Regenerative receivers have a long history in radio dating back to the first days of the vacuum tube. Though they were once very popular among amateurs and professionals alike, they are now thought of as mere play toys instead of legitimate receivers. Most of this reputation is due to design short cuts that have been implemented over the years. They perform very well when properly designed and constructed, contrary to popular belief. I have enjoyed building regenerative receivers for a long time. They are very simple to construct and they seem to suck the signals right out of the "ether". For information regarding the theory of operation check out some of the links at the bottom of this page. "Ancient" radio books are my favorite sources for regenerative receiver theory as the information hasn't been muddled with in the name of cheapness or to promote the super heterodyne:-) I will provide more information as time allows.
I built this particular design on a piece of double-sided copper circuit board using ugly construction in one evening and was utterly amazed at how well it performed. There was very little frequency drift even with the open air construction. Strong AM shortwave broadcast stations didn't pull the frequency or cause any noticeable blocking effects. It certainly took two hands to operate this rig effectively but with a little practice it became a pleasure to use. After working out the details I drew the schematic and have been planning on building the rig in a more robust fashion.
The main impetus for designing and building this rig was to improve upon previous regenerative receivers that I had built while still keeping the circuit simple. I don't claim any of the design concepts since most of them are older then me and my father. I simply studied the old theories and pieced it all together.
Below is a revised schematic of a regenerative receiver that I built. The original schematic I had posted was of my actual circuit based upon parts I had on hand and this revised schematic is the same circuit with more details and with a few component changes to values that are more readily available.
This receiver will receive AM, CW, SSB, and FM (with slope detection) and has excellent sensitivity, selectivity, and frequency stability. I once was able to cut the sidebands off of WWV at 10 MHz when the regeneration was tuned just below the point of oscillation. It covers the lower half of the HF spectrum and the coil can be easily modified to cover your favorite bands.
Construction is pretty straight forward. I would recommend point-to-point construction over a ground plane. I constructed mine over a piece unetched copper circuit card. This provides a very good ground plane and minimizes ground loop problems. The circuit card is securely fastened to a 1/2 inch thick piece of oak using small, brass wood screws. The oak is used as the receiver base.
Component values are not critical except those associated with the audio filter U2. The voltage regulator used for U1 can be any regulator to provide about 6 to 9 volts. MPF102 should be okay to use for Q1 and Q2 although I have not tried it. For the audio section after R5, you can use option A if you want to use the audio stage that I designed or you can use option B if you want to use your own audio amplifier. (Thanks go to Charles Kitchin, N1TEV for informing me that I still need 5k to 10k of resistor in series with the 0.1uF capacitor if an external audio amplifier is used.)
There is nothing sacred in my design about using the TLE2061 BiFET op amp in the audio stage. I just happened to have one in the junk box and it had very good specifications. The amp is used as a single stage bandpass filter. You will probably want to tailor the gain and cutoff frequencies to your own preferences. The midband gain is basically R8/R5. In this design the voltage gain is 18 (25dB). If you do change these resistors the capacitors C15 and C13 will have to be changed also to keep the same frequency response. This circuit has a calculated low frequency cutoff of 159 Hz and a high frequency cutoff of 2947 Hz. The low frequency cutoff point occurs when R5 = XC13, or fL = 1 / (2*pi*R5*C13). The high frequency cutoff point occurs when R8 = XC15, of fH = 1 / (2*pi*R8*C15).
C9 and C10 resonant in parallel with L2 and determine the frequency at which the receiver is tuned to. Choose C9 to select the desired operating frequency range and choose C10 to allow a band spread of the desired resolution. The throttle capacitor C8 should be whatever value provides a smooth control of the regeneration. I found 50pF to just about right. With the throttle capacitor in the drain circuit controlling the regeneration I can use a regulated supply which is a vast improvement over most potentiometer controlled designs. This keeps the supply voltage consistent at all levels of regeneration, therefore aiding in stability. Make sure the coil windings are wound in the correct orientation or else you may have problem obtaining regeneration (positive feedback).
You will notice that I do not use a volume potentiometer. I use the input attenuator capacitor to control the overall signal level. This method prevents input overload, frequency pulling, and other associated strong signal effects. The capacitor allows a very nice and smooth level control and it never gets scratchy, unlike some potentiometers. An insulated bushing is recommended for the input attenuator capacitor C6 to minimize residual signal pickup. Keep the coil assembly away from metals such as iron to minimize losses. A rigid aluminum front panel that has a low impedance connection to the circuit ground is recommended to prevent hand capacitance effects.
If you build a regenerative receiver send me a note and let me know how it goes!