Saturday, November 14, 2009

A Place In Fabland?

Several months ago, I wrote a series of posts discussing small-scale manufacturing as part of a strategy of adaptation to economic collapse due to Peak Oil. My position then was that small-scale manufacturing would primarily be employed to make the simple low-tech tools needed for a much simpler life. In this role, it would enable us to continue to have reasonable access to things such as hand tools and bicycles. I had not seriously considered small-scale manufacturing as a means of maintaining widespread access to the gadgets that define modern life in advanced industrial society. But that was before I knew much about the global community of “fabbers.” As I wrote my earlier series, I devoted a small amount of space to the fabber phenomenon, but I didn't have time to do it justice. A couple of news articles over the last month have caused me to turn back again for a more complete exploration of this subject.

According to Wikipedia, a digital fabricator (also known as a “fabber” or “fab”) is basically a “small, self-contained factory that can make objects described by digital data.” According to many enthusiasts, fabbers have great potential for democratization of the means of production in industrial society. This is because of the following advances:

  • The invention of small, inexpensive machines capable of producing three-dimensional parts

  • The digital definition of three-dimensional part manufacture as an act of three-dimensional printing

  • Increases in computational power of consumer electronics, including PC's and printers

  • And lastly, the invention and widespread availability of new materials that can be easily formed, machined and “printed” into parts, in ordinary, non-clean room environments.

All of these things are now being combined into machines that promise to do for manufacturing what cheap and powerful consumer electronics have done for media. As powerful and inexpensive consumer electronics have combined with the Internet to turn everyone into a potential creator of entertainment or news or art, so the digital fabber revolution promises to turn everyone into a potential creator of useful manufactured goods. Just as the consumer electronics revolution has weakened the power of traditional producers of media, so the fabber revolution has the power to displace traditional, capital-intensive, large-scale manufacture of goods.

Thus some fabber enthusiasts tout these machines as technological miracles that will enable every garage to be a high-tech small-scale manufacturer of high-tech products. These devices are put forward as the definitive answer to our present economic collapse, and the key to continued prosperity over the long haul. But are they all these things after all? Are they any of these things? If fabbers are the miracle that their enthusiasts claim, this leads to a near-term future that potentially looks quite different from the darker future of enforced simplicity and technological retreat envisioned by many collapse-watchers.

What role will the fabber revolution play in the near-term future of industrial society, particularly in the First World? How will the fulfillment of that role affect our society farther on, over the next several centuries? Do fabbers have the potential to preserve widespread access to highly advanced manufactured goods? Or are there limits on all advanced industrial activity that ensure a collapse of industrial society? I'm sure that everyone has their own, instinctive, gut-level answer to these questions. Yet such gut-level responses must be examined to determine whether they are fact-based or merely faith (or sometimes, wish) based.

I don't know that I will be able to offer a definitive answer to these questions. But I thought a good starting point would be to lay out what we already know about fabbers, and to put forth pertinent questions that would have to be answered in order to accurately define the true potential of fabbers in dealing with our present collapse. My observations and pertinent questions are listed below, in outline form:

  1. What can fabbers make now? (These are things whose manufacture has been reliably and repeatedly demonstrated.)

  • Gross machine parts made of plastic and some metals

  • Rudimentary control components, such as “printed circuits”

  • Objects d'art

  1. What things can fabbers not make now?

  • Ultrapure microelectronic substrates (that is, substrates made from inorganic materials like silicon)

  • Inorganic microelectronic circuits (maybe a fabber will be developed that can do this, but it requires creating ultrapure “clean room” conditions inside the average Joe's garage)

  • Other fabbers. (They can make most of the machine tool parts, but they can't yet make the microelectronics used in control of fabbers.)

  • Note: if fabbers are only practically useful when they have great computing power (needed for rapidly fabricating complex parts in 3D), then one won't be able to use a fabber to build another fabber until a fabber can also produce all of its own control circuitry and microprocessors.

  1. What will fabbers will need in order to be self-replicating (or build their own replacements), with present-generation computational abilities?

  • Feedstocks of ultrapure materials

  • A source of electric energy

  1. Ultrapure inorganic materials as a restricting condition

  • Energy, Silicon and the Siemens process (and other processes). (All processes now used for purifying silicon and associated dopants, and combining these materials into appropriate semiconductors, require large amounts of energy. As access to fossil-fuel energy declines, these processes will become increasingly expensive.)

  • Other microelectronic ingredients, like dopants, are increasingly scarce

  • Less-pure forms of these materials are less and less remarkable, until in the limit, they are no more remarkable than the natural states of these materials. Useless for high-speed electronics below a certain level of purity. (Example: a galena crystal, commonly found in nature, can be used to build a crude AM radio receiver. But it takes much purer materials to build high-speed, high-performance microelectronics.)

  • If energy is the limiting factor in producing these materials, energy is a limiting factor in a “fabber” revolution.

  • Has any work been done in recycling microelectronics, beyond simply reclaiming the metals used in them? What is the energy cost in extracting and re-purifying the silicon, metals, dopants contained in an IC or larger chip?

  1. Question: In a resource-constrained future, can suitably fast microelectronics be printed using less exotic materials? Can these be easily programmed to provide the sort of production control currently exercised by the electronics in today's fabbers?

  • Answer: polymeric organic semiconductor materials are being developed for use in possible thin-film, printable microelectronics.

  • Transistors and integrated circuits have been made with these organic semiconductors. How fast can they be made to operate? Current silicon-based MOSFET's can be switched at speeds well over 1 GHz (one billion cycles per second). Can organic transistors and microcircuits be made as fast? At present, they are not. (Circuit speed is a factor in processor speed, and thus in the speed with which a fabber controlled by such a circuit can turn out complex parts.)

  • Are there impending resource limits on organic semiconductors?

6. In the spirit of the Precautionary Principle, are there any moral or ethical or other downsides to the fabber revolution? Are there potential negative outcomes or uses of this technology that haven't been widely forseen?

Anyway, those are the points for consideration that I was able to think of in a short time. Over the next several weeks, I may try to take a stab at a few of them, as time allows. Unfortunately, time doesn't allow this weekend, as I worked a bit extra on Friday and I have to go in again tomorrow for a few hours. If anyone else wants to take a stab at tackling these questions, feel free.

For further reading, check out these links:

No comments: