Can you get an older copy of the ARRL Handbook (pre 1990 or so)? It's loaded with oscillator circuits.Please help me design hartley and collpits oscillator for frequency 100KHz using transistor BC107A.
.....we come to that sticky wicket of loaded vs unloaded Q. We need to extract a certain amount of power from our tank circuit in order to maintain oscillation. Depending on the input impedance of our active device (in the case of a mosfet, the problem isn't as severe), we are going to LOAD DOWN our tank circuit with the input admittance of the device in question. BJT's have rather low input impedances...you have to take this into consideration when designing the tank circuit. By using LARGE values of C, you tend to "swamp out" the low impedance of the transistor with a known value of capacitance. This does a couple of things for you. It maintains a high overall Q, necessary for frequency stability....and it also bypasses a large part of the variable capacitance "Varactor action" of the emitter base junction....ordinarily a major contributor of phase noise. Back when I first started building oscillators, nobody even knew what phase noise was...now it's known to be a major factor in communications quality...(or lack thereof).I can give you a few PRACTICAL pointers on oscillator building....from my many years in the trenches designing (or trying to, anyway!) stable VFO's for amateur radio transmitters.
For Colpitts oscillators you want a very large C/L ratio in your tank circuit. This may seem counterintuitive based on the conventional thinking of circuit Q, where the more reactance you have compared to circuit resistance, the more of it you have. However......
(Continued next post)......
by Jake Hertz
by Don Wilcher