To celebrate the twentieth anniversary of their Trinitron line of televisions, Sony launched the KX-45ED1. At forty three inches the screen on this particular model made it the largest tube television in the world, and it came with the kind of price tag that if you need to ask…you can’t afford it (likely around $100,000 USD today). Three decades later, only two of these mythical displays were thought to exist and [shank] chronicled his quest to acquire one of the last remaining “Big Boys” in the mini documentary below.

As it turns out, one of these gigantic tube televisions was located on the second floor of a restaurant in Japan still sitting in the same place it was installed in 1989. It hadn’t moved in the intervening decades, because the television and its specialized support stand weighed over 500 pounds. Having an object that heavy physically moved down a flight of stairs would seem to be the most formidable challenge for most, but compounding the issue for [shank] was that the building housing this colossal CRT was set to be permanently closed in less than a week.

With next to no time to arrange an international flight, [shank] utilized the power of internet to ask for help from anyone currently living near the “Big Boy” CRT’s soon-to-be final resting place. It just so happened that a fellow retro tech enthusiast based in Japan saw the post, and traveled over an hour by train at a moment’s notice to aid [shank]. The heartwarming story of total strangers united by a common interest of preserving a rare piece of tech history is certainly worth a watch. Let alone the goofy size comparison footage of the smallest CRT display sitting on top of the biggest one.

For more on tube TVs and the like, check out this article by Dave on retro gaming on CRT displays.

[Corelatus] said recently that “someone” asked them to identify the phone signals in the 1982 film The Wall, based on the Pink Floyd song of the same name. We suspect that, like us, that someone might have been more just the hacker part of the brain asserting itself. Regardless, the detective work is fascinating, and you can learn a lot of gory details about phone network in-band signaling from the post.

The analysis is a bit more difficult because of the year the film was made. At that time, different countries used slightly different tone signaling standards. So after generating a spectrogram, the job was to match the tones with known standards to see which one best fit the data.

The signal was not common DTMF, as you might have guessed. Instead, it was a standard known as SS5. In addition to the tones being correct, the audio clip seemed to obey the SS5 protocol. SS5 was the technology attacked by the infamous blue box back when hacking often meant phone phreaking.

The same phone call appears on the album, and others have analyzed it with some even deeper detective work. For example, the call was made in 1979 from a recording studio by [James Guthrie], who called his own phone in the UK, where his next-door neighbor had instructions to hang up on the operator repeatedly.

If you want to see and hear the entire clip (which has several phone-related audio bits in it), watch the video below. The sequence of SS5 tones occurs at 3:13.

Usually, when we hear tones in music, we think of Morse code. As for phone phreaking, we hear it’s moved to street kiosks.

Game Boys have a link cable that lets two of them play together. You know, to battle with a friend’s Pokemon and stuff like that. But who says that it should be limited to transmitting only what Big N wants you to?

[Chromalock] wrote a custom GB program that takes in data over the link cable, and displays it on the screen as video, as fast as it can be sent. Add in a microcontroller, a level shifter, and software on the big computer side, and you can hook up your Game Boy Color as a normal video device and send it anything you want, from a webcam to any program that outputs video.

Well, almost. The biggest limitation is the data link cable, of course. On the older Game Boys, the link cable is apparently only good for 8 kHz, while the Color models can pull a not-quite-blistering 512 kHz. Still, that’s enough for 60 fps in a low-res black and white mode, or a slow, screen-tearing high-res color experience. You pick your poison.

There are gotchas that have to do with the way the GB displays palettes that get left as “to-do” on the software side. There is room for improvement in hardware too. (GB Link looks like SPI to us, and we’d bet you can push the speeds even higher with clever GB-side code.) In short, this is an awesome demo that just invites further hacking.

If you want to know more about the Game Boy to get started, and maybe even if you don’t, you absolutely must watch The Ultimate Game Boy Talk. Trust us on this one.

Although the US’ Moon landings were mostly made famous by the fact that it featured real-life human beings bunny hopping across the lunar surface, they weren’t there just for a refreshing stroll over the lunar regolith in deep vacuum. Starting with an early experimental kit (EASEP) that was part of the Apollo 11 mission, the Apollo 12 through Apollo 17 were provided with the full ALSEP (Apollo Lunar Surface Experiments Package). It’s this latter which is the subject of a video by [Our Own Devices].

Despite the Apollo missions featuring only one actual scientist (Harrison Schmitt, geologist), these Bendix-manufactured ALSEPs were modular, portable laboratories for running experiments on the moon, with each experiment carefully prepared by scientists back on Earth. Powered by a SNAP-27 radioisotope generator (RTG), each ALSEP also featured the same Central Station command module and transceiver. Each Apollo mission starting with 12 carried a new set of experimental modules which the astronauts would set up once on the lunar surface, following the deployment procedure for that particular set of modules.

Although the connection with the ALSEPs was terminated after the funding for the Apollo project was ended by US Congress, their transceivers remained active until they ran out of power, but not before they provided years worth of scientific data on many aspects on the Moon, including its subsurface characteristics and exposure to charged particles from the Sun. These would provide most of our knowledge of our Moon until the recent string of lunar landings by robotic explorers.

Heading image: Apollo Lunar Surface Experiments Package of the Apollo 16 mission (Credit: NASA)

You can buy motorized camera sliders off-the-shelf, but they’re pretty costly. Alternatively, you can make one yourself, and it’s not even that hard if you’re kitted out with a 3D printer. [Creative 3D Printing] did just that with a nifty design that adds rotation into the mix. Check it out in the video below.

The basic slider is built out of 3D-printed components and some good old aluminum extrusion. A small 12-volt motor trucks the camera cart back and forth using a leadscrew. It’s torquey enough and slow enough that there isn’t much need for more advanced control—the motor just does the job. There’s also a limit switch set up to trigger a neat auto-reverse function.

The neat part, though, is the rotational mechanism. A smooth steel rod is attached to the slider’s housing, which can be set up in a straight line or aligned diagonally if desired. In the latter case, it rotates the mounting on the camera cart via a crank, panning the camera as it moves along the slider’s trajectory.

It’s a mechanically sophisticated design and quite unlike most of the camera sliders we feature around these parts.

Rope-climbing robots are the stuff of engineering dreams. As kids, didn’t we all clutter our family home with constructions of towers and strings – Meccano, or Lego – to have ziplines spanning entire rooms? Good for the youngsters of today, this has been included in school curricula. At the University of Illinois, the ME 370 students have been given the task of building a robot that can hang from a rope and walk across it—without damaging the rope. The final projects show not only how to approach tricky design problems, but also the creative solutions they stumbled upon.

Imagine a tiny, rope-climbing walker in your workshop—what could you create?

The project is full of opportunities for those thinking out of the box. It’s all about the balance between innovation and practicality: the students have to come up with a solution that can move at least 2 meters per minute, fits in a shoebox, and has some creative flair—no wheels allowed! The constraints provide an extra layer of challenge, but that’s where the fun lies. Some students use inverted walkers, others take on a more creature-like approach. The clever use of motors and batteries shows just how far simple tech can go when combined with a bit of engineering magic.

This project is a fantastic reminder that even small, seemingly simple design challenges can lead to fascinating creations. It invites us adults to play, and by that, we learn: a win-win situation. You can find the original article here, or grab some popcorn and watch the video below.

There are many retrocomputer emulation projects out there, and given the relative fragility of the original machines as they enter their fifth decade, emulation seems to be the most common way to play 8-bit games. It’s easy enough to load one on your modern computer, but there are plenty of hardware options, too. “The computer we’d have done anything for back in 1983” seems to be a phrase many of them bring to mind, but it’s so appropriate because they keep getting better. Take [Stormbytes1970]’s Pi Pico-powered Sinclair ZX Spectrum mini laptop (Spanish language, Google Translate link), for example. It’s a slightly chunky netbook that’s a ZX Spectrum, and it has a far better keyboard than the original.

On the PCB is the Pico, the power supply circuitry, an SD card, and a speaker. But it’s when the board is flipped over that the interesting stuff starts. In place of the squidgy rubber keyboard of yore, it has a proper keyboard,. We’re not entirely sure which switch it uses, but it appears to be a decent one, nevertheless. The enclosure is a slick 3D-printed sub-netbook for retro gaming on the go. Sadly, it won’t edit Hackaday, so we won’t be slipping one in the pack next time we go on the road, but we like it a lot.

It’s not the first Spectrum laptop we’ve covered, but we think it has upped the ante over the last one. If you just want the Spectrum’s BASIC language experience, you can try a modern version that runs natively on your PC.

