Skywave

Radio concepts: Bands and modes

Sat, 05 Oct 2019 00:00:00 +0000

This is the first in a series of beginner-friendly posts on radio concepts. In amateur radio, you’ll hear a lot of terms like “VHF”, “2 meters”, “FM”, “single sideband” and so forth. These give essential information about a radio signal. So, what do all the terms mean? I’ll try to keep it simple, with some more technical info in links and footnotes.

Frequency and wavelength

The most basic property of a radio wave is its frequency. As a wave moves past an antenna, it creates a positive voltage, then a negative voltage, then back to positive and so on.¹ The number of complete cycles per second is the frequency, measured in hertz (Hz). So a radio wave with frequency 500 kHz (500,000 Hz) goes from positive to negative and back 500,000 times per second. A FM radio station on 96.5 MHz completes a cycle 96,500,000 times per second.

Instead of frequency we can also talk about wavelength, which is the distance between consecutive wave peaks (or troughs) in space, at a single instant in time.

(image source)

Radio waves travel at the speed of light,² so there is a strict relationship between frequency and wavelength:

$\begin{align*} \textrm{wavelength} & = \frac{\textrm{speed of light}}{\textrm{frequency}} \\\\ \textrm{frequency} & = \frac{\textrm{speed of light}}{\textrm{wavelength}} \end{align*}$

Because the speed of light is very close to 300,000,000 meters per second, we can use this approximation:

$\begin{align*} \textrm{wavelength in meters} & = \frac{300}{\textrm{frequency in MHz}} \\\\ \textrm{frequency in MHz} & = \frac{300}{\textrm{wavelength in meters}} \end{align*}$

For example, that FM radio station at 96.5 MHz has a wavelength of approximately 300 / 96.5 = 3.1 meters. WiFi at 2.4 GHz (2400 MHz) has a wavelength around 300 / 2400 = 0.125 meters = 12.5 centimeters.

Wavelength is particularly important because the size of an antenna, generally speaking, is proportional to the wavelength of the signals it needs to transmit or receive. Longer wavelengths (lower frequencies) require larger antennas.

Bands

It is useful to group ranges of frequencies into bands, because different frequencies have different physical properties and applications. People have found uses for radio waves all the way from a few Hz to hundreds of GHz and beyond.³ However, the most widely-used frequencies in ham radio fall into three general ranges:

HF (“high frequency”). Officially⁴ covers the range 3 – 30 Mhz. The special thing about HF radio waves is that they reflect off of the ionosphere, a layer of Earth’s upper atmosphere. By bouncing between Earth’s surface and the ionosphere, HF signals can travel extremely long distances, far beyond the horizon.

(image source)

For long distance radio, you can’t beat HF. But the disadvantages are many: large antennas, expensive radios, lots of knowledge and time investment required.

HF propagation depends on many factors such as time of day, solar flares, and the 11-year sunspot cycle. A HF frequency that works great for global communication might stop working just a few minutes later. Currently (2019) we are near a sunspot minimum, which makes HF more challenging; however there is still a lot of fun to be had.

Besides amateur radio, you will also find shortwave broadcasts, numbers stations, military radar systems and other stuff in this range. CB radio sits at the high end of HF, just below the amateur 10 meter band.
VHF (“very high frequency”), spanning 30 – 300 MHz. VHF is mainly used for local communication and, unlike HF, is suitable for handheld radios. VHF usually does not bounce off the ionosphere, and will not go over the horizon, although dedicated VHF enthusiasts find ways around that. VHF is also limited by line-of-sight; it will not travel through obstructions such as hills.

Besides hams, other users of VHF include FM broadcasts, some TV channels, air traffic control, MURS, police/fire/EMS and lots more.
UHF (“ultra high frequency”), spanning 300 MHz – 3 GHz. The lower end of UHF is similar to VHF in characteristics and uses. It is the home of FRS/GMRS in the USA, and the similar PMR446 in Europe.

The upper end of UHF includes cell phones and WiFi. Hams have privileges on the 2.4 GHz frequencies used by WiFi, and can operate WiFi hardware at much higher power, subject to amateur radio rules.

