Copasetic Flow

CW Simon: Simon, but, you know, for Morse code Remember Simon from the ’90s? The handheld game with four buttons that played a sequence you had to copy back from memory? Each round added one more step, and you kept going until you finally made a mistake. KO6BTY built a phone game, Project TouCans CW Simon , using Gas Town that does the same thing, but for Morse code. Part of the inspiration for CW Simon was W1REX telling KO6BTY and I that he’d worked on Simon 2. You can hear more about that on Rex’s episode on HRWB. Why CW Simon Exists KO6BTY found that keying CW was helping her learn Morse more quickly, but she and I didn't always have enough time to get on the air together. CW Simon grew out of that. It is a quick way to get more sending practice in, one short round at a time. CW Simon gives you a way to spend more time sending, not just listening. Long Island CW Club points out: “At least a quarter of practice should be sending.” How CW Simo...

  Carbon-fiber masts don’t have to be the villain. For Straight Key Month I ran a quarter-wave 20-meter vertical two ways—first with 12-gauge wire taped to the outside of a Goture carbon-fiber mast, then with 18-gauge magnet wire laced inside the mast. Result: both versions delivered S9 into Utah from downtown San Francisco, and the internal-wire build was far stealthier for city operating. I finally got to do a long term—a few hours—test of the TouCans and a Stick vertical antenna with magnet wire on the inside this week! The interior magnet wire setup worked just as well as the same antenna rigged with 12 gauge wire taped to the outside of the carbon fiber mast did. The rig was S9 into the Utah SDR that morning. Here's a look at the antenna setup and results. There are more text details below the video.. Construction To construct the antenna I removed the brass bottom of the carbon mast, removed the upermost segment, and then inserted 18 guague magnet wire through the...

 An AM broadcast station was loud in my headphones for about two minutes. Then, TouCans mysterious noise cacnellation circuit charged, and the POTA was off and running. According to the POTA website, I haven't activated US-0757 , aka the San Francisco Maritime National Historical Park , in about eight months. It was nice to be back! We'd had about a week's worth of atomspheric river rains, so it was pretty phenomenal to have this view during the activation. The temperature was in the low 60s which was very pleasant. I didn't go for a swim this time, but I'm going to get back to that soon as well. View of the Golden Gate Bridge from over my shoulder The transit ride—SF MUNI 54 to BART to San Francisco MUNI 30—was a piece of cake as always. The video below shows part of the route as seen from the bus window. I just now realized that I copied my opeining into the video twice. You didn't imagine it. Sorry about that :) I stopped for lunch at the Buena Vista I...

 Project TouCans and kitchen window antenna are continuing to work well during straight key month! This is just a quick update. The rig layout can be seen in the video below.

 I made eight QSOs on my first hour long straight key month shift operating as K3Y/6! Points of note: I'm still operating five watts from the kitchen window in San Francisco. The ' shiny F2 ' cutoff was not disproven. The critical frequency between San Francisco and Utah did go below 7,750 kHz and there was only one QSO that went east of Louisiana out to New York. That happened when the fof2 value was about 8200 kHz. Does the Data Fit with ND7K in Arizona? I noticed that while Utah had disappeared again, I was still being spotted in Arizona at 01:00 UTC. Looking at the fof2 values to the spotting station, ND7K revealed that this spot conformed to my evolving rule of thumb. The midpoint fof2 is at about 8,000 kHz at the time, so above 7,750 kHz. Note for future use: The rig's freuqncy/fof2 = 14.0574/7.750 = 1.8138 The factor is for use with this figure.

 I made my first QSO on the Project Toucans CW practice app last week and it was DX to Brazil with  PY2UIA , Henrique. He also commented here on the blog . Henrique suggested the iambic keyer would be easier to use if its keying speed was adjustable. That hadn't occured to me since I usually use the straight/cootie key version. And, now it is! On the iambic version of the practice app, the gap+/- size adjustment has been changed to wpm+/- buttons: The keys + and - reduce the dit time, (the standardized time all the other iambic times are based on), by 5 ms, or increase by 5 ms respectively. Give the new controls a try on the iambic app , make a few QSOs of your own, and tell us what other features you’d like to see in the comments.

