[Webinar Recording] Expert Design Tips and Tricks for 3D Printing

Greg Paulsen: Hey everybody, we're gonna get started right on the hour today. We got a good amount to cover and some really cool things to talk about. So, yeah, welcome to today's live event. I'm Greg Paulsen from Xometry, and I'm excited for you joining us today. We're going to spend the next hour discussing design best practices for additive manufacturing. Our goal is to give you practical, actionable advice, and to help you optimize your parts, reduce costs, and get the most out of industrial 3D printing technologies.

I'm joined by my co-host, I'm not sure which direction I'm pointing, but he's either up, up, down, left, or right of me right now for you, Senior Solutions Engineer Matt Schmidt, for additive manufacturing. He brings a ton of hands-on experience. I'm thrilled to be able to share these design principles with you all today.

So, let's just talk about the agenda really quick. We're gonna start with some brief introductions, we're gonna learn a little bit about Xometry, and then move into our core 3D printing topic. I want us to first shift our paradigm into an additive manufacturing mindset, thinking about how your design will grow and what that really means. And then we're gonna show some examples and go into specific design tips and tricks. I also have some news and some resources that we're gonna share at the end, and that's going to be followed by our live Q&A.

You, the audience, are participants. You have the opportunity to submit questions at any time using the Q&A tab on the Zoom webinar platform. Matt and I will try to get to as many questions as possible at the tail end of this webinar. Any overflow, submit away, we're gonna tackle those after the event. Finally, this is all recorded. We're gonna send an email with our blog post link, that's hosting the recording. It'll have a summary and some of the questions and answers that we had during this event. That way, you can watch it again, or share it with your colleagues.

So, let's just jump right in. First off, again, I'll reintroduce myself. My name is Greg Paulsen, I'm Director of Category Management here at Xometry, focusing on additive manufacturing. I've been in the advanced and additive manufacturing industry for 16, 17 plus years. And I have really fallen in love with making parts, understanding materials designed for manufacturing processes, and the different processes and the outcomes of those.

Joining me today is Matt Schmidt, our Senior Solutions Engineer, and Matt, do you want to give a quick introduction?

Matt Schmidt: Sure, absolutely. Great to be aboard. Thank you very much. Again, Matt Schmidt, I am the Senior Solutions Engineer for Xometry, focusing on additive manufacturing. I have been involved with new product development and additive manufacturing for quite a while. I don't want to date myself too much, but my primary roles and the responsibilities here are to help our customers with the various, you know, designed for additive manufacturing, the various print platforms, materials and processes that go along with making additive parts. Great to be here.

Greg Paulsen: Yeah, absolutely, and thanks for joining us, Matt. And I think what's kind of cool about this, if you look at Matt's and my background, we see more than just additive, right? So we also kind of see it at a holistic, like, where it works, and, you know, where we suggest it, as a solution.

So, speaking of that, we work in a multidisciplinary company. We, our digital manufacturing marketplace called Xometry. Founded in late 2013, really started making parts for customers. In 2014, and actually went public. Xometry under NASDAQ in June 2021, so it's been a very exciting journey. And over the course of our marketplace, we've now are serving over 71,000. Active customers, which is super exciting. And what's great is every single time I do a webinar. Our quarterly reports will come out, and I'll be able to up that number, which is always fantastic.

We're two-sided marketplace, so we are working with the buyers, those who need stuff made, when you guys have a digital interface for that. We work with our suppliers, those who can produce those goods on demand. We celebrate the diversity of the manufacturing supply base, both domestic and globally. And we're able to offer solutions across the board from low-volume work all the way up to scaled and sustained production. And that's over dozens of manufacturing processes. Our platform marketplace, like I said, is that two-sided buyer and seller. What really knits it together is our AI integration. We have a platform, it's called the Xometry Instant Quoting Engine® for our customers. It's a single, one-stop shop where you can go and upload your technical files, get pricing and lead time right away. We actually use AI to predict and give you real-time costing information. On your project, you get to change the settings, configure your quote, pricing and lead time will update. And when you're ready, press buy, and then it'll go to our network. Our network is, you know, is advanced suppliers. We actually use AI to matchmake. The project scope with those who best complete that work. And we're able to make that match. Through both our domestic and global needs, depending on what your needs are for your project. So it's a really dynamic system, it kind of gives you that one-click feel. That you would have an e-commerce platform, only it's your custom manufactured parts.

But like I said, I wanted to introduce us, I wanted to talk a little bit about Xometry. But there's a lot of meat and potatoes today here, folks. So we're gonna be talking about why 3D print, and then the design tips for 3D printing.

So, first off, some of you may not be in the 3D printing world, some of you may be learning about it right now. Or may just need a little refresher course of what additive manufacturing really means. And additive manufacturing is all about growing parts. So, on the contrary, when you think about subtractive, like a CNC operation, you're taking a stock material. And then you're cutting it down to size. I take a sheet stock, I use a laser to cut it down, I take a, you know, an end mill or a drill. A reduced material, working to get my final part. With additive manufacturing. A part design, a 3D model, is interpreted and builds software, often called slicing software, because it's layer by layer. And it looks at how it'll gradually grow those parts. Using a base material in a controlled setting.

Those materials can be something like a liquid resin. So, stereo lithography is a great example for that, and usually have, like, kind of liquid resin smoother look to it. Powder bed, so think, like, a flour-fine powder for plastics or metals. To grow parts, typically using lasers or some heat process. And then filament base. A lot of you may be used to this if you think 3D printer, you may think about that printer that's nearby you right now on your desktop, where you have that spool. Of plastic filament melting down and zigzagging back and forth. Very popular, we call it FDM. But each one of those has its raw material. The raw material kind of executes the output, and we're gonna go into more details about that later.

So what drives cost, though? It's the making of the parts, right? It's the volume of material used. So, if you're making a brick, you're paying for that full brick. Right? But if you're making something that's kind of lighter weight and latticed out, that reduction of volume using that part will reduce the price of that part. Machine overhead. We are typically running industrial additive manufacturing. So instead of machines that cost hundreds of dollars, most of our industrial equipment costs hundreds of thousands of dollars. So it has a labor rate, it has a factory footprint, it has staff behind it. And that overhead can often be the driver of costs, especially in prints that. Take a while to make, so if you have a 100-hour print, for example, you're gonna be paying that labor rate against it, and it's probably going to be more expensive than the material itself. That you're running those parts in. And lastly, secondary finishing. Some finishing, like tumbling or automated bead blasting, or even chemical vapor smoothing. Is an automated batch-based process. Other finishing, anything that I'm sanding, I'm painting, I'm drilling, I'm installing inserts. You're paying a labor rate against, and that'll drive costs on a part-by-part basis, so it's definitely something to consider.

