Monday, 31 October 2016

Toomas Karmo (= VA3KMZ) (Part B): Remarks on Radiotelegraphy

Anticlockwise from top right corner: Operations clocks (formerly synchronized with a University of Toronto Network Time Protocol (NTP) server, but as of recent days synchronized a little more robustly, with the NTP server (as usual in my setup, I make civil time (here EDT) green, UTC red); system "thermometers"; real thermometers, showing the temperatures of my two cores (I think I applied the thermal grease less effectively under Core 0 than under Core 1, when getting the CPU onto the motherboard in 2013 January); the MFJ-416, with squared-paper practice sheet, and additionally with headphones barely visible (the screen is showing, in its simulation-of-conversations mode, GM OM TNX FER B, as part of GM OM TNX FER BUZZ - in other words, in the delightful lingo which ham radiotelegraphy inherits from the days of the Charleston, the Tin Lizzie, and the coonskin coat, "Good morning, Old Man, thanks for the buzz" - "buzz" here meaning "radio call"); my 1960-vintage Heathkit Q-multiplier, for separating closely adjacent stations at  the receiver, and my modern MFJ antenna analyzer, for detecting feedline-rig impedance mismatches; more system-monitoring stuff, showing how in the spirit of pope Francis's encyclical "Laudato Si" I run just 2 GB of RAM (well, I'd like to run more, but buying it is worrisome, installing it tedious); a cunningly selected excerpt from my study notes, in a Debian GNU/Linux xterm; a specially moving picture of a radio room at prayer, on the USA "Night of Nights" in which old shore-to-ship stations are briefly brought back into operation (from, gallery of images from the year 2014); my Heathkit 1960-vintage Cheyenne transmitter and Comanche receiver (not currently operational), with a landline telegraph key purchased rather cheaply at some recent Ontario hamfest; my 1980s fully transistorized  Kenwood transceiver (not currently operational), with hand microphone plugged in and no telegraph key plugged in.  Significant equipment not shown here: my rather low-quality currently operational 1980 Realistic DX-300 general comms receiver, and the (execrable, yet beloved) currently operational 1965-vintage all-valves general comms receiver handed down to me by Dad. 

Quality assessment: 

On the 5-point scale current in Estonia, and surely in nearby nations, and familiar to observers of the academic arrangements of the late, unlamented, Union of Soviet Socialist Republics (applying the easy and lax standards Kmo deploys in his grubby imaginary "Aleksandr Stepanovitsh Popovi nimeline sangarliku raadio instituut" (the "Alexandr Stepanovitch Popov Institute of Heroic Radio") and his grubby imaginary "Nikolai Ivanovitsh Lobatshevski nimeline sotsalitsliku matemaatika instituut" (the "Nicolai Ivanovich Lobachevsky Institute of Socialist Mathematics") - where, on the lax and easy grading philosophy of the twin Institutes, 1/5 is "epic fail", 2/5 is "failure not so disastrous as to be epic", 3'5 is "mediocre pass", 4.5 is "good", and 5/5 is "excellent"): 5/5. Justification: Kmo had time to make the necessary points to adequate, or even gratifying,  length. 

Revision history:

  • 20161101T1739Z/version 2.3.0: Kmo made a handful of adjustments at the border between the purely cosmetic and the mildly substantive, including an adjustment correctly documenting the "Night of Nights" radio-room-at-prayer image.. He reserved the right to make minor, nonsubstantive, purely cosmetic teaks over the coming 48 hours, as here-undoumented uploads 2.3.1, 2.3.2, 2.3.3., ... . 
  • 20161101T0419Z/version 2.2.0: Kmo not only made cosmetic tweaks but corrected three or four or so technical errors (for instance, now correctly stating the number of lines in the MFJ-418 LCD display). He reserved the right to make minor, nonsubstantive, purely cosmetic tweaks over the coming 48 hours, as here-undocumented uploads 2.2.1, 2.2.2, 2.2.3, ... . 
  • 20161101T0325z/version 2.1.0: Kmo added a photo. He reserved the right to make minor, nonsubstantive, purely cosmetic tweaks over the coming 48 hours, as here-undocumented uploads 2.1.1, 2.1.2, 2.1.3, ... . 
  • 20161101T0211Z/version 2.0.0: Kmo uploaded better version, replacing outline with polished prose (and adding a few remarks over and above what he had in the outline). He hoped in the next hour to add a top-of-posting photo, and then to set to work on minor, nonsubsantive, purely cosmetic polishing. -  
  • 20161101t0001Z/version 1.0.0; Kmo uploaded initial version. He had to leave most of the work in mere outline form, under time pressure. He hoped to convert the outline into reasonably finished prose over the next 2 hours. 

  • [CAUTION: A bug in the blogger software has in some past weeks shown a propensity to insert inappropriate whitespace at some late points in some of my posted essays. If a screen seems to end in empty space, keep scrolling down. The end of the posting is not reached until the usual blogger "Posted by Toomas (Tom) Karmo at" appears.]

    As I wrote last week in "Part A" of this two-part essay, I have been fortunate to receive, pretty much out of the blue, an e-mail from an Ontario ham-radio student of radiotelegraphy. 

    My correspondent wrote on or around 2016-10-16, reporting that he was now studying for the Industry Canada "Basic" ham-radio examination. (Upon passing this, the government gives him a callsign and a licence. If he achieves a high grade, his licence carries full operating privileges. If his grade is of a passing level without being specially high, his licence and callsign bar him from some of the lower - and in intercontinental work preferred - frequencies, until he is  again examined, now possibly in Morse.) 

