NTC Thermistor To ThingSpeak Meter Makes A Great IoT Starter Project

December 20, 2022 by Donald Papp 5 Comments

There are a lot of IoT solutions and frameworks out there, and [Davide] demonstrates how to make a simple data logging and tracking application with his ESP8266-to-ThingSpeak project, which reads up to four NTC (negative temperature coefficient) thermistors and sends the data to ThinkSpeak over WiFi.

IoT can be a pretty deep rabbit hole, so if you’re looking for a simple project to demonstrate the working parts and provide a starting point, the project’s GitHub repository might help you get started. We’ve also seen ThingSpeak used to track toilet paper sheet usage, which is a nice demonstration of how to interface to a physical object with moving parts.

On the other hand, if you find reading NTC thermistors to be the more interesting part, you’re in luck because [Davide] has more information about that along with a modified ESP8266 Arduino library. Watch a tour of his temperature logging hardware in action in the video, embedded below.

Continue reading “NTC Thermistor To ThingSpeak Meter Makes A Great IoT Starter Project” →

Seriously, Don’t Buy This Mopping Robot

December 20, 2022 by Donald Papp 18 Comments

The original Roomba robotic vacuum cleaner led to loads of clones and lookalikes over the years, and one of them is the ALEE mopping “robot”. [Raymond] tears it down and reveals what’s inside. Turns out it contains mostly regret! Although it does host some design cleverness in its own way.

Technically the ALEE, which cost [Raymond] a cool $85 USD, is not a robot since it has no sensors. And unless a dragging a wet cloth pad kept moist by a crude drip reservoir counts as “mopping”, it’s not much of a mop, either.

This one-motor unit (and tiny battery) is responsible for both motion and direction control. There are no sensors.

There is one interesting aspect to this thing, and it’s to do with the drive system and direction control. The whole thing is driven by a single motor, and not a very powerful one. The center of the robot has a pair of wheels that are both driven at the same rate and speed, and the wheel assembly can pivot around its axis. That’s about it. There are not even any bump sensors of any kind.

So how does this thing move, let alone change direction to (poorly) emulate an original Roomba-like crisscross pattern? The control board appears to have one job: if the motor stalls, reverse direction. That, combined with the fact that the drive unit can pivot and the enclosure is dragging a wet rag, appears to be all the chaos that’s needed to turn bonking into a wall into an undefined direction change.

It’s not great performance, but it sure is some impressive cost-cutting. You can see it bonk around unimpressively in a short video, embedded below the page break.

Just to be clear, [Raymond] knows perfectly well what he’s in for when he obtains cheap tech items from overseas retailers for teardowns. The ALEE does have some mildly interesting secrets to share, but overall, it really wasn’t worth it. Sometimes cheap tech has hacker potential, but there’s no such potential here. Seriously, don’t buy this thing.

Continue reading “Seriously, Don’t Buy This Mopping Robot” →

See The ATARI GEM Desktop Running On A Portable Word Processor… Thing

December 20, 2022 by Donald Papp 14 Comments

Get ready for vintage computing aplenty in [David Given]’s project to port EmuTOS to the AlphaSmart Dana. He’s got it all on video, too. All 38 hours of it over 13 episodes!

The GEM desktop, as seen on the Atari ST line of computers.

[David]’s fork of EmuTOS is an open source version of the Atari TOS, which is itself the 68000-based OS for the Atari ST line of computers.

As for the AlphaSmart Dana, it is a roughly twenty-year-old portable word processor thing with pen input which runs a version of PalmOS. It’s a slightly oddball piece of hardware, but quite capable in its own way. A match obviously made in heaven? It is if you have [David]’s skill and drive!

To get EmuTOS working on the Dana, the first step was figuring out how to find and work with the Dana’s debug port, using it to get direct access to the CPU while bypassing the boot ROM. Turns out that the Dana’s 68000-compatible processor has a handy feature: by manipulating the right pin, one can remote-control the CPU (to a certain extent) via the UARTs. That’s the entry point for a whole lot of hacking that ultimately results in firing up the GEM desktop on the Dana, and being able to run (some) original Atari ST software. Probably the biggest issue is that the screen size isn’t a great match for what the OS expects, but it works.

Continue reading “See The ATARI GEM Desktop Running On A Portable Word Processor… Thing” →

Snail Mail Notifier’s Simple Power Management To Maximize Battery Life

December 19, 2022 by Donald Papp 33 Comments

There are no weird, specialized components nor esoteric sleep mode tricks behind the long battery life of [Zak]’s WiFi mail slot watcher. Just some sensible design and clever focus on the device’s purpose: to send an HTTP request whenever it detects that the front door’s mail slot has been opened. The HTTP request is what kicks off useful notifications, but it’s the hardware design that’s really worth a peek.

The watcher’s main components are a ESP-M2 WiFi module, a reed switch, and a single lithium cell. Here’s how it works at a high level: when the mail slot is opened (detected by the reed switch), the ESP module is powered up just long enough to connect to the local WiFi network and send a single HTTP request, after which it shuts back down. The whole process takes between four and ten seconds.