We… when you think about scaling in additive manufacturing. But what can I do with this? I don't need a tool. I don't need to tool up, like, injection mold to make plastic parts, or, you know, get large stocks of material and then machine down, in, you know, with the CNC machine. This is the winner when it comes to low-volume work and prototyping. You're directly, digitally manufacturing your parts when you're using 3D printing. That 3D file is being interpreted into build setup software. That build setup software is sending that information to a printer. And the printer will do its best to make your part. Because of this, the turnaround's very quick. You don't need a minimum, so you can make… you could serve a customer of one with 3D printing. And you get parts within days, typically, on order. And the designs, because I am growing those parts, and Matt will talk a little bit. More about this later, you can be more function-oriented. In CNC machining, leaving material behind is good, because it's less work, less cutting to be done. In additive manufacturing, you could really make, specific designs. And make decisions on your designs to orient towards function. Which is very cool. And again, because it is just a 3D CAD model, it's the lowest barrier to entry next to one other process that we offer.

Sheet cutting technically wins, because sheet cutting, which is a, I think, laser cutting of a flat sheet. Just requires a line drawing, a DXF file. But you add a third dimension to that, and now you've accessed the entire 3D printing landscape.

So, this is how we see the world at Xometry, right? The world of manufacturing, this is plastics and rubber. But we see manufacturing as a continuum. Because we're looking at a diverse amount of shops and services, we're able to kind of see where different technologies. Lie within this continuum. If you look from left to right, I'm looking at low volume, think, like, one-offs. Or prototyping. Up to, kind of, the dozens of parts, dozens of hundreds of parts, and then moving all the way up to high-volume production.

If you pause right now and take a peek around you, not my desk included, but take a peek at your desk. You may find that most of the plastic components on there are probably one of these processes, like injection molded, extruded, compression molded. I forgot thermoforming, but thermoforming would be a great example there. As well. So, ultimately, if you're serving, like, the hundreds of thousands or millions, you're typically in that world. But there's a lot of service needs that hit this middle ground, where additive manufacturing can often meet or surpass. Where you maybe typically urethane casting, which is great for that 5 to 50s unit, or using plastic CNC machining, you may be able to find alternatives. And advantages in additive manufacturing in that realm.

And this is just a quick list of technologies offered at Xometry. So when you go to our Instant Quoting Engine you can see, you know, CNC machining, sheet metal, sheet cutting, you know, very bread and butter for us, as well as additive manufacturing. And then the formative processes. Think, plastic injection molding, metal die casting, metal stamping. And urethane casting extrusion processes. So regardless of where you are in your product development lifecycle, whether you're still making that part, or you're looking for a new supplier of something you're sustaining. We usually have an option to meet you there.

But I am talking about 3D printing, which turns out is not just one thing. It's a lot of things. So with 3D printing, we have different verticals of technology. What they have in common is that. Additive manufacturing side, so I'm growing parts using a digital file. The most popular processes typically are these ones on the left here. We'll talk more about those as well. But we usually think of them in different categories, plastic and metal. As well as under plastics, we typically have thermoplastics, so something like ABS or nylon, where. When I apply heat, it goes into a liquid state, it cools, and now it's a new shape. That's gonna be a thermoplastic print. Thermosets. I think if you've ever had a… had a cavity filled, where they have to shine that UV light in after they put some, you know, that liquid or paste in there. That's gonna be something like a thermoset material, where it hardens and catalyzes with exothermic reaction. So think about resin photopolymer printing, for example. Very good example of a thermoset or resin-based 3D printing process.

Both these metals use a powder-based process, but they have different ways of making the metal part solid. Direct metal laser sintering, uses a laser to selectively melt areas. On a build chamber and build apart bottom-to-top with a very fine metal powder, kind of being fused down to a build bed. And growing from there. Metal binder jetting uses a glue binder, and then has secondary processes, so it'll make the shape. Kind of gluing later by layer, creating a green part. And then you go through several stages in order to take that green part, move it to a furnace stage. And then it'll actually sinter into a metal part.

We see that these four on the side tend to be the most popular amongst our users at Xometry. Because they have a lot of advantages when it comes to cost competitiveness. Design flexibility, and choices in materials and performance. SLS and MJF are very similar to each other. They both are… can print mostly in nylons, but they also have polypropylene and TPU and filled materials. But they have a unique ability to. Nest in multiple dimensions, so I… I don't just have to think about my parts stacked on a table, and saying, that's my build. I could actually put many parts, building 30 to 300 parts per machine per night. So you have a lot of production viability.

FDM is a… this fused deposition modeling that kind of spool a filament, that's printing those parts, has that boutique aspect where I could actually choose the plastic and the color that I need for that. And if I need to scale, I just get more machines. I make an army of them. To produce parts on demand. And then stereolithography is the OG. 3D printing technology, so SLA. Is a, is a technology that will use UV light. To cure liquid resin on a layer-by-layer basis. And there's been this engineering resin revolution that's been happening over the past 5 years. Where I used to think all SLA was just brittle stuff that, you know. You have a once and it breaks on you. Now you have real high performance, even UV withstanding materials that are being built within this process.

But every single one of these, and Matt and I were talking about this when we were practicing, every single one of these is our own webinar on it by themselves. So, I will move on, but always happy to answer questions about these processes later. So… yeah, with that, I'm gonna go in the design fundamentals and, and. Then, Matt is gonna help kind of cohes it together. A little bit.

So, we've been talking a lot about growing parts, and when I have this little animation here, kind of showing. A, you know, a part looking at it from a section view, going on that layer-by-layer basis. I find myself doing this a lot. When I'm changing my mindset, I'm making a part for 3D printing, making that design. I'm using a section view a lot, and kind of thinking about it by orientation. How am I intending to design my features so they grow with each other using design best practices? You're gonna get in this mindset a lot when you're designing, your 3D printer parts.