    My correspondent additionally reported having assembled a low-power ("QRP") "Pixie" radiotelegraphy transceiver, equipped with a crystal for the 40-metre ham band frequency of 7.023 MHz. And my correspondent expressed the intention of building an additional, more flexible, rig. 

    The news of his Pixie took me by surprise. What is a Pixie? Remedying my ignorance today, via Google, I learn the following:
    • The total financial outlay is low, even given that the Pixie sits in the undemanding QRP class - possibly a mere 20 USD, admittedly excluding (surely) the antenna, the headphones, and the batteries or other DC power supply. 
    • Within this tiny rig sit just two transistors. 
    • RF power outputs of between 100 mW and 500 mW are feasible. (So one must, with a reasonable antenna, be at least in principle able to work at least some contacts, say from Toronto down into the USA.) 
    • One unfortunately is liable to hear about 200 kHz of band simultaneously. 
    All the just-cited points, save the last, are positive. The last of the points - the negative one - I glean from a a remark of USA ham, and reviewer, NG9D at 200 kHz is bad. One thereby picks up not only stations at or near 7.023 MHz, but also stations across pretty much the whole of 40-metre ham radiotelegraphy - generating cacophony in the headphones if, as is typical, the band has several loudly audible stations all transmitting at once.  

    Here, as always in life, I obsess over things that could go wrong. One cringes over that 200 kHz. I recall how in the old 2012-era University of Toronto Hart House Amateur Radio Club shack I would filter aggressively, finding it advantageous to restrict my receiving window-into-band to a width of 100 Hz or so - in other words, to just one-two-thousandth of the Pixie's alarmingly wide 200 kHz. In general, we do get what we pay for - in my correspondent's case, a 20 USD rig; and at the old Hart House club, at the other end of the scale, an ICOM IC-756PROII  costing roughly 100 times more. 

    While applauding my correspondent's initiative and pluck, I do therefore find myself echoing some sentiments of USA ham K3AN. K3AN appraised the Pixie on 2009-05-28, at I would caution against recommending such pursuits to the newcomers to our ranks, despite the attractively low cost of entry. Let them start with a 100-Watt multiband rig, where they can easily make contacts, including DX contacts. As they become more immersed in Amateur Radio and begin looking for new worlds to conquer, then kit building and QRP are just two of the many avenues open to them.

    It may well be that the investment in a Pixie will ultimately prove worthwhile. The Pixie might eventually prove instructive, at a somewhat advanced level of study, as a tutorial in rig design, for instance in oscillator concepts. 

    I now add for my correspondent's benefit that the study of Morse is concisely discussed in Dave Finley's book Morse Code: Breaking the Barrier. The book is available in the Toronto Reference Library through a "Stack Request" for call number 621.383 F38. I read the book from cover to cover, perhaps once and perhaps (for all I can recall) more than once, rather early in my studies. 

    In learning Morse, I, like the just-cited book author, have found the MFJ-418 "Pocket Morse Tutor" helpful, despite its stiff cost (as advertised today at, 89.95 USD). There are, to be sure, computer programs which teach Morse. However, the MFJ-418 encourages copying with pen and paper, rather than with keyboard. This machine thus sticks more closely than computers do to the spirit of radiotelegraphy, in which one strives for long-distance communication through an elegant deployment - a deployment in which the hardware complexity, and the concomitant problems of hardware repair, get minimized. 

    Admittedly, there are limits to how low we want to go on hardware. 

    I would suggest that the "sweet spot" is one on which we avoid today's silly "memories" and other user-friendly (i.e., down-dumbing) front-panel elaborations; and in which we avoid surface-mount components and most integrated-circuit chips, thereby keeping our rig repairable in even the small workshop; and in which we achieve rock-solid frequency stability (if necessary, even putting the pertinent crystal into a little temperature-regulated oven, as was done with a celebrated top-of-the-line Morse-appropriate rig in the 1960s); and in which our front panel displays operating frequencies digitally (well, here some integrated-circuit chips are handy), to the fanatically tight precision of plus-minus 1 Hz; and in which we have a plain and rugged spectrum scope, perhaps even in a box separate from the rig; and in which we can select our automatic gain control not just to "FAST" and "SLOW" but even to "OFF"; and in which we can do lots of filtering, for instance by imposing a notch; and in which we can adjust our beat-frequency oscillator manually on both sides of zero, without that silly user-friendly, dumbed-down, selecting from "USB" and "LSB" and "AM" and "CW"; and in which we have an elaborate antenna-tuning matchbox, separate from the rig itself. 

    The "sweet spot" is not in all respects set by archaic technologies. I thus note not with approbation but with a shudder the conditions under which another friend-or-relative, "CDEF", worked, a generation before PQRS. CDEF retailed home receivers in Tartu, in southern Estonia, as "RET Raadio", on one or more occasions himself building a high-grade receiver. I believe the last few of CDEF's drawers of 1930s components, including even two or three or six or more valves, are still in the attic of CDEF's still-living daughter. At any rate I saw them there, with CDEF's young civil-engineer great-great-grandson, in the summer of 2010. In the 1918-1920 Estonian War of Independence, CDEF operated in the military, with spark. 

    They used spark on the Titanic. It is basically a transmission of key-modulated static, hogging (as a heavily  varying carrier, far from a single sine wave, mathematically must) a wide frequency range. Spark was obsolete by the 1920s, with oscillators now generating clean single-sinusoidal wavetrains. We may piously hope that no matter how deep our future Dark Ages may prove, some centuries from now, spark is forever deemed obsolete. 

    The MFJ-418 at any rate hits a "sweet spot" in the biggish world of Morse training aids, deploying just the right amount of digital computing hardware, and providing just the right amouing of display on its little (two-line) LCD screen. 

