The Art & Skill of Radio-Telegraphy

-Second Revised Edition-
William G. Pierpont N0HFF

Previous Chapter Table of Contents Next Chapter

Chapter 15 - Timing

Timing Is The Heart Of The Code

Timing is the heart of the code: there is no code without it. Clear intelligibility depends upon right proportions.  However, it is true that some distortions are less unintelligible than others, and people can learn to read that sort of stuff -- but is it justifiable?  Control of timing rests entirely with the sending operator.

For this reason, attention to careful timing is first needed when the student starts to practice with a manual key, especially a straight key, though also with most other types.  This is one reason why some good teachers discourage the use of anything but a keyboard by a beginner.  Most modern teachers agree that it is important not to specifically mention the subject of timing until the student has learned the alphabet and numbers so well by hearing them that he recognizes their patterns without hesitation.

Some teachers recommend that other than using "dits" and "dahs" to verbalize characters, they should not be time-analyzed at all in teaching, but that it all be done intuitively by sound.  On the other hand, some excellent teachers of the past (before keyers and keyboards) have insisted on teaching precise timing, in terms of its elements, from the very first.  Accurate timing is vital, but it must never distract the student from the basic recognition of characters by their essential unity of patterning: it must not lead to his breaking down the characteristic rhythm of the characters by analyzing them into components.


The basic unit of code timing is the Baud, which is the duration of one dit (or "dot"), denoted here by 1 for the "on" signal, and by (the equal unit) 0 (zero) for silence, the "off" signal. The basic contrasting signal to the "dit" is the dah which has a duration of 3 units (111).  It is obvious that each dit and each dah must be separated both before and after by at least one unit silence (0) in order to be distinguishable: this (one unit) is the normal spacing between parts of a character.  Normal spacing between characters within a word (or group) is three units (000), and between words (or groups) is seven units (0000000).

Punctuation marks normally follow the last word with only one character space (000) between.  It is these components of time, signal "on", short or long, and "off", which produce the patterns or rhythms which distinguish one character from another.  We must learn to hear these patterns, sense them, feel them, and this is best done by hearing well-sent code.  In actual practice, individual operators may and do deviate somewhat from the standards given above.  This may be for emphasis or because of communication conditions, as well as unconscious individual variations.


In the perception of rhythm by the human ear the precise duration of sounds is, within fairly wide limits, unimportant. If the longer signals (i.e., dahs) are substantially longer than the shorter ones (i.e., dits), the ear will be satisfied.  While our judgment of the duration of brief sounds is poor, we can judge the relative length of brief silence intervals much more accurately.

It has been said: "If you take care of the spaces, the 'marks' will take care of themselves." Spacing, the periods of silence between parts of a character, between characters in a word or group, and between words, is critical to good receiving.  Sloppy or hastily sent code can be a terror to receive and understand. (Beyond some speed the persistence-of-hearing effect tends to fill in the small spaces and make us unable to consciously recognize characters.)    [In American Morse with its three different lengths of dashes, each successively longer one was taught as being twice as long as the next shorter one  -- an amount which is clearly "hearable" under almost any conditions. In practice, however, because they used a telegraph sounder which marks the start of a signal by one kind of click, and its end by a different-sounding click, with silence in between, these durations were often shortened without confusion for reasons stated above.  The same thing was true for its internally spaced characters.]

Code "translators", microcircuitry for converting code into print, break down when sending is poor or interference is severe.  The human ear and mind, however, can copy rotten code far better than any machine.  The "ear" is a forgiving organ: by mental compensation we can quickly recognize and read stuff as passable code, which if it were recorded on paper tape would show its glaring defects.  In the presence of interfering signals and static, and to a large extent during fading, the "ear" can be trained to pick out a very weak signal and read it well.  (Chapter   11)


From the very beginning of telegraphy, as soon as the art began to  spread, the individuality of operators became apparent. Little peculiarities in sending stood out to identify each one, just as voice quality and style do in speaking.  Mostly these were subtle little things which did not distract from easy intelligibility.  But they did involve aspects of timing and rhythm.  We hear them today on the CW bands among amateurs using hand keys just as they did among all operators in the past.