The other side of this is that additive manufacturing machines, they are. Doing their darndest to make your 3D digital design. But you cannot stop them. And tune a feature, like you could in CNC machining. So, for example, say a hole is a very tight tolerance. I can't, run that operation, stop the mill, go inside, measure. Adjust my, you know, adjust my offsets, and then cut again, and check it out. Out of manufacturing, the machines are calibrated using, essentially, calibration pieces, calibration cubes. And then they're producing… then you're producing those parts with, standard scale and offsets to them. They… there are machines that are more accurate than others, but all of them are considered producing a near-net shape. And that's important when you think about your tolerances and how you're creating your prints for 3D prints.

And lastly, it turns out we're not in space, gravity still exists. Support structure very common within additive manufacturing, so most processes require support structures. And what they are are features that are sacrificial, that are built. With your part, to hold overhangs. So if your part is, like, a T-shape, for instance. And I'm building it kind of, like, as if it's standing up like a piece of text. I'd be building sacrificial features that are growing up… going up to hold that overhang of the T, of the T at the top there. Then they get removed after the build. This picture here is probably the worst example of showing a lot of really strong supports, but it's very visual for a presentation to show these little nubs here where that sacrificial support structure was removed.

And we talk about these base materials a lot, and again, this is getting yourselves in the mind of additive manufacturing. Resolution is very process-dependent. I'm going… I like to use my Sharpie and mechanical pencil example, when I talk to my customer… customers, because of how. This may affect your design, and even, like, the design tips we're going to talk about later. How you may, for example, in a powder bed process. Want to make a larger hole than something that's… you're able to do. In a resin 3D printing process. So, I always like to think about powder bed fusion, where I'm putting a lot of heat against a metal or plastic part, or plastic material to fuse it together. As a Sharpie fine point. It's got decent resolution, so that's why I use a fine point and not a marker. But if you linger anywhere, you're gonna bleed. Like, you're gonna get that… that ink bleed. So, if you look at my Sharpie fine point, that little circle right there, it's kind of filled in. Like, it's. Even though there wasn't supposed to be material there. That bleeding, that extra heat that's radiating out there, actually would fuse that up. In a process that's… especially at a very, very small scale.

When I look at UV-cured resins. The resin itself, like the liquid resin, typically has a liquid look at the end. And you have really great detail and clean edges, so it's typically the best resolution. Processes are gonna be resin-based, 3D printing. And then lastly, that filament-based 3D printing, it can sweat the small stuff, so it's… I don't even loop on this, right? I kind of go close together and pinch and then move out. I am zigzagging a filament, a melted filament, back and forth. To create my layers on a layer-by-layer basis. And, and with that, it has more course features, but one of the great things it can do is it can build large, broad mechanical structures really well. But if you have something that's very high detail, or has very fine features they may be sacrificed. And I'll show you what I mean in just a second here.

So, this slide, I'm calling it the "Who's That's Pokemon" slide, but we're gonna go through our different examples of powder bed, resin, and filament. So, first off, let's talk about multi-jet fusion. So, this is a plastic powder bed fusion process. The parts start with a white nylon powder. Layer by layer, it's fusing that, using essentially an ink mixed with heat to create, create a selective fusion. And you can kind of see, like, if you imagine this T right here is 10 millimeters tall. You can see how the finer details, like, there's actually a smaller text there that didn't even resolve on this, but you see the small features kind of have a rounded look. About this, and it's a little fuzzy, but I could still read it. You know, it's still pretty good, but it has some fuzzy features to it.

When I look at this compared to a liquid photopolymer. I have much crisper details. So, in this case, again, that's 10 millimeters, 15 millimeters, 7.5. And, 5 here, but you can see that the smallest text actually did resolve in this. I don't recommend making all your text like this, but you can get away with it better in SLA. But you can see that you have much more crispy features, from the get-go with this. With this process, so better feature detail resolution.

And then lastly, and I don't… I'm not knocking on FDM, by the way, because it's an extremely versatile process, and we print a lot of parts at FDM. But if you're doing small text, you know, 7mm text or smaller. It probably isn't the best process for you, because you could see. That. That kind of line stroke, right? So you could see that filament and how it's deposited. On these parts, and the more. The more gradual that curvature is, the less steep the slope is. The more coarse it looks, and as it actually is going more vertical, you tend to get some better results on it. This is probably, like, a worst-case scenario print, but I really used it to emphasize the differences between the different outputs. So depending on your part design, you may be going to one process loop. Versus the other.

And that brings us into this paradigm of function and finish. Every single 3D printing process. Has its own outcome. It has its own set of materials, its own detail resolution to it. So when we start thinking about what I want to do, like, what is my goal? For my project, we're gonna be talking about, we're gonna be talking about what processes should I start to down-select or look at. When I'm thinking about my 3D printing design. If I'm looking for something, like, aesthetic, cosmetic prototype. I'm usually looking at what details I want to do. Like, what… let's spin the CAD, let's talk about those, those parts. If I see small text or features, I'm probably going to put you towards a resin-based 3D printing process. Clarity, by the way, resin-based prints can. Make clear parts. So it's really the only place to go for that. If you… you can print in clear filaments on FDM, for example, but you get a very diffused part for that. You know, smoother, smoother parts, you have a few different options available, including some secondary processing for SLS, MJF. But anything that's just general prototyping, honestly, SLS and MJF. We could do a whole webinar on when not to choose it, because 90% of the time, those are probably going to be your go-to's, if they fit into the building area.

So, in this case, we have an example of, I want to clear part, you know, SLA. Clearview, Quick Clear, works really well. On the other side of this, we… a lot of what our work is, is with mechanical engineers. We're making stuff that does stuff, right? So we're looking for functional parts. We have a few down selects here because we have. Our menu's very well oriented towards mechanical components. The first question is, is it going to be metal, right? Because I have… only have a couple metal options, so metal's a very quick down select. The second is, do I need… is this going to be rubbery? Like, is it a short A value? We have a few different processes that have a short A value, so we could print rubber or rubber-like components. If it is just going to be a rigid plastic. Then the question is, what's it gonna do? Doesn't need to withstand heat over 150 Celsius. Does it need to be flexible? I mean, ductility, does it need to pop into place and pop out? Chemical resistance, impact resistance, all of that can be a down select criteria for you, and then under each one of these umbrellas. We have a variety of materials that can fit your need.