Ham radio enthusiasts, people looking to borrow their neighbors’ WiFi, and those interested in decoding signals from things like weather satellites will often grab an old satellite TV antenna and repurpose it. Customers have been leaving these services for years, so they’re pretty widely available. But for handheld operation, these metal dishes can get quite cumbersome. A 3D-printed satellite dish like this one is lightweight and small enough to be held, enabling some interesting satellite tracking activities with just a few other parts needed.

Although we see his projects often, [saveitforparts] did not design this antenna, instead downloading the design from [t0nito] on Thingiverse. [saveitforparts] does know his way around a satellite antenna, though, so he is exactly the kind of person who would put something like this through its paces and use it for his own needs. There were a few hiccups with the print, but with all the 3D printed parts completed, the metal mesh added to the dish, and a correctly polarized helical antenna formed into the print to receive the signals, it was ready to point at the sky.

The results for the day of testing were incredibly promising. Compared to a second satellite antenna with an automatic tracker, the handheld 3D-printed version captured nearly all of the information sent from the satellite in orbit. [saveitforparts] plans to build a tracker for this small dish to improve it even further. He’s been able to find some satellite trackers from junked hardware in some unusual places as well. Antennas seem to be a ripe area for 3D printing.

[Lonyelon] wanted to build an anniversary gift for his girlfriend. He decided to say it with e-Paper, a wise choice given its persistence and longevity.

The project is an anniversary calendar. It displays a counter of the total time the couple has been together, measured in years, months, days, and hours—so it’s remarkably precise. [Lonyelon] also programmed it to display additional counters to create plenty of additional fun anniversaries—the couple can celebrate milestones like their 1000th day together, for example. It also cycles through a range of cute messages and displays photos of the couple together.

The code is on Github for the curious. The build is based around a LilyGO e-Paper display with an onboard ESP32 microcontroller. [Lonyelon] paired this with a 2,500 mAh battery. It lasts for ages because the device is programmed to update only every 20 minutes, spending the rest of its time in deep sleep. Since it’s an e-Paper display, it uses zero power when it’s not being updated, so it’s the perfect technology for this application.

It’s a simple project that comes from the heart—the core of any beautiful gift. In fact, some of the coolest projects we feature were built as gifts for romantic partners, family members, or even our fellow hackers. If you’ve been cooking up your own neat build, please let us know on the tipsline!

Multiplication on a common microcontroller is easy. But division is much more difficult. Even with hardware assistance, a 32-bit division on a modern 64-bit x86 CPU can run between 9 and 15 cycles. Doing array processing with SIMD (single instruction multiple data)  instructions like AVX or NEON often don’t offer division at all (although the RISC-V vector extensions do). However, many processors support floating point division. Does it make sense to use floating point division to replace simpler division? According to [Wojciech Mula] in a recent post, the answer is yes.

The plan is simple: cast the 8-bit numbers into 32-bit integers and then to floating point numbers. These can be divided in bulk via the SIMD instructions and then converted in reverse to the 8-bit result. You can find several code examples on GitHub.

Since modern processors have several SIMD instructions, the post takes the time to benchmark many different variations of a program dividing in a loop. The basic program is the reference and, thus, has a “speed factor” of 1. Unrolling the loop, a common loop optimization technique, doesn’t help much and, on some CPUs, can make the loop slower.

Converting to floating point and using AVX2 sped the program up by a factor of 8X to 11X, depending on the CPU.  Some of the processors supported AVX512, which also offered considerable speed-ups.

This is one of those examples of why profiling is so important. If you’d had asked us if converting integer division to floating point might make a program run faster, we’d have bet the answer was no, but we’d have been wrong.

As CPUs get more complex, optimizing gets a lot less intuitive. If you are interested in things like AVX-512, we’ve got you covered.

Camera sliders are a popular project for makers—especially those who document their projects on video. They’re fun and accessible to build, and they can really create some beautiful shots. [Lechnology] set about to follow in this fine tradition and built a rather capable example of his own. Check it out in the video below.

The slider relies on V-slot rails, perhaps most familiar for their heavy use in modern 3D printers. The rails are paired with a 3D-printed camera carriage, which runs on smooth rubber rollers. A chunky stepper motor provides drive via a toothed belt. Trinamic motor controllers were chosen for their step interpolation feature, making the motion much smoother.

