3D printing circuit boards, photography-resistent clothing, and wireless DNI.

Feb 13, 2015

Now that I've had a couple of days to sleep and get most of my brain operational again, how about some stuff that other parts of me have stumbled across?

Building your own electronics is pretty difficult. The actual electrical engineering aside you still have to cut, etch, and drill your own printed circuit boards which is a lengthy and sometimes frustrating task. Doubly so when multi layer circuit boards are involved because they're so fiddly and easy to get wrong. There is one open source project that I know of called the Rabbit Pronto which is a RepRap print head for fabbing circuit boards but it might be a little too experimental for the tastes of some. This constitutes a serious holdup to people being able to fabricate their own computers but that's a separate issue. Enter the Voltera, a rapid prototyping machine for circuitry. Currently clocking in at $237,061us on Kickstarter and still going, the Voltera isn't quite a 3D printer in that it doesn't seem possible to fabricate circuit boards completely from scratch with it, you still need a static baseplate. However, what the Voltera does do is lay down successive layers of conductive and insulating inks on top of the fibreglass board until your entire circuit has been printed out. If surface mount technology is how you roll (and that's increasingly the only game in town) you won't have to worry about drilling holes for components' leads but there is nothing preventing through-hole designs. The firmware is designed to accept industry standard Gerber files so users aren't necessarily tied down to any one CAD package. Even more interesting is that the Voltera includes a solder paste head, so after the board's done it'll lay that out for you as well so that components can be positioned appropriately. Additionally, the bed of the Voltera implements reflow soldering, which means that after the components are positioned the temperature can be slowly raised until the solder paste cooks down and solid electrical connections are made - no more toaster oven. All but one of the Batch-2 runs of the Voltera are spoken for already so if you really want one you'd best jump on it, else you're going to have to wait for them to go into general manufacture.

Privacy runs fast through our fingers in the twenty-first century. If it's not security cameras on the street recording everything and everyone walking by. If it's not securicams it's drones (public and private sector both) on surveillance runs. If it's not drones, sometimes it's people with cameras and smartphones photographing people who really don't want pictures taken (cases in point the photography policies of many hacker cons). In other words, paparazzi are no longer a problem exclusive to the rich and famous. Enter Steve Wheeler of Betabrand, a company that crowdsources and lets people vote on clothing designs as its think tank strategy; projects with good prospects enter a crowdfunding phase so early adopters can gain access to them. If something does really well, the something goes into mass production. Their latest project (which is doing surprisingly well) is called Flashback - anti-photography clothing that reflects so much light into the lens that only the clothes can be seen. Flashback clothing works the same way as the high-visibility vests and strips that urban bicyclists wear by using glass nanospheres bonded to the fabric itself to form what amounts to a flexible, highly reflective surface that refracts as much light as possible. Currently there are only four pieces, a hooded jacket, a scarf, blazer, and trousers but depending on how things go the clothing line might grow. The Wired article I've linked to has a couple of "during the photograph" pictures but their crowdfunding page has execellent before-flash/after-flash pictures. There is some skepticism about how well they actually work (especially from professional photographers) but after reading a bit about the theory it seems sound to me, and I'm considering rounding up all of the reflective strips my cow-orkers wear to do a couple of "Will it or won't it?" pictures over lunch as an experiment. If exotic clothing is your thing you might want to keep an eye on this brand, though you'll pay close to designer prices for their wares.

The slow and steady march toward direct neural interface - creating a bi-directional link between the brain and computer hardware - proceeds apace. In 2011 Dr. Eberhard Fetz was given a $1mus, three year grant to advance his work on implantable neuroprosthetics. Now we have the CerePlex-W, an implantable neural activity receiver which wirelessly transmits its data to nearby computer systems which can act upon those commands. Currently it's on sale only on the research market for use with simian test subjects, but the Braingate Consortium is in talks with the US FDA to begin clinical human trials some time in the near future. The CerePlex-W is a wireless device broadcasting at 30 milliwatts of power so it can be picked up just a meter or two away, yet it's able to transmit data at a speed of 48 megabits per second, princely bandwidth for broadcasting the activity of the cerebral cortex indeed. Whatever is connected to the receiver can use the command signal however it wishes, from manipulating a cursor on a screen all the way to... that's a good question. Entering characters? Driving a wheelchair around? Using a robotic arm to move stuff around? The mind boggles, especially when you take into account the possibility of setting up a tech chain: If you can type, you can both program and send e-mail to vendors and have stuff hooked up for you, and then write the software to control it, and then use the hardware to do other things, and then still other things, and build better prostheses... The device is described as being about the size of an automobile gascap and is not fully contained, which is to say that it still has to have a persistent opening through the skin and skull to connect to an electrode grid placed atop the subject's brain. Major surgery is, of course still required to position the electrode grid on one of the motor cortices. Still, output bandwidth of this device aside it represents a remarkable breakthrough in that it's so small. After ten years of hard work all of the signal processing is done on board without needing to be plugged into racks of computers to do the number crunching. There isn't any word yet on when FDA trials will begin but you can be once they do all hell's going to break loose. Time to start saving our pennies...