Four Tubes, One Bottle, No Waiting

In 1960, the writing was on the wall: Tubes were being bested in every area by transistors, which did not require heater power, did not generate as much heat, and were a tiny fraction of the size. GE's engineers made an interesting decision: To give tubes one more shot in the war with transistors, and push off the inevitable end of the vacuum path for a few more years.

What GE did was combine multiple common tube types into "fat" tubes—as many as four in a single glass envelope, all heated from the same filament. The idea was to reduce the amount of power required to heat the tubes and the space they required on the circuit board, as well as the associated costs of multiple sockets. In a very crude way, GE was applying the concept of integrated circuits to tubes, and if solid-state technlogy had not advanced as quickly as it did in the first half of the 1960s, the Compactron idea might have been developed further, with even more tubes in a single envelope, with perhaps resistors and capacitors to implement resistance coupling networks for amplifiers. We'll never know, but Robert Casey has an interesting page speculating on a sort of "super-Compactron" implementing all tubes required for an "All-American Five" table radio in a single envelope. This was never done, but it would certainly have been possible.

In truth, Compactrons were designed almost completely for the color TV market. Table radios and Hi-Fi equipment represented much simpler design challenges for electrical engineers. (Some high-end Scott AM/FM tuners used Compactron tubes, but not many.) Color TVs are tough nuts to crack in many respects, and the transistors of the time (1960) were not quite as good as tubes, especially for high-voltage and VHF/UHF circuits. So GE's engineers took a look at the various elements of Color TV circuitry, and started combining compatible tube elements into single glass bottles.

The result is the catalog of Compactron tubes we know today. A good example is the 6AF11, a three-tube portmanteau consisting of a high-mu triode for AGC keyer service, a medium-mu triode for sync separator service, and a sharp-cutoff pentode for video amplifier service. The 6T9 is a hi-mu triode and power pentode combo representing the entire audio section of a simple TV set. (There was no stereo sound in TV technology at that time.) These sound very application-specific, but they're still just tubes, and they can be used for lots more than the TV circuits that they were designed for. The 6T9, for example, can work at RF frequencies up to 30 MHz, and there are a number of circuits using it for simple amateur radio CW transmitters, with the triode acting as a crystal oscillator and the power pentode as a 5W RF power amplifier. The 6AF11 is used in a popular regenerative all-band receiver circuit designed at GE and published in Popular Electronics in 1963.

Compactron tubes are available in huge numbers, and quite cheap, for this reason: They were manufactured in huge numbers right before the bottom fell out of the tube market. Immense quantities survived, and now you can get extremely useful tubes like the 6AF11 and the 6T9 for a couple of bucks each, which is a fraction of what they cost (in inflation-adjusted dollars) in 1963.

Is This Really a Win?

So. Is stuffing as many tubes as possible in one glass bottle really a win? I think so, but there are a couple of downsides to using Compactrons in hobby projects:

• The mix of tubes is not always precisely what you need, and if it isn't, one or more of those tube assemblies inside the bottle won't be useful. You may end up designing for what you have rather than with what you need, and while that's a subtle difference (and not always a bad thing) it's a real engineering challenge.
• Point-to-point wiring to Compactron sockets can be tricky. Compactrons were developed late enough in tube history to have been intended for printed circuit wiring, not solder tab sockets. Components pile up around thoser solder tabs, making it absolutely crucial to plan the order in which components are soldered in, lest components soldered first block access to tabs needed later. Also, the solder tabs are very close together, making short circuits very easy to do without extreme care, especially when using larger components (think: one or two watt resistors) with larger gauge leads. The photo below (of my partially completed 6T9 stereo amplifier) will give you a sense for what I mean—and not all pins on the 6T9 sockets are used!

• If you have some experience with tubes, keep in mind that the individual tube sections within a Compactron are often electrically identical to older, single-section tubes with similar functions. The gated beam discriminator section for a 6Z10 Compactron, for example, is identical to the 6BN6. If you have any 6BN6 circuits or experience, you can apply that knowledge when using the 6Z10—and the power pentode section is identical to that of a 6T9. Get a late tube-era data book (Antique Electronic Supply has them) and study it!

