Broadcast Engineer at BellMedia, Computer history buff, compulsive deprecated, disparate hardware hoarder, R/C, robots, arduino, RF, and everything in between.
5737 stories
·
5 followers

Should’ve Used a 555 — Or 276 of Them

1 Share

When asked to whip up a simple egg timer, most of us could probably come up with a quick design based on the ubiquitous 555 timer. Add a couple of passives around the little eight-pin DIP, put an LED on it to show when time runs out, and maybe even add a pot for variable timing intervals if we’re feeling fancy. Heck, many of us could do it from memory.

So why exactly did [Jesse Farrell] manage to do essentially the same thing using a whopping 276 555s? Easy — because why not? Originally started as an entry in the latest iteration of our 555 Contest, [Jesse]’s goal was simple — build a functional timer with a digital display using nothing but 555s and the necessary passives. He ended up needing a few transistors and diodes to pull it off, but that’s a minor concession when you consider how many chips he replaced with 555s, including counters, decoders, multiplexers, and display drivers. All these chips were built up from basic logic gates, a latch, and a flip-flop, all made from one or more 555s, or variants like the 556 or 558.

As one can imagine, 276 chips take a lot of real estate, and it took eleven PCBs to complete the timer. A main board acts as the timer’s control panel as well as serving as a motherboard for ten other cards, each devoted to a different block of functions. It’s all neat and tidy, and very well-executed, which is in keeping with the excellent documentation [Jesse] produced. The whole thing is wonderfully, needlessly complex, and we couldn’t be more tickled to feature it.

Read the whole story
tekvax
3 days ago
reply
Burlington, Ontario
Share this story
Delete

A Deeper Dive Into Reverse Engineering with a CT Scanner

1 Share

We’ve recently got a look at how [Ken Shirriff] used an industrial CT scanner as a reverse engineering tool. The results were spectacular, with pictures that clearly showed the internal arrangement of parts that haven’t seen the light of day since the module was potted back in the 60s. And now, [Ken]’s cohort [Curious Marc] has dropped a video with more detail on the wonderful machine, plus deep dives into more Apollo-era hardware

If you liked seeing the stills [Ken] used to reverse engineer the obscure flip-flop module, you’re going to love seeing [Marc] using the Lumafield scanner’s 3D software to non-destructively examine several Apollo artifacts. First to enter the sample chamber of the CT scanner was a sealed module called the Central Timing Equipment, which served as the master clock for the Apollo Command Module. The box’s magnesium case proved to be no barrier to the CT scanner’s beam, and the 3D model that was built up from a series of 2D images was astonishingly detailed. The best part about the virtual models is the ability to slice through them in any plane — [Marc] used this feature to hunt down the clock’s quartz crystal.

[Marc]’s Apollo gyroscope was next up, and the look inside the sealed case was very revealing. The details of the mechanical construction were stunning, right down to the bearings supporting the gyro rotor. A power supply module that had seen better days also got the treatment; its scans revealed the exploded capacitor responsible for its rough outward appearance. All the scan data are publicly available on Lumafield’s website, although you’ll need to create an account if you want to play with the models.

As for the scanner itself: is it something that could be built at home? Perhaps. We’ve seen plenty of homebrew X-ray machines, and even a CT scanner or two. Let us know if you tackle a build like this — we’d love to get a look inside.

Read the whole story
tekvax
3 days ago
reply
Burlington, Ontario
Share this story
Delete

Want To Use a Classic Mac Mouse On a Modern Computer? No? Here’s How To Do It Anyway

1 Share

Need to hook a classic Mac mouse up to your modern machine with the help of a DIY USB adapter? [John Floren] has you covered. [John]’s solution uses a board with an ATmega32U4 microcontroller on it to connect to the Mac mouse on one end, and emulate a USB HID (Human Interface Device) on the other. A modern machine therefore recognizes it like it would any other USB input device.

Why is this necessary? The connector on the classic Mac mouse may look like a familiar DE-9 connector, but it is not an RS-232 device and wouldn’t work if it were plugged into a 9-pin serial port. The classic Mac mouse uses a different pinout, and doesn’t have much for brains on the inside. It relies on the host computer to read its encoders and button states directly.

This project is actually a bit of an update to a piece of earlier work [John] did in making a vintage Depraz mouse work with modern systems. He suspected that it wouldn’t take much to have it also work with a classic Mac mouse, and he was right — all it took was updating the pin connections and adding some pull-up resistors. The source code and design files are on GitHub.

Even if one does not particularly want to use a classic Mac mouse for daily work, there’s definitely value in this kind of thing for those who deal in vintage hardware: it allows one to function-check old peripherals without having to fire up a vintage machine.

