Stuck at 10-12 wpm on the way to 20: what actually breaks the plateau?
- Replies 2
- Views 48
- Created
- Last Reply
Top Posters In This Topic
-
Admin_Terry 1 post
-
Mod_Mike 1 post
A better way to browse. Learn more.
A full-screen app on your home screen with push notifications, badges and more.
Opening a thread for people working on the 5 to 20 wpm stretch. The early climb to about 10 wpm tends to feel steady. Then many learners stall somewhere around 10 to 14 wpm, and progress stops for weeks even though practice hours haven't dropped. Here is a rough map of the problem, and I'd like to hear what has worked for others.
The plateau often has a mechanical explanation. At 5 wpm, a dit is 240 ms (dit length in ms = 1200 / wpm). At 20 wpm it is 60 ms. Somewhere in between, you can no longer consciously count dits and dahs and translate them into a letter. The brain has to switch from analyzing the character to recognizing it as a single sound pattern, like a spoken syllable. That switch won't happen if you keep slowing down to analyze. It usually needs characters sent at a speed slightly too fast for you to copy comfortably.
That is the reasoning behind two common training approaches. The Koch method starts with the characters at full speed (usually 18 to 20 wpm) and adds one new character at a time as you reach roughly 90 percent accuracy. Farnsworth timing also keeps the characters fast but stretches the gaps between characters and words, so you have time to think. Both avoid the habit of learning at 5 wpm with long, slow characters and then having to relearn everything at speed. If your practice material is sent at a true 8 wpm with normal spacing, consider whether you've effectively trained yourself to count.
Some questions for the thread:
1. Head copy. When did you stop writing every letter down, and how did you make the change? One approach is to start with callsigns and common short words, since those are the most predictable, and keep paper for numbers and names. Another is to write nothing during a practice session and check afterward against the source text.
2. Drills. Which do you find more valuable: random five-character groups, real words, callsign lists, or full text? Callsign drills have a clear payoff because they train fast recognition of mixed letters and numbers, which random words don't. Common-word drills (the, and, for, name, rig, ant, wx, QTH) build the whole-word recognition that matters once you hit QSOs.
3. Software versus ear. Decoders such as fldigi, CW Skimmer, or phone apps are good for confirming what you heard, and poor as a substitute. They struggle with QSB, noise, and hand-sent timing, and if you watch the screen you stop developing your own ability to listen. A reasonable compromise is to copy first, then check with the decoder.
4. Keyer and paddles. Iambic mode A and B behave differently on the final squeeze, and your timing can feel wrong if you've set up the paddles one way and learned on another. Contact gap, paddle tension, sidetone (most people settle somewhere in the 500 to 700 Hz range), and weighting (default is about 3:1 dah to dit) all affect how clean your sending is. Poor sending sounds sloppy to the person copying you, and it also makes your own copy worse because you're training your ear on bad rhythm.
5. On-air practice. Slow CW nets, contest exchanges, and simple rag chews are all useful. Contests are good because the exchange is predictable: callsign, a signal report, a serial number or a zone. You hear the same short patterns repeatedly. If someone sends faster than you can copy, QRS (please send more slowly) is a normal request, and no one should mind it.
What does your weekly schedule look like, and which element of it has made the biggest difference?
Link to comment
https://www.hamradiobase.com/forums/topic/11145-stuck-at-10-12-wpm-on-the-way-to-20-what-actually-breaks-the-plateau/Share on other sites