Although it might seem like there was a sudden step change from analog to digital sometime in the late 1900s, it was actually a slow, gradual change from things like record players to iPods or from magnetic tape to hard disk drives. Some of these changes happened slowly within the same piece of hardware, too. Take the Sony DXC-3000A, a broadcast camera from the 1980s. Although it outputs an analog signal, this actually has a discrete pixel CCD sensor capturing video. [Colby] decided to finish the digitization of this camera and converted it to output HDMI instead of the analog signal it was built for.

The analog signals it outputs are those that many of us are familiar with, though: composite video. This was an analog standard that only recently vanished from consumer electronics, and has a bit of a bad reputation that [Colby] thinks is mostly undeserved. But since so many semi-modern things had analog video outputs like these, inspiration was taken from a Wii mod chip that converts these consoles to HDMI. Unfortunately his first trials with one of these had confused colors, but it led him to a related chip which more easily outputted the correct colors. With a new PCB in hand with this chip, a Feather RP2040, and an HDMI port the camera is readily outputting digital video that any modern hardware can receive.

Besides being an interesting build, the project highlights a few other things. First of all, this Sony camera has a complete set of schematics, a manual meant for the end user, and almost complete user serviceability built in by design. In our modern world of planned obsolescence, religious devotion to proprietary software and hardware, and general user-unfriendliness this 1980s design is a breath of fresh air, and perhaps one of the reasons that so many people are converting old analog cameras to digital instead of buying modern equipment.

The increasing dominance of lithium cells in the market place leave our trusty NiMH cells in a rough spot. Sure, you can still get a chargers for the AAs in your life, but it’s old tech and not particularly stylish. That’s where [Maximilian Kern] comes in, whose SPINC project was recently featured in IEEE Spectrum— so you know it has to be good.

With the high-resolution LCD, the styling of this device reminds us a little bit of the Pi-Mac-Nano— and anything that makes you think of a classic Macintosh gets automatic style points. There’s something reminiscent of an ammunition clip in the way batteries are fed into the top and let out the bottom of the machine.

[Maximilian] thought of the, ah, less-detail-oriented amongst us with this one, as the dedicated charging IC he chose (why reinvent the wheel?) is connected to an H-bridge to allow the charger to be agnostic as to orientation. That’s a nice touch. An internal servo grabs each battery in turn to stick into the charging circuit, and deposits it into the bottom of the device once it is charged. The LCD screen lets you monitor the status of the battery as it charges, while doubling as a handy desk clock (that’s where the RP2040 comes in). It is, of course powered by a USB-C port as all things are these days, but [Maximilian] is just drawing from the 5V line instead of making proper use of USB-C Power Delivery. (An earlier draft of this article asserted incorrectly that the device used USB-C-PD.)  Fast-charging upto 1A is enabled, but you might want to go slower to keep your cells lasting as long as possible. Firmware, gerbers and STLs are available on GitHub under a GPL-3.0 license– so if you’re still using NiCads or want to bring this design into the glorious lithium future, you can consider yourself welcome to.

We recently featured a AA rundown, and for now, it looks like NiMH is still the best bang for your buck, which means this project will remain relevant for a few years yet. Of course, we didn’t expect the IEEE to steer us wrong.

Thanks to [George Graves] for the tip.

A PCB business card is a great way for electrical engineers to impress employers with their design skills, but the software they run can be just as impressive as the card itself. As a programmer with an interest in embedded machine learning, [Dave McKinnon] wanted a card that showcased his skills, so he designed one that runs voice recognition.

[Dave] specifically wanted to run a neural network on his card, but needed to make it small enough to run on a microcontroller. Voice recognition looked like a good fit for this, since audio can be represented with relatively little data, a microphone is cheap and easy to add to a circuit board, and there was already an example of someone running such a voice recognition network on an Arduino. To fit the neural network into 46 kB, it only distinguishes the words “one” through “nine,” and displays its guess on an LED seven-segment display. [Dave] first prototyped the system with an Arduino, then designed the circuit board around an RP2040.

The switch from Arduino to the RP2040 brought with it a mysterious change: it would usually recognize the word “eight,” but none of the other numbers. After much investigation, it turned out that the new circuit was presenting samples at a much higher rate than the older one had, which was throwing the network off. [Dave] increased the sampling period and had the user speak the numbers slowly, which solved the issue.

The microcontroller was well chosen; the RP2040 is good enough for machine learning that there are dev boards explicitly designed for it, and even comparatively less powerful Arduino boards can do surprisingly good voice recognition. On the hardware side, [Dave] cited some of the Linux business cards we’ve seen as inspiration.