As you can see, the terminology is archaic; “high frequency” is actually towards the low end of frequencies in use today!

Amateur bands

When hams speak of “bands” we usually mean a specific range of frequencies reserved for amateur use. These are identified by approximate wavelength, e.g. “80 meter band” or “70 centimeter band”. You’ll need to consult a chart to find the exact frequency limits and privileges, which vary by country and by license type.

(excerpt from ARRL band chart)

The most important bands for local communication are 2 meters and 70 centimeters. Most handheld radios, including the ubiquitous Baofengs, will support both. Because of line-of-sight limitations, repeaters are very common on these bands. A repeater is a fixed station on a hilltop or tall building, which receives a signal on one frequency and simultaneously retransmits it on another. Because the repeater has a better location, better antenna and more power than your handheld radio, it can vastly extend your effective range — sometimes to 100 miles or more. A full write-up about repeaters is out of scope here, but you can find a wealth of information online.

(own work)

To complicate things, hams will sometimes say “VHF” to mean 2 meters and “UHF” to mean 70 centimeters, even though there are other amateur bands within VHF and UHF as defined above.

On HF a whole bunch of bands are available. The best choice at any point in time depends on many factors such as time of day, ionospheric conditions, the space available for antennas and the license privileges of the operator. At the current sunspot minimum, lower frequencies like 40 and 20 meters are open on a daily basis, while the higher frequencies like 10 meters are only sporadically open for long-distance communication.

Voice modes

A constant radio wave on a single frequency doesn’t convey any information beyond “I’m here!”. In order to send a message — be it voice, text, or something else — we need to vary the radio wave over time. This is where modulation modes come in.

Amplitude modulation (AM) is one simple way to transmit audio. The amplitude (strength) of the radio wave varies according to the amplitude of the sound wave. Because AM transmitters are easy to build and AM receivers are even simpler, AM was the first voice mode. It is still used for broadcasting and some other things, but has largely been replaced by FM and SSB on ham radio (see below).

Frequency modulation (FM) is another audio mode. Instead of varying amplitude over time, we vary frequency. FM produces clear audio and is the dominant mode for local communication on VHF and UHF. Most handheld radios do FM and only FM. The main downside is that a FM signal has a large bandwidth: it occupies a large range of frequencies. For this reason and others, FM is not used on HF bands.

(image source)

Single sideband (SSB) is a more efficient form of AM. Although AM is different from FM, the varying amplitude nonetheless spreads the signal over a range of frequencies.

(image source)

In this waterfall plot of an AM signal, the horizontal axis is frequency, the vertical axis is time, and the color indicates signal strength. The strong red signal in the center is the carrier and contains no information by itself. Also, there are two copies of the audio signal: one above the carrier, the other below. These are the upper and lower sidebands. In other words, the AM signal has a lot of redundancy. By transmitting just one sideband and no carrier, we can save on both power and bandwidth.

(same image source)

This waterfall shows an upper sideband (USB) signal. The lower sideband (LSB) would be the same, except flipped left-to-right. For historical reasons, LSB is used on the lower HF bands (up to 40 meters) while USB is used at higher frequencies.⁵

The main disadvantage of SSB is that it requires more complex electronics in both the transmitter and receiver, which is why SSB did not catch on until the 1960s. With modern electronics it is easy to send and receive SSB. Aside from a few diehard AM users, SSB has taken over as the voice mode for HF ham radio.

Digital voice modes encode audio as a stream of bits before sending it. There are a number of digital modes in use, mainly on VHF and up. I don’t do digital voice so I don’t have much to say about it!

Other modes

Continuous wave (CW) is the oldest⁶ radio mode. It works by switching a transmitter on and off to represent the dots and dashes of Morse code. Although Morse proficiency is no longer required to get a ham license in most countries, it is still a very popular mode. CW occupies a very small (though nonzero) bandwidth, and works well even when signals are weak, because the human ear is very good at picking out the rhythmic beeps of Morse code. CW-only radio kits are the most affordable way to get into HF.

The downside of CW, or upside depending on how you look at it, is that you have to learn Morse code! Skilled CW operators don’t hear individual dots and dashes; they recognize the rhythms of letters or entire words and phrases. It’s like learning a spoken language or a percussion instrument. Many tools and resources are available for learning Morse code, too many to list here.

Digital data modes allow sending digital data over ham radio. Typically this involves an audio connection between a radio and a computer’s sound card. The computer translates data to and from audio tones, much like an old-school phone-line modem. Sometimes a dedicated device takes the place of the computer, or the functionality might be built into the radio itself.

A large number of digital modes are used for different purposes. Some of the more popular ones are FT8, JS8, RTTY, PSK31, and APRS.

Slow-scan television (SSTV) is a way of sending images over ham radio. Despite the term “television”, we are talking about grainy still images sent over the course of several seconds or minutes. If you miss the excitement of waiting for JPEGs to load over dial-up internet then this is the mode for you. Because images are sent so slowly, they fit into the same bandwidth as a single SSB voice signal.

Amateur television (ATV), by contrast, is full-motion video and audio sent using the same technology as commercial broadcast TV. This requires a lot of bandwidth (6 MHz) so it’s only done on the higher-frequency bands (70 centimeters and up) where there is more room.

That’s it for now! Hope you found this informative and stay tuned for more content. There is lots more to say about modes so check out Wikipedia or the ARRL website if you’d like to know more.

Footnotes

The radio wave itself consists of an electric field and a magnetic field, oscillating together, hence “electromagnetic (EM) radiation”. These are vector fields, meaning that they have both a magnitude (strength) and a direction at each point in space. The voltage detected by your radio is a scalar quantity — magnitude only, whether positive or negative. Usually it’s produced by the electric field, which induces an electrical current in your antenna. But there are some antennas, such as magnetic loop antennas, where the magnetic field does the work instead. ↩
The actual speed depends on what material the light or radio wave is moving through. Light is fastest in a vaccum, and about 0.03% slower in air. Radio waves travel up to 1/3 slower in wire or coaxial cable. The ratio of the speed in a material to the speed in vacuum is called the velocity factor. It’s important to account for velocity factor when designing antennas. In optics, the reciprocal of the velocity factor is called the index of refraction. ↩
The light we see is also an electromagnetic wave, at a really high frequency like 300 THz (300,000 GHz). At even higher frequencies you get ultraviolet light, then X-rays, then gamma rays. In between radio frequencies and infrared light, there is a “terahertz gap” where neither radio technology nor optical technology works very well, making it difficult to utilize those frequencies. ↩
A lot of ham HF equipment can transmit down to 1.8 MHz, the bottom of the 160 meter band, even though that is technically medium frequency and not HF. Some can also transmit on 6 meters, at the low end of VHF. 6 meters is known as the “magic band” because it’s quite unpredictable and has characteristics of both HF and VHF. ↩
And on 60 meters, which is a special case in many respects. ↩
Actually, not quite. The very first radio transmitters used an electric spark to excite a resonant circuit, kind of like hitting a bell with a hammer. This produces a damped wave that fades out until the next spark happens. Damped waves are extremely wasteful of bandwidth and are no longer allowed. Once continuous wave took over, the term CW became synonymous with Morse code itself. But a CW signal is not actually continuous; it switches on and off with the pattern of Morse dots and dashes. ↩

APRS with the Yaesu FT-2900R

Mon, 23 Sep 2019 00:00:00 +0000

This is a short write-up about using APRS with the Yaesu FT-2900R mobile radio. For those who aren’t familiar, APRS is a global network which allows sending GPS position updates, text messages, and other data over ham radio, mainly 2 meter FM. APRS “digipeaters” relay these messages by radio and “IGates” send them to the Internet, allowing global communication.

Here’s the radio in action, with a Mobilinkd TNC3 at right (more on this little device later).

The FT-2900R seems to be a great choice for APRS. I receive lots of packets and even at 5 watts, my packets get through to several digipeaters/IGates in the area. The FT-2900R can output up to 75 watts; while this is much more than is typically used for APRS, it could be useful in remote areas far away from other APRS nodes. Despite its high output power, the FT-2900R is fanless; the entire case serves as a giant heatsink. As a result, it is built like a tank and will hold up in demanding environments. Finally, it is affordable on the used market; I got two for $200 total on /r/hamfest.