Read the whole story
tekvax
3 days ago
reply
Burlington, Ontario
Share this story
Delete

Looking back on The Source, the first online "information utility," through teletype transcripts

1 Share

While working on a project to refurbish a 1970s teletype machine, Ars Technica writer, Cameron Kaiser, unearthed a manual and a cache of teletype printouts from The Source, the online service that predated Prodigy, Delphi, and QuantumLink.

The discover created a nostalgic gateway into the emerging online world of 1979:

TCA launched The Source at COMDEX in June 1979.

Read the rest
Read the whole story
tekvax
12 days ago
reply
Burlington, Ontario
Share this story
Delete

Hackable $20 Modem Combines LTE And Pi Zero W2 Power

1 Share
The modem in question plugged into a black powerbank.

[extrowerk] tells us about a new hacker-friendly device – a $20 LTE modem stick with a quadcore CPU and WiFi, capable of running fully-featured Linux distributions. This discovery hinges on a mountain of work by a Chinese hacker [HandsomeYingYan], who’s figured out this stick runs Android, hacked its bootloader, tweaked a Linux kernel for it and created a Debian distribution for the stick – calling this the OpenStick project. [extrowerk]’s writeup translates the [HandsomeYingYan]’s tutorial for us and makes a few more useful notes. With this writeup in hand, we have unlocked a whole new SBC to use in our projects – at a surprisingly low price!

At times when even the simplest Pi Zero is unobtainium (yet again!), this is a wonderful find. For a bit over the price of a Zero 2W, you get a computer with a similar CPU (4-core 1GHz A53-based Qualcomm MSM8916), same amount of RAM, 4GB storage, WiFi – and an LTE modem. You can stick this one into a powerbank or a wallwart and run it at a remote location, make it into a home automation hub, or perhaps, process some CPU-intensive tasks in a small footprint. You can even get them with a microSD slot for extra storage – or perhaps, even extra GPIOs? You’re not getting a soldering-friendly GPIO header, but it has a few LEDs and, apparently, a UART header, so it’s not all bad. As [extrowerk] points out, this is basically a mobile phone in a stick form factor, but without the display and the battery.

The modem with its cover taken off, showing the chips on its board.Now, there’s caveats. [extrowerk] points out that you should buy the modem with the appropriate LTE bands for your country – and that’s not the only thing to watch out for. A friend of ours recently obtained a visually identical modem; when we got news of this hack, she disassembled it for us – finding out that it was equipped with a far more limited CPU, the MDM9600. That is an LTE modem chip, and its functions are limited to performing USB 4G stick duty with some basic WiFi features. Judging by a popular mobile device reverse-engineering forum’s investigations (Russian, translated), looks like the earlier versions of this modem came with the way more limited MDM9600 SoC, not able to run Linux like the stick we’re interested in does. If you like this modem and understandably want to procure a few, see if you can make sure you’ll get MSM8916 and not the MDM9600.

Days of using WiFi routers to power our robots are long gone since the advent of Raspberry Pi, but we still remember them fondly, and we’re glad to see a router stick with the Pi Zero 2W oomph. We’ve been hacking at such sticks for over half a decade now, most of them OpenWRT-based, some as small as an SD card reader. Now, when SBCs are hard to procure, this could be a perfect fit for one of your next projects

Read the whole story
tekvax
12 days ago
reply
Burlington, Ontario
Share this story
Delete

Converting an 80s Typewriter Into a Linux Terminal

1 Share

Typewriters may be long past their heyday, but just because PCs, word processor software, and cheap printers have made them largely obsolete doesn’t mean the world is better off without them. Using a typewriter is a rich sensory experience, from the feel of the keys under your fingers that even the clickiest of PC keyboards can’t compare with, to the weirdly universal sound of the type hitting paper.

So if life hands you a typewriter, why not put it back to work? That’s exactly what [Artillect] did by converting an 80s typewriter into a Linux terminal. The typewriter is a Brother AX-25, one of those electronic typewriters that predated word processing software and had a daisy wheel printhead, a small LCD display, and a whopping 8k of memory for editing documents. [Artillect] started his build by figuring out which keys mapped to which characters in the typewriter’s 8×11 matrix, and then turning an Arduino and two multiplexers loose on the driving the print head. The typewriter’s keyboard is yet used for input, as the project is still very much in the prototyping phase, so a Raspberry Pi acts as a serial monitor between the typewriter and a laptop. The video below has a good overview of the wiring and the software, and shows the typewriter banging out Linux command line output.

For now, [Artillect]’s typewriter acts basically like an old-school teletype. There’s plenty of room to take this further; we’d love to see this turned into a cyberdeck complete with a built-in printer, for instance. But even just as a proof of concept, this is pretty great, and you can be sure we’ll be trolling the thrift stores and yard sales looking for old typewriters.

Read the whole story
tekvax
12 days ago
reply
Burlington, Ontario
Share this story
Delete
Next Page of Stories