You can get just about any gear reduction you want using conventional gears. But when you need to get a certain reduction in a very small space with minimal to no backlash, you might find a wave drive very useful. [Mishin Machine] shows us how to build one with (mostly) 3D printed components.
The video does a great job of explaining the basics of the design. Right off the bat, we’ll say this one isn’t fully printed—it relies on off-the-shelf steel ball bearings. It’s easy to understand why. When you need strong, smooth-rolling parts, it’s hard to print competitive spheres in plastic at home. Plastic BBs will work too, though, as will various off-the-shelf cylindrical rollers. The rest is mostly 3D printed, so with the right design, you can whip up a wave drive to suit whatever packaging requirements you might have.
Combined with a stepper motor and the right off-the-shelf parts, you can build a high-reduction gearbox that can withstand high torque and should have reasonable longevity despite being assembled with many printed components.
We’ve seen other interesting gear reductions before, too.
Less parts than cycloidal drive, that’s nice. The efficiency is probably not great due to the rolling resistance of the rollers, but for many applications that doesn’t matter.
As with any 3D printed gearbox it will not last long in continuous use, but that’s not a problem for many hobby and prototype projects.
Good for educating one on wave drives which might come in handy for the next ‘wave’ of roboticist.
How is it less parts? For a basic cycloidal drive all you need is the outer housing with fixed rollers (the rollers on a cycloidal drive don’t need to rotate), the centre axle, cycloidal disc and output shaft.
For the wave gearbox you need the outer housing, centre axle, roller race, output and all the rollers or ball bearings.
I read that as “warp” drive, anyone else?
In that case i would have known from the title alone that it is clickbait.
Because everybody knows you need a replicator for that.
Let’s say we gonna make it out of carbon steel, the biggest challenge will be the steel ball/roller cage which basically is the output, any play between steel balls/rollers and the cage will result backlash which obviously is not desirable.
I was very interested in this reducer concept at first glance, but later on, I figured out that any play between the cage and steel balls/rollers would result in backlash, which is undesirable. Machining such a cage, which is also the output shaft, isn’t an easy task since it requires a very tight tolerance…
Not only that, but the rollers will be sliding (not rolling) against the cage when they transfer torque, which isn’t exactly ideal.
However, seeing how the rollers are always in contact with the bearing, they are always arranged in a perfect circle, so it would be possible to make a carrier with pins for the rollers. That cage would follow the wave, so it would need something to transfer the torque to a shaft, without adding much complexity or backlash.
I wonder if you could make a two-stage gearbox with a small one inside the wave generator bearing? You could have the cage fixed and the outer ring be the input to the bigger wave generator.
Oldham coupling for this Wave Drive:
https://www.youtube.com/shorts/0yJb5-nMu08
https://www.youtube.com/shorts/llkCui9tJxs