Manufacture It Yourself
The nineties promised us desktop everything. In the new millennium, that promise is really starting to deliver. One of the most exciting developments bringing power to the individual home office or workshop is the "desktop fabricator," a sort of printer for 3-dimensional objects.
Science Fiction writer Bruce Sterling has been raving about developments in this area – nicknamed "Fab Labs" — for about seven years. But frankly, I could never quite wrap my head around it as an already existing phenomenon, and so I didn’t give it much attention. But when I heard about Evan Malone’s open source Fab@home project to "freely distribute designs and software that allow anyone to build and operate their own open-architecture, multimaterial desktop Standard Freeform Fabricator" (SFF) it was clearly time to put an end to my own ignorance by talking (via email) to the man directly.
Malone worked in applied physics at Fermilab in Batavia, Illinois as part of the "Proton Driver " proton synchotron conceptual design team. As a Masters student at Cornell, he worked on Cornell’s World Champion Robocub Autonomous Robotic Soccer Project. His current work on SFF is part of his Doctoral research in Mechanical Engineering, also at Cornell.
h+: Call me dense, but I’ve read about Fab Labs before, and when people talk about it, they don’t seem to talk about the physical materials you need to have at hand to "print" a 3 dimensional object. I’m always left with this futuristic nanotech vision where you punch in the code and the machine sort of builds you an object.
So tell me a bit about that aspect of it. Do you have a machine with the components of a large variety of material objects and then you send it your info, or do you feed it the materials and then it organizes it for you or what?
The idea behind Fab@Home and multi-material freeform fabrication in general is to make a single machine which can go directly from level 2 (raw materials) to level 4 (finished products) – an automated fabrication and assembly machine
EVAN MALONE: Traditional manufacturing can be decomposed, roughly, into a pyramid of 4 types of processes:
- Extraction: Raw materials are extracted from nature or through recycling.
- Refinement: Raw materials are processed and formed into stock material – billet, rod, pellets, gases, liquids, etc.
- Fabrication: Stock materials are fabricated into parts – carefully formed shapes of a single material.
- Assembly: Parts of different types and materials are assembled into products.
The idea behind Fab@Home and multi-material freeform fabrication in general is to make a single machine which can go directly from level 2 (raw materials) to level 4 (finished products) – an automated fabrication and assembly machine. Fab@Home uses a simple syringe pump technique to deposit materials. Refined raw materials are supplied to the machine in a pure form or formulated as “inks” which can be readily dispensed by the machine from a disposable syringe.
The basic workflow is as follows:
- You decide what you want to make, design it with some sort of 3D design software, and export it as a common file type (STL).
- Using the Fab@Home software, you import your STL file(s), assign “material properties” to them, and then the software processes the model into a machine executable manufacturing plan.
- You load the machine with syringes of the appropriate materials and command the machine to start.
- The machine moves the syringe along paths as material is dispensed, laying down strands of material, layer by layer building up the 3-dimensional shape of the object
- If the object involves multiple materials, the machine automatically switches between multiple mounted syringes, or asks you to load the appropriate syringe.
- You may, optionally, manually drop in parts that are traditionally manufactured – e.g. an LED, IC, or battery, during the course of building, and embed them within an otherwise “fabbed” object.
“Material properties” are parameters that the system uses to control the flow of material as it is dispensed. When you are trying out a new material, you need to have the machine make some test patterns, then use these to adjust the parameters for the particular material. These parameters are stored for future use, and you can then choose to apply any stored parameters to any object.
h+: Would it be fair to say that this is for hobbyists at this point? When do we see home manufacturing on a reasonably broad scale?
EM: Interestingly, Fab@Home is working well for a couple of different groups – yes, hobbyists are playing with the machines, but secondary school and university educators are using them to teach design and manufacturing, and researchers are using the machines to explore new means of controlling the shape of materials for biological tissue engineering, handicapped assistive technologies, and robotics. There are more than 130 machines around the world; we’re still waiting to see what most of the users will do with them. I suspect that home manufacturing will gradually creep into being over the next decade. I expect that the demand for the technology will be in the art studios and hobbyists garages, for making replacement parts and for inventing new art forms and devices, in the research laboratories, and also in the kitchen for dessert making and decoration. I suspect that the hobbyists, artists, and researchers will start to show some impressive achievements with the technology, using it to do otherwise impossible things, and this will drive new R&D and investment in fabber technology. There is a solid decade of work required before the appropriate set of raw materials is developed and the control software intelligence is developed that will allow home users to build really exciting products.
h+: Please say a bit about the degree to which this might be putting manufacturing power into the hands of individuals, and conversely, is this a threat to any currently existing industries?
EM: On its face, Fab@Home is exactly about putting manufacturing power directly in the hands of individuals. However, I don’t think personal fabrication is ready to compete with traditional product manufacturing quite yet. Practically speaking, we have hoped that the Fab@Home project would increase public awareness of rapid prototyping / 3D printing, and would build a hacker community interested in improving the technology. Both of these things seem to be happening, to our delight. I suspect that there will be a performance/price threshold for personal fabrication technology, hopefully achieved via open-source effort, after which users will be able to focus on product designs without having to worry about babying the printer. At this point, creativity, competition, and collaboration will drive demand for ever better fabbers. Until then, these systems require some skill and patience to work with.
After this threshold (hopefully 3-5 years), I suspect that the technology will primarily be used to develop new types of products, rather than competing directly with existing manufacturing. On the other hand, the advantages of Fab@Home’s multi-material fabrication relative to traditional manufacturing are packaging freedom and total customization – blending function and form to a far greater extent than possible with traditional manufacturing, and allowing each product to be unique. I see wearable devices, prosthetics, hearing aids, implants, custom battery packs, surveillance devices, and also tissue engineering as the applications for which fabbers will be better suited than traditional approaches.
h+: Some of us first heard about the FabLab idea via Neil Gershefeld’s project with the Center for Bits and Atoms. Are you associated with that project? Competitors? Are there differences in ideas or realization that you could share with us?
EM: I have been involved in Prof. Gershenfeld’s FabLab project for a few years, and Prof. Gershenfeld is partly responsible for our creating the Fab@Home project. Prof. Gershenfeld and Prof. Lipson (co-creator of the Fab@Home project) realized that personal fabricators might be a very powerful tool to include in a FabLab, and might even be built using the facilities of a FabLab. We were invited by Prof. Gershenfeld to show our very first Fab@Home prototype at the opening of a FabLab in South Africa. Prof. Gershenfeld envisions and is pursuing an implementation of personal fabricators built entirely using the technologies developed in the FabLabs – the technologies include: ATTiny microcontrollers, Internet communication protocols, their own motor driver designs, and more. We are continuing with the COTS approach because I believe that we can meet a lower price point and a lower skill point for the end users more readily this way. Still, we continue to share ideas and meet annually.