The slider doesn’t just move linearly, either. It can rotate the camera, too, since it has an additional motor in the carriage itself. In a nice retro touch, the wires for this motor are run with an old coiled telephone cable. It’s perfect for the job since it easily extends and retracts with the slider’s motion. Controlling everything is an Arduino, with speed and rotational modes set via a tiny screen and a rotary encoder control.

It’s a very complete build, and it performs well too. The video it produces is deliciously smooth. We’ve featured some other great camera sliders over the years, too. If you want to dig into Trinamic drivers, we can get you started.

Early Monday morning, while many of us will be putting the finishing touches — or just beginning, ahem — on our Christmas preparations, solar scientists will hold their collective breath as they wait for word from the Parker Solar Probe’s record-setting passage through the sun’s atmosphere. The probe, which has been in a highly elliptical solar orbit since its 2018 launch, has been getting occasional gravitational nudges by close encounters with Venus. This has moved the perihelion ever closer to the sun’s surface, and on Monday morning it will make its closest approach yet, a mere 6.1 million kilometers from the roiling photosphere. That will put it inside the corona, the sun’s extremely energetic atmosphere, which we normally only see during total eclipses. Traveling at almost 700,000 kilometers per hour, it won’t be there very long, and it’ll be doing everything it needs to do autonomously since the high-energy plasma of the corona and the eight-light-minute distance makes remote control impossible. It’ll be a few days before communications are re-established and the data downloaded, which will make a nice present for the solar science community to unwrap.

While Parker has been in a similar position on previous orbits and even managed a fortuitous transit of a coronal mass ejection, this pass will be closer and faster than any previous approach. It’s the speed that really grabs our attention, though, as Parker will be traveling at a small but significant fraction of the speed of light for a bit. That makes us wonder if there was any need for mission planners to allow for relativistic effects. We’d imagine so; satellite navigation systems need to take relativity into account to work, and they don’t move anywhere near as fast as Parker. Time will be running slower for Parker at those speeds, and it sure seems like that could muck things up, especially regarding autonomous operation.

Ever since the seminal work of Cameron, Hamilton, Schwarzenegger, et al, it has been taken as canon that the end of humanity will come about when the moral equivalent of SkyNet becomes self-aware and launches all the missiles at once to blot us out with a few minutes of thermonuclear fire. But it looks like AI might be trying to raise an army of grumpy teenagers if this lawsuit over violence-inciting chatbots is any indication. The federal product liability lawsuit targets Character.AI, an outfit that creates LLM-powered chatbots for kids, for allegedly telling kids to do some pretty sketchy stuff. You can read the details in the story, but suffice it to say that one of the chatbots was none too pleased with someone’s parents for imposing screen time rules and hinted rather strongly about how the child should deal with them. The chat logs of that interaction and others that are part of the suit are pretty dark, but probably no darker than the advice that most teenagers would get online from their carbon-based friends. That’s the thing about chatbots; when an LLM is trained with online interactions, you pretty much know what’s going to come out.

In today’s “Who could have seen that coming?” segment, we have a story about how drivers are hacked by digital license plates and are keen to avoid tolls and tickets. The exploit for one specific brand of plate, Reviver, and while it does require physical access to the plates, it doesn’t take much more than the standard reverse engineering tools and skills to pull off. Once the plates are jailbroken — an ironic term given that license plate manufacturing has historically been a prison industry — the displayed numbers can be changed at will with a smartphone app. The worst part about this is that the vulnerability is baked right into the silicon, so there’s nothing to be patched; the plates would have to be recalled, and different hardware would need to be reissued. We’ve been skeptical about the need for these plates from the beginning and questioned why anyone would pay extra for them (last item). But maybe the ability to dump your traffic cam violations into someone else’s lap is worth the extra $20 a month.

And finally, this local news story from Great Falls, Montana, is a timely reminder of how machine tools can mess up your life if you let them. Machinist Butch Olson was alone at work in his machine shop back on December 6 when the sleeve of his jacket got caught in a lathe. The powerful machine pulled his arm in and threatened to turn him to a bloody pulp, but somehow, he managed to brace himself against the bed. He fought the lathe for 20 whole minutes before the motor finally gave out, which let him disentangle himself and get some help. He ended up with a broken back, four fractured ribs, and an arm that looks “like hamburger” according to his sister. That’s a high price to pay, but at least Butch gets to brag that he fought a lathe and won.