What I'm Doing Here

This page is a compendium of what I know about Compactron tubes, as well as any Web links or citations in paper literature about them. As I assemble circuits of my own using Compactron tubes, I will post the schematics and photos here. Check back from time to time! And if you have anything to contribute, drop me a note at:

Compactron Tube Listing

Below is a list of all the generally useful 6V and 12V filament Compactron tubes that I know about. I'm omitting Compactrons with odd series-string filament voltages (8V, 15V, 17V, 21V, etc.) because they require more fooling around to use. I'm also omitting some of the tubes that are extremely specific to color TV service, like shunt regulators, just to avoid cluttering up the list with devices that hobbyists are not likely to use. There are a handful of types that I am omitting because I can't find good online data for them; the 6KE6 sweep tube is one of these.

Links to tube data are (mostly) to the wonderful NJ7P Tube Data Site. A (D) at the end of a tube description is a link to a PDF of the tube's full data sheet, which generally contains characteristic curves for the device. (Keep in mind that these are page scans and are often very large files, sometimes over 1 MB.) The PDF data sheets are (mostly) from Frank Philipse's tube data sheet pages. Tubes for which no links are present at Compactrons that I know exist (generally by finding them in tube-tester data books) but for which I have not yet found data online.

If you know of any other useful 6V or 12V Compactron tubes not listed here, please let me know!

There is also a photoconductive sensor cell manufactured (circa 1963) by GE in a Compactron envelope. It's type Z-2946, though I have found out nothing about it other than the fact that it exists. If you ever run across a data sheet, please let me know.

Compactron tube sockets (both solder tab and printed circuit types) are reasonably common at Antique Electronic Supply and on eBay.

Safety Issues for High-Voltage Tubes

It's important to say right here up front: The voltages you must use in Compactron tube work can kill you. Years ago, the dangers of high voltages were common knowledge among electronics people, because high voltage was present in virtually everything you could call "electronic." However, over the past forty years, semiconductor technology operating at 12V or less has taken over hobby electronics. We've gotten very used to poking at circuit lashups with our bare fingers to try and spot bad solder joints or to detect overheating components. To stay alive, you're going to have to learn a whole new set of techniques, and work by them religiously:

• If you sit on a stool at your bench, make sure the stool's legs are insulated from the floor using rubber "crutch tips" or something similar. In general, if you can find an all-wooden stool to sit on, so much the better.
• Put nonconductive plastic knobs on all controls.
• Once power is on to a project, don't touch anything "behind the panel" with your fingers.
• Don't probe around inside projects with metal tools unless you're double-sure that power is unplugged. Don't settle for "off." Even if the handles are insulated and you don't get a shock from the circuit, if you bridge high voltage to negative or ground, you can damage the tool.
• Even after power is unplugged, before you begin sticking your fingers into a project, bleed off any charge that might remain in any filter capacitors. I made up a tool which is simply a 27,000-ohm, 10 watt wire-wound power resistor mounted on a length of wood dowel, with the resistor's leads sticking straight out for about an inch or a little more. This should be good for caps holding as much as 550VDC. To use it, bend the resistor's leads so that you can easily bridge the hot lug on the cap to ground. Then bring the resistor's leads in contact with the hot lug and ground and hold it there for about ten seconds. By that time, no dangerous charge will be left in the cap.

Drawing Your Own Tube Schematics with Visio

If you modify published tube circuits before you build them, it's always a good idea to redraw them so you don't forget how the circuit on the bench differs from the one in that old article in QST. There's a very nice drawing program called Visio (now owned by Microsoft) that I've been using to draw schematics for over ten years now. Older versions can be had on eBay for as little as $20, and in fact the version I use—Visio 2000—is now six years old and still perfectly useful. (The latest versions require product activation, which I cannot abide, but in truth there's little in the newest Visio versions that isn't in Visio 2000. Just don't use Visio 1.0.)

I created a stencil file full of tube pinouts, including all the common tube types (dual diode, triode, tetrode, pentode, and so on) and all you need to do is drag them off the stencil and start connecting the leads. (Visio comes with stencils for all the common components like resistors, capacitors, and inductors.)

