DIY Fume Extractor With ATtiny13 Speed Control

Let’s be honest, commercially-available soldering fume extractors are cheap enough that you probably don’t need to build one yourself. But it still makes for a good starter project, especially if you go out of your way to really flex your maker muscles like [Arnov Sharma] did with this tidy build.

All the hallmarks of modern hardware making are on display here — you’ve got the 3D printed enclosure, a motor salvaged from a cheap toy quadcopter, and a custom PCB which uses the ATtiny13 and an AO4406 MOSFET to implement a PWM speed control.

The first press of the button starts the motor off at max speed, but keep pushing it, and the motor’s speed will ramp down until it turns off entirely. There’s even a TP4056 charge controller to top off the internal 18650 cell when the fume extractor is connected to a USB power source.

Is it over-engineered? Perhaps. But projects like these are a great opportunity to practice your skills, whether it’s PCB design or creating bespoke 3D printed enclosures. In the era of cheap 32-bit microcontrollers, it’s also refreshing to see hackers still dragging the ATtiny from time to time.

Continue reading “DIY Fume Extractor With ATtiny13 Speed Control”

Pretty Pico Macropad Eases Transition Back To Office

[Thomas “Mel” Maillioux] is no stranger to the custom mechanical keyboard game — and faced with having to return to the office, they decided to whip up a sweet little macropad to both commemorate the occasion and make work a bit easier.

This cotton candy-colored block of beauty was designed to pay homage to [Mel]’s favorite joystick, the TRS-80 self-centering number with the single red, square button, and it looks fantastic. They started this journey by studying the key legends on their laptop to determine which macros might serve them best, based on which legends were the most worn.

Fortunately, all the macros they wanted to use — lock the workstation, save the current, active file, minimize/restore all windows, snap windows to the right or left, and volume control — are all macro’d already within Windows, so that made things rather easy.

Hardware-wise, it doesn’t get much easier than a Raspberry Pi Pico, some mechanical switches, an old USB cable, and donor CAT5 pairs so it looks pretty inside and out. Plus, the handy rotary encoder volume knob will mute and un-mute when pressed. We think the snap-fit enclosure looks great, and it needs no supports to boot. If you want to make your own, be sure to check out the repo.

Okay, we lied: macro pad making does get easier, provided you have access to a 3D-printer.

Via MKKC Discord

Machine Learning App Remembers Names So You Don’t Have To

Depending on your point of view, real-life conversations with strangers can either be refreshing or terrifying. Some of us are glib and at ease in new social situations, while others are sure that the slightest flub will haunt them forever. And perhaps chief among these conversational faux pas is forgetting the name of the person who just introduced themselves a few seconds before.

Rather than commit himself to a jail of shame on such occasions, [Caleb] fought back with this only slightly creepy name-recalling smartphone app. The non-zero creep factor comes from the fact that, as [Caleb] points out, the app crosses lines that most of us would find unacceptable if Google or Amazon did it — like listening to your every conversation. It does this not to direct ads to you based on your conversations, but to fish out the name of your interlocutor from the natural flow of the conversation.

It turns out to be a tricky problem, even with the help of named-entity recognition (NER), which basically looks for the names of things in natural text. Apache OpenNLP, the NER library used here, works well at pulling out names, but figuring out whether they’re part of an introduction or just a bit of gossip about a third party is where [Caleb] put the bulk of his coding effort. That, and trying to make the whole thing at least a little privacy-respecting. See the video below for a demo.

To be sure, this doesn’t do much more than a simple, ‘remind me of your name again?’ would, but without the embarrassment. It’s still pretty cool though, and we’re especially jazzed to learn about NER and the tons of applications for it. Those are projects for a future day, though. We’re just glad to see that [Caleb] has moved on from monitoring the bodily functions of his dog and his kid. At least for now.

Continue reading “Machine Learning App Remembers Names So You Don’t Have To”

Nevermore Is What You Get When Engineers Design Air Filters For 3D Printers

What happens when an air filter for 3D printers gets designed by engineers with a passion for function, a refusal to compromise, and a desire to do without bad smells or fumes? You get the Nevermore, a design for a recirculating active-carbon filtration system to deal with VOCs (volatile organic compounds) from 3D printing.

3D-printable parts and an easy-to-fill chamber for bulk-activated carbon make this recirculating air filter for VOCs a smart, space-saving design.

The Nevermore Micro (and larger Nevermore Max) were originally intended to complement the Voron 3D printer design, but are made such that they can be used with just about anything else. These filters use 3D-printable parts, and are designed to be easily filled (and refilled) using bulk-activated carbon instead of some kind of proprietary pre-packed filter like most commercial offerings. The Voron project is all about a printer without compromises, and the Nevermore comes from that same design ethos.

