2026-01-12 05:00:51
EDM (Electrical Discharge Machining) is one of those specialised manufacturing processes that are traditionally expensive and therefore somewhat underrepresented in the DIY and hacker scenes. It’s with great delight that we present EnderSpark, a solution to not one but two problems. The first problem is how to perform CNC operations on hard-to-machine materials such as hardened metals (without breaking the bank). The second problem is what to do with all those broken and forgotten previous-generation Creality Ender 3D printers we know you have stashed away.
To be honest, there isn’t much to a cheap 3D printer, and once you ditch the bed and extruder assembly, you aren’t left with a lot. Anyway, the first job was to add a 51:1 reduction gearbox between the NEMA 17 motors and the drive pullies, giving the much-needed boost to positional accuracy. Next, the X and Y axes were beefed up with a pair of inexpensive MGN12H linear rails to help them cope with the weight of the water bath.
The majority of the work is in the wire feeder assembly, which was constructed around a custom-machined aluminium plate.
It’s not lost on us how the original RepRap bootstrapping concept could be applied here: a basic frame made externally in a low-cost material, then using the machine to cut a much thicker, stronger copy for its own upgrade. The main guide nozzle is an off-the-shelf ruby part surrounded by a 3D printed water-cooling jacket. To maximise power transfer from the wire into the electrically conductive workpiece material, the top part of the wire feeder, including the wire itself, is one electrode, and the entire bottom part of the frame is electrically isolated from it. The bottom part pulls the ‘consumed’ stock wire through the nozzle above and keeps it under tension, sending it onward to the waste spool.
Electrically speaking, the project is based on stock Ender electronics, with an additional power driver stage to send capacitor-discharge-derived pulses down the wire from the 48V power supply, up to 10A, generating the needed tiny sparks as the wire is advanced into the electrically grounded workpiece. Industrial machines operate around twice this voltage, but safety is a big issue with a DIY machine. Not to mention 48V and water don’t make the best of friends. Speaking of water, it needs to be de-ionised to reduce dielectric loss, but ionic contamination will build up over time, so it needs to be regularly changed.
Software-wise, the machine is running on G-code, so all that is needed is a custom plugin for Fusion 360 to turn the extracted toolpath (they’re using the Wazer water cutter profile as a basis) into G-code, with knowledge of the material. There aren’t too many variables to play with there.
In the future, a few things are being considered. Adding closed-loop control of the pulse energy would be straightforward, but controlling the horizontal feed rate would be a little trickier to implement with a pure G-code approach. We’ll keep an eye on the project and report back any advances!
If you’re thinking you’ve seen this sort of thing before, you’re right. Here’s another DIY EDM machine, and another, and finally, a Kickstarter we covered a while back that converts any 3D printer into a wire EDM.
Thanks [irox] for the tip!
2026-01-12 02:00:19
At some point during our primary school careers, most of us probably constructed a simple compass, often by floating a magnetized needle on a cork in a cup of water. The water in such a configuration not only lets the needle spin without friction, but also dampens out (so to speak) the needle’s tendency to swing back and forth across the north-south line. Liquid-filled compasses use the same principle, but even well-made compasses can develop bubbles when exposed to temperature or pressure variations. Rather than accept this unsightly state of affairs, [The Map Reading Company] designed a new kind of liquid-free, inductively-damped compass.
It’s hard to design a compass that settles quickly, even if it uses a strong magnet, because the Earth’s own magnetic field is just so weak, and the stronger the internal magnet is, the more likely it is to be thrown off by nearby magnetic objects. As a result, they tend to swing, overshoot, and oscillate around their final orientation for some time. Most compasses use liquid to damp this, but a few, mostly military compasses, use a conductive baseplate instead: as the magnet moves, it induces eddy currents in the baseplate, which create a weak magnetic field opposing its motion, slowing the magnet down. Inductively-damped compasses don’t get bubbles, but they don’t let you see a map through the baseplate. [The Map Reading Company] dealt with this by making the baseplate transparent and surrounding the compass needle with a ring of high-conductivity copper alloy. This gave him a clear baseplate compass for easy map reading which would never develop bubbles. It’s a simple hack, and should be easy to replicate, but it still seems to be a new design. In fact, [The Map Reading Company] is releasing most of the design to the public domain. Anyone can build this design.
If this prompts your interest in compasses, check out the Earth inductor compass. We’ve also seen a visualization of the eddy currents that damp these oscillations, and even seen them used to drive a bike.
Thanks to [Mel] for the tip!
2026-01-11 23:00:56
Continuing his reverse-engineering of the Intel 8087, [Ken Shirriff] covers the conditional tests that are implemented in the microcode of this floating point processing unit (FPU). This microcode contains the details on how to perform the many types of specialized instructions, like cos and arctan, all of which decode into many microcode ops. These micro ops are executed by the microcode engine, which [Ken] will cover in more detail in an upcoming article, but which is effectively its own CPU.
Conditional instructions are implemented in hardware, integrating the states of various functional blocks across the die, ranging from the instruction decoder to a register. Here, the evaluation is performed as close as possible to the source of said parameter to save on wiring.
Implementing this circuitry are multiplexers, with an example shown in the top die shot image. Depending on the local conditions, any of four pass transistors is energized, passing through that input. Not shown in the die shot image are the inverters or buffers that are required with the use of pass transistors to amplify the signal, since pass transistors do not provide that feature.
Despite how firmly obsolete the 8087 is today, it still provides an amazing learning opportunity for anyone interested in ASIC design, which is why it’s so great that [Ken] and his fellow reverse-engineering enthusiasts keep plugging away at recovering all this knowledge.
2026-01-11 20:00:07
[Ronan] likes 35mm film photography, but the world, of course, has gone digital. He picked up an Epson FilmScan 200 for about €10. This wonder device from 1997 promised to convert 35mm film to digital at 1200 DPI resolution. But there was a catch: it connects via SCSI. Worse, the drivers were forever locked to Windows 95/98 and Mac System 7/8.
In a surprise twist, though, [Ronan] recently resurrected a Mac SE/30 with the requisite SCSI port and the System 7 OS. Problem solved? Not quite. The official software is a plugin for Photoshop. So the obvious answer is to write new software to interact with the device.
First, of course, you have to figure out how the device works. A service manual provided clues that, as far as the SCSI bus knew, the device wasn’t a scanner at all, but a processor. The processor, though, used SCSI as a simple pipe to handle Epson’s standard “ESC/I” protocol.
Armed with that information and a knowledge of the Mac’s SCSI Manager API, the rest is just coding. Well, that is until [Ronan] tried to scan the other five negatives in the six-negative film carrier. He was frustrated until he found an old patched SANE driver for the scanner from 2002. By looking at how it worked, he was able to figure out how to switch to the other negatives.
Color scanning also took a little coaxing. The scanner returns three monochrome images, one for each color channel. Some assembly, then, is required. In the end, though, the project was a complete success. Can’t find a FilmScan 200? Don’t have a SCSI port? There’s always the roll-your-own approach.
2026-01-11 17:00:28
Emulating older computers on microcontrollers has been a staple of retrocomputing for many years now, with most 8-bit and some 16-bit machines available on Atmel, ARM, or ESP32 platforms. But there’s always been a horsepower limit, a point beyond which a microcontroller is no longer enough, and a “proper” computer is needed. One of those barriers now appears to have been broken, as microcontroller-based emulation moves into the 32-bit era. [Amcchord] has the Basilisk II emulator ported to the ESP32-P4 platform, providing a 68040 Mac able to run OS8.1. This early-1990s-spec machine might not seem like much in 2026, but it represents a major step forward.
The hardware it uses is the M5Stack Tab5, and it provides an emulated Mac with up to 16 MB of memory. Remember, in 1992 this would have been a high-spec machine. It manages a 15 frames per second refresh rate, which is adequate for productivity applications. The emulator uses the Tab5’s touchscreen to emulate the Mac mouse alongside support for USB input devices. To 1990 hackers, it’s almost the Mac tablet you didn’t know you would want in the future.
We like this project, both because it’s advancing the art of emulation on microcontrollers, and also because it delivers a computer that’s useful for some of the things you might have done with a Mac in 1992 and could even do today. Pulling this out on the train back then would have blown people’s minds. There’s even a chance that MacOS on something like this would turn a few heads in 2026. It’s certainly not the first emulated Mac we’ve seen though.
2026-01-11 14:00:41
The Tamagotchi Connection is a series of Tamagotchi toys that took the original portable pet concept and mixed things up with a wireless connection, which allowed you to interact with the pets of other proud Tamagotchi owners. This wireless connection is implemented using an infrared transceiver, somewhat like IrDA, but as [Zach Resmer] discovered while reverse-engineering this connection, it’s actually what is called ‘Nearly NEC’ by [Natalie Silvanovich], who has a GitHub repository full of related Tamagotchi hacking tools and ROM dumps.
With the protocol figured out, creating a transceiver for low-bitrate infrared communication isn’t particularly hard. In this case, it was implemented using an RP2040 MCU and an appropriate IR LED and receiver pair. This Tamagometer project was also implemented as an app for the Flipper Zero, and a custom PCB called the Pico TamaBadge by [Daniel Weidman].
There’s a web application associated with [Zach]’s project using a Web Serial-enabled browser (i.e. Chrome). The serial protocol is somewhat documented in the patent for the device’s connection feature, which makes it relatively easy to implement yourself.
