3D Printing & Imaging Science
Bet You Can’t Find a Use for This 11-Million-to-1 Gearbox

featured image gearbox

It’s a solution looking for a problem, and hopefully you can find the problem.

The solution that puzzle inventor Oskar van Deventer has come up with is a fist-sized 11,373,076:1 gearbox. That’s no typo — to get the output to rotate only once would require over 11 million rotations of the input shaft. Why? He’s not sure; he wants that to be your job.



With the normal (“spur”) gears we all think of for basic gears, changing the speed of a shaft is space intensive. For example, to change the speed by 5x one gear must be five times the diameter of the other. This is okay for small reductions but is cumbersome for larger ratios. For this reason, multiple stages are often implemented.

The typical way of achieving high gear reductions is to use either a planetary gearset (one or two stages of which are in most electric drills) or a worm gear (as in a sewing machine).

Headed in a different direction, Van Deventer has a new type of gear he calls “grinder gears” that comprise an inner gear with one fewer tooth than the outer gear, causing it to wobble its way through the rotation. For the engineers, it’s a variation of a cycloidal drive. The secret is that these gears allow a geardown based on the amount of teeth instead of their physical size differences.

What’s Inside the Gearbox?

First off of the crank is a dual-stage planetary; this powers a dual-stage grinder. The grinders move in opposite directions, but are one tooth off of each other themselves (27 vs 26 teeth, with the inner gears 26 and 25 teeth respectively). What the yellow gear does to the green gear is almost completely undone by the orange gear (but not quite) leaving the orange gear to vary by infinitesimal amounts relative to the yellow. Adding up all the gear types and stages yields the 11,373,076:1 ratio.

Possible Applications

The inventor has not come up with any, so, to get started let us look at what it would be tempting but unsuitable for.

Precision: One might think that because the output moves so gradually that it would be ultra-precise. While the net movement would indeed be small and you could count on it to be accurate, the amount of play in the design would devour its precision. On average it would be amazing, but not usefully reliable at any specific point.

Torque: “Gear up for speed, gear down for torque”, so this must have astronomical torque, right? Well, in theory there is enough mechanical advantage for a single man to win a tug-of-war against all of Manhattan Island — when fabricated from imaginary materials. In reality, the plastic the gearbox is made from (or steel, or tungsten, or carbon nanotubes) will hit its limit many orders of magnitude before that mechanical advantage could ever be applied. Van Deventer does point out that the design actually handles torque better than most other gear systems, but nowhere near enough.

Vibration: Oskar shuts this one down himself. Grinder gears grind and sound like dogs on a bone. Even if the net wobble in the gearbox is eliminated, it still shakes and shudders.

Long-Term Measurements: Perhaps this could be used to measure or affect something that spins very fast, over a very long period of time? 24/7 spinning as fast as one can would result in not even one output rotation over the course of an entire month. Material strengths here again are in play; it seems far more likely that the gearbox would wear itself out (“grinder” is an apt description) long before a single rotation could ever be completed.

It still reduces speed, so, whatever purpose it may someday serve would probably revolve around an application where speed change alone is valued and not its typical tradeoffs of torque and precision.

MM - Grinder Gears 1

Product Before Market

This gearbox is an interesting case study of Maker culture and the accessibility of technology. We live in a world where an idea that does something that has never been done before can be built at home and shared without cost (so that it can be reproduced by anyone else also at home), all without it having any use or any place in the market. Outside of art, that is something we did not have a few years ago.

Most inventions are driven by a need to solve a problem for a certain price. Burdening innovation with the chains of both usefulness and profitability has merits for efficiency in a factory, but the investment here is minimal.

Divorcing these three components and letting crowdsourcing handle the unsolved pieces means we will get to see a lot more unbridled innovation in the future. That is going to mean a lot of chaff (and perhaps in the end this gearbox is among it), but it also means we are going to achieve things we otherwise would not have.

If you can fill in a piece of the puzzle, Oskar would love for you to leave him a comment.

52 thoughts on “Bet You Can’t Find a Use for This 11-Million-to-1 Gearbox

  1. What about very small movements like moving a microscope slide a billionth of an inch. You dont care about accuracy you just want to make a tiny change in viewing angle while looking through the microscope. It could slop around a little on the way to the final viewing position could it not?

    1. It would definitely make their design more compact, but like Matt mentioned in the article “grinder gear” is a very appropriate name. Someone would have to find a solution to it wearing itself smooth before it made any meaningful measurement on that scale.

  2. With a gear ratio that high, the natural solution would be to hook it up to something incredibly fast that has previously had no way of being exploited. Normal gearboxes turn high speed from a motor engine into slower speeds to turn the wheels. With a gearing this astronomical, the high speed could perhaps come at the biological or molecular order – generating low voltage electricity from circulating blood, perhaps?

  3. I used to work on a Kyropoulos sapphire grower that pulls the crystal at the rate of 0.1 mm per hour for 14 days. Using a 5500 rpm motor at a 11 million to 1 reduction, it comes out to 0.03 rev per hour. Using a 5 mm per rev lead screw comes out to 0.15 mm per hour. Might work… although the vibration from the motor/gearbox needs to be isolated, as crystals like to grow in serene environment.

  4. You could attach it to a multi layered stack of water and oil sheets and these would slowly change with the eccentricity of the output.

    Would make for an interesting ceiling thing in a dentist’s office.

  5. You could use it to design a locking mechanism that could only be (reliably) opened with a high speed motion controller. A simple proof of concept could be made by attaching the slow end to the dial of a Masterlock combination lock.

  6. reverse it, … find a way to move the final gear first. take it from 11,373,076:1 , to 1:11,373,076 .. replace the handle with, essentially an alternator. or build some thing similar to create power….

  7. If you had more than one working in parallell, would that help to recover the precision?

  8. I’ll tell you applications .. if you can make this smaller still.. waayyyy smaller..

  9. Use as a energy storage device.

    For example: Take a big block of lead (or anything that is heavy) and lift it up using this device really slowly with cheap energy (solar, wind.. whatever).

    Then when you need power, you simply drop the dead weight. This in turn turns the device in reverse, fast enough to spin an alternator.

  10. wouldn’t it wear out before you could get to a single rotation of the output?

  11. Reverse the drive, and use for wind turbine. The prop on the slow end the the turbine on the fast end.

    1. I’m not sure but wouldn’t it be insanely difficult to move the slow end? If torque is stepped up from the crank in original configuration, wouldn’t trying to turn the output shaft be insanely difficult?

  12. I’m no expert, but my understanding is that the reason that turbine/jet engines never really took off (no pun intended) for automotive use is the giant size and weight of the gearbox that it took to reduce 30,000+ rpm of the turbine shaft down to the 100rpm range of the wheels. could it be scaled up to deliver the necessary torque to drive a car?

  13. Fine adjustments. Build your gear ratio to suit the minimum unit of adjustment you need.

  14. Use it to drive the axises of a 3D printer. A little bit of play won’t be an issue for any particular point, but having the position be extremely accurate on average will help make reliable prints.

  15. I would use the setup in the video as a pully system or a winch. Able to pull heavy loads with minimal force needed to rotate it. And it could have different speeds depending on which side is locked.

  16. Simple – use it to do minimal motion friction reduction. Let me give an example.

    Ever had to lift a house to replace a support stantion? It’s a bitch. The problem is lifting the house *just* enough to relieve the weight on the post. With this little f’er made from stronger stuff, you could replace all those expensive jacks that they use now. Simple add a drill attachment to it. Stick it in, let your drill run for a bit, then pull out the stantion. Now I can see how this might be overkill for a normal stickbuilt, but doing a foundation repair on a skyscraper, where an inch lift in the foundation could tip the entire thing? This f’er is a lifesaver.

  17. Looks like it would make a decent engine rev counter. consider an engine that needs an oil change every 500 hours and rotates at 1000rpm, in an hour that is 60000 rotations in 100 hours that is 6million so at 11 million rotations for a 11million to 1 ratio that would indicate 183 hours for one output shaft rotation. But trouble is any practical use will wear the input out fairly quickly.
    You could use that shaft to power an automatic greaser as afaik those currently on the market need an air compressor to power them.

  18. Wind or water power? Something where turning a high-torque generator would be valuable?

  19. Moving an item to keep pace with very slow processes (I was thinking say the rate of sinkage of houses on soft ground)

  20. If the handle could be reversed so that one turn of the handle causes the output of over 11 million revolutions….then attach a motor that can spin at approximately 26 meters per second (which is doable) then the output would be a physical mechanism spinning at the speed of light! Which brings the Van Stokum Cylinder into play. The Van Stokum cylinder is a type of time machine based on an immense cylinder spinning at near-light speed. The physicist W. J. van Stokum realized in 1937 that such an object would effectively stir spacetime as if it were syrup, dragging it along as the cylinder turned. What Van Stokum didn’t realize is that circumnavigating such a cylinder can lead to closed time-like paths. Anyone orbiting the cylinder in the direction of the spin would be caught in the current and, from the perspective of a distant observer, exceed the speed of light and thus travel back in time. Circling the cylinder in the other direction with just the right trajectory would project the subject into the future. The Van Stokum time machine is based on the Lense-Thiring effect.

  21. One application that requires extreme ratios such as this, are the wheels on the transporters for rockets at Cape Kennedy (Cape Canaveral). The Saturn rockets were the largest, transported upright, and would take hours to travel just a few miles (1? 2? 4?).

    If you can use this in the opposite manner (that is, convert one rotation into several million) then you build a spring-powered car, powered by nanotube springs. Theoretically, as shown by MIT back a few years ago, carbon nanotube springs have a greater energy density (power to weight) than lithium batteries, and approaching gasoline.

Comments are closed.


Matt has a passion for education and unconventional engineering. He delights most in projects that remove barriers to curiosity and allow makers to explore new hobbies without committing to them. He is active in any hacker or makerspace activity he can get his hands on and has more creation-related pastimes than a healthy adult probably should.

View more articles by Matt Freund