You can download the stencil file here. It has stencils for the several earlier versions of Visio, but the Visio 5 stencil works fine with Visio 2000 and later versions.

Compactron Tube Material on the Web

• Multi-Function Compactrons Promise 2-Tube Radio. This is the news release from GE that started it all, way back in July, 1960. The most interesting thing is that the two tubes that GE's engineers promise in the news release (and are said to be available in sample quantities) never went into production, as best I know.

• Audio Amplifiers from TV Compactrons. A nice simple 2-watt mono circuit from Wade Stanfill that may be constructed using any of a (long) list of Compactron tubes. Wade provides plenty of tips for jiggering the circuit based on what you have in terms of transformers. If you want to try a simple Compactron audio amp and have a little experience "rolling your own" with tubes this is a very good place to start.

• Min-Amp. By Jon Stanley. A simple 1W audio amp built on an Altoids box, with a second Altoids box housing the HV power supply. The tube is a 6LU8.

• A 6EM7 Headphone Amp. An experimental high-fidelity audio amp intended to drive headphones rather than speakers.

• Compactron AA5 AM Tube Radio by Bob Casey WA2ISE. Bob re-implements the classic All-American Five AM receiver using three Compactrons. The three tubes are the 11AR11, the 8B10, and the 38HK7. Lots of theory, graphs, and practical advice.

Compactron Projects in Books and Magazines

This list of articles and citations were all originally published in print books and magazines, but a very few have been scanned and mounted on the Web. You can find some of these publications on eBay, or on the used book services like ABE Books.

• "Introducing the Compactron," Popular Electronics, August 1961, P 45. GE's Receiving Tube division fielded a number of articles on the tubes at about this time, writing by staff engineers, and this was the first. It uses a 6D10 triple triode (basically, three identical trodes very similar electrically to those in a 12AT7) in a three-stage circuit covering the FM broadcast band, with the stages used as an RF amp, superregenerative detector, and audio amp. A similar but more sophisticated circuit was published the very next month, also from GE.

• "Compactron VHF Receiver," Popular Electronics, September 1961, P 45. A 6D10 triple triode tube provides an RF amp, superregen detector, and audio amp. The article claims that the circuit will detect both AM and FM circuits, but narrow band FM doesn't detect nearly as well with a "rushbox" as AM. The article is also by an engineer from GE, which was pushing Compactrons very hard at that point.

• "Compactron Frequency Calibrator," Popular Electronics, July 1962, P. 75. By the legendary Herb S. Brier W9EGQ, this circuit uses a single 6D10 triple triode to generate marker signals either every 100 KHz or every 10 KHz, depending on a switch setting. One triode section is used as a 100 KHz crystal oscillator, and the other two act as a multivibrator oscillator that modulates the 100 KHz output such that 10 KHz heterodynes are generated as well, up through about 30 MHz.

• "KCS AmateurBand Receiver." GE Ham News. Part 1, Nov/Dec 1962. Part 2, Spring 1963. This receiver used Compactron tubes, but I do not have the GE Ham News issues in which it appears, and can't tell you anything about it. If someone who has it could send me good scans, I will post them.

• "One Receiver, All Bands," Popular Electronics, January 1963, P 39. A 6AF11 dual triode/sharp-cutoff pentode provides a regenerative detector and two stages of audio. I actually built this in 1964, and found it problematic, though how much of that was my youthful inexperience I don't know. (Photos of my 1964 unit are here and here.) It should work modestly well, but the plug-in coil forms cited are no longer available and you'll have to recalculate the coils for whatever forms you can find. It was also written by a GE staff engineer. The same circuit was published in the 1965 GE Hobby Manual, in case you have that in a pile somewhere. The article has been scanned and posted online, at this page. Look for "6AF11: P. Hatfield". The Hobby Manual article has also been scanned and posted here.