My antenna for this test is a MFJ-1729 magnetic mount antenna on top of a metal baking tray, which acts as a primitive ground plane. A roll-up J-pole is another convenient option for portable use. For a directional link to a faraway digipeater or IGate, a homemade yagi might work well.

The Mobilinkd connects to my Android phone over Bluetooth. Using APRSdroid I can send position reports, see a list or map of other stations, and send and receive text messages. I can even use SMS and email through appropriate gateway services. This could be quite useful in a remote area with no cell coverage, or in an emergency where the cell network is down. (Note that, like everything on ham radio, APRS messages are public information.)

Because my position updates reach an IGate, my location appears on aprs.fi.

In addition to the map, a lot of other station information is available on the website. As you can see, a number of nodes are able to hear me — the top 3 at only 5 watts.

The cable

The big disadvantage of the FT-2900R in this application is that I had to build a strange custom cable to connect it to the TNC. APRS packets are encoded as audio tones sent over FM radio. The TNC is a modem¹; it modulates bytes into tones and demodulates the incoming tones back to bytes, very much like the modems we once used for dial-up internet. Its interface to the radio consists of bidirectional analog audio, plus a “push-to-talk” (PTT) signal which tells the radio when to transmit. Newer Yaesu radios have a dedicated “DATA” jack on the rear which provides all of these signals; however, the FT-2900R does not.

The workaround is this three-headed cable…

…which plugs into the front-panel microphone jack…

…and also the rear speaker jack.

These two connections are spliced onto the cable which connects to the TNC. A piece of heat-shrink tubing covers the splice.

Once zip-tied to the chassis, the wiring is relatively tidy.

Building the cable

The cable was not too difficult to make. If you have a bit of soldering experience, it should take only an hour or so. I started with a RJ12 patch cable, a 3.5mm TS plug for the speaker jack, and a 3.5mm TRRS plug for the TNC.

Here I have just finished splicing the three microphone-jack connections (audio, PTT, and ground), the two speaker-jack connections (audio and ground) and the TNC side of the cable. The three wires at the top left remain unconnected; they connect to the pins on the RJ12 jack that aren’t used for this project.

Bundled up and ready for the outer heat shrink.

The hardest part was soldering the TRRS plug, especially the inner pin. After all four wires were connected, I crimped the tabs (bottom left of connector) around the cable. I did the same on the speaker plug (not shown).

I made notes about the color of each wire so I could connect everything correctly. The printouts are from this page and the Mobilinkd TNC3 manual.

Future plans

My long-term plan is to replace the TNC with a Raspberry Pi running Direwolf and put everything in an ammo can. Then I will have a rugged, self-contained digipeater that I can deploy in a remote area or for emergency use. I can, for example, carry my handheld radio and the Mobilinkd and use APRS whenever I am within radio range of the portable digipeater.

For power I’ll use my solar battery box or any other 12V (really ≈13V) power supply. The digipeater can also report its own status (battery voltage, temperature, etc.) over APRS. I’ll include an external WiFi antenna for IGate functionality when an Internet connection is available. Other possibilities include a GPS receiver and 3G/LTE connectivity.

Of course, the cable I made is not limited to APRS. I could use it for AX.25 networking, DTMF-based remote control of some device, a voice mailbox system or anything else that involves a computer sending and receiving audio on 2 meter FM.

Although everything seems to be working great, I do worry about RF interference into the audio cables. If that becomes a problem, I will try shielded audio cables. A metal ammo can, with openings much smaller than 2 meters, will also provide some shielding.

Thanks for reading and stay tuned for updates on this project!

Footnotes

TNC stands for terminal node controller. These were originally quite feature-rich devices, but over time more and more functionality has been moved into the host — in this case, my Android phone. The Mobilinkd TNC3 is just a modem which speaks a very simple frame-based serial protocol with the host. It doesn’t know anything about APRS per se. ↩

Solar battery box in an ammo can

Mon, 15 Jul 2019 00:00:00 +0000

Hi, I’m Keegan, callsign KM6PNI! Welcome to my new blog. It’s been a long time since I blogged, and I decided to get back into it by documenting some of my amateur radio projects. For my first post, I will show off a solar-charging battery box that I built for portable radio adventures and emergency power.

