Fabbing Among The Stars
To What Extent Can we Rely Upon Computers?
Yes I do believe star travel is possible someday if very, very difficult and expensive. I doubt however we’d bother to establish colonies on extrasolar planets if those planets weren’t a good deal more like earth than most of the real-estate we’ve found in our own solar system. A fabber meant to operate on an Earthlike world therefore would be less likely to be used in hermetically-sealed pressure structures then one used on the moon and would more likely turn out agricultural tools, biomass processing equipment, air breathing engines and from whatever the local equivalent of trees might be, structural units. We’d also want simple yet versatile and fairly powerful remote-control or pilotable mobile machines to dig trenches, lift rocks, position beams, transport ore, scout terrain.
Any fabber able to turn out all of these machines either wholly or piecemeal must be able at minimum to build up solid components from electrically conductive elements, electrical insulators and ferro-magnetics, aluminum, silica and iron or steel would be good examples of each, respectively. The fabber would also need to be able to introduce controlled amounts of impurities or do pants into silicon wafers.
Before we go on let’s take a look at what kinds of things we’d be likely to need within the first few days, weeks, months after landing on a planet orbiting another star. Right up there with food, water and shelter we’ll want energy right away. We might arrive with handheld fusion torches or pocket-sized antimatter reactors but it could happen that such devices would require a larger technical infrastructure to manufacture than could fit inside a starship of a few thousand ton capacity. Photo-voltaic cells are both comparatively simple and within the envelope of a reasonably good fabber. Since we’d be unlikely to mount a colonization expedition on a planet without breathable air, water in quantity and visible light we should be able to use solar cells there. You need some sort of back-up for dark side hours either batteries or some type of heat engine generator operating on stored thermal energy or chemical fuel. In order to get direct use of our solar cells we’ll need electric motors. Anything that can fab a motor could also do a good job on generators, piston engines, batteries and the essential parts of wind turbines. It should also be able to manufacture it’s own constituent parts.
Besides the economy of sending an all-purpose manufacturing device with a colonization expedition, such a fabber would also provide some cultural security for the descendants of the expedition. Should future generations lose the technical knowledge brought from Sol System by their progenitor, yet retain the skill or lore of turning out copies of established components from fabbers, the culture could retain a level of technological competence which could allow it to conserve resources and avoid the terrible feast-starve annual cycle so common inaboriginal cultures. It’s my position that a Native American or an Asian nomad would be able to understand, given the chance, an internal combustion engine or a Stirling cycle and could learn through inspection and experience, how to make use of an electric motor, light or radio. It’s less likely at least to me, that the wherewithal to manufacture computer chips and program digital systems could be understood by tribesfolk ignorant of algebra and integers greater than twenty.
What alternative might there be to the digital computer for running something as complex as a fabber? To answer this question we might review the history of television. Before the digital age a great deal of electronic communication went on without anything very much resembling a computer. timing circuits and matrices of photo-electric dots caused moving pictures taken one place to appear on a screen somewhere else and it was all done with technology possible around 1920. You couldn’t record a TV show in those days as we do now but you could save the film that was broadcast over the high frequency FM waves.
There is an analogous situation in which solid objects can be copied if not manipulated in all the ways possible with a computer file. Most fabbed artifacts are generated under the direction of a program drawing information from a 3-D CAD file, translating data into substance. The files themselves may be drawn on the computer with graphics programs. They may be generated by scanning a solid object. They may be generated by a combination of the two techniques, a model of some kind which is first scanned then tweaked on the computer to introduce some designer-conceived refinements. We could take a solid object and scan it “destructively,” using a laser for instance sweeping back and forth, incrementing from forward to back, using very brief pulses to tick off bits of object. A detector would tell with each laser pulse if some solid has been burned away and if yes, a signal would be sent to a fabber device as a command to deposit an increment on a new build. In this way a new artifact is being built up as another one is destroyed. If we have two fabbers hooked up to the scanner we can produce two copies while one is being destroyed. This seems a bit wasteful of material but it relies upon technology which is fairly macroscopic and much less fragile than modern computers.
Inquiring a bit more deeply into how a system of this sort might work, we’ll consider the mass spectrograph. This device is used to analyze the composition of a chemical sample by ionizing it within an electric field and detecting the amount of each element which falls in various regions on a plate or other deposition surface. Each atom when disassociated from others with which it was formerly joined, now existing in the ionized state has a certain charge corresponding to the number of electrons which have been torn away, I.E. One, two, three etc. and a characteristic mass as well. The electric field can pull upon the atom with a force proportional to the charge of the atom and the amount of acceleration thereby imparted to the atom is in turn proportional to the atom’s mass. (From knowledge More A) Between the strength of the pull and the inertia of the mass any given type of atom will consistently travel a particular distance and come to rest at a given spot in an electric field of a given strength.
In order to sort out all of the atoms in a large object we’d need to spend a lot of time since atoms are very small and there are so many of them. We could however destructively analyze the composition of a large object by taking a tiny point sample, then burning away a chunk, point sampling again and so-on, throwing most of the mass away. Again this sounds wasteful and hopefully we’ll develop ways of recovering streams of vaporized or ionized material to be recycled but there’s perhaps an easier way to go about the problem without wasting much valuable material.
