Transcription
3D printing an entire house: computer files automatically translated into physical objects. Even full-scale houses out of concrete. Making buildings like this promises to convert digital designs easily and faithfully into real physical structures in the world. This sounds like a dream, except there's quite a few challenges that are getting in the way, and frankly, the buildings made like this, they look a little weird.
But companies across the world, they're trying to solve these problems by racing to make the process scalable, affordable, sustainable, and palatable to people that would just prefer living in a more traditional-looking home. Your next guest launched a 3D printed home business to fight the affordability crisis. His company, all Quist 3D, completed the first ever owner-occupied printed home in Williamsburg, Virginia, in 2021, and their success could shake up everything, changing how we design and make buildings forever.
The pace of innovation in the construction industry is unbearably slow. When comparing it to other industries like automobiles or digital technologies like, like phones or computers, for instance, this slow pace it's usually blamed on the complexity of making buildings where every site is unique. And then couple that with how the trades and building product manufacturers work together, or don't, the building construction industry it just cannot be vertically integrated like those more agile industries. Cars and phones, they sell thousands or millions of units, which make unique technologically advanced manufacturing solutions much more plausible. These roadblocks have stymied countless promising innovations over the years. Elon Musk disrupted the auto industry, even space flight, but his ventures into construction, they've repeatedly failed. But 3D printing houses promises to cut across all these challenges.
It works by using a large computer-controlled gantry, or robot machine. Precision motors control movement and the X, Y, and Z coordinates while a deposits a special concrete mixture. It lays down a series of layers, one on top of the other, over and over until the building is complete. And in some ways, this is similar to how a brick house might be made, but instead of a human bricklayer performing the repetitive task of placing each brick module, this robot just repeatedly adds layers of a much finer mixture. Bricks and grout are substituted for cement and aggregate. I use this analogy between bricks and 3D printing concrete, and I'll keep coming back to it because bricks have certain limitations. For instance, you can only stack bricks on top of themselves or on top of other things. If there's no brick underneath to support the next one, it just falls over. So you can't make openings in walls using just bricks; you have to use a lentil, usually made of steel or concrete, to be able to span over the void and then transfer the loads down to the ground.
3D printing has similar limitations. You can only pile wet concrete on top of other wet concrete or something else. And just like the brick house, 3D printing openings is a challenge. Maybe you have to place something to hold the layers that would otherwise just be floating in the air. Then, as the layers build up and they cure over time, the whole thing could be strong enough to span over the opening, or you could just leave the steel member in there as a lentil.
Well, those limitations, they're not a big deal. There's something that bricks can do that 3D printing buildings cannot do, and that's make a sharp 90° outside corner. It's hard to stop and start the flow of the concrete cleanly; it always leaves this mark. So to keep the flow continuous, you have to round the outside corners and hide the times when the flow stops, and you hide them inside of the thickness of the wall. That's where no one will see it, and that's why all the examples of the 3D printed houses have those flowing outside curvy shapes. That's only one way that this new construction technology will change the design and the way that we configure our buildings, but to understand all the other ways, we need to look at the various stages of the design and construction process that this offers.
It begins in the computer where the architect lays out the rooms at the broadest of scales. This can be pretty similar to a traditional house using the usual relationships between the bedrooms, living rooms, and kitchens, but the details are much different. And even at this early stage, the designer needs to be able to understand the limitations of the machines and the materials of construction to make sure that the forms that are being modeled in the computer can actually be translated.
Once the design is complete, the file passes through a program that translates all of the surfaces of the digital model into a series of instructions that the printer can follow. This includes when, where, and how fast everything should be moving. This can be used to drive simulators that will detect any potential problems and then also offer estimates on how long the project will take. If that all checks out, we can start setting everything up on site. The 3D printer is brought to the site in pieces on a truck. The whole thing is modular so that no piece is larger than one will easily fit on a truck bed. Once everything is there, it can take about a day or so to set up the machinery. The gantry starts with the foundation blocks that need to be as level as possible, and from there everything is erected, shimmed, and then bolted into place. It's critical that everything is perfectly square and level, or the machines they could jam, or it might deposit materials in ways that disrupts the final product and just doesn't look good.