The last thing to talk about is when we talk about these 3D printing platforms, they are platforms, right? They're physical spaces that have capacity limits. This is a general overview. I know, like, Matt, you've… we've raised our hand on this, and you can comment on that in a second as well. But in general, like, DLS and DMLS platforms kind of cap out at that 9-10 inch range. On part size. Select laser sintering and MJF has a sweet spot around that 12 to 13 inch range, or it's, um. Or smaller, as well as metal binder jettings, kind of in that world. And, SLA and PolyJet, those resin-based processes, can go larger, so 24-inch is kind of like a sweet spot cap-out for that, but FDM typically has the largest format. That's because it has a lot of stability, it's building on a big, flat build plate. And we could easily… we could casually print parts up to 36 inches, with the machines that we have. And Matt, I'm not sure if there's any… anything else you want to say on that… on this slide here.

Matt Schmidt: No, yeah, we pretty much categorize, the systems and the materials into into general, commercially available, off-the-shelf type solutions that we can rapidly provide. That are repeatable, reliable, always available. Yes, there are machines that are much larger, and can do much different things. We have some of those. If you don't see anything here in this slide stack, doesn't mean we do not have access to it. Just means we might need a consultation to talk about it a bit. Feel free to contact us.

Greg Paulsen: Yeah, I got a drawer full of really cool parts for processes that we don't offer right now, and you know, we want to add to our marketplace as they mature. So yeah, that's, you know, I think… I think this is your rodeo now, so, like, how do we add value?

Matt Schmidt: Yeah, oh yeah, I need to take a breather now, thank you. Right. So, so as Greg was explaining here in the past, you know, some different… the different options, the different materials, different platforms, there might seem like a lot of options in 3D printing. But, you know. I'll be the first to say that just because you can print doesn't mean you should. We need to have, you know, there needs to be a driver to print parts.

Looking forward. As additive manufacturing is just another tool in that toolbox. That's why we carry over 20 different manufacturing processes, because additive is just one of those tools. But you have to think about using additive moving forward into a design, not looking backward. It's always better to look forward. So when you look forward, you can add value to your design with the built-in adhere advantages of additive manufacturing.

Like lightweighting parts, huge in aerospace. Using optimization software to create advantages in buy-to-fly ratios. Custom lattice structures that can lightweight parts while keeping them. Maintaining superior strength. With that comes design complexity. Lattice structures, gyroid features, things like that, they are not simple figures, they are not easy to manufacture. Hence, Advantage Additive Manufacturing. Customization. Pretty simple. Photo there of customization, it's just the Xometry logo. But that just goes to show, you know, you can serialize parts. You can create patient-specific parts, you can add logos to your parts, you can add. Scan codes and things like that to your parts for all different types of serialization, so customization is very handy with additive manufacturing, adding more value.

Design freedom. Just because you have one idea in your mind doesn't mean you can't print 3 or 4. Let's, you know, look at generative design. We're gonna talk about that really big with these CAD packages out there nowadays. Creating lots of iterations of different types of design configurations. And with, lucky with additive manufacturing. It's rapid. We won't forget about that term, rapid prototyping. We can print a lot of parts really fast, so you can go through iterations faster. Get to your end goal, get to market faster. And then also, let's not forget about park consolidation. I once heard from somebody in additive manufacturing it wouldn't be an additive presentation without mentioning the GE fuel nozzle. The GE fuel nozzle was a huge part consolidation with DMLS. That added value with additive manufacturing, huge value. So, that's just a great example of part of consolidation for additive manufacturing.

Greg Paulsen: By the way, I always… when we did rapid prototyping my early career, one of the things we did for that multiple design iteration is we call it the shotgun approach, right? So you can… we were working at… on grips and things, and we would basically print 5 or 6 slightly varied iterations of that in one build overnight. Send it over to our stakeholders, get their feedback, like, that same day, do some modifications, print a few more. And it helped us really narrow down. On that need very quickly, before we got in the state of production. So it's just super, super useful to do.

Matt Schmidt: Yep. And you can see, you know, as you're looking at. Your design intent, where you're going to be in the future, is it going to be 3D printing? Greg had that great slide earlier that kind of showed additive manufacturing right in the middle. Of that, the road to production, so to speak, slide. Where we're right in the middle of that production range. And, you know, how those costs shake out is going to be important to your costed bill of material at the end of the day. So, you know, we're looking at. Kind of complexity versus cost? In 3D printing, in traditional manufacturing, so we've got. Just a simple 3x3x2 volume type of geometry. And we've got a traditional CNC, and then two additive processes down. Down the column, and then across the… across the row, we've got various volumes of infill. For those various geometries. And what's interesting is, as you can see with CNC. Obviously, like Greg mentioned earlier, toolpaths cost money. So, the solid block of material is going to be very inexpensive in CNC. And there used to be the presentation pitch, I keep on going back to my historical notes from over the years. Complexity is free and additive. Not exactly, but, you know, it does prove a point here where you're moving across with CNC, and you're adding complexity, you're adding cost. If you look at DMLS or SLS in the additive process. As you move to the right, you're decreasing in cost. It's adding complexity. It's not exactly free, but you're removing runtime, you're removing material volume, you're removing energy consumption. Therefore, parts become cheaper. That's your scale of economy with additive manufacturing and production. It will bottom out pretty quickly. But as additive continues to industrialize, this will improve.

Greg Paulsen: Yeah, absolutely. And I always thought it was interesting, so I have this shelled in the middle here, so this is just… less materials, so you see this drop, right here, because it's just lower volume than the quasi-lattice, and yeah, just that volume really does drive the cost there.

So, we're gonna get some specifics, now, talking about design tips and tricks. Something to note is I have a lot on these pages, I'm gonna talk about… talk more about the overview, and how… what we see on the Xometry side. But we can always, rewind and go back, especially in the Q&A section, to talk a little bit more about a specific design tip or trick. Because we see a huge variety of these, and I do want to talk with the biggest pain points that we have when we receive parts.