As mentioned, the power control isn’t managed by any unusual components; it comes down to a NAND gate with a single inverted input, and a MIC5504 3.3 V regulator responsible for feeding the ESP board. The logic gate controls whether the voltage regulator is enabled or disabled, and therefore whether the microcontroller receives any power at all. Most of the time the regulator is disabled, but when the reed switch triggers, its input to the NAND gate is pulled low and the regulator is turned on, booting up the ESP board.

In order to stay on, the first thing the ESP board does is use a GPIO pin to drive the inverted input of the NAND gate high in order to keep the regulator enabled, and it has a window of about half a second to do this. Once the HTTP request is sent (and the battery voltage sensed), the ESP board pulls that pin low, disabling the regulator and turning itself off until the reed switch once again begins the process.

After seven months of use, the battery has dropped from 4.2 V to 3.9 V, so there’s plenty of life left. The project’s GitHub repository has the necessary code if you’d like to apply some of its ideas to your own projects. Alternately, you may wish to consider supercapacitors and solar in lieu of batteries. Even if ultra-level power savings isn’t your bag, when WiFi and networking is involved, there are software-level opportunities to be more efficient. Even a judicious 1 ms delay can save a surprising amount of power in the right circumstances.

Mini Cheetah Clone Teardown, By None Other Than Original Designer

December 19, 2022 by Donald Papp 21 Comments

[Ben Katz] designed the original MIT Mini Cheetah robot, which easily captured attention and imagination with its decidedly un-robotic movements and backflips. Not long after [Ben]’s masters thesis went online, clones of the actuators started to show up at overseas sellers, and a few months after that, clones of the whole robot. [Ben] recently had the opportunity to disassemble just such a clone by Dogotix and see what was inside.

Amusingly, one of the first things he noticed is that the “feet” are still just off-the-shelf squash balls, same as his original mini cheetah design. As for the rest of the leg, inside is a belt that goes past some tensioners, connecting the knee joint to an actuator in the shoulder.

As one may expect, these parts are subject to a fair bit of stress, so they have to be sturdy. This design allows for slender yet strong legs without putting an actuator in the knee joint, and you may recall we’ve seen a similar robot gain the ability to stand with the addition of a rigid brace.

It’s interesting to read [Ben]’s thoughts as he disassembles and photographs the unit, and you’ll have to read his post to catch them all. But in the meantime, why not take a moment to see how a neighbor’s curious sheep react to the robot in the video embedded below? The robot botches a backflip due to a low battery, but the sheep seem suitably impressed anyway.

Continue reading “Mini Cheetah Clone Teardown, By None Other Than Original Designer” →

Image-Generating AI Can Texture An Entire 3D Scene In Blender

December 18, 2022 by Donald Papp 26 Comments

[Carson Katri] has a fantastic solution to easily add textures to 3D scenes in Blender: have an image-generating AI create the texture on demand, and do it for you.

It’s not perfect — the odd door or window feature might suffer from a lack of right angles — but it’s pretty amazing.

As shown here, two featureless blocks on a featureless plain become run-down buildings by wrapping the 3D objects in a suitable image. It’s all done with the help of the Dream Textures add-on for Blender.

The solution uses Stable Diffusion to generate a texture for a scene based on a text prompt (e.g. “sci-fi abandoned buildings”), and leverages an understanding of a scene’s depth for best results. The AI-generated results aren’t always entirely perfect, but the process is pretty amazing. Not to mention fantastically fast compared to creating from scratch.

AI image generation capabilities are progressing at a breakneck pace, and giving people access to tools that can be run locally is what drives interesting and useful applications like this one here.

Curious to know more about how systems like Stable Diffusion work? Here’s a pretty good technical primer, and the Washington Post recently published a less-technical (but accurate) interactive article explaining how AI image generators work, as well as the impact they are having.

Antweight Combat Robot Tips, Shared From Experience

December 18, 2022 by Donald Papp 1 Comment

[Harry]’s newest robot, the MotherLoader V2, looks fantastic but was ultimately more of a learning experience and test bed for some experimental features. Luckily for us, [Harry] created a lengthy write-up detailing everything that he tried and revised.

3D printing and aluminum both feature heavily in antweight robots, in part because when contestants are limited to 150 grams it’s safe to say that every bit counts. We recommend reading [Harry]’s entire article to get all the details, but here are some of the bigger takeaways.

Treads provide a lot of contact surface, but there are a lot of ways they can go wrong. Pliability and grip have to be good matches for the robot’s design, otherwise the tread might bunch up or otherwise perform poorly when trying to maneuver. [Harry] had several dud efforts, but ended up with a great result by borrowing an idea from another competitor: composite tracks.

These have an inner track printed from flexible TPU filament, and an outer layer formed by casting silicone directly onto the 3D printed core. It’s a somewhat involved process, but the result is a durable and custom-fitted inner track on the inside, and a softer grip outside. Best of both worlds, and easily tailored to match requirements.

Speaking of TPU, [Harry] discovered that it can be worth printing structural parts with TPU. While ABS is usually the material of choice for durable components, printing solid parts in TPU has a lot to recommend it when it comes to 150 gram robots. Not only can TPU parts be stiff enough to hold up structurally, but they can really take a beating and happily spring back into shape afterwards.

We’ve seen [Harry]’s work before on antweight combat robots, and it’s always nice to peek behind the scenes and gaze into the details. Especially for processes like this, where failures are far more educational than successes.