RepRap was the origin of pushing hobby 3D printing boundaries, and here we see a RepRap scaled down to the smallest detail. [Vik Olliver] over at the RepRap blog has been working on getting a printer working printing down to the level of micron accuracy.
The printer is constructed using 3D printed flexures similar to the OpenFlexure microscope. Two flexures create the XYZ movement required for the tiny movements needed for micron level printing. While still in the stages of printing simple objects, the microscopic scale of printing is incredible.
Despite the best efforts of the RepRap community over the last twenty years, self-replicating 3D printers have remained a stubbornly elusive goal, largely due to the difficulty of printing electronics. [Brian Minnick]’s fully-printed 3D printer could eventually change that, and he’s already solved an impressive number of technical challenges in the process.
[Brian]’s first step was to make a 3D-printable motor. Instead of the more conventional stepper motors, he designed a fully 3D-printed 3-pole brushed motor. The motor coils are made from solder paste, which the printer applies using a custom syringe-based extruder. The paste is then sintered at a moderate temperature, resulting in traces with a resistivity as low as 0.001 Ω mm, low enough to make effective magnetic coils.
S4 slicer uses the path from any point (here, Benchy’s prow) as its basis…
This non-planar slicer is built into a Jupyter notebook, which follows a relatively simple algorithm to automatically generate non-planar toolpaths for any model. It does this by first generating a tetrahedral mesh of the model and then calculating the shortest possible path through the model from any given tetrahedron to the print bed. Even with non-planar printing, you need to print from the print-bed up (or out).
Quite a lot of math is done to use these paths to calculate a deformation mesh, and we’ll leave that to [Joshua] to explain in his video below. After applying the deformation, he slices the resulting mesh in Cura, before the G-code goes back to Jupyter to be re-transformed, restoring the shape of the original mesh.
… to generate deformed models for slicing, like this.
So yes, it is G-code bending as others have demonstrated before, but in a reproducible, streamlined, and straightforward workflow. Indeed, [Josh] credits much of the work to earlier work on the S^3-Slicer, which inspired much of the logic and the name behind his S4 slicer. (Not S4 as in “more than S^3” but S4 as a contraction of “Simplified S^3”). Once again, open source allows for incremental innovation.
It is admittedly a computationally intensive process, and [Joshua] uses a simplified model of Benchy for this demo. This seems exactly the sort of thing we’d like to burn compute power on, though.
There are a number of reasons you might want to build your own smart speaker virtual assistant. Usually, getting your weather forecast from a snarky, malicious AI potato isn’t one of them, unless you’re a huge Portal fan like [Binh Pham].
[Binh Pham] built the potato incarnation of GLaDOS from the Portal 2 video game with the help of a ReSpeaker Light kit, an ESP32-based board designed for speech recognition and voice control, and as an interface for home assistant running on a Raspberry Pi.
He resisted the temptation to use a real potato as an enclosure and wisely opted instead to print one from a 3D file he found on Thingiverse of the original GLaDOS potato. Providing the assistant with the iconic synthetic voice of GLaDOS was a matter of repackaging an existing voice model for use with Home Assistant.
Of course all of this attention to detail would be for naught if you had to refer to the assistant as “Google” or “Alexa” to get its attention. A bit of custom modelling and on-device wake word detection, and the cyborg tuber was ready to switch lights on and off with it’s signature sinister wit.
If you want to get started in microfluidic robotics, [soiboi soft’s] salamander is probably too complex for a first project. But it is impressive, and we bet you’ll learn something about making this kind of robot in the video below.
The pneumatic muscles are very impressive. They have eight possible positions using three sources of pressure. This seems like one of those things that would have been nearly impossible to fabricate in a home lab a few decades ago and now seems almost trivial. Well, maybe trivial isn’t the right word, but you know what we mean.
The soft robots use layers of microfluidic channels that can be made with a 3D printer. Watching these squishy muscles move in an organic way is fascinating. For right now, the little salamander-like ‘bot has a leash of tubes, but [soiboi] plans to make a self-contained version at some point.
We know you’ve seen them: the time-lapses that show a 3D print coming together layer-by-layer without the extruder taking up half the frame. It takes a little extra work compared to just pointing a camera at the build plate, but it’s worth it to see your prints materialize like magic.
Usually these are done with a plugin for OctoPrint, but with all due respect to that phenomenal project, it’s a lot to get set up if you just want to take some pretty pictures. Which is why [Whopper Printing] put together the LayerLapse. This small PCB is designed to trigger your DSLR or mirrorless camera once its remotely-mounted hall effect sensor detects the presence of a magnet.
The remote hall effect sensor.
The idea is that you just need to stick a small magnet to your extruder, add a bit of extra G-code that will park it over the sensor at the end of each layer, and you’re good to go. There’s even a spare GPIO pin broken out should you want to trigger something else on each layer of your print. Admittedly we can’t think of anything else right now that would make sense, other than some other type of camera, but we’re sure some creative folks out there could put this feature to use.
Currently, [Whopper Printing] is selling the LayerLapse as a finished product, though it does sound like a kit version is in the works. There’s also instructions for building a DIY version of the hardware using your microcontroller of choice. Whether you buy or build the hardware, the firmware is available under the MIT license for your tinkering pleasure.
Being hardware hackers, we appreciate the stand-alone nature of this solution. But if you’re already controlling your printer through OctoPrint, you’re probably better off just setting up one of the available time-lapse plugins.
[kida] has a highly innovative set of 3D-printable, musical fidget toys that play classic video game tunes. Of course there’s the classic Super Mario ditty, but there’s loads more. How they work is pretty nifty, and makes great use of a 3D printer’s strengths.
To play the device one uses a finger to drag a tab (or striker) across the top, and as it does so it twangs vertical tines one-by-one. Each tine emits a particular note — defined by how tall the thicker part is — and plays a short tune as a result. Each one plays a preprogrammed melody, with the tempo and timing up to the user. Listen to them in action in the videos embedded just under the page break!
There are some really clever bits to the design. One is that the gadget is made in two halves, which effectively doubles the notes one can fit into the space. Another is that it’s designed so that holding it against something like a tabletop makes it louder because the surface acts like a sounding board. Finally, the design is easily modified so making new tunes is easy. [kida]’s original design has loads of non-videogame tunes (like the Jeopardy! waiting theme) as well as full instructions on making your very own versions.
Fidget toys are a niche all their own when it comes to 3D printed devices. The fidget knife has a satisfying snap action to it, and this printable linear toggle design is practically a fidget toy all on its own.
![Benchy, printed upside down on [Josh's] Core R-Theta printer.](https://hackaday.com/wp-content/uploads/2025/04/upside-down-benchy-nonplaner-e1745086286623.jpg?w=600&h=450)
