[LCWO LOGO]  

Login

User name:
Password:


Language
Български Português brasileiro
Bosanski Català
繁體中文 Česky
Dansk Deutsch
English Español
Suomi Français
Ελληνικά Hrvatski
Magyar Italiano
日本語 Bahasa Melayu
Nederlands Norsk
Polski Português
Română Русский
සිංහල Slovenščina
Srpski Svenska
ภาษาไทย Türkçe
Українська 简体中文

Who is online? (18)


LCWO Discussion Forum [Atom LCWO Forum Feed]

This is a simple discussion forum for LCWO users. Feel free to use it for any kind of discussion related to this website.

Thread: Learning CW: what has worked for me and what hasn't

Back to the Forum

AuthorText


Posted: 2018-12-31 07:18
I am 56 years old and started learning CW on LCWO in January 2018 (I had been a member before but had no time to practice).

At the end of one year of daily practice, I have reached the point where I can send and receive CW at 20/20 mentally (sending up to 25wpm with a paddle and copying in my head at 20 wpm). My goal was to, initially, establish a solid working speed (20/20) and then increase it to the limit of my capabilities.

What I have found is that:

1. Studying one lesson a day at the beginning and moving on, as the long as the results are reasonable is the way to go. The big problem for most of us seems to be "wear and tear": acquisition of code with the full-blown Koch method and moving on to the next lesson with results over 90% seems to take for ever and people drop out.
I would suggest that the "once over lightly" approach is more conducive to results. In my personal case, after listening over and over to a lesson I would get anything between 88-98% (sometimes I only got 74%). I just moved on, on the assumption that things would "gel" and they generally did.

2. One should start as fast as one can (I learned at 20/10 throughout). Even if one doesn't hear the difference between the characters initially, it will happen.

3. It is, indeed, critical to practise every day, without exception. Several (3-5) short 15mn sessions have been the most beneficial for myself.

4. Using MorseMachine, to drill the difficult letters, in conjunction with the lessons is extraordinarily helpful.

5. Once the code has been thoroughly learned, the occasional day off seems to help in consolidating acquisition.

6. I put a lot of effort in Word training on LCWO to increase my speed, and, while I reached 40 wpm at my 43rd attempt and was able to continue at that speed, I found it didn't translate into a better capacity to copy code in real life situations where one must copy continuously long strings of text.

7. What has been the most helpful is the interactive mode pioneered by the K44 (as the "Echo mode") and the CW Machine( as the "Flashcard Mode"). The CW Machine program generates random words, combinations of letters or numbers and sends them to one and one must send them back using either a straight key or a paddle. The implementation of this mode on the CW Machine is far superior to that on the K44 since, with the CW Machine Trainer one can choose to receive words from a word list, numbers or alphanumeric combinations, whereas the Echo mode on the K44 is limited to alphanumeric combinations and doesn't support sending words from a word list.
I started training using the CW Machine and its training program with word lists of the most common 100 words of the English language and now the most common 1,000 words and have found this immensely helpful.

8. If I could, with all due respect, make one suggestion to our kind administrator (I don't know if this is feasible or not): would there be any way to support similar interactive capacities on LCWO so that people who don't have access to keyers such as the K44 or Morse Machine be able to train in this way?

Preferably with a real straight key or paddle modified for the purpose and not a mouse?

9. Last but not least, what I would like to say to other members who are just starting is: it can be done! One just has to stick to it.

There is a very evocative German term my late father used frequently (I don't know whether it is still part of modern German usage): one needs "Sitzfleisch" (literally, "sitting flesh", i.e. the capacity to sit and work).



Posted: 2019-01-07 21:12
Thanks for dealing with us your experience

Concerning 20/10 I have the impression that it is too fast for most people, progress after lesson 10 is too low, they get demotivated and stop till January 1 of the next year..

So may be the instant-satisfactory-generation is better off with starting the course with 20/5

It is possibly your misundertanding the next lesson is open when you constantly copy 90% or over correct. Only ONE score of 90 or higher is the decision boundary.

In higher numbered lessons >20 the new character is hardly present, so it is a good thing to exercise the last learned 5 characters in custom code exercises.

Morse machine does not exercise pipelining which is essential when you want to decode faster then QRSS.

Without investing in expensive machinery, you may exercise sending with a decoder, or even better play a known mp3-text, may be one two three and so on, with 10 wpm and transmit it with a key/battery/beeper synchronous.

Copy text lines not and words up to 40 wpm is due to the fact you need more time to decode the word thàn the wordspacing allows. More exercising words only will solve that problem, another better method is using ARRL mp3 exerciding files of 15 minutes, when you play the same file every day till you copy 100% you learn autimatically most used words and decoding text in your head.
http://www.arrl.org/40-wpm-code-archive


Posted: 2019-01-11 13:20
I have something to add to the excellent suggestions. I also found 20/10 was crazy fast for me. Eventually, I settled on 20/6, and worked my way through the 40 lessons. When I found myself stuck, and unable to reach 90% (or close to it) I would hang-out on Morse Machine and just casually type away at the letters for 10 minutes a day. I went a whole month like that, when I got stuck somewhere around lesson 23, and eventually got over the hump.

I have now gotten back into it at 20/10, from the beginning (and I found I could do 20/15 for the first several letters). I'm still just putzing along at about 10 mins of study a day, and am now on lesson 19 after about 3 weeks at 20/10. I blew through 18, 19, and 20 today, and it was easy to get over 90% each time. After I finish 20/10, I'll move on to 20/15 (or perhaps word practice, instead).

QSO's are still alien to me. I get frustrated every time I turn on the radio, because I live in Asia, and a good percentage of the CW here is in other languages (or has weird codes that are local ID numbers, and annoying). It seems really difficult to find a "normal" callsign, as well. Everybody is a something something slash something something local code something. Makes no sense, sometimes. I'm hoping to work into where I can work US ops on 20m and 40m. I know I can hear them (I have a 2-el yagi at 75 feet, here in Seoul).


Posted: 2019-01-13 09:10
Thanks to nonagenarian for the excellent suggestions. I am thinking of starting to take dictation of texts in CW from a known source, such as the ARRL archives so kindly mentioned and transmitting them back for deciphering by software such as CW Trainer. Just as I had to do as a child in Switzerland when learning French.


Posted: 2019-01-13 14:01
[quote=nonagenarian]
[SNIP]
. . . another better method is using ARRL mp3 exerciding files of 15 minutes, when you play the same file every day till you copy 100% you learn autimatically most used words and decoding text in your head.
http://www.arrl.org/40-wpm-code-archive[/quote]


I think listening to morse mp3s etc
( eg on your cellphone when you are in a queue, or instead of watching TV )
is a very good idea,
but
if you listen to the same file all the time you will soon learn which letter is coming up next, which rather spoils it.

Fabian published a utility for generating mp3 from text, so you can use this to make 4 or 5 ( or more ) files of the same words in different orders and play them all back in random order.

If you use your cellphone instead of just a "dumb" mp3 player, then an App which randomly takes from a list of words and morses them would be better . .


Posted: 2019-01-14 06:35
Thank you for your suggestion. I already use pretty much daily the "Convert Text to CW" utility on LCWO to convert news articles of interest to CW and listen to them.

I downloaded ebook2cw, the program Fabian wrote, to convert ebooks to CW but have been unable to use it (I run Windows 7). Hopefully, I will be able to find a way to make it work as I would like to listen to longer items than news articles in the NYT or the Guardian.


Posted: 2019-01-14 12:51
I just downloaded the mp3 only version on a windows7 64 bits computer. Works fine.

Possibly you use an old compiled version, in Help look for compatibility and you will find instructions. However just download the file from the indicated website, mp3 only Install it and start it in a dos window, followed by the name of the textfile to translate.


Posted: 2019-01-15 09:37
Success! Just converted Stevenson's "Dr. Jekyll and Mr Hyde" to mp3 using the GUI.

Vielen Dank, Fabian, für alles was du gemacht hat!


Posted: 2019-01-19 05:00
FYI Project Gutenberg is a great source of input text, although since it from literature you won't get too much number copy.

nonagenarian: I took me a while to realize that as I progressed through lessons that indeed the weighting of characters was I guess I would say non-existent or random. I did some short time interval runs on day an had a few cases where the new character was only sent a few times - I was quite surprised. I have never seen Fabian's code, but I assume that since the play time intervals are fixed, that at the start of sending he has a bufffer of the entire session, if that's true, its very simple in software to divide total characters calculated to be sent at given speed by the number of distinct characters in the lesson to know how many of each to send; then you fill a data structure with in input, such as "KKKKKMMMMMUUUUU....." and then randomly select from the first structure to a new structure that will be the send buffer. This way it is pseudo random in order, but very close to equal weighting. I'd guess this is the technique G4FON uses to enable his option of "extra stressing" the latest character.

Well all these developers are providing many diverse techniques to help all of learn cw. We appreciate the effort.

If someone ever finds the one silver bullet to learning, please send on my way.

73


Posted: 2019-01-19 14:01
AFAIR (R=remember) Fabian DJ1YFK started this website with equal probability for each character in a lesson, but later on he puts some emphasis on the new presented character.

But due to the relative less presented new character in higher numbered lessons (above lesson 10) and the advice to start with 20/10, I am seriously afraid there are a lot of people quiting the course, that should have collected the whole set of characters when the advised default was 20/5

I am just home of celebration the 93 birthday of a local ham. Driving a car with icy weather and the low sun straight in my face took all my vision away and I stopped my car. Too dangerous to proceed.

The ham I attended his birthday said to me: Your metabolism is fast, mine slow, slow metabolism includes delayed reaction time, that leads to being bad in learning Morse code. (He is solid copy 15 wpm). I learned it with the eish5 tmo ch 0 method 80 years ago, when possibly Koch and farnsborough were not yet born, I was around 15 years old. Just by listening to a machine tape of a coastal station with a cq callsign and list of frequencies they were listening, in an infinite loop, and a book presenting the characters and their Morse code in dots and dashes. I worked up to 20 wpm by listening to the coastal station daily news bulletin in a couple of months , and life took possession of all my time. Untill I attended a local hamfest 10 years ago, did obviously still 20 wpm and decided to train on this website and ARRL, G4FON off line and JustLearnMorsecode. (The last one is less suited for QRQ training because the leading and traling edge of marks is substracted from the nominal mark time)

I worked up in a couple of years to 40 wpm solid copy by head, and a reasonable understanding copy of plain text between 45 and 50 wpm.

Will be because my metabolism is fast, but I am also passed 90 years old, so I invite other people present on this website, to present themselves in my user group nonagenarians (== people with 89


Posted: 2019-01-19 14:07
Error due to website allowed text: with age greater than 89 and smaller then 100 years old

Back to the Forum

You must be logged in to post a message.

$Id: forum.php 62 2015-01-12 17:34:44Z fabian $