Many of us grew up with dreams of piloting a forklift one day. Sadly, most warehouses take a dim view of horseplay with these machines, so few of us get to live out those fantasies. Playing with this desk-sized RC model from [ProfessorBoots] is probably a safer way to get those kicks instead. You can check it out in the video below.

The 3D-printed body of the forklift is the first thing you see. It’s great quality, and it instantly puts you in mind of the real thing. The build is true to the dynamics of a real forklift, too, with proper rear steering. Inside, there’s a custom circuit board hosting an ESP32 that serves as the brain of the operation. Its onboard wireless hardware allows remote control of the forklift via a smartphone app, PS4 controller, or many other options. It controls the drive motors and steering servo, along with another motor driving a threaded rod to move the forks up and down. The whole thing is powered by two Fenix 16340 batteries—small lithium-ion cells that can be recharged with an integral micro USB port.

The project video is very thorough about the design and build. It’s worth watching just to understand the specifics of how forklifts actually raise their forks up and down. It’s good stuff.

This forklift is just the latest RC build from [ProfessorBoots]. He’s done great work in this space before, like this charming skid steer and incredibly complex crane.

Are you familiar with Huffman encoding? That’s where you pick shorter codes for more frequent letters. Morse code is the same way, in that the most-used letters are the shortest. [Matheus Richard] had the same idea for optimizing your workflow in the Linux shell. The idea is to measure what commands you use the most and make them shorter.

If you use zsh, it is easy to find out what commands you are using the most. If you use bash, [Matheus] helpfully offers a command to give you a similar result (the original post limits the list to the last entry which we are sure is a typo):

history | awk '{CMD[$2]++;count++;}END { for (a in CMD)print CMD[a] " " CMD[a]/count*100 "% " a;}' | grep -v "./" | column -c3 -s " " -t | sort -nr | nl | head -n10

Once you know the commands you use the most, you can use your shell’s aliasing or scripts to shorten them up. For example, in [Matheus’] case, git was a very common command. This led to aliases:

alias gc="git commit --verbose"
alias gp="git push"
alias gprom="git pull --rebase origin main"

Not only does this save typing, but you lessen your chance for typos: “git comit”, for example. Another idea is to alias your common errors, for example setting an alias for “git” as “gti”.

These are small things, but they are definitely time savers. Be sure to read the rest of the post, as there are a number of other optimization ideas. [Matheus] definitely has a thing for zsh, but there are many other shells out there. Some of them are even evolving towards more modern programming languages.

The ESP32 is s remarkably powerful microcontroller, where its dual-core processor and relatively high clock speed can do some impressive work. But getting this microcontroller designed for embedded systems to do tasks that would generally be given to a much more powerful PC-type computer takes a little bit more willpower. Inspired by his dog, [Folkert] decided to program an ESP32 to play chess, a famously challenging task for computer scientists in the past. He calls this ESP32 chess system Dog.

One of the other major limitations of this platform for a task like this is memory. The ESP32 [Folkert] is using only has 320 kB of RAM, so things like the transposition table have to fit in even less space than that. With modern desktop computers often having 32 or 64 GB, this is a fairly significant challenge, especially for a memory-intensive task like a chess engine. But with the engine running on the microcontroller it’s ready to play, either in text mode or with something that can use the Universal Chess Interface (UCI). A set of LEDs on the board lets the user know what’s going on while gameplay is taking place.

The UCI also enables Dog to play online at lichess.org, and [Folkert] has included a link on the project page where others can play with his microcontroller chess system this way through the Internet. It has a pretty respectable Elo rating at around 2100 as well, so don’t think that just because it’s a small platform that the wins will come easy. If you’d prefer your chess engines to run on retro hardware, take a look at this build, which also uses an ESP32 but puts it to work by running old Commodore chess equipment from the 80s. Of course, you can play chess on even less hardware. It has been done.

Vintage hi-fi gear can be very attractive, particularly compared to modern stuff. However, when this stuff starts getting into its third or fourth decade after production, things start to wear out. Chief among them—the little incandescent bulbs that light up the dials with such a beautiful glow. [Piffpaffpoltrie] was suffering just this problem on an old Technics amp, and decided to go for a more modern upgrade.

