The Screamer Is Just Like The Clapper But Even More Annoying

Remember The Clapper? It was a home automation tool (of sorts) that let you turn appliances on and off by clapping. [Kevin O’Connor] has built something rather similar, if more terrifying. It’s called The Screamer.

The build is based around a Sonoff S31 smart switch. [Kevin] selected an off-the-shelf device because he wanted something that was safe to use with mains power out of the box. But specifically, he selected the S31 because it has an ESP8266 inside that’s easy to reprogram with the aid of ESPHome. He ended up hooking up a whole extra ESP32 with an INMP441 microphone over I2S to do the scream detection. This was achieved with a simple algorithm that looked for high amplitude noises with lots of energy in the 1000 – 4000 Hz frequency range. When a scream is detected, it flips a GPIO pin which is detected by the S31, which then toggles the state of the smart switch in turn. Job done.

It’s a simple project that does exactly what it says on the tin. It’s The Screamer! If you’d like to learn more about the original Clapper that inspired this build, we’ve looked at that before, too. Meanwhile, if you’re cooking up your own excuses to scream at the lights and walls in your home, please only notify the tipsline if it has something vaguely to do with electronics or hackery.

ChatGPT & Me. ChatGPT Is Me!

For a while now part of my email signature has been a quote from a Hackaday commenter insinuating that an article I wrote was created by a “Dumb AI”. You have my sincerest promise that I am a humble meatbag scribe just like the rest of you, indeed one currently nursing a sore shoulder due to a sporting injury, so I found the comment funny in a way its writer probably didn’t intend. Like many in tech, I maintain a skepticism about the future role of large-language-model generative AI, and have resisted the urge to drink the Kool-Aid you will see liberally flowing at the moment.

Hackaday Is Part Of The Machine

As you’ll no doubt be aware, these large language models work by gathering a vast corpus of text, and doing their computational tricks to generate their output by inferring from that data. They can thus create an artwork in the style of a painter who receives no reward for the image, or a book in the voice of an author who may be struggling to make ends meet. From the viewpoint of content creators and intellectual property owners, it’s theft on a grand scale, and you’ll find plenty of legal battles seeking to establish the boundaries of the field.

Anyway, once an LLM has enough text from a particular source, it can do a pretty good job of writing in that style. ChatGPT for example has doubtless crawled the whole of Hackaday, and since I’ve written thousands of articles in my nearly a decade here, it’s got a significant corpus of my work. Could it write in my style? As it turns out, yes it can, but not exactly. I set out to test its forging skill. Continue reading “ChatGPT & Me. ChatGPT Is Me!”

Hackaday Podcast Episode 321: Learn You Some 3DP, Let The Wookie Win, Or Design A Thinkpad Motherboard Anew

Join Hackaday Editors Elliot Williams and Tom Nardi as they take a whirlwind tour of the best and brightest hacks of the last week. This episode starts off with an update about that Soviet Venus lander that’s been buzzing the planet, then moves on to best practices for designing 3D printed parts, giving Chrome OS devices a new lease on life, and a unique display technology that brings a Star Wars prop to life.

You’ll also hear about designing new motherboards for beloved old computers, why you might want to put your calipers on a flatbed scanner, and a NASA science satellite that’s putting in double duty as a wartime reporter. Finally, they’ll cover the interesting physics of meteor burst communications, and the latest developments in the ongoing libogc license kerfuffle.

Check out the links below if you want to follow along, and as always, tell us what you think about this episode in the comments!

Download in DRM-free MP3.

Continue reading “Hackaday Podcast Episode 321: Learn You Some 3DP, Let The Wookie Win, Or Design A Thinkpad Motherboard Anew”

Series of purple and red mechanisms are stretched from left to right. Almost like arrows pointing right.

Compliant Mechanism Shrinks Instead Of Stretching

Intuitively, you think that everything that you stretch will pull back, but you wouldn’t expect a couple of pieces of plastic to win. Yet, researchers over at [AMOLF] have figured out a way to make a mechanism that will eventually shrink once you pull it enough.

Named “Counter-snapping instabilities”, the mechanism is made out of the main sub-components that act together to stretch a certain amount until a threshold is met. Then the units work together and contract until they’re shorter than their initial length. This is possible by using compliant joints that make up each of the units. We’ve seen a similar concept in robotics.

The picture reads "Excessive vibrations? / It tames them by itself... / ... by switching them off! Bridge undergoing harmonic oscillation about to crumble on the left and mechanisms on the right.

Potentially this may be used as a unidirectional actuator, allowing movement inch by inch. In addition, one application mentioned may be somewhat surprising: damping. If a structure or body is oscillating through a positive feedback loop it may continue till it becomes uncontrollable. If these units are used, after a certain threshold of oscillation the units will lock and retract, therefore stopping further escalation.

Made possible by the wonders of compliant mechanics, these shrinking instabilities show a clever solution to some potential niche applications. If you want to explore the exciting world of compliance further, don’t be scared to check out this easy to print blaster design!

Continue reading “Compliant Mechanism Shrinks Instead Of Stretching”

This Week In Security: Lingering Spectre, Deep Fakes, And CoreAudio

Spectre lives. We’ve got two separate pieces of research, each finding new processor primitives that allow Spectre-style memory leaks. Before we dive into the details of the new techniques, let’s quickly remind ourselves what Spectre is. Modern CPUs use a variety of clever tricks to execute code faster, and one of the stumbling blocks is memory latency. When a program reaches a branch in execution, the program will proceed in one of two possible directions, and it’s often a value from memory that determines which branch is taken. Rather than wait for the memory to be fetched, modern CPUs will predict which branch execution will take, and speculatively execute the code down that branch. Once the memory is fetched and the branch is properly evaluated, the speculatively executed code is rewound if the guess was wrong, or made authoritative if the guess was correct. Spectre is the realization that incorrect branch prediction can change the contents of the CPU cache, and those changes can be detected through cache timing measurements. The end result is that arbitrary system memory can be leaked from a low privileged or even sandboxed user process.

In response to Spectre, OS developers and CPU designers have added domain isolation protections, that prevent branch prediction poisoning in an attack process from affecting the branch prediction in the kernel or another process. Training Solo is the clever idea from VUSec that branch prediction poisoning could just be done from within the kernel space, and avoid any domain switching at all. That can be done through cBPF, the classic Berkeley Packet Filter (BPF) kernel VM. By default, all users on a Linux system can run cBPF code, throwing the doors back open for Spectre shenanigans. There’s also an address collision attack where an unrelated branch can be used to train a target branch. Researchers also discovered a pair of CVEs in Intel’s CPUs, where prediction training was broken in specific cases, allowing for a wild 17 kB/sec memory leak.

Continue reading “This Week In Security: Lingering Spectre, Deep Fakes, And CoreAudio”

There are a number of metal cylinders displayed in a line. Each cylinder has a rectangular brass plate mounted to each end, and these brass plates stand upright, with the metal cylinders held horizontally between them.

Home-casting Thermoelectric Alloys

If you want to convert heat into electrical power, it’s hard to find a simpler method than a thermoelectric generator. The Seebeck effect means that the junction of two dissimilar conductors will produce a voltage potential when heated, but the same effect also applies to certain alloys, even without a junction. [Simplifier] has been trying to find the best maker-friendly thermoelectric alloys, and recently shared the results of some extensive experimentation.

The experiments investigated a variety of bismuth alloys, and tried to determine the effects of adding lead, antimony, tin, and zinc. [Simplifier] mixed together each alloy in an electric furnace, cast it into a cylindrical mold, machined the resulting rod to a uniform length, and used tin-bismuth solder to connect each end to a brass electrode. To test each composition, one end of the cylinder was cooled with ice while the other was held in boiling water, then resistance was measured under this known temperature gradient. According to the Wiedemann-Franz law, this was enough information to approximate the metal’s thermal conductivity.

Armed with the necessary data, [Simplifier] was able to calculate each alloy’s thermoelectric efficiency coefficient. The results showed some useful information: antimony is a useful additive at about 5% by weight, tin and lead created relatively good thermoelectric materials with opposite polarities, and zinc was useful only to improve the mechanical properties at the expense of efficiency. Even in the best case, the thermoelectric efficiency didn’t exceed 6.9%, which is nonetheless quite respectable for a homemade material.

This project is a great deal more accessible for an amateur than previous thermoelectric material research we’ve covered, and a bit more efficient than another home project we’ve seen. If you just want to get straight to power generation, check out this project.

Internals of ding-dong doorbell.

Wireless Doorbell Extension Features Home-Wound Coil

Today in the it’s-surprising-that-it-works department we have a ding dong doorbell extension from [Ajoy Raman].

What [Ajoy] wanted to do was to extend the range of his existing doorbell so that he could hear it in his workshop. His plan of attack was to buy a new wireless doorbell and then interface its transmitter with his existing doorbell. But his approach is something others might not have considered if they had have been tasked with this job, and it’s surprising to learn that it works!

What he’s done is wrap a new coil around the ding dong doorbell’s solenoid. When the solenoid activates, a small voltage is induced into the coil. This then gets run into the wireless doorbell transmitter power supply (instead of its battery) via a rectifier diode and a filter capacitor. The wireless doorbell transmitter — having also had its push-button shorted out — operates for long enough from this induced electrical pulse to transmit the signal to the receiver. To be clear: the wireless transmitter is fully powered by the pulse from the coil around the solenoid. Brilliant! Nice hack!

We weren’t sure how reliable the transmitter would be when taken out of the lab and installed in the house so we checked in with [Ajoy] to find out. It’s in production now and operating well at a distance of around 50 feet!

Of course we’ve published heaps of doorbell hacks here on Hackaday before, such as this Bluetooth Low Energy (BLE) doorbell and this light-flashing doorbell. Have you hacked your own doorbell? Let us know on the tips line!

Continue reading “Wireless Doorbell Extension Features Home-Wound Coil”