Capabilities

I will describe the internals later, but basically this is a lead-acid battery and a solar charge controller, installed in a military-surplus ammo can.

It provides 12 volt power through four Powerpole connectors, as well as a car accessory socket. It also provides two USB Quick Charge ports. Between the USB ports is a built-in voltmeter.

Here you can see the battery charging my phone, my laptop, and my Yaesu FT-60R handheld radio. For the laptop, I’m using a USB-C car charger brick, which I modified with a Powerpole connector. The FT-60R charges through a Powerpole-to-barrel-jack cable that I made, and the phone is ordinary USB Quick Charge.

Here the box is powering my Yaesu FT-891 and a Raspberry Pi 3 (in the plastic box at top). The RPi interfaces with the radio so I can send and receive digital messages. With this equipment plus an antenna, I can send a message around the world, using no infrastructure and only a few watts of power. I’ll write another article about the RPi digital-mode setup at some point.

On the right side of the box, there are two additional Powerpole connectors and a pair of circuit breakers (full wiring diagram below). These connectors are used to charge the battery. Here I have a 12V solar panel plugged in to the connector marked “SOL”.

I’ve connected an in-line power meter, which shows that my panel is providing 38 watts. Whatever power is not consumed by loads goes into recharging the battery.

This is one of my solar panels. It is rated for 100 watts maximum, in direct sunlight. At some point I will buy a folding panel that is easier to transport to the field.

Most of the devices I want to use in the field run from 12V DC. But I can also run AC devices using an inverter. How about a solar-powered box fan on a hot day at the campsite? It draws a lot of power (60 watts on low speed) but it does work!

To charge the box from wall power, I use a battery charger from the auto parts store, modified with a Powerpole connector. The charger connects to the “BAT” side of the power input port. This bypasses the solar charge controller and connects directly to the battery through a 30 amp fuse. In this configuration, both circuit breakers need to be switched off.

Capacity test

The big question is, how long will this battery run my equipment? I did an experiment using a 5 meter LED light strip as a test load.

In an emergency this would provide all the light I need for my living room. It also makes a nice decoration for my bicycle. :) The lights drew 1 A (± 10%) of current throughout the test. Since the battery is rated for 35 Ah, this corresponds to a C-rate of 1/35 ≈ 0.029 C.

I measured the voltage over time using my Rigol DS1054Z oscilloscope. The scope has an Ethernet port and I was able to query the voltage over the network using a small Python script.

Ignoring the fast drop in the first few minutes, the voltage decreased steadily from 12.9 to 11.9 over the course of 25 hours. The linear graph is expected, because we did not get close to exhausting the battery’s stored energy. At some point though, the discharge curve goes nonlinear and the voltage starts to drop rapidly. I don’t have a datasheet for my battery, but this one is similar.

Let’s look at the 0.05C curve. It hits 11.9 volts after approximately 12 hours, and 10.8 volts at 20 hours. I’ll consider 10.8 to be fully discharged; going any lower would likely damage the battery. Extrapolating to my experiment suggests I could run this 1 amp load for almost 42 hours. This is not a rigorous calculation, but it confirms that the nominal rating of 35 amp-hours is realistic for a 1 amp load.

At a higher discharge rate, I will get less total energy. And it’s less stressful on the battery if I don’t go as low as 10.8 volts. Anyway, this shows that I could run my LED lights for several evenings, or use my HF radio all day, without any solar charging.

Internals

Here’s the wiring diagram:

I used circuit breakers on the battery and load connections, because I expect that plugging in excessive loads will be the most likely cause of overcurrent. The breakers are externally accessible and also function as the power switches for the whole box. The internal fuses are not going to blow unless a component fails or I plug something in wrong.

The charge controller will protect the battery from damage due to over- or under-voltage. I have the low voltage threshold set at 11.0 volts, which is a safe though not terribly friendly discharge point.

The connector labeled “external charger” allows direct access to the battery terminals. Besides charging from the wall, this has several other uses:

I can check the battery voltage when the box is off.
In case a breaker or the charge controller fails, I can run a load directly from the battery.
I can place an external battery in parallel with the internal one. There are some problems that can occur when doing this, but it would be useful in a pinch.
By adding one internal switch, I could isolate the internal battery and run the box from only an external battery.