Ever tune across the HF bands and wonder just how many Morse code operators are on the air at that moment? Thanks to WW5TH’s ingenious setup that reads directly from the Reverse Beacon Network (RBN), you can see it in near real time. The RBN ’s worldwide network of “skimmers” automatically decodes CW signals, giving a fascinating snapshot of CW activity across the globe.  Pat wrote the code up here and you can find the latest data here . Here's an example from this morning. This quick visualization from W5WTH shows how data from the Reverse Beacon Network can reveal real-time CW activity. It’s a great way to see where the bands are hopping and how propagation changes minute by minute. Check out Pat’s GitHub for code details, and watch the data to spot when Morse action is hopping. Want to dive deeper? Try building your own CW dashboard from RBN data using Chat GPT , or explore similar projects in ReverseBeacon.net and Parks on the Air to see who...

 One of the first things I realized about my RockMite, (the tranciever at the heart of Project TouCans, our low power, Morse code only, ham radio), is that the sidetone output is, shall we say, LOUD! This works for me since my hearing isn't great and I'm frequently operating from urban environments where there's lots of ambient QRM in the air around me: horns, crowds, sirens, and so on. I'm often reminded of that the sidetone is loud by other posts I find00 about working on old RockMites. The author usually starts by mentioning that theyauthor turned down the sidetone volume by changing the circuit. The post that comes to the top of my mind is  Taming the Wild Rockmite  by G. Forrest Cook. It has a great section on boosting the output power Rockmite's that I've put to good use. In any event, while working on training an AI to decode Morse code, I started looking at spectograms of the Project TouCans audio output. Here's what I saw  Those vertical bands that...

 Yesterday I posted about ChatGPT5 writing a TensorFlow CW decoder. It's actually decoding clean CW from its training data, but when I feed it audio recorded from an actual HF receiver, nope. I asked ChatGPT to write me a script to inspect the recorded audio and found out why this morning. The first thing I should point out is that this particular model is learning Morse code the wrong way. It is defineitly comparing dots and dashes. It's based on a vision neural net and so... The second thing I should point out is that it's learning Morse code faster than I did, so, you know, that's pretty impressive. OK. Here's why the AI isn't understanding RF-borne CW yet. Here's a spectrogram that corresponds to its self created training data. This particular one is from the video I posted about yesterday. The video is included again below so you can listen to the code as well. This is fairly impressive in that you can in fact actually read out the dits and dahs. Now, ...

 This week's AI project is to create an AI Morse code decoder. I've been working with the new ChatGPT 5 model since late last week. I've asked a few different models if they could understand Morse code. ChatGPT 5 couldn't. Gemini couldn't. That's when it occurred to me that this was probably the perfect time to learn how to use TensorFlow to make an AI. So, I changed my question. I asked ChatGPT 5, "If I wanted to setup a model that learned Morse code using Google's Tensor engines, could you describe the entire process and output the code for me?" To which it promptly, (what an awesome pun!), replied, "Heck yes—that’s a super fun project. Here’s a complete, practical path to a TPU-accelerated Morse code recognizer using TensorFlow + CTC (Connectionist Temporal Classification). It generates synthetic Morse audio (with realistic timing/noise/tempo wobble), trains a small CRNN on log-mel spectrograms, and decodes with greedy CTC. You can run it ...

 Everything old is new again. OK, I just said that to sound fancy. Actually though, the Project TouCans CW histogram , (you know for measuring the consistency of your dits and dahs while sending Morse Code), uses WebUSB and Web Audio , (respectively), to implement a connection to the serial port, where it listens to a Halikey, and to generate a sidetone. That's the old part.  Here's the new part, (to me anyway). It turns out both of those web capabilities are either documented by or a result of something called Project Fugu . I stumbled across the Google Develper site while browsing, (pun totally intended), through the extensions in my Chrome browser. It turns out Chrome Apps are going away, (finally a Google technology I didn't use is disappearing), and should be replaced by, (in Google's estimation), Web Apps, (which are referenced as Progressive Web Apps by Google Gemini.)  The capabilities available are pretty cool and include Web Bluetooth and a File System...

