PI4ZLB

It is popular to blame new technology for killing things. The Internet killed newspapers. Video killed the radio star. Is FT8, a new digital technology, poised to kill off ham radio? The community seems evenly divided. In an online poll, 52% of people responding says FT8 is damaging ham radio.  But ham operator [K5SDR] has an excellent blog post about how he thinks FT8 is going to save ham radio instead.

If you already have an opinion, you have probably already raced down to the comments to share your thoughts. I’ll be honest, I think what we are seeing is a transformation of ham radio and like most transformations, it is probably both killing parts of ham radio and saving others. But if you are still here, let’s talk a little bit about what’s going on in ham radio right now and how it relates to the FT8 question. Oddly enough, our story starts with the strange lack of sunspots that we’ve been experiencing lately.

Classic Ham Radio

I’ve been a ham radio operator since 1977. The hobby has changed a lot over the years. I can remember as a teenager making a phone call from my car and everyone was amazed. Ham radio covers a lot of ground, but “traditional” ham radio is operating a station on the HF bands — 3.5 MHz to 30 MHz — and talking to people all over the world. That kind of ham radio is suffering right now for a few reasons. First, HF propagation largely depends on sunspots and sunspots tend to ebb and peak on an 11-year cycle. Right now we are in a deep low part of the cycle and even the last few peaks have not been very good and no one knows why.

I’ve often thought that if Marconi and the others had started experimenting with radio during a sunspot low, they might have decided radio wasn’t very practical. With low sunspot activity, higher frequencies don’t propagate well at all. Lower frequencies might get through, but those require much larger antennas and that causes another problem.

At the height of classic ham radio, every ham wanted a beam antenna or a cubical quad or some other type of rotating directional antenna. Being able to swing an antenna at a particular direction brings more power to bear on the receiver and also helps you receive the other station. The problem is, the antenna elements are typically about a half wavelength in size. So at 20 meters, the elements are about 10 meters in size. You can shorten them a little using some tricks but you pay a price for that in performance. At 10 meters, though, the size is quite manageable. Many hams had directional antennas for the 20, 15, and 10 meter bands (all-in-one antennas called tribanders). A very few would have something for 40 meters — despite Mosley’s description of its 40-20-15 antenna as “vest pocket”, but that was pretty exotic. At 80 meters, mechanically rotating directional antennas are all but unheard of.

So when propagation is bad you should go to lower frequencies, but that means larger antennas. Worse still, the last few decades have seen an increasing hostility to ham radio antennas with city governments, home owner’s associations, and similar. People living in apartments or condos have the same kind of problem. So the number of hams who can even put up a tribander or any sort of visible antenna has dropped significantly.

So here you are with your radio. The bands are bad, and your small hidden antenna is not very good at any band that might work. What do you do?

Voice is Wasteful

One historical answer to this problem was to quit talking and start using Morse code. For a variety of reasons, Morse code will get through when there isn’t enough power, antennas, or propagation to send voice communications. A skilled operator can pull a Morse code signal out of noise that you would swear is just noise. But what if you aren’t a skilled operator? Bring in a skilled computer.

Some hams have always experimented with digital operation, mostly with war-surplus teletype machines. Sending data digitally is almost as good as sending Morse code and it is easy to type and read a printout compared to manually sending and receiving code. Sure, computers can read code, but since a human is sending it, it is likely to not be perfect copy unless the software is very smart and can adjust to slight variations like a human operator can.

Then came a digital mode called PSK31. It was a low-bandwidth slow digital protocol that used a computer’s soundcard to both send and receive. The computer could pull data out of what you would swear was nothing. There was some error correcting and other technical features that made PSK31 possibly better than Morse code for disadvantaged operations even by very skilled operators.

There are other similar digital modes, but most of them have not really caught on in the way that PSK31 has. Until FT8.

So FT8?

FT8 is a digital mode, too. It was specifically created to work well in really bad situations like meteor scatter or moonbounce. To maximize the chances of success, each FT8 packet holds 13 characters and takes 13 seconds to send. The protocol depends on a highly synchronized clock and every minute is divided into 15-second slots. Because of this FT8 contacts are highly structured and short. It’s like Twitter on sleeping pills. You won’t use FT8 to talk about your new motorcycle with your friend in Spain.

