3D Printed Houses: State of the Art

Concreef does not offer 3D printed houses and has not built one. This page exists because the question is asked constantly, and because the public conversation about printed housing is unusually detached from what is actually being constructed. What follows is a description of the state of the art internationally, and of the specific things that would have to be true before printed housing became a credible offer from a workshop like this one.

What can be printed

In built printed housing, what gets printed is the vertical wall envelope. The standard form is a double-skin cavity wall: two parallel extruded beads connected by ribs, with the cavity filled with insulation, with reinforcement and grout, or with a poured concrete core. Internal partitions are sometimes printed as well.

Everything else is conventional. Foundations are cast in the usual way. Floor slabs are cast or precast. Roofs are timber or steel. Windows, doors, lintels, insulation, waterproofing, electrical and plumbing services, screeds, plaster and all finishes are built by ordinary trades. The great majority of completed examples worldwide are single-storey, and two-storey printed buildings remain unusual.

Companies active in construction printing internationally, among them COBOD, WASP, PERI, ICON and XtreeE, are named here only as industry context. None of them is a Concreef partner, supplier of services to Concreef, or connected to Concreef in any way beyond the fact that Concreef's own machine was manufactured by WASP.

The realistic picture is that examples exist across Europe and North America, that they are few relative to the attention they receive, and that most of them are demonstrations, research projects or small pilot schemes rather than routine housing production.

Design considerations

Printed housing designs tend to look alike, and the reason is structural rather than stylistic. Curved and folded plan shapes are stiffer in the fresh state, print more reliably and avoid sharp corners where the nozzle has to decelerate. Straight orthogonal walls are printable but less forgiving.

Openings are a persistent design problem. Every window and door interrupts the continuous print path and creates a span that the printed material cannot make unaided, so lintels are precast, steel, or formed as a gradual corbelled arch. Arrangement of openings therefore affects buildability, not just elevation.

The wall cross-section carries several jobs at once. Skin thickness and rib spacing determine fresh stability and hardened capacity; the cavity has to accommodate insulation, and often reinforcement, and the ribs bridging the insulation are thermal weak points. Optimising the section for structure and for thermal performance pulls in opposite directions.

Services must be planned into cavities before printing. Chasing a thin printed skin afterwards removes material the structure was relying on.

And the whole design has to be coordinated with conventional trades who will build the floors, roof and finishes, which is an organisational problem as much as a technical one.

Materials and durability

Printable mortars for housing are fine-aggregate, binder-rich and admixture-controlled, designed to be pumpable, extrudable and stiff enough to stand immediately. That composition is a compromise. It delivers the process behaviour needed, but it also means a printed wall is not automatically a lower-carbon wall. The literature on digital fabrication with cement-based materials repeatedly makes this point: material savings from optimised geometry can be offset by the higher binder content per cubic metre, and the result depends on the mix and the comparison baseline. Claims of large carbon reductions from printed housing should be read with that in mind.

Durability in a building envelope is governed by the layer interfaces. Water tends to travel along them, which affects freeze-thaw performance and, where steel is present, corrosion risk. Cover to embedded steel is harder to guarantee in a printed element. In practice, printed housing walls are usually rendered, clad or otherwise protected on the exposed face rather than left bare.

Long-term performance data is thin, simply because there are few buildings and they are recent. That is not a criticism of the technology, it is a statement about the evidence available.

Process and lead time

The sequence for a printed house is: structural and architectural design, structural approval, foundations and slab cast conventionally, printer set up and calibrated on site with power and material supply arranged, walls printed in stages with curing between sections, reinforcement and cavity operations completed, openings and lintels placed, then the conventional build takes over for roof, services, insulation, windows and finishes.

The public narrative usually measures only the wall printing stage. That is the shortest stage. Design and permitting typically dominate a printed house programme, because the approval is bespoke. Foundations, roof, services and finishes take the same time they take in any house. Weather governs on-site printing, since printable mortars are temperature sensitive and rain during printing is a serious problem, so a temporary shelter is often required.

The honest summary is that the printing stage is fast and the project is not.

Limitations

Reinforcement is the first and largest unsolved problem. Extrusion printing cannot accommodate a conventional rebar cage. The alternatives, fibres, bars placed into cavities and grouted, horizontal bars laid between layers, a conventionally reinforced poured core, or post-tensioning, all work in specific cases but none has become standard design practice with code support behind it.

Standards are the second. There is no harmonised European product or design standard for extrusion-printed structural walls, so each project is permitted as a special case, with the cost, delay and risk that implies.

Seismic requirements are the third, and they matter directly in Bulgaria. Ductility comes from reinforcement, so the reinforcement gap becomes a seismic design gap.

Beyond those, the familiar material limitations apply: anisotropy, cold joints at interrupted layers, loose tolerances at interfaces with manufactured windows and doors, and a ridged surface that usually needs treatment. Machine capacity and single-phase power limit the scale of what a small workshop can attempt. And the economics are unproven at small volumes, where mobilisation, setup and bespoke engineering are spread over very few square metres.

Concreef context

To be explicit: Concreef has not built a house, does not sell printed houses, and does not have the validated mix, the structural design, the reinforcement strategy, the permitting route, the machine capacity or the site logistics that a house would require. The work done so far is a small series of test wall sections about a metre across, produced during 2026 in a rented workshop in Sofia on a Crane WASP machine with an LDM XXL twin-screw extruder, together with material trials on a printable premix and on an own mix being developed from local raw materials.

The interest in printed housing is real, and following the field closely is part of why the workshop exists. But describing printed housing as an available service would be false, and the gap between a metre-wide test wall and a permitted dwelling is measured in engineering programmes, not in enthusiasm. Enquiries about printed housing are answered with this position stated first.