August 19, 2016
Joshua Pearce of Michigan Technological University shares the developments making 3-D printers less expensive and easier to use.
PwC: Joshua, can you please describe your market positioning in the 3-D printing ecosystem?
Joshua Pearce: Sure. My group and our work represent the open source trends in 3-D printing. We’re coming at it from the low-cost side and using global collaboration to drive innovation. Our printers in general cost less than $500, and either they are put together from kits or they print themselves. It’s called RepRap, short for self-replicating rapid prototype—a technology that was originally developed in Britain but has now spread around the world.
PwC: Do you think the lowcost printers will be in every home in the future?
Joshua Pearce: Yes, most homes. We’ve done some initial studies on the economics of using a 3-D printer at home. We looked at 20 household products that can be printed in fewer than 25 hours—so shorter than a weekend. Our analysis concluded that if the quality of the product was good enough, then owning a printer could offset the costs of purchasing equivalent products by somewhere between $300 to $2,000, depending on the objects and the amount of usage. Since 3-D printing allows a person to customize objects, I think we can go with the higher end of that range. There are now hundreds of thousands of similar open source designs of printable objects.
The economics get even better if we take into account some recycling. We developed RecycleBot, which can recycle some of the plastic waste as feedstock for an FFF [fused filament fabrication] printer. The RecycleBot takes milk jugs and other household plastic waste and turns them into feedstock for the 3-D printer. Commercial filament today is at least $35 per kilogram. If a person uses recycled plastic containers, not counting labor, the material cost is about 10 cents per kilogram, so basically free. This option is a win-win economically and environmentally.
A typical household can easily recover the cost of a printer in less than a year and also have a positive impact on the environment. These are all strong incentives for owning a 3-D printer.
PwC: What are the challenges to overcome before we see these in every home?
Joshua Pearce: There’s a long list of challenges. Currently, I’ll say that all the low-end printers—they don’t just work. That is, you need to have some technical competency. You don’t need an engineering degree, but you need to be able to tinker and work with tools and bolts and so on. These printers all need considerably more personal upkeep than people are accustomed to with appliances. The devices require constant tweaking to maintain the sweet spot for printing goodquality parts. That requirement is one of the reasons the popularity of 3-D printers is limited mostly to hobbyists and engineers today.
PwC: You talked about RepRap printers that use plastic as feedstock. What about metal printing on the desktop—are we there yet?
Joshua Pearce: There have been metal welding robots for some time. Last year we developed a RepRap that could print steel when attached to a MIG [metal inert gas] welder. The design source is the same: a geometrical STL [STereoLithography] data file. We are shooting current back and forth between a metal tip and a grounded metal print bed. The tip is melting and dropping small drops of metal many times a second as it builds the layer. Then we move the build platform in the vertical direction and repeat to create a 3-D metal object.
PwC: Today, printers take hours to print an object of any complexity and size. Is that likely to change in the future? Are we close to any physical limits?
Joshua Pearce: I don’t know over the long term, but I can talk about near-term potential. The Deltabot printers are at least twice as fast as the Cartesian printers, because they have much lighter heads.(1) Right now, only a couple of Delta-style printers are on the market. I predict that won’t be true much longer. The Delta printers will basically take over all the Cartesian printers, because they have some significant benefits, one of which is speed.
Also, as the low-end printers mature, they get features already in the high-end professional printers that can speed up performance. For example, the printers no longer need to decelerate around curves, because designers can take that into account in the software. I’m not sure what the absolute physical limits are, but there is clearly room for improvement from where we are today, even without going to fundamentally different physical processes.
“There is clearly room for improvement from where we are today, even without going to fundamentally different physical processes.”
PwC: What other features of highend printers should we see in lowend printers in the near future?
Joshua Pearce: The high-end printers have feedback control, and we should see that in lowend printers soon. They also use a different process to print metal—laser sintering— and the patents on that process are running out. So I would expect rapid innovation to occur in 3-D printers that use laser sintering, sort of what happened with the RepRap and FFF method. As soon as the patents expired and the open source community got their hands on FFF method, the innovation accelerated substantially.
Another feature is self-calibration. We want the printers to self-calibrate, rather than users calibrating the printers by hand, which I think is beyond what normal users are willing to do. If a user can just press a button and the printer knows where it is, that starts to move low-end 3-D printing into the realm of realistic large-scale deployment.