    It is particularly pleasant to operate the MFJ-418 with headphones, copying onto square-ruled paper in one colour of ink, and then correcting one's work in a different colour. This is one of the various points I have illustrated in the photo at the start of today's blog post. 

    On using the MFJ-418 over the years, I have found it advisable to keep a special log in my Debian GNU/Linux box, showing cumulative time-invested and current level of proficiency, and also showing occasional applications of my MFJ-418 home-desk work within the more intimidating environment of a shack. Here is an excerpt. I perhaps need not comment on my various cryptic remarks regarding level of proficiency, stations worked, and the like, except to remark that "skccgroup" and "morseresource" are tutorial Web sites, respectively and (useful supplements, as opposed to replacements, for the specially and distinctively useful MFJ-418); that "QSO" is ham jargon for "conversation"; that I have for a year or two been drilling at 22 wpm; that "DIES PERDITA" and "HEBDOMA PERDITA" are Latin for "lost day" and "lost week", respectively; that the "vigu" in my more recent log entries is Estonian for "number of errors"; and that the (phonetically disyllabic) "teene" is Estonian for "number indicating merit". Although I have a little mental-arithmetic routine for computing this number, I need not here weary my readers with details on the routine. For present purposes, it suffices to say that "merit" declines not only when I make a mistake in copying but also when I have to listen to an MFJ-418 multi-"word" string of Morse code twice, in my ongoing struggle to avoid mistakes in copying: 

    20100712=00h33->0000h33__callsigns@intr04=sets20&errs02 hurraa
                             __increase in self-confidence,
                               partly as consequence of visiting
                               hamfest in Milton 2010-07-10-a
                             __for the first time did a set
                               in just one row (no need to repeat)
    20100713=00h36->0001h09__callsigns@intr04=sets19&errs00 hurraa
                             __12 sets are 2-row and error-free
                             __00 sets are 1-row and error-free
    20101004=00h49->0031h06__sheet of lttrs+nrmls@intr05=sets22&errs01
                             __14 sets are 2-row and error-free
                             __01 set  is  1-row and error-free
                             __so fig-of-merit = 14+2 = 16
                           __half-sheet of  words@intr05
    20101005=00h45->0031h51__half-sheet of lttrs+nrmls@intr05
                           __sheet of words@intr05 w/11 impeccable lines
    20101007=01h03->0032h54__sheet words@intr05 w/16 impeccable lines
    20101008=00h36->0033h30__sheet words@intr05 w/12 impeccable lines
    20101009=00h51->0034h21__sheet words@intr05 w/11 impeccable lines
                           __sheet of lttrs+nrmls@intr05=sets16&errs00
                             __07 sets are 2-row and error-free
                             __01 set  is  1-row and error-free
                             __so fig-of-merit = 7+2 = 9
    20101010=00h51->0035h12__listened in shack
                             (_for first time got essentially full
                               substance of a long slab of a QSO,
                               with operators in New York State
                               and New Jersey, sending slowly
    20110225=00h28->00100h53__sheet of FCC chars
                               __07 sets are 2-row and error-free
                               __00 sets are 1-row and error-free
    20110227=01h00->000101h53__listened in shack rather successfully
                               (_copied fairly confidently from
                                 both sides of QSO between AK and WV
                                 (_perhaps my best shack session so far)
                                __am also now more confident with
                                  bandpass filtering for cw,
                                  having on this day reviewed IC-756-II
                                  manual for double bandpass tuning etc
    20110228=00h11->000102h04__apprx 35% (unmeasured) of sheet of FCC chars
    20110301=00h28->000102h32__sheet of FCC chars
                               __06 sets are 2-row and error-free
    20110402=00h36->0000117h34__copied skccgroup gettysburg_address_15.mp3
                                __approx 2 hopelessly missed words,
                                  approx 4 words-good-enough-to-guess,
                                  and various single-character errors
                              __copied from morseresource quotes
    20110605=01h23->0000141h50__had first adequately successful QSO,
                                w/K3WWP in Pennsylvania LAVS DEO
    20110904=01h15->00000160h58__listened in HHARC shack
                                 __had CW QSOs w/Italy, Balearics, Quito
                                   __LAVS DEO

