The race is on in the architectural world to 3-D print houses and other buildings. It seems like every few weeks, there’s another “first” announced, although most of these efforts seem fairly similar: they are scaled-up versions of the desktop 3-D printers that extrude a thin bead of molten plastic and layer up a complex shape, only in the scaled up version, the hot plastic is replaced by wet concrete. The main boundary being pushed is how to use 3-D printing more, and faster, to make more of the house.
By contrast, Branch Technology, of Chattanooga, TN, is trying to use 3-D printing less. The type of additive manufacture they pursue doesn’t even look like desktop 3-D printing. Rather than layer up continuously from a flat surface, Branch is one of a handful of systems building in 3-D space using a robot arm, creating a minimal matrix of carbon fiber composite that defines a wall’s shape and structural properties.
The company was founded by architect Platt Boyd, who was inspired, among other things, by a 3-D doodling pen, a handheld device that extrudes a thin string of a stiff thermoplastic, and allows the user to “doodle” in 3-D space. Boyd and his team developed a hardware and software method for building 3-dimensional structures of composite materials using a robot arm to squeeze out a matrix, line-by-line, that is largely open space.
Boyd calls his method Cellular Fabrication of C-Fab. The Branch website explains that this is building nature’s way, but it doesn’t explain what aspect of nature it mimics. The structural frames they are creating look like they might be relating to the branching patterns of tree-limbs, and the company’s name is “Branch Technology” so one is tempted to assume that it’s mimicking tree structure. That assumption is wrong.
So we asked Platt Boyd. “It’s cellular structure.” Many living things, he explains, are composed of cells where none of the materials have structural stiffness, but instead have a boundary-defining material (a membrane) which, when filled with all the stuff that makes up the cell, combine to form a robust, self-supporting structure.
“You see how natural materials come together on a cellular basis, in the cellular structure, the material that fills the cell provides the strength. In our muscles, it’s water pressure, blood pressure. Using an economical material like water or other readily available material within the bounds of a higher strength material. You’re using geometry to maximize strength. You minimize material use, but the form is free to become almost anything. You see these things in the natural world, but in construction, we haven’t used those materials in those ways. It opens up possibilities beyond the capabilities of conventional construction. We can begin to make things more like the beauty you see in the natural world, free-form, not rectilinear, more organic in shape. That’s really what spawned all of this, observing how structure is made in the natural world.”
Branch’s system extrudes a matrix that describes a multi-cell object, and then ‘fleshes it out’ with cheaper, fast-applied materials. “The bounds of that cellular structure are normal, economical construction materials. Foam, then concrete, then gypsum or some other surface material.”
But it’s the free-form nature of this capability that really sparks the imagination. Branch’s demonstration-structures are willowy fantasias of compound curves, the kind of thing that would be either impossible or prohibitively expensive to construct by, say, conventional concrete forming.
The idea is to use 3-D printing minimally, purely to create an openwork structure that defines the shape of the wall, and then use conventional building materials such as foam and concrete, to complete the wall’s structure and finish. This approach implicitly acknowledges the major weakness of current 3-D printing: it’s slow, and slow is expensive. Moreover, the more exotic materials like carbon fiber are very expensive in and of themselves. Branch’s strategy is to 3-D print the outlines, and use inexpensive, traditional materials to complete un-traditional shapes. “One of the advantages to composite technology is that you can finish it in any number of ways. I think that’s one of the major things that differentiates us from other 3-D printing processes. Our 3-D printing technology gives us the ability to access forms and geometry that traditional building tech might have had difficulty doing, or be very expensive to get to that place. We’re really trying to minimize the amount of 3-D printing we need to create these forms.”
The engineering of the composite structure is performed by Branch’s software, which optimizes the structure. Execution is perfomed by a robot arm with Branch’s extrusion head mounted on it. They started with a tabletop robot arm, a Kuka Agilus provided to them by Kuka in the early stages of the company’s development. That arm, with a 3/6 foot reach, could build a wall 4’h x 4’w x 3.6’d. They have since graduated to a track-mounted Kuka industrial robot, a KR 90, mounted on a 10-m long track that can build walls up to 25 feet wide by 58 feet long,
Their current objective is interior partitions, but they are also testing methods of making load-bearing walls. Even the bare composite framework is surprisingly robust. Boyd load-tested an early prototype made of unreinforced plastic and found it able to support many times its own weight. The addition of fiber reinforcement and foam fill made it even stronger. “One of the things that’s exciting is the strengths that are available. Just plastic and spray foam was stronger than wood construction when compression tested.”
Adding a layer of spray-applied concrete up their game by a quantum leap, into the realm of load-bearing walls. “With the matrix fully embedded in concrete, it performs like 3000 PSI concrete,” says Boyd. “We have further testing to see if the 3-D printed matrix can act in lieu of traditional reinforcement.” Boyd tested this version by standing one wheel of his pickup truck on it.
Branch’s current composition uses ABS polymer with short fibers – carbon fiber, glass fiber, or others – in a random orientation. It is unlike shell-type composite structures that gain strength by engineering the orientation of the fibers or the fabric-weave, and more like traditional trusswork that has been set free of the constriction of rectilinear elements.
“We consider ourselves a construction company,” explains Boyd. “We look at 3-D printing as our enabling technology. The customers we’re trying to satisfy are in the construction industry. 3-D printing is very nascent in its understanding of materials. We can easily look to other industries in terms of materials to understand the properties. The reason why 3-D printing is the cornerstone of this type of fabrication is the realization that with 3-D printing, complexity doesn’t come at any significant penalty compared to a simple structure. In that way, we’re replicating nature: it’s almost as easy to be a complex and hierarchical structure as well as very homogeneous and uniform. That complexity is unlocked.”
“We are currently focused on doing interior work. Basically now, we’re bordering on art, but our long-term goals is building robust systems.” They continue testing, seeking to meet construction standards and building codes for load-bearing structures. They are working with resin manufacturers to find a binder with thermal expansion similar to that of concrete.
The company does not want to keep their technology in-house-either: they are eager to attract architects who want break free and use their system to build the next generation of innovative design. To that end, they’re sponsoring a contest to design a house, with a prize of $10,000 and all the 3-D printed walls. Branch is currently working on an 18-foot-tall project for the Atlanta Museum of Design, working with Pete Cason (designer of Pentagon 9/11 memorial).
All images courtesy of Branch Technology