The number one reason why we call any customer. After an order is submitted in additive manufacturing is a thin feature. That's gonna be, typically, you know, a thin wall features or, or, like, thin pin, features. Or gaps and holes that won't resolve properly within the process selected. These are some general guidelines, but something really to look at is we consider walls in two different ways. Unsupported and supported. An unsupportive wallet would just be a direct cantilever. It's… we really wanted those to be at least 1.2 millimeters thick, 2.5 millimeters is, is. You know, better for a more coarse process, like FDM, especially large format printing. But just know that walls below half a millimeter. They won't resolve or easily damage. So does this mean that printers can't make parts under half a millimeter? No. But what it means is most parts are not benchmark models. Most parts are not. Cool little test to show off the capabilities of a printer or something. Most parts have some function to them. They need to be handled, they need to be installed. And keeping within these guidelines is going to let you make a functional part that won't break the second you open up the box.

Supportive walls can be a lot more forgiving, so when you think about even designing walls with a curvature, or a T-shape, or a C shape, or a X shape. You are adding additional beefy structure to that design. You can go a little bit more broad, or sorry, a little bit thinner on supported walls, because you have, you have something else that's helping it along, helping it resolve and survive. I have this little design here, kind of showing a thicker wall. It does… requires less of a backbone. In fact, this could probably stand alone just fine, but as you go along with the thinner walls, you're building more of a support for it, so less cantilevered material. And more… more support for it to resolve and survive, handling and use.

Holes. A few things you need to look at is detail resolution, as well as the process in holes. If you have a thermal process, hosts had to shrink a little bit with them, so if you have a hole smaller than 1 millimeter. It may… it may not resolve fully, or you may not be able to stick that pin, or have that use. If it's a vent, maybe you're okay. But if something more functional and mechanical, you definitely want to use, go above a millimeter and work those guidelines. The larger aspect ratio of that, the larger the hole you want to have, because there are things, like, it could just be one location where there's a little bit of a shrink… a little bit more of a shrunken hole. In that case, that could ruin your entire application. For making those parts.

Gaps and clearances have very similar roles, and if it's easier to clear, you can be a little bit more forgiving in your design. If not using the general guidelines, about half to 1 millimeter is usually a good minimum. On what you're using. But. In general, if you're designing for injection mold design. You know, coring out those objects, using ribs for stiffness and strength instead of… instead of thickness. All those design features are going to help you. With additive manufacturing as well. So, a mental model, especially with that uniformity. For… or for ejection molding, works really well here. You don't need to worry about undercuts or draft too much, but they can help.

So, when I look at features and text, for example, text is just a series of thin gaps, thin walls, and these wires or pins. So, think about text as features, not just… not font itself. But usually minimum feature size, 1.5 millimeters is usually my go-to, and if I need to stretch the limits, I'll go a little bit thinner to a half a millimeter or so. But before… but beyond that, you're gonna have detail loss. Like, a good example is that. That powder bed fusion design I showed earlier, you kind of have more of that fuzziness that comes out of it.

Wires and pins, they can have trouble printing themselves. Especially if they're too thin, so 1.5 millimeters is a good minimum. But look at that aspect ratio. So, like, the 5X, 5X aspect ratio there. The thicker the better, or adding something that protects them. Because it's not just, can my printer do that? Again, most people aren't printing benchmark models. Most people are using 3D printing to do something with it. You want something that protects those features. Because a wire pin is very vulnerable, it could snap like a pencil, if it gets hit the wrong way. Sometimes I call a pin in a wrong place on a 3D print design that's otherwise perfect, a godpin, because if it breaks, your entire part is scrapped. That's where I ask, like, maybe you should, you know, maybe you should make that into a sacrificial feature, or use a metal pen and make a hole there. You know, something to, to take that vulnerability away in your design.

So we talked a little bit about text, but engraving and boss seem kind of the same thing. Half a millimeter is usually enough to get you definition. One and a half millimeters is. Is pretty good, but you don't really need to go too much further than that if you're doing either engraving or embossing, if you're just trying to show a logo or a feature. And just know, if I'm thinking text. I also think, Sans Serif Tech, so aerial font, for example, is fantastic.

And, like I said, we're going over these. I see some questions coming in, so please keep on asking away as well.

So, when we talk about features, the other thing we need to consider is that support structure, right? So we're talking… we've been talking about growing parts in your net shapes. And support structure. I… I can build features on a design that do not require support. This is something where usually I'm using a 45 degree angle. Will help out a lot, so I have, for example, a little kickstand design here. My kickstand, just notice that it's a bunch of 45 triangles, so. I have these features building off naturally, I have text on this, my text is popping off at a 45 degree, I have my. Logo on it, my logo's popping off at 45 degrees, so if I build this this orientation here. Everything is a self-supporting structure. I could get away with a little bit of an overhang, no more than 2 millimeters. Bridging, so if I have. Something to think a little keyhole or archway or something that's gonna be, printed. With that bridge parallel to the build surface of a park. That's, you don't really want to span more than 10 millimeters, but even that, like, we may require support structures for. Just know that this is more important, for example, in metals. Where metal… your support structures are metal, too, and they're maybe harder to remove.

The last thing, I'm putting never down on knife edges. Additive manufacturing is a process where you're using a smaller base to build something up. Think about it as a bunch of little Legos. How many Legos do you need to stack together to get a knife edge? You'll never get there, right? So when we… when we try to grow a knife edge, like something that's overly acute. In a design, you're going to have either a feathered look or unresolved features to it. I recommend popping a half millimeter chamfer on those edges to get a cleaner, more consistent look that's gonna look the same. Bill to build, and that design.

Lastly, just general tips, so how do I kind of clean up my design on this? Fillet, everything is really a really good design. Role. So, typically, if I have something that's cantilevered off of my part, if I don't have a fillet there. Then I have acute stress points right at that joint. If I put a fillet there, it just helps even out the stress, making that feature much more robust.

I also add drafts, so if I'm making my text, I usually add, like, a 3 to 6 degree draft on my text, or any protruding feature. It still gets me the cosmetic needs that I want for my design. But it gives me a wider base, so it kind of beefs up the base of that design to allow me to have finer features at the end where you're going to be perceiving that. Draft is a really great tool, especially when you're making text, text in your design, or any, type of logo.

And clearance channels are something that's really important for liquid-based processes, as well as powder-based processes. I typically am post-processing with blasting. Like, I'm blasting. Beads, down into these parts to clear them out, or I'm using, like, a water jet or something. To get out that material. If I have a blind boss, so, you know, a screw hole that just ends there. It has uncured or unsintered material within that. If I'm trying to blast it out, it's… all I'm going to be doing is packing that material down to the base, and I may need to use a toothpick or something to just help clear that out. If you add a little exit hole right at that base, it's gonna give a channel to let air flow. And clean out those whole areas. So, it's really great when you think about designing for additive, to add those exit holes, and because we have that free-form design. It's… it's not a bother to the part design as well.

