A Bicycle Powered By A Different Kind Of Eddy

When you think of a bicycle and an Eddy, you’d be forgiven for thinking first of Eddy Merckx, one of the most successful competitive cyclists to ever live. But this bicycle, modified by [Tom Stanton] as shown in the video below the break, has been modified by ditching its direct drive gearing in favor of using the friction-like eddy currents between magnets and copper to transfer power to the wheel.

Before even beginning to construct a mechanism for powering the bicycle, [Tom] had to figure out the basics: what kind of materials could be used for a metal disk? The answer, after much testing, turned out to be copper. What kind of magnets work best, and in what formation? Expensive high grade, aligned North to South pole for added eddy-dragging goodness. Would the mechanism work with any efficiency?

The end result is interesting to watch, and it’s not exactly as you’d have expected. Yes, eddy currents drive the copper hub, but at a 100 RPM difference. Where does all of that energy go? Hint: not to the wheel, and certainly not into propelling the bicycle. All in all it’s a fantastic experiment with unpredictable results.

If bicycle based bumbling about bakes your biscuits, you might appreciate this tennis-ball-enhanced ride too.

31 thoughts on “A Bicycle Powered By A Different Kind Of Eddy

  1. There is a small engineering niche for magnet drives: they can transmit motion across sealed systems, they have built in torque limiting, and great wear characteristics. Driving a bike with one is really cool.

    1. Respectfully disagree. Magnetic couplings are terribly inefficient. All they can do is dissipate some fraction of the power being applied, with the fraction increasing as the difference between the input and output speeds increases. There is no conversion of speed into torque, or vice-versa. A far better approach would be to mount a brushless generator on the pedals and run that through a high-efficiency DC-DC converter to supply variable power to a brushless hub motor. That would have the built in torque limiting and great wear characteristics you applaud, without all the waste.
      We live in a time of electric cars with smart electronics. This video was just a clickbait stunt. The results were completely predictable.

      1. Or don’t be daft to eddy currents to transfer torque. Just put magnets on both disks and let them lock to each other. That’s the torque-limiting transmission, with no slip, no losses, like a synchronous motor.

        1. This is how I repurposed a dead belt sander to a bench top sander. I intended to run it with a cheap $20 Horror Fright hammer drill. Apparently the motor ran pretty fast originally, I couldn’t get a useful speed unless I drove the belt spindle directly, otherwise it was geared down too much. There’s no way to attach anything to this spindle, and I tried as hard as I dared to get it apart. A really strong loss-prevention magnet gripped the gear, and an appropriate socket in the drill, all mounted to some plywood, and it works pretty good. I think the magnet may be too strong to provide safety slippage though. The $20 drill is full speed or off, so I’m using a residential dimmer switch for speed control

  2. Tom didn’t really think it through, as usual.
    Pound for pound, aluminum has half the resistance of copper. At a tenth the price.
    But doing it just for the clicks gives you a free pass to engineer like that.

    1. Didn’t think through it? He spent over a minute of the video specifically addressing every point you claimed he didn’t consider regarding the difference between copper and aluminum. Another knee-jerk criticism in the comment section which I’ve come to expect, as usual.

      1. Ummm… Replacing an efficient chain drive with gearing with something that is lossy and no way to implement gearing doesn’t look like he thought it out.

        Results matter.

        1. The question is: did he really think this would produce a decent mechanism or did he just want to experiment and see for himself what it would do?
          I think the latter.

          And exactly thát separates the true hacker from the keyboardwarriors.

      2. Gary: Yes, he did mention cost and weight, but failed to interpret it correctly and used copper anyway. He didn’t think it through. Or, more kindly, he was optimizing for different things than loss or efficiency or practicality or mass, and just neglected to mention what.

    2. Sure he did – copper packs in the greater conductivity for a physically small gap regardless the mass. This wasn’t meant to be as efficient as a normal transmission, but I think he was happily surprised that it wasn’t more terrible.

    3. Like Gary wrote, Tom actually did address it. He not only did a great set of experiments, he also made a working prototype. Good video. If you don’t like the project don’t watch it.

