Year: 2020

New Depths For IMac Repair

March 3, 2020 by 17 Comments

There’s not much economic sense in fixing a decade-old desktop computer, especially when it’s the fancy type with the screen integrated into the body of the computer, and the screen is the thing that’s broken. Luckily for [JnsBn] aka [BEAN] the computer in question was still functional with a second monitor, so he decided to implement a cheap repair to get the screen working again by making it see-through.

The only part of the screen that was broken was the backlight, which is separate from the display unit itself. In order to view at least something on the screen without an expensive replacement part, he decided to remove the backlight altogether but leave the display unit installed. With a strip of LEDs around the edge, the screen was visible again in addition to the inner depths of the computer. After a coat of white Plasti Dip on the inside of the computer, it made for an interesting effect and made the computer’s display useful again.

While none of us, including the creator, recommend coating the inside of an iMac with Plasti Dip due to the risk of fire and/or other catastrophic failure, there’s not much to lose otherwise. Just don’t shove this one into the wall. Continue reading “New Depths For IMac Repair” →

Train All The Things Contest Update

March 3, 2020 by 2 Comments

Back in January when we announced the Train All the Things contest, we weren’t sure what kind of entries we’d see. Machine learning is a huge and rapidly evolving field, after all, and the traditional barriers that computationally intensive processes face have been falling just as rapidly. Constraints are fading away, and we want you to explore this wild new world and show us what you come up with.

Where Do You Run Your Algorithms?

To give your effort a little structure, we’ve come up with four broad categories:

  • Machine Learning on the Edge
    • Edge computing, where systems reach out to cloud resources but run locally, is all the rage. It allows you to leverage the power of other people’s computers the cloud for training a model, which is then executed locally. Edge computing is a great way to keep your data local.
  • Machine Learning on the Gateway
    • Pi’s, old routers, what-have-yous – we’ve all got a bunch of devices laying around that bridge space between your local world and the cloud. What can you come up with that takes advantage of this unique computing environment?
  • Machine Learning in the Cloud
    • Forget about subtle — this category unleashes the power of the cloud for your application. Whether it’s Google, Azure, or AWS, show us what you can do with all that raw horsepower at your disposal.
  • Artificial Intelligence Blinky
    • Everyone’s “hardware ‘Hello, world'” is blinking an LED, and this is the machine learning version of that. We want you to use a simple microprocessor to run a machine learning algorithm. Amaze us with what you can make an Arduino do.

These Hackers Trained Their Projects, You Should Too!

We’re a little more than a month into the contest. We’ve seen some interesting entries bit of course we’re hungry for more! Here are a few that have caught our eye so far:

  • Intelligent Bat Detector – [Tegwyn☠Twmffat] has bats in his… backyard, so he built this Jetson Nano-powered device to capture their calls and classify them by species. It’s a fascinating adventure at the intersection of biology and machine learning.
  • Blackjack Robot – RAIN MAN 2.0 is [Evan Juras]’ cure for the casino adage of “The house always wins.” We wouldn’t try taking the Raspberry Pi card counter to Vegas, but it’s a great example of what YOLO can do.
  • AI-enabled Glasses – AI meets AR in ShAIdes, [Nick Bild]’s sunglasses equipped with a camera and Nano to provide a user interface to the world. Wave your hand over a lamp and it turns off. Brilliant!

You’ve got till noon Pacific time on April 7, 2020 to get your entry in, and four winners from each of the four categories will be awarded a $100 Tindie gift card, courtesy of our sponsor Digi-Key. It’s time to ramp up your machine learning efforts and get a project entered! We’d love to see more examples of straight cloud AI applications, and the AI blinky category remains wide open at this point. Get in there and give machine learning a try!

Inputs Of Interest: My First Aggressively Ergonomic Keyboard

March 3, 2020 by 59 Comments

Ever since my RSI surgery, I’ve had to resort to using what I call my compromise keyboard — a wireless rubber dome affair with a gentle curvature to the keys. It’s far from perfect, but it has allowed me to continue to type when I thought I wouldn’t be able to anymore.

This keyboard has served me well, but it’s been nearly three years since the surgery, and I wanted to go back to a nice, clicky keyboard. So a few weeks ago, I dusted off my 1991 IBM Model M. Heck, I did more than that — I ordered a semi-weird hex socket (7/32″) so I could open it up and clean it properly.

And then I used it for half a day or so. It was glorious to hear the buckling springs singing again, but I couldn’t ignore the strain I felt in my pinkies and ring fingers after just a few hours. I knew I had to stop and retire it for good if I wanted to keep being able to type.

Continue reading “Inputs Of Interest: My First Aggressively Ergonomic Keyboard” →

Just When You Thought There Was Nothing New In Nixie Clocks…

March 3, 2020 by 9 Comments

Nixie clocks have become such a staple in our community as to have become mundane. They’re pretty, but show us something new! It seems [Marcin Saj] has done just that with his offering, because with a bank of 18 IN-2 Nixie tubes he’s telling the time –  but in binary rather than the usual decimal.

The tubes are arranged in three banks of six, the upper registering hours, the middle minutes, and seconds on the lowest. Each one only uses two digits, as you might expect from a binary device they are 0 and 1. Behind is a large PCB with the Nixie sockets, and on the back of that in sockets are a pair of Nixie driver boards, a real-time clock module, temperature sensor module, PSU module, and either a Particle Photon or an Arduino Nano IoT.  This two-option set-up for the choice of dev board is unusual, and there is code for both of them in the GitHub repository.

The result is eye-catching and unusual, and certainly a departure from the usual Nixie digital clock. Hackaday readers are probably more likely than the average Joe or Jane to be able to read binary at a glance, watching it in action in the video below the break is an interesting exercise in testing one’s binary-aptitude.

