Author: Dan Maloney

3211 Articles

Blame It On The Sockets: Forensic Analysis Of The Arecibo Collapse

August 29, 2023 by 58 Comments

Nearly three years after the rapid unplanned disassembly of the Arecibo radio telescope, we finally have a culprit in the collapse: bad sockets.

In case you somehow missed it, back in 2020 we started getting ominous reports that the cables supporting the 900-ton instrument platform above the 300-meter primary reflector of what was at the time the world’s largest radio telescope were slowly coming undone. From the first sign of problems in August, when the first broken cable smashed a hole in the reflector, to the failure of a second cable in November, it surely seemed like Arecibo’s days were numbered, and that it would fall victim to all the other bad luck we seemed to be rapidly accruing in that fateful year. The inevitable finally happened on December 1, when over-stressed cables on support tower four finally gave way, sending the platform on a graceful swing into the side of the natural depression that cradled the reflector, damaging the telescope beyond all hope of repair.

The long run-up to the telescope’s final act had a silver lining in that it provided engineers and scientists with a chance to carefully observe the failure in real-time. So there was no real mystery as to what happened, at least from a big-picture perspective. But one always wants to know the fine-scale details of such failures, a task which fell to forensic investigation firm Thornton Tomasetti. They enlisted the help of the Columbia University Strength of Materials lab, which sent pieces of the failed cable to the Oak Ridge National Laboratory’s High Flux Isotope reactor for neutron imaging, which is like an X-ray study but uses streams of neutrons that interact with the material’s nuclei rather than their electrons.

The full report (PDF) reveals five proximate causes for the collapse, chief of which is “[T]he manual and inconsistent splay of the wires during cable socketing,” which we take to mean that the individual strands of the cables were not spread out correctly before the molten zinc “spelter socket” was molded around them. The resulting shear stress caused the zinc to slowly flow around the cable strands, letting them slip out of the surrounding steel socket and — well, you can watch the rest below for yourself.

As is usually the case with such failures, there are multiple causes, all of which are covered in the 300+ page report. But being able to pin the bulk of the failure on a single, easily understood — and easily addressed — defect is comforting, in a way. It’s cold comfort to astronomers and Arecibo staff, perhaps, but at least it’s a lesson that might prevent future failures of cable-supported structures.

Continue reading “Blame It On The Sockets: Forensic Analysis Of The Arecibo Collapse” →

Hackaday Links Column Banner

Hackaday Links: August 27, 2023

August 27, 2023 by 26 Comments

We mentioned last week how robotaxi provider Cruise was having a no-good, very bad week, after one of their driverless taxis picked a fight with a semi, and it was revealed that amorous San Franciscans were taking advantage of the privacy afforded by not having a driver in the front seat. It appears that we weren’t the only ones to notice all the bad news, since California’s Department of Motor Vehicles issued an order to the company to cut its robotaxi fleet in half. The regulatory move comes after a recent Cruise collision with a fire truck, which injured a passenger in the taxi. Curiously, the DMV order stipulates that Cruise can only operate 50 vehicles during the day, while allowing 150 vehicles at night. We’d have thought the opposite would make more sense, since driving at night is generally more difficult than during daylight hours. But perhaps the logic is that the streets are less crowded at night, whereas daytime is a more target-rich environment.

Continue reading “Hackaday Links: August 27, 2023” →

RPDot: The RP2040 Dev Board Barely Bigger Than The Chip

August 25, 2023 by 33 Comments

Is [William Herr]’s RPDot actually the world’s smallest RP2040 dev board? We can’t say for sure, but at 10 mm on a side, we’d say it has a pretty good shot at the record.