• "License-Free FM Transmitter," Electronics Illustrated, July 1964, P 47. Harry Kolbe presents a low-power, hi-fidelity FM BCB transmitter using a 6C10 triple-triode Compactron for its audio stages, and an ECC85/6AQ8 dual triode as a Colpitts oscillator and reactance tube modulator. Circuits like this used to be common: One section of the ECC85 becomes part of the capacitive voltage divider governing the oscillator frequency of the other half. By changing the audio input to the reactance triode, its reactance changes, thus changing the oscillator's frequency. Add an antenna and you've got a transmitter with a 500-foot range. Note: This is a pretty complex circuit for a magazine construction article!

• "Integrated Stereo Amplifier," Electronics Illustrated, July 1964, P. 42. Walter Morrow gives us an ambitious (for a pulp magazine) stereo amplifier project that for each channel uses A 6C10 Compactron for the preamp stages, half a 12AX7 for the driver, and an ECLL800 for the phase inverter driving two power pentodes in push-pull, delivering ten watts to the speakers. The ECL800 is interesting all by itself: It's a species of Compactron in an ordinary 9-pin miniature package containing a triode and two power pentodes. This is a difficult (and expensive) project, but if you want to move beyond a very simple tube amp it might not be a bad way to go, but plan on spending big on the transformers.

• "The Stereo S'Lector," by Alton B. Otis, Jr, Popular Electronics, September 1964, P 73. The author uses a high-gain triple-triode 6D10 to sense the presence of the 19 KHz subcarrier on stereo FM broadcast signals, and trip a 4PDT relay when a subcarrier is sensed. This may seem odd, but FM stereo was very new in 1964, and separate multiplex adapters were often added on to existing hi-fi FM tuners to receive a stereo signal. Subcarrier sense relays were present in high-end FM tuners and did basically the same thing. The only odd part is the JW Miller 1354 19 KHz oscillator coil, and I've seen these for sale here and there. Not a very practical project in 2008, but an interesting circuit nonetheless.

• "The Compact BCer" by Homer Davidson, Electronics Illustrated, September 1964, P 32. Relatively simple AM broadcast regen (not shortwave!) using a single 6AL11 pentode/beam power pentode tube as a detector and speaker amplifier. The main coil is a slug-adjustable Fifties-style ferrite loopstick with a tickler winding added. The newer, non-adjustable transistor radio loopsticks will probably work in the circuit, but as they are not adjustable, you may have to pull windings to resonate an arbitrary variable capacitor. (Look on eBay for loopstick antennas. Most I've seen are non-adjustable, but you just might get lucky.)

• "One Compactron Regulated Power Supply" GE Hobby Manual, 1965, P 185. A fascinating circuit using a 6JZ8 medium-mu triode/power pentode. It's a variable voltage power supply that will give you 150-250 volts regulated, sourcing a maximum of 60 milliamps. Given that solid-state regulators capable of 250V output are fairly rare (though I have one somewhere) this would be a useful and educational gadget to build.

• "Two Compactron Stereo Amplifier" GE Hobby Manual, 1965, P 101. Nice amp using two 6T9 triode/power pentode tubes. Designed for high-level crystal phonograph input, but you can jigger the input network to take almost any kind of audio. I built this in 2005 for amplifying computer audio, and it works beautifully. I added a balance control. My modified schematic can be found here (Flash SWF format) with photos here.

• "A 10-Watt, One-Tube Transmitter" QST, March 1971, P 25. Another great Lew McCoy circuit. I've built this one twice, once point-to-point and another using printed circuits. It's a very simple crystal-controlled CW transmitter using the 6T9 triode/power pentode tube that will give you ten watts input and about five-six watts output. Like many QST circuits of the time it uses Miniductor stock, but I calculated and wound my own plug-in coils on vitamin bottles and it worked just fine. The circuit was also published in the 1973 ARRL Handbook.

• "Build a Vacuum Tube Amp" Nuts & Volts, August 2004, P 50. Simple (maybe too simple) circuit using two 6T9 triode/power pentode tubes. There's no balance or tone controls, since the amp was designed to amplify audio from a computer sound board. The author makes a nice parts kit and circuit boards available for it, and if you're not experienced with tube construction, this would be an excellent place to start. See the author's Web site.

Got anything I've missed? Contact me and I'll add it!