However, because the information is digital and of limited format, a typical exchange is that one operation calls CQ. Another operator notices and clicks on the first station in their display. Now their computers exchange basic information like location and signal strength. And then the contact is done.

The Good, The Bad…

If your goal is to “work” a lot of countries, or states, or islands, or any of the other entities hams try to get awards for, then this is great. It favors getting the minimum data through under the worst conditions. If you want to use ham radio to learn about other people and cultures, this doesn’t help because you just can’t say all that much. The truth is, though, that having long casual conversations with people very far away doesn’t happen as much as you’d think anyway.

[K5SDR’s] point, though, is that right now HF ham radio is on the brink of disaster even without FT8. The bands are bad and with antennas restricted, there isn’t much to do for a lot of hams. FT8 lets them get on the air. Purists complain it doesn’t take skill. But honestly, we’ve heard that before. Automated Morse code gear didn’t ruin ham radio. Nor did the availability of store-bought equipment.

Besides, this is all classic ham radio. There’s plenty of other things to do: emergency preparedness, radio control, propagation experimentation, and TV or image transmissions, just to name a few. If those don’t excite you, there’s moonbounce and satellites (even one orbiting the moon), so there’s always something to get involved with. The frontier is moving, and ham radio is moving with it, or at least maybe it should be.

When I started working in a video production house in the early 1980s, it quickly became apparent that there was a lot of snobbery in terms of equipment. These were the days when the home video market was taking off; the Format War had been fought and won by VHS, and consumer-grade VCRs were flying off the shelves and into living rooms. Most of that gear was cheap stuff, built to a price point and destined to fail sooner rather than later, like most consumer gear. In our shop, surrounded by our Ikegami cameras and Sony 3/4″ tape decks, we derided this equipment as “ReggieVision” gear. We were young.

For me, one thing that set pro gear apart from the consumer stuff was the type of connectors it had on the back panel. If a VCR had only the bog-standard F-connectors like those found on cable TV boxes along with RCA jacks for video in and out, I knew it was junk. To impress me, it had to have BNC connectors; that was the hallmark of pro-grade gear.

I may have been snooty, but I wasn’t really wrong. A look at coaxial connectors in general and the design decisions that went into the now-familiar BNC connector offers some insight into why my snobbery was at least partially justified.

Keeping the Impedance

The connector that would eventually become known as the BNC connector when it was invented in the 1950s has its roots in two separate connectors developed in the 1930s and 1940s for the burgeoning radio and telephone industries. When it comes to wires and connectors for DC and low-frequency AC circuits, pretty much anything that will carry the current and provide a firm mechanical connection will do. But once a circuit is into the radio frequency range it’s a different story. At such frequencies coaxial cable is preferred for transmission line, and any connectors inserted into the line need to be engineered to minimize changes in impedance, which could cause reflection of the signal and generate standing waves that can cause damage.

Paul Neil, an electrical engineer who had been with the Bell Company since 1916, was well versed in RF systems. In the 1940s he identified a need for a coaxial connector capable of working well at microwave frequencies and designed the Type N connector. Like all coaxial connectors, it was designed to present as little change in the characteristic impedance of the feedline as possible by keeping the spacing between the center conductor connection and the outer shell as close to the feedline dimensions as possible. Neil’s connector had a female threaded outer shell on the plug that mated with male threads on the matching socket, and was designed to be weatherproof. The N connector took its name from Neil’s last initial and is still in use to this day.