I'm not sure, if you had a… Matt, if you had anything to comment on this… those designs. I know, like, we went through some of those, especially on metals.

Matt Schmidt: Yeah, it's increasingly important, you know, to consider support structures when it comes to metals. And considering part orientation. When it comes to your DFAM approach, your design for additive manufacturing approach, which will help you. A look at, support structures, how they can be removed. Where are they going to be contained, and if you can get away with. With utilizing them or not utilizing them. But, we'll talk about a tool coming up here that, can help eliminate some of that for you.

Greg Paulsen: Absolutely. And, yeah, we'll get there in a second, but the last thing is there's stuff that you can do after your print. So, everything I've been talking about so far is, will my print. Make, and will it survive the post-processing? So you can do, what you can do better, like, in file preparation, you can add textures. To your print, but then, secondary things, I could do finishes, like adding cosmetics, like dye or chemical vapor smoothing. So what is a texture? Texture is something that's digitally applied. This is different than texturing a tool for ejector mold, for example, where you're. You're adding that feature to the tool. This is something where your CAD file will have those features to it.

This mouse, actually, I was playing around with a Formlabs tool that came out called Meshi, and I added some snakeskin to my mouse cover here. And that… so I designed my original design, which you see on the right-hand side. And then I applied a digital texture to it, and what it can really do is it can help mask features, so it can help mask. You know, coarse layer lines on your design. And give you something that has more of a professional cosmetic look to it. Even a fine matte texture, especially with a resin print, can give you a really cool look to that, and something that gets applied. Before you load the part up. To those designs.

Dye or paint can give you the colors that you're looking for, and chemical vapor smoothing can also, help take some of these plastic parts. And bring them to the next level, getting them almost an injection mold-like quality to them. The other note is, when I think about threads, threads are, again, a series of thin walls and thin gaps. Things happen there, but what can I do to mitigate that in my 3D print design? Using threaded inserts. So, when you're designing for a 3D print, you may move to use, for example, brass screw to expand, or heat-set inserts are very, very common, and it's something that auto-quotes again on Xometry's site.

So… I'll go over this, you know, pretty quickly, and again, also, I see questions coming in, so please keep on asking questions, that's awesome.

So CAD files, we usually like to accept something under 100MB. If it's over 100 megabytes, you probably have exported it at a too. Two small, are too high of resolution. Like, you have a resolution that won't actually change the outcome in the physical printing world. We do like, step files. Over other file formats, because the world is your oyster, you can use STEP for all sorts of manufacturing processes. But if you do need to do a mesh file. The format 3MF is really great, because it can hold, first off, it has a smaller file size. Towards complexity, like, it's much smaller than an STL file. It also can hold stuff like color data, so you can add color and textures directly into a 3MF file. And we have different printing processes that can make those types of parts.

Other things is when you're uploading a technical drawing, if you do have a technical drawing for your need, use our manufacturing standards. You can find that at Xometry.com forward slash manufacturing standards. That's gonna show you the general outcomes, and that's… should be your title block. You shouldn't be using CNC tolerances against a 3D printed part. We want to make sure you manage your expectations. And on this list here, I mean, we've seen a lot of great outcomes with Fusion, SOLIDWORKS, Onshape, NX. Great CAD software out there. If you are accessing free software, just make sure that you can export it as a step file, because step, again, is a universal, it's like the PDF of 3D file formats.

And, and for 3D printing tools. I have, like, I have my toolbar. GrabCAD print and Formlabs PreForm right now, alongside with other professional, tools. Because they helped me simulate and visualize a build of something. For a very quick gut check. They're free to download, and you can go and take a look at, for example, FDM or other processes. And Formlabs Meshy is an online tool, and then Microsoft 3D Viewer, believe it or not, great way to verify. Parts, especially if you add color. Matt, do you have any comments on that? What do you see?

Matt Schmidt: Yeah, absolutely. A simple software like GrabCAD is going to enable. The part to be sliced in Create Support. And generate a support preview for you that you can step through slice by slice. To find out where you might have issues with your print. Thin wall conditions, trapped support. Too much support, too little support, under, you know, downskin angles. This, you know, step through, slice by slice will help you. Creates a better 3D printed part, and a better 3D printed design.

Greg Paulsen: Yeah, absolutely. And I do have to say, usually if you have a desktop-free printer at your work or home. You are typically our better designer for additive, because you've probably already broke it apart by accident, or had a build fail because of a design feature. So that empathy built with having a desktop printer can actually. Go a long way in your design process using more industrial equipment.

Matt Schmidt: Alright, so let me take this one. Let's add some automation and some horsepower to your design. So, if you're, you know, if you're fortunate enough to have, like, a seat of, you know, Siemens NX additive manufacturing software suite. You would be empowered with some really awesome tools, and part of that is generative design. So, generative design is going to be, you know, defined as a computational, iterative design process. It's gonna use computing horsepower and algorithms to create a wide variety of CAD models and possibilities, CAD and shapes. That are driven by user-defined specifications. Material, space limitations, loading characteristics. Could be part orientation, static, static mounting points, dynamics, things like that. This is going to automatically generate these different CAD models for you. That can be optimized for specifically additive manufacturing. You can see some of the topology optimization come in as the iterations go from left to right. On that, on that mount. So, you can see, you know, this is only possible in a 3D printing situation. And, on the upper right-hand corner there, we've got an example from Formlabs. That shows a welded assembly that is transformed into an additive manufacturing. Component that is consolidating those parts, optimizing material usage. And then coming out with, you know, a really organic design that can only be, you know, 3D printed. This is where you're adding value to additive manufacturing. And creates huge value. Again. NX is a great, you know, platform for this, but there are others, Fusion 360 has one. You know, there are others that you can find that will help you with a generative design technique. And focus on additive manufacturing.

Greg Paulsen: Yeah, absolutely, and what's great is there. That list is getting longer every year. Because I'm a big proponent of generative design, but accessibility. Is always a challenge. So right now, it's very much, seasoned engineers are having access to these types of tools. But even, like I said, like Matt mentioned. Fusion has a generative add-on now. It costs something, but it still gets you some really cool results. But it's that cool idea of saying, here's the thing I wanted to do, here's the areas it needs to attach to, here's my keep-away zones. Make it. Like, it's… it's prompt engineering for a 3D space.

