In this chapter of the Zero To Maker workshop, we're looking at mechanical design, but oriented for makers. So anything that you need to know about material selection, joining parts together, physics and more. If you're new to this workshop, Jared and myself will be taking you on a fast-paced and practical journey to learn a wide variety of maker skills so that you have the tools and knowledge to make anything. Follow along as we develop our own projects and share insights into the process.

When making our projects, there are a lot of important design choices we make that make the project, well, a lot better. And these choices are based on years and years of issues and bad design choices that we've made. And that's what we'll be doing today. We'll go through some of the choices we've made in our projects and try and give you some of the condensed mechanical wisdom that we think is important for every maker to know.

First things first, what materials should your project be made of? So my project is Leo. It's an omnidirectional robotics platform that navigates around the house. And the spine of my project are these two aluminium extrusions here, which connects all the 3D printed parts together. After years of making things, I've come to settle on this formula here for my projects. And I see it so commonly used in the maker community. 3D print as much as possible. And if it's large enough, connect it all with this 2020 aluminium extrusion.

For Fab Academy, we had a mid-semester assignment where we had to make a machine and we built an Esky or a cooler on wheels. And that was just a whole bunch of motors and a cooler all mounted on a 3D printed and 2020 aluminium frame. It's just a great formula and I really like it. Here's why.

The first thing to talk about is this 2020 aluminium extrusion. I could have used steel, but I didn't because aluminium is about three times lighter. It looks much nicer and it's so much easier to work with. I might've opted for steel if I needed it to be stronger, or maybe if I was carrying some heavy loads, or it was going to be subject to some hefty impacts. But for this robot, aluminium is going to be strong enough. Aluminium also comes in this 2020 extrusion profile and this lets you use T-nuts to attach things to it in a really professional way. But the biggest reason I like using it is that it is much easier to work with. I only have simple hand tools to cut metal and it takes much more effort to cut and fabricate with steel than it is a softer metal like aluminium. And that's why I also like working with 3D printed parts. Once it's printed out, it's super easy to work with. A lot of glues work with plastics. It's easy to sand down and it's meltable.

Liam's about to talk about how it happened, but some of the wheels on our Esky robot got a little bit warped, but with just a little bit of heat, we managed to soften and straighten it up. When I printed out these purple pieces that connect to the 2020 extrusion, I realized that I goofed up and made the holes a little bit too small. It would be a pain to make this bigger with a file and you can't exactly drill a square hole. So I just softened it up with a heat gun and rammed the 2020 extrusion through it to make it a little bit bigger. It's not ideal, but it's a way better option than reprinting a part. And that's why having a meltable plastic is really fun to work with.

So when deciding on a material, think, will it be strong enough? Does weight matter? And how easy is it to work with? Another consideration is the environment that you deploy your project in. It plays a massive role in the materials you use and how you design parts. Now in my project, Plant Pulse, which is an automated agricultural monitoring and watering system, where I'll have an enclosure that needs to sit outside in the elements for months on end. A hidden, but major challenge that the enclosure has to face is UV light, which can seriously degrade most plastics, making it brittle and eventually crack and crumble. This part will be printed and we'll have to choose a suitable UV-resistant material. PLA won't last too long against the sun. PETG is a bit better, but a specialty plastic called ASA is perfect for the constant battering of the sun. So I'll be using it. A much more obvious challenge is heat, which obviously melts and warps plastic. And you'd be surprised, a lot of plastic starts softening at lower temperatures than you think. For instance, Jad left the Esky robot wheels in his car. It was only a 30-degree autumn day, but since it was inside the car and a dark color, the temperature of the wheels reached 55-degrees, softening the PLA and making them warp a bit. If the wheels were printed in a more heat-resistant plastic like PETG, we would have been okay. In my project, the enclosure will see about 60-degrees max, so PLA is definitely off the cards. PETG might be suitable, but ASA is also heat resistant and it'll solve the UV challenge. So we'll use that. And being outside means rain. You don't have to be a seasoned maker to know electronics and metal shouldn't be wet. The first idea has to wrap everything up in a watertight enclosure. The easiest and most common way to do this is use a Tupperware container. It's quick, dirty, and it gets the job done. And they provide a waterproof seal and latches to keep it all closed. We can also make our own custom 3D-printed enclosures and laser cut parts, which is what I'll do. But there is an issue. 3D-printed parts are porous, meaning water can seep through it. You can coat your prints and seal them to make them a little bit more waterproof. Now making it completely watertight is a lot of effort. Instead, we can control how the water will move around our enclosure. Of course, combining these two techniques of preventing and controlling will yield us the best results. My project is in a pretty harsh environment, but yours might not be. So it's always worth thinking about what environmental challenges your project will have to deal with.

