Stability, Accuracy, Purity – The “1929-Type” Station

Hiram Percy Maxim long considered precise frequency control to be the biggest challenge in amateur radio.1 All at once it was set to become everyone’s concern when the new international treaty would take effect in 1929. “As we climb up into the super frequencies, as we do when we use forty meters and below, frequency precision becomes a problem of the first magnitude,” he wrote in Summer 1928. But he was certain that amateurs could devise new techniques and figure out how to meet this challenge as they had with so many others. Anticipating an international backlash should amateurs stray outside the bands, he urged everyone to “For goodness’ sake get inside your band and stay inside.”

A new radio world dawned on New Year’s Day. “January 1st marks the dividing line between the old and the new in amateur radio,” declared Warner in his opening editorial in January 1929 QST.2 Although amateur radio was the victim of “stupidity and crass greed, and fear and lack of knowledge of us” at the Washington Convention of 1927,3 it nevertheless had secured a permanent position among the recognized group of worldwide radio services, and retained a set of exclusive frequency bands, albeit smaller than they had been. Progress in transmitter and receiver design in just the past year was a cause for great optimism about continuing “to exact from amateur radio that full measure of enjoyment which for so many years has held us in thrall to it.” In fact, the convention may have been a blessing in disguise, he contended, that would spur amateurs to improve the sharpness and purity of their transmitted signals, the selectivity of their receivers, and the accuracy with which frequency was determined in both—all things that should have been done long ago.

The last two years had produced a new law, an international radio treaty and new federal regulations. Together they were dramatically changing how amateurs viewed the radio spectrum. Maxim’s concerns with frequency precision were well justified, and while precise frequency determination was still quite a challenge, amateurs had to at least be sure they were hitting the new bands.

To help manage the frequency control problem, the ARRL announced in November 1928 a new field appointment called Official Frequency Station.4 The currently nominal 2% accuracy in frequency was simply inadequate on the higher bands. On 40 meters, for example, this amounted to 140 kHz, which meant that a station could operate only between 7,140 and 7,160 and still be confident of being in the band. And on 20 meters that level of inaccuracy made it impossible to operate anywhere in the band and still be sure of being inside its edges!

The new service called standard frequency transmissions from the Bureau of Standards would maintain an accuracy of 0.1% to 1%, which helped improve the situation somewhat. An ARRL OFS would be capable of 0.5% accuracy—that’s 35 kHz in the 40 meter band, about as good as any amateur station was capable of in 1929. Don Wallace, W6AM, led the committee on Official Frequency Stations as its chairman.

But accurately measuring frequency was not enough. Everyone together had to fit into those bands. Broad and wandering signals could no longer be tolerated. Amateurs had seen this before, of course; it was the same reason spark had been outlawed.

Transmitters must had to become more stable and produce signals that were no wider than necessary, and receivers had to become more selective and not produce signals of their own. And both had to become much more accurate. Amateurs began to refer to a new, yet-to-be-invented class of equipment as 1929-type because that would be the first year on the other side of Warner’s technological dividing line separating old from new.

QST Technical Editor Robert Kruse had begun laying the groundwork for 1929 in a series of articles that ran through 1927 and 1928, beginning with a review of the basics of vacuum tube circuits,5 including oscillators for transmitting and receiving. Hartley, Meissner, Armstrong, Colpitts and others—he explained them all. Because he worked each circuit up from simple, minimal versions and then derived one by tinkering with another, explaining the tradeoffs involved in each one, Kruse’s series is instructional and relevant even today.

The next year, in Spring 1929, QST began a one-year series of articles focusing on improving equipment and called it the Technical Development Program. For transmitters this included power supply filtering, rectifiers, and crystal oscillators; for receivers, calibrated wave meters, bandpass filters, and shielded inter-stage transformers; and for both, band switching using interchangeable, pluggable coils, and more precise, repeatable frequency selection or band spread. And of particular interest to both broadcasters and amateurs was the development of increasingly sophisticated techniques for radiotelephone, particularly 100% modulation and linear amplifiers.6

To recognize progress Handy began listing stations nominated by ARRL Section Managers as having the best signals observed on the air.7 To qualify, a station had to be heard by two independent observers and over several different times—a precursor to the A-1 Operator Club.

BT sep sm

To further encourage the development of better equipment, the ARRL announced in early 1929 that a trophy cup and cash prizes ($25, $15, and $10) would be awarded the following January for the best photos and descriptions of 1929-type stations8 as judged by the editors of QST. Competition was open to ARRL members in all countries, except for the headquarters staff. Their hope was that amateurs would start with the techniques published recently in the series of QST articles and move beyond them in interesting and effective new ways. Each month the best submission up to that date would be chosen for publication. Winners for 1929 would be selected from those published during the year.

The features to be considered by the judges included (as expressed in QST):

  • Ingenuity employed in design, construction and arrangement
  • The transmitter
  • The receiver
  • Power supply for the transmitter
  • The antenna system
  • Change-over arrangements
  • Provisions for monitoring transmitter output for quality
  • Provisions for accurately knowing the emitted frequency of the transmitter
  • Provisions for working on different bands
  • Provisions for rapid and accurate change of frequency within each band
  • The goodness of the keying system
  • Workmanship
  • Extent to which the apparatus is “homemade”
  • Interest and intelligibility of the descriptive manuscript and illustrations

In January, H. L. O’Heffernan, G5BY, was declared the winner, with W8CEO, W1WV and W8BQ placing next in order.9 O’Heffernan had also been the top scorer in both the 1927 and 1928 International Relay Parties as 5BY before prefixes were in use. After the second event, he scrapped his entire relay-party-winning station and set out that spring to construct new equipment specifically designed to meet the requirements of a 1929-type station, with expenses considered “of secondary consideration,” and got it on the air that November, more than one month early.10

His transmitter used a crystal controlled oscillator for stability and accuracy, but did not trade off flexibility, the usual sacrifice. It was built with two pairs of crystals, one pair at each end of the band, which could be selected by a rotary switch. The crystal sockets and switch were all homemade like the rest of the station. This arrangement allowed him to shift between two frequencies to avoid interference without retuning, in whichever end of the band he happened to be operating. The oscillator produced a 3,500 kHz signal, which fed a cascade of one or two frequency doubling stages, thus selecting either the 7-MHz or 14-MHz band. Interchangeable tank coils enabled band selection in the second doubler and final amplifier stages.