So, great time to remind you all, submit your questions. I got a few more things here, and we're gonna jump into questions and answers in about 2 minutes here. But yeah, we've talked about this. We make a lot of parts. We make so many parts all the time in over, 8 different manufacturing. Additive manufacturing processes. And we also have that holistic view of other processes, so we can help you graduate if you need to. Or design and optimize if you're looking to stick with additive for a high, you know, high-performance production parts.

The best way to get started is Xometry's Instant Quoting Engine®. Go to Xometry.com, you can log in, it's a secure, upload, so you could upload your 3D file. And pricing happens in seconds. So that is, you know, 97% of the time, you're gonna be going and using instant quoting engine and starting your quote. You could go get your pricing, configure, your options. And press buy. And it's a very e-commerce checkout for something that's usually a headache for, for manual sourcing.

We also have our request a manual quote option. So we have a manual quoting team. And Matt sees a lot of these, too, so, but that's something where it's gonna… we maybe help you set up the quote. Or we're turning around that quote because of a custom finish, or a custom inspection requirement. Or material that we have access to, but may not be instant quoting on our menu. We're always happy to explore those options with you.

And then the third part, and this is how it usually happens after we've already gone through steps one and two with you. Is… if you're in a production-level program, say you're a drone company, or an autonomous vehicle company, and we've helped you build. Certain… certain aspects of this. Okay, I need to serve my customers, I have demands, needed for my assembly plants. How do we build a structured schedule? That's where we may work with partner engaged quoting to, to reach into our network, find the best solution for you. And… and help engineer that sale to make it work with great outcomes.

And we have a lot of resources. Like I said, every single process is its own design guide, its own webinar, its own everything. So we… we like to put a lot of resources on Xometry.com. We have, of course, webinars and recordings, like, we're gonna have for this event here. We just launched our design guide for stereolithography, so that is a brand new, freshly published. Republished, I should say, but updated for 2025. So that's available right now. You can go to our stereotography capabilities page, and also our email follow-up, we're gonna have a link to that as well. And you can always check out our blogs, product updates, or case studies.

I know, Matt has a really great resource tool as well. We have these, a quick design references page, so if you go to Xometry.com, go into search and click the… or search for a quick design references. You're gonna see these one-pagers per process that we have for you that's super convenient.

And just lastly, if you haven't checked us out in the last month, check us out. We've lowered a lot of pricing. June has been a 3D printing month for us. We've lowered our pricing on select laser sintering. We've added a more competitive offerings, using Formlabs for SLA, added more competitive offerings under FDM, including international options. And lower the pricing on Multi Jet Fusion. So our most popular processes have all seen decreased pricing. So we're really excited to have that. Oh, thanks, yeah, that just. Put out a link here. And lastly, you know, check us out. Check out the website, and we also have a. A, a code for you, ADDITIVEWEBINAR10, that you can use on your next order. And with that, let's jump into some questions here. So… I have a question, does shrinkage… does shrinkage affect. Are the shriegus Effect final parts? Do you want to touch that bet?

Matt Schmidt: Sure. So, any type… yeah, of course, polymers shrink. Especially when they're heated and they're cooled. The slicing software and the computational software that's provided by the OEM has an automatic. Scalability factor for the parts when it comes to printing and cooling and post-processing. So you don't need to worry about that on your end. You submit the part that you want to see. It's very WYSIWYG. What you see is what you get. So, once you send us a CAD model, that's what we're printing. Within parameter tolerances.

Greg Paulsen: Yeah, absolutely, and that's what we hold ourselves to, is that manufacturing standards, that's… that's gonna be your… to manage your expectations of what you're gonna get when you press print. So absolutely, look at the standards over the worry about what this process is gonna do. We manage a lot of that, digitally on our side.

Alright, so, I have a question, so it's a question on warping SLS parts, but they're talking about metal, so I'm gonna say, like, a DMLM or DMLS here. Large flat parts printed in stainless steel 316 have a tendency to warp. If I have ribs on the 3D model and plan a machine the way myself.

Matt Schmidt: After I received the part, will that help, or will the part spring back once the ribs are gone? So the key is here, this is orientation related. It's got a lot to do with the geometry, and we talked about ratios of walls and ribs and things like that. So, I mean, the thinner you get, you know, the more difficult that's going to be to keep straight, because, you know, ultimately it's. It's using a lot of heat. I mean, that is a mini foundry going on inside of a DMLS machine. That is reaching 3, 4000F. So there's a lot of heat that's going on inside of that machine. So, every time you slice that model, there is volume to that slice across that broad surface that you just created or you're laying down. So if you imagine, if you just wanted to create a piece of sheet metal. That was a quarter inch thick. Sure, we can print that piece of quarter-inch thick sheet metal, but you don't want to print it flat. So you're slicing all that volume at once, because what happens is you're going to repeat that process. Layer by layer by layer, heat upon heat upon heat upon heat. And more… the more that happens, the more that is going to create internal stress. And, you know, deform that part. So you want to angle that part up on an… tip it up on an angle, the angle of repose. It'll be an acute angle, it might need to drive some support, but in this case, you are taking less volume per slice, you've got less heat per slice. And you will get much better result. If you have to. Stand it straight up, and, you know, ask us for advice on how we can help you with orientation to create a better part.

Same thing goes for the polymer side. There's no difference, it's still heat and polymers, it's just heat and metals.

Greg Paulsen: Yeah, absolutely. And something that I realized is, because you're building on a later-by-later basis, you know. When that… that first part of that part. I may have started. 20 hours before the last part of that part's made, right? So there's a lot of heat stuff that happens in between. That's where supporting strategy and orientation play such a big role.

You're also talking about, like, you know, parts bringing back. Parts can be in a stress condition, metal parts, you can run an annealing cycle. A lot of times, larger parts, we're gonna do that anyways. On the build plate… build plate to de-stress them in a fixed position before the supports are removed. Not as important on smaller parts, but for larger parts, yeah, they could dance on you.

And I have seen, you know, when we talk about. Printing bricks, and why you don't want to print bricks. Stress is so powerful. I've… I remember printing a tool block. For someone, this is back in, like, 2015 or something. And it ripped itself out in half. It literally… be stressed so much that it ripped apart… it ripped its own body in half. And what we ended up doing was actually coring and latticing it, and then it printed perfectly.

