How The WS2812 Is Made

[Scotty Allen] of Strange Parts is no stranger to Chinese factory tours, but this one is now our favorite. He visits the font of all WS2812s, World Semi, and takes a good look at the machines that make two million LEDs per day.

The big deal with the WS2812s, and all of the similar addressable LEDs that have followed them, is that they have a logic chip inside the LED that enables all the magic. And that means die-bonding bare-die ICs into each blinky. Watching all of the machines pick, place, glue, and melt bond wire is pretty awesome. Don’t miss the demo of the tape-and-frame. And would you believe that they test each smart LED before they kick it out the door? There’s a machine that clocks some data in and reads it back out the other side.

Do we take the addressable LED for granted today? Probably. But if you watch this video, maybe you’ll at least know what goes into making one, and the next time you’re blinking all over the place, you’ll spill a little for the epoxy-squirting machine. After all, the WS2812 is the LED that prompted us to ask, three years ago, if we could live without one.
Continue reading “How The WS2812 Is Made”

A giemsa stained blood smear from a person with beta thalassemia (Credit: Dr Graham Beards, Wikimedia Commons)

First CRISPR-Based Therapies For Sickle Cell Disease And Beta Thalassemia Approved In The UK

The gene-therapy-based treatment called Casgevy was recently approved in the UK, making it the first time that a treatment based on the CRISPR-Cas9 gene editing tool has been authorized for medical treatments. During the clinical trials, a number of patients were enrolled with either sickle cell disease (SCD) or β thalassemia, both of which are blood disorders that affect the production of healthy red blood cells. Of the 45 who enrolled for the SCD trial, 29 were evaluated in the initial 12-month efficacy assessment, with 28 of those found to be still free of the severe pain crises that characterizes SCD. For the β thalassemia trial, 42 patients were evaluated and 39 were still free of the need for red blood cell transfusions and iron chelation after the 12-month period, with the remaining three showing a marked reduction in the need for these.

Both of these blood disorders are inherited via recessive genes, meaning that in the case of SCD two abnormal copies of the β-globin (HBB) gene are required to trigger the disorder. For β thalassemia a person can be a carrier or have a variety of symptoms based on the nature of the two sets of mutated genes that involve the production of HbA (adult hemoglobin), with the severest form (β thalassemia major) requiring the patient to undergo regular transfusions. Both types of conditions have severe repercussions on overall health and longevity, with few individuals living to the age of 60.

The way that the Casgevy treatment works involves taking stem cells out of the bone marrow of the patient, after which the CRISPR-Cas9 tool is used to target the BCL11A gene and cut it out completely. This particular gene is instrumental in the switch from fetal γ globin (HBG1, HBG2) to adult β globin form. Effectively this modification causes the resulting cells to produce fetal-type hemoglobin (HbF) instead of adult HbA which would have the mutations involved in the blood disorder.

For the final step in the treatment, the modified stem cells have to be inserted back into the patient’s bone marrow, which requires another treatment to make the bone marrow susceptible to hosting the new cells. After this the patient will ideally be cured, as the stem cells produce new, HbF-producing cells that go on to create healthy hemoglobin. Although safety and costs (~US$2M per patient) considerations of such a CRISPR-Cas9 gene therapy may give pause, this has to be put against the prospect of 40-60 years of intensive symptom management.

Currently, the US FDA as well as the EU’s EMA are also looking at possibly approving the treatment, which might open the gates for similar gene-therapies.

Top image: A giemsa stained blood smear from a person with beta thalassemia. Note the lack of coloring. (Credit: Dr Graham Beards, Wikimedia Commons)

Hackaday Links Column Banner

Hackaday Links: November 19, 2023

Two RUDs are better than one, right? That might be the line on Saturday morning’s briefly spectacular second attempt by SpaceX to launch their Starship vehicle atop a Super Heavy booster, which ended with the “rapid unscheduled disassembly” of both vehicles. The first attempt, back in April, had trouble from the get-go, including the rapid unscheduled partial disassembly of their Stage Zero launch pad, followed by rapid but completely predictable disassembly of a lot of camera gear and an unfortunate minivan thanks to flying chunks of concrete.

Starship’s first “hot” separation

Engineering changes helped keep Stage Zero more or less intact this time, and the Super Heavy booster performed flawlessly — for about three minutes. It was at that point, right after the start of the new “hot staging” process, where Starship’s six engines light before the booster actually drops away, that the problems started. The booster made a rapid flip maneuver to get into the correct attitude for burn-back and landing before disappearing in a massive ball of flame.

Reports are that the flight termination system did the deed, but it’s not yet exactly clear why. Ditto the Starship, which was also snuffed by the FTS after continuing to fly for about another five minutes. Still in all, the SpaceX crew seem to be ecstatic about the results, which is understandable for a company with a “move fast, break things” culture. Nailed it.

Continue reading “Hackaday Links: November 19, 2023”

Hacking Apple’s Magic Mouse To Fix Its Worst Flaws

The Magic Mouse was first released by Apple in 2009 and was a major departure from previous designs. It was sleek, low-profile, and featured a touch pad on the top for gestures. Although the first generation was powered by two AA batteries and didn’t lead to much commentary, the 2015 redesign caused a lot of scathing memes and worse, mostly due to the rechargeable battery and the Lightning charging port that had been located on its bottom, leading to Dead Magic Mouse syndrome when you wanted to charge it. Since then myriad hackers have tried to fix the Magic Mouse’s issues, with [Ivan Kuleshov]’s recent attempt being perhaps the most straightforward and possibly successful.

