The components that make up the oscillator section of the KnightSMiTe are Q1, R1, R2, R7, C1-C3, Y1 and L1. The output of the oscillator is coupled to the following stage (Q2) via capacitor C4. ** Q1 ** Q1, a 2N2222 in this design, can be any general purpose NPN bipolar transistor. In an oscillator, the active device (usually a transistor or FET) provides gain which must be greater than unity at the operating frequency. In addition, a feedback network couples a portion of the output signal back to the input. The energy coupled back must be in phase with and slightly larger than that which produced it at the input in order for the system to oscillate and run continuously. A resonator or resonating network used in the feedback path sets the frequency of oscillation. Oscillators are named for their designers and for a unique feature of their architecture... usually the manner in which feedback is introduced. The KnightSMiTe uses the popular Colpitts configuration identified by the capacitive voltage divider (C2 and C3) between the base and emitter used for feedback. A crystal, used to set the frequency of operation, behaves as an inductor and series resonates with the network that consists of C1, C2, C3 and L1. C1 is variable and provides a means to alter the resonant frequency of the network slightly. A crystal appears inductive only for a very small region of frequency (between series and parallel resonance) and capacitive elsewhere. Thus, the frequency of oscillation can only be altered to within the limits of the two resonance modes. In the KnightSMiTe, the oscillator runs continuously during both transmit and receive. In the transmitter, it supplies the signal which when amplified becomes the transmitted carrier. In the receiver, it serves as a local oscillator to Q2 where it mixes with the signal intercepted by the antenna to produce a difference frequency in the audio spectrum. That signal is then amplified by the audio amplifier U1 (the LM-386) to a sufficient level to drive a pair of low impedance headphones or a small speaker. ** R1 ** R1 is only used for a moment... at startup. It injects a turn-on bias boost into the base of Q1 to kick start the oscillator when power is first applied. In some cases it's not required, but having it guarantees oscillator startup. It should be a high value to minimize loading of the crystal which would otherwise serve to reduce tuning range or inhibit oscillation altogether. It should be low enough to provide a sufficient "surge" of base emitter current to get the oscillator up and running. ** C2 and C3 ** Once started, oscillation is sustained by feedback to the base of Q1 from the output taken from the emitter via the capacitive voltage divider consisting of C2 and C3. C2 works with C3 as an autotransformer (the primary and secondary share a common path - C3) to provide the positive emitter to base feedback (i.e. the output at the emitter is in phase with the input at the base) required for Q1 to sustain oscillation. The ratio of C2 to C3 establishes the amount of feedback. Too little feedback (C2 small relative to C3) will not adequately sustain oscillation. Too much feedback (C3 large relative to C2) will degrade oscillator stability and increase harmonic content at its output. An abrupt frequency shift or chirp may also result as the oscillator is keyed. ** R2 ** R2 provides DC degenerative feedback at the emitter of Q1 and sets the source impedance for both the feedback autotransformer (C2 and C3) and the output load presented by the input of Q2 via the coupling capacitor C4. Its value is somewhat non-critical. A larger value for R2 would increase the source impedance and make the oscillator sensitive to frequency pulling (called load pulling) due to the changing input impedance of the power amplifier when it's keyed. The result would be an objectionable "chirp" in the transmitted output waveform. Making R2 smaller would increase the current drain in Q1 making the oscillator less power efficient. Decreasing it significantly would cause an increase in Q1's operating temperature (high collector current), reduce its reliability and degrade frequency stability. The values of C2 and C3 would have to be scaled inversely with a change in value of R2 to maintain the same degree of feedback employed for sustained oscillation. Taken to an extreme this could result in spurious oscillations (undesired frequencies) or even destruction of the crystal due to excess drive. ** R7 ** R7 is a zero ohm jumper providing a return path to ground for the oscillator. It was originally implemented to facilitate keeping the wiring on one side of the board. That need was eliminated but we decided that keeping the jumper would provide flexibility to users who might wish to pursue optional uses of the transceiver. If this jumper is removed, the oscillator can be keyed on and off by connecting a key across this junction. This may be desired if only the transmitter section of the KnightSMiTe is used and an independent receiver is employed. Removing R7 also permits independent use of the audio amplifier by disabling Q1 and Q2. ** C1 ** C1 works in conjunction with L1 to vary the operating frequency of the KnightSMiTe. To some degree, its value is critical. Maximum adjustment range is realized with a capacitor having a low minimum value capacitance (on the order of 3 pfd or less, while maximum values greater than 50 pfd will do little to improve the frequency pulling range of the oscillator. Although C1 can be used to some extent as an RIT (receive incremental tuning), SMT capacitors are not sufficiently robust to survive in this environment. A varactor makes a better choice for an RIT control but will require a carefully regulated bias supply and increase the complexity of the design. Oscillator frequency is highest for both receive and transmit when C1 is at minimum capacitance. The frequency shift between transmit and receive however is smallest at this end of its range... on the order of 100 to 200 Hz . With C1 at maximum capacitance (lowest frequency), the shift between transmit and receive increases to approximately 500 to 700 Hz. In both cases the oscillator frequency increases when switching from receive to transmit. This means that when a station zero's your frequency, you will hear a tone between 100 and 700 Hz depending on where C1 is set. Since the KnightSMiTe frequency increases when switching to transmit, a responding station will likely be above your receive oscillator (i.e. in your upper sideband) by the same amount. The difference frequency is the tone you hear. ** Operators Note ** Once contact is made, ask the station to move his TX frequency up 500 Hz or so. This will produce a more pleasing tone in your receiver. Don't ask them to QSY or you may lose them. QSY generally implies "Both" stations change frequency. You only want "them" to move while you stay put. If he/she decreases their TX frequency, you'll receive a lower pitch until they move far enough into your lower sideband that you hear them emerge once again.. ** L1 ** As with C1, the value of L1 is somewhat critical. Reducing it will reduce the tuning range of the oscillator, while increasing it may result in the oscillator ceasing operation. L1 provides an inductive reactance in series with the crystal Y1 to compensate for (cancel) the contribution of the series modal capacitance of the crystal at series resonance. This technique enables the frequency of crystal oscillators to be "pushed" up in frequency while C1 serves to offset the reactance of L1 and "pull" the frequency lower. Together they maximize the range at which the frequency can be shifted. The KnightSMiTe is tunable in excess of 1.5 kHz in this manner. ** Y1 ** The Crystal (Y1) controls the operating frequency of the KnightSMiTe. Any 80 meter Crystal should work well in this transceiver. A crystal series resonant at 3686.4 kHz is provided with the KnightSMiTe kit. ** C4 ** C4 provides interstage coupling from the output of the oscillator to the input of Q2. It needs to be sufficiently large to permit Q2 to be driven to full output power yet small enough to prevent the loading presented by Q2's input from pulling the oscillator frequency during transmit. Its value is not critical but should remain small consistent with the drive requirement of Q2 for efficient Class-C operation at the desired operating supply voltage.
If you are interested in the Knightlites and Amateur Radio contact Paul at