(Written by Eddie Krassenstein with input, dialogue, images and recommendations by Neil Jansen)
We all know what consumer level 3D printers are capable of. They can print out objects using a handful of non-conductive materials. If you need that cool looking plastic earring, or that completely customizable iPhone case, or that unique little toy for the little-ones, 3D printers have pretty much conquered the task of creating these. With this said, however, many futurists envision a world where 3D printers will be capable of much more. One day, perhaps an entire laptop or MP3 player could be manufactured in the comfort of your own home, on a machine not much bigger than a 3D printer. This used to be a dream, but one team is taking a step in the right direction.
The story begins with Neil Jansen and Karl Lew. Many years ago, these hardware hackers were building electronic projects in their homes, and were struggling to figure out a way to manufacture small runs of boards for friends, family, and other interested people on the internet. Putting them together by hand required a stereo microscope, steady hands, and a lot of patience. Factories that build electronics were out of the question, because they only build in large quantities. It would take them longer to program the machines and set them up than it would be to run a few boards. So the economics only work out if the quantity goes up, or the amount payed for each board increases. It was not looking good.
As the 3D printing revolution happened, these two men realized that their problems could be solved by using the same technology that makes RepRap printers possible. Desktop fabrication was a cool buzzword and a noble problem to devote their lives to. As it turns out, electronics manufacturing was actually one of the original goals of the RepRap project. Adrian Bowyer, the founder of RepRap, originally envisioned a machine that could replicate all of its parts. That means 3D printing its frame, gears, couplers, and even the electronics that control the machine.
In the early days, 3D printing the plastic parts presented a few problems, but they were easily solved through the scientific method and lots of trial and error. Automated assembly of electronic circuit boards, on the other hand, has always been a very large and complicated problem. The RepRap team was so successful at Fused Filament Fabrication, that they forgot all about the original goal of electronics replicability. The FirePick Delta team has picked up where they left off. They have broken the problem down into very small pieces and attacked them one by one, methodically, and are now very close to meeting one of Adrian Bowyer’s original goals of the RepRap project. This also brings them closer to their own goals of being able to make smalls runs and prototype electronic boards, for business and pleasure.
Their machine uses a Delta configuration, instead of the more conventional Cartesian configuration. There is a new wave of 3D printers that use the Delta configuration, and for good reason. They’re faster and cheaper than conventional machines, and are much easier to build. The astute reader will notice that it uses the original “rotational” delta design per the now-expired Delta patent, rather than the “linear tripod” design that the current 3D printer Deltas use. This allows for a rectangular frame instead of a triangular or round frame, at the cost of some Z travel. A fair tradeoff for this machine, that works primarily with rectangular boards and rectangular parts.
There are quite a few challenging problems that are faced when trying to make a machine assemble electronics:
- Placement: How do you pick the electronic parts up, and put them down safely? Normally a vacuum system is used. But the parts also need to be rotated precisely, because parts are oriented at different angles on the board.
- Tool changes: Assuming that a vacuum nozzle is needed to pick up the parts, then many sizes will be needed because the parts vary from fractions of a millimeter up to around 50mm. There are also other tools needed, like dispensing solder paste and of course, a 3D printing attachment for making fixtures
- Accuracy: Is it accurate enough to place the electronic parts?
- Component Feeders: How you do get the parts into the machine?
- Soldering: How do you solder the parts?
- ESD Safety: Plastic and acrylic parts build up a nasty static discharge, and that’s something that can kill semiconductor parts. The FirePick Delta machine will be 3D printed from ESD-safe conductive plastic that uses carbon nanotechnology to provide a safe electrical path to ground, thereby preventing any ESD buildup.
- Computer vision: How do you recognize parts and features on the board, and make corrections during part placement?
- Software: How do you tie all of these pieces together? How do you make them connected? The FirePick Delta software runs on a Raspberry Pi, and can be controlled from any web browser, including tablets and iPhones. It also allows multiple machines to be networked together, and allows them to work cooperatively.
- Cost: Finally, how do you do this cheaply? Commercial machines can cost more than a Ferrari.
The FirePick Delta team has successfully solved each of these problems. Now they’re trying to build their final prototype and will spend a while refining it. In the meantime, it works as a 3D printer, and they will be using it to print prototype pieces for their own machine out of ESD-safe conductive plastic.
“The first version of the machine is mainly intended to either 3D print or assemble circuit boards, which are currently separate steps,” Neil tells 3DPrint.com. “So you could print out an enclosure, and then assemble a circuit board, and then hand-assemble the two with a screwdriver.” “It’s possible that someone in the future will figure out how to do [both operations] at the same time, using a mixture of conductive and resistive materials, but currently I do not believe the technology is there, at least at the hobbyist level.”
