Time machine HOWTO: How to win the ancient QRP contest

Some time ago digging the web for information on The Radio Club of Hartford, I came across this article in QST magazine (September issue, 1921). In a nutshell, it is about a QRP contest as it should be - "Do More With Less" and then measure how big this "More" is and how much spare "Less" were used. I briefly read this article, something about some kind of reward, some competitions, the names - not very exciting at first sight, but one remarkable thing caught my attention - The Winner, transmitter that won a first prize. This TX has turned my stereotypes upside down (once again): how should it work? where is a rectifier? grounded plate tank? keying the grid-leak?? I have to investigate, or even better - to make my own, based on this article, design and run such a prehistoric radio on the air!
So, is that contest over? I do not think so, not yet, not for me..

---

The Radio Club of Hartford (affiliated) recently conducted a very interesting contest in the building of C.W. sets. Several months ago the rules were drawn up and about a dozen members entered. The idea was to devise a simple inexpensive C.W. set of low power, preferably operating from 110 volt lighting current, to supersede the spark coil in the small stations about town. The sharper wave and greater distance with reduced interference made this very desirable, and the Radio Club of Hartford is to be commended for instituting steps that can well be followed by other cities in the reduction of QRM.

One of the members of the club offered a silver loving cup as a grand prize, and in addition there were five 5-watt power tubes offered by the club. The rules called for the award of the prizes to the men having the highest scores on the following basis:

Overall electrical efficiency	50%
Workmanship	20%
Ingenuity in construction	15%
Economy in cost	15%
100%

The awards were made at the final meeting of the club before closing for the summer, and eight contestants were on hand with their sets. These were of every imaginable description, but mostly following the general idea of a small base bearing a vertical panel carrying the controls, with the apparatus behind.

Considerable ingenuity was displayed in the source of power. Several of the sets used step-down ("toy") transformers on the 110 volt supply, the low potential current thus obtained being used both to light the filament and to operate a spark coil with regular interrupter, the secondary voltage of which was dropped by a shunt condenser and then fed to the plate of the tube. These sets of course would operate from a storage battery equally well.

It was a condition of the contest that the operating wave length should not exceed 200 meters, and thru faulty design only two sets were able to achieve this - those submitted by J. C. Randall, 1ANQ, Dist. Supt., and F. H. Schnell, 1MO, our Traffic Manager, who is incidentally vice president of the club. Both of these sets, however, were able to get down to 180 meters, altho readings were taken at 200. A phantom antenna was used, consisting of a 12-ohm resistance and a mica condenser of .0005 mfd. capacity, in series with a Jewell thermo-couple ammeter.

Diode Spark Killer for Semiautomatic Key

swing hell yes
lots of swing is a pain to copy
but
a bit of swing, is a horse of a different color
add a bit of a chirp and some drift
cans on head
pre war receiver
turn down the lights
sit by the glow of the tubes and dial lamps
  © W8ZNX

I'm not sentimental, I'm just as romantic as you are, so I agreed with Mac W8ZNX that "a bit of a chirp and some drift" is not a drawback, it is an attribute, trait of a telegraph radio station, especially the Glowbug one. Just a bit, not too much, and how big this bit should be, I prefer to choose myself.

Well, I used to use a Straight Key, then learned a BUG, and everything was fine until I used a cathode keying of a rock bound oscillator or a differential keying of my Magic Eye MOPA transmitter. When I tried to manipulate the high impedance grid circuit of the TNT oscillator, I got a pretty nasty instability of note, specifically at the dots generated by semi-automatic key (dot contact spring, you know).

Quick patch and further explanation:

---

---

As I see it, the dot contact spring produce a micro spark, discharging the bypass capacitor sporadically, so that every time the exciter starts to oscillate, it has unpredictable remaining voltage on the tube grid. This effect is negligible on dashes (non bouncing contact), as well as in case of cathode keying (low impedance circuit).
Anyway, additional germanium diode on the BUG contacts solved the problem and I got exactly what I want - "a bit of a chirp and some drift", not too much.

