Host of the Soldersmoke podcast, [Bill Meara], contributed this guest post.
While the rest of the world is moving toward high speed broadband, some hams—including one Nobel Prize winner—are going in exactly the opposite direction. Our ‘QRSS’ mode makes use of an unusual mixture of modern digital signal processing (DSP), ancient Morse code, and simple homebrewed transmitters. Very narrow bandwidth is desirable because this reduces the noise in the radio communication channel, greatly improving the S/N ratio. But Shannon’s communication theory tells us that narrow bandwidth comes with a cost: slow data rates. In QRSS, beacon transmitters using only milliwatts churn out slow speed Morse ID signals on 10.140 MHz that are routinely picked up by DSP-based receivers on the other side of the globe. Many of the receivers, ‘grabbers’, have visual outputs that are available online in real time. QRSS has been getting a lot of attention on the Soldersmoke podcast and on the Soldersmoke Blog. For more information check out this overview and the hardware involved. Here’s a gallery of received signals.
Airgapping refers to running a machine or machines without connections to external networks. Literally, a gap of air exists between the machine and the outside world. These measures present a challenge to those wishing to exfiltrate data from such a machine, leading to some creative hacks. [Jacek] has recently been experimenting with leaking data via Ethernet adapters.
The hack builds on [Jacek]’s earlier work with the Raspberry Pi 4, in which the onboard adapter is rapidly switched between 10 and 100 Megabit modes to create a signal that can be picked up via radio up to 100 meters away. Since then, [Jacek] determined the Raspberry Pi 4, or at least his particular one, seems to be very leaky of RF energy from the Ethernet port. He decided to delve deeper by trying the same hack out on other hardware.
Using a pair of Dell laptops connected back to back with an Ethernet cable, the same speed-switching trick was employed. However, most hardware takes longer to switch speeds than the Pi 4; usually on the order of 2-5 seconds. This limited the signalling speed, but [Jacek] was able to set this up to exfiltrate data using QRSS, also known as very slow speed Morse code. The best result was picking up a signal from 10 meters away, although [Jacek] suspects this could be improved with better antenna hardware.
Overseas factories can be sort of a mythical topic. News articles remind us that Flex (née Flextronics) employs nearly 200 thousand employees worldwide or that Foxconn is up to nearly a million. It must take an Apple-level of insider knowledge and capital to organize such a behemoth workforce, certainly something well past the level of cottage hardware manufacturing. And the manufacturing floor itself must be a temple to bead blasted aluminum and 20 axis robotic arms gleefully tossing products together. Right?
Well… the reality is a little different. The special sauce turns out to be people who are well trained for the task at hand and it doesn’t require a $1,000,000,000,000 market cap to get there.
[Adam leeb] was recently overseas to help out with the production ramp for one of his products and took a set of fantastic videos that walk us through an archetypical asian factory.
The Room
I’ve been to several factories and for me the weirdest part of the archetype is the soul crushing windowless conference room which is where every tour begins. Check out this one on the left. If you ever find yourself in a factory you will also find a room like this. It will have weird snacks and bottles of water and a shiny wood-esque table. It will be your home for many, many more hours than you ever dreamed. It’s actually possible there’s just one conference room in the universe and in the slice of spacetime where you visit it happens to be in your factory.
Ok, less metaphysics. It’s amazing to watch the myriad steps and people involved in taking one product from zero to retail-ready. [adam] gives us a well narrated overview of the steps to go from a single bare board to the fully assembled product. From The Conference Room he travels to The Floor and walks us through rows of operators performing their various tasks. If you’ve been reading for a while you will recognize the pick and place machines, the ovens, and the pogo pin test fixtures. But it’s a treat to go beyond that to see the physical product that houses the boards come together as well.
Check out [adam]’s videos after the break. The first deals with the assembly and test of his product, and the second covers the assembly of the circuit boards inside which is broadly referred to as SMT. Watching the second video you may notice the funny (and typical) contrast between the extremely automated SMT process and everything else.
