[Build Some Stuff] created an unusual spiral clock that’s almost entirely made from laser-cut wood, even the curved and bendy parts.

The clock works by using a stepper motor and gear to rotate the clock’s face, which consists of a large dial with a spiral structure. Upon this spiral ramp rolls a ball, whose position relative to the printed numbers indicates the time. Each number is an hour, so if the ball is halfway between six and seven, it’s 6:30. At the center of the spiral is a hole, which drops the ball back down to the twelve at the beginning of the spiral so the cycle can repeat.

The video (embedded below) demonstrates the design elements and construction of the clock in greater detail, and of particular interest is how the curved wall of the spiral structure consists of a big living hinge, a way to allow mostly rigid materials to flex far beyond what they are used to. Laser cutting is well-suited to creating living hinges, but it’s a technique applicable to 3D printing, as well.

Thanks to [Kelton] for the tip!

What’s the worst part about packaging up a whole lot of the same basic thing? It might just be applying the various warning stickers to the outside of the shipping box. Luckily, [Mr Innovative] has built an open-source automatic sticker dispenser that does the peeling for you, while advancing the roll one at a time quite satisfyingly.

This tidy build is made primarily of 20×20 extruded aluminium and stainless steel smooth rod. All the yellow bits are 3D printed. The brains of this operation is an Arduino Nano, with an A4988 stepper motor driver controlling a NEMA17.

Our favorite part of this build is the IR sensor pair arranged below the ready sticker. It detects when a sticker is removed, then the stepper advances the roll by one sticker height. The waste is collected on a spool underneath.

Between the video and the instructions, [Mr Innovative] has made it quite simple to build one for yourself. Definitely check this one out after the break.

[Mr Innovative] may as well go by [Mr. Automation]. Check out this automated wire prep machine from a few years ago.

DIY e-bikes are often easy to spot. If they’re not built out of something insane like an old washing machine motor, the more subtle kits that are generally used still stand out when compared to a non-assisted bike. The motors tend to be hub- or mid-drive systems with visible wires leading to a bulky battery, all of which stand out when you know what to look for. To get a stealthy ebike that looks basically the same as a standard bicycle is only possible with proprietary name-brand solutions that don’t lend themselves to owner repair or modification, but this one has at least been adapted for use with an open source motor controller.

The bike in use here is a model called the Curt from Estonian ebike builder Ampler, which is notable in that it looks indistinguishable from a regular bicycle with the exception of the small 36-volt, 350-watt hub motor somewhat hidden in the rear wheel. [BB8] decided based on no reason in particular to replace the proprietary motor controller with one based on VESC, an open-source electric motor controller for all kinds of motors even beyond ebikes. Installed on a tiny Arduino, it fits inside the bike’s downtube to keep the stealthy look and can get the bike comfortably up to around 35 kph. It’s also been programmed to turn on the bike’s lights if the pedals are spun backwards, and this method is also used to change the pedal assist level, meaning less buttons and other user-interface devices on the handlebars.

[BB8] has been working on this for a while, and although the bulk of a working ebike controller is there, it still doesn’t support the torque sensing pedals included with this bike. We’re presuming that this is still a work-in-progress as the Arduino and associated code easily interfaces with all the other sensors available on this bike. Hopefully this open-source motor controller finds its way into other proprietary systems as well, since a lot of these ebikes can turn into massive paperweights if the companies who originally built them go out of business or simply decide to stop supporting older models. Of course, you can avoid this issue entirely by building your own ebike from spare parts.

Thanks to [Arnoud] for the tip!

As fun as claw games are, the jaws are always disappointingly weak, and you usually end up with bupkis. What if the jaws were completely within your  control? That’s the idea behind [Upside Down Labs]’ muscle-controlled servo claw game.

While electromyography (EMG) is great for identifying neuro-muscular abnormalities and allows for amazing prosthetic limbs to work, it can also be used for fun. As you’ll see in the video after the break, accurate block-stacking (and possible candy-grabbing) depends on teamwork and tensed muscles.

Though the user provides the muscle, the brains behind this operation is an Arduino Uno with a Muscle BioAmp shield stacked on top, which [Upside Down Labs] also created. This shield makes it ridiculously easy to connect EMG sensors and other I²C devices like screens and, well, servo claws. From there, it’s really just a matter of printing the claw, connecting it to a 9g servo, and using an accompanying kit to prepare the skin and connect the muscles to the Arduino. Be sure to check it out in tense block-stacking action after the break.

If you want to listen in on your muscles, look no further than the BioAmp EMG Pill.

If you have an iota of musicality, you’ve no doubt noticed that you can play music using glass bottles, especially if you have several of different sizes and fill them with varying levels of water. But what if you wanted to accompany yourself on the bottles? Well, then you’d need to build a bottle-playing robot.

First, [Jens Maker Adventures] wrote a song and condensed it down to eight notes. With a whole lot of tinkling with a butter knife against their collection of wine and other bottles, [Jens] was able to figure out the lowest note for a given bottle by filing it with water, and the highest note by emptying it out.

With the bottle notes selected, the original plan was to strike the bottles with sticks. As it turned out, 9g servos weren’t up to the task, so he went with solenoids instead. Using Boxes.py, he was able to parameterize a just-right bottle holder to allow for arranging the bottles in a circle and striking them from the inside, all while hiding the Arduino and the solenoid driver board. Be sure to check it out after the break.

Don’t have a bunch of bottles lying around? You can use an Arduino to play the glasses.

It’s really quite unfortunate that Hackaday/Supplyframe employees and their families are not allowed to place in the 2023 Halloween Hackfest, because our own [Tom Nardi] has thrown down a costume gauntlet with his kids’ proton pack conversion.

Starting with an inert off-the-shelf toy from 2021, [Tom] set out to make the thing more awesome in every way possible. For one thing, it’s blue, and outside of the short-lived animated series The Real Ghostbusters, who ever heard of a blue proton pack? So one major change was to paint it matte black and age it with the old silver rub ‘n buff technique. And of course, add all the necessary stickers.

[Tom] added plenty of blinkenlights, all running off of an Arduino Nano clone and a pair of 18650s. He got lucky with the whole power cell thing, because an 8 x 5050 RGB LED stick fits there perfectly and looks great behind a PETG diffusing lens. He also drilled out and lit up the cyclotron, because what’s a proton pack without that? There’s even a 7-segment LED voltmeter so Dad can check the power level throughout the night.

Finally, he had to do a bit of engineering to make the thing actually wearable by his daughter. A frame made of square aluminium tubing adds strength, and a new pair of padded straps make it comfortable. Be sure to check it out in action after the break.

What’s a Ghostbusters costume without a PKE meter? 

Have you ever wanted to be the best Super Mario Brothers speedrunner, but you just couldn’t do the frame-perfect inputs? Fear not, because [Gregory Strike] is here to save the day with his automatic NES controller!

In his previous video, [Greg] already made an automatic controller that plays a sequence of inputs at the perfect time, but it still failed some of the frame-perfect tricks. So what gives? Deviation in the timing of the NES itself gives, as he shows how the NES doesn’t sample inputs at exactly the same time every frame. To account for this, he used the latch signal, which starts the controller reading process as a time reference, and replaced his digital “mixtape” with a more time-flexible Arduino. After the modification, he shows it pulling off frame-perfect inputs every time he plays Super Mario Brothers.

But if you have a controller that can do frame-perfect inputs and it can be connected to a computer, you can connect the controller to the internet! That’s right, [Greg] created a Twitch bot that tells the Arduino exactly what inputs to send, which then relays it to the NES. It accepts simple sequences of inputs via chat, and you can try it out right now on [Greg]’s Twitch stream.

This project shows promising results, and we think it’s possible to do much more with its internet connection. We’re certainly looking forward to what [Greg] decides to make next.

Video after the break.

DIY electric personal vehicles are a field where even hobbyists can meaningfully innovate, and that’s demonstrated by the Mega-Wheelie, a self-balancing one-wheeled skateboard constructed as an experiment in traversing off-road conditions.

[John Dingley] and [Nick Thatcher] have been building and testing self-balancing electric vehicles since 2008, with a beach being a common testing ground. They suspected that a larger wheel was the key to working better on rough ground and dry sand and tested this idea by creating a skateboard with a single wheel. A very big, very wide wheel, in fact.

The Mega-Wheelie houses a 24V LiFePO4 battery pack, 450 W gearmotor with chain and sprocket drive, SyRen motor controller from Dimension Engineering, Arduino microcontroller, and an inertial measurement unit to enable the self-balancing function. Steering is done by leaning, and the handheld controller is just a dead man’s switch that disables the vehicle if the person piloting it lets go.

Design-wise, a device like this has a few challenging constraints. A big wheel is essential for performance but takes up space that could otherwise be used for things like batteries. Also, the platform upon which the pilot stands needs to be as low to the ground as possible for maximum stability. Otherwise, it’s too easy to fall sideways. On the other hand, one must balance this against the need for sufficient ground clearance.

In the end, how well did it work? Well enough to warrant a future version, says [John]. We can’t wait to see what that looks like, considering their past 3000 W unicycle’s only limitation was “personal courage” and featured a slick mechanism that shifted the pilot’s weight subtly to aid steering. A video of the Mega-Wheelie (and a more recent unicycle design) is embedded just below the page break.

And just for reference, here is some of [John]’s previous work on a self-balancing unicycle design.