So we have all these parts for our project. What's the best way to join them together? Let's start with permanent joints. My robot uses these mecanum wheels and I have these 3D-printed parts that I want to attach to it. The most obvious way to do this is with glue, which is what I did. One side I attached with CA glue and it's a good choice because it's cheap, it bonds instantly, it's easy to use and it's really strong. I'd probably give it like a 7 out of 10 in terms of strength. However, I went to glue the other side and it just wouldn't stick. And that's because this side of the wheel was a little bit curved. It wasn't perfectly flat. And CA glue only works well on nice and flat surfaces. It doesn't expand to fill any of those gaps. So I did a really bad thing and I hot glued them together. It is not ideal, but because of how thick it is, hot glue can fill in those gaps. This is probably not the best option, but I only did it because I needed to test out to see if these wheels will work. And hot glue is obviously a very quick process. However, it is messy and it isn't very strong. I'd probably give it like a 3 out of 10 in strength. And I can probably actually just rip these things apart if I give it a good... Oh, far out. That is actually really strong. That's a lot of hot glue. That might've been a really good bond, but usually I would be able to rip these apart. Hot glue isn't really a professional adhesive to use, but it is good for just temporarily tacking things down. The better choice would be epoxy. Two-part epoxy is cheap. It is a little bit messy and can be hazardous. So follow all appropriate safety precautions, but it can fill those gaps. And when it fully sets after a few days, it will probably be stronger than the wheel or the 3D printed part itself. It's probably a 9 out of 10 in strength. And those are the glues that I use 80% of the time, but of course there are tons more available for specific jobs or needs. Just ensure that you check it is appropriate for the material you are joining. For example, we wanted to use liquid nails to glue a 3D-printed part to the cooler on our Esky robot, but that adhesive didn't work on the plastic that the cooler was made out of. It doesn't work on polyurethane. Another permanent way to fasten parts is with welding. And I'm not only talking about traditional metal-on-metal welding, but welding 3D printed parts together. Of all the 3D printing tricks I've learned over the years, this has to be the handiest of them all. You can use a cheap and nasty 3D pen to weld parts together, or you can use a soldering iron to kind of stitch them together. This will definitely ruin your tip, so use an old or a spare tip for it. I've got a dedicated cheap soldering iron just for welding plastics. And if done right, it can be as strong as epoxy. So those are some ways to permanently join parts, but what are some other ways to join things non-permanently? Let's start with something that everyone knows, bolts. Most of the time you'll drill a hole all the way through, or maybe you'll model it into your print and put a nut on the other end. But we can get way more creative with this. If you're making something from scratch, we can design something called a captive nut into our 3D print. You can put a slot or a channel and that'll hold it nicely. Similarly, for laser-cut parts, we can cut a hole that matches the nut's outline and sandwich this between additional layers. We might not even need a nut. Jared's project screws the bolt right into the printed plastic. He made the hole a little bit smaller, and as you screw it in, it threads the hole. But this isn't super strong. A better way to add a thread to a 3D printed part is using the heat set insert. They're super strong, look professional, and if you're using 3D-printed parts, you should learn how to use them. But there are more fasteners than just bolts. Other fasteners include Velcro or hook and loop are perfect to keep parts weakly in place. Elastic bands are also great for lashing a couple of parts together. Zip ties are worth an honorable mention, perfect for joining wires, whole parts, and anything in between. And don't disregard the humble magnet. While usually not super strong, there is something to be said about hiding them inside a part and it magically snapping into place. And there's honorable mentions to parts that you can just buy or design into your projects, like clasps, buttons, latches, and zips. All of these methods are great. For my enclosure, I'll go with using the plain old bolt, skipping some of the heat set inserts because I'll be making quite a few of these and I'll save a lot of time. But I'll definitely be sure to use other methods to mount electronics and do cable management.

Let's end this video by talking about some important lessons when using wheels. So I've got my mecanum wheels here, and these are just fancy wheels that allow for omnidirectional movement. Somebody literally reinvented the wheel only about 50 years ago. And your first thought might be to mount them directly to the motor shaft, but that might be a bad idea. For a robot this size and weight, there will be a lot of force acting on these wheels, and it's hard to make a mechanical connection strong enough to hold them without bending or breaking off. On a smaller and lighter robot, this is absolutely fine. There is no issue with doing this whatsoever. But for this big boy, when he starts weighing, you know, a few kilograms or so, I instead let the wheel spin freely on this supported shaft that can hold the weight. And then I connect a motor to it and drive the wheel on that shaft. And there's lots of ways you can do this with the gears like this or with a pulley and belt system. And because I want this to spin freely, I mounted it on bearings, which is something that every maker should learn to use. You can get them for cheap and they are incredibly easy to use in 3D prints. You don't need to use bearings for everything though. The arms of my robot, they're just a bolt with some lubricant. Now these motors come in different versions that have different gearings, and they can have a max speed between 10,000 RPM, all the way down to 10 RPM. So how did I know which one to choose? Well, if you know the size of your wheel, you can figure out how fast the robot is going to move. Let's say I have a motor that spins at 120 RPM, divide that by 60 seconds, and that's two revolutions per second. Next, I need to find out the circumference of my wheel. I'm an engineer, and I still ask Google this because it's just easier to do. So if the circumference of my wheel is 25 centimetres and it spins twice a second, that means my robot is going to move 50 centimetres a second. Now you can convert that to kilometers or miles per hour or whatever unit you want, but I know how big 50 centimetres is and I know how long a second is. So it's really easy to visualize how fast that robot is going to move with that motor. And I went through this process with different gearings of this motor to find that I wanted 330 RPM, which is about 1.4 meters per second. Just be aware that as you gear the motor to increase speed, you are losing torque or how much oomph the motors can put out. If you want a deep dive into motors, we have a video linked below for you.

Well, that was a really fun dive into each of our projects, and there was lots of other topics we could have covered here. We probably could have done an entire series on just this chapter alone and just our projects and the lessons we've learned. Hopefully you can take home some of these lessons we've learned over the years. And if you have any other maker tips, drop them in the comments below. Thanks for watching.