This is an important theme, I think, for assessing the overall promise and the limitations of 3D printed construction, and it has to do with the idea of tolerance. Different parts of buildings operate within a variety of tolerance levels; that's the word that's used to describe the acceptable variability of dimensions for a building component. For instance, it's totally acceptable for a brick wall to be anywhere within 20 mm from where it's drawn in the architect's drawings. It's okay for the wall to be 20 mm too far this way or this way; it's just the nature of how we build things in an imperfect world. There's a discrepancy between where the perfect place is for it and where it can actually be. And when humans are piecing the building together over time and in parts, it's relatively easy to compensate within the construction process and mask those imperfections with trim or other materials that are easier to work with or have finer tolerances. But computers and robots don't yet operate with the same ranges of acceptable tolerance; it's more difficult for them to just adapt to the imperfections of the real world. In concrete, the preferred material of 3D printing, requires a lot of tolerance for standard construction practices. A concrete wall can be within 2.5 mm of where it's supposed to be and still be considered to be in the right place. For a computer, without this tolerance calculated in, 2.5 mm can spell disaster.
Despite that challenge, one of the payoffs of getting this whole process right is that it only takes about three or four people around 24 or 48 hours to complete the printing of an entire house. Contrast this with bricklaying, for instance, which might take, I don't know, a couple of weeks for a human to place 8 to 10,000 bricks in a typical house. And like bricks, the concrete shell is only one component, though, of many within the complete building. You have to supplement this infrastructure with other things like window units and millwork and built-ins, plumbing and electrical, and things like that. This is again where the challenge of tolerance comes in. The rest of the building industry is tailored to a certain way of constructing things that leaves smooth, straight, and level locations for things like cabinets and windows to go in. Trying to fit in a perfectly extruded aluminum window into a lumpy concrete opening leaves a pretty significant gap between the two. This needs to be filled in with something, usually, which isn't the most durable or clean-looking solution. So detailing how 3D printed concrete meets these other elements can be a real challenge.
But if we go back to the printing itself, everything comes down to this concrete being extruded through a nozzle. There's a lot of chemistry and physics that is at play here that ultimately determine whether the final house will be a lumpy mess or graced with smooth, elegant surfaces. You have the concrete itself, which is actually a very special blend of it. Critical factors such as how well it flows through the system and how supportive it is are dependent on the precise mixture of just a few ingredients. These are the cement, or the binder; the aggregate, or the stuff that you're binding; and any additional ingredients that are usually called admixtures. And these impact things like drying time or the ductility of the material. All of these ingredients are for trade-offs with ranges within which their use becomes an asset or a liability. For instance, larger aggregate material can be printed faster, and the larger, coarser rocks actually ensure that there's better bonding between the printed layers. This is actually the most important factor for determining the overall strength of the house in the end, but if the aggregate is too large, it'll clog up the nozzle. So you want to get the sizes just right: large enough without being too large. Admixtures help with balancing things like buildability, workability, and extrudability. Fly ash is actually the main admixture for high-performance 3D printing concrete. It allows the concrete to remain in the sweet spot of workability for a longer period, and it actually makes the overall thing more durable in the end. However, large amounts of fly ash can lead to slower development of strength and buildability, which is why it's often mixed with other things like clay to retain shape stability while it all sets.
This concrete passes through a head, and it also has a significant impact on how things will turn out in the end. Some heads are round, while others are more rectangular in shape. Sometimes they have a little rake to spread things out, but this can create a buildup at the edges, which leads to a rougher texture in the end. For the rectangular nozzle, the head itself needs to rotate so that the broad side of the head stays perpendicular to the vector of motion. With most 3D printed houses today, the concrete is left exposed on both the outside and the inside. The concrete is also the structural material, helping to hold everything up. And in typical construction, these tasks are done with very different materials that are tailored to perform well for their specific application. Drywall, for instance, makes a great interior surface, but you don't put it on the outside, and you certainly don't ask it to help hold up the whole building. But with 3D printed concrete, the extruded layered walls have to do it all, and there are obvious compromises here. For instance, if anything gets a little off in the construction process on site, you're forever haunted by a memorial of this imperfection forever.
