As I write this, Supercon 2023 is in full swing down in Pasadena — 80 degrees and sunny at the moment, as opposed to 50 and pouring rain where I am, not that I’m bitter. Luckily, though, we can all follow along with the proceedings thanks to the livestreams on the Hackaday channel, which of course will all be available once they’re edited in case you miss anything live. There are a ton of interesting talks coming up, so there’ll be a lot to catch up on when the dust settles. And that won’t be far from now; by the time this post publishes, Supercon will be all but over, which makes it the Thanksgiving dinner of cons — all that work and it’s over in just a few minutes.
Author: Dan Maloney3300 Articles
Lessons In Printer Poop Recycling
The fundamental problem with multi-color 3D printing using a single hotend is that they poop an awful lot. Every time they change filaments, they’ve got to purge the single nozzle, which results in a huge number of technicolor “purge poops” which on some machines are even ejected out a chute at the back of the printer. The jokes practically write themselves.
What’s not a joke, though, is the sheer mass of plastic waste this can produce. [Stefan] from CNC Kitchen managed to generate over a kilo of printer poop for a 500-gram multi-color print. So he set about looking for ways to turn printer poops back into filament, with interesting results. The tests are based around a commercial lab-scale filament extruder, a 3Devo Composer, but should apply to almost any filament extruder, even the homebrew ones. A few process tips quickly became evident. First, purge poops are too big and stringy (ick) to feed directly into a filament extruder, so shredding was necessary.
Second, everything needs to be very clean — no cross-contamination with plastics other than PLA, no metal bits in the chopped-up plastic bits, and most importantly, no water contamination. [Stefan]’s first batch of recycled filament came from purge poops that had been sitting around a while, and sucked a lot of water vapor from the air. A treatment in a heated vacuum chamber seems to help, but what worked best was using purge poops hot and fresh from a print run. Again, ick.
[Stefan] eventually got a process down that produced decent, usable filament that would jam the printer or result in poor print quality. It even had a pretty nice color, which of course is totally dependent on the mix of colors you start with. Granted, not everyone has access to a fancy filament extruder like his, so this may not be practical for everyone, but it at least shows that there’s a path to reducing the waste stream from any printer, especially multi-material ones.
Wooden Game Boy Is A Challenging Intro CNC Project
[Sebastian] describes himself as “a total noob” when it comes to CNC, so in an attempt to get to know his new CNC router, he chose about the most complex possible project — replicating an original Game Boy case in wood. And spoiler alert: he nailed it.
Of course, he did have a few things going for him. At least from a straight woodworking perspective, it’s hard to go wrong by choosing walnut as your material. Then again, it can be unforgiving at times, and picky about tooling, which is probably why [Sebastian] used nine different tools to get the job done. But where he upped the difficulty level was in reproducing so many of the details of the original injection-molded plastic case. There are top and bottom shells, each of which has to be milled from both sides. This makes registration tricky when the parts are flipped. Specific indexing holes were used for that, along with the old “blue tape and CA glue” fixturing trick, which seemed to work quite well. For our money, though, the best bit is the lettering on the front face, which was milled out with an engraving bit and then filled with a spritz of white spray paint. A surfacing bit then came along to knock the overspray down, leaving labels that contrast beautifully with the dark wood. Gorgeous!
It wasn’t all easy sailing, though. There are just some things plastic can do that wood can’t, like holding screw threads in small studs without splitting. So, the case had to be glued shut once the mix of salvaged and new components went in. Still, it looks fabulous, and [Sebastian] says what we see in the video below is the one and only piece. Pretty sweet for the first try. Surprisingly, it doesn’t seem as if we’ve seen a wooden Game Boy before — a wooden NES, sure, but not a Game Boy.
Continue reading “Wooden Game Boy Is A Challenging Intro CNC Project”
Just How Dodgy Are Cheap USB Chargers Anyway?
Aside from apparently having both the ability to reproduce on their own and simultaneously never being around when you need one, USB chargers seem innocuous enough. The specs are simple: convert mains voltage to 5 volts, and don’t kill anyone while doing it. Both specs are typically met by most designs, but judging by [DiodeGoneWild]’s latest USB charger teardown, the latter only just barely, and with a whole lot of luck.
The sad state of plug-in USB power supplies is one of [DiodeGoneWild]’s pet gripes, and deservedly so. Most USB chargers cram a lot of electronics into a mighty small volume, and are built to a price point, meaning that something has to give in the design. In the case of the two units he tears apart in the video below, it’s pretty clear where the compromises are. Neither unit met the specs on the label in terms of current supplied and voltage regulation, even the apparently more capable quick charger, which is the first to go under the knife. The PCB within holds some alarming surprises, like the minimal physical isolation between the mains part of the circuit and the low-voltage section, but the real treat is the Schottky diode that gets up to 170°C under full load. Safety tip: when you smell plastic burning, throw the thing out.
