Oak Ridge National Laboratory's robotic prosthesis looks like something out of medieval times—a hand clad in chain mail more appropriate for wielding a broadsword than a mug of coffee. Both the underlying skeleton and thin, meshlike skin are made of titanium to make the hand durable and dexterous while also keeping it lightweight. The powerful miniature hydraulics that move the fingers rely on a network of ducts integrated into the prosthesis's structure—no drilled holes, hoses or couplings required.
Yet what makes this robot hand special is not what it can make or do but rather how it was made and what it represents. Conceived on a computer and assembled from a few dozen printed parts by so-called additive manufacturing, more popularly known as 3-D printing, Oak Ridge's invention offers a glimpse into the future of manufacturing—a future where previously impossible designs can be printed to order in a matter of hours.
“You're looking at a very, very complex design that has internal hydraulic tubing that can be run in excess of 3,000 pounds per square inch,” says Craig Blue, director of Oak Ridge's energy materials program. “You have meshing to make it a lightweight structure, putting material only where you need it. There's no technology today, other than additive manufacturing, that can make that [robotic hand].”
As 3-D printing matures to the point where it can make complex machinery that can't be made any other way, big-volume manufacturers such as Boeing and GE are starting to apply the technology to their advanced product lines. Instead of the old approach of carving a usable part out of a large block of material, additive manufacturing builds an object up layer by layer. This shift in thinking has the potential to affect every facet of manufacturing—from prototype design to mass-produced product.
Yet technical challenges continue to bedevil 3-D printing. Compared with ordinary subtractive manufacturing, additive manufacturing can be slow, the fit and finish of its materials inconsistent. Further, 3-D printers have trouble building objects out of multiple kinds of materials, and they cannot yet integrate electronics without frying the circuits.
Researchers are working hard to overcome these limitations—and few doubt that for customizable, small-volume applications, additive manufacturing has tremendous power. As the technology expands into the mass marketplace, 3-D printing could begin to power a widespread manufacturing revolution.
The origins of 3-D printing stretch back to the late 1980s, when start-up companies and academics—most notably at the University of Texas at Austin—invented machines that could build three-dimensional models of digital designs in minutes. For decades those systems and similar types, which first cost around $175,000, gained notoriety for their ability to help inventors and engineers rapidly and relatively inexpensively produce their prototypes.
Since then, 3-D printing has taken two paths. At one extreme, hobbyists and would-be entrepreneurs can whip up plastic models using machines that cost $2,000 and less. These kitchen-counter devices allow users to invent new objects—a technology that has invited comparisons between 3-D printing and personal computers. “In the same way the Internet, the cloud and open-source software have allowed small teams to live on ramen noodles for six months and build an app, post it and see if anyone is interested, we're beginning to see that same phenomenon spread to manufactured goods,” says Tom Kalil, deputy director of technology and innovation at the White House's Office of Science and Technology Policy.
At the other extreme, large manufacturers are cultivating advanced, industrial-strength approaches to produce aircraft parts and biomedical devices such as replacement hips. The machines required to do this cost upward of $30,000, with laser-based appliances that make high-quality metal products selling for as much as $1 million. These printers can use polymers, metals or other materials in liquid or powder form. Objects begin as digital files, enabling designers to tweak their work before the building begins, with little impact on cost.
Printing in 3-D could replace certain conventional mass-production processes such as casting, molding and machining by 2030, especially in the case of short production runs or manufacturers aiming for more customized products, according to the “Global Trends 2030: Alternative Worlds” report issued last November by the National Intelligence Council, a team of analysts supporting the Office of the Director of National Intelligence. Aerospace companies are at the forefront of this trend. GE Aviation, which has been making aircraft engines for nearly a century, recently bought two suppliers that specialize in making aircraft parts via additive manufacturing processes. Boeing already uses 3-D printing to make more than 22,000 parts used on its civilian and military aircraft.
Such companies are discovering that 3-D printing can also be more efficient than conventional production, both in terms of energy and materials. “If you're machining a part, it's not unusual that 80 to 90 percent of the block [of material] you start with can end up as chips or scraps on the floor,” says Terry Wohlers, principal consultant and president of Wohlers Associates, an additive manufacturing consulting firm in Fort Collins, Colo.
Breaking the Mold
Despite these advantages, manufacturers still largely think of 3-D printing as a way of making prototypes rather than industrial-grade products. The reasons are threefold: slow speeds, inconsistent quality and the difficulty of building complex objects.
Foremost, additive processes are relatively slow, depending on the level of detail required. Oak Ridge engineers, led by principal designer Lonnie Love, spent 24 hours making the parts for their 1.3-pound robotic hand and another 16 hours assembling it. (They are developing hardware that will print the entire prosthesis in a single piece.) “If you were building something the size of a softball and you wanted fine features and definition, you could envision that taking six to eight hours to build,” says Richard Martukanitz, co-director of the Center for Innovative Materials Processing through Direct Digital Deposition at Pennsylvania State University. At those speeds, building thousands of units using 3-D printers would take years.
Some additive manufacturing systems work faster. Those developed for the U.S. Navy can deposit 20 to 40 pounds of material per hour. In this case, however, speed comes at the expense of feature definition, which is poor; finished parts also require postprocess machining, Martukanitz says. To speed things up, researchers are working on systems that print at variable speeds—quickly laying bulk material but slowing down when a part requires more detail. “This is getting a lot of attention now because people are seeing the limitations of the additive process from a productivity standpoint,” he notes.
Another option to boost speed is to distribute workloads across several manufacturing facilities. Yet this approach requires a higher level of standardization than what currently exists. A critical component of a GE jet engine should look, feel and perform reliably, regardless of how or where GE (or one of its suppliers) makes it. ASTM International, formerly known as the American Society for Testing and Materials, is one organization active in developing standards for 3-D printing, although its work is in the early stages.
Scientists are also trying to make self-monitoring 3-D printers that could quickly churn out consistent designs. The system would analyze high-speed video of the object as it is built or use infrared thermography to detect flaws, then immediately correct those imperfections without stopping the building process, Blue says. “You download the plans for the part to the printer and get the perfect part every time,” he adds.
The growing complexity of products—which increasingly incorporate many different materials, along with embedded electronics—poses a further challenge to 3-D printing. One approach is to develop 3-D printers with multiple extrusion heads, each depositing a different type of material. One of those heads could be used to embed wires directly into a device as it is being built.
Researchers at Oak Ridge, the University of Texas at El Paso's W. M. Keck Center for 3D Innovation, and elsewhere are designing 3-D printers that can also print circuitry. The challenge has been to avoid overheating and damaging electrical components while adding on the surrounding layers of plastic or metal. Investigators are testing ways to print insulating material around electrical components to protect them. “Within the next decade you'll see coupling of printed electronics with other materials,” Blue says.
Taken together, such advances hold good promise for Oak Ridge's robotic hand, not to mention anyone using it. The scientists envision a time when doctors will be able to scan a person's healthy hand, make a mirror image of it electronically and then print the new, preassembled prosthesis ready to use.