Essentially, the Magic Mouse has two major flaws: ergonomics and the worst possible location of the charging port. Although both 3D models and commercial products exist to alleviate the former issue – and some of these even add wireless charging in between mousing sessions – all attempts to relocate the charging port were met by failure, as the Magic Mouse cannot be both charged and used at the same time due to how Apple designed the circuit.

What [Ivan] did differently is that aside from tweaking some existing 3D models for Magic Mouse extensions to his liking, he also fixed the charging issue by avoiding Apple’s circuitry altogether and adding a USB-C port in the process. He also added a TP4056-based charging module, directly soldered to the battery’s terminals, that will top off the battery when plugged in. During experimentation on a live Magic Mouse, this led to the battery charge reported in MacOS increasing correspondingly. More or less, at least.

The 3D printed shell isn’t just a wrapper around the original mouse either, but splits the squat rodent into its upper and lower sections, so that the optical sensor isn’t suspended off the surface, while also keeping the touch-sensitive top section where it should be. According to [Ivan] the project files will be made available on his GitHub account in the near future.

Fixing Astronomy In The Blink Of An Eye

If you’ve ever set a telescope up in your backyard, you probably learned how quick any kind of lighting ruins your observation. In fact, a recent study found that every year, about 10% of the stars that were visible the previous year disappear in the mishmash of light scattering through the atmosphere. A company called StealthTransit has a solution: blink the lights in a controlled way. They have an animated video explaining the concept.

The technology, named DarkSkyProtector, assumes there is LED lighting and that the light’s owner (or manufacturer) will put a simple device in line that causes the LED to blink imperceptibly. As you might guess, the telescope — presumably some giant observatory uses a GPS receiver to synchronize and then images only when the LED lights all turn off. That presumes, of course, that you have a significant number of lights under control.

It is hard to imagine every city and home having astronomy-safe lighting. However, we can imagine a university installing a lighting system on its campus to protect night viewing. The system underwent a test in the Caucasus mountains using a 24-inch telescope and was apparently quite successful with a shutter rate of about 150 Hz. We weren’t clear if each LED control module has to have a GPS-disciplined time source, but it seems like you’d have to. However, the post talks about how the bulbs wouldn’t cost more to make than conventional ones, so maybe they don’t have anything fancy in them.

You can see satellites in the day with some tech tricks. Want to check out observatories? Hit the road. Or, get time on a telescope with Skynet University.

Two-Channel Guitar Stomp Box Makes Momentary Switches Latching

When we first saw [Maarten Tromp]’s article about a “momentary latching switch” for guitar effects pedals, we have to admit to being a bit confused. When it comes to push-button switches, “momentary” and “latching” seem to be at odds with each other, with different mechanisms inside the switch to turn one into the other. What gives?

As it turns out, [Maarten]’s build makes perfect sense when you consider the demands of a musical performance. Guitar effects pedals, or “stomp boxes,” are often added to the output of electric guitars and other instruments to change the signals in some musically interesting way. The trouble is, sometimes you only need an effect for a few bars, and the push-on, push-off switches on many effects pedals make that awkward.

[Maarten]’s idea was to build a stomp box with momentary switches that act as inputs to an ATtiny2313 microcontroller rather than directly controlling the effect. That way, a bit of code can determine how long the switch is tapped, and activate a relay to do the actual switching accordingly. A short tap of the button tells the microcontroller to latch the relay closed until another tap comes along; a long press means that the relay is held open only as long as the button is held down.

Yes, he could have used a 555, a fact which [Maarten] readily acknowledges, but with some loss of flexibility; he currently has the threshold set at 250 milliseconds, which works for his performance style. Changing it would be a snap in code, as would toggling the latching logic. A microcontroller also makes expansion from the two-channel setup shown here easier.

Looking for more effects pedal action? We’ve got a bunch — a tube-amp tremolo, an Arduino Mega multipedal, a digital delay line. Take your pick!

Pineberry Pi HatDrive: Using NVMe SSDs With The Raspberry Pi 5

When the Raspberry Pi 5 launched, many were left chomping at the bit after seeing the PCIe FPC connector alongside the promise that an ‘NVMe SSD HAT would be forthcoming’. Although the official Raspberry Pi NVMe HAT is still a long while off, the Polish company Pineberry Pi is ramping up to release its Top & Bottom versions of its very wittily called HatDrive.

They sent a prototype to [Jeff Geerling], who has been putting his grubby mitts all over them before putting together a video showing off the HatDrive Top, which can accept 2230 and 2242 size NVMe drives.

The primary goal of adding an NVMe drive to the RPi is of course to get rid of those slow and fragile SD cards. Although the SD card standard supports near-NVMe-like speeds with UHS-III, the Raspberry Pi 5 bottoms out at UHS-I, around 100 MB/s. Despite this, using an NVMe drive for booting still takes some work, as [Jeff] lays out in a clear article. Most of this involves tweaking the /boot/config.txt file to enable external PCIe support, editing the onboard EEPROM to change the boot order (in lieu of having a PC-like BIOS screen) and getting the OS image flashed onto the NVMe drive you intend to boot from.

Although things seem to work fine during [Jeff]’s testing, some caveats remain, such as the RPi 5 officially supporting only PCIe Gen 2 x1, with Gen 3 possible, but with potential data integrity issues. There’s also the fundamental limit of having only a single lane of PCIe available. If that’s no problem, then Pineberry Pi offers the aforementioned HatDrive Top for traditional HAT-style mounting, and a Bottom version that can accept up to 2280 format NVMe SSDs. Including the provided ribbon cables, you can order the Top and Bottom for €20 and €25.99 respectively, with the first batch to ship in early December.

Continue reading “Pineberry Pi HatDrive: Using NVMe SSDs With The Raspberry Pi 5”