WARNING: Can reduce clicks and improve the signal even if you use a cathode keying or rock bound oscillator.

Experimental QRP transmitter by UB5DT, "Radio" magazine, April 1967

For those of my friends tube radio builders, who asked me about "Soviet novice transmitter", "XUSSR QRP TX" (as it pronounced in telegraph),"something authentic, but not too complex" etc, such a RIG should be a good choice - easy to build and clearly adjustable design, small amount of widespread parts, a lot of potential operational fun (two possible output sockets, changeable configuration, very uncommon CW note, i believe). And, after all - a frumpish bald telephonist on the cover as a genuine sign of Soviet authenticity.

---

Experimental QRP Transmitter by I. Tsapiv (UB5DT):

This simple two-stage telegraph transmitter has been developed for a QRP experiments on the amateur HF bands.

The rock bound exciter on the 6П3С (6L6-GB) vacuum tube utilizes circuit known as the "Three-point" (or Tritet in common notation). The crystal controls the frequency of oscillation - 3510kc in this case. Signal output is taken from a separate tuned tank circuit L1C4 in the plate circuit of exciter; this tank should be tuned to 3510kc or 7020kc.

The second stage on the 6П7С (6BG6) tube acts as an amplifier or frequency multiplier (doubler or tripler) depends on an inductance of the L2 coil and bias voltage on the signal grid of this tube.

If you would like to use this transmitter on 80m band as well as on 40m with a noticeable lowered power, you have to connect an antenna to the "A1" connector and pull the amplifier tube (6BG6) out from the socket. For the general power output on 7020kc and QRP operations on 14040kc and 21060kc that tube has to be back and antenna switched to the "A2" output connector.

Measured Power:

frequency	single stage (6П3С)	both stages
3510kc	8 W	-
7020kc	3 W	18 W
14040kc	-	6...8 W
21060kc	-	1,5...2,5 W

Several interesting QSO's  has been obtained using this transmitter in October 1966. For example, UA0MX and K2AGU on 14Mc, and F9MC on 21Mc with a power level as low as 1.7 Watt.

GKE-100 (ГКЭ-100) Short Wave Generator Tetrode

This out-of-the-way power vacuum tube was widespread, but relatively rarely used by radio amateurs in theirs home-made transmitters, leastwise I could not find any publication of a practical TX circuit or typical telegraph and AM operation ratings in accessible to me post-1950's HAM radio sources. Perhaps it can be explained by the fact that there was a lot of much more usable tubes of the same power, which became available in those years, but such a lack of information puzzles me anyway.
At the same time, among the pirates, GKE-100 tube was very popular, though not considered as the best way to spend a 1500 Volts plate voltage (in comparison with more powerful GK-71 for example).

TNT Uncovered - Flea-Power (QRP) Telegraph Transmitter in the Pilot Radio's "Radio Design" magazine, 1930's

I found this article in the Lloyds Dipsy Dumpster - site, where you can find a lot of rare vintage radio articles, manuals and data sheets.

It is remarkable, that "Pilot Radio & Tube Corporation", well known as a manufacturer of the great Super-Wasp receivers, was a publisher of a Ham-Radio literature, handbooks and magazines, as well as many of early radio manufacturing companies. As for me, I like that sort of advertising much more, than modern spam in my mailbox and if it would be possible, it should be a nice choice to make this QRP transmitter using a native Pilot's parts only.

Anyway, when I've read this article, one interesting fact has drawn my attention - I can recognize the TNT when I see a TNT design, but in the text this transmitter has been described as a TPTG.. Moreover, there is a grid capacitor on the assembles view, but with a opposite description below: "condenser is not needed". So, in the article we can see that mysterious moment, when as the chrysalis becomes the butterfly, TPTG transmitter was transformed into a trendy TNT.