  3. Eddy Merckx put out over 450 watts continuous in his famous 1-hour effort.

    If even half of that went to eddy current losses, and the disk didn’t cool through air or radiation through that hour, it would have melted.

    And he would not have set the record.

  4. Seems most commenters are completely missing the point.
    The video is about being inquisitive about the physical world, experimenting, and learning. It’s not proposing a new means of propelling a bike.

    1. Yep. That’s exactly why I wrote this up. It’s not about knowing it won’t work well as a bicycle transmission. Duh.

      It’s about doing real world experimenting and learning something in the process- things *aside* from the “it won’t work” conclusion. Experimentation can teach you things you’d never learn otherwise, even if you know it’ll be a failure from the start.

    2. It seems you are missing the point.
      The video exists to get clicks, make some youtube and sponsor revenue, and write off the CNC router and Form 3 printer as a business expense.
      The eddy current story is the MacGuffin.
      It’s’ a fine video, well produced and interesting. But make no mistake: It’s manufactured as entertainment, masquerading as experimenting and learning.

      1. “It’s manufactured as entertainment, masquerading as experimenting and learning.”
        There is no reason why it cannot be both entertaining and an experiment. He has many great videos and if these make him money to do more experiments and make more videos that is a win-win to me. If you don’t like it, don’t watch it.

  5. Indeed. Eddy current brakes are sometimes used for load testing of motors (I once played with a very small one at school, where it was connected to a DC motor and used for teaching about PID loops and stability. Using an eddy current brake as a transmission part is just daft. I predict efficiency is very low, somewhere between 0% (if the wheel is blocked) to maybe 20% at a very light load.

    1. I would take it as a dynamic brake system, for speed regulation on longer downhill grades. Or just an E-bike that can put the energy back into the battery, but sadly both options seem not to be common on the market.

  6. I developed a magnet motor that produces usable torque and RPM and can easily drive a generator or whatever you need. Nobody seems interested in free green energy production. EVs will no longer have to be plugged in and buildings can be 100% off grid. Please feel free to contact me for further information

  7. Seems like he was doing it because he wanted to make an eddy current coupling, and used the bike as a test platform. The people who knew it wouldn’t be as good as a regular bike, congratulations. But there’s more to learn than just whether it works as well as something else – I was interested to see that this was not unusable, although it wasn’t great.

    I do of course believe in the absence of cheap room temperature superconductors you want something that doesn’t rely on eddy currents at all even in this format; it’s easy enough to embed magnets in a middle disc, but maybe something iron based with good magnetic properties could be made resembling an induction motor, minus the electromagnetic coils. (Magnetic couplings are well established; I say “maybe” because I’m thinking about what could be a fun DIY.)

        1. I stand corrected. The steel in a rotor increases the flux in the aluminum conductors, but is deliberately laminated to produce no eddy currents, and produces no torque by itself. It’s the aluminum cast into that structure that is the component that conducts the current and produces the force.

          1. That’s fair. Without the conductors it wouldn’t be induction, no matter if it resembled it, and I should be more clear. That said, with the conductors properly included, it seems like a good way to go.

            The outside of a squirrel cage induction motor driven by a VFD is just an object that produces a magnetic field inside itself that rotates at a variable speed. Depending on the design of the rotor inside, the torque vs speed curve can be varied a fair amount, and the power efficiency is pretty good. So if you rotate the magnetic field by physically moving some magnets around the inner rotor, and that rotor continues to work as it usually does… If the output starts to slip to e.g. 80% of the no-load speed, the torque shoots up to maximum which can be quite a bit more than nominal. And pedaling at different speeds will change what particular ground speed the 100% mark actually is.

            If the efficiency is as good as usual, then unlike anything with eddy currents or friction, the effect would be just like a torque converter, shortening the effective gear ratio under load without shifting. Of course, it would be harder to do properly than either a locked coupling or one with just eddy currents, but still.

  8. He could have increased the magnetic flux density by adding a “backing iron” of high magnetic permeability material to the opposite side of the magnets. Going even more extreme, he could have oriented the square magnets into what’s called a Halbach array.

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.