Making Everything In The Desktop 3-D Workshop

When Defense Distributed tested a functional pistol it fabricated out of plastics with a three-dimensional printer and published the blueprints online last spring, the self-described anarchist group earned swift condemnation from critics who worried that easily accessible, unregulated, and undetectable firearms would make the problem of gun violence even worse.

Sens. Charles E. Schumer (D-N.Y.) and Bill Nelson (D-Fla.) introduced legislation in June to extend the Undetectable Firearms Act of 1988 to include a ban on homemade 3-D-printed gun parts. On CNN, Schumer called Defense Distributed’s publication of the gun plan a “reckless act.” He warned that “a terrorist or a felon can make a gun in the comfort of their home—not even leaving their home—and do terrible damage with it.” 3-D printers are small enough to sit on a desktop and fabricate parts by extruding molten plastic.

Journalists from an Israeli television station were among the 100,000 people to download the computer-aided design, or CAD, files for the gun. The reporters printed the firearm and tested it. They also brought parts of the gun they made­—undetected—into a press conference with Israeli Prime Minister Benjamin Netanyahu.

The New South Wales Police Force, in Australia, fabricated a gun using a $1,700 3-D printer. Upon testing, the gun burst into pieces. “Not only are they illegal,” said Commissioner Andrew P. Scipione, “but they are enormously dangerous, both to the person you are choosing to use them against and to yourself.”

The truth is, to date, 3-D printers can produce only minimally functional firearms. Such guns, even their developers admit, aren’t nearly as reliable as guns that are professionally manufactured. The plastics most commonly used in hobbyist-grade 3-D printers—polylactic acid (PLA) and acrylonitrile butadiene styrene (ABS)—aren’t up to the task of containing the extreme forces involved in firing a bullet. That has customarily been a job for metal and specially designed polymers.

But that could change as 3-D machines improve, more materials for them become available, and the designs of the firearms evolve. Moreover, 3-D printers have already made semiautomatic weapon parts of high quality, a development that may prove more dangerous than a crude polymer gun that can fire a single round.

The making of guns and gun parts has grabbed a lot of attention, but it is a small corner of the nascent 3-D-printing world. Regular hobbyists are gaining access to a technology that was once used primarily by professional engineers. They are using 3-D printers to realize in physical form the art, toys, and contraptions that previously existed only in their imaginations. Major retailers—already taking this new market seriously—are offering the printing machines. Staples began to sell 3-D Systems’ Cube for $1,300 earlier this year. RadioShack offers the Afinia H-Series printer for $1,600.

People in the broader 3-D-printing community don’t readily opine about those who are using the technology to develop firearms. And when they do, the gun makers are regarded as interlopers, co-opting a useful technology for an unintended purpose. “These technologies, if you can use them to build anything, then you can build anything with them,” says Ryan Wicker, professor of mechanical engineering and director of the W.M. Keck Center for 3D Innovation at the University of Texas, El Paso. Components for jet aircraft and custom-made medical implants are potentially more challenging and constructive applications for 3-D printers, he notes.

Rod Strand, sales director for printer maker LulzBot, seconds the sentiment. “While we don’t condone the printing of potentially harmful items, we can’t control what people use their 3-D printers for,” he says. “We do hope that people are using them for more innovative things than finding creative ways to make guns.”

MakerBot, a New York City-based 3-D-printing start-up that was recently acquired by the professional-grade printing firm Stratasys, aims to create a friendly environment for people who want to get acquainted with 3-D printing. The firm opened the first store dedicated to 3-D printing last fall in Manhattan. It sells Maker­Bot’s own Replicator 2, assembled across the East River in Brooklyn, for $2,200, as well as PLA filament to feed the machines for $48 per 1-kg spool.

Multiple machines set up in the store churn out demonstration models such as bolts, chains, and figurines. The process the machines use, called fused filament fabrication (FFF), is slow and meticulous. The robotic nozzles build the items by drawing them in molten polymer layer by layer. An object the size of an orange takes hours to produce.

The store gets steady traffic of tourists, curiosity seekers, and serious customers. A salesman assured C&EN that the outlet sells printers every day. The location also serves as a hub of 3-D-printing activity, including classes for users getting started in 3-D design.

Beyond the store, MakerBot hosts Thing­iverse, an online community for 3-D-printing hobbyists who want to share CAD files of their designs. The website offers templates for hundreds of sculptures, toys, and machines, some of which required immense creativity and countless hours of engineering to develop.

One of these machines is the Turbo Entabulator. It is the brainchild of Chris Fenton, an electrical engineer whose day job is building supercomputers. He is also an avid tinkerer. One of his recent projects was building a miniature version of the Cray, a 1970s supercomputer.

Fashioned from ABS and printed on a MakerBot machine, the Entabulator is an attempt to make the slowest, most impractical computer possible, just for the fun of it. It is a mechanical computer that uses punch cards to tabulate the Fibonacci series up to the number eight. Fenton has designed dozens of mechanical parts from scratch.

“You could make it out of injection-molded plastic, but you would never build the molds because they would cost you hundreds of thousands of dollars,” he says. “It is a machine that almost can’t exist unless you have something really low cost like a 3-D printer to make parts.”

The polymer choices for FFF machines are limited, especially for the entry-level models. PLA and ABS are the plastics most commonly used. “They have relatively low processing temperatures, and they are also rather inexpensive,” Wicker says.

The two polymers offer different properties, explains Strand. PLA is more brittle than ABS. PLA is also harder to print with because it is less viscous, making it difficult to maintain the integrity of the column of material that comes out of the nozzle. The drawback of ABS, on the other hand, is that it shrinks as it cools, potentially creating warps and other imperfections in the finished product. A machine that processes ABS needs to have a heated bed to maintain a constant temperature as it builds the part.

Sturdier plastics are available for use in FFF machines. LulzBot offers experimental grades of nylon and polycarbonate. Nylon has a very low viscosity. “It is not something you are going to want to try the first time you create a 3-D print,” Strand says. This is also true of polycarbonate, which adds another complication: It needs to be processed at higher temperatures.

Commercial-grade FFF machines from other suppliers can handle even more exotic resins, such as polyether imide, polyether ether ketone, and polyether ketone ketone. But such machines can cost more than $100,000.

Personal printers, those that sell for less than $5,000, may soon become big business, according to Terry Wohlers, president of the market research firm Wohlers Associates. Such printers comprised 6.4% of machine sales of $618 million in 2012. But the personal market is growing: 46.3% in 2012 versus 28.6% for the rest of the market.

Wohlers observes that the community of interested dabblers in 3-D technology is a positive portent of things to come, harking back to the early days of personal computers in the 1970s. “It is not a big market, but it is growing, and it will continue to grow, especially as those systems bridge the gap to higher-quality, professional-grade systems. And they will,” he says.

It’s an offshoot of this community of hobbyists that is producing 3-D-printed weapons. One key player is Michael Guslick. Back in 2011, Guslick, an engineer from Milwaukee, was looking around Thingiverse for objects to print on his secondhand 3-D printer. He found plans for a magazine follower, a plate that attaches to a spring in a rifle magazine. Guslick wondered whether 3-D printing could yield more mechanically complex gun parts.

Guslick has a background ideally suited for such a project. In addition to being an engineer, he is a self-taught machinist. He has a side business modifying paintball guns. He has even tried his hand at conventional metal-based gunsmithing.

Guslick is also a capable practitioner of 3-D printing. To circumvent the high cost of the spools of ABS sold by Stratasys for use in its machines, he purchased a third-party spool that was the same 0.07-inch gauge. His test prints were a mess. Not all ABS, he learned, is the same. He decided to make his own filament. Ashland Distribution sold him a 55-lb bag of SABIC Innovative Plastics’ MG47 resin, which he deemed close to the genuine Stratasys article, and he found a local extruder to form it into filament. The results were much better.

For the gun project, Guslick set his sights on a key part of the AR-15. The gun is a variant of the M-16 used by the U.S. military during the Vietnam War era. It was also the gun used in the mass shooting in Newtown, Conn.

Motivated, he says, not by politics but by curiosity, Guslick wanted to make the lower receiver, the part of the AR-15 that other components—the buttstock, magazine, and grip—attach to. It also houses the trigger mechanism. Guslick likens the lower receiver to an automobile chassis. It is also the regulated component of the gun. A buyer needs a background check to get one. All the other components of the gun can be acquired without such controls.

Guslick found a CAD file for the lower receiver on a metal-machining website. He modified the design to reinforce areas in the part he thought might be vulnerable to mechanical stress. After printing it, he had to do some finishing steps, such as drilling and reaming, by hand.

After assembling the rest of the gun around the printed receiver, Guslick tested it using .22 caliber rounds, which are much less powerful than the .223 Remington rounds normally used in AR-15s. The receiver worked fine.

The gun didn’t perform nearly as well using the .223 rounds. The gun had cycling issues, meaning the cartridges weren’t being fed and extracted properly. The receiver was flexing too much when the gun was fired, which Guslick blames on ABS’s low modulus, or stiffness. “It certainly wasn’t stiff enough to operate reliably,” he says.

Defense Distributed did its own print of Guslick’s design. However, the organization used a photopolymer-based printing system, not FFF. The thermoset resin used in the process is stiffer, but it doesn’t have the strength of a thermoplastic. The difference showed. During testing, Defense Distributed’s lower receiver cycled well, but it cracked after a few rounds.

Defense Distributed then modified Guslick’s design to compensate for the shortcomings of the polymer it was using. The new receiver managed to fire more than 600 rounds.

If that result seems chilling, it’s meant to be. Defense Distributed wants to make gun laws obsolete through 3-D-printing technology. “We see this as a political project,” says Cody Wilson, a UT Austin law student and the group’s founder. “It begins and ends in ideology.”

Defense Distributed’s flagship project was to develop a fully functional 3-D-printed firearm. In addition to Wilson, who has no technical background, two others worked on the group’s gun. One was an electrical engineer and a gun enthusiast. Another had experience in 3-D printing.

The idea of plastics in guns isn’t novel. Plastics have been used in guns for decades, most famously in the frame of the Glock 17 handgun, now favored by police departments around the world. The material used to make the Glock is said to be a glass-filled nylon resin, though the company, invoking trade secrets, refuses to confirm this.

Whatever the material is, a gun made with it “can shoot until it is smoking, and it still works fine,” says Doug Robinson, a member of Glock’s technical services department. Some critical components of the gun, such as the barrel, are made of metal.

It is the 3-D printing and the choice of material, a grade of ABS available from Stratasys, that make Defense Distributed’s project unique. Overall, the program took about eight months to complete. About six of those months were spent getting a license to manufacture firearms from the Bureau of Alcohol, Tobacco, Firearms & Explosives. This, Wilson says, was the “surest cover” for the enterprise. It is legal to make firearms for personal use, but there are limits to what can be manufactured and distributed.

Defense Distributed still confronted obstacles. Stratasys recalled a printer that the group was leasing when it caught wind of what Wilson was up to. The group had to buy a secondhand commercial-grade Stratasys machine for about $8,000.

While the group was awaiting the license, which arrived in March, Defense Distributed made magazines for rifles and tested barrels for the printed gun. The key hurdle was maximizing barrel strength. In 3-D printing, the orientation of the threads of resin is an important determinant of strength.

With that in mind, the group settled on printing the barrel in concentric circles. The group also treated the ABS barrels with acetone vapor, a common trick among 3-D-printing hobbyists, to smooth out the bore and help prevent barrel failure.

It took just two months to design the gun, namely the grip and the firing mechanism, around the barrel. The only metal parts were a nail used as a firing pin and a steel plate inserted into the device so it would comply with the Undetectable Firearms Act.

Defense Distributed tested the gun, which they named the Liberator, using .380 ACP rounds, the same small bullets used in James Bond’s Walther PPK—most states don’t require a license to buy ammunition.

The Liberator worked, at least in the sense that it managed to fire a bullet. The plastic around the cartridge expanded so much that the team had to pry out the spent shells between shots. And every successive discharge deformed the gun more. The best that Defense Distributed was able to accomplish in subsequent testing is five rounds without switching parts.

Guslick and a friend printed their own version of the Liberator on a hobbyist-grade LulzBot printer. They made a couple of minor modifications to the design, such as using metal pins, instead of plastic, to hold the gun together.

The results were similar to Defense Distributed’s. They were able to get nine rounds out of the gun before it became too deformed to use. But bashing out the spent rounds with a rock, replacing screws, and constantly readjusting the firing pin took time—three hours to shoot the nine rounds.

Guslick measured the muzzle velocity. The .380 round travels at about half the velocity of a bullet fired out of a conventional pistol of the same caliber. He estimates that it carries only half as much energy as a Major League Baseball pitcher’s fastball. The ABS bore expands so much upon firing, he believes, that gas escapes around the bullet, robbing it of velocity.

Though clearly unreliable, the Liberator isn’t as bad as some of the videos circulating around the Internet would suggest, Wilson maintains. He accuses the creators of the videos, such as the police in Australia, of an “agitprop propaganda strategy.” He speculates that they deliberately designed the tests to fail by using PLA, firing bigger rounds, or not smoothing out the bore. “We have never had failures like some of the ones we are seeing on video,” he says.

Wilson does acknowledge that the functioning AR-15 lower receiver flouts guns laws more insidiously than the Liberator does anyway.

Guslick agrees the Liberator has some shortcomings: “3-D printing a gun is a very expensive and slow method to get an inferior product,” he says. “It is highly impractical. You can make a far superior gun with about $15 worth of components from the plumbing aisle of your local hardware store. So printing is really the long way around.”

Still, there is a community developing 3-D-printed guns on defcad.com, a forum sponsored by Defense Distributed. It is similar to Thingiverse, which long ago booted firearms-related activities off its site. On Defcad, variants of the Liberator are already evolving, much as the AR-15 lower receiver had before.

One user, who goes by the handle Cana­diangunnut, made a rifle version that fires a .22 round. He has no engineering background and became acclimated to 3-D-design software and printing only this year. “Since I was a kid, I have always built stuff with my hands and been mechanically inclined,” he tells C&EN.

Canadiangunnut used the forum to bounce around ideas and get advice. “Yikes! I never thought of the barrel breaking and becoming a projectile,” he remarks in one post. “I have not acetoned the barrel yet. How safe is it to bring that stuff to a boil?”

The first rifle that Canadiangunnut made cracked when it was fired. His subsequent design had a thicker barrel. The new gun managed to get 14 rounds off before the barrel started to split apart.

Just months after Defense Distributed gained notoriety with the Liberator, Canadiangunnut designed a much improved gun, setting a target for future Defcad users.