Toaster.
Coffee-maker. Blender.
Three-dimensional printer?

By Michael Gillis

That’s the kitchen of the future, according to Hod Lipson, Cornell assistant professor of mechanical and aerospace engineering, and graduate student Evan Malone, who see inexpensive fabricators, able to create objects on demand, one day joining the stable of well-worn home appliances.

Cornell’s Fab@Home is a project with that goal in mind. Developed in the Computational Synthesis Laboratory, it could transform manufacturing by eliminating the need for the mass production of common items.

Whether it’s a toothbrush, a fork, a shoe or an action figure you need, Lipson said a trip to the store won’t be necessary. Instead, it will be as simple as pressing “print” and waiting for your home fabber to build what you need, before your eyes, in the comfort of your own kitchen.
 
“It’s a revolution about to happen,” Lipson said. “You could fabricate relatively complicated objects on your desktop. That could change the way we design and consume products in an incredible way.”

And Fab@Home is catching on fast. The blueprints, software, and drivers are available online at www.fabathome.org, and the components can be had for about $2,000. A New Mexico company, Koba, even sells all the parts ready to assemble for less than $3,000, and it is struggling to keep up with the demand.

Graduate student Evan Malone and MAE Prof. Hod Lipson at work on their fabber design in the Computational Synthesis Lab in Upson Hall.           University Photo

That has sparked interest beyond Cornell. Home “fabbers” are under construction all over the globe and scientists and hobbyists alike are experimenting with Fab@Home machines, pushing the project’s boundaries. A high school student in Kentucky, Noy Schaal, recently used a Cornell-built fabber to build a chocolate replica of her home state, winning first place in her school’s science fair. Biology researchers at Vanderbilt and Rockefeller universities are employing Cornell-built fabbers. The government of South Africa recently asked to see Fab@Home at work in local “fab labs,” and the Science Museum of London has added one to its permanent collection, using it in an exhibit on the history of plastics. Fabbers have been built by computer-aided design and manufacturing instructors at the University of Washington, doctors in Austria, a fine arts student in Chicago, researchers in Japan, and hobbyists in Massachusetts and the U.K. Lipson stresses it was decided early on that Fab@Home would evolve as an open-source endeavor. Shifting production away from corporations and specialized engineers will “democratize innovation,” he believes.

Anyone with a screwdriver, soldering iron, a little imagination, and a home fabber will be able to design and create items otherwise cost prohibitive to produce in small quantities.

Once that happens, printing a toothbrush or a shoe is only the beginning. That same toothbrush can be designed to fit the ergonomics of a specific hand. A shoe can be crafted to fit the contours of the foot. Malone has shown that fabbers can produce working batteries, artificial muscles, transistors, and even living biological tissues.

Eventually this capability will trickle down to the home version, allowing more complex objects with working parts to be printed in the kitchen and emerge from the fabber fully functional.

A prototype for change
Lipson’s earlier work in robotics, where machines were given the tools of evolution and motivation to self-replicate, is what led him to Fab@Home.

From a CAD model of a rubber squeeze bulb, the fabber creates one of silicone                                     Provided

While experimenting with “autonomous adaptation in behavior and morphology of robotic systems,” Lipson stirred up a “primordial soup” for machines, providing the building blocks of robotic life and letting machines build themselves—evolve. That happened in simulation, where oddly-shaped objects formed and strived for self-replication, the ultimate evolutionary reward.

Lipson said some of these strange machines were later constructed, but could not be assembled without manually adding individual components, unlike the simulations. That’s when Lipson turned to rapid prototyping technology, which allows objects to be “printed” from raw material.

“I said, wouldn’t it be great if I could have a rapid prototyper machine that could print the entire thing, if it could print the wires, the actuators, the batteries, all from raw materials?” Lipson explains. “That was the impetus to exploring this idea of multi-material rapid prototyping.”

Rapid-prototyping technology has been tapped in industrial and commercial labs for more than a decade. It is called upon to fabricate models of everything from cell-phone shells to car parts, but differs from the home fabber in both cost and size. Industrial machines

The fabber makes a replica of a Lego tire.             Provided

are typically expensive and severely limited in the materials they can work with, usually one at a time. In addition, industrial and commercial applications call for more sophisticated and specialized technology, including high-powered lasers and accurate thermal processing, and discourage users from experimenting with their own materials.

Home fabbers are intended to work more like an ink-jet printer, but instead of ink, they squeeze out a variety of substances—everything from cheese to silicone—to create three-dimensional objects, one layer at a time. Those substances are first poured into syringes, which can be swapped out of the fabber to alternate the material. Some home fabbers have already been customized to hold multiple syringes. Some day, Lipson sees people buying cartridges for their home fabbers the same way they buy ink now.

The future is now
Fab@Home may be following the path of a machine that changed the course of history, Lipson points out.

The 1975 edition of Popular Electronics introduced its readership to the MITS Altair 8800, which many people believe to be the first home computer. Although the Altair did require some expertise, it opened the door to a home computing movement. More importantly to Lipson, anyone with a little know-how could hack the Altair and make modifications.

Critics at the time doubted the machine would register at all, suggesting there was no interest or demand for a computer in the home.

But once people had access to the inner workings of the Altair and discovered how to modify it, there was suddenly a need for software, accessories, and games, and interest exploded.