Meanwhile, an engineer at Amphenol Corporation was working on his own design. Carl Concelman’s connector, similarly dubbed the Type C connector, used the same approach to reduce impedance changes in coaxial connections. However, he chose to make his connector quick-disconnect; rather than tediously screwing and unscrewing the outer shell, the C connector had a bayonet connection. The outer shell of the socket had lugs diametrically opposed on its outer surface. These lugs would mate with the long arm of L-shaped grooves machined into the inner surface of the outer shell on the plug. The shell would be rotated to move the lugs into the short arm of the groove, locking the two connectors together mechanically and electrically.385

Best of Both Worlds

Both the N and the C connectors enjoyed success in the marketplace, but neither was ideal. The N connector was slimmer in profile than the C but had all that pesky threading and unthreading to deal with; the C connector has that nice quick-disconnect but was bulky. In addition, neither connector was particularly easy to manufacture as each required some fairly fancy machining. With those shortcomings in mind, an engineer at the Hazeltine Electronics Corporation named Octavio Salati came up with his own design. It would have the bayonet locking feature of the C connector and the slimmer profile of the N connector. It used the same techniques as both connectors to minimize reflections due to inline impedance changes.

Salati’s connector was patented in 1951 with the unexciting name “Electrical Connector.” Unlike its predecessors, it would not be dubbed the “S-Connector” but, in a gentlemanly gesture, it was called the BNC, for “Bayonet Neil-Conselman.” To support the RF work for which it was originally designed, the connector had a 50-ohm characteristic impedance; later, a 75-ohm version was made for the television industry. The connector is usable up to around 11 GHz, although it’s not ideal past 4 GHz or so owing to the slots cut into the conductor for the outer shield, which start to radiate signals.

The BNC connector has seen widespread acceptance as a coaxial connector in industries far beyond its original target markets. From public service communications to scope probes to computer networking, Salati’s design, and by extension both Neil’s and Conselman’s, has delivered solid performance for the past sixty years.

Ham radio operators bouncing signals off the moon have become old hat. But a ham radio transmitter on the Chinese Longjiang-2 satellite is orbiting the moon and has sent back pictures of the Earth and the dark side of the moon. The transceiver’s main purpose is to allow hams to downlink telemetry and relay messages via lunar orbit.

While the photo was received by the Dwingeloo radio telescope, reports are that other hams also picked up the signal. The entire affair has drawn in hams around the world. Some of the communications use a modulation scheme devised by [Joe Taylor, K1JT] who also happens to be a recipient of a Nobel prize for his work with pulsars. The Dwingeloo telescope has several ham radio operators including [PA3FXB] and [PE1CHQ].

You can find technical particulars about the satellite on its web page. There are also GNU Radio receivers and information about tracking. If you want to listen in, you’ll need some gear, but it looks very doable. The same page details several successful ham radio stations including those from [PY2SDR], [CD3NDC], [PY4ZBZ], [N6RFM], and many others. While the Dwingeloo telescope is a 25-meter dish, most of the stations have more conventional looking Yagi or helical antennas.

If your Mandarin is up to it, there is live telemetry on that page, too. You might have more luck with the pictures.

For working conventional satellites, you often need an agile antenna. We suspect the lunar orbiting satellite appears to move less, but you’ll have other problems with more noise and weak signals. Although hams have been bouncing signals off the moon for decades, they’ve only recently started bouncing them off airplanes.

AI is currently popular, so [Chirs Lam] figured he’d stimulate some interest in amateur radio by using it to pull call signs from radio signals processed using SDR. As you’ll see, the AI did just okay so [Chris] augmented it with an algorithm invented for gene sequencing.

His experiment was simple enough. He picked up a Baofeng handheld radio transceiver to transmit messages containing a call sign and some speech. He then used a 0.5 meter antenna to receive it and a little connecting hardware and a NooElec SDR dongle to get it into his laptop. There he used SDRSharp to process the messages and output a WAV file. He then passed that on to the AI, Google’s Cloud Speech-to-Text service, to convert it to text.

Despite speaking his words one at a time and making an effort to pronounce them clearly, the result wasn’t great. In his example, only the first two words of the call sign and actual message were correct. Perhaps if the AI had been trained on actual off-air conversations with background noise, it would have been done better. It’s not quite the same issue, but we’re reminded of those MIT researchers who fooled Google’s Inception image recognizer into thinking that a turtle was a gun.

Rather than train his own AI, [Chris’s] clever solution was to turn to the Smith-Waterman algorithm. This is the same algorithm used for finding similar nucleic acid sequences when analyzing genes. It allowed him to use a list of correct call signs to find the best match for what the AI did come up with. As you can see in the video below, it got the call signs right.