PI4ZLB

Before swearing my fealty to the Jolly Wrencher, I wrote for several other sites, creating more or less the same sort of content I do now. In fact, the topical overlap was enough that occasionally those articles would get picked up here on Hackaday. One of those articles, which graced the pages of this site a little more than seven years ago, was Getting Started with RTL-SDR. The original linked article has long since disappeared, and the site it was hosted on is now apparently dedicated to Nintendo games, but you can probably get the gist of what it was about from the title alone.

When I wrote that article in 2012, the RTL-SDR project and its community were still in their infancy. It took some real digging to find out which TV tuners based on the Realtek RTL2832U were supported, what adapters you needed to connect more capable antennas, and how to compile all the software necessary to get them listening outside of their advertised frequency range. It wasn’t exactly the most user-friendly experience, and when it was all said and done, you were left largely to your own devices. If you didn’t know how to create your own receivers in GNU Radio, there wasn’t a whole lot you could do other than eavesdrop on hams or tune into local FM broadcasts.

Nearly a decade later, things have changed dramatically. The RTL-SDR hardware and software has itself improved enormously, but perhaps more importantly, the success of the project has kicked off something of a revolution in the software defined radio (SDR) world. Prior to 2012, SDRs were certainly not unobtainable, but they were considerably more expensive. Back then, the most comparable device on the market would have been the FUNcube dongle, a nearly $200 USD receiver that was actually designed for receiving data from CubeSats. Anything cheaper than that was likely to be a kit, and often operated within a narrower range of frequencies.

Today, we would argue that an RTL-SDR receiver is a must-have tool. For the cost of a cheap set of screwdrivers, you can gain access to a world that not so long ago would have been all but hidden to the amateur hacker. Let’s take a closer look at a few obvious ways that everyone’s favorite low-cost SDR has helped free the RF hacking genie from its bottle in the last few years.

Hardware Evolution

Even though the project is called RTL-SDR, the Realtek RTL2832U chip is in reality just half of the equation; it’s a USB demodulator chip that needs to be paired with a tuner to function. In the early days, there were a number of different tuners in use, and figuring out which one you were getting was a pretty big deal. The Elonics E4000 was the most desirable tuner as it had the widest frequency range, but it could be difficult to know ahead of time what you were getting.

The packaging and documentation were all but useless; either the manufacturer didn’t bother to include the information, or if they did, it would often become outdated as new revisions of the product were produced. The only way to be sure about what you were getting was to see if somebody had already purchased that particular model and reported on their findings. Luckily, the tuners were cheap enough that you could buy a couple and experiment. In those days, it wasn’t uncommon to find RTL-SDR compatible devices for less than $10 from import sites.

Opening up a contemporary RTL2832U+E4000 receiver, we can see they were relatively simple affairs. The flimsy plastic case doesn’t do much to prevent interference, and the Belling-Lee connector connector is intended for use with a traditional TV antenna. Note this particular model features an IR receiver so the user could change TV channels with the included remote; a reminder of what this device was actually built for.

These days, you don’t need to wade through pages of nearly identical looking USB TV tuners to find compatible hardware. There are now several RTL2832U-based receivers which are specifically designed for RTL-SDR use, generally selling for around $30. These devices not only address the shortcomings of the original hardware offerings, but in many cases add in new capabilities that simply wouldn’t have made sense to include back when they were just for watching TV on your computer.

Here we have the “RTL-SDR Blog v3” receiver, which is one of the most popular “next generation” RTL-SDR receivers. The plastic case has been replaced with an aluminum one that not only reduces interference, but helps the board dissipate heat while in operation. The crystal has been upgraded to a temperature compensated oscillator (TCXO) which helps reduce temperature drift. The R820T2 tuner is paired with a standard SMA antenna connector, and both it and the RTL2832U have some unused pins broken out if you’re looking to get into developing modifications or expansions to the core hardware.

Software Library

The improvements to the base RTL-SDR hardware are welcome, and it’s nice to not have to worry about whether or not the receiver you’ve purchased is actually going to work with the drivers, but realistically those changes mainly benefit the more hardcore users who are pushing the edge of the envelope. If you’re just looking to sniff some 433 MHz thermometers, you don’t exactly need a TCXO. For most users, the biggest improvements have come in the software side of things.

For one, the RTL-SDR package is almost certainly going to be in the repository of your favorite GNU/Linux distribution. Unless you need some bleeding edge feature, you won’t have to compile the driver and userland tools from source anymore. The same will generally be true for the SDR graphical frontend, namely gqrx by Alexandru Csete. Those two packages are enough to get you on the air and browsing for interesting signals, but that’s just the beginning. The rise of cheap SDRs has inspired a number of fantastic new software packages that are light-years ahead of what was available previously.

Certainly one of the best examples is Universal Radio Hacker, an all-in-one tool that lets you search for, capture, and ultimately decode wireless signals. Whether it’s a known protocol for which it already has a built-in decoder, or something entirely new that you need to reverse engineer, Universal Radio Hacker is a powerful tool for literally pulling binary data out of thin air. Those looking to reverse unknown wireless protocols should also take a look at inspectrum, another tool developed in the last few years that can be used to analyze captured waveforms.

If you’re more interested in the practical application of these radios, there have also been a number of very impressive “turn-key” applications developed that leverage the high availability of low-cost SDRs. One such project is dump1090, a ADS-B decoder that was specifically developed for use with the RTL-SDR. With a distributed network of receivers, the software has allowed the community to democratize flight tracking through the creation of open data aircraft databases.

The Gift of Inspiration

In the years since its inception, the RTL-SDR project has become the de facto “first step” for anyone looking to experiment with radio. It’s cheap, it’s easy, and since the hardware is incapable of transmission, you don’t have to worry about accidentally running afoul of the FCC or your local equivalent. Honestly, it’s difficult to think of a valid reason not to add one of these little USB receivers to your bag of tricks; even if you only use it once, it will more than pay for itself.

Ultimately, this is the greatest achievement of the RTL-SDR project. It drove the entry barrier for radio experimentation and hacking so low that it’s spawned a whole new era. From the unique vantage point offered by Hackaday, we can see the sharp uptick of RF projects that correspond to the introduction of an easy to use and extremely affordable software defined radio. People who might never have owned a “real” radio beyond the one in their car can now peel back the layers of obscurity that in the past kept the vast majority of us off the airwaves. This is a very exciting time for wireless hacking, and things are only going to get more interesting from here on out. Long live RTL-SDR!

An errant wire snipping across the wrong electrical pins spells the release of your magic smoke. Even if you are lucky, stray parts are the root of boundless malfunctions from disruptive to deadly. [TheRainHarvester] shares his trick for covering an Arduino Nano with some scrap plastic most of us have sitting in the recycling bin. The video is also after the break. He calls this potting, but we would argue it is a custom-made cover.

The hack is to cut a bit of plastic from food container lids, often HDPE or plastic #2. Trim a piece of it a tad larger than your unprotected board, and find a way to hold it in place so you can blast it with a heat gun. When we try this at one of our Hackaday remote labs and apply a dab of hot glue between the board and some green plastic it works well. The video suggests a metal jig which would be logical when making more than one. YouTube commenter and tip submitter [Keith o] suggests a vacuum former for a tighter fit, and we wouldn’t mind seeing custom window cutouts for access to critical board segments such as DIP switches or trimmers.

We understand why shorted wires are a problem, especially when you daisy-chain three power supplies as happened in one of [TheRainHarvester]’s previous videos.

Hundreds of years from now, the story of humanity’s inevitable spread across the solar system will be a collection of engineering problems solved, some probably in heroic fashion. We’ve already tackled a lot of these problems in our first furtive steps into the wider galaxy. Our engineering solutions have taken humans to the Moon and back, but that’s as far as we’ve been able to send our fragile and precious selves.

While we figure out how to solve the problems keeping us trapped in the Earth-Moon system, we’ve sent fleets of robotic emissaries to do our exploration by proxy, to make the observations we need to frame the next set of engineering problems to be solved. But as we reach further out into the solar system and beyond, our exploration capabilities are increasingly suffering from communications bottlenecks that restrict how much data we can ship back to Earth.

We need to find a way to send vast amounts of data back as quickly as possible using as few resources as possible on both ends of the communications link. Doing so may mean turning away from traditional radio communications and going way, way up the dial and developing practical means for communicating with X-rays.

The Tyranny of Physics

The essential problems with deep space communications come from two sources – the inverse-square law and information theory. The inverse-square law states that the amount of energy at the receiving end of a radio communications link is inversely proportional to the square of the distance to the transmitter. Basically, radio waves spread out from the source and at very great distances tend to diminish into the background noise. That’s why deep-space communications networks tend to have large antennas on both ends of the link, to gather and focus as much of the weak signal as possible, as well as to be able to transmit a powerful and narrowly focused beam.

Information theory tells us that more data can be packed into higher frequency signals than lower frequencies. Early satellites didn’t need much bandwidth to do their jobs, so VHF and UHF radios were generally sufficient. But as spacecraft became more sophisticated and the amount of data they needed to send back increased, their communications links began shifting gradually up the electromagnetic spectrum into the microwave region. The Voyager probes, currently in interstellar space, have an uplink using 2.1 GHz for the relatively low-bandwidth tasks of vehicle control, with a downlink at 8.1 GHz, reflecting the increased bandwidth needed to send scientific data back to Earth.