Not everything needs to be printed all at once in a single pass, though. You can break up the house into different parts, and some parts can be pre-built off site or just in a different orientation that will end up in the final house. You can make it upside down, for instance, then flip it over. These pieces can be hoisted into place when it's time for them in the buildup of the overall structure. So this is how you make things like cantilevers, for instance, where there's nothing to support it underneath, or maybe certain parts of the building, they're not even 3D printed at all, maybe like the porch or, or a pitched roof, which wouldn't make any sense to print.
That said, with 3D printed houses, you usually get an inner layer and an outer layer of extrusion. These are tied together with more layers of concrete that are printed back and forth, or they use these metal ties that help string them together. So this is a small-scale example of how the walls can be built. Uh, we, we do the exterior beads, and we come back and we do the infill of the structure, and all of this aids in the overall strength of the wall and it offers space for insulation in between. For reference, this too is similar to the way that a brick wall might be made. Inside the cavity is filled with polystyrene foam that fills in the gaps and it insulates the whole assembly. Water could also get into this gap at some point, so the bottom is actually sloped with holes cut in the outside layer to be able to manage for drainage. So there's all of these design limitations, which to me seem pretty manageable, but I think the greatest challenge with 3D printing technology, at least as it is right now, has to do with the relative inflexibility of the building after construction. If anything happens to the 3D printed part of the structure, it's almost impossible to be able to repair it back to its original state. What are you going to do? Bring out that giant machine? Also, the machine needs direct access from above to do its job; you're not going to take off the roof to be able to fix it. Additions or renovations are all the same thing.
Another limitation that I see right now is that the companies that are using this technology are attempting to become more vertically integrated, more so than a typical construction companies they might have in-house architects; they manage the machines, and they oversee the entire construction process. This makes for a really closed system. An architect can't just dabble with 3D printed buildings, for instance, so the barrier to entry is just really high. And what if the company that made your house goes out of business? You'll never be able to make repairs or additions that fit in with the original. Buildings are made with modules and layers for a reason. Each layer is tailored to the task that it needs to perform, and it works in concert with the other layers of the assembly. The modularity ensures that any part can just be replaced when needed and not leave any visible scars. Thinking of a building as one large seamless object is completely antithetical to the realities that buildings face. Components of buildings could very easily be 3D printed and likely are these days, but at that scale, the scale of a component, it's often more economical to fabricate it in other ways. It's easy to create a reusable mold to cast concrete into, for instance. It doesn't make sense to 3D print a building part that could just be made more easily in another way. And in this sense, I find it curious that 3D printing in other industries is usually a method for making prototypes of things; it's not more efficient for large-scale manufacturing. But with buildings, because each one is more like a prototype, the promise of 3D printing is more about scaling up and reproducibility.
This modular approach to fabrication is also true for other things like razors, but most companies, they're not innovating fabrication methods to provide a better shave. Take cartridge blades, for instance; you find them in almost every grocery store, but this concoction is really just devised to make more money for the company. They charge up to $2 per cartridge and then convince you that it's special because it has five blades and a lubricating strip. The next thing you know, you're paying hundreds of dollars a year for something that you don't really need. The truth is that one blade is plenty if you use it correctly.
One company that's doing it right is today's sponsor, Henson Shaving. Henson turns the traditional business model upside down. Instead of designing a cheap old handle and charging a fortune for blades, they developed a precision-engineered razor that supports the entire width of the single blade at a depth of only 27 microns. This means that there's no way for the blade to flex or bend, and these are the things that cause skin irritation. Henson's expertise comes from making parts for satellites and space probes; they've engineered stuff that's on Mars right now. So when I say that 30° is the perfect angle for a blade to shave the skin, you can trust them. The razor not only provides a better shave than cartridge blades, but it's also way cheaper. Each blade costs only 10 cents. You can go through 20 blades for the price of one cartridge blade. And here's the best part: if you use promo code Stuart Hicks at checkout, you'll get a package of 100 blades for free. Just go to Hensonshaving.tocom, pick out the razor that you like, add the 100 pack of blades to your order, and when you enter the promo code, the price of the blades will be deducted. You probably won't have to spend another dime on shaving for the rest of the entire year. I absolutely love my razor, a high-quality, well-made personal item, so treat yourself or get one as a gift for someone that also appreciates quality design. Once again, the tens and shaving.tocom, get 100 blades for free with promo code Stuart Hicks. Links are down below. Enjoy, and as always, thanks for watching.