3D printing is emerging as a popular technology in the space industry, where it increases the potential to reduce spacecraft mass and cut launch costs. But will it ever be possible to print the rockets themselves?
General techniques of additive manufacturing (what we now call 3D printing) have become common knowledge in the engineering community. But to most of us it means watching a desk-sized printer slowly extruding plastic to form a model toy, key-fob or similar inconsequential item. The 'resolution' of the device tends to be poor, leading to visible stepping in curved surfaces.
Despite this impression, additive manufacturing is fast becoming an indispensable component of the modern-day industrial process, especially for niche applications such as space. In 2014, the International Space Station received its first experimental 3D printer, which immediately earned its keep by printing a spare part for itself. Later, after an astronaut mentioned he could do with a socket wrench, a design was quickly developed, transmitted to the station and produced in orbit.
While you could argue that the astronaut's plastic wrench was closer to a Christmas cracker toy than something an earthbound car mechanic would find in the toolbox, space hardware that owes its existence to additive manufacturing is currently speeding its way to Jupiter, the largest planet in the solar system. The material of choice is not plastic, but that stalwart of aerospace: titanium.
The company responsible for the early adoption of the technology is aerospace giant Lockheed Martin which, far from being in it for the novelty, is "streamlining satellite production with 3D titanium printing to lower cycle times and reduce cost". According to the president of Lockheed Martin Space Systems, Rick Ambrose, 3D printing can reduce production times from months to days: "it could take a year to build a propulsion system. But printing can reduce this to days or weeks".
The way the process works is explained in a Lockheed graphic entitled '3D Printing 101' (the number has nothing to do with George Orwell's famous room, but refers to introductory courses in further education), which describes it in five easy steps. 1: load design data to the printer controller. 2: load titanium powder into the printer and evacuate the chamber. 3: wipe the powder across a platform and melt with an electron beam. 4 'lower the platform and repeat until the object is formed. And 5: cool the part and return excess powder to the hopper.
Titanium, typically alloyed with other metals, is widely used in the aerospace industry, largely because of its relatively low density, which is key to saving weight. Typical space applications include brackets, thrust frames and other structural components. But 3D printers can use a variety of materials, which according to Lockheed Martin's senior vice president & chief technology officer, Ray Johnson, is "a big part of the benefit of this new way of production".
The company currently uses 3D printers in two main ways: for prototyping parts, using polymers, and for printing "flight-ready parts" in titanium, aluminium and Inconel. (Inconel is a trademarked nickel-chromium 'superalloy' well suited to the high-pressure and high-temperature environments of space exploration.)
Lockheed Martin Space Systems (LMSS) has produced more than 300 parts through additive manufacturing, Johnson says, "most of them used for prototyping during the development phase". Sometimes, he explains, the parts are used as a substitute for a flight component, to accelerate the build, "while we wait for the actual item to be delivered"; alternatively, they help to demonstrate new concepts as part of a transition from more traditional manufacturing methods.
Actual flight-model components built by LMSS include an Inconel pressure vent, used in the flight test of Nasa's new Orion capsule on 5 December 2014, and a set of eight structural brackets on the Juno spacecraft, which is currently en route to Jupiter.
These may seem like minor parts, but Johnson confirms that the size of space-qualified parts is growing "from brackets the size of playing cards to propellant tanks printed by room-sized machines". It is this increase in component size that brings "significant savings", he adds. For example, satellite propellant tanks could take up to 21 months to produce with today's manufacturing methods, such as titanium forging, but additive manufacturing technology can reduce this "to less than six months, and with a cost saving of over 50 per cent".
In fact, titanium tanks are the most complicated items the company has produced so far. "It may sound simple - empty fuel canisters - but it's not," says Johnson. "These tanks operate in space under incredible internal pressure, and also need to survive high-g loading and other environmental stressors related to launch."
Thinking in 3D
While the evolutionary improvement of industrial processes to save time and money is embedded in engineering culture, so too is the desire to find different and better ways to do things. Today, we know them as disruptive technologies or disruptive innovations.
In addition to the evolutionary developments, LMSS is marrying precision modelling and simulation with additive manufacturing to create complex parts that are impossible to manufacture traditionally. According to Johnson, "features like durable lattice work, intricate textures and organic shapes are all possible".
He uses an antenna-reflector fitting as an example. Based on customary design and manufacturing processes, the part was evolved by gradually stripping away material to save weight, reducing the mass from 395g to around 80g in the process. However, by designing from scratch for additive manufacturing, the resultant part weighed only 40g. "It looks less like the original part", he says, "but we don't care about the shape. It's the functionality that's important".
It also reinvigorates the design process itself, because engineers can concentrate more on the functionality of a part than its shape. To encourage this, LMSS deploys its design engineers to the factory floor to work side-by-side with manufacturing engineers and "learn what additive manufacturing is truly capable of", says Johnson. "This hands-on approach stimulates creativity and encourages our design engineers to think differently about their designs".