Replacing the original bulb with a like unit was undesirable—even if many last for decades, [Piffpaffpoltrie] didn’t want to have to tackle this job again in the future. Instead, an LED swap was the order of the day. A short strip of warm-white LEDs seemed to be the perfect solution, with three LEDs in series being just about right for the 11-volt supply used for the original bulbs. The only problem was that the stereo supplied the bulbs with AC, not DC. Thus, a quick bridge rectifier circuit was thrown in, along with some series resistors. This wrangled the voltage into a straighter line and delivered the right voltage level to drive the LEDs nicely and smoothly.

The result is a nicely-illuminated set of power meters on this vintage Technics amp. We’ve seen some neat LED swaps in the past, too, including this tricky motorcycle lamp upgrade. Meanwhile, if you’re slogging it out to bring your vintage gear more up to date, consider dropping us a note on the tipsline.

While you tend to think of diamonds as ornamental gemstones, diamonds also have many important industrial uses, and many of those diamonds are now synthetic polycrystalline diamonds. How are they made? [JerryRigEverything] takes us behind the scenes at a diamond manufacturing facility, something you don’t get to see every day. Check out the giant presses that exert about a million pounds of pressure in the video below.

The process starts with diamond powder, which is just what it sounds like. Although you can get real diamond powder, most uses today start with synthetic diamonds. The powder has many uses in cosmetics and as an abrasive. But the video will combine it with cobalt and table salt to form diamond shapes.

The salt is a high-temperature electrode. The process requires temperatures of nearly 1400C (2500F) and a lot of pressure. Common talc, some metal electrodes, and a heater tube are also used in the process.

The press can convert a little diamond dust into a diamond in about 10 minutes. However, because the machines are so dangerous, they are each set in their own blast room, which is sealed when the press is in operation.

While this press was — no pun intended — impressive, we’ve seen bigger. Nothing like this will show up in your garage anytime soon, although, as the video shows, you can buy 3D printer nozzles made from the material. As for a press, you might have to just settle for an arbor press.

Altermagnetism is effectively a hybrid form of ferromagnetism and antiferromagnetism that might become very useful in magnetic storage as well as spintronics in general. In order to practically use it, we first need to be able to control the creation of these altermagnets, which is what researchers have now taken the first steps towards. The research paper by [O. J. Amin] et al. was published earlier this month in Nature. It builds upon the team’s earlier research, including the detection of altermagnetism in manganese telluride (MnTe). This new study uses the same material but uses a photoemission electron microscope (PEEM) with X-rays to image these nanoscale altermagnetic structures.

Additionally, the spin orientation of these altermagnetic structures within the MnTe was controlled using microstructure patterning and thermal cycling in magnetic fields. The micropatterning with electron beam lithography enabled the creation of large single-domain altermagnetic structures, which is promising for further research. As noted in the outlook section by the researchers, this part of the research is still very much about creating the basic means to use altermagnetism, even for something as seemingly straightforward as data storage. In this particular study, the reading (imaging) mechanism was an expensive PEEM setup with the X-rays produced by a synchrotron.

Honestly, we still struggle to figure out plain old magnetism. Obviously, there’s more to it than that.

Heading image: Illustrative models of collinear ferromagnetism, antiferromagnetism, and altermagnetism in crystal-structure real space and nonrelativistic electronic-structure momentum space. (Credit: Libor Šmejkal et al., Phys. Rev. X, 2022)

[CNLohr] has been tinkering with some fun parts of late. He’d found out that ordinary LCD screens could be used as simple touch sensors, and he had to try it for himself. He ended up building a little doohickey that combined USB C, an LCD display, and a touch interface, all for under a buck. You can check out the video below.

The key to this build was the CH32V003 CPU. It’s a RISC-V microcontroller that runs at a healthy 48 MHz, and it costs just 10 cents in reasonable quantities. A PCB etched to mate with a USB C cable eliminates the need for a connector.

[CNLohr] then gave the board a three-digit 7-segment LCD display from Aliexpress, which can be had for around 21 cents if you buy 100 or more. He then figured out how to drive the LCDs with a nifty trick that let the microcontroller use the display as a crude touch sensor. All in all, the total bill of materials for one of these things comes out somewhere under a dollar in quantity.

It’s mostly a random assemblage of tech glued together for a demo, but it’s a fun project. It’s worth checking out even if it’s just to learn how to create an integral USB C port on your own PCBs. The way it’s achieved with the etched contacts and milled-out tabs is pure elegance. Files are on Github for the curious.

We’ve featured a ton of [CNLohr’s] work over the years; the clear keytar was a glowing highlight, as were his early discoveries in the depths of the ESP8266.