Building A Local Network With LoRaWAN

At its core, the Internet is really just a bunch of computers networked together. There’s no reason that there can’t be other separate networks of computers, or that we all have to tie every computer we have to The One Internet To Rule Them All. In fact, for a lot of embedded systems, it doesn’t make much sense to give them a full network stack and Cat6e Ethernet just to report a few details about themselves. Enter LoRaWAN, a wireless LAN that uses extremely low power for Internet-of-Things devices, and an implementation of one of these networks in an urban environment.

The core of the build is the LoRaWAN gateway which sits at the top of a tall building to maximize the wireless range of all of the other devices. It’s running ChirpStack on the software side and uses a Kerlink Wigrid station to broadcast. The reported range is a little over 9 km with this setup. Other gateways can also be added, and the individual LoRa modules can report to any available gateway. From there, the gateways all communicate back to the central server and the information can be sent out to the wider network, Internet or otherwise.

The project’s creator [mihai.cuciuc] notes that this sort of solution might not be best for everyone. There are other wide area networks available, but using LoRaWAN like this would be likely to scale better as more and more devices are added to the network. For some other ways that LoRa can be used to great effect, take a look at this project which builds an off-grid communications network with it.

A man playing an accordion-like instrument made from two Commodore 64s

The Commodordion Turns Two C64s Into A Single Instrument

One of the main reasons the Commodore 64 became an icon of the 1980s was its MOS 6581 “SID” sound chip that gave it audio capabilities well beyond those of other microcomputers of the 8-bit era. The SID became something of a legend by itself among chiptune enthusiasts, and several electronic instruments have been designed that generate their sound through a SID chip. Not many of those look anything like traditional musical instruments however, so we’re delighted to see [Linus Åkesson]’s new project: two Commodore 64s joined back-to-back using a bellows to form a wonderful new instrument called the Commodordion. It can be played in a similar way one plays a traditional accordion: melodies are played with the right hand, chords with the left, and volume is adjusted by varying the pressure in the bellows.

An accordion-like instrument made from two Commodore 64sThe two computers are basically unmodified, and boot Commodore BASIC like they normally would. A custom circuit board emulates a cassette player and provides the software to be loaded into memory. Both computers run the same program and can be switched between the right-hand and left-hand role by pressing a specific key combination. The software in question is called Qwertuoso, and basically maps notes and various features of the SID chip to keys on the Commodore’s keyboard.

Of course, it’s the bellows that makes this instrument a true member of the accordion family. Made from 5.25″ floppy disks and sticky tape, it forms a more-or-less air-tight system linking the two computers. The airflow in the bellows is measured through a microphone placed next to the air intake: the amount of noise generated is roughly proportional to the amount of air being expelled or inhaled. This information is then used to modulate the volume generated by the two SID chips.

By [Linus]’s own admission it’s not the most ergonomic of instruments, so we’re doubly impressed by the amount of skill he demonstrates while playing it in the video embedded below. It’s not the first time either that he has turned a Commodore 64 into a musical instrument: he previously built a church organ and a theremin. While the Commodordion may look complicated, it’s actually much simpler in construction than a mechanical accordion.

Continue reading “The Commodordion Turns Two C64s Into A Single Instrument”

A widescreen slate-style cyberdeck with a small keyboard sits in front of a cassette deck stereo. Headphones sit to the left of the deck and an old Casio calculator watch is to the right.

2022 Cyberdeck Contest: QAZ Personal Terminal

The slabtop form factor has had a resurgence in the cyberdeck community, and [Greg Leo] has designed the QAZ Personal Terminal to be about as small as a slabtop could be while still having full-sized keys.

Since the device is using a 35% QAZ keyboard as its primary input device, [Leo] has helpfully given a quick overview of how text is input in the video below. Coupled with that surprisingly popular 4:1 LCD screen we’ve seen elsewhere, this cyberdeck looks like a modern interpretation of a TRS-80 Model 100. The Banana Pi powering the QAZ Personal Terminal is running Debian with spectrwm, a tiling window manager making arranging windows a breeze with either a mouse or keyboard. The integrated mouse layer on the keyboard means you don’t need a separate mouse at all if you don’t want to spoil the 1980s mobile chic.

[Leo] has another video all about doing calculus on this cyberdeck with the math shortcuts integrated into the keyboard. Fractions, exponents, and common Greek letters are demonstrated. We can see this being a really great note-taking device for engineering and math courses if you wanted something more portable than a laptop.

It’s hard to get very far hacking without a little math. For more math-focused input devices, check out the Mathboard or the MCM/70.

Continue reading “2022 Cyberdeck Contest: QAZ Personal Terminal”

When [Elon] Says No, Just Reverse Engineer The Starlink Signal

We all know that it’s sometimes better to beg forgiveness than ask permission to do something, and we’ll venture a guess that more than a few of us have taken that advice to heart on occasion. But [Todd Humphreys] got the order of operations a bit mixed up with his attempt to leverage the Starlink network as a backup to the Global Positioning System, and ended up doing some interesting reverse engineering work as a result.

The story goes that [Todd] and his team at the University of Texas Austin’s Radionavigation Lab, on behalf of their sponsors in the US Army, approached Starlink about cooperating on a project to make their low-Earth orbit constellation provide position, navigation, and timing capabilities. Although initially interested in the project, Starlink honcho [Elon Musk] put the brakes on things, leaving [Todd]’s team high and dry. Not to be dissuaded, they bought a Starlink user terminal, built what amounts to a small radiotelescope — although we’ve seen something similar done with just an RTL-SDR — and proceeded to reverse-engineer the structure of Starlink’s Ku-band downlink signal. The paper (PDF link) on their findings is densely packed with details, such as the fact that Starlink uses an orthogonal frequency-division multiplexing (OFDM) scheme.