Though we need a solid object to break down, one which has as many separate constituents as the object we want to fabricate and use, the model need not be made from the Same constituents. If we’re building an object made of copper, iron and silicon for instance we might use a model built up from droplets of molten sodium carbonate, zinc oxide and calcium hydroxide or any of these or other reasonably innocuous compounds with small amounts of different additives in each, to represent distinct materials. As the model is destructed the mass spectrometer would detect the additive molecules while the calcium, sodium potassium zinc, whatever could probably be trapped magnetically or just run through a long narrow metal channel from whence they might later be recovered, reprocessed and separated.
In copying a needed component or device we’d make one or more real working copies and at least one “model” copy from dummy materials. As long as the increments in all of the models are the same size and as long as the deposition units for each fabber uses the same material each time a given material in the real model is laid down, it won’t matter to the system if it’s making real components or dummy models.
Besides conductor, insulator and ferromagnetic for an electrical device we’d also need a scaffolding material, something to support overhangs as cross sections expand from bottom to top ina given build. We’ll need a lubricant material to interface between separate components ina single build should we wish to generate integrated moving parts. We’ll need five matter emitters at least and we’ll know ahead of time what sorts of materials we’ll be using so the model materials can be selected accordingly. I suspect for actual components we’d use something like the fused deposition system the vacuum chamber laser or electron beam devices which might use either wire or beads of molten material so we can build up layers each comprised of two or more elements. Alternatively we might use a powder-based technique with the materials laid down like different inks in a multi-color photo-copier, by means of charged rollers or an inverted pallet.
A destructive master-based fabbing system would differ from present techniques in another important way. In most systems we use today, we have the entire design file available to us so we can build the bottom layer just as it’s seen in the computer graphic file. With a destructive technique we’d pretty much need to tick off the top layer on the master letting it become the bottom layer of the copy which will be built upside down with respect to the master or the master must be upside down and suitably supported from above which comes to about the same thing. The point is, the master must be designed to such a way as to be able to render an inverted copy. The most conceptually simple way to do this is to form a structure of some throw-away material around the master so if you turned it upside down it would have a flat surface on which to rest. then we’d arrange for the same or another cheap material to be used as support structure for the copy which would emerge upside down, with other support material as needed, able to be separated from the support structure which material could be used over and over again. In this way a bit of semi-skilled artisanry would allow an essentially blind automaton to copy one artifact from another.
Our automaton must at simplest level move a 3-D positioning system three of them at once actually, master, copy and dummy, increment by increment, along a line then over to the next line and eventually down to the next layer and so on throughout the solid object.
With each increment a signal from a particular region of the mass spectrometer would trigger the appropriate emitter to fire. These are not trivial operations but within the capability of early 20th-century television technology possibly with the exception of laser, electron beam and deposition techniques.
Chief among the things the fabber will turn out early on, will be parts for other fabbers and they will never cease to do so. Over a number of generations the colonists will encounter “drift” meaning the tendency of successive copying from one edition to the next and so to the next and on and on to accumulate mistakes, divergences, change from the original. In order to prevent this I suspect an original computerized fabber will be used to create reference models for as long as possible. Even if this Seed System breaks down entirely or is destroyed; drift can be kept at a hopefully acceptable minimum by hooking a fabber to three or more controller devices. With a fairly simple set of circuitry the replicator could average the distance or time increment dictated by the controllers and deposit a dab of matter or an energy pulse when two or more of the controllers sent simultaneous commands. It makes the electronics a little more complicated but we’ll be able to fab capacitors, vacuum tubes, potentiometers, timing circuits, probably transistors.
I suspect we could with 22nd-century technology transplant a self-sustaining self-replicating technology at about the level of 1930 or so using fabbers controlled with sensors and circuitry common in that era and producible with the fabbers described here. In this way a manageable cargo sent across interstellar distances could continue to turn out household utensils, agricultural and carpentry tools, engines, stoves, radios, powered articulated limbs indefinitely. The transplanted technology would be limited in what it could produce but colonists need not build sky scrapers, super tankers and intercontinental ballistic missiles at least not for a few generations after landfall. On one or more continents rich in mineral and biological resources a comfortable way of life could be achieved in less time than it took the English Colonists in Virginia and Massachusetts Bay. Perhaps pioneers to other stars will in time set aside the size-limited design, frozen and admittedly cumbersome fabber-based industry and go after a means of doing business resembling more the industries of the 20th Century than those which took them to the stars. If so, we’d hope they’d build from the ground up with one eye on industrial pollution and the other on overpopulation but we can only hope. It could be however that most of the settlers will embrace the sense of independence a fabber for everyone could impart and they may decide to spread outward at a modest but sophisticated level of technology rather than reaching for the sky or reshaping the face of their new planet. Whether we could realistically expect a band of Lakota or Apache tribespeople to transport fabbers and the mineral extraction equipment needed to feed them across the plains or desert on painted ponies, a band of Gypsies with caravan wagons or a Viking ship plying strange coasts might be believable. A civilization once, starborne might opt for a more pastoral existence than one bound by cities or farmsteads and isn’t the urge to explore and seek beyond the next horizon what brought them here in the first place? A nomadic band or agrarian village would require a technical priesthood to carry out the more opaque activities not readily accessible to the average community member but so it his always been to one extent or another from the herbalist or midwife of the Paleolithic Age to the engineers and technicians of today.