Lipson hopes Fab@Home will catch on in much the same way.
“We need to nucleate this revolution by making a build-your-own rapid prototyping machine,” Lipson said.

For that to work, Lipson adds, it needs to be low cost, functional, hackable, and multimaterial.

“That’s what we set out to do with Fab@Home,” Lipson said. “We just put it up on the Web—we didn’t do any publicity—and it just exploded.”
In just a few months, the Fab@Home Web site logged 5.5 million page downloads from more than 350,000 unique visitors. Some are more driven than curious, and are working to expand the machine’s capabilities. In addition, users are experimenting with a whole host of materials, including Play-Doh, wax, cheese, cake frosting, and gypsum.
“It’s driven by hands-on people,” Lipson said.

Proponents realize there is still much to do and the machine, in any version so far, is far from perfect, especially when compared to its industrial equivalent.

“It does not compare well, in terms of accuracy and speed,” Lipson said. “So our machine is slower and less accurate. It’s a little like comparing the Altair 8800 to the mainframe of the day.”
The idea, Lipson adds, is to make Fab@Home accessible to all. “People will help perfect it,” he said.

One of those people is Malone, who designed and built the first fabber—the Fab@Home Model 1—in Lipson’s lab in the summer of 2006.

Malone remembers being drawn to Lipson’s vision of robots and self-replication. Malone hopes his fabber will make a big leap toward that vision, which is how he intends to earn his Ph.D. He anticipates that by next winter, his research fabber will construct an entire robot able to emerge from the machine on its own accord, replete with circuitry, actuators, and batteries.

That lofty goal aside, Malone eventually found himself attracted to the idea of making multi-material fabrication technology available to everyone, which hit home on a trip to South Africa to demonstrate Fab@Home.

Managers at the Soshanguve FabLab in South Africa get to see a fabber in action.          Provided

“The personal interaction with the people down there made the empowerment aspect of this project much more practical and immediate,” Malone said.

Although Malone is looking forward to completing his doctorate and moving on, he does see a continuing role for himself with Fab@Home, at least as an adviser and possibly as a part of a business venture.

The inner workings
When Malone compiled a list of materials for the first Fab@Home, he said he wanted to make sure all of the necessary components were available to anyone eager to build one. Once all those materials are in hand, Malone estimates someone can build a Fab@Home over a weekend.

The “printer’s” chassis consists of 40 acrylic pieces, laser-cut to specification, and a variety of linear stepper motors, bearings, rails, and fasteners. Mounted on this are several subassemblies, including a motorized table that moves up and down to accommodate layers of deposited material, the positioning system that controls the deposition syringe in the horizontal plane, and the “syringe tool” which holds the syringe and the motor that depresses it. Fab@Home uses standard syringe barrels, which are disposable and accommodate a wide variety of materials, according to Lipson.

After the table and positioning system are installed, the parts that process commands and help shape objects in the fabber are added: a 4-axis amplifier, a microcontroller, limit switches, and a breakout board and cables. The firmware for the unit’s microcontroller can be downloaded from Fab@Home’s Web site.

A few days and about $2,000 later, you have a fabber.

The simplicity of Fab@Home is one reason it continues to generate interest, but another strong component is the open-source philosophy behind it.

Some people, for example, have already suggested how to reduce the price by assembling the motors in house, although that admittedly requires some engineering skills.

Lipson and Malone created the software that imports an object’s three-dimensional information and controls the fabber to build it. Lipson points out that  the software uses standard computer-aided design formats to read information, which helps ensure compatibility. It is written only for Windows at the moment.

So what’s been “fabbed” so far?

Everything from a squirt bottle, watchband, propeller, batteries, a working flashlight with printed wiring and switch, and even a Darth Vader head. Many other objects have been designed using chocolate, cheese, and cake frosting.

Lipson said another Web site, www.3Dprintables.org, will soon include downloadable blueprints for many objects able to be created in Fab@Home and other rapid prototyping machines.

Desktop future
Lipson and Malone see exciting possibilities for the future of desktop manufacturing.

The simple objects being created today will lead the way to more complex creations, they believe. In the same way the digital revolution allowed music and images to be downloaded and shared, Lipson sees home fabrication being as easy as downloading a design and printing it at home. As the technology evolves, and multiple materials can be printed in the kitchen fabber, objects of greater complexity will be feasible.

Instead of buying an iPod online and waiting for it to be shipped, for example, Lipson sees a future where consumers will download the blueprints and print it at home.

For Lipson, the most dramatic possibility for home fabrication is changing the way we invent and create. He said it harkens back, ironically, to a time of custom creation and manufacturing’s soul. Convincing society that makes sense may take more time, however.

The Fab@Home Model 1 builds a watchband with an embedded watch.             Provided

“It will take at least a generation of engineers to escape the idea that in order for something to be cheap, it needs to be mass produced,” Lipson said.

But perhaps that greatest hope for home fabrication is the imagination. That, Lipson said, could mean a new age of invention.

“Forget about companies designing these things,” Lipson said. “It will be ordinary people designing these things. It’s a liberating process. It liberates talent that is now only obstructed by the tremendous barriers of resources and skills that prevent most ideas from being realized.”

All in the comfort of your own kitchen.