Innovation will also occur on the cost front, the simplicity of use, the ease of building the printer, and so on. Right now we can’t print in the micron resolution on high-end sintering machines, but that doesn’t mean that two years from now, I won’t have a powder or liquid printer on the desktop in my lab to do just that.
“I would expect rapid innovation to occur in 3-D printers that use laser sintering, sort of what happened with the RepRap and FFF method. As soon as the patents expired and the open source community got their hands on FFF method, the innovation accelerated substantially.”
PwC: A key limitation facing the industry is the ability to print with multiple materials. What are the future prospects on this front?
Joshua Pearce: I think that journey is just starting. In the low-cost printers using the FFF method, the printers have multiple heads, so those printers can use two colors or two materials— like a flexible plastic and a stiff plastic. The development on this front is also being held back by patents. For example, I know one 3-D printer company that uses two heads and faces the challenge that printhead two smears the work of printhead one, spoiling the product. The obvious solution to that problem is to have one of the heads move out of the way, or even have both moveable on the vertical axis (z-axis) independently. Despite this solution being obvious, my understanding is that such features are all covered by patents, which are in various stages of expiring. As these expire, we would start to see little lever arms or other mechanisms that move multiple heads and we will see many more materials used.
PwC: It seems that there are not many standards to provide confidence in the performance of materials?
Joshua Pearce: Indeed, the lack of standards is a significant issue that should be addressed so the industry can move forward. We did one of the first 3-D printing mechanical test studies—a study on the tensile strength of a RepRap-style printer. We asked a group of RepRap users to print with a standard and send us a print of the design of a dog bone. We used our printers and printed the same designs, too. Then we tested them all for strength characteristics.
Our tests indicated that different systems produced different outcomes even though they used the same material and design. So there is a third variable that must be accounted for: it’s not just the structure that someone is making or the material that someone is using, but also the particular printer that is used and how it is used. There is much more complexity than a machinist just using a laser to cut some raw material.
The good news is that a lot can be governed by software. I think there will be a feedback loop soon on even the lowend printers. So, for instance, it will ensure that the actual temperature of the plastic coming out is hot enough to get the type of curing wanted. I think this feedback loop will arrive in the next two years.
“The lack of standards is a significant issue that should be addressed so the industry can move forward.”
PwC: Another area of great interest is printing complete systems. You have a project to print electronics and conductive material. What are you trying to make?
Joshua Pearce: My big goal is to be able to print an Arduino [a microcontroller board], which is the brain behind the 3-D printers and basically behind all the scientific equipment we use in the lab—our open source lab.
For example, we have a handheld water tester, and right now we’re still buying a $20 Arduino and simple circuitry to control it. If we could print the Arduino board in the case of the tester itself, we could make the whole setup a lot smaller. Also, we could customize and take away parts of the Arduino board we don’t need. We could make a simplified Arduino-compatible electronic microcontroller for ourselves and then just drop in the chips. Already our open-source water tester replaces a $2,000 tool with a $50 printed one. When we are done, the overall cost will be less than $20 for a scientific device that can replace several different $2,000 to $5,000 proprietary tools.
PwC: As you know, Moore’s law was a great way to understand and predict the growth of the computer industry. Is there something similar for the 3-D printing industry?
Joshua Pearce: I think one of the best examples of a catalyst for 3-D printing is the exponential growth of free and open source digital designs. Right now, a major limiting factor for a normal consumer is that they don’t know CAD [computer-aided design]. Many companies are working on simple and easy-touse software tools to create designs. In reality, for most things customers want something really well done and professionally designed.
Hundreds or maybe thousands of these professional people are sharing their designs.
We tracked the number of designs that are available for free, and they are growing exponentially. There are already several hundred thousand designs that any user can download and print, and we are surely headed to millions. Whether a person owns a printer, has one at school, or rents time on one, there is a good chance that what someone wants to print is already designed or close to it. They can then customize and make it personal or better, but the basic starting point already brings them very far.
Every new design adds more value to owning or having access to a 3-D printer—so it is pretty clear we are just at the start of a fundamental shift in the way production works.
1 Cartesian printers use the three dimensions of x, y, and z. The printhead moves in the xy horizontal plane, and the build platform moves in vertical dimension z. Delta printers use parallelograms in the arms, so the printhead moves in all three dimensions—like the end of a pick-and-place robot.