome to convert to SI units.

        They all precess nicely. Fun to watch.

    1. A pendulum’s period is pretty rigidly determined by the length of the swing. So perhaps this would work if you could choke up or feed out the pendulum arm.

      If you attempt to speed it up by “hitting” it harder with the pulse, it just swings higher. If you attempt to slow it by hitting it with a retarding pulse, it just doesn’t swing as high. So you might speed up or slow down one swing, then the rest are unaffected.

      1. Actually, the pendulum period *is* slightly dependent upon the amplitude. As the amplitude becomes higher, the period gets longer. The limit of this is obviously at +/- 180 degrees, where the period become infinite.

        You *could* tune the pendulum period by giving it a harder kick to slow it down, and let the amplitude decay to speed it up.

        For an initial amplitude of (say) 0.1 radian (5.7 degrees), a 10% change in amplitude will yield a period shift of about 11 seconds per day: a perfectly reasonable amount to fine tune a pendulum and sync it to something else.

  7. I remember toys like this from when I was a kid. Toys like that didn’t last very long in my house… Taking it apart, I recall finding a surprisingly simple one-transistor circuit with very few discretes — far simpler than the one shown here.

    Google provides something that looks similar to what I recall: https://www.sciencetronics.com/geocities/bilder/images/electronics/projects/pendulum_sch.gif
    (parent page is https://www.sciencetronics.com/geocities/electronics/projects/pendulum.html )

  8. Great video, article, and comments! I was wondering if anyone was aware if a kit was sold anywhere that included this circuitry and coil? I am working on a project prototype that requires a swinging arm and this looks like a great way to accomplish it. Thanks!

  9. Hello,
    I’m building swinging sticks with this solution for movement. Stickcs are from brass and I need stronger magnetic field.
    Is it possible to add in circuit potentiometer to regulate magnetic field and what will be effect of biger coil 1.3″ diameter, 1″ lenght around M10 screw?
    I’m not so good in electronic and I need help.
    Thank you

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.