Not that it really matters, mind you — the technical feat of building a fully functional dev board that’s only 3 mm longer on each side than the main chip is the kind of stuff we love to see. [William] says he took inspiration from the [SolderParty] RP2040 Stamp, which at one inch (25.4 mm) on a side is gigantic compared to the RPDot. Getting the RP2040 and all the support components, which include an 8MB QSPI Flash chip, a 3V3 LDO, a handful of 0201 passives, and even a pair of pushbuttons, required quite a lot of design tweaking. He started his PCB design as a four-layer board; while six layers would have made things easier, the budget wouldn’t allow such extravagance for a prototype. Still, he somehow managed to stuff everything in the allotted space and send the designs off — only to get back defective boards.

After reordering from a different vendor, the real fun began. Most of the components went on the front side of the board and were reflowed using a hot plate. The RP2040 itself needed to go on the back side, which required gentle hot air reflow so as not to disrupt the other side of the board. The results look pretty good, although those castellated edges look a little worse for the wear. Still, for someone who only ever worked with 0402 components before, it’s pretty impressive.

[William] says he’s going to open-source the designs as well as make some available for sale. We’ll be looking out for those and other developments, but for now, it’s just pretty cool to see such SMD heroics.

Doppler Speed Sensor Puts FFT And AGC To Work

August 24, 2023 by 6 Comments

Some people hate to revisit projects that are done and dusted. We get that; it’s a little like reading a book you’ve already read when there are so many others to choose from. But rereading a book sometimes reveals subtle nuances you missed the first time around, and revisiting projects can be much the same, as with this new and improved Doppler radar speed sensor.

We seem to have been remiss in writing up [Limpkin]’s last go-around with the CDM324 microwave module, a 24-GHz transceiver that you can pick up on the cheap from the usual sources, but we’ve featured this handy little module in plenty of other projects. [Limpkin]’s current project uses the same module to create a Doppler speed sensor, but with a little more sophistication all around. Whereas the original used a simple comparator to output a square wave that’s proportional to the Doppler shift and displayed the speed on a simple terminal session, version two takes a different tack.

First, [Limpkin] opted to implement the whole sensor in hardware. The front end is quite different — an op-amp with 84 dB of gain followed by an automatic gain control (AGC) stage built from a MAX9814 microphone preamp. Extraction of the speed from the module output is left to an STM32F301 running an FFT algorithm on the signal coming out of the analog circuit, which essentially picks out the biggest peak in the spectrum and calculates the Doppler shift from that, displaying the results on an LCD display.

Of course, as a [Limpkin] project, there’s a lot more to it than just that. The write-up is very detailed, going down a few enjoyable rabbit holes like characterizing the amplification chain and diving into the details of Johnson-Nyquist noise to chase down stray oscillations. There’s some great stuff here, and it’s well worth a deep read; there’s also the video below that lets you see (and hear) what’s going on.

Continue reading “Doppler Speed Sensor Puts FFT And AGC To Work” →

Guitar Distortion With Diodes In Code, Not Hardware

August 23, 2023 by 14 Comments

Guitarists will do just about anything to get just the right sound out of their setup, including purposely introducing all manner of distortion into the signal. It seems counter-intuitive, but it works, at least when it’s done right. But what exactly is going on with the signal? And is there a way to simulate it? Of course there is, and all it takes is a little math and some Arduino code.

Now, there are a lot of different techniques for modifying the signal from an electric guitar, but perhaps the simplest is the humble diode clipping circuit. It just uses an op-amp with antiparallel diodes either in series in the feedback loop or shunting the output to ground. The diodes clip the tops and bottoms off of the sine waves, turning them into something closer to a square wave, adding those extra harmonics that really fatten the sound. It’s a simple hack that’s easy to implement in hardware, enough so that distortion pedals galore are commercially available.

In the video below, [Sebastian] explains that this distortion is also pretty easy to reproduce algorithmically. He breaks down the math behind this, which is actually pretty approachable — a step function with a linear part, a quadratic section, and a hard-clipping function. He also derives a second, natural exponent step function from the Schockley diode equation that is less computationally demanding. To implement these models, [Sebastian] chose an Arduino GIGA R1 WiFi, using an ADC to digitize the guitar signal and devoting a DAC to each of the two algorithms. Each distortion effect has its own charms; we prefer the harsher step function over the exponential algorithm, but different strokes.

