[IMSAI Guy] has an old HP 3488A Switch Control Unit that he wants to dismantle for parts ( see video below the break ). The 3488A is pretty simple as far as HP test equipment goes — a chassis that can hold various types of relay cards and is programmable over GPIB. He notes up front that these are plentiful and inexpensive in the used test equipment market. Continue reading “HP 3488A Teardown, Dismantled For Parts”
Whether it’s because they’re concerned about worsening pollution or the now endemic variants of COVID-19, a whole lot of people have found themselves in the market for a home air quality monitor thee last couple of years. IKEA noted this trend awhile back, and released the VINDRIKTNING sensor to capitalize on the trend.
The device must have sold pretty well, because last month the Swedish flat-packer unveiled the considerably more capable (and more expensive) VINDSTYRKA. Now thanks to the efforts of [Oleksii Kutuzov] we’ve got a fantastic teardown of the new gadget, and some more information on the improvements IKEA made over its predecessor.
Certainly the most obvious upgrade is the addition of an LCD readout that displays temperature, humidity, and how many particulates the device detected in the air. There’s even a “traffic light” colored indicator to show at a glance how bad your air supply is. The other big change is the addition of wireless, though unlike the WiFi hacks we saw for the VINDRIKTNING, this built-in capability uses Zigbee and is designed to plug into IKEA’s own home automation ecosystem.
Speaking of those hacks, a GitHub user by the name of [MaartenL] chimes in to say they’ve managed to hook an ESP32 up to test pads on the VINDSTYRKA motherboard, allowing the parasitic microcontroller to read the device’s sensors and report their data on the network over a service like MQTT, without impacting the sensor’s normal operations. This is how the first hacks on the older VINDRIKTNING were pulled off, so sounds like a promising start.
But even if you aren’t looking to modify the device from its original configuration (how did you find this website?), it seems pretty clear the VINDSTYRKA is a well-built piece of kit that will serve you and your family well. Which is more than what could be said for some of the cheapo environmental sensors flooding the market.
Thanks to [killergeek] for the tip.
You’ve probably seen three-way bulbs. You know, the ones that can go dim or bright with each turn of a switch. [Brian Dipert] wondered how the LED version of these works, and now that he tore one apart, you can find out, too. The old light bulbs were easy to figure out. They had two filaments, one brighter than the other. Switching on the first filament provided some light, and the second gave off more light. The final position lit both filaments at once for an even brighter light.
LED or filament, three-way bulbs have a special base. While a normal Edison-base bulb has the threaded part as the neutral and a center contact for the live wire, a three-way bulb has an extra hot contact ring between the threaded part and the center contact. Obviously, a compatible LED bulb will need this same interface, but will work differently inside.
Inside the LED, [Brian] found two rings of LEDs that took the place of the filaments. He was able to identify all the ICs and devices on the board except one, an MT7712S. If you can read Mandarin, we think this is the datasheet for it.
We weren’t sure what [Brian] would find inside. After all, you could just sense which contacts had voltage and dim the LEDs using PWM. It probably wouldn’t take any less circuitry. LED lighting is everywhere these days, and maybe they don’t all work the same, but you have to admit, using two strings of LEDs is reasonably faithful to the old-fashioned bulbs.
Sometimes LED bulbs are different depending on where you buy them. We were promised LED bulbs would never burn out. Of course, they do, but you can usually scrounge some LEDs from them.
[Big Clive] picked up a keychain battery to charge his phone and found out that it was no bargain. Due to a wiring mistake, the unit was wired backward, delivering -5 V instead of 5 V. The good news is that it gave him an excuse to tear the thing open and see what was inside. You can see the video of the teardown below.
The PCB had the correct terminals marked G and 5 V, it’s just that the red wire for the USB connector was attached to G, and the black wire was connected to 5 V. Somewhat surprisingly, the overall circuit and PCB design was pretty good. It was simply a mistake in manufacturing and, of course, shows a complete lack of quality assurance testing.
The circuit was essentially right out of the data sheet, but it was faithfully reproduced. We should probably test anything like this before plugging it into a device, but we typically don’t. Does our phone protect against reverse polarity? Don’t know, and we don’t want to find out. [Clive] also noted that the battery capacity was overstated as well, but frankly, we’ve come to expect that with cheap gadgets like this.