Despite my attempts to cut wires short, the inside is a bit of a rat’s nest.

I mounted the busbars on top of the battery using heavy-duty Velcro. The battery is also attached to the ammo can using Velcro. The charge controller is basically not visible in this photo; it’s situated vertically to the left of the battery, with some foam to fill empty spaces and prevent things from sliding around. I included storage for spare fuses (small box at top right).

Building it

Bill of materials:

Military surplus ammo can from Alameda Army Navy Surplus. Approximately 6”W × 12”L × 10”D exterior dimensions. This one was used for aircraft flares.
ML35-12 12V 35Ah AGM battery
Busbar pair, which included the battery terminal covers
30A MPPT solar charge controller from a previous project
3 × panel mount Powerpole housing
Panel mount dual USB QC 3.0 + voltmeter. Similar units are available with USB-C. However, they can’t provide enough power to charge my laptop, which is why I got the external brick.
Panel mount car accessory socket
Square panel-mount plate
2 × 30A circuit breaker
4 × 12 AWG automotive-type fuse holders and fuses
Plastic box for spare fuses (from a bike tube repair kit)
Lots of 12 AWG wire and crimp connectors
Miscellaneous hardware
Car battery charger

Total cost, not including the solar panel, was around 300 USD. This is cheaper than a comparable AGM battery box from Goal Zero, although lacking a few of its features.

Here is the battery, with card for scale:

When I got the ammo can it looked like this:

I decided to spruce it up with some Rust-Oleum Universal Flat Soft Iron spray paint. Nice and shiny, waiting for the paint to dry:

I had to do some stripping, sanding and re-painting to fix some areas that didn’t come out smooth the first time. It’s still not a perfect paint job but good enough for me!

I installed the plastic panel-mount plate and then drilled the large holes with a 1 1/8” hole saw. I’m not that great with the hole saw and I ended up chewing up the plastic plate a bit. But most of it isn’t visible once the components are installed, and I still think cutting with the plate on was a good idea to ensure the proper fit.

As you may have noticed, I installed the plate sideways. This was deliberate, because I wanted the protective covers to open to the sides and not top/bottom. However, I didn’t realize until later that this meant the voltmeter would be sideways too. Oh well.

I used a Dremel to cut rectangular holes for the breaker terminals:

Here’s the charge controller before installation.

I trimmed those leads as part of wiring everything up.

Once everything was installed and working, it was time to finish painting. I wrote the labels with Sharpie using a stencil, and then went over them freehand with an acrylic paint pen.

The connectors are masked off so I can spray the whole thing with a clear coat finish. However, I found that the clear coat stripped the spray paint in some places, which required further touch-ups. (Of course it hadn’t done this on my test sample with the same metal and paint….)

On the other, blank side I wanted to add a little decoration, so I painted a diagonal rainbow, vaguely inspired by the ZX Spectrum.

Rainbows are pretty, and I did this project mostly during Pride month here in San Francisco so I guess I had rainbows on my mind!

I clear-coated this side too. This is actually where I had the most trouble with spray paint coming off. So I repainted the metalic grey after that happened, and clear-coated just the acrylic rainbow stripes. All told there was a lot of painting, masking, cleaning, re-painting, and waiting, but I’m really happy with the result!

Further thoughts

I chose a lead-acid battery because they are cheap, easy to work with, and durable. The downside is that they’re heavy! The whole box weighs 30 pounds. I can carry it a few blocks to and from my car, but I wouldn’t want to hike a mile with it. I have some lithium iron phosphate cells that I may work into another battery box or an upgrade to this one.

Because of the weight, carrying this box by the attached metal handle is not too comfortable. My wife (also a ham) is going to make a leather cover for the handle.

The circuit breakers stick out from the side of the box, and can be damaged if the box bumps into stuff. I already had to replace a breaker that I broke during construction. I plan to add some kind of shield but haven’t decided on a design yet.

That’s it! Hope you enjoyed my first post, and stay tuned for more projects!