Jeff Duntemann's Contrapositive Diary Rotating Header Image

August 28th, 2011:

Thermionic Detectors

The air conditioning condensate drain clogged earlier this week, and it may have been dripping water inside the air handler for several days. I only noticed it once it started spilling out onto the utility room floor. So we shut the damned thing off until Monday, which is as soon as the HVAC guys can get out here to poke at it.

Oh, and it’s been hot. 2003 was hotter here, but 2011 is right up there. My internal thermionic detector was detecting some thermic excess, so midafternoon yesterday I went downstairs to my shop, where it’s 5 degrees cooler. After spinning the dials to make sure nobody was having any fun on 10 meters that I wasn’t having, I sat at the bench and looked around. Something, somewhere had to need fixing.

Oh. Yes, of course. Back in 2008 I built a 1-tube FM BCB receiver pretty much on a lark, from the Popular Electronics article in which GE introduced the Compactron tube. The article, “Introducing the Compactron,” by Phil Hatfield W9GFS, talked about some of the challenges of stuffing as many active elements into a 12-pin bottle as could be done. This was the last gasp of the vacuum tube, but GE gave it a good shot, and in August 1961 having as many as four active devices in one envelope was still pretty hot.


50 years ago…just about now. The example circuit was an FM BCB receiver using a single 6D10 triple triode tube. I had all the parts (even the tube) lying around, so on a whim I built it.

It worked, but it had several defects. One was touchy tuning due to hand capacitance. This is always a problem when both sides of the tuning cap are above RF ground. Another problem was that the tuning range was too wide (about 65 MHz – 120 MHz) and the entire FM band was crammed into about 20 degrees of rotation of the shaft. The last problem was that the regeneration control was just, well, nuts.

The first problem was beyond help; 50 years ago we just lived with hand capacitance. I tried to fix the second problem in 2008 but didn’t improve matters much. I’d painted myself into a corner with the physical arrangement of the tuning coil, and couldn’t easily swap it out. The third problem was still undiagnosed when I set the radio aside to do other things.

So I pulled it down off the shelf, patched it into my bench supply, and powered it up. Still worked; regeneration still nuts. I spun it, listened, spun it some more, and listened some more. (Why do I hear that song with “She wears short-shorts; I wear T-shirts…” every damned time I turn on the radio?) Regeneration was erratic, happening at different points on the pot, and sometimes didn’t happen at all. Alas, I was overanalyzing the operation of the superregenerative detector, thinking maybe I was driving the detector too far in one direction or another. Occam hit me in the head with a frying pan about then: The regeneration pot was bad.

Duhh. The wiper was making only marginal contact with the resistive element. Not sure why. Don’t care that much. Fished a new 50K pot out of the drawer, soldered it in, and wham! Regeneration worked as it should.

Having fixed one out of the three problems, I was on a roll, and being on a roll…I started to build another radio. John Baumann KB7NRN has a design for a similar detector with a grounded rotor. I’ve been intending to build it since, well, 2008, and had most of the parts set aside in a milk jug. So with the upstairs mostly uninhabitable until Monday, I got to work downstairs last night, and have gotten pretty far along on the RF portion of the circuit.


This is only a lashup; once I get the circuit working as I want, I’m going to build another instance on a conventional chassis with its own power supply. It’s going to take a few more evenings to get to the smoke-test stage, but I’ll let you know the rest of it comes along.