    20111231=00h30->0000208h12__FCC faults0, merit41

                                __so 2q=87.5%
                              __fastwords 5setx10rows=82%
    20120101=01h30->0000209h42__listened in HHARC shack
                              __CW QSO 00h10 w/WA6RZ HURRRAAA
                              (_total radio handson this day = 03h28)
    20120109=00h21->0000213h38__FCC faults1, merit42-2=40d
    20120110=00h22->0000214h00__FCC faults2, merit48-4=44d
    20120111=00h17->0000214h17__I now begin drilling FCC chars at 20 wpm
                                rather than at 18 wpm
                                (_geopol news (Iran) impels me to ramp up)
                              __FCC faults0, merit37[sic 37]
    20120112=00h29->0000214h46__FCC faults1, merit36-2=34d
                              __I now begin drilling mock QSOs
                                at 20 wpm rather than at 18 wpm
                              __mock QSO
                              __I now begin drilling fastwords
                                (_this is a drill in which I do not
                                  write down what I hear)
                                at 20 wpm rather than at 18 wpm
                              __fastwords 5setx10rows=80%[sic 80%]
                              (_separate from this time accounting
                                is my ((SNIP)) viewing of
                                Morse training vids:
                                  __ca 1944
                                  __showing coin-wrist,
                                    emphasizing that wrist and index finger
                                    are what do the work,
                                    emphasizing that wrist is to be relaxed
                                  __ca 1966
    20120408=01h27->0000246h47__listened in HHARC shack
                                __short Ukraine CW QSO HURRRRAAAAA!!!!!!!
                                  __UR7HA = {Yuri.N.Golyanik}
    20130826=00h20->000344h49__FCC faults1, merit54-2=52d (Zpro7) HUZZZZAAAA
    20130827=00h22->000345h11__FCC faults1, merit45-2=43d (1proJ)
    20130828=00h20->000345h31__FCC faults0, merit38[sic38]
    20150720=00h25->000366h09__FCC faults0, merit48[sic48]
    20150721=00h16->000366h25__FCC faults0, merit36[sic36]
    20150722 DIES PERDITA
    20150723=00h20->000366h45__FCC faults4, merit36-8=28d
    20150724=00h22->000367h07__FCC faults0, merit38[sic38]
    20150725 DIES PERDITA
    POLICY: From now on I seek to do something every week,
            but NOT necessarily every day. I thus log lost weeks,
            in my "HEBDOMA PERDITA" formalism,
            without logging lost days.
    20160725=00h08->00379h41__FCC vigu0, teene05[sic05]
    20161001=00h16->00382h00__FCC vigu1, teene23-2=21d (S pro H)
    20161008=00h14->00382h14__FCC vigu0, teene28[sic28]
    20161015=00h15->00382h29__FCC vigu0, teene23[sic23]
    20161022=00h20->00382h49__FCC vigu0, teene33[sic33]


    Everyone learning Morse has for decades faced, at the very outset, the question, "Koch Method or Farnsworth Method"? Details can be had by Googling on the string koch method versus farnsworth method. Here it suffices to say that both methods strive to avoid a known problem for Morse students, the "Speed Plateau". 

    There is some temptation to learn Morse naively, progressing from mastery of the whole FCC set of about 44 characters at, say, a slow 6 words per minute up to medium-speed 12 or 14 wpm, up to the rather fast 18 wpm, and up ultimately to the duly professional 24 wpm. I suspect my friend-or-relative PQRS, whom I discussed last week, proceeded in this way during the Hitler war, in his ultimately successful progression to the Wehrmacht's daunting 24-wpm target. Although the Koch method was current in Reich radio-training facilities, PQRS probably had no access to such luxuries.  But the wartime success of PQRS notwithstanding, it is said that some people get stuck at perhaps 10 wpm or so, finding for a long time that they just cannot copy even at the moderately useful 14 wpm. 

    To avoid the bottleneck, the Koch method drills a a sufficiently stiff speed, say 12 wpm, from the beginning. In the first day or two or four or so, one starts with just two characters, for instance K ("dah-di-dah") and M "(dah-dah"), generated by one's teaching machine in such strings as K MK MM KKM KMMMK MMKMKMKMKKK (supposing the machine to be set, as is reasonable, for six random-length "words" per string). In the next days, one adds a third character, say R, so that one's machine is now briskly playing into the headphones such strings as KMR RRMK RKR MMMMRKM RKKKKKM MMKR. Some days later, one adds a fourth character, say S, so that the strings become brisk-sounding randomized sequences like MKSR RSKM MRSSSK SSKRM MSK KSRS

    One then begins to learn - or I, at any rate, began to learn - distressing things about the human brain. 

    Why is it that when one has mastered, say, 20 of the 44 or so FCC-prescribed characters appropriate in English-language radiotelegraphy, the addition of a 21st character proves so dramatically painful? Why is it that suddenly, with the introduction of just one new character to the growing set, the whole brain is thrown into a dramatic confusion, from which it takes several days to recover? These are days which must be endured patiently, before the pen once again writes down accurately, on that suddenly discouraging square-ruled sheet, what the machine has been showing on its (temporarily concealed) two-line LCD display and sounding into the headphones. It is as though the brain is filing recipe cards, under some perverse rules which cause its entire accumulated file to be scrambled as soon as just one additional card gets dumped into the growing set.

    Grimly fearing an outright learning disability (not for nothing have I been diagnosed with a mild form of autism), I stopped, then resumed, then stopped, then resumed, and so on, over a period of years, dragging out what I should have achieved in under one year. I read, or first read, the Finley book on 2004-11-29. I first set up the MFJ-418 on 2004-12-14, drilling on random perhaps-13-wpm "words" made up from the two-character set B W. Finally (as can be inferred from the work log displayed above), I succeeded, conducting my first really adequate conversation, or "QSO", with K3WWP in rather-nearby Pennsylvania on 2011-06-05.

    I do not think I have found any learning experience - certainly not classical Greek, perhaps not even multivariate real analysis from Spivak's universally dreaded Calculus on Manifolds - in this particular way disheartening, to the particular point of inducing an outright suspicion of neuropathology.

    Perhaps Farnsworth, then, is the better bet.

    In the case of Farnsworth, the student learns most or all of the FCC character set at the outset, hearing the characters always at a reasonably brisk dits-and-dahs tempo, but at first with long pauses between characters. As it might be:


    (Here I write dots not for dits but for stretches of silence.) As the months go buy, the teaching machine is progressively directed to tighten up the spacing, to a point at which the student perhaps hears


    finally graduating to the nicely professional


    On both the Koch and the Farnsworth approaches, the student is (correctly) forced to hear characters as units, like drumming routines, rather than as discrete dits and dahs. C, for instance, is both under Koch and under Farnsworth not a laborious and slow


    but a nice, crisp, drummer-in-the-orchestra-pit


    One suspects that many students of Morse have private mnemonics for hearing characters as units, i.e., as tight drum routines. The Web literature somewhere suggests hearing D, or "dah-di-dit", as "dog did it". In this spirit I for my part make Z a personal tribute to actress Zsa Zsa Gábor (1917-), as "dah-dah-dit-dit", for "Zsa Zsa did it." In this same spirit, I make P into "The Pope saw it" ("di-dah-dah-dit"). The sound had better (say I) enjoy some mnemonic connection with "love" (or, as Elvis Presley would say, with "luv"). And so I turn that "di-dah-di-dit" into "Get OVER it."