The second charger didn’t fare any better; although it didn’t overheat, that’s mainly because it shut itself off before it could deliver a fraction of its rated 1 amp output. The PCB construction was shoddy in the extreme, with a squiggly trace standing in for a proper fuse and a fraction of a millimeter separation between primary and secondary traces. The flyback transformer was a treat, too; who doesn’t want to rely on a whisper-thin layer of cheap lacquer to keep mains voltage out of your phone?
All in all, these designs are horrible, and we have to thank [DiodeGoneWild] for the nightmares we’ll have whenever we plug into one of these things from now on. On the other hand, this was a great introduction to switch-mode power supply designs, and what not to do with our own builds. Continue reading “Just How Dodgy Are Cheap USB Chargers Anyway?”
Big Red Button Puts Toddler In Command Of Chromecast
Controversial position: the world needs more buttons. We’ve gotten so far away from physical interfaces like buttons, knobs, and switches in favor of sleek but sterile touch-screen “controls” that when we see something like this big red button so toddlers can start a TV show, we just have to latch onto the story and see what it’s all about.
As it turns out, the big red button itself is probably the least interesting part of [Mads Chr. Olesen] build. The real meat of the project is the reverse engineering effort needed to get Chromecast to start the show. As [Mads] explains, once upon a time a simple GET
request to a URL was all it took to do so, but no more; Google has repeatedly nerfed the Chromecast API over the years, enough that [Mads] had some digging to do.
Luckily, pyChromecast is a thing, but using it for DRTV, a streaming service of the Danish Broadcasting Corporation, required figuring out the AppID of the DRTV app. It looks like [Mads] used Wireshark to sniff traffic to and from the Chromecast, and netlog-viewer to analyze the capture. That and a little Developer Tools action in Chrome led to all the information needed to modify pyChromecast to support DRTV. The rest of the project consisted of building a box for the huge red arcade button and wiring it up to a Wemos D1. A Raspberry Pi actually talks to the Chromecast, and now the toddler is able to call up his favorite show and pause and restart it at will, no parent required.
We appreciate the reverse engineering heroics [Mads] displays here, which provide good general lessons for other purposes. It’s been a while since we’ve seen a Chromecast physical interface build, too, so we appreciate the refresher.
Partial Relay-Based Calculator Puts The Click Where It Counts
It looks like [Michal Zalewski] is raising the next generation the right way. First, his eldest son asks for help building a one-bit computer from discrete transistors. Not to be left behind, his little brother then asked for help with an even more retro project, which resulted in this partially relay-based calculator. Maybe there is some hope for the future.
Now, purists will no doubt notice the ATmega64 microcontroller sitting in the middle of the main PCB on this project and cry “Foul!” But perfect is the enemy of done, and as [Michal] explains, at $6 a pop for the Omron relays he and his son chose, there’s only so far you can go with relay logic before you’re taking out a second mortgage. So the relays are limited to the ALU of the calculator, along with the drivers for the six seven-segment LED displays. The microcontroller is just there for housekeeping functions like scanning the keyboard and decoding digits. All the actual calculations are in the relay logic, not silicon. And we’d be remiss not to praise his son’s stylistic choices for this design — that it uses relays with clear covers, and that it has single-sided PCBs with curvy, hand-drawn traces traces that look hand-drawn on old-school yellow substrate. [Michal]’s heart must swell with pride to have fathered someone with such exquisite taste.
For his part, [Mikal] did some really good documentation for this build, including excellent descriptions of Boolean math with half- and full-adders and how relays are used to create the basic logic gates that comprise them. The calculator itself is still a work in progress, with microcontroller code still in development, but it’s working enough that you can enjoy the display driver’s clickiness in the video below. If that doesn’t do it for you, we’ve got other relay calculators to scratch that click itch. Continue reading “Partial Relay-Based Calculator Puts The Click Where It Counts”
Pocketable Yagi Antenna Really Shoots For Distance
For amateur radio operators, the quest for the perfect antenna never seems to end. Perhaps that’s because our requirements are always changing. We never quite seem to get to one design that can do everything. This copper-foil Yagi antenna might not do everything, but it really seems to tick off the boxes for gain and directionality along with ultra-portability.
If you’ve been following [Ben Eadie (VE6SFX)]’s trip down the rabbit hole of lightweight antenna building, you’ll recall that he’s already knocked off a J-pole antenna and a stealthy mobile slot antenna using little more than copper foil tape. Both of those designs performed great, but [Ben] had bigger fish to fry: he wanted to build a directional antenna for the 2-meter band and go for distance. The traditional Yagi-Uda is generally the preferred design for beam antennas, but they tend to be bulky and cumbersome. But with a roll of copper foil tape [Ben] was able to lay out a three-element Yagi on a sheet of Tyvek wrap. Reinforced with some packing tape and stiffened with a couple of fiberglass rods attached to a 3D printed handle, and it was ready to go.
[Ben]’s field test results were most impressive. Not only was he able to open up repeaters up to 90 km away, but he was getting good signal reports to boot. He was even able to reach a repeater 150 km distant, just barely though. Still, that’s mighty impressive performance from something that looks like a Union Jack and rolls up to fit in a pocket.
Continue reading “Pocketable Yagi Antenna Really Shoots For Distance”