This isn’t, of course, the first phone charger teardown we’ve seen. This probably isn’t as deadly as the USB killer, but we still wouldn’t want to risk it.
No matter your age or background, there’s an excellent chance you’ll recognize the Nintendo Entertainment System (NES) at first glance. The iconic 8-bit system not only revitalized the gaming industry, but helped to establish the “blueprint” of console gaming for decades to come. It’s a machine so legendary and transformative that even today, it enjoys a considerable following. Some appreciate the more austere approach to gaming from a bygone era, while others are fascinated with the functional aspects of console.
The NesHacker YouTube channel is an excellent example of that latter group. Host [Ryan] explores the ins and outs of the NES as a platform, with a leaning towards the software techniques used to push the system’s 6502 processor to the limits. Even if you aren’t terribly interested in gaming, the videos on assembly programming and optimization are well worth a watch for anyone writing code for vintage hardware.
There is a bit of a paradox when it comes to miniaturization. When electronics replaced mechanical devices, it was often the case that the electronic version was smaller. When transistors and, later, ICs, came around, things got smaller still. However, as things shrink to microscopic scales, transistors don’t work well, and you often find — full circle — mechanical devices. [Breaking Taps] has an investigation of a MEMS chip. MEMS is short for Micro Electromechanical Systems, which operate in a decidedly mechanical way. You can see the video, which has some gorgeous electron microscopy, below. The best part, though, is the 3D-printed macroscale mechanisms that let you see how the pieces work.
Decapsulating the MPU-6050 was challenging. We usually mill a cavity on the top of an IC and use fuming nitric on a hot plate (under a fume hood) to remove the remaining epoxy. However, the construction of these chips has two pieces of silicon sandwiched together, so you need to fully expose the die to split them apart, so our usual method might not work so well. Splitting them open, though, damaged parts of the chip, so the video shows a composite of several devices.
[Mike Harrison] has an upcoming project which will combine a large number of flip-dot displays salvaged from buses. [Mike] thought he knew how these things worked, and had a prototype PCB made right away. But while the PCB was being manufactured, he started digging deeper into the flip-dot’s flipping mechanism.
As he dismantled one of the flip-dots, he realized there was a lot going on under the hood than he realized. The dots are bistable — staying put when power is removed. This is achieved with a U-shaped electromagnet. The polarity of a driving pulse applied to the coil determines which way to flip the dot and saturates the electromagnet’s core in the process. Thus saturated, each dot is held in the desired position because the black side of the dot is made from magnetic material. But wait, there’s more — on further inspection, [Mike] discovered another permanent magnet mounted in the base. He’s not certain, but thinks its job is to speed up the flipping action.
Besides curiosity, the reason [Mike] is studying these so closely is that he wants to build a different driver circuit to have better and faster control. He sets out to better understand the pulse waveform requirements by instrumenting a flip-dot and varying the pulse width and voltage. He determines you can get away with about 500 us pulses at 24 V, or 1 ms at 12 V, much better that the 10 ms he originally assumed. These waveforms result in about 60 to 70 ms flip times. We especially enjoyed the slow-motion video comparing the flip at different voltages at 16:55 in the video after the break.
[Mike] still has to come up with the optimum driving circuit. He has tentatively has settled on a WD6208 driver chip from LCSC for $0.04/ea. Next he will determine the optimum technique to scale this up, deciding whether going for individual pixel control or a multiple sub-array blocks. There are mechanical issues, as well. He’s going to have to saw off the top and bottom margin of each panel. Reluctant to unsolder the 8500+ joints on each panel, his current idea is to solder new controller boards directly onto the back of the existing panels.
This video is a must-watch if you’re working on drivers for your flip-dot display project, and we eagerly look forward to any future updates from [Mike]. We also wrote about a project that repurposed similar panels a couple of years ago. There are a few details that [Mike] hasn’t figured out, so if you know more about how these flip-dots work, let us know in the comments below.
Continue reading “A Close Look At How Flip-Dot Displays Really Work”