    Having learned Morse, one then has to consider how to retain the knowledge, in a busy life in which hands-on radio must generally get overshadowed by more pressing concerns. In my particular case, for instance, there are pressing concerns, within radio and yet outside the practical shack, in mathematical-physics-of-radio. Outside radio, there are also pressing duties, notably in astronomy.

    Perhaps practical skill in Morse, like any complex reflex, stays in the brain indefinitely. It is not, however, prudent to bet on this - especially in a world that might descend in a scant 24-month or 12-month spiral into outright civil chaos, with ham-radio expertise suddenly a mattter of public safety (even as it suddenly was in, say, the Britain of 1939).

    As can be seen from the last few of the log excerpts above, I for my part now drill just once a week, for about a quarter hour, doing just the most important of the various possible kinds of drill - the sequences not of intelligible words but of mere "words", assembled by my trusty MFJ-418, to random length, from random FCC characters.

    Two such "words" are thus




    The machine seems in my case to vacillate in a truly random, gratifying, way between "words" one character long and "words" running through various intermediate lengths, right up to 8 or so characters.

    I find that under this undemanding schedule, my proficiency level at 22 wpm drops to below what it used to be when I drilled at 22 wpm six times a week, and yet does stabilize within the range of the acceptable. In the case of a large civic emergency, such as a global Internet collapse, it would on the strength of this retained skill be possible to enter a radiotelegraphy shack at once, ready to copy all but the faster stations, and with the brain poised to regain its modest stretch of lost ground in just a couple of weeks - I presume to the relief of, for instance, government ministries seeking radio volunteers for their rapidly forming civil-defence units.


    My correspondent adds that he would like to have "long, slow" conversations with me eventually. To this I have replied, in e-mail, in a cheerful affirmative, while noting that I have yet to bring my own private collection of ham gear into an operational state. (I have a 1960-vintage Heathkit "Cheyenne" and "Comanche" transmitter-receiver pair, with separate single-valve Q-multiplier appropriate for the Comanche receiver. I also have a 1980s all-solid-state Kenwood TS-440S transceiver. The Kenwood was an outright donation, perhaps now 13 or so months ago, from an extraordinarily kind Toronto ham whom I met through half-day annual radio-support operations at the Canadian National Exhibition. Rounding out my little navy, in which the Kenwood is of course the flagship,  are (a) my workhorse low-quality general-comms receiver from 1980, a functioning Realistic DX-300; and (b) the truly wretched, yet beloved, valve circa-1965 Hallicrafters S-120 handed down to me by Dad. It was on the Realistic DX-300, poor though it is, that I followed the Falklands news in 1982, when living in Melbourne (reception of a BBC relay proved easy even in the daytime). It was on the S-120 that my Dad and I monitored Radio Prague from Nova Scotia, in the tragic August of 1968. Finally, as so-to-speak supply vessels in the fleet, there is the MJF-418, plus a modern MFJ antenna-and-feedline analyzer. - This fleet I am resolved to handle with maximal care, eventually bequeathing it via last-will-and-testament to some appropriate institution or individual, within the general Estonian radio world.)

    If others reading my blog can help my correspondent, for instance by entering into their own long, slow Morse conversations with him, I would urge them to get in touch with the two of us. Such readers could start by e-mailing me as Toomas(dot)Karmo(at)gmail(dot)com, citing today's blog posting, and asking me to effect an e-mail introduction.

    [This concludes this (two-installment) essay.] 
    No comments:

    Monday, 24 October 2016

    Toomas Karmo (= VA3KMZ) (Part A): Remarks on Radiotelegraphy

    Quality assessment: 

    On the 5-point scale current in Estonia, and surely in nearby nations, and familiar to observers of the academic arrangements of the late, unlamented, Union of Soviet Socialist Republics (applying the easy and lax standards Kmo deploys in his grubby imaginary "Aleksandr Stepanovitsh Popovi nimeline sangarliku raadio instituut" (the "Alexandr Stepanovitch Popov Institute of Heroic Radio") and his grubby imaginary "Nikolai Ivanovitsh Lobatshevski nimeline sotsalitsliku matemaatika instituut" (the "Nicolai Ivanovich Lobachevsky Institute of Socialist Mathematics") - where, on the lax and easy grading philosophy of the twin Institutes, 1/5 is "epic fail", 2/5 is "failure not so disastrous as to be epic", 3'5 is "mediocre pass", 4.5 is "good", and 5/5 is "excellent"): 4/5. Justification: Kmo had time to make the necessary points to adequate length. 

    Revision history:

    • 20161025T0159Z/version 2.1.0: Kmo added a paragraph or so on the Ark Two PSK31 effort.  - He reserved the right to make further, nonsubstantive, merely cosmetic, changes over the coming 48 hours, as here-undocumented versions "2.1.1", '2.1.2", "2.1.3", ... . 
    • 20161025T0142Z/version 2.0.0: Kmo uploaded improved version, the point-form outline now converted to reasonably finished prose. 
    • 20161025T0002Z/version 1.0.0: Kmo uploaded initial  version. He had to leave most of the work in mere outline form, under time pressure. He hoped to convert the outline to reasonably finished prose over the next 2 hours.