Dennis Little, vice president of production at LMSS, agrees. "Our experience has been that engineers depend heavily on the left side of their brains, the hemisphere that favours the logical, sequential and analytical. 3D models and designs engage the right side, the hemisphere responsible for more creative and holistic thinking. When our engineers engage both their left and right brain, we are realising geometrically complex designs, features and parts never seen before."
Johnson thinks additive manufacturing is "opening new doors where design engineers innately think in three dimensions" and boasts that the company's engineers can now print "nearly any component they conceive".
The technology is a "critical component of'our end-to-end digital manufacturing environment", says Johnson, an environment the company calls its Digital Tapestry, since it "weaves together the entire product development lifecycle, from conceptualisation to product realisation". The Digital Tapestry is designed to provide a "seamless digital environment" to keep the digital data "intact" throughout the process.
To make this work, not just for LMSS but across each of Lockheed Martin's business areas, the company has established a corporate Production Council composed of leading individuals from each of the areas, which range from defence to information systems. A potential challenge with any disruptive innovation is the transition from laboratory to production and thereafter to the marketplace. As new technologies arise, explains Johnson, qualification can be difficult and time-consuming. "We are addressing this by ensuring that all of our business areas have access to the same quality systems and machines".
The national organisation tasked with encouraging the adoption of additive manufacturing is America Makes. Its director Ed Morris puts the case succinctly: "3D printing is a game changer because it has a whole new set of rules. When you change the rules, you change the game". By way of explanation, he adds that since 3D printing uses only the material required for the finished part, it "radically reduces" both waste and production time "which combine to yield a lower product cost".
Johnson is equally effusive: "The next generation of engineers won't be limited to the constraints of machining-based manufacturing", he opines. "Additive manufacturing is a huge leap forward in our vision for the factory of the future."
3D or not 3D, that is the question...
Technology is a subject often spoiled by hype: remember the promise that nuclear electricity would be too cheap to meter? Or the expectation that man would be on Mars by the late 1980s? Or that computers and associated automation would allow us all to retire to a life of luxury?
What about 3D printing? What are we to make of news stories about printing food on the space station or the European Space Agency's concept for a 3D printed moonbase? Are these just examples of over-hyped imaginative ideas or grounded in reality? Perhaps a bit of both.
For a start, both ideas are dependent on what is known in the printer world as 'feedstock', in other words the stuff you feed your printer. In the first case, the feedstock must be an edible, presumably water-soluble substance, while in the second it is lunar regolith (soil or rock) converted into a form of 'lunar concrete'.
While it might seem cool to transmit the design for a candy bar to the ISS and 'print it' in situ, instead of packing it onto a delivery capsule in the old-fashioned way, the mass of the feedstock and the water (and the mass of the printer) still has to be transported into orbit. And as for printing moonbases, isn't this just automated concrete laying? Admittedly, it's a challenging and clever system that can build a moonbase without direct human intervention. But is this really 3D printing, or a misappropriation of an industry term?
The idea of building up a structure or component from stock material has been realised in the form of plaster-of-Paris models and fibre-glass boats for decades. In the aerospace industry, the technique was adapted to 'lay-up' composite structures for radomes and antenna reflectors. In parallel, computer-based design and computer-controlled machines have made the transformation of digital design files into products a daily reality. So is additive manufacturing simply an extension of CAD/CAM, a software-driven lay-up?
If the technology had peaked with the novelty keyfob, or even the printed socket wrench, this might be a fair assessment, but consider Lockheed Martin's concept of 'pointwise composition control'.
The project started with a recognition that thermoplastic polymers should form the 'matrix' of future materials, because they were easily melted and could be deposited a layer at a time. They could also be compounded with a wide variety of reinforcing materials, such as carbon fibre and nanotubes, to control final properties. It didn't take long to realise that these new compounds used as feedstock for a 3D printer could produce "advanced design geometries that were not previously possible". Moreover, engineers could not only control the composition and functionality of a part, but also the composition at any location within the part. This is the nub of pointwise composition control.
A practical example that would benefit from the process is an aircraft nose radome, which typically houses a radar: the part must be light and strong, but its forward section must be transparent to radio waves. Another component printed to illustrate the potential and "capture the imagination of designers" features a circular gridded aperture at one end and a square grid at the other, with a colour transition from blue to green, the shade being an analogue for changing thermal or electrical properties.
But that's only the start of the concept, which foresees building an entire vehicle, a hand-launched UAV perhaps, in a continuous process, with printed electronics embedded in the wings and solvent-resistant material for the fuel tank liners. A multi-axis robot coupled to feedstock mixing and blending equipment provides flexibility and access to a wider manufacturing area. Lockheed Martin says it's simply a matter of changing the properties of the feedstock on the fly.
For larger aircraft, the firm declares, you just need a larger factory and more robots, some floor-mounted, some on gantries, all working in concert. Perhaps printed rocket engines are closer than we think.