For as stunning an engineering achievement as Voyager has been, and notwithstanding the fact that it’s still working more than 40 years after launch, its radio gear only barely supports its interstellar mission. To be fair, Voyager was never meant to last this long, and every bit of data that makes it back to Earth is just icing on the cake. But for future missions specifically designed for interstellar space, sending back enough data to make such missions feasible will require more bandwidth.

Small, Bright, and Fast

In late April, NASA is sending a pallet of gear up to the ISS, and one of the experiments stashed in the cargo is meant to explore the potential for X-ray communications, or XCOM, for deep space. The Modulated X-Ray Source (MXS) is a compact X-ray transmitter that will be mounted outside the space station. The receiver for this experiment is already installed; the Neutron Star Interior Composition Explorer (NICER) has been gathering X-ray spectra from neutron stars since 2017, while also gathering data about the potential for using X-ray pulsars as navigational beacons in a sort of “Galactic Positioning System”.

MXS is an interesting instrument. When one thinks of making X-rays, the natural tendency is to assume a traditional hot-cathode vacuum tube, where electrons are boiled off a filament and accelerated by an electric field in the range of 100 kilovolts to slam into a tungsten anode, would be used. But vacuum tubes like those found in a hospital X-ray suite aren’t the best space travelers, and even when ruggedized they’re too bulky and heavy to send upstairs.

So NASA researchers developed a more spaceflight-friendly X-ray generator. Rather than heating a filament to generate electrons, the X-ray source in MXS uses creates photoelectrons by bombarding a magnesium photocathode with UV light from LEDs. The few photoelectrons produced then enter an electron amplifier, an off-the-shelf component found in mass spectrometers that uses specially shaped chambers coated with a thin layer of semiconducting material. Each incident electron liberates a few secondary photoelectrons, which bounce off the other wall of the multiplier to create more electrons, greatly amplifying the signal. The huge stream of electrons is then accelerated by a 10 kV field to collide with the target anode and produce X-rays.

While the MXS source sounds similar to a hot-cathode tube, there are important differences. First, the source can be made cheaply from off-the-shelf components and a 3D-printed metal enclosure. The whole assembly weighs only about 160 grams, fits in the palm of a hand, and has no unusual power or temperature control requirements. The big difference, though, is with how fast the X-rays can be turned on and off. A glowing filament can only heat up and cool down so quickly, meaning that effective modulation of X-ray from hot-cathode sources is difficult. In the MXS, X-rays are produced only when the UV LEDs are on, and those can be switching very quickly, in the sub-nanosecond range. The ability to modulate an X-ray beam lead to data rates in the gigabits per second range, greatly enhancing our ability to move data around in space.

What’s more, X-rays can be more tightly collimated than radio waves or even light, which is also being experimented with for space communications. The tighter X-ray beam spreads out less, making transmission more power efficient and reception easier by virtue of the strong signal from relatively bright transmitters.

Although the distance between the MXS and NICER in these XCOM experiments is only about 50 meters, they stand to position us for much better bandwidth for deep space communications. The MXS source itself has a lot of potential applications beyond XCOM too, from cheap, lightweight, low-power medical imaging on Earth and in space, navigational beacons for spacecraft, and even advanced chemical analysis by X-ray spectroscopy

We’ve seen lots of hacks about capturing weather images from the satellites whizzing over our heads, but this nicely written how-to from [Eric Sorensen] takes a different approach. Rather than capturing images from polar satellites that pass overhead a few times a day, this article looks at capturing images from GOES-17, a geostationary satellite that looks down on the Pacific Ocean. The fact that it is a geostationary satellite means that it captures the same view all the time, so you can capture awesome time-lapse videos of the weather. 

The fact that GOES-17 is a geostationary satellite means that it is a bit more involved. While polar satellites that orbit at an altitude of 800km or so can be received with a random piece of wire, the 35,800 km altitude of geostationary satellites means that you need a better antenna. That doesn’t have to be that expensive, though: [Eric] used a $100 parabolic antenna and a $100 Airspy Mini SDR receiver connected to an Ubuntu laptop running some open source software to receive and decode the 1.7GHz signal of the satellite.

The other trick is to figure out where to point the dish. Because it is a geostationary satellite, this part has to be done carefully, as the parabolic antenna has only a small receiving angle. [Eric] designed a 3D-printed mount that fits onto a tripod for his antenna.

Capturing satellite weather images is a fascinating thing to do, and this adds another level of interest, as the images show the full disc of the earth. Capture a series over time, and you can see storms spin around and across the ocean, and see just how complicated they are.

