I recently designed and put together the following super-essential regen receiver:
Rc,Re and Gain are potentiometers only so that I could get the biasing corrected. Sure enough the receiver works and picks up around 5 or so stations (here where I live, Northern Germany, the reception is not optimal).
First of all, I have two questions on this design:
-
Which considerations go into determining the ratio between antenna/tuning coil and tickler coil turns? On a related note, what is a good number of turns if I would want to separate the antenna coil and the tuning coil?
-
When I go from a very loud station to a weaker station there is a defined amount of time in which the sound is low and then rises slowly. I suppose is one of the capacitors that needs to readjust to a different voltage but which one is the main suspet?
Secondly I would like to improve the design but I'd like to keep it "modular". What I mean is that I want to add pieces to it to improve it without having to redesign the circuit from scratch. In this spirit, the tuning capacitor, the coil and the regen control are attached to the "main circuit" through crocodile clips, so that they can be exchanged or additional parts can be put inbetween or instead. So my question is, what would be good additions to the design?
Lastly, ideally I would like to have a multi-tap setup to receive multiple bands. Will the tickler coil work indipendently from which tap I take the signal if it sits on top or near the beginning and the first tap of the "main" coil (I don't know if its clear but I can try to paint it)?
As a final note, I am aware of the designs of Kitchin but unfortunately I don't have a good and cheap source of JFETs so I will stick with the diode-resistor-capacitor detector.
Sorry for the long post!
Best Answer
There are a couple of things that could be improved:
The tuning coil needs matching to the input impedance of the transistor - as shown the transistor will heavily damp the resonance causing the selectivity to be low. The normal way is to have a tap on the coil - probably about 1/5 to 1/10 of the way up from ground.
You need a path for DC at the input of the detector diode between C3 and the diode. A 10 k resistor would be a reasonable starting point. Currently it is relying on the leakage current of the diode.
The RF will be rectified by the diode and charge up C3 such the the diode only just conducts - this could be the delay you are seeing. C3 doesn't need to be anywhere near as large as 10uF - it only has to pass the RF so 0.01uF or 1000pF would be my first choice. You could also add a small amount of forward bias to the detector with a high resistor to the +9v - maybe 470K or so would be a starting point.
You could also use a voltage double detector by using a diode instead of a resistor to ground.
There are many web sites with example circuits - there are also many in old magazines from the 1950's and 60s. Many are online now - one particularly good effort is at American Radio History. One of my favorite magazines for this type of circuit can be found in Radio Constructor - an English magazine - Radio Constructor