Kudos to [Sebastian] for this easy-to-understand treatment of what could otherwise be a difficult subject to digest. We didn’t really expect that a guitar distortion pedal would lead down the rabbit hole to diode theory and digital signal processing, but we’re glad it did.

Continue reading “Guitar Distortion With Diodes In Code, Not Hardware” →

Start Your Semiconductor Fab With This DIY Tube Furnace

August 22, 2023 by 3 Comments

Most of us are content to get our semiconductors from the usual sources, happily abstracting away the complexity locked within those little epoxy blobs. But eventually, you might get the itch to roll your own semiconductors, in which case you’ll need to start gearing up. And one of the first tools you’ll need is likely to be something like this DIY tube furnace.

For the uninitiated, [ProjectsInFlight] helpfully explains in the video below just what a tube furnace is and why you’d need one to start working with semiconductors. Perhaps unsurprisingly, a tube furnace is just a tube that gets really, really hot — like 1,200° C. In addition to the extreme heat, commercial furnaces are often set up to seal off the ends of the tube to create specific conditions within, such as an inert gas atmosphere or even a vacuum. The combination of heat and atmospheric control allows the budding fabricator to transform silicon wafers using chemical and physical processes.

[ProjectsInFlight]’s tube furnace started with a length of heat-resistant quartz glass tubing and a small tub of sodium silicate refractory cement, from the plumbing section of any home store. The tube was given a thin coat of cement and dried in a low oven before wrapping it with nichrome wire. The wrapped tube got another, thicker layer of silicate cement and an insulating wrap of alumina ceramic wool before applying power to cure everything at 1,000° C. The cured tube then went into a custom-built sheet steel enclosure with plenty of extra insulation, along with an Arduino and a solid-state relay to control the furnace. The video below concludes with testing the furnace by growing a silicon dioxide coating on a scrap of silicon wafer. This was helped along by the injection of a few whisps of water vapor while ramping the furnace temperature up, and the results are easily visible.

[ProjectsInFlight] still needs to add seals to the tube to control the atmosphere in there, an upgrade we’ll be on the lookout for. It’s already a great start, although it might take a while to catch up to our friend [Sam Zeloof].

Continue reading “Start Your Semiconductor Fab With This DIY Tube Furnace” →

Mass Production 3D Printing Hack Chat

August 21, 2023 by 13 Comments

Join us on Wednesday, August 23 at noon Pacific for the Mass Production 3D Printing Hack Chat with Gabe Bentz!

We’ll take a wild guess and say that right now, within arm’s length of wherever you’re reading this, there’s something that was produced by injection molding. Look around; it’s there someplace, and whatever it is, thousands or perhaps millions of other identical artifacts were produced along with it, all by squeezing hot plastic into intricately machined metal tools.

It’s not much of an overstatement to say that, for good or for ill, the world is made from injection-molded plastic. But not every product can support the often considerable up-front costs associated with injection molding. The tooling needed is often remarkably complicated and correspondingly expensive, and running the machines that actually do the molding is expensive and highly specialized. Unless you’re committed to making a lot of parts, injection molding might just be out of your league.

join-hack-chatBut does that mean that medium-sized runs of parts are out of luck? Not at all! Gabe Bentz, founder and CEO of Slant 3D, is passionate about filling the manufacturing void where injection molding is prohibitive, either by virtue of start-up costs or because the part design is just not possible to manufacture. His massive print farms are busy day in and day out cranking out parts for customers that otherwise couldn’t be made. So if you’ve ever wondered what it takes to run a print farm, and what kinds of design considerations make a part a candidate for mass production by 3D printing, drop by the chat and we’ll see what he has to tell us.

Our Hack Chats are live community events in the Hackaday.io Hack Chat group messaging. This week we’ll be sitting down on Wednesday, August 23 at 12:00 PM Pacific time. If time zones have you tied up, we have a handy time zone converter.