    [CAUTION: A bug in the blogger software has in some past weeks shown a propensity to insert inappropriate whitespace at some late points in some of my posted essays. If a screen seems to end in empty space, keep scrolling down. The end of the posting is not reached until the usual blogger "Posted by Toomas (Tom) Karmo at" appears.]

    I have had two distinctively welcome e-mails this October.

    One was from a member of Ontario's private "Ark Two" disaster-preparation team, in response to a minor e-mail I had for my part sent to Ark Two, after perhaps two or three years of neglected communications. 

    This is the Ark Two which is promoted at, and which is further explained at It is an interesting venture, some would say on the fringes of respectability - a grouping of survivalists, or "preppers", fearing thermonuclear holocaust, and unlike some inappropriately selfish "preppers" resolved in the face of this fear to do whatever good they can for the wider community. 

    My other distinctively welcome e-mail came pretty much out of the blue, from an Ontario individual who had noted my comments on John Michael Greer's blog at and had decided to write me. This correspondent made some specific reference to radiotelegraphy. 


    As a physical installation, Ark Two is beyond frightening. Here we have, just a few tens of kilometres to the northwest of the Toronto conurbation, the Sovyetsky Soyuz. There is that quiet access road, those high gates - inducing perhaps that sickening feeling that we have been here before, in some previous life, under Stalin. There is that heavy bunker door. The door once opened, there is that gently sloping entrance tunnel, and then that vast sequence of rooms. One of them is a radio broadcasting studio, empty of equipment when I saw it. 

    I recall the clammy atmosphere in the well-buried erstwhile schoolbuses which are those endless rooms: the subterranean cool, and perhaps even dripping, Ontario April, while up above, beyond those low curving ceilings, where the birds still sing, there blazes an Ontario June. 

    I sometimes picture the end of civilization, should its end prove sudden and thermonuclear. One stands in Richmond Hill near Major Mackenzie Drive, looking south, toward Toronto. Above the southern horizon rises the mushroom stalk, dark as any peacetime thunderhead. Above that - easy, in fact, to pick out from Richmond Hill, twenty kilometres north of Ground Zero - is the equally dark mushroom cap. From this cap a black rain is already falling onto what must at grade level, in the Toronto Bay-and-Front-Streets core, be a firestorm. One knows that in a few minutes, Richmond Hill radiation levels will start their inexorable rise, inducing first the ordinary vomiting, then the other symptoms. One is happy enough at this early point to be facing probable death in Richmond Hill, as opposed to possible survival in Ark Two.


    My specific attempts to help Ark Two have involved PSK31, the contemporary ham-radio replacement for traditional late-1940s-onward ham radioteletype. The Ark Two team has been trying to develop a maximally simple PSK31 rig, for use by minimally trained personnel in post-holocaust conditions. One bottleneck has been their (problematic) effort to develop a modest RF amplifier, to boost their transmitter output. Had I the time and knowledge, I would look with a favourable eye on the effort, while also promoting a separate, parallel, effort to tune the antenna. A PSK31 transmitter in a national emergency is liable to be run into something quite crude, like an improvised dipole, or even an improvised random-wire aerial. The situation parallels what the western Allies encountered in the Hitler war, dispatching low-power suitcase-sized Morse transmitters with their undercover agents into occupied Europe, and confronting the reality that agents would have to put up inprovised antennas (even indoor antennas). For the Allies, it was important to get a match between transmitter output impedance and impedance of antenna-with-feedline. Their solution (particulars are on YouTube, in some 1940s spy-radio discussion) was simple, and I think Ark Two might want to ponder it: no elaborate, fragile, standing-wave-ratio meter, to detect undesirable reflection of signal along feedline back to transmitter in the undesirable event of an impedance mismatch; rather, one somehow wires into or near the feed line one or more low-power lamps, whose staying dark or lighting up (I am vague on details here) provides sufficient warning of gross impedance mismatch.

    In general, I confess, I try to avoid entanglements with PSK31, since that particular technology presupposes access to a functioning computer, running a graphical display. We may expect computers to become scarce soon after a nuclear war.

    These days, my thoughts are more on Ark Two than they have been in recent years. Partly (this is why I e-mailed Ark Two lately), I am concerned by the cyber-warfare analysis of the properly reticent, and yet properly informed, Harvard writer Bruce Schneier, at (blogging under the general heading "Schneier on Security", under the 2016-09-13 title "Someone Is Learning How to Take Down the Internet").

    And partly I am alarmed by what I have learned lately from a meticulous unofficial observer, and what I accordingly posted on this blog last week (under the heading "(Part A): J.M.Greer on Popular Geology, Chronocentrism, and Deep Time"). I may as well reproduce last week's scrap of news:

    In the second half of 2016, the principal Russian system for the manufacturing of domestic consent - namely, the Russian-language television, directed at Russian viewers - has (I am told) shifted its tone in a civil-defence direction. There is now (I am told) bizarre official talk of food reserves in St Petersburg. As of the 2016 Russian autumn, it is claimed that there exists a wartime-emergency bread stock, sufficient for the entire city population for an entire 30 days, assuming a drawdown rate of 300 grams per person per day. (The roughly 10 percent or 15 percent of the population which dislikes the government finds this funny, I am told: the current joke, I am told, is that 300 grams per person per day is ever so much more than the 125 grams per person per day available in the 1941-09-08/1944-01-27 Siege of Leningrad.) There is also (I am told), as of this 2016 Russian autumn, strange official domestic-television talk of shelter space in Moscow. The subway, more formally the Московский метрополитен, is being proclaimed to viewers as capable of sheltering every Muscovite in the event of nuclear war.