The keying scheme used a technique introduced two years earlier in QST for eliminating key clicks and back-wave.11 12 A capacitor coupling the oscillator output to the first frequency doubler had one of its plates mounted to the arm of a telegraph sounder that was energized by the keying circuit. When the key was pressed, the capacitor plates came together pressing on its mica insulator for maximum capacitance, allowing the signal to pass. When the key was lifted, the capacitor plates would separate and cause the signal to die out. A similar sounder-manipulated capacitor was inserted in the output lead connecting to the antenna. Together, they served to soften the keying characteristic to completely eliminate clicks (or thumps, as they were called), suppressed the key-up signal (back-wave), and allowed the oscillator to run uninterrupted and therefore maintain a very stable frequency. A third relay operated from the same keying line switched the antenna between transmitter and receiver for break-in operation. His full wave Zeppelin antenna for 14 MHz operated as a half wave on 7 MHz and was fed with parallel feeders spaced 8 inches apart with paraffin-soaked wooden insulators.

While the resulting signal may have been free of clicks and thumps, one can imagine the racket inside the operating room as the transmitter was keyed. But since it was entirely controlled by this system of relays, it could easily be operated remotely, one of O’Heffernan’s design intentions. His relatively portable receiver could be quickly disconnected from the antenna and batteries in the radio room, set up on a table at bed side with a duplicate power supply and separate receiving antenna. In a matter of minutes he could be operating while in bed (and did so), sacrificing only his ability to change transmit frequency. While not strictly a 1929-type station requirement, this may have been a breakthrough in mitigating operator fatigue (or at least accommodating it).

Power for the transmitter was derived from the 200-volt, 50-Hz residential electrical service using a collection of transformers to power filaments, transmitter signal generation tubes and final amplifier. The most critical power source for a clean signal was the high-voltage supply for the oscillator—the only one to use a full-wave vacuum tube rectifier with capacitors and chokes for filtering. This gave the station its pure CW note. The other tubes had their own transformers and chemical (or electrolytic) rectifiers made from two-pound jam jars filled with an electrolyte solution of borax and distilled water. Twenty-five jars in series on each side of the 1400-volt transformer secondary together with filter capacitors supplied DC for the amplifier stage, and another setup using fifteen on each side of the lower HV supply for the other tubes, totaling eighty jars in all—some serious home brewing! Most of the power supply was located in the basement below the operating room. A bank of batteries supplied the grid bias.

O’Heffernan’s four-tube regenerative receiver was built directly from a design in QST 13 that contained several 1929-type features such as increased selectivity and finer tuning on the higher frequencies. But since most transmitters on the air were still unstable in one way or another, increasing selectivity to help eliminate QRM became counterproductive beyond a point where the received signal would not stay within the passband. (It’s unclear exactly how narrow his threshold was, but may have been around 1 or 2 kHz.)

Separate “midget” variable capacitors were used for changing bands, in a home made plug-in arrangement designed to preserve calibration. Each higher-frequency band capacitor had plates removed to reduce its capacitance in order to spread out the tuning rate. A commercial tuning mechanism with a drum dial permitted these capacitors to be interchanged relatively easily without disturbing the front panel control.

Rounding out the station was the dual purpose signal monitor and frequency meter, an essential component in a 1929-type station. As a monitor, it allowed O’Heffernan to hear his signal while transmitting. When calibrated to the transmitter’s crystals, it could be used as a heterodyne frequency meter with a claimed accuracy of 0.1%.

The next generation of amateur radio station had arrived, prompted yet again by the booming proliferation of radio services and the resulting international tightening of amateur regulations.

AR sep sm

de W2PA

  1. H. P. Maxim, “Frequency Precision,” QST, July 1928, 8.
  2. K. B. Warner, Editorial, QST, January 1929, 7.
  3. See “Treaty” and “Family Harmonics”
  4. “Official Frequency Stations,” QST, November 1928, 68.
  5. R. S. Kruse, “How Our Tube Circuits Work,” Parts 1, 2, 3, 4, QST, December 1926, 9, January 1927, 27, February 1927, 9., March 1927, 38.
  6. C. B. DeSoto, “Two Hundred Meters and Down,” The American Radio Relay League, Inc., 1936, 118.
  7. F. E. Handy, “High Grade Stations – 1929 Signals,” The Communications Department, QST, April 1929, V (addendum pages).
  8. K. B. Warner, “We Open a Station-Description Contest,” QST, March 1929, 37.
  9. J. J. Lamb, “G5BY Wins 1929 Station Description Contest Cup,” QST, January 1930, 17.
  10. “G5BY,” QST, October 1929, 23.
  11. Oscillators were especially unstable when switching on and off. One solution was to simply shift the frequency somewhere else where it would appear as an inverted CW. This other signal was called back-wave. See, for example the chapter “Crossings III—Accolades.”
  12. A. G. Shafer, “Keying the Amplifier,” QST, July 1927, 33.
  13. R. A. Hull, “High-Frequency Receivers for the Coming Year,” QST, November 1928, 9.

Leave a Reply