It’s important to note that their goal was not to break encryption or sniff in on user data; rather, they wanted access to the synchronization and timing signals embedded in the Starlink data structures. By using this data along with the publically available ephemera for each satellite, it’s possible to quickly calculate the exact distance to multiple satellites and determine the receiver’s location to within 30 meters. It’s not as good as some GPS-Starlink hacks we’ve seen, but it’s still pretty good in a pinch. Besides, the reverse engineering work here is well worth a read.

Thanks to [Adrian] for the tip!

Velomobile Gets Electric Assist

What do you get when you throw all accepted bicycle designs out the window and start fresh? Well, it might look a bit like [Saukki’s] velomobile.

Most bikes come in a fairly standard, instantly-recognizable shape which has been popular for over a century now. While it’s a vast improvement over its predecessor, the penny-farthing bicycle, there’s no reason that a bike needs to have this two-triangle frame shape other than that a pretentious bicycle racing standards group says they have to. If you want to throw their completely arbitrary rulebook out of the window, though, you can build much more efficient, faster bikes like recumbents or even full-fairing velomobiles. And if you want to go even faster than that, you can always add a standard ebike motor kit to one.

This is a lot harder than putting a motor on a normal bicycle. Bicycles tend to have standardized parts and sizes, and [Saukki]’s velomobile is far from the standard bike. First, he needed custom mounts for the display and also for the battery, which he needed to make extra wide so its weight wouldn’t rip through the carbon fiber body. The emergency brake lever motor cutoff needed to be dismantled to work with his control system too, and finally the mid-drive motor needed a custom mount as well. It’s a TSDZ2 motor that comes with torque-sensing pedal assist.

The changes didn’t stop there. The velomobile max speed is much higher than a standard bike. This called for some gear ratio changes, in the form of a monster 60-tooth chain ring.

This leads to the one major problem with this build which is that the velomobile can achieve such high speeds on its own that the electric assist cuts out for most of the ride. There is a legal requirement over much of Europe that e-bikes only have pedal assist (without a throttle) and that they stop assisting above a specific speed. But if you want to build an e-bike that pushes the boundary of the law instead of strictly adhering to it, take a look at this one which uses a motor from a washing machine.

Continue reading “Velomobile Gets Electric Assist”

Retrofitting Robots

Al Williams wrote up a neat thought piece on why we are so fascinated with robots that come in the shape of people, rather than robots that come in the shape of whatever it is that they’re supposed to be doing. Al is partly convinced that sci-fi is partly responsible, and that it shapes people’s expectations of what robots look like.

What sparked the whole thought train was the ROAR (robot-on-a-rail) style robot arms that have been popping up, at least in the press, as robot fry cooks. As the name suggests, it’s a robot arm on a rail that moves back and forth across a frying surface and uses CV algorithms to sense and flip burgers. Yes, a burger-flipping robot arm. Al asks why they didn’t just design the flipper into the stovetop, like you would expect with any other assembly line.

In this particular case, I think it’s a matter of economics. The burger chains already have an environment that’s designed around human operators flipping the burgers. A robot arm on a rail is simply the cheapest way of automating the task that fits in with the current ergonomics. The robot arm works like a human arm because it has to work in an environment designed for the human arm.

Could you redesign a new automatic burger-flipping system to be more space efficient or more reliable? Probably. If you did, would you end up with a humanoid arm? Not necessarily. But this is about patching robotics into a non-robotic flow, and that means they’re going to have to play by our rules. I’m not going to deny the cool factor of having a robot arm flip burgers, but my guess is that it’s actually the path of least resistance.

It feels kind of strange to think of a sci-fi timeline where the human-looking robots come first, and eventually get replaced by purpose-built intelligent machines that look nothing like us as the environments get entire overhauls, but that may be the way it’s going to play out. Life doesn’t always imitate art.

Trying To Build The World’s Fastest Roomba

A lot of people complain that Roombas are unreliable, poor at their job, or just plain annoying. Few people complain they’re not fast enough in a straight line. Regardless, [electrosync] set about building the world’s fastest Roomba for his own personal satisfaction.

For this challenge, [electrosync] set his own rules. The build must look like a Roomba, use two drive wheels, and one motor per wheel. It also has to maintain its vacuum functionality. After stripping down a used Roomba, he set about carving out space in the chassis for upgraded hardware. Brushed DC 775 motors were selected for the drivetrain, and these run through a 3:1 planetary reduction gearbox. 3D-printed mounts were then used to install the new motors in the existing chassis. New 3D-printed wheels completed the drivetrain. The original Ni-MH cells were replaced with a pair of 3-cell lithium polymer batteries for more power.

Measured with a Bluetooth GPS device, the upgraded Roomba achieved an impressive 36 km/h (22 MPH). With new wheel designs clad in urethane rubber and an improved anti-wheelie device, it hit a mighty 49 km/h (30 MPH). Adding 4-cell batteries pushed things further to 57 km/h (35 MPH), but the Roomba became difficult to control.

The gauntlet has been thrown down. Do you think you can build a faster Roomba? Time to get hacking! Video after the break.

Continue reading “Trying To Build The World’s Fastest Roomba”