Matt Schmidt: Yeah, it's funny to say. You almost use injection molding principles even when it comes to metals.

Greg Paulsen: Absolutely, absolutely. Metal DMLS on a diet, way to go.

So, let's see. So, David, so David asked a few questions, but I actually see something about keying together, and I'm gonna focus on that for right now. When I'm… when I'm building keyways or slots, I think it's really important to understand those manufacturing standards. So, say I have a hole with a key in it, or a spline feature. We know typically in heat-based processes, those holes are going to shrink. So our design guides typically have a set offset, like offset by 4 thousandths of an inch, or 6 thousandths. In order to make that hole a little bit larger. Because we know it's gonna shrink, and that's usually gonna be a better fit than just printing it. As is. If you have multiple parts keying, sometimes it takes a little bit of tuning to work those together. Or you want to tolerance them in a forgiving manner to do that. Do not design flush to flush. Don't design to flush to flush in regular manufacturing either, but it's. There are ways to do that, but sometimes it does require a prototype of your prototype. Something really cool that you could do is slice your file and just print that feature, and see how that feature works, so you don't need to. Pay for the full print if you're just testing a small piece of that feature there, and that's something that's very easy to do with additive.

Matt Schmidt: Great, absolutely. And a quick tip, pro tip on that. Document everything you do. On a detailed drawing, document what material you used, what machine you used, what process. What the material, what part orientation, pack density. Document as much as you can to make your part repeatable and reliable in the future, so you don't have to reset and do it all over again.

Greg Paulsen: Absolutely. Alright, I got one, so you offer stereography, do you also produce ceramic parts? If so, what resins floor are you manufacture, and is it manufactured in-house or external suppliers? So, true ceramic printing, we have some discovery on that, but it is… it is very touchy because ceramic parts often will shrink about 25% or more from their net, and what I'm talking about is. True ceramics, like, going to be in very, very high heat conditions. But a bread-and-butter everyday use. Actually, I got. Stacked right behind me is a ceramic-filled resins for stereotography. So all these auto-quote in Xometry's site.

These are really good for, kind of. Test jigs, fixtures, if you're doing. Like, what's it called? A wind tunnel testing, for example. Don't expect flex, but expect very, very high stiffness, so I have. Accura Bluestone here, I have, Formlabs, what's this, 4K?

Matt Schmidt: Rigid 10K, yeah, this Formlabs Rigid 10K.

Greg Paulsen: And then I, you know, I have, SOMOS Perform. So we have different brands as a marketplace, we can bring competitors in one place, and you get to press buy with them. But, you know, if you hear. You got much more of a ceramic-y pick to them. So they're gonna have a lower heat. Deflection than a true ceramic, but still higher than most, than most polymers would have. For those needs. But it's a great option. That's not quite true ceramic, but true ceramics are a little bit tricky on a part-by-part basis. A lot of times, they require tuning and iterations to get. That part one may actually be part five. So this is a little bit tougher to launch.

Alright, let's see. So, Andres had a question about, components within the U.S, how much is typically manufactured in the U.S.? Honestly, we just started adding international options for the first time on additive manufacturing this year. So wholeheartedly, our manufacturing base is U.S. Domestic, and we've added some international options for SLS and FDM. Recently. But we have a very robust manufacturing base. In fact, actually, within the U.S.. We work a lot with, export-restricted items, so ITAR and EAR work. And that's just a… that's a checkbox on our website. So, we work very well with whatever your needs are. You get to see where your parts are being made, like, so you can choose… you explicitly are choosing domestic or international. We're never gonna surprise you with that. That's something… that's part of the power that we have. With our instant quoting engine. All right? And, let me see. Oh, I got a good one for you, Matt. So. For large parts, is it more cost-effective to print on a large format machine, or break the part in the sections?

Matt Schmidt: With mating features later, and then glue or sand or bond? Well, that would depend on a couple of, specifications or, you know, requirements, one being tolerances, of course, when you. Print in different pieces, you're going to have a different stack-up rate. You know, if parts are plus-minus 15 thousandths for each part. You're gonna have much more stack up than building, you know, a larger part. With such processes like FDM. FDM, you can take advantage of custom infill percentages. So with large parts, that's a huge cost savings when it comes to printing one large part versus smaller parts. You will have… less volume, material volume, going into that one part, so your cost would be less there. And then also, of course, you've got less labor into post-processing each individual part, rather than one large one. Again, we've got some large systems out there we can accommodate some big things. We can also do split and bond, you know, where, you know, just ABS with ABS cement works great. That's what it was made for. So, we can do a various array of assemblies, or, you know, individual parts. And large parts as well. So, we can help you draw a line in the sand and figure out what that cost. You know, competitive nature would be between each.

Greg Paulsen: Yeah, and something I recommend there is upload your full design. Upload the part as you hope to receive it. And what we can do is work through a strategy. We could even often split and bond on our side and develop the strategy, so you may not need to do that. And sometimes that's better for us to have that control, because we may be putting offsets and other things that we know is going to perform well for that process that you may not have that tribal knowledge about. So, always upload what you want, and this goes back to that previous question about shrinkage as well. Upload the design you expect to install. Don't try to offset those features or anything. And, you know, and then cross your fingers, because we're actually going to be measuring to your CAD, you know, for those features there.

What was I gonna say? I had one more, one more statement. So, split and bond. On that side, like, one of the things I had to build one time was a really large round piece. And it just happened to be outside of the Fortus FDM platform, so. Fortus goes 36 inches by 24 inches on the planar side, and it was beyond that. Instead of tilting up on the side, that may have affected my roundness. By tilting up on the side, so actually the roundness was more secure by doing a split and bond. So, sometimes just spinning the CAD will help us find the right solution. Because you're outside of the hypotheticals, and you're into, like, real application. So uploading that CAD is super important.

So, you guys were at time. Thank you so much for the questions. We're always available to answer any more questions, and you can reach out. We're gonna send an email follow-up. With this recording, a couple days from now. Please take a look at ADDITIVEWEBINAR10 for a great discount on your next project. And again, thanks so much, Matt, for joining me today, and thank you all for attending. I really appreciate it.

Matt Schmidt: Absolutely. Thanks, everybody.