Remember the infamous ARRL "Single Tube 75-watt Transmitter"? Ever seen THIS??

Discussion in 'Homebrew and Kit Projects' started by W7UUU, Oct 2, 2017.

ad: L-HROutlet
ad: l-rl
ad: Left-3
ad: Subscribe
ad: MessiPaoloni-1
ad: L-MFJ
ad: l-assoc
ad: Left-2
  1. W7UUU

    W7UUU QRZ Lifetime Member #133 Life Member Platinum Subscriber QRZ Page

    I stumbled upon this during a search for something unrelated. In all my years I'd never seen this before, and I'm wondering if I'm not alone in that.

    "One-Tube XTAL rig gives 150-watt output on CW" :eek:

    The image I captured is likely too small to read the text so here it is transcribed:


    May be plate-modulated with 60% efficiency for carrier output of 100 watts

    The long-cherished amateur dream of a one-tube crystal-controlled 'phone or CW transmitter comes very close to being realized with the new RCA-813 beam power tetrode. In plate-modulated service, 100% modulation can be obtained with good linearity, low distortion, and a carrier output of 100 watts! In CW telegraphy service, excellent keying can be accomplished in the screen circuit and a power output of 150 watts can be obtained! In neither case is the r-f crystal current excessive.

    The circuit used to achieve these truly remarkable results is the Reinartz tetrode crystal oscillator arrangement, shown in diagram UC-14. The data give above for plate modulated operation was obtained at a frequency of 3990 kcs...." etc. etc. The article continues on to a second page - suggesting using this to drive a KW amplifier.

    I really wonder if anyone ever built one of these?!

    I gotta admit it would be fun to try - I have a couple 813 tubes laying around. Not to mention plenty of B&W link coils for various bands. But the 1500 volts at 180 mA might be a challenge :) I could run it a lot less (850 I have on hand) just for the kicks of having an 813 as an oscillator :)

    Just thought others might get a kick out of this.

    Full 2-page PDF is attached.


    150-watt single tube RCA 813 transmitter.JPG

    Attached Files:

    N7ANN, N2EY, AC0OB and 2 others like this.
  2. K9STH

    K9STH Ham Member Volunteer Moderator QRZ Page

    There were similar articles using the 814 and other tubes besides the 6L6, 6V6, 807 and even the 6146. For a number of editions, there was a 6146 oscillator in the ARRL Handbooks. I even built a few of those for new Novice Class operators when I was in high school.

    Being in the RCA "Ham Tips", you can bet that a number of the 813 transmitters were built and that they did work!

    I have a Nuvistor 432 MHz converter that appears in the "Ham Tips" and it definitely does work.

    Glen, K9STH
    KX4OM and W7UUU like this.
  3. N0TZU

    N0TZU Platinum Subscriber Platinum Subscriber QRZ Page

    K is a 2500V keying relay, "do not use an ordinary key in this position under any circumstances" :eek::eek::eek:
    W7UUU likes this.
  4. N2EY

    N2EY Ham Member QRZ Page

    Remember always that pre-WW2 rigs used the much bigger round crystals that plugged into a 5 pin tube socket. They could take much more crystal current than an FT-243. (The FT-243 was specifically developed during WW2 to use less precious quartz).

    The 813 is nominally a 500 watt tube, so running one at 250 watts input is an easy life for it. Note that the transmitter only operates on the crystal frequency, no doubling or tripling. The rig may save a few parts, but it needs a 1500 volt supply!

    IMHO, such rigs are best left in the history books. Building a simple MOPA isn't that much more complex and the results are so much better.

    A 6AG7 driving a pair of 807s will give 100 watts of clean CW with 750 volts on the 807 plates and 300 or so volts for the 6AG7. Easier, cheaper, smaller, safer, and you can use the same rocks on 2 bands (straight through and doubling).

    73 de Jim, N2EY
    KX4OM, WA7PRC, W4KJG and 1 other person like this.
  5. W7UUU

    W7UUU QRZ Lifetime Member #133 Life Member Platinum Subscriber QRZ Page

    Just fun to think about isn't it :) That RCA article put a smile on my face on a day I really needed one.

    N7ANN, K9ASE and N2EY like this.
  6. W0AAT

    W0AAT XML Subscriber QRZ Page

    I have a 435 volt xformer at 350 mils, add a voltage doubler... ~1200 volts... nope I am not doing this. I have an aversion for high voltage these days! Neck nerve damage means my hands make sudden unintended movements at times.
  7. N2EY

    N2EY Ham Member QRZ Page

    If you liked that one, you're gonna LOVE this one....

    QST for September, 1932, starting on page 17. "Thirty Three Watts Per Dollar from a Type '52" by W6CUH.

    The "Type '52" is the 852 tube. Look up the ratings on that bottle......

    Note that the author's design kinda pushes the ratings a bit. Also note how many ways there are to kill yourself in the transmitter, with 4500 volts.

    73 de Jim, N2EY
  8. K2CAJ

    K2CAJ XML Subscriber QRZ Page

    Suitable for very, very short CW transmissions:

    *bzzt* "DAAH!"
    N4MFA likes this.
  9. W7UUU

    W7UUU QRZ Lifetime Member #133 Life Member Platinum Subscriber QRZ Page

    Oh my! I just looked up the article - what an insane idea! I love how he describes removing every other plate in the capacitors to get to a safe arc-over spacing! The "Tuning Precautions" section is particularly hair-raising. I cannot imagine QST or any other publication running an article such as this today.

    One curiosity: the author describes output being tested at 14.4 MHz. I realize of course that until 1947 the upper end of 20 was in fact 14.4 but why would one test and proclaim operation at the exact band edge? That QST would publish that, even in 1932, is puzzling - clearly wasn't legal then any more than putting a KW signal out on 14.000 or 14.350

    N7ANN likes this.
  10. N2EY

    N2EY Ham Member QRZ Page

    Actually quite a common practice. What's a bit much is that he had to remove all the insulation and devise a mounting system because the frame would arc and smoke....

    Here are the typical maximum recommended CW operation ratings on the 852 (per tube), followed by W6CUH's operating values (in parenthesis) in the push-pull setup:

    Plate voltage - 3000 volts (4500)
    Plate current - 85 mA (112.5)
    Plate input - 255 watts (500)
    Power output - 165 watts (425)
    Plate dissipation - 100 watts (75)
    Grid current - 15 mA (50)

    The extremely high claimed efficiency (85%!) is the result of very high plate voltage and being driven very hard. Note the grid leak resistors - the driver stage grid leak is a 100 watt unit and the final stage grid leak is a 200 watt unit! When you need to develop over 1000 volts of grid bias, with grid current of 50 mils or so per tube, the wattage adds up. (Note also the 2 ohm parasitic suppressors...)

    The pictures in the archive scans aren't that great. I have the actual magazine, and the clearer pictures scare the bejeebus out of me. Not only is everything out in the open but the only insulation between the tuning capacitor shaft (which has 4500 volts on it) and the operator is the knob itself.

    OTOH, because primary keying is used, there is no B+ until the key is closed.

    Note the antenna connection. There's no link or capacitor - the single-wire feeder from the Windom-type antenna simply clips onto the tank coil. Impedance match is achieved by choosing the right tap point. What could POSSIBLY go wrong?

    And yet W6CUH used it safely, because he went on to write more articles for QST in later years.

    Some of the descriptions in the text are fun to read - and may require some translation. For example:

    "Primary keying was chosen because of its many advantages. In the first place it completely eliminates key clicks, no mean achievement with powers on the order of a kilowatt. The filter condenser is relieved of excessive peak voltages, obviating the use of a bleeder resistor, itself a large item at a voltage of 4500."

    "The plate supply itself consists of a 2 kw. transformer delivering up to 4500 volts each side of of the center tap, a 10,000 volt 10 amp mercury arc with a conventional keep-alive system, and a 2 uF 12,000 volt mica filter condenser". (That's a mercury arc, not mercury vapor - with a mica filter capacitor. I don't think Mouser or Digi-Key carries them.)

    "The antenna feed line should then be clipped on the tank coil about about two turns either side of the center tap. Otherwise the application of full voltage to the Type '52's will cause the tank condenser to arc over despite its 12,000 volt rating" (I suppose he learned that the hard way).

    "RFC - 150 turns of No. 30 enameled wire space-wound on a glass toothbrush holder"

    (the scary part is that I know what all of that stuff means. I wonder what size Tungars he used in the keep-alive?)

    In those days, everyone knew that the frequencies were approximate and would be kept inside the band. Since this was a CW only setup, "14.4" actually meant "14.399" or so.

    Note the harmonic relationship of the three frequencies he mentions. He probably had just one crystal, slightly below 3.6 MHz. And, in those days, (1936 ARRL Handbook) the 'phone part of 20 was 14.150 to 14.250, 75 meters was 3.9 to 4.0, and 40 was all-CW - no 'phone at all.

    150 watts from an 813 as a power oscillator at 1500 volts? Oh puh-leeze.....

    73 de Jim, N2EY

Share This Page