The Next Wave: Open Source 3D Printing

09/13/2011 10:00 am EST


Josh Wolfe

Editor, Forbes/Wolfe Emerging Tech Report

The 3-D printing movement may sound a bit like science fiction, but it’s already a reality…and one of the pioneers is not only breaking new ground in the field, but sharing it openly with anyone who is interested in building their own 3-D printer, says Josh Wolfe of the Forbes/Wolfe Emerging Tech Report.

Dr. Adrian Bowyer is a senior lecturer in the Mechanical Engineering Department at the University of Bath. He is a founder and director of RepRap Ltd. and the inventor of the RepRap Project—a worldwide, open source, self-copying 3D printer.

Dr. Bowyer joined the Mathematics Department at the University of Bath in 1977. His main areas of research are geometric modeling and computing, the application of computers to manufacturing, and the engineering use of biology. He is a co-inventor of a principal algorithm for computing Voronoi diagrams.

Why don’t we start with an overview of your work?
During my time here at Bath University, I’ve worked on the application of computers to manufacturing engineering, on the geometry of engineering shapes inside computers, and in the area of biomimetics, the science of applying ideas from evolved systems into engineering.

Recently I’ve been working on a biomimetic invention: a 3D printer known as the RepRap project.

How does biology relate to developing hardware—and how did you choose to focus on 3D printers?
Biology, fundamentally, is the study of things that copy themselves. I’ve always been interested in applying that concept to machines. 3D printing technology seemed sufficiently versatile that it ought to be possible to construct a printer that could make a copy of itself.

Also within biology, I studied systems that exhibit symbiosis, or mutualism. The classic example is the mutualism between insects and flowering plants: plants make nectar to bribe the insects to pollinate them, and both species benefit.

When I designed the RepRap machine, I tried to mimic that mutualism, and link the machine with the most obvious species choice: human beings. To create a mutualistic relationship, the RepRap machine gives the human beings useful stuff—essentially consumer goods.

In return, human beings give the RepRap machine their assembly skills. When the RepRap machine copies itself, it makes a kit of parts, and humans put the parts together. The humans are equivalent to the insects, and the RepRap machines are equivalent to the flowers.

Did the goal of building a self-replicating machine evolve out of your work studying biology, or was it a totally separate problem?
It was somewhere halfway between the two. The idea of a machine copying itself is not new; it goes back to Samuel Butler in the 19th century and John Von Neumann in the 20th.

Actually making one, however, was a bit of challenge. I approached the project not on the basis of a piece of technology, but on the basis as a piece of biology.

Anything able to copy itself should be subject to conventional Darwinian laws—therefore, I needed a design that would exhibit what biologists call an evolutionarily stable strategy—meaning there is a balance, an equilibrium, between the opposing forces. There’s no point in designing something that’s not evolutionarily stable, because by definition, it won’t succeed.

How has the outside world reacted to the success of the RepRap project?
We’ve seen a flourishing of companies making 3D printers priced from several hundred dollars, most notable perhaps being MakerBot in New York (which I helped to found and am a shareholder in), and Bits From Bytes in the UK (acquired by 3D Systems (DDD). But these are conventional printers; they don’t reproduce themselves.

These non-reproducing 3D printers actually have a very interesting characteristic, compared to RepRaps. Most non-reproducing 3D printers can make a RepRap, but RepRaps don’t make them.

So each conventional 3D printer only has to make one RepRap machine to create an equal number of RepRaps, but then the RepRaps can continue to make themselves. From an evolutionary perspective, you can see what that does to the population dynamics.

NEXT: The Dazzling Future


What’s your long-term vision for the possibilities, beyond prototypes and garage hobbyists, for 3D printers?
Conventional printing makes for a good analogy. Think about the production of Forbes magazine, when tens of thousands of copies are printed on a big press in a factory. On a smaller scale, most of us also run our own printing press with our own inkjet printers.

Those little machines could never handle the printing of Forbes magazine, but they produce perfectly adequate prints for our own needs. When technology is in the hands of individuals, suddenly speed doesn’t matter very much.

Imagine making some completely trivial everyday plastic object, like a coat hook. In a conventional industrial setting, an injection-molding machine could churn out thousands of coat hooks every hour, and they’d get distributed to hardware stores where they’d be marked up and sold.

That’s the conventional model for making stuff. Now, if everyone had their own 3D printer, then they wouldn’t need to drive to the store to buy a coat hook; they would just download a coat hook design and print it on their 3D printer. They don’t mind if it takes an hourthey just go out and have lunch and come home to have a coat hook.

Do you see limits on the complexity of what could be printed through these machines?
The trick here is to miniaturize the idea of self-replication. Again, we turn to biology to illustrate how it could work.

The reason why a tree can grow is because it’s composed of billions of tiny elements, each of which is copying itself. Once you’ve got something that’s capable of copying itself, it’s capable of growing the number exponentially, which is why you and I can weigh 60 kg or more. It’s only because we’re composed of things that replicate that we become complicated structures, put together in a reasonable amount of time.

Any manufacturing technology that could achieve similar results will need to use that same trick, which is another reason why self-replication is important. How long it takes to get there is anybody’s guess. Maybe 20 years, or maybe it will be like the fusion reactor, always 40 years away into the future.

From a market perspective, what becomes more valuable: the piece of hardware or the data for the designs?
It seems to me that the greatest value lies in the ownership of the data. However, the whole idea that one can own a piece of information is intrinsically flawed, because information does not behave like either energy or matter.

Energy and matter are both conserved. If you’ve got three kilograms and you give me one kilogram, then you’ve only got two—that’s the way the universe works. But if you give me a piece of information, you’ve still got it, and I’ve got a copy of it. Information, unlike matter and energy, is not conserved.

It only makes sense, ultimately, to talk about ownership for things in the world that are conserved, like matter and energy. We generate legal constraints around the ownership of information, which ultimately fail, because you can’t keep information contained.

An example is the MP3 format, used for music. Laws restrict us from duplicating copyrighted material and yet every 17-year-old on the planet has 30 gigabytes of illegally downloaded MP3s on their hard drive. That law is completely pointless.

Increasingly, as hardware moves in the direction of being created by moving information around, then the idea of protecting that information is going to go the same way as music has gone. Congress and Parliament can go blue in the face buzzing about it; it won’t make a blind bit of difference to what actually happens in the world.

Subscribe to the Forbes/Wolfe Emerging Tech Report here…

Related Reading:

Related Articles on MARKETS