One Man’s Journey To Build Portable Concrete 3D Printer Produces Its First Tiny House

[Alex Le Roux] want to 3D print houses.  Rather than all the trouble we go through now, the contractor would make a foundation, set-up the 3D printer, feed it concrete, and go to lunch.

It’s by no means the first concrete printer we’ve covered, but the progress he’s made is really interesting. It also doesn’t hurt that he’s claimed to make the first livable structure in the United States. We’re not qualified to verify that statement, maybe a reader can help out, but that’s pretty cool!

The printer is a very scaled gantry system. To avoid having an extremely heavy frame, the eventual design assumes that the concrete will be pumped up to the extruder; for now he is just shoveling it into a funnel as the printer needs it. The extruder appears to be auger based, pushing concrete out of a nozzle. The gantry contains the X and Z. It rides on rails pinned to the ground which function as the Y. This is a good solution that will jive well with most of the skills that construction workers already have.

Having a look inside the controls box we can see that it’s a RAMPS board with the step and direction outputs fed into larger stepper drivers, the laptop is even running pronterface. It seems like he is generating his STLs with Sketch-Up.

[Alex] is working on version three of his printer. He’s also looking for people who would like a small house printed. We assume it’s pretty hard to test the printer after you’ve filled your yard with tiny houses. If you’d like one get in touch with him via the email on his page. His next goal is to print a fully up to code house in Michigan. We’ll certainly be following [Alex]’s tumblr to see what kind of progress he makes next!

71 thoughts on “One Man’s Journey To Build Portable Concrete 3D Printer Produces Its First Tiny House

    1. the gantry. ever been clocked by a boat mast? Neither have I but a machine like this could probably do it twice in the same amount of time a boat would do it once. safety first!

    2. This is a fantastic first step, it is obvious to see how this is just a scaled down test. Shits people focus on the irrelevant parts of this post and not on the significance of this test. Great job, keep up the amazing work. And filter out the posts like this person who clearly just wanted to see their own text on a wall and see if they get a response.

    1. No setup time for the more time-consuming molds, can likely be operated by one person instead of a team, if properly set up will output exactly what’s in the file.

        1. If you don’t want to make the same thing every time, molds are not as quick. I would assume that if you are 3D printing houses, you won’t make them all the same because it’s easy to change them.

          1. Molds are just wood planks mounted together, it is easy to make different square shapes. And require no 3d design skills at all.
            Arcs, round corners, any kind of round XY feature is were the printer would shine.

      1. As a kid it was a daydream of mine to build a bricklaying house robot after seeing an automatic domino stacker for big domino tips.
        This cement extruder is like a plastic extruding printer, good for automated CAD one off copies, though the printer is less portable and the product less disposable.
        This generation is going to be crap like our current FDM printers vs even bad injection molding; but this is a first step and maybe a dead end on the road to smart cheap building structures. The question is if it is cheaper tp ‘print’ or flatpack structures when the energy required to make cement vs aluminum and/or plastic & foam is calculated.
        as for fusing/adhesion of layers it surely cant be worse than cement and brick though without rebar I would still be concerned on strength.

        1. It’s an interesting concept and we’ll see how it matures. Right now, the modern CAD-designed factory built house is pretty damn efficient. Computer/designer cut the model into component pieces which are preassembled on the factory floor (wire/plumbing/insulation/drywall). These are then shipped to the site and assembled by plan. Because they’re built on jigs, things are nice and square and fit together real nice like.

  1. Success in the experiment, yes. Success in proof of concept, yes. But feasible, no. Useable, no. Does not have correct rebar structure etc. would be best to build a wall extruder with rebar, vs a 3D printer approached. We have brick extruders , highway extruders, because they are better than 3D printer. Nice idea, wrong application.

        1. Michigan rarely get any earthquake and there hasn’t been any indication that Michigan could get earthquake strong enough to do more than crack china and rattle door. Also plenty of area in Michigan are hilly and one could aim for higher ground to avoid flood. Fire is the only hazard but concrete doesn’t burn and it’d take insane heat to damage concrete.

      1. Prove it.

        Concrete has a really low tensile strength – a few hundred pounds per square inch. It forms cracks too easily. Wood can be significantly higher (1000-2000 psi) if the load is parallel to the grain.

          1. My criticism here was just with the comment that this shed is stronger. It’s really just poured concrete – there’s no way it’s going to be that strong.

            I’m still curious about your “shown to improve strength” comment, considering straight in that link is “Generally fibers do not increase the flexural strength of concrete, and so cannot replace moment–resisting or structural steel reinforcement. Indeed, some fibers actually reduce the strength of concrete.”

            Plus, in addition, I don’t think fiber reinforcement would work well at all in an extrusion system, because the fibers wouldn’t cross the layer boundary. So cracks would just form at the layer boundaries.

          2. I’ve read an interesting article lately that speaks on the subject of steel reinforcement for concrete. I haven’t had time to verify the thesis of the article, so take it as second hand knowledge. It stated that the use of steel reinforcement in concrete dramatically shortens the total life of our structures. Whereas the aqueducts and colosseum have lasted nearly two thousand years the maximum age without intervention of a modern steel reinforced construction is only 100 years due to rust. I’m very interested in the applications of 3D printed concrete
            structures without reinforcement. Could we use modern high performance concretes with clever design to get thousand year structures? A lot of older methods like columns, arches, etc are difficult (and use more concrete), but perhaps with 3D printing we can at least remove the difficult part.

          3. The big problem with steel-reinforced concrete is “steel”, not “reinforced” — it mainly has to do with rust. If you could afford to use aluminum rebar, you could get the benefits without sacrificing longevity.

