A 'Universal' Regenerative Detector
that receives AM, SW, or FM!
Oct 30, 2004 by Rick Andersen
Here is an amazingly versatile circuit-- look closely and you'll see that it's really the same detector found in my article on the Superregen for FM VHF Aircraft. Turns out that this same Superregen, when used in "normal" regen mode instead of in superregen mode, makes a damn good alternative to the more well-known Armstrong and Hartley -type regens. In my opinion, it makes the best little AM broadcast-band regenerative I've ever played around with.
All you have to do is scale 3 or 4 components' values to get the different bands and modes.
Some history: Ever since my vacuum tube days in the early 1970's when I first became a Novice class ham radio operator, I've built and tinkered with the ubiquitous Armstrong-style Regen: the one with the small "tickler coil" winding that magnetically induced a small amount of signal from the plate circuit back into the main tuning coil in the grid circuit. You can find this design just about anywhere on the Net if you do a search on Regenerative Receivers. The amount of Regeneration or positive feedback was usually controlled by a variable capacitor, separate from the Main Tuning cap; this Regen control cap shunted some of the feedback to ground, allowing the operator to thereby vary the sensitivity of the detector.
When I started learning about the transistor version of the old tube Regen, I learned that some transistor circuits didn't use a variable cap for the regen control; instead, they used a potentiometer to vary the amount of base bias to the transistor, or even the level of collector voltage from the battery.
Then there was the Hartley-style Regen, based on the Hartley Oscillator, where there is no need for a separate Tickler winding, since the main inductor is tapped about 1/3 of the way up from the ground end of the coil. This trick, too, creates a positive feedback condition--- In general, any standard oscillator circuit can be used as a Regen receiving circuit by backing off the feedback so that the oscillator just stops oscillating; at this point it acts as a high gain, narrow bandwidth Bandpass Filter resonating at the frequency you've tuned it to with the main tuning capacitor.
After I had built my first FM/VHF Aircraft receiver (see the article elsewhere on my web pages), it occurred to me that I could, if I wanted to, decrease the gain so that the radio pops out of the loud, hissing Superregenerative mode. What I didn't realize, until I tried this, was that the receiver was now in plain vanilla Regenerative mode, which I could verify by hearing faint heterodyne whistles as I tuned back and forth. Hmmm. I wonder how well this receiver would perform as a "straight" regen...
After some experimentation (always fun for me), I came up with a table of component values that were scaled for each frequency band. I found that, in addition to the obvious scaling of the main inductor L and the main tuning cap Ct necessary to retune the FM/VHF radio to lower bands, I needed to similarly scale the RF choke RFC and the feedback capacitor Cf that sits across the emitter and collector leads of the 2N3904 NPN transistor.
I found that taking the feedback cap up as large as 470 pF or even .001 uF (1000 pF) allowed the detector to work both in the Shortwave/World-Band range of approximately 5-15 MHz (with the parts I had on hand), and that replacing that tiny 6-turn choke coil (in the FM/VHF circuit) with a heftier 10 millihenry, molded inductor available from Electronix Express and other mail-order firms, was what I needed to get the radio to work down as low as the AM broadcast band, 540-1700 KHz.
I also became a believer in magnetic loop antennas!
Take a look at the component values, in the schematic below, and you'll notice that (for Shortwave reception) I specify a square loop of stiff copper wire, 1.5 feet (18 inches) on a side, standing up vertically above the radio box. You don't have to do it this way; you can install or wind a 5-6 uH inductor for "L", and then attach a 20" whip antenna (through a 27 or 33 pF capacitor) to the top of the tuning cap, and get acceptable reception. However, I discovered that a one-turn loop of stiff wire (my buddy had some 'brazing rods' used in welding; he soldered them into a square for me),
broken open at the bottom of the square to provide the two antenna leads going to the radio (in place of the component "L"), made a really great magnetic antenna. [In case you don't know, loop antennas are more directional than electric-field whips, and pick up less electrical noise.]
The only things to watch out for are that you don't allow the wire in the loop to short to any exposed, grounded metal (since the loop is at 9 volts DC potential), and that you keep things mechanically rigid. This last point is important since the Regen will wobble its tuning frequency if you're listening in oscillating mode to some CW or SSB, and the square loop is wobbling physically.
An AM Broadcast Receiver built from this circuit in a balsa-wood box
For the AM broadcast version of this versatile circuit, I went to a craft store and bought some thick balsa-like wood and glue, and made me a box. Mounted a 365 pF tuning cap through a hole drilled in the front panel, and installed a Regen potentiometer also. The regen circuit was built "Ugly Construction"-style on a 2x3 inch piece of copper-clad circuit board. Instead of the Radio Shack amplified speaker shown in the schematic, I rolled my own little Class AB audio amp, using a 2N3904 preamp ahead of a 2N3904 / 2N3906 complementary ('push-pull') pair amplifier.
The Volume control is the Regeneration control. I mounted a 3" speaker in the front panel, behind a series of drilled holes.
The inductor (Loop Antenna), in this AM radio, borrowed from a long and glorious history of AM tube radios from the 1950s and '60's, is a flat, oblong coil of somewhere between 17 and 25 turns of #22 wire, glued vertically against the inside of the rear wall of the wooden box. I say "somewhere between 17 and 25 turns" because the inductance varies with the tightness of the winding, the exact size of the coil, etc. Mine is about 3x6, maybe 3x7" and I adjusted the turns until I was able to receive our local Harrisburg, PA powerhouse, WHP AM 580 KHz, near the bottom of the dial (tuning cap almost fully meshed).
The receiver pulls in stations amazingly well with this 3x6 inch loop antenna, in my rural, hilly valley about 25 miles outside of Harrisburg, where TV reception is almost nonexistent.
At night, the performance is just amazing, for such a simple radio. As you tune up the band toward 1700 KHz, you need to back off the regeneration more and more to keep the set from squealing in oscillation; conversely the low end needs quite a bit more regen feedback in order to hear anything. Right at the edge of oscillation, the selectivity can be very sharp!
One enhancement that I mentioned in the article on Crystal receivers is that you can boost the sensitivity of any of these simple radios by running a ground wire to between a point at circuit ground, and, say, a cold water pipe or that little screw that holds the face plate onto your wall electric socket. But really, this radio is so sensitive that you don't need to do this, especially if you live closer to a large city than I do.
I know, I say this for EVERY article: Try building this circuit if you want to experience some of the Regen's famous sensitivity-- and this one doesn't need fancy tickler- or tapped-coils. The feedback path is through Cf (collector to emitter) and in essence this circuit resembles a variation on the Colpitts Oscillator.
Oh-- Did I mention that this EXACT SAME CIRCUIT is used in simple FM wireless mikes also?
Yes-- It's a transmitter too! Now do you see why I call this a "Universal" receiver?