THE REFLEX/REGENERATIVE AM RECEIVER

The last phone number that I remember at 1247-36th Street was LEnnox 2-3838. I seem to recall at least one phone number change so that our parents could regain control of the phone system from my teenage sisters.

The radio building started there. There were many kludges: Starting with tearing apart the bias coils from dynamic speakers to use the enameled wire to wind coils on various round things like mailing tubes and oatmeal cartons. Building a "foxhole radio" using a rusty razor blade, safety pin and a piece of pencil lead for a rectifying semiconductor. Then there was the first "real radio"--both Wayne and I received a Heathkit crystal radio set (kit) in the mail one day. Pop had, as usual, taken his (small) soldering iron to work with him. We were so uncontainably enthusiastic that we made do with a 200W soldering iron that was available (taking great care to meticulously heatsink the leads of the Germanium rectifier which had to be soldered in each kit). They actually worked. I'm still surprised that we didn't fry everything with that humongous iron.

Ahh, the zillions of variations. We were stuck with 2000 ohm impedance headphones which loaded the crystal radios so much that their Q, selectivity and sensitivity were pathetic and we could only receive a couple of nearby stations with a longwire antenna. Then transistors became available. The first ones were the Raytheon CK722 (of which Carl actually bought a couple of NEW!) and the General Electric 2N107. We're talking crude here, but I suppose that they actually exhibited current gain.

Now we get to go through all the variations of crystal set front end with single and dual stage transistor audio amplifiers. Still poor Q, selectivity and sensitivity, but louder and an added battery expense.

It was after we moved to Dutton (the telephone exchange was MYtrle--the last "named" phone number I knew) that I had my solid state epiphany--the Sylvania 2N229. WOW! Here was a transistor cheap enough for me to buy new and it was HOT! It provided high gain and reasonable transition frequency. This is where I reached the zenith of my "simple radio" career--a circuit from some electronics magazine which was a "reflex/regenerative" design. What a neat trick--a regenerative detector which had the rectified audio fed back through the same transistor. The sensitivity was awesome, the Q excellent, and with some tweaking I could get very respectable selectivity for a regenerative receiver. I could receive local stations without an external antenna! After sundown with an external antenna I could copy AM radio stations from all over North America. I think it was with this set that I stumbled across WLS Chicago (even though WLS was pumping 50,000 Watts, receiving it was problematic because their propagation pattern was off in another direction—no doubt to protect the local stations because WLS was already in the next century in terms of programming and special audio effects--if one had a choice between WLS and the local yokels, there would be no choice).

When I could buy a Japanese 6 transistor superhetrodyne radio for $10, I stopped building my own AM radios. I still have fond memories of swinging at Dutton Shadyside Park with my trusty Japanese radio propped against one of the uprights of the swingset--using the metal of the swingset structure as a boost antenna, allowing me to receive WLS reliably in the daytime.

The "AM radio era" seems like such a long time in retrospect but couldn't have been much more that 4 years (9-13 years of age).

Postscript: in the 70's I did some library searches for the circuit of the reflex/regen radio to no avail. In `99 I spent a bit of time on the internet trying to track it down (figuring that such a novel circuit would certainly have resurfaced in the intervening years)--still no go. At that point, I sat down with the information that I had on reflex receivers, regenerative receivers (and my misty memory) and drew up a circuit which was my best guess about what I'd built in the early 60’s. I didn't assign any component values--just drew the squiggles representing the components and their connections. I also made some general notes to myself about the various considerations of the circuit elements. In 2001 I sat down and figured out the component values and built it again (using techniques similar to those of my youth--things screwed and glued to a piece of wood) and it worked first try--a one transistor time machine!