Interview: Fosters’ Xavier de Kestelier – Unlocking 3D printing’s potential

Xavier de Kestelier, partner at Foster + Partners’ Specialist Modelling Group, on working with Skanska to develop a commercial 3D printing technology and winning second prize in NASA’s competition to print a settlement on Mars.

We’ve all heard the buzz around 3D printing, but is it all just hype or are commercial applications for construction projects just around the corner?

In the past few years 3D printing has been used extensively for creating architectural models, but it is still very early days in terms of creating construction components.

According to the Gartner Hype Cycle for 3D printing (pictured above), which gauges market hype and maturity for different technologies, industrial 3D printing is close to the “Peak of inflated Expectations”, ie early publicity has produced a number of success stories, but also many failures. Gartner forecasts that it will take five to 10 years for it to reach the “Plateau of Productivity”, when mainstream adoption starts to take off and market applicability and relevance are clearly paying off.

I believe it is likely to follow a similar path to 3D printing for prototyping, which was introduced about 10 years ago, when Foster + Partners bought its first 3D printer, and has now reached the Plateau of Productivity.

What are the obstacles to 3D printing at a macro scale?

Scaling up the technology is tricky, volumetric calculations show that if you double the size of something it takes around eight times as long to print and costs eight times as much. 3D printing is already expensive, so scaling it up is no trivial matter and for the time being construction applications are likely to focus on small-scale components.

If 3D printing is not going to be a fad, you really have to look at the real advantages it offers over alternative technologies, in terms of making construction cheaper, or better quality. Construction can learn a lot from other industries, like medical science, where 3D printing has been used to create quite small parts, like dental and surgical implants, each one unique and designed to fit the specific body of an individual.

In building terms, that might mean 3D printing smaller unique brackets or node connections, such as those produced by Arup for a research project in the Netherlands. [Where selective laser sintering was used to produce unique metal nodes for a proposed street lighting scheme].

One major advantage is that printing complex curving geometry costs exactly the same as printing linear geometry, so any geometric element of a design essentially comes for free.

A few years ago we ran some 3D printing tests in concrete with the School of Civil and Building engineering at Loughborough University, looking at how a concrete wall could be designed differently when there is no extra cost in making it complex. We looked at adding acoustic treatments to the surface of the wall and printing in cavities to run services through.

At the end of 2014, Fosters signed a collaboration agreement with Loughborough, main contractor Skanska, robotics firm ABB, materials manufacturer Tarmac, and others to develop IMCRC’s lab-based concrete 3D printing technology to a commercially viable scale.

What progress has been made on this?

We are in the process of finalising the details, there are lots of aspects to work out, such as the completely new process of translating a digital model into concrete printing.

The consortium is interesting because it is an entire vertically integrated supply chain working together on a new process, which almost never happens. It means that all processes are linked up and when we design we are not just defining an object’s volume and geometry, we are also helping design the 3D printer tool part, the specific functionality of the robot and the G Codes that will run it.

In contrast, traditional construction makes it difficult to know, at the start of the design process, what the manufacturing possibilities are, or what a machine really can or cannot do. That aspect is left to the main contractor when it asks a fabricator to propose a solution.

What other 3D printing research is showing strong potential?

Massachusetts Institute of Technology’s Medi Lab has built an additive manufacturing machine able to 3D print transparent glass, which the team believes could be used to create architectural components or even entire building facades. It highlights the potential that exists for printing with a whole range of different materials.

China has famously pioneered the 3D printing of entire buildings. Are they ahead of the game?

The techniques being used might be cheaper and faster, but design-wise they are not really exploring new ground and printing entire sections of wall is really just emulating current construction techniques and not exploiting the real potential 3D printing has to offer.

Foster + Partners came second in NASA’s 3D Printed Habitat Challenge to design a settlement on Mars constructed by robots. When might a scheme like this be built in reality?

That’s hard to say, our proposal took advantage of a new printing technology that has been proven feasible in the lab, which involves microwaving batches of dust together to create dome structures, which on Mars would be designed to protect astronauts from gamma radiation.

The 3D Printed Habitat Challenge has three phases, the first looked at the overall concept of printing a settlement, the second will involve constructing the robotic system needed to build the structures and the third will involve printing examples of the habitat. We’re currently considering whether to enter phase two.

How intrinsic to construction will 3D printing be in 20 to 30 years?

It will be very intrinsic, on the one hand I see it as being used for high level, expensive components with complex geometries and integrated services, on the other, I can imagine a small contractor carrying a 3D printer in the back of his van to create brackets and connections on the job site.

When extending my house recently, I worked with my contractor to design and 3D print a set of brackets to raise my bath above the floor, because brackets were not included with the bath. The process took just one day to design, print and install, and the contractor was amazed at the potential.

That said, we are quite far away from a scenario where every part of a building is 3D printed, including the range of materials needed for structure, insulation etc. Even in more advanced manufacturing sectors, like F1 car design, no one tries to print everything. Ultimately, 3D printing is just one manufacturing technique among many, it is good at some things, not so good at others, and certainly not the solution to all the world’s problems.

MIT’s Medi Lab has built an additive manufacturing machine able to 3D print transparent glass, which the team believes could be used to create architectural components or even entire building facades. It highlights the potential that exists for printing with a whole range of different materials.– Xavier de Kestelier, Foster + Partners

Story for BIM+? Get in touch via email: [email protected]


  1. Ok then John print me a full size building and I’ll believe you, talk about a load of nonsense.

Comments are closed.

Latest articles in Technology