    This pair of alarming snippets of news-from-experts notwithstanding, I continue to direct my focus away from World War Three. We have lived with the threat of World War Three long enough, and from the 1949-08-29 Semipalatinsk A-bomb test right up to 1991 in a more intense form than at present. (The late August of 1991 seemed grim for a couple of days.) To my own mind - I might be wrong - the more terrible threat is, rather, the prospect of a slow decay in institutions, in a process consuming five or ten generations, and not punctuated by anything as dramatic as all-out nuclear war. Eventually, we get the illiteracy, the slumlords and warlords, everywhere outside the remaining gated enclaves - our electricity on for four hours a day if we are lucky; our sewers out of commission; our water distributed now to mere neighbourhood spigots, for collection with buckets; and our children lucky if they can read slowly, moving their lips, taking guesses at the longer words. We are eventually, that is to say, back in the Rome of the Vandals, Lombards, or Goths, except that this time it is with guns. No doubt occasional drones will mar our skies. No doubt a few all-terrain vehicles, burning expensive ethanol for The Man, will gallop and buck down our unlit, potholed streets.

    We get a possible taste of what is coming, so far as communications infrastructure is concerned, from John Michael Greer's commenter "LatheChuck", writing over server timestamp "10/22/16, 5:35 PM". LatheChuck refers, among other things, to my own ongoing mathematical-physics-of-radio efforts:

    I imagine that my local warband leader would see the value in keeping a guild of radiomen inside the gates. You can't teach radiotelegraphy in a week (or even in a year), especially if your raw material (potential substitutes from the warband) first must be taught reading, writing, and arithmetic! The basics of transistors and transmission lines will be esoteric enough, let alone Toomas's project to put ham radio on a firm physical basis (including Special Relativity). Then there's the mental training needed to make characters out of the dits and daahs (and vice versa). Though ciphers are forbidden by FCC regulations, it's hard to imagine those regulations being enforced in a warband society, so skill with ciphers would also be valuable.

    Talented radio engineers have already been sought out by Mexican drug trafficking organizations. Whether they're retained (detained) for future use, or disposed of after each job, we'll only know if bodies are discovered.

    "According to a report by Animal Politico, an independent Mexico-based investigative media company, about the so-called “slaves of narco,” 36 communications specialists had gone missing in the region between 2008 and 2012."


    Before taking up this week's welcome incoming correspondence on radiotelegraphy, I will expand a little on LatheChuck's remarks about my programme of writing:
    • My hope is to produce, eventually, a series of "white papers", to support people who are training themselves in the mathematical-physics-of-radio. These will be people who may some day have to design equipment. 
    • I hope to handle rigorously some questions often slurred over, for instance "With respect to what zero-voltage reference point do we consider the voltage distribution, at any one instant, along the classical dipole antenna?" (Hmmm....not quite sure how to handle this one at the moment. I did know, I think, some years back!) 
    • As part of this, I hope to treat div, grad, and curl in Maxwell's equations, rigorously - going carefully into (for example) the derivation of the formula for div-in-cylindrical-coordinates (ahhh, must review this, as work from months ago!); or again (a different example) showing how scalar line integrals, vector line integrals (as in "work"), scalar surface integrals, and vector surface integrals (as in "flux"), and indeed the humbler Riemann integrals of univariate calculus, are all instances of one general limit-of-progressively-finer-Riemann-sums idea. 
    • In particular, I hope to amplify some of my present notational ideas, on which the 1930s Alonzo Church mathematical-logic "lambda calculus" is used to clarify operations-on-functions, such as the taking-of-a-div. (This is a notational programme already implemented at MIT, I gather with great success, by Sussman and Wisdom - with some initial details available from 
    • Also as part of this, I hope to put Maxwell's equations into their proper Special-Relativity setting (eventually working out to my own satisfaction, and explaining to others, the tensor formalism in which an electrostatic field in a rest frame of a charged particle becomes a magnetic field in a frame moving relative to that rest frame). 
    • As an underpinning for eventual careful work with div, grad, and curl, I hope to improve my grasp of real multivariate analysis. 
    • And as a preparation for that, I am in turn at present trying to train myself in rudiments of toplogy. (When I write "at present", I have in mind that this is an initiative begun on 2016-09-07, and budgeted at 200 hours desk-time, and ongoing. As of 2016-10-21, my log shows an investment of 69 hours, 5 minutes.)

    My correspondent this past week mentioned radiotelegraphy. 

    LatheChuck's words quoted above, with their sad reference to conditions among today's Mexico druglords, call to mind the experience of my relative or friend "PQRS" in the Hitler war.

    Estonians unwilling to support Stalin in that war had a few options, none of them good. Most or all of the options got duly tried within my own little circle of friends and relatives.

    One could altogether dodge military involvements. - Yes, Virginia, as there is indeed a Santa Claus, so also is this seemingly impossible thing indeed possible. When the big European Soap Opera was at last over, or at least was winding down, a sometime Latin teacher was greeted in Tallinn by an old friend. - "Ah, Ants (or whatever), you have survived everything: how did you do it?" - "Oh, the usual. I hid out in the woods ... but you know, in the woods, korrektsus täitsa puudub, there is absolutely no correctness." (Korrektsus, or "correctness", is a virtue rightly held in high regard, through much of the Nordic and Teutonic world.)

    Or (a second option), one could eventually join the 1944-through-1950s (or later) anti-Soviet guerillas, taking to the woods in a manner not fully pacifist, with one's prewar concern for korrektsus I presume largely abandoned.