A Nevermore filter sits inside the build chamber, and works by recirculating air inside while passing it through the activated carbon. The idea is that by concentrating on dealing with the problem at the source inside a relatively small build chamber, one doesn’t need a lot of airflow. A small recirculating air filter can do the job efficiently, though for best results, the build chamber should be as sealed as possible.

One interesting caution is that it seems not all activated carbon is the same, and it is absolutely crucial to use only acid-free, steam-activated (not acid-washed) carbon in a recirculating filter like the Nevermore. There are horrifying photos of oxidized metal surfaces resulting from using acid-residue carbon, some of which took only minutes to occur. Thankfully, there are pointers to trusted sources for the known-good stuff.

It’s known that 3D printing results in chemical and particle emissions. These differ significantly depending on both material and type of printer, but it’s enough of an issue to warrant attention. One deals with particulates with something like a HEPA filter, but VOCs require a carbon filter. This is where the Nevermore comes in. Active carbon filters will wear out simply from exposure to the air, so if one is serious about cleaning VOCs when printing, it is definitely worth looking into bulk carbon with a design like the Nevermore.

Hackaday Prize 2022: Treadmill Becomes Human-Powered Generator

Running on a treadmill is a great way to workout, but what if the effort you put in could be put to use? This treadmill generator from [Amitabh Shrivastava] does just that.

The build starts with a regular old treadmill, which has a motor inside typically used to power the tread. Instead, the motor’s control electronics were removed, and it was repurposed to work as a generator. The output from the treadmill’s DC motor was fed directly to a DC-DC converter. This was then fed to an inverter that generates 120 V AC, which can power appliances that use up to 20-25W based on [tinkrmind’s] running performance.

It’s a fun way to generate power during a workout. If you don’t want your treadmill’s monitor to die in the middle of a Friends rerun, you’ll have to dig deep on those long runs. We’ve seen similar builds before too, with exercise bikes being a popular method of generating electricity. In fact, that’s [Amitabh]’s next project! Video after the break.

Continue reading “Hackaday Prize 2022: Treadmill Becomes Human-Powered Generator”

This Pico-W IoT Starter Project Gets You Into Home Assistant Quick As A Flash

Many of us hacker types with some hardware knowledge and a smattering of embedded experience would like to get into home automation, but there can be quite a learning curve. If you’re looking for a hackable starting point; something to deploy, learn about and then later expand upon, then look no further than the PicoW Home Assistant Starter project from [Danilo Campos].

The project is based upon the arduino-pico core, which supports a whole pile of RP2040-based boards, so you don’t need to restrict yourself to the “official” Pico-W, so long as you have working networking, Wi-Fi or otherwise. Integration is provided by the arduino-home-assistant library, which acts as the bridge between your sensors and other widgets, MQTT, and thence the network beyond. Events and sensor data on the end-point are packaged up with MQTT and published out to the broker via the network provided, all for minimal initial effort. Once you’ve got the basic connectivity to your Home Assistant instance working, there are many code examples in the arduino-home-assistant GitHub page to give you a helping start to connect whatever tickles your fancy.

It turns out we’ve covered HA quite a bit on these fair pages, like for example, these sweet automated window blinds. Another hack uses load cells under the bed legs to detect if someone is in bed or not, and if this isn’t your thing, maybe your idea of a home assistant is a bit more like this one?

Streamlining The Toolchain

Sometimes I try to do something magical, and it works. Most of the time this happens because other people have done a good part of the work for me, and shared it. I just cobble a bunch of existing tools into a flow that fits my needs. But the sum of all the parts is often less than the whole, when too many of the steps involve human intervention. Tools made for people simply keep the people in the loop.

For instance, I wanted to take a drawing that my son made into a stamp, by way of a CNC machine and whatever scrap wood we have kicking around in the basement. It’s easy enough, really. Take the photo, maybe use a little tweaking in GIMP to get the levels right, export it into Inkscape for the line detection and maybe even make the GCode right there, or take it off to any convenient SVG-to-GCode tool.

While this works straight out of the box for me, it turns out that’s because I have experience with all of the sub-tools. First, it helps a lot if you get the exposure right in the first place, and that’s not trivial when your camera’s light meter is aiming for grey, but the drawing is on white paper. Knowing this, you could set it up to always overexpose, I guess.

Still, there’s some experience needed in post-processing. If you haven’t played around with both image processing and image editing software, you don’t know how they’re going to interact. And finally, there are more parameters to tweak to get the CNC milling done than a beginner should have to decide.

In short, I had a toolchain up and running in a jiffy, and that’s a success. But in terms of passing it on to my son, it was a failure because he would have to learn way too many sub-tools to make it work for him. Bummer. I’m left wondering if I can streamline all of the parts to work together well enough, or whether I’m simply needed in the loop.