Rolling. We're back in the factory. Been so long.

This is a random positioning machine and this episode of The Factory is the story of this device. This is easily one of the most complicated things I've ever 3D printed. A random positioning machine will continuously reorient a specimen. In our case, it's intended for plants. It'll continuously reorient that plant specimen so that the plant doesn't know which way up is. This is kind of like being in space. It's kind of like being in microgravity without having to leave the Earth's surface.

And so this episode of The Factory is the story of this curious little machine. The Powerhouse Museum were running a program for schools where students would be creating experiments as if they're gonna go into space. And a lot of the experiments that students were pitching were things like plant growth, crystal growth, things of that nature, things that these machines are kind of perfect for. And so we were brought into the mix to design and build this device, create kits for students to assemble, and then students can take their experiments and run them in the classroom and simulate being in space.

So naturally, it had to be pretty affordable and had to use technologies that schools can replicate, things like 3D printing. Now, we didn't invent this idea. There's a machine called a Klinostat and this looks like the patent diagram for it. And this is essentially a pot plant tipped sideways, driven by what looks like a motor. So that will just rotate the plant in one direction. And presumably that's good enough for most situations.

Random positioning machines exist in one form or another. This looks like a kind of custom built or university lab scale machine. And this is closer toWhat we're working with is a two-axis random positioning machine. The machine we initially saw was expensive and specialized, so we needed to come up with a more accessible design. Our design consists of a major outer frame that rotates in one axis, and an inner frame that rotates in another axis. To enable the rotation of the inner frame, we use a slip ring, which allows the wires to rotate continuously without twisting.

In the major assembly, we have a breadboard with two servo cables. One servo is fixed to the outer frame, while the other servo is driven on the inner frame. The signal from the servo on the inner frame needs to pass through the rotating connection provided by the slip ring.

To prototype this project, I started by 3D modeling a proof of concept single axis. This helped me visualize and understand the project better. The first iteration consisted of an inner rotating experiment platform with a solid drive gear. At this stage, I hadn't yet considered the second axis and was mainly focusing on getting familiar with CAD.

The next iteration of the design is more mature. It includes a complete inner frame with an experiment platform and the beginnings of a drive system. The metal gear servos are pressed with a small spur gear, which is a parametric design template that I found to be very useful.

Overall, this project is a classic prototyping project that required some experimentation and refinement.Easy. It wasn't an intimidating design at all, but that just presses straight on. Like no, no real fastening needed. Might add the fastener just to help draw it onto the splined servo shaft.

And here we're experimenting with the ideas of now coupling this inner frame to the outer frame. So we just have a pivot and a much larger pivot because we need to get the servo wires through this hole in the shaft.

So now we have three flavors of gear, the small spur gear, the inner large spur gear, and now the larger outer spur gear to go over that square shaft. And finally, a static outer frame to take the inner frame and accept the slip ring.

So in the mature design, the slip ring is actually used as a bearing surface. Probably a bit of a no-no with slip rings, but like we're dealing with very, very low torques here. So the slip ring is actually pinned into the inner frame and it kind of acts like the axle for it. Both an electrical slip ring and a mechanical bearing.

The outer frame has these fixed hard points to mount the servos and we're just using these self-tapping screws that come with the servos to connect them to that frame.

The gears are necessary because we want to gear down the angular velocity. If we just had a direct drive from the servo, I think these things would be spinning way too fast. And we also want to magnify the torque.

A lot of the work in this project was really in the tuning of fits. Big problems are lots of small problems, right? So the general shape of each platform, it's basically the same problem repeated. You've got something spinning inside of something else and then you step out and you've got something spinning inside of something else. A lot of how this is tuned comes down to the fit.We need to have sliding fits for all the shafts, but press fits for all the things that can't move. Things like the gears on the ends of the shaft. So there's a little bit of post-processing required for these kinds of prints.

You can see the square end for this shaft where a gear gets pressed on is printed in such a way that it has this kind of like elephant footing on the end. So there's a little bit of tuning that we need to do, but it's not so bad.

So I assembled the first prototype, had it running on the bench for a few minutes. And then when I looked over to it, only one of the axes was turning, which was very unusual. I think it must've been the outer axis was continuing to turn, but the inner axis had frozen.

In my initial prototype, I had the slip ring being pinched by the outer frame. So the part that should be spinning was instead seizing in the outer frame and the slip ring basically failed. The inner frame kept turning and turning and twisting the wires because the slip ring couldn't turn and it just drove itself to destruction.

Kind of disheartening, but really not a big deal. We just need to widen that hole to get a good clearance and print another prototype.

So I printed another prototype, got it off the bed, snapped it apart, reassembled the second prototype. And we did a soak test. We did a life cycle test, which was super annoying to have in the office. So we eventually wound up putting it in like a closed room, but I let this soak test run for a week.

So as we left it running in the corner of the office, I would occasionally just remember that I was meant to check in on it and kind of have a bit of a look and it seemed to be going all right.

So life cycle testing is successful, but the windowlooking for to build your own random positioning machine. And if you have any questions or need any assistance, feel free to reach out to our support team. We're always here to help.Sure, here is the transcript formatted into paragraphs:

"Gonna need. Claire has released the documentation so we can forward that onto you. The whole thing is under a Creative Commons license. Just go ham at it. There are the step files, the STLs that are ready for printing, the code for the Raspberry Pi Pico. Claire's immaculate instructions. Everything's on the repo. Go have fun with it.

The experiment platform is designed to have an interference fit with these. They're not test tubes. They're cultivation tubes. These things. And also to be wide enough to accept a very small Petri dish that you can just tape on. So also include links of where I source these to.

I'm not gonna lie. I found this project pretty intimidating at the outset. It's a simple enough idea to describe, to have something that rotates and something else that rotates. But there's always the devils in the details. We discussed using, instead of gears, we discussed using rubber bands and pulleys. And I thought the risk there for slippage might be a little high. And also we would then have to curate getting the right size rubber bands. I'm sure it would have worked quite well. It probably would have worked even better. Maybe even been a bit quieter. There's no backlash. It would maybe run a bit more silently. But I found a gear generator. So I used the gear generator to make these.

Oh yeah, I guess it's just the depth that I felt in creating this is just a testament that like big problems are lots of small problems. Working through my bin of prototyping material, I just kind of worked from the inside out. I started with the smallest problem and then just built the next layer out of it. And it didn't actually take that many tries to get there in the end.

It's pretty wild what high school kids get up to."These days, I didn't do anything like this when I was in high school, but it's just a sign of the times. The improving technology, 3D printing is accessible. So thanks to the Powerhouse Museum for coordinating something cool and thanks to Claire for delivering it.

If you make your own random positioning machine, drop us a line. We'd love to see what you get up to with it.

Until next time, thanks for watching.