    Or (a third option) one could try to join the British forces. But as it turned out, the Chamberlain government, at any rate as represented via embassy or consular authorities to Estonians in Switzerland, was not in the 1939 autumn set up to take in such volunteers. What happened after 1939, I do not know.

    As a fourth option, one could oppose Stalin under the Finnish flag, in Finnish uniform. This was a comparatively civilized alternative. One thereby did not do much for Estonia until the late summer of 1944, when Estonia's "Finnish boys" got repatriated. However, for a couple of years prior to that I guess one did help Finland survive, the breach in Finland's Mannerheim Line notwithstanding.

    Finally, one could oppose the Red Army under the German flag, wearing a German uniform with some kind  of dinky little Estonian badge or badges. This was of all the options the most appalling. (Unless outright suicide be added to the list; I know of no suicides.) Perhaps, however, the German-led Narva-front resistance from the early spring of 1944 until the disastrous September was one of the factors saving Finland from a summertime seaborne invasion, and so helping underpin Finland's survival. At any rate I propose this, as I believe some other writer or writers in essence have, by way of a topic for the historians of tactics.

    Additionally, this German-flag, German-uniform resistance, particularly under Narva, made possible, by for two seasons stalling the Red Army advance, the safe September exodus of something like 70,000 Estonians. The unpleasant form of resistance thereby facilitated the survival in exile of an Estonian government - it left office on 1992-01-29, acknowledging the restoration of independence by a home government - and of various other institutions-in-exile.

    My friend or relative "PQRS" chose that final, unpleasant option. (Well, he was under conscription, so perhaps talk of "choosing" is in his case not fully apt.) He was able to take the option (well, quasi-option) in an interestingly specialized form.

    PQRS had always been interested in technical matters. As a quite young child, he had risked life and limb by attempting an aeroplane flight, driving his little wheeled waggon down a sloping roof. (His Dad's pocket watch had served as an instrument, on his small instrument panel.) He had also, I suspect at a rather early stage, upset his parents by creating a hole in the floor, so that he could get a good earth (a good ground) for his crystal receiver. He had made an arc lamp as part of a projection microscope, energizing his arc from the (direct-current, I suspect 240 V) power supply of his little town, for the edification of schoolmates. And working from instructions in a peacetime radio-enthusiast magazine, he had built himself a ham-class transmitter.

    It was, to be sure, unfortunate that his peacetime transmitter got finished rather late, in 1939. This vagary in timing made its use not fully advisable. Estonia was forced that autumn to admit a few tens of thousands of Red Army personnel, under the duress of a diplomatic ultimatum. Just a few months later, Estonia voted to request admission to the USSR as a constituent Republic, and after a few days' deliberation Stalin granted what was requested. (When I write here  "voted" and "deliberation", I mean of course "kinda-sorta", as in that nice Mel Brooks 1980s rap parody of the Reich. As I remember it from YouTube, it  ran something like this: "So we had an election/ Well, kinda-sorta/ Say hello/ To the New Order.")

    Now, in 1943 or so, with Stalin's occupation of Estonia temporarily thrust aside by Hitler, and with the increasingly beleaguered Hitler conscripting Estonians, it was necessary for PQRS to take decisions. Conventional military service, killing people? Not. Mindful of his radio background, PQRS trained in Morse, to the high and severe standards of the Wehrmacht.

    PQRS subsequently told me that the requirement was 24 words per minute. This is about as fast as you can feasibly go with a traditional "straight key" on the transmitting side. On the receiving side, it is hard to copy with pencil and paper at any speed much faster than this, at least if you want to continue being accurate and legible, as you had better be when traffic is encrypted.

    PQRS passed his Wehrmacht radio Morse Code exam, I imagine to his family's relief. From this point onward, PQRS's career became interesting. He was directed to transcribe Red Army Morse transmissions in the field, walking his papers over to a decryption room or decryption tent barred to mere Estonians.

    This Red Army radio-monitoring work was not even very hard, at least as PQRS explained it to me around 2002. For the Red Army (I think confronted with a need to train lots of operators quickly) confined its transmissions to strings of digits, using no letters at all!

    Since decryption got done right in the field - admittedly, behind doors or tent flaps closed to PQRS, since PQRS was unfortunately not of the "Master Race" - this side, too, of operations was perhaps not daunting.

    At any rate, PQRS got through the entire Soap Opera, I suspect without ever having to bash anyone up.


    I should perhaps add in parentheses my dear Mum's reminiscences of her own papermail efforts with PQRS.

    It was for certain reasons advisable for Mum and her own Mum (my dear Grandma) to find PQRS. But this was in the chaos of late 1944 or 1945, when Mum and Grandma were in flight - first to the north of Germany, then to Denmark. Mum, as she explained it to me late in her life, exercised both of the two obvious options. One was a formal missing-persons query via the Red Cross. And - she explained to me in the kitchen, in 2000 or so - the Red Cross are "still looking". The other was the Reich military postal service, the Feldpost. The Feldpost delivered Mum's letter with their customary efficiency, evidently not being thrown off their routine by the various inconveniences attending a military collapse.


    Radio, then, is of importance when things break down, be it in contemporary Mexico or in the Hitler war.

    Radiotelegraphy has a place of special honour in radio, involving as it does the most basic of equipment, with the most basic of designs and the smallest number of components.  It was therefore good for me to learn of my correspondent's work in radiotelegraphy.

    [To be continued, and probably concluded, as "Part B", probably next week, probably in an upload in the four-hour UTC interval 20161101T0001Z/20161101T0401Z.]

    Newer Posts Older Posts Home