We’re on a sort of vacation in Tanzania at the moment and staying in a modest hotel away from the tourist and government district. It’s a district of small shops selling the same things and guys repairing washing machines out on the sidewalk. The guys repairing washing machines are more than happy to talk. Everybody’s amazingly friendly here, the hotel guy grilled us for an hour about our home state. But I really didn’t expect to end up in a conversation about computer vision.

In search of some yogurt and maybe something cooler to wear, we went on a little walk away from the hotel. With incredible luck we found a robotics shop a few blocks away. Mecktonix is a shop about two meters each way, stuffed full of Arduinos, robots, electronics components, servos, and random computer gear, overseen by [Yohanna “Joe” Harembo]. Nearby is another space with a laser engraver and 3D printer. The tiny space doesn’t stop them from being busy. A constant stream of automotive tech students from the nearby National Institute of Transport shuffle in for advice and parts for class assigned projects.

In between students, Joe demos an autonomous car he’s working on. In classic hacker fashion, he first has to reattach the motor driver board and various sensors, but then he demos the car and its problem –  the video frame rate is very slow. We dive in with him and try to get some profiling using time.monotonic_ns(). He’s never done profiling before, so this is a big eye opener. He’s only processing one video frame every 4.3 seconds, using YOLO on a Pi 3, and yup, that’s the problem.  I suggest he change to gamut detection or a Pi 4.

Gamut Detection

If you’re not familiar with gamut detection, it’s one of the simplest of all computer vision techniques, so it’s easy to implement on slow processors and almost trivial to code. Basically, it’s “look for a color”.  If you want your robot to follow you, wear a lime green T shirt. Now the robot just has to look for lime green. Same for catching a ball or following a line. The algorithm is simple – convert each pixel to HSV, where hue corresponds to the direction around the color wheel,  saturation corresponds to how concentrated the color is, and value how bright. Brightness depends on the lighting, so you can throw value away and just set limits for H and S. Anything within those limits is part of our target. The box formed by those limits is our “gamut”.

There are a couple speedups you can apply as well. First, ask yourself how much resolution you need from the camera. If you only want to track a green T shirt that’s never less than 24 pixels on screen, turn the resolution down by a factor of six each way and look for four-pixel T shirts. You now have 1/36th as much data to process and your algorithm runs 36 times faster. If you can’t control the camera resolution, you can shrink the image or just sample every nth pixel. Second, you can often ask for a YUV or YIQ image from the camera. Discard Y and set your limits in IQ or UV coordinates. It’s about the same as HSV.

Joe’s eating this up – he’s had limited chances to talk with somebody else who is into computer vision. As we write this, he’s still trying with YOLO, but at lower resolution. If it doesn’t work he’ll try gamut detection. And it’s not his only project. Passenger carrying motorcycles called pikipikis are common here. A student has a project to enforce passengers wearing a helmet, and we fiddle with the student’s project.

The Dar es Salaam Scene

There’s other tech happening in Tanzania too. A few blocks away is [Ruta Electronics], a similar sugar cube sized shop developing smart meters. Everything from cases to PCB etching happens in the tiny shop. Downtown there are a few tech startups. There’s a fab lab, mostly oriented towards children. And on a quiet side street off the main drag, there’s a tiny shop with three guys who are hacking like crazy.

For us, we’ve had a chance to make a friend from a different culture and play with a robot car together — what could be better?  When you’re traveling, are you on the lookout for other hackers or hackerspaces? It’s worth the effort and brings our community together in a way that even the internet can’t.

With how cheap they’re getting, everyone seems to be jumping on the resin printer bandwagon. They may not be able to fully replace your trusty old FDM printer, but for certain jobs, they just can’t be beaten. Sadly though, creating those smooth time-lapse videos of your prints isn’t quite as easy to do as it is on their filament-based counterparts.

Not as easy, perhaps, but not impossible. [Fraens] found a way to make time-lapses on any resin printer, and in a wonderfully hacky way. First, you need to find a smartphone, which shouldn’t be too hard, given how often we all tend to upgrade. [Fraens] recommends replacing the standard camera app on the phone with Open Camera, to prevent it from closing during the long intervals with nothing happening. The camera is triggered by any readily available Bluetooth dongle, which is connected via a simple transistor circuit to an Arduino output. To trigger the shutter, a light-dependent resistor (LDR) is connected to one of the microcontroller’s inputs. The LDR is placed inside the bed of the resin printer — an Anycubic Photon in this case — where light from the UV panel used to cross-link the resin can fall on it. A simple bit of Arduino code triggers the Bluetooth dongle at the right moment, capturing a series of stills which are later stitched together using DaVinci Resolve.

The short video below shows the results, which look pretty good to us. There are other ways to do this, of course, but we find the simplicity of this method pleasing.