If you are looking for a simpler way to get started in receiving weather satellite images, check out this guide to converting an old TV antenna and USB receiver to capture images from polar satellites.

This is an exciting day for me — we finally get to build some ham radio gear! To me, building gear is the big attraction of amateur radio as a hobby. Sure, it’s cool to buy a radio, even a cheap one, and be able to hit a repeater that you think is unreachable. Or on the other end of the money spectrum, using a Yaesu or Kenwood HF rig with a linear amp and big beam antenna to work someone in Antartica must be pretty cool, too. But neither of those feats require much in the way of electronics knowledge or skill, and at the end of the day, that’s why I got into amateur radio in the first place — to learn more about electronics.

To get my homebrewer’s feet wet, I chose perhaps the simplest of ham radio projects: dummy loads. Every ham eventually needs a dummy load, which is basically a circuit that looks like an antenna to a transmitter but dissipates the energy as heat instead of radiating it an appreciable distance. They allow operators to test gear and make adjustments while staying legal on emission. Al Williams covered the basics of dummy loads a few years back in case you need a little more background.

We’ll be building two dummy loads: a lower-power one specifically for my handy talkies (HTs) will be the subject of this article, while a bigger, oil-filled “cantenna” load for use with higher power transmitters will follow. Neither of my designs is original, of course; borrowing circuits from other hams is expected, after all. But I did put my own twist on each, and you should do the same thing. These builds are covered in depth on my Hackaday.io page, but join me below for the gist on a good one: the L’il Dummy.

L’il Dummy

As Al points out in the article linked above, a dummy load is just a resistive element that matches the characteristic impedance of the transmitter’s antenna connection. In almost every case, that’s going to be 50 ohms. The reason that the load needs to be as resistive as possible is that it needs to continue looking like a flat 50-ohm load no matter what frequency is applied to it. Any inductive or capacitive elements in the load will make it more reactive, changing the impedance as the input frequency changes. This could lead to RF power getting reflected back into the final amplifier transistors in the transmitter, possibly damaging them or destroying them altogether. Not what you’re looking for.

That means our resistive elements need to be as non-inductive as possible. But, they also need to be able to dissipate a lot of power. The HT dummy load, which I’ve dubbed L’il Dummy, needs to handle the 5 to perhaps 8 watts an HT can output. Trouble is, power resistors in that range are often wirewound, and a coil of wire will have too much inductance. We’ll need to be clever in sourcing components.

The circuit for L’il Dummy is hardly worth a schematic – it’s just an SMA jack with a 50-ohm resistor across the outer ground and the inner conductor. I chose to build the circuit on an RF Biscuit board. This is an open-source design that enables all kinds of handy little RF circuits — attenuators, filters, and as in this case, dummy loads. The resistive element I chose was a thick-film SMT device capable of dissipating 35 watts – way more than enough for this job. That and an edge-mount SMA jack should have been all I needed to make a working dummy load.

To my surprise, once I soldered the resistor to the RF Biscuit board, the dummy load was almost as good an antenna as the stock rubber ducky on my Baofeng HT. I was able to hit a local repeater through the dummy load without any issues. Clearly not a good design. To correct it, I put the whole thing into an enclosure made from 1″ copper pipe. Not cheap stuff, but not too bad, and I like the look of polished copper. Soldering the whole case together was a challenge that my big Weller soldering gun wasn’t up to, and trying to get everything heated up enough with a propane torch without overdoing the heat was a fun time.

Testing on a Budget

Now for the $50 question: does it work? I tested the resistance with a DMM and it comes out to just about 49 ohms, which is close enough in my book. But that’s DC resistance; what about impedance? I don’t have an antenna analyzer, so I trolled around and found a simple method for measuring impedance with only a function generator and an oscilloscope. My scope has a 20-MHz function generator built in, so I whipped up a quick test jig from a BNC jack and an SMA jack, connected in series through a leftover 1000-ohm resistor.

Applying a sine wave into the dummy load, measuring peak-to-peak voltages on each side of the resistance, and doing a little math is all that’s needed to characterize the impedance from 2.5 MHz to 20 MHz. The math is simple:

with V1 being the voltage across the input, V2 being the voltage across the output, and Rref being the actual value of the series resistance, which I measured at 998 Ohms.

And the results are pretty close to 50 Ohms, and flat across the tested band

I wish I could measure it at VHF and UHF frequencies, but that will have to wait until I get a function generator that goes up to 400 MHz or so. I doubt very much that a $50 budget would cover that, though.

Next Time

I had intended to cover both L’il Dummy and its bigger, somewhat smarter brother in one article, but I still have some testing to do on Big Dummy. I’ll cover that next time, and after that we’ll move onto measuring the output of a cheap Chinese HT and perhaps building a filter to clean it up.