 I've been casting about trying to find the format for Project TouCans video QSLs I liked the best, and I've finally got it! Today, in the Mt. Moriah BLM National Conservation Area, (US-9692), I learned how to do split screen while also recording the screen, and I'm kind of in love with the way these look! In one panel, you can see the controls for Project TouCans. Over the audio, you can hear the QSO and my commentary as I work. In the other panel, you can see the view out the front of back of the phone which is kinda gorgeous on Mt. Moriah! It's just like Project TouCans. It mixes really old tech—the rig—with brand new tech—split screen phones, html/JavaScript control panels, embedded controllers; the last two things at least partially built by an LLM at this point—and looks kinda awesome... to me, anyway! Here's the first QSL. It doesn't have the polish of the ones I make at home—the gang and I are camping near Great Basin National Park—but I love the format...

 The Soldersmoke blog pointed out an interesting post that was a focal point of a larger book review for "Agent Sonya: The Spy Next Door". The post's author was quite convinced that the key protagonist of the nonfiction tome could not and did not construct her own QRP communication rig in the 1930s as stated by the book's author. I beleive Agent Sonya could have constructed the rig  More about the book on the publisher's site The Soldersmoke post The focal review post The 1936 Radio Handbook So, as I was saying: I, for one, believe. I'll admit up front, I didn't have time to read the entire linked post. The following assumes that the issue was Sonya being able to build a transmitter in 1939 for use from Switzerland to Moscow. The short version? It should have been the picture of simplicity.  As an aside and a bit of a bona fides, we routinely QSO from San Francisco to Texas and beyond. That's the same distance as Switzerland to Moscow. I grew up in a ...

 TouCans is back up and running! The battery noise is also just gone! One issue caused both the noise and the key stoppage. The ground wire connection from the keyer relay was in the process of breaking away during the fixes last week. This led to the noise I heard in the rig. It also led to the wire, of course, actually detaching which kept the keyer from working. KO6BTY soldered the ground connection back this morning, and TouCans is back up and running well! Here's what the noise before the fix. This is what a wire breaking in TouCans does. The new battery is doing a good job. Here's how the RBN saw TouCans right after the fix.

 Did you know there's a Morse keying world championship? Neither did I, but Martin, a ham radio op from the UK does. He's compiled a list of episodes from the BBC Sounds podcast that leans into the radio-esque, but contains other interesting topics including a family who, well.. here's a quote from Martin's blog: In 1966, Roy Bates occupied a disused military platform in the North Sea, and moved his family aboard.The next year he declared it to be the sovereign Principality of Sealand. Martin also did a cool thing and started a list of internet links referred to in the GQRP Sprat journal . It looks like it would be a lot of work, but it's a welcome relief from typing in addresses from the printed pages of the journal. I love that the journal is printed by the way, don't get me wrong.

Use any key but 'space' or 'esc' as a CW Key The app provides keyer sidetone on key down. As you key, it maintains two histograms: the one on top, (blue bars), measures the the lengths of your dits and dahs; the one on bottom, (green bars), measures the length of your pauses between dits and dahs, your pauses between letters, and your pauses between words. You can try to use the space key, but on my browser, it makes the scroll bar roll down the page. Halikey instructions are included below. Demo Video Halikey instructions To use the app with a Halikey, simply plug your halikey into your comupter, then your key into the other side of the Halikey. Then click 'Start Halikey'. You'll be presented with a dialog like the following For my computer, the Halikey turns up as 'USB Serial Port (COM3)' When I'm uncertain, I just unplug the Halikey refresh the page, note which port is missing in the new list, hook the Halikey back up to th...