Thursday, October 28, 2010

A One-Tetrode CW Station

Hi Myrton,
The rig that you worked last night was the result of a chance find at the Deerfield, New Hampshire hamfest a few weeks ago. Dunno why, but I paid a buck for a neat-looking old vacuum tube. It was a 24A, made by RCA. Turns out, this is the direct successor of the UY-224; the first American-made, indirectly heated, screen-grid tube. My example bears the inscription, "Wards Airline" on the base. I suppose it was an OEM item for one of those radios. This tube was state-of-the-art in 1929 but pretty much obsolete by 1935.

The circuits used here are entirely run-of-the-mill. For example, I'm sure you could sketch the grid-leak/ticker-feedback regenerative receiver circuit in your sleep. The transmitter is a crystal-controlled, "Miller" oscillator. I initially tried 58Vdc on the transmitter plate (the maximum voltage obtainable with my bench power supply), but I could only tease out 4mW of power at 3.5MHz. I then placed a "wall wart" transformer back-to-back with a filament transformer to produce 150Vdc. A bridge rectifier made from 1N4007's is followed by a 47uF ripple filter. The transmitter output power increased to 50-60mW with the anode supply potential at 150Vdc.

This tube is clearly "played-out." In fact, my attempt to "rejuvenate" the thoriated-tungsten emitter only brought about a temporary increase in cathode emission. Still, the old girl had enough kick left in her for eight CW QSOs (please see list below).

I used three DIP-type, DPDT relays to switch the tube electrodes (grid, screen and anode) between the transmitter and the receiver circuitry. A hand-thrown toggle switch controls two of these relays. The third relay "keys" the RF output power between a 50 Ohm resistor and the antenna. I first tried keying the oscillator cathode return but it chirped like a Chickadee. Allowing the oscillator to free-run helped quite a lot on that account.

Some of my QRP pals have built roughly the same circuit, only using a junction field-effect-transistor (JFET) in place of my old vacuum tube. One complaint that I've heard about the JFET version is due to the temperature rise of the transistor while transmitting. Apparently, this results in an annoying frequency-drift once the set is switched back to "receive." However, my fire-bottle runs hot enough in receive to swamp any differential anode dissipation. Upon throwing the transmit/receive switch my receiver plops right back on frequency.

I had one more pleasant surprise on the initial shake-down of this circuit. Hams have had trouble spotting their transmit frequency on regenerative receivers since the days of Hiram Percy Maxim. A strong signal simply chokes the sensitive regenerative detector. I've read how some of the fellows would place a wash tub over their receiver. Others used a low-power auxiliary VFO as an aid for spotting their transmit frequency.

I've twice heard mention of a (1950's?) spotting technique for use with crystal controlled transmitters. The "trick" involved building the set such that the quartz crystal is very weakly coupled across the receiver antenna terminals. Tuning the receiver through the crystal's series-resonant frequency, a sharp "blip" was supposedly heard in the headphones. Well, that's exactly what I heard as I first tuned my 24A genny across the CW band! I quickly learned to tune to the low-frequency side of this marker. It works perfectly.

The receiver board is located in the foreground of the above photograph. The transmitter/relay board is in the middle. I couldn't find a five-pin tube socket at the hamfest, so I drilled some press-fit holes in the plywood and hard-wired the pins.

I borrowed a power supply from the office to energize the heater, as I didn't have anything readily available to generate 2.5V @ 1.75A. One could practically toast miniature marshmallows on this receiving tube. It uses nearly 4.4w of heater power in order to produce 60mW of RF output; horribly inefficient, but it certainly makes a warm and pleasant glow!

I'll be pulling this rig to pieces as the power supply has to go back to work today. I guess I've had my dollar's worth of fun! A couple of my contacts last night ran over a half-hour. It's amazing what you can do with half-a-hundred milliwatts when the band cooperates.


  1. W4AG, Hillsborough, NC, 1039km, 599/449
  2. N1UIY, Athol, MA, 172km, 599/549
  3. KD1JV, Randolph, NH, 130km, 569/549, "...insane set up you got there Mike, but it works!"
  4. N1GKE, Hope, RI, 278km, 589/579, 30 minute QSO; I peaked at S-9
  5. W4AG, Hillsborough, NC, 1039km, 599/559, Thanks for my best DX, Stan
  6. WA1JHV, Chepachet, RI, 257km, 579/559, 43 minute QSO; I peaked at S-7
  7. NY3A, Glen Rock, PA, 586km, 599/459
  8. KA2REY, Whitehall, NY, 78km, 599/559
Mike, AA1TJ

1 comment:

  1. This is the funnest rig I've seen in a LOOOooong while! What a great tribute to the period technology -and to AA1TJ's peculiar and wonder-full warpage!