His machine is capable of doing two of the most important and difficult tasks in the process of assembling working electronic devices. While it’s not quite to the point of simply being able to press a button and have it print out an entire working device on its own, it certainly shows us how close we are to one day having a machine that will print and assemble working electronic devices at the touch of a button.
Neil plans on building this machine, and making it capable of several tasks that are difficult or impossible to perform by hand. The FirePick Delta will feature an auto tool changer, that will be able to perform multiple tasks. Here are some of the tools that Jansen envisions on his machine:
- 3D Printing Hotend – Like that of your typical delta style 3D printer.
- SMT Pick-and-Place vacuum nozzle – Provides the ability to pick up tiny electronic parts (SMT) and place them in a particular location on a circuit board.
- Solder paste dispensing syringe – This will allow the machine to place solder paste consistently on the SMT pads
- Hot air rework – Allows for the reuse of of expensive electronics boards by slowly melting solder, to allow for removal of old chips. With the right software, it also might be capable of soldering the entire board, without the need for a dedicated reflow oven (Neil currently uses a toaster oven to melt the solder).
- End mill / drilling attachment for PCB milling and other light-duty jobs.
- In-Circuit programmer for Atmega processors – Fully-automated Arduino-Compatible bootloader programming
- Pen plotter
- Dial indicator for verifying Z leveling
- Laser element for photoplotting photo-sensitive boards
They have designed the software and hardware to be more of a framework, so that others can come along and add functionality that previously didn’t exist. Their computer vision framework has been written in such a way that even a liberal arts major could write an application, to say, sort a pile of parts by color and size. It’s all done through REST APIs and JSON notation, which any web developer weenie would understand. You don’t have to be an EE graduate from MIT to develop applications, or to use the machine.
“We want to make electronics manufacturing easy, for people that need custom electronics, but have a focus or passion somewhere,” Neil explained. “For example, a musician wanting to make or modify an electronic instrument, or a video game programmer who wants to make a new type of game controller, or a pilot with an idea for a new UAV platform. When you realize that most creative ideas are from people in a unique field that are trying to apply skills from a different field.. then you see that having a method to take an idea from concept to reality in a rapid way is really empowering.”
Empowering it is, and empowering it could be in leading us to a more self-sufficient society that is capable of making the things we need on demand, without depending on others. Neil truly believes that the third industrial revolution is here now, and there will be a technical renaissance because of the influx of rapid prototyping machines in today’s world. “Having plastic and metal pieces are not enough,” he explains. “We need to be able to ‘rapid-prototype’ the brains [of objects] as well.”
Neil and his team plan launching a Kickstarter campaign for the FirePick Delta, hopefully by Q2 of 2015. Currently for the barebones machine, it costs Neil approximately $300 (without accessories). However, he believes that he can bring the costs down by lining up parts to be sourced from China.
“I believe that I can offer a barebones kit for $300 and not go broke,” he tells us. “Assembled kits would be considerably more because we have to build and support them. We’ll also offer machines in the several-thousand dollar range to small businesses and makerspaces. They’ll be built better, will include all the accessories, and will come bundled with real tech support and other perks that they would need. This product will help small business in the USA that want to make their stuff locally rather than going to China. This will create jobs and prevent intellectual theft by factories that counterfeit American products after they get the source documents and plans.”
He also told us that he believes that this machine would work well in Chinese factories because they are cheap and capable, and can work in parallel to manufacture things, an approach that has only recently started to pick up steam. He truly believes that the FirePick Delta has the potential to really make a huge impact, and that it could cause an industry shakeup, where older companies might end up dying out, or be forced to regroup, as newer companies come about. He compares the potential to that of the
RepRap movement of DIY 3D printer builders, which has caused larger corporations such as Stratasys to change their courses.
While Neil’s team is working on this hardware solution, they are also working on a Thingiverse-like ecosystem for the sharing of electronic projects
“We’re trying to create an ecosystem that better handles the source files of an open-source hardware project,” he told us. “Kind of like a Thingiverse but with focus on making sure that the files are stored in a relevant way, so that anyone can build it by loading it into their machine without a lot of fuss. It would also source the components for them, and also make it easy to branch and merge existing projects in this format. We believe that since every open source project can be built on other open-source projects, this will lower the barrier to making wonderful things without re-inventing the wheel.”
What do you think? Would a machine that can basically build an electronic device from the ground up be something you’d purchase? How far off do you think we are from being able to develop full working electronic devices on a 3D printer-like machine? Discuss in the FirePick Delta discussion thread on 3DPB.com
Be sure to check out the video where Jansen explains this project in more detail below: