The goal of this project was to build a superhet broadcast band receiver using all 12 volt space-charge tubes. Click here for the schematic in PDF
The space-charge tube was developed during the late 50's to early 60's mainly as a transitional device for car radios to replace the traditional high voltage vacuum tubes. Transistors were coming on the scene, but they were expensive and the upper operating frequency was limited. These tubes operated with 12 volts on both the filaments and plates which eliminated the vibrators and other devices for providing the high voltage. Their operating frequency was well above the broadcast band. One of the drawbacks was the high current demanded by the filaments. However, since the power source was a car battery, it really wasn't a problem. Also, with only 12 volts on the plates, at low current, it was impossible to obtain enough audio power output to drive a car speaker system. The solution was to use a transistor as the speaker driver. This was a good application for the low frequency, high power germanium devices that were available at the time. The result was a hybrid receiver using both vacuum tubes and transistors.
While the power transistor worked well as the audio power device, it still required a substantial amount of drive to obtain maximum output. In several designs, a version of a power space-charge tube was used to drive the transistor. Although referred to as "power", these tubes could only provide power measured in milliwatts. An example of these is the 12K5, a power tetrode that can deliver up 50 milliwatts to a speaker. Not much to speak about, but I wanted an "all space-charge" design, so I had to work around it. My solution was to use two of these tubes in parallel to increase the power and provide a better match to the output transformer I was using. Also, use a sensitive speaker with a large magnet. And, finally, house the radio in a case to take advantage of the acoustics to improve the volume. The result is a radio with more than enough over-all volume for normal listening. In fact, my wife is always saying: "turn down that radio", if that tells you something!
Top view with cover removed
I used a common 4 watt, 70 volt line transformer for the audio output. (as seen on left)
Since I wanted a good performing receiver, I included a RF amp along with two IF stages which brought the tube count to seven.
View of ferrite bar antenna
I also wanted a built-in antenna, so I used a ferrite bar that is mounted in the top rear of the cabinet. I used two round 3/8 inch bars of Delrin, a flat piece of HDPE and three black tie wraps to support it.
Most of the RF critical parts were salvaged from an old tube portable radio that used a 67 1/2 volt "B" battery. The case was a total loss so I decided to part it out because it had a three-section variable capacitor and matching RF and Oscillator coils. I was able to use the ferrite bar antenna as well. Having a set of these matched RF coils and capacitor is critical if you want everything to track. They only needed minor component changes in the circuit to work with the space-charge tubes.
Closeup of the homebrew dial
One thing I didn't like about using the salvaged variable capacitor was the resulting non-linear dial. Notice the crowding between 1300 and 1600. In contrast, check out the dial on my "One-Transistor Radio" that uses a true straight-line tuning capacitor.
Back view showing power connector
The IF transformers are some universal replacements for AA5 sets that I happen to have.
View of the dial and string mechanism
The 4 inch tuning wheel, along with a slide-rule dial, provided a nice, smooth combination.
The bottom is relatively uncluttered since I used mostly 1/4 watt resistors and low-voltage capacitors.
The RF coil is seen at the top right with the Oscillator coil below it. The coil in the upper left is the power line choke that is really only needed if the set is powered from an automobile to reduce engine noise. I had one, so what the heck, I used it.
One thing you might notice in the schematic is the rather large resistor values in the cathodes of the IF stages. It turns out that the 12AF6's have a really high gain, so I used the high values in order to tame the circuits. Even with these values, the over-all gain of the IF is high and it is very stable.
Tune up was easy. First, kill the oscillator stage by grounding the cathode of the 12AD6. Power up the set, then inject a signal by loosely coupling a wire from the signal generator next to the ferrite bar antenna. Connect an oscilloscope with a X 10 probe to the plate of the 12AD6 and make the adjustments while looking for a peak of the signal while injecting minimum signal from the generator. Not shown, in the schematic, are trimmer capacitors mounted on the side of each section of the tuning capacitor. The ferrite bar antenna tracking cannot be changed other than setting the top end of the dial with the associated trimmer capacitor. I set it at 1710 kcs with the tuning set to the high end of the dial. Next adjust the RF coil slug for the low end of the dial which is 530 kcs, the same as the ferrite bar. Then adjust the high end with the RF trimmer capacitor to 1710 kcs. You might have to go back and forth with these adjustments to get it correct because one effects the other. Now, remove the cathode short and turn off the signal generator. Remove the probe from the oscilloscope and connect it to a frequency counter. With the tuning set to the low end of the dial, adjust the Oscillator slug to read 985 kcs on the frequency counter. Then set the dial to the high end and adjust the Oscillator trimmer to read 2165 kcs. As before, go back and forth with these adjustments to obtain the correct settings. Last, align the IF stages by first removing the probe from the plate connection and again, kill the oscillator by grounding the cathode of the 12AD6. Also, disable the AVC by grounding the low side of the 2nd IF transformer's secondary winding. Connect an oscilloscope to the high side of the volume control and inject a 455 kc, modulated with 1 kc, signal into the grid (pin 7) of the 12AD6. Start out with a high enough signal to see a demodulated 1 kc sine wave on the oscilloscope. Now adjust both the top and bottom slugs in all of the IF transformers for a peak while reducing the signal generator level as needed. The final adjustments will result in a recovered sine wave with only microvolts of injected signal.
I was originally going to stager-tune the IF's to obtain good fidelity. However, using the simple "peak everything" method works very well and gives pretty good audio response with sharp selectivity, which is nice for DXing.
The case is made mostly of 1/8, 3/16 and 1/2 inch Garolite.
The final results of this set were surprising. The sensitivity is excellent, especially at the low end of the band and the selectivity is also very good. Other than the audio, it is as good as my GE Super Radio II.
The set draws about 1.8 amps at 12.6 volts. Not exactly the radio you would want to take into a fallout shelter to last for months on one battery! However, with a 12 volt, 7 AH lead-acid battery, it lasts long enough for me, between charges. It also works well on a small 12 volt switching supply I have. I like the battery operation better because there is no line noise and I can take it out on the back porch for some night time listening.
The 12 volt space-charge tubes do have their advantages. Among them is the low plate voltage which is safe to handle and also allows for the use of low voltage components. Also, as an added bonus, these tubes really glow in the dark and produce warmth!
Click here to return to the MODERN HOMEBREW page
© 2006 - 2008 ~ Michael R. Starcher ~ All Rights Reserved