Adrian Bowyer on The Future of 3-D Printing
Manufacturing will be more like farming, says founder of the self-replicating 3-D printer RepRap.
It’s easy to understand the allure of what 3-D printing seems to promise: on-demand, fully customizable, distributed manufacturing. Someone steal your bike seat? No worries, download the template for one and print yourself another on your desktop 3-D printer. Want a unique gift for your partner’s birthday? Design a picture frame from scratch and print it, as well as a charming photo of you two, using your 3-D and inkjet printers respectively. Need an extra dining chair for a big family dinner? Your at-home 3-D printer might be too small for the job, but the Kinkos of the future right down the street can help.
These were the kinds of scenarios that 3-D printing enthusiasts began fantasizing about in the late 2000s. Before then, 3-D printers – which use a variety of techniques to essentially melt, assemble, and harden materials (mainly plastic) into their desired shapes – had been operating in the industrial shadows for decades, primarily producing prototypes for large manufacturers like Boeing.
That changed in 2005 with the launch of RepRap. Short for “replicating rapid prototyper,” the open-source project had a deceptively straightforward goal: design an affordable 3-D printer that could produce its own parts. Along the way, it inspired a number of spinoffs, like MakerBot, whose founders were previously involved in RepRap and which became the premier brand of desktop 3-D printer, as well as most laypeople’s introduction to the technology.
RepRap was started by 66-year-old English engineer and mathematician Adrian Bowyer while he was a lecturer at the University of Bath, from which he has since retired. Bowyer is humble when credited with ushering in a new era of 3-D printing – “I hesitate to say that I’m responsible for everything because that sounds a bit like me blowing my own trumpet,” he says – but he acknowledges the role RepRap has played in lowering the cost of the technology by “a factor of about one hundred,” in his estimation. I had the pleasure of speaking with Bowyer at length over the phone about mind-boggling three-dimensional electrical circuitry, the process by which increasingly accessible technology meets ever more adept users, and his theory that manufacturing will become a lot more like agriculture. Our conversation, reproduced below, has been edited for clarity and brevity.
How would you describe the current state of desktop 3-D printing? Is it just hobbyists tinkering?
It’s moved beyond that. People who have use for it but are not interested in the technology, quite a large number of them have systems. Good examples of such people might be artists and sculptors. Obviously they have a certain amount of technical knowledge because they need that to do their traditional activity, but nonetheless, their interest in 3-D printing is a way of producing three-dimensional objects for their expressive art. Chefs are quite interested in it for printing food. It’s beginning to move into places where people aren’t just hobbyists and tinkerers who are doing it. And of course the people who are not hobbyists and tinkerers are the ones who are driving the ease of user interfaces and the improvements in the software that makes it much more simple for them to get their head around it.
Technically, the machines are in a state of transition, in that for the last year or so we started to see machines that can work with multiple materials…. The real revolution will happen when we’ve got low-cost machines that are capable of dealing with four or five materials of radically different electrical conductivity, thermal conductivity, mechanical strength, that sort of thing. And then you can build really complicated objects. If you ask me to put my finger on what the killer app will be, it will be printing with multiple different materials.
Because that will allow you to make finished goods?
Yes, finished goods, true, but just take electrical conductors as an example. The machines of course work with plastic, which by and large are electrical insulators. Now as soon as you can work with a really good electrical conductor as well, you can start to build three-dimensional electrical circuitry.
All the electronics that we use are essentially two-dimensional. You get profoundly greater ease of design and reduction in constraints on what you can do when you increase the number of dimensions in which you’re working, and going from two to three is a very, very big step when it comes to laying out electrical devices, because suddenly the whole problem of interconnecting everything almost goes completely away. If you’re making a two-dimensional printed circuit board, well, it has to have two sides so you can run tracks across each other without them touching…. All of that sort of consideration goes away the minute you flip up a dimension into three. To make a long story short, [3-D printing with materials that are] electrical conductors will allow a whole vast number of things to be done, and they’re literally things that we don’t have any other technology to do, even in the most expensive production facilities, let alone people in their homes.
What advancements are necessary for the printing of three-dimensional electrical circuitry?
It’s more a material problem than a problem of the design of the machines. The ideal material would be one that behaved like a plastic but that conducted electricity like a metal, and we simply don’t have a material like that at the moment. My guess would be somewhere between three and six years’ time we’ll have something which we can easily and reliably print with and which can conduct electricity almost as well as a metal.
Many people seem to believe that 3-D printing holds the promise of a new industrial revolution based on on-demand, customized, distributed manufacturing. Considering where the technology is at currently, do you think that’s a reasonable perspective?
It’s possible that that might happen, but of course I’m very old. I’m so old that I can remember the microcomputer revolution of the late 1970s and early ’80s. Those who acquired small computers in 1978 had to solder them together ourselves, and indeed I did. That wasn’t going to be done by everybody. And then gradually they got easier and easier to use, and gradually the designs improved. You know, they started out where you had to buy a kit and then you could build a computer which consisted basically of a circuit board and a keyboard on a bit of wire with no plastic case around it. Then Apple came along and introduced the Apple IIe. That took 15 years or so, and progress in that field is still happening now. Looking back, nobody says the computer was a flash in the pan and it wasn’t going to go anywhere because we’ve got the benefit of history. 3-D printing, the low-cost end of 3-D printing now, is like where computing was in 1983. It looks promising, it looks as if it is going to spread, but nobody can quite tell. I think it’s possible, but I’m not going to say it’s definite.
How long do you think that process would take? Fifteen years like the spread of microcomputers?
Probably quicker because of those computers, because communication is now so much easier for everybody. What that means is technology tends to improve more quickly because ideas spread more quickly. I suspect that we’re now in a position where hundreds of thousands of people around the world have their own 3-D printers. Hundreds of thousands is not the same as, well, if you count mobile phones, billions of people have their own computer. Hundreds of thousands to billions is a big gap, but it’s a gap that can be bridged quite quickly with exponential growth, which is, of course, what happened with computers and seems to be happening with 3-D printers. But [3-D printers are] not really easy enough to use for people who are not in some way technically oriented yet, just like the early computers. You don’t need to be an engineer to use a mobile phone. You don’t have to be particularly smart to do it. With 3-D printing you do have to be sort of smart to use the machine. Well not so much smart, but have a certain amount of instinct and feel for how engineering works. Improvements are going to increase the spread [of 3-D printers] as they do with every technology because they make it more accessible. How long it will take, I don’t know. Ten years seems like a good guess to me.
In ten years, do you imagine that we’ll have in-house desktop printers or something like Kinkos?
That second alternative is a staging post on the way, I think. Take conventional printing, for example, the sort of printing that produces books and newspapers. Along came the laser printer, which [at first] was [still] incredibly expensive. I can remember when they used to cost $10,000. People started setting up bureaus in main streets and high streets all over the world, so you could go in and get something nicely printed very, very quickly. The cost reduced even further and now those of us who need to print things can have something costing $100 in our house. And I think 3-D printing will go the same way. There are lots of examples like this. Laundry, for example. It used to be that citizens would send their laundry off to the town to be washed, where they had all the machines for doing it. But nobody does that now because we’ve got a robot in our kitchen to do it. The technology gets cheaper, cheaper, cheaper, and eventually people say, “Well, it’s only a few hundred dollars, I can afford to invest that and have it just sitting on a shelf 80 percent of the time,” which is, of course, what you do with your washing machine. The most interesting thing about the washing machine is not when it’s working, it’s the enormous amount of time it sits idle.
So you do believe that there will be a 3-D printer in every household?
I’d say about the same proportion as there are inkjet printers.
Will the machines become easier to use? Or will the public become more adept?
A combination of both, I’m sure. It’s unquestionably the case that the tablet now is much easier to use than the Apple IIe in the 1980s. The software is easier to use, the design is improved. But also, if you just pluck an average person off the street and ask them what an IP address is and how to set up an ethernet in their home, they’ll have a rough idea of how to do it. If you’d done that to someone 30 years ago, they wouldn’t have had a clue what you were talking about. The skill level in the population rises and the ease of use of the technology gets better and better. These two things tend to meet in the middle.
Besides the use of multiple materials, what would you predict about the near future of 3-D printing?
Significant improvements in user interfaces and ease of use of the machines, having the machines monitor their internal behavior in much more sophisticated ways. Given the cheapness of current electronics, having them look at their internal behavior in such a way that they can correct themselves on the fly and reduce the amount of skill needed to operate them, I think that will be quite important. Do you remember – going back to conventional document printers – do you remember when you had to print out a test piece and say, “Oh, this one matches this one.” That’s just one tiny example. You have to do that with 3-D printers all over the place all the time at the moment, but of course now when you buy a [conventional] printer, the printer internalizes all of that. It does it with a little internal camera and it prints the test page, but it also looks at it itself and figures out how to fix itself. That sort of thing.
And this will come about as a matter of course through better design?
Oh, I don’t doubt it. People are working on this all over the world. Because RepRap is open source, a lot of this information is open source and it’s being developed by the open source community, so it’s generally available.
How does that open source development affect the future of 3-D printing?
If I have a plant I can give you a seed and no money changes hands. This is another example of that kind of thing. I suspect that the whole of technology is becoming far more like farming. It’s very interesting if you look at the world of companies and technology. I did a little bit of research on this for a talk I had to give a few years ago. You look at the Forbes list of the biggest companies in the world…. that list is very interesting. It’s very interesting because if you ask Forbes what the world’s biggest industry is, they say agriculture. And nowhere in that list of big companies is there any agricultural company. That’s very strange when you think about it. Why is it that of the world’s biggest companies, none of them is involved in the world’s biggest industry? That is a very odd thing indeed and it screams out for an explanation. I think the explanation is because agriculture, farming, is based entirely on things that copy themselves. Everything that’s done in agriculture depends on objects that copy themselves. Because of that, there’s this continual entry into agriculture of people who have almost no resources but nonetheless can do it because the requirements to get started are essentially zero. Given a handful of seeds you can have tens of square meters of crops one year and you can have a field full the next year. That type of growth is impossible except when you’re dealing with things that can copy themselves.
Back to 3-D printing: RepRap is entirely about making 3-D printers that can copy themselves. That’s got to beat every technology in the end because it’s simply the exponential growth that comes from self-copying. As I say, very interesting that agriculture doesn’t feature in the top companies in the world yet it’s the world’s biggest industry. That’s because it’s self-copying. I think that self-copying 3-D printers are going to make the whole of manufacturing much more like agriculture.