It wasn’t long ago that you needed to know Morse code to be a ham radio operator. That requirement has gone in most places, but code is still useful and many hams use it, especially hams that like to hack. Now, hams are using the Raspberry Pi to receive highly readable Morse code using very low power. The software is QrssPiG and it can process audio or use a cheap SDR dongle.
There are a few reasons code performs better than voice and many other modes. First, building transmitters for Morse is very simple. In addition, Morse code is highly readable, even under poor conditions. This is partly because it is extremely narrow bandwidth and partly because your brain is an amazing signal processor.
Like most communication methods, the slower you go the easier it is to get a signal through. In ham radio parlance, QRS means “send slower”, so QRSS has come to mean mean “send very slowly”. So hams are using very slow code, and listening for it using computerized methods. Because the data rate is so slow, the computer has time to do extreme methods to recover the signal — essentially, it can employ an extremely narrow filter. Having a QRSS signal detected around the world from a transmitter running much less than a watt is quite common. You can see a video introduction to the mode from [K6BFA] and [KI4WKZ], below.
[Scott Harden] is working on a research project involving optogenetics. From what we were able to piece together optogenetics is like this: someone genetically modifies a mouse to have cell behaviors which can activated by light sensitive proteins. The mice then have a frikin’ lasers mounted on their heads, but pointing inwards towards their brains not out towards Mr. Bond’s.
Naturally, to make any guesses about the resulting output behavior from the mouse the input light has to be very controlled and exact. [Scott] had a laser and he had a driver, but he didn’t have a controller to fire the pulses. To make things more difficult, the research was already underway and the controller had to be built
The expensive laser driver had a bizarre output of maybe positive 28 volts or, perhaps, negative 28 volts… at eight amps. It was an industry standard in a very small industry. He didn’t have a really good way to measure or verify this without either destroying his measuring equipment or the laser driver. So he decided to just build a voltage-agnostic input on his controller. As a bonus the opto-isolated input would protect the expensive controller.
The kind of travesty that can occur when [Scott] doesn’t have access to nice project boxes.The output is handled by an ATtiny85. He admits that a 555 circuit could generate the signal he needed, but to get a precision pulse it was easier to just hook up a microcontroller to a crystal and know that it’s 100% correct. Otherwise he’d have to spend all day with an oscilloscope fiddling with potentiometers. Only a few Hackaday readers relish the thought as a relaxing Sunday afternoon.
He packaged everything in a nice project box. He keeps them on hand to prevent him from building circuits on whatever he can find. Adding some tricks from the ham-radio hobby made the box look very professional. He was pleased and surprised to find that the box worked on his first try.
We’ve reported on “space” balloons before. Heck, some of us have even launched a few. Usually they go way up in the air, take some cool pictures, and land within driving (and retrieving) distance the same afternoon. You get often amazing photos and bragging rights that you took them for the low, low price of a really big helium balloon and a fill.
But what if you shrunk everything down? Over the last few years, [Andy, VK3YT] has been launching ever smaller and lighter balloons with very low power ham radio payloads. So no camera and no photos, but the payback is that he’s launching payloads that weigh around thirteen grams complete with GPS, radio, solar cell, and batteries. They can stay up for weeks and go really far. We’d love to see some construction details beyond the minimalistic “Solar powered party balloon, 25mW TX”. But that about sums it up.
[m0xpd] got his hands on an inexpensive AD9850 DDS Module from eBay but needed a way to control it. He took inspiration from the projects that used a PIC microcontroller, but decided to add his own twist by using a Raspberry Pi to build a multi-mode beacon transmitter.
At the center of this breadboarded circuit lies the green AD9850 module. To its left is a level converter he built to get the 3.3V levels from the RPi board to work with the rest of the 5V hardware. The signal then feeds into a QRP amplifier and a low pass filter.
He didn’t start from square one when it came time to write the code for the RPi. Instead he grabbed an Arduino sketch for the very same DDS and ported it over to Python. The first test signal was his call sign sent in Morse code at QRSS speeds. But he also managed to get Hellschreiber messages working, making it a multiple-mode device.
![The kind of travesty that can occur when [Stefan Kiese] doesn't have access to nice project boxes.](https://hackaday.com/wp-content/uploads/2016/07/img_3466.jpg)