Meanwhile if binary Nixies are too commonplace, how about binary neon lamps?

Continue reading “Just When You Thought There Was Nothing New In Nixie Clocks…” →

Expanding, And Eventually Replacing, The International Space Station

March 3, 2020 by 99 Comments

Aboard the International Space Station (ISS), humanity has managed to maintain an uninterrupted foothold in low Earth orbit for just shy of 20 years. There are people reading these words who have had the ISS orbiting overhead for their entire lives, the first generation born into a truly spacefaring civilization.

But as the saying goes, what goes up must eventually come down. The ISS is at too low of an altitude to remain in orbit indefinitely, and core modules of the structure are already operating years beyond their original design lifetimes. As difficult a decision as it might be for the countries involved, in the not too distant future the $150 billion orbiting outpost will have to be abandoned.

Naturally there’s some debate as to how far off that day is. NASA officially plans to support the Station until at least 2024, and an extension to 2028 or 2030 is considered very likely. Political tensions have made it difficult to get a similar commitment out of the Russian space agency, Roscosmos, but its expected they’ll continue crewing and maintaining their segment as long as NASA does the same. Afterwards, it’s possible Roscosmos will attempt to salvage some of their modules from the ISS so they can be used on a future station.

This close to retirement, any new ISS modules would need to be designed and launched on an exceptionally short timescale. With NASA’s efforts and budget currently focused on the Moon and beyond, the agency has recently turned to private industry for proposals on how they can get the most out of the time that’s left. Unfortunately several of the companies that were in the running to develop commercial Station modules have since backed out, but there’s at least one partner that still seems intent on following through: Axiom.

With management made up of former astronauts and space professionals, including NASA’s former ISS Manager Michael Suffredini and Administrator Charles Bolden, the company boasts a better than average understanding of what it takes to succeed in low Earth orbit. About a month ago, this operational experience helped secure Axiom’s selection by NASA to develop a new habitable module for the US side of the Station by 2024.

While the agreement technically only covers a single module, Axiom hasn’t been shy about their plans going forward. Once that first module is installed and operational, they plan on getting NASA approval to launch several new modules branching off of it. Ultimately, they hope that their “wing” of the International Space Station can be detached and become its own independent commercial station by the end of the decade.

Continue reading “Expanding, And Eventually Replacing, The International Space Station” →

A Honeycomb Patching Robot Powered By Arduino

March 3, 2020 by 13 Comments

No, it’s not the kind of honeycomb you’re probably thinking of. We’re talking about the lightweight panels commonly used in aerospace applications. Apparently they’re rather prone to dents and other damage during handling, so Boeing teamed up with students from the California State University to come up with a way to automate the time-consuming repair process.

The resulting machine, which you can see in action after the break, is a phenomenal piece of engineering. But more than that, it’s an impressive use of off-the-shelf components. The only thing more fascinating than seeing this robotic machine perform its artful repairs is counting how many of its core components you’ve got laying around the shop.

Built from aluminum extrusion, powered by an Arduino Due, and spinning a Dewalt cut-off tool that looks like it was just picked it up from Home Depot, you could easily source most of the hardware yourself. Assuming you needed to automatically repair aerospace-grade honeycomb panels, anyway.

At the heart of this project is a rotating “turret” that holds all the tools required for the repair. After the turret is homed and the condition of all the cutting tools is verified, a hole is drilled into the top of the damaged cell. A small tool is then carefully angled into the hole (a little trick that is mechanical poetry in motion) to deburr the hole, and a vacuum is used to suck out any of the filings created by the previous operations. Finally a nozzle is moved into position and the void is filled with expanding foam.

Boeing says it takes up to four hours for a human to perform this same repair. Frankly, that seems a little crazy to us. But then again if we were the ones tasked with repairing a structural panel for a communications satellite or aircraft worth hundreds of millions of dollars, we’d probably take our time too. The video is obviously sped up so it’s hard to say exactly how long this automated process takes, but it doesn’t seem like it could be much more than a few minutes from start to finish.

Continue reading “A Honeycomb Patching Robot Powered By Arduino” →

Evolution Of A Backpack VR System

March 3, 2020 by 8 Comments

Persistence is what a hacker needs to make it to their goal. That’s exactly what it took for [Erik] to make an untethered VR backpack system.

Starting way back in the Spring of 2019, [Erik] began working on an untethered VR system. Sure, the Oculus Quest was coming out, but it wouldn’t be compatible with the game library of PC based systems. [Erik] decided he wanted the best of both worlds, so he decided to build a backpack that carries a computer powerful enough to drive the Rift S.

The initial system was to use a cut-up backpack, an HP mini PC with an external Nvidia 1060 GPU, and a basic DC-DC converter. The result? Just about nothing worked. The HP’s boot process didn’t play well with an external GPU.

[Erik] went through several iterations of this project. He switched over to a standard PC motherboard and tried a few different DC-DC converters. He settled on a device from HDPLEX rated at 200 watts continuous. The converter plugs directly into a standard 24-pin ATX motherboard power connector and isn’t much larger than the connector itself.

The old backpack with its added padding and wood frame gave way to a Zotac VR go backpack. Only the straps and frame of the Zotac are used, with [Erik’s] custom parts mounted using plywood and 3D printed parts. The outer frame is aluminum, with acrylic panels.

Power comes from 7000 mAH LiFe batteries, with each pack providing an hour of runtime. The Backpack can hold two packs though, so wiring them up in parallel should double that runtime.

We have to say this is an extremely well-documented build. [Erik] explains how he chose each component and the advantages (and pitfalls) of the choices he made. An example would be the RAM he picked. He chose DDR4 with a higher spec than he needed, just so he could undervolt the parts for longer run-times.

Not everything in VR is fun and games though – you can ditch that monitor and go with a VR desktop.