3
$\begingroup$

I've been learning to decode cw for a while now, mainly by using the Word training feature of https://lcwo.net . I use this rather than the exercises with random characters, both because it's more fun to hear words and because a hand injury makes it hard to type fast enough.

I've heard that it's good to learn at a "high" speed, like 20 wpm to get the feel for how each symbol sounds rather than counting dits and dah's. Now here's the thing: I find the easiest speed to hear is somewhere at 30+ wpm, while the 20 wpm words are harder by comparison! At 20 wpm, I hear dits and dah's, and I just can't "hear" the separate units as easily.

My question is: Should I try to practice more at the slower 20 wpm, or can i just go on practicing at becoming more confident at 30+ wpm, and hope the slower speed will come naturally later?

I can't decode longer texts at 30 wpm without longer spaces between the words, but I know it's common to learn with letters at high speed with gaps between them, and maybe this could be effective by a similar logic... ?

Many guides give warnings like "5 wpm and 20 wpm are different skills, so learn the right one". I find nothing on my problem, but I worry slightly I might be doing work I'll have to redo later.

$\endgroup$
2
  • 1
    $\begingroup$ How does LCWO teach you to copy call signs? They don't seem to be "words" in the conventional sense, so doesn't learning to copy call signs equate to learning to copy individual letters? $\endgroup$
    – Brian K1LI
    Commented Mar 20, 2021 at 12:11
  • $\begingroup$ Don't slow down. I'm in the same boat. I find copying words instead of chars a lot easier and more satisfying. It's like comparing to speak vs spelling. The way you learn it is like a baby learns a language. There's no better way. You can still learn to spell later. Also watch Kurt Zoglmann's Morse Code Speed vs Proficiency: youtu.be/5BlhhBK1CBw $\endgroup$
    – robsn
    Commented Feb 10, 2022 at 12:26

5 Answers 5

5
$\begingroup$

It is one thing to know the English word "car".

It is another thing to be able to understand this word in all the ways it may be spoken. American English will end it with a rhotic "r" (/kɑɹ/), while a RP accent will omit the "r" entirely and just say /kɑː/. Some people speak fast, some people slowly. Sometimes you will have to understand the word in a noisy room.

In other words, vocabulary and proficiency are different skills.

Currently you are focused on learning vocabulary. You will gain proficiency later. I would not worry too much about the settings you are using at this stage in your learning. Proficiency comes from practice, not the settings you are using in your training program now.

$\endgroup$
4
$\begingroup$

If you want to actually successfully operate, then you should (eventually) practice copying Morse Code at the full range of WPM speeds you expect to hear over the air. Most Morse Code practice software allows you to change the WPM, try from maybe 12-20 WPM for SKN (straight key nights) up to 24-40 WPM for CW contesting. If one practices on only slow characters, it will be hard (if not impossible) to copy at higher WPM speeds. This is likely at least partially true for the inverse as well. If you vary your practice WPM rate, you will likely be more successful once you get on the air.

For QSOs (especially for DX), don't forget to learn to reliably copy random callsigns, not just common words and abbreviations.

$\endgroup$
3
  • 2
    $\begingroup$ In training neural nets, this type of variation in training is called data augmentation, and is required to prevent overfitting, which greatly reduces the ability of a neural net to generalize. The cortex is not the same as an ML CNN, but there are similarities. $\endgroup$
    – hotpaw2
    Commented Mar 20, 2021 at 16:49
  • $\begingroup$ Callsigns is something I've thought in the back of my mind I probably should do. You brought it forward! I'll start in the next session! $\endgroup$
    – EdvinW
    Commented Mar 20, 2021 at 21:21
  • $\begingroup$ I use rufzXP for practice copying call signs. It has been a help for contesting. $\endgroup$
    – Deepstop
    Commented Apr 4, 2021 at 21:17
3
$\begingroup$

CW is an interesting discipline. There are two distinct elements, encoding and decoding. You reference decoding yet refer to encoding guidelines.

Learning CW, I was taught to associate the DIT, DAH and REST as music, both in tempo and beat. The principle being to process a combination of sounds reflexively not analytically, as you stated in your initial post. It is more efficient to recognize how each symbol “sounds” rather than counting dits and dah’s. Progressing in that thought, it is more efficient to recognize how common or repetitive words sound i.e., rhythmically. This applies specifically to the “Q” code shorthand such as QRV; QRZ; QRM; QRA and so on. One can see how the ability to decode CW will quickly outpace the ability to encode CW.

Encoding CW by nature of mechanical transfer is much slower when doing it by hand. CW is all about timing, the accuracy in timing should take precedence over speed. You are probably aware there are a couple different methods, Koch; Farnsworth; Paris and others. Everything about CW is proper timing/spacing.

The goal then, learning to encode CW with metronomic precision. Anyone listening will enjoy exchanging information with you. Sloppy timing increases both errors in decoding and frustration for the decoder. The most important thing to remember in the discipline of CW, its always about the ability of the decoder. Listening to clearly articulated 5wpm may be frustrating because you can decode much faster. Yet you can only transmit as fast as the receiver can decode. While you maybe competent to transmit at 20wpm if the receiver can only decode 5wpm that is all you can send. The continually QRS will slow efficient communication just as quickly if not more so then the poor discipling of timing/spacing.

$\endgroup$
1
  • $\begingroup$ Hello Ron, and welcome to this site! Thanks for a nice first answer, we look forward to seeing more of you here. :-) $\endgroup$ Commented Mar 20, 2021 at 20:12
3
$\begingroup$

I learned it years ago by listening to the dits and Dahas spacing with the spacing much slower. I would recommend go as fast as where you can make out the different dits and dahas. Then as your Think about it" time gets shorter, decrease the spacing pad. For example, the individual letters at 10 or 15 wpm but the spacing is 5 wpm. I also hit a plateau every now and then. I put it down for a week or two and go back to it. Good luck and don't get discouraged. It helps to find a ham learning CW also. I once called up another ham on the phone so we could interact.

$\endgroup$
1
  • 1
    $\begingroup$ Hello Lowell, and welcome to this site! :-) $\endgroup$ Commented Apr 1, 2021 at 21:52
1
$\begingroup$

If you listen on the air, you'll hear Morse QSOs at a wide range of speeds. These days probably most code is formed by keyers, which form individual characters perfectly, but you'll also hear people sending code with straight keys and bugs, which can have very different timing. In particular, many operators who use bugs send code with fast closely-spaced dits and relatively-long dahs. In addition to wide variety in speed and timing, real QSOs on HF bands often feature noise, sending mistakes, interference, fading, weak signals, and so on. All this makes copying real QSOs often much more difficult than copying practice recordings.

If your goal is to learn Morse code so you can make Morse QSOs, then my advice would be to abandon practice copying 30+ WPM code for now. You'll need to learn to walk before you can run, even if "running" is easier with practice recordings. I'd recommend practicing copying 15 WPM code. Most QSOs at speeds slower than that is sent using the Farnsworth method, with the characters sent at 15 WPM or so with longer spacing between the characters and words, so there's not much need to copy characters slower than that. If you can copy 30+ WPM code already then you may be a natural, so your speed may quickly increase once you start making real QSOs.

By the way, for me one of the key skills for making faster Morse QSOs was learning to "throw away the pencil" and copy what the other operator was sending in my head, rather than writing or typing each character as it's copied. But that's a skill for later; before trying to learn that I'd recommend just making a bunch of QSOs at moderate speeds. In other words, make sure you've mastered the basic skills first.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .