EXCELLENT ARTICLE ON QRQ CW FROM FRED W3NJZ

[Chuck aa0hw]

OPERATING CW AT SPEEDS Of more than 60 WPM

Written by Fred Ryan, W3NJZ, FISTS # 10202

Greetings! Although I have been a CW-only operator for almost 60 years, I just joined the FISTS club, realizing that it wasprobably the sole remaining organization active in promoting CW operating. Sorry to be so tardy in recognizing that. Upon joining FISTS, I asked NancyKott if an article on high-speed CW might be of interest to the FISTS members. She encouraged me to write this article, as manyhad inquired about how CW operating at high-speeds is accomplished. The following is about how I accomplished that. In discussions with other high-speed operators, it appears that they used similar techniques. Their approaches differ in some respects however, and if you will obtain a copy of Bill Pierpont's latest printing of his book, "The Art and Skill ofRadio-Telegraphy”, third revised edition, itcontains an appendix on high-speed code with the viewpoints of others included. Bill was a great teacher, and I hope to explain things about high-speed code operating that will be a fraction as good as his excellent text. CW operators lost a good friend last year when Bill passed away. My article will be broken into two sections: comprehending high-speed code, the second, sending high-speed code. COMPREHENDING HIGH-SPEED CODE Let me first define some of the terms that I will be using. "High-speed" means rates over 60 wpm, and up to well over 100 wpm. "Comprehension, means understanding the code "by ear" without the use of computers. It differs from code where a hard copy is required, such as in the old copying contests or in the relaying of messages. It is of a conversational nature, and is frankly, the most enjoyable form of CW in that the information exchange is greatly enhanced and approaches normal conversational speeds. It requires work to achieve, but in my opinion, is not all that difficult to master. If you have average CW skills and sufficient desire, it can be accomplished. When many of us first learned code, itwas by memorizing the combinations of dotsand dashes and relating them to the various letters and numbers. As pointed out by Bill Pierpont in his book, that is a terrible way tolearn code.One should learn the sounds of the letters and not dwell on the dots and dashes. Once one learns the sounds of the letters and constructs words from them, letter by letter, and puts down the pencil, speed increases up to around 55 wpm are just a matter of practice to achieve. One then runs into a brick wall and further improvement becomes more and more difficult. Many hams have run into thatbarrier and just thought that they had reached their personal limit. That simply is not so. At about 55 wpm it becomes difficult to chain the letters together to form words as the letters are just arriving too quickly toaccomplish that. Another style ofcomprehending code then starts making sense. You must learn to stop concentrating on the individual letters and let your brain dothe work of generating the word. You are, ofcourse, still aware of the individual letters, but you let the brain tell you what the wordis. It's exactly like learning a new language.Most of you already do that subconsciously on common words like "the", "and", etc.

That process must be extended to all words. It can only be accomplished over about 60 wpm, as the letters are arriving too slowly at lower speeds to allow one to recognize the "sound” of the words. The barrier at 55 wpm can be difficultto overcome if you have had years of practice sticking to the comprehension of individual letters. For me, it took about two years to go from 55 wpm to over 70 wpm. Others have done it much quicker than that. The surprising thing is that, using this technique; one can comprehend code much easier over 60 wpm than below. In fact, the comprehension becomes easier and easier as the speed increases beyond 60 wpm, not harder, as the "sound" of the words becomes easier to grasp because of the higher speed of the arriving letters. That isexactly the way we learn to talk. The words are spoken and we learn to recognize the word's "sound". If we had to figure out wordsby having someone say them slowly, letter by letter, we would be largely illiterate. The spelling of the words comes at a later time in order to allow writing and reading. It is not important for speech, however. Shortly afterlearning to comprehend words in learning to speak, the words are put into a flow of thought, such as a sentence. The application of this process to comprehending C W well over 100 wpm will be discussed later. Making the transition from constructing words, letter-by-letter, to comprehending them directly takes effort. I can't suggest that you listen in on other high-speed operators to improve that skill, as there are few still active on the bands. George Hart, W1NJM has ceased operating his high-speed code practice sessions.When I made the transition there were more high-speed groups active than presently, and I would tape their sessions and play them back over and over. Some of us are still active a few times a week close to 7.032 MHz, or during the winter, close to 3.532 MHz. That is the FOG, or fast operator's group, and typicallyoperates between 55 and 70 wpm. Another group that is currently inactive, but will hopefully resume activity, is the SOB group, or the speed operator's bunch. When in operation, their speeds vary from 70 wpm to over 100 wpm. They operate on frequenciesclose to the FOG frequencies. In the absence of many active high-speed groups, I have a couple of suggestions for breaking that "brick wall" at55 wpm. Make up some tapes or computer documents with your own messages and play them back, over and over. When they become too well known, make up new ones. One thing I recommend is to create a list of words commonly used in conversations. Instead of just making up a text of them at, say 60 wpm, use an analogy of the Farnsworth method and send the words at 70 wpm, but allow large spacing between the words. This can be done using a memory keyer such as made by MFJ, or one of the computer CW programs for sale that allow for that. Repeat the same word over again two or three times before moving on to the next word-It is impossible to comprehend words at that speed by constructing them from the individual letters. That forces you to grasp the entire wordinstead. It is not an easy transition to make, but stick with it. After you feel that you can comprehend code over 60 wpm, try joining into one of the active high-speed groups.When you feel ready to take that step, another problem comes into play. In order to comprehend the words directly you must be completely relaxed. It is easy to become intimidated by the other operators. Once you cease to be relaxed, all is lost. You will misscomprehending an important word, tighten up, and drop back to comprehending letters. All you can hope for is to run into someunderstanding high-speed operators who have gone through the same learning experience, and who really want you tobecome one of them. I was very fortunate to have had some very patient high-speed operators to help me along, Tom, W4BQF, Larry, N8LN, and Ken, W9LRV.

Hopefully there are some such people within your reception area. I would like to make a few final comments about high-speed code comprehension. I know of no high-speed code operator who uses a computer program to comprehend the code for them and display it on a screen. There are two reasons for that. The very most important reason is that it would remove all of the enjoyment of operating CW. I love the CW language, and using it gives me great pleasure. Reducing it to a computer-generated copy would bring no joy to any CW operator. It would be just like operating RTTY or digital modes. The second reason, completely subservient to the first, is that the comprehension programs that are on the market are completely outclassed by the human brain in understanding CW, through the typical band conditions of heavy QRM and QRN. Anyone who uses a computer to comprehend the code soon loses interest and drops out. Finally, a few comments on what I think the ultimate speed of comprehension of CW is. In my opinion it is from 200 to 300 wpm, similar to our limits of speech. That limit, of course, is greatly reduced if we are trying to explain something very complicated and technical. One must drop down in speed considerably to form your thoughts on such subjects, and for the other operator to understand what you are saying. That isalso true when speaking. Learning to comprehend words will get you up to around 100 wpm. To get faster than that you have to stop paying attention to the individual words, and just follow the flow of thought, just as we do in ordinaryconversation. I have not yet accomplished that, probably due to the absence of operators who can send quickly enough.One person who has is Tom, W4BQF. He can follow the flow of thought up to 140 wpm. You have to realize that we spent many years learning to speak quickly.Speaking quickly is much easier thansending CW quickly, so that comparable amounts of time would be required to get to over 200 wpm, and with nobody capable ofoperating a keyboard that quickly, you get no practice. In the subsequent sending section, one suggestion for achieving that will be discussed. One might ask, "why bother mastering CW at speeds over 60 wpm?". In addition to its being much more enjoyable than low speed CW, I have always found that achieving something new is one way of keeping young, It always bothered me to find that others could do something that I couldn't do. Learning to do new things, like high-speed CW, keeps the brain from atrophying. Another reason is that CW is a superior language, compared to speaking. It is digital in format and, if proper spelling isused, comprehension is unaffected by things like bad pronunciation, dialects, or just poor speech. If it is sent perfectly, it is comprehended perfectly. Try that with speaking to some people in the English language!

SENDING HIGH SPEED CODE Written by Fred Ryan, W3NJZ, FISTS #10202 In my previous article on operating high-speed code, the comprehension of it was described. Now the sending part will becovered. It is the harder task to master in high-speed CW operating. Like many, I started out using a straight key and reached a top speed of less than 30 wpm. Moving on to using a bug, the top speed peaked out a little over 40 wpm. Next came the paddles, where my maximum sending speed peakedout around 60 wpm. The instrument presently useful forsending over 60 wpm is the keyboard, and it can be used successfully to send well over 100 wpm. Having gone through thissuccession of sending means and gettingproficient in all of them, I have to tell you thatsending high-speed code on a keyboard over 70 wpm is BY FAR the most difficult of them all. It is not just about having the fingers type quickly, but you must organize your thoughts, get the spelling right, tell the
fingers what to send, and do this with perfect cadence with no gaps between words longerthan the standard spacing. Some CW operators look down on using a keyboard to send code. All I can say to them is to show me that they can master its use over 70 wpm. They will soon find that using paddles at 55 wpm is child's play compared to that. An expert typist capable of transcribing text at over 100 wpm, would be totally unable to accomplish the other skills required in sending a high-speed QSO without extensive practice. One should also recognize that even when some are using a keyboard at low speeds, you don't know the physical limitations of the operators sending. It maybe all that they are capable of doing at that time. We can't afford to lose any C W operators, and no one should look down on,and possibly discourage, an operator for using whatever is within their capabilities forsending. EQUIPMENT USED FOR SENDING HIGH-SPEED CODE One must obtain a high qualitykeyboard. There are a lot of junk keyboards floating around. They exhibit large variations in the mechanical precision of their actions,with lots of wobble in the keys and variationsin operating pressure from key to key, and often an excessively low operating pressure. My favorites are the older IBM keyboards, obtainable at computer or hamflea markets for next to nothing. They have very good precision in the action; with real springs under the keys instead of air filledplastic bubble springs. Most of them can have the tops of the keys removed for non-essential keys, reducing the number of inadvertent errors produced when a finger brushes against them. The IBM model M allows you to completely remove those keys, making errors with them impossible. It also raises the level of the number keys, reducing the possibility of hitting a number while going for a letter in the upper row of letters. I use a Model M. This error problem goes back to the original spacing chosen for the keys back in the 1800s, when women with small fingers were the typical typists. That spacing is far too small for use by men with large fingers, and since a source of wider key spacing keyboards is to my knowledge, not available, modifying a keyboard by reducing the chance of errors isall that is left. There are two choices in the arrangements of the keys on the keyboard, the common "qwerty" arrangement, and the Dvorak. If you are just starting to use a keyboard you should consider learning itsuse with the Dvorak arrangement. If you are already a competent typist, stick with the common letter arrangement. Your top sending speed will be limited by factors other than the ease of finger movement. Various pieces of code generating equipment are available. One can use an MFJ unit, or a computer running one of the software programs for generating code, such as CWTERM (CW 500) or YPLOG*. The advantage of the computer programs is that they will display the code as you type it. That allows you to pick up sending errorswhen your fingers tell you that you might have made one. If you are typing much faster than the text going out, you have time to correct the error. Not all rigs are capable of operating at high-speeds. The later Ten Tec rigs are one example of rigs that will not key over 70 wpm, although their earlier, non phase lock loop frequency controlled rigs, are. Look into the manufacturer's claims before buying a high-speed rig. At high-speeds it is important to put out a strong signal, or the comprehension becomes difficult. A combination of QRO and antenna system is often necessary. HOW TO TYPE HIGH-SPEED CODE If you are fortunate, you already type by the "touch typing" method. That means that you don't look at the keys to see which one to hit next. If you are like many, you just picked up a keyboard that came along with a
computer and started typing by looking at the letters, to see which one to poke. That is not an approach that you can use to send high-speed code, and you are going to have to learn "touch typing". In typing classes,they force you into learning the location ofthe proper keys by eliminating the symbols on top of each key. Taking a typing class isa good way to learn the proper way to use a keyboard, but many will find that they can teach that to themselves. In addition to enabling you to send high-speed code, you will find it coming in handy when typing an E-mail or creating a document. You should be capable of typing atleast 10 wpm faster than the text you are sending out, keeping the buffer filled aheadof the transmitting text. That allows you tohave perfect cadence in the code. It also allows you some short time to correct sending errors if your message appears on a screen. The space bar must be used rigorously so that words are not run together. Once you have practiced sufficiently to get your typing speed over 60 wpm, try making up imaginary conversations at progressively higher and higher speeds. That teaches you to get your thought train going fast enough to make up the message, get the spelling correct, and type it without the buffer going empty. There is no simple way to accomplish this and considerable practice will be necessary to achieve high-speed sending. Unfortunately, the side tone that you like to listen to when you are sending is now a distraction, as you hear the side tone ofletters that you typed earlier, not the ones that you are presently sending. Asmentioned before, you should be typing atleast 10 wpm faster than the text being transmitted, so that the side tone is sending you irrelevant information on whether or not you are typing the correct letter in real time. A side tone that corresponds to your real time typing and not the buffer delayed transmission could be used, but I don't thinkit would be of much use at very high-speeds (over 70 wpm). It is best to turn off the side tone completely over 70 wpm as it can actually prevent you from sending the higher speeds. One can get some on-the-air practice with high-speed sending by using it even at low speeds like 40 wpm. If I am sending at40 wpm, I will type in bursts of over 70 wpm just to get typing practice. The buffer and code program will slow those 70+wpm bursts to the proper speed. Getting from sending at 70wpm to over 100 wpm is just a matter of more and more practice, but progress will continue. From my experience,for typing over 70 wpm you should; A, turnoff the side tone completely, B. stop looking at the screen except maybe a glance now and then when your fingers tell you that you might have made a sending error, C. spend all of your energy on organizing your thoughts, getting the spelling correctly, and especially, constantly visualizing where the fingers are going to go next. THE FUTURE OF HIGH-SPEED CODE Continued activity in high-speed code depends on getting new operators on board. The number of high-speed operators in this country is down to a handful, and will disappear before long unless new members will join in. That is the principal purpose for my writing this article. I make the followingoffer; if anyone reaches that "brick wall" near 55 wpm and can also touch type over 55 wpm, contact me at QRZ.com and I will work with you to break that "brick wall". Once you have achieved that, you will quickly see how much more enjoyable high-speed CWoperating is, compared to QRS. It was mentioned earlier that it wasmy opinion that high-speed C W operating could be extended beyond 200 wpm, if a quicker way than using a keyboard to send it could be found. The logical candidate to replace the keyboard is computer voice recognition. These programs get better every year, and could be used to convert your speech into CW, with the upper speed limit being limited by how fast you can talkI'm sure that some would reject that as not being "true CW". Those same might say that
using a keyboard, paddles, or a bug, is not "true CW". To me, the heart of CW is the comprehension of it, and will use any means possible to achieve higher QSO speeds. I hope that you found these two articles of interest. “73” Fred, W3NJZ