For many operators there was a certain pride in this.  However, there is a danger here also, because some operators deliberately created peculiar styles of sending as a sort of trade mark. When such distortion reaches a certain point and becomes habitual, intelligibility suffers.  We hear some of these operators today on the air. They do not seem to realize, or perhaps even care about the difficulty they cause.  With the advent of the "double speed key", also called the "side swiper" or the "cootie key", a key which is operated by sidewise movement, with one contact on each side, a new set of peculiar styles of sending arose.  Sidewise instead of up and down motion helped relieve some forms of fatigue, but also the peculiar motion patterns developed a different timing pattern, one that is sometimes hard to copy.

The use of "bugs" , semi-automatic keys (the best known being the "Vibroplex") which soon became very popular, also gave rise to various personal sending peculiarities unless the operator was careful.                              "SWINGS"  One of the most interesting developments in disturbed timing of hand sending was the rise of so-called "swings."  Swing has to do with a change in the normal rhythm of sending, sometimes described as a change in symmetry or lack of it: a peculiar way of forming the characters.  Swings most commonly developed among marine operators within a close-knit group having a large volume of specialized communications.  Thus we have the names "Banana boat swing", "Lake Erie swing", "Cuban swing", etc.

The operators of the large United Fruit Co. were especially noted for this.  Some have claimed that swings developed as a most effective way of copying the early day spark signals (which sounded so much like static) through heavy static.  The basic principle of "sea-going swing" was to exaggerate the spacing between letters when a letter ending with a dah was followed by one beginning with a dah, and similarly for one ending in a dit when the next began with a dit.  The spacing before and after an  E  within a word was often made a bit longer for clarity.  Exaggerated dah lengths were common also in the attempt to improve readability: e.g., the first dah in C was generally dragged out slightly.

Other individual rhythmic disturbances were common also, such as drawing out the second dah in "Q" (which we often hear on the air today).  In order to avoid confusion in the midst of typically heavy Gulf of Mexico static, sending the call signs of two main shore stations was modified: -- the P of WPA was made with long dahs, while the space between A and X of WAX was exaggerated and the dahs of X were lengthened.  This stopped the confusion.  In later years such swings were found necessary for intelligibility in low frequency marine work when signals were barely audible.  Some said "Banana boat" swing developed from call letters KFUC, the general call for all United Fruit Co. ships.  Others suggested the rolling motion of boats contributed to forming it. The name "Cuban swing" or "Latin swing" came from the way most Cuban and Mexican operators ran their words together.  Sometimes it must have been quite deliberate -- just to try to be individualistic, such as  a jerk in forming  H P C S 4 5 Y Q; a lengthening of one of the dahs a bit in  J, 1, etc.; any "funny" little stroke.  But these things made them hard to copy by other operators.

Early in 1936 the Eastern Air Lines (EAL) communications supervisor decided to develop an EAL swing for its operators. He dreamed up the idea of modifying a "bug" by moving the stationary dot post a half inch forward.  This produced a swing like none ever heard before.  The operators did not like it and soon repositioned the post, but it unconsciously influenced the sending of many of them ever after.  Recently operators in a foreign navy were found almost impossible to understand at first because of a peculiar rhythm taught by their telegraph instructors.

Over the years, peculiarities of this sort have often been observed in other parts of the world as well.  These, too, would have to be called "swings".  Swings.  The earliest comment found so far about swing is from Radio News Dec. 1921 p.565: "The American Radio Operator" (commercial and shipboard):  criticizes "the cultivation of a fancy or eccentric style of sending, believed clever in originality, but causes the receiving operator to make more effort to copy than usual.  He introduces a jerk in his H's, P's, C's, 3's, 4's, 5's, Y's, and Q's and makes one of the dashes of J and 1, etc., a trifle longer than the rest... A tricky swing he makes as an effort to acquire the 'funny' stroke as he goes on.  -- Consider the other operator!"

Previous Chapter Table of Contents Next Chapter

The Art &Skill of Radio-Telegraphy-Second Revised Edition-
©William G. Pierpont N0HFF
This page last updated August 02, 1998

Modifications and compile by Thom LaCosta - K3HRN - December 2004