            As for 3d printing with reinforcement, it’s a tough problem. One option is putting reinforcement up first, then printing around it, but this requires clever nozzle design at least. Another easier approach is to use the 3d printer to print two “skins” of concrete at the inside and outside of a wall, then put reinforcing framework between them, and pour concrete in the gap. But the skins need to be tied together (just like forms in conventional work) to resist hydraulic pressure from the poured concrete, so that’s not easy either.

          4. Actually, I just googled to see if anyone had found an economical solution to the aluminum-concrete corrosion problem. Turns out Aluminum Bronze is the cool new rebar material, so maybe look into that.

  2. I was asked to provide the mechanical engineering for a portable concrete printer last year. I did my homework and called a few experts in concrete, building construction, and the kind of low cost relief housing the device was intended to provide for the developing world.

    I eventually declined the project because the people I talked to outlined some major problems:
    1. You tend to want structures to flex predictably with load or completely resist it. Cinder block walls with the right mortar can flex some over a long distance without cracking. Reinforced concrete walls that are thick enough can substantially prevent it. Printed walls offer the worst of both worlds, where they’re likely to delaminate if thin, and prohibitively expensive to print thick.

    2. The promise of a 3d concrete printer is often low cost housing, but what is meant by low cost? To print consistently enough to not collapse over the course of a few years of weather changes, the mixed concrete would have to have a very predictable viscosity. This means a precise mix, fed very accurately through the extruder, at a very well known rate so as to not completely jam the machine as it sets. This means the operators would have to be pretty skilled and the concrete a particular mix made by a factory with the necessary resources, which likely rules out anywhere close to the developing nation this particular project was pitched for.

    You can make cinder blocks and bricks out of almost nothing and arrange them into a very stable structure without much training. With a little maintenance, those structures can last a decade and more in even the harshest conditions. These printers would need to be manufactured from the ground up, people trained to use them, shipping to the location, a plot prepared, operators overseeing them for the few days it would take to print any structure worth having, and even then the structure needs additional operations (like a roof, reinforcement, flooring, insulation, etc) to be livable. These costs stack up fast.

    3. The crux comes down to who benefits from a concrete printer. Starvation is a problem across the planet, but opening up a new cookie factory or producing a machine that prints pancakes doesn’t solve the fundamental problem. The resources to make housing are everywhere. The problem is often that the infrastructure where housing is inadequate is totally messed up. Sometimes all the skilled workers have been wiped out generation after generation with starvation or disease until no one has the knowhow to build lasting structures. Sometimes it’s that a population boom has happened so fast that there’s not enough housing to fit everyone in desirable areas, but people are too impoverished to move away from where there is subsistence work available (I experienced this situation while working on a city planning project for an overcrowded area on Bangkok). Sometimes it’s that the weather is so inclement there is a perpetual cycle of people being displaced by flooding or typhoons.

    Homelessness isn’t often due to a lack of houses, but a complex web of factors. There is likely more housing available in Manhattan than is needed to provide a home for every homeless person on the island ( ).

    This interview ( ) with UNICEF’s Chief of Water on the failed Play Pump project is really informative. It sheds light on how it takes a deep understanding of the problems at play when people are facing a dire situation to come up with a solution that addresses the root cause.

    1. I’m not an expert, and I can understand why you passed on the project because posting a solution to issues online is much easier than actually implementing a solution. But since I’m online, and not an expert here are some thoughts:

      1) dual extruder, one for concrete, one for mortar, print concrete bricks right on top of layers of mortar.
      2) Mix the concrete on site with another automated machine that can manage such a thing… that in and of itself is likely a Really involved project with it’s own list of complications.
      2.5) (A plot needs to be prepared) interchangeable heads on the printer to allow it to work as a backhoe to prepare/level the site.
      3) “The resources to make housing are everywhere” Skip the concrete use clay/dirt/mud and other natural sources.

      All and all 3d printed houses are not likely for the developing world… and there is the (completely solvable) challenge of plumbing and wiring concrete houses in the first world where such amenities are basically required.

      Where I see the technology to be used most heavily is in subdivision creation in the US and other developed countries… places where labor costs are extremely high and time to complete the project is usually pressing.
      Places where I can setup an automated backhoe to prepare a site, move the backhoe to the next site and setup a printer/crane/gantry to start working on the first site…
      Also places where people are particular about houses not being identical, but instead being built bespoke.

      New gated community developments might make use of tools like this… where all houses are the same material, but the layout might be different, where columns, arches, and curved walls are stylistic choices that the owners make, but the price point means they need to be built without the aid of the actual skilled craftsmen of yore.

      I’m sure there are challenges to overcome, but none of them seem to be insurmountable at this point.

  3. would be cool to fit the whole apparatus on a 40 ft trailer. just park the thing next to the construction site, hit a button and the thing unfolds like a carnival ride possibly out to about 30 ft (giving something like a 25×35′ build area, height likely limited to overpass clearances, but maybe it can expand upwards). it would require a bare minimum of manual assembly and calibration, and then building can start immediately. of course you still have to prep the build site, and thats not quite a push button job.

  4. Okay he produced a crude outhouse with no windows, vents, flat roof, etc. No rebar to keep up structural strength.

    Whoop de doo.

    It’s something a kid would design.

    A person could do the same with hardibacker board, and wood frame and have something better.

      1. That is a rather silly argument. So build on top of your argument, let’s build a house out of scotch tape.
        Now you can prove that is is a bad idea, only if you build a yet another house out of scotch tape?

  5. While a gantry system could have its advantages, I’d like to see some 4D-printing robots that draw inspiration from mound-building termites, wasps, and other eusocial arthropods.

  6. I’m not sure of the economics of this.
    I just recently witnessed a regular 160m2 family house build.
    It took 4 weeks to build the foundation, and only 2 days to build the walls (2.8 meter high).

    At the first glance, it is not the walls, which needs to be optimised out….

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.