COLPITTS OSCILLATOR

Both the Armstrong and the Hartley oscillators have a tendency to be unstable in
frequency because of junction capacitance. In comparison, the COLPITTS OSCILLATOR has
fairly good frequency stability, is easy to tune, and can be used for a wide range of
frequencies. The large value of split capacitance is in parallel with the junctions and
minimizes the effect on frequency stability.

The Colpitts oscillator is very similar to the shunt-fed Hartley oscillator, except
that two capacitors are used in the tank circuit instead of a tapped coil (figure 2-15).
The Hartley oscillator has a tap between two coils, while the Colpitts has a tap between
two capacitors. You can change the frequency of the Colpitts either by varying the
inductance of the coil or by varying the capacitance of the two capacitors in the tank
circuit. Notice that no coupling capacitor is used between the tank circuit and the base
of Q1. Capacitors C1 and C2 of the tank circuit are in parallel with the input and the
output interelement capacitance (capacitance between emitter, base, and collector) of the
transistor. Thus the input and the output capacitive effect can be minimized on the tank
circuit and better frequency stability can be obtained than with the Armstrong or the
Hartley oscillator.

Figure 2-15. - Colpitts oscillator.

Figure 2-16 shows a common-base Colpitts oscillator using a pnp transistor as the
amplifying device. Notice in this version of the Colpitts oscillator that regenerative
feedback is obtained from the tank circuit and applied to the emitter. Base bias is
provided by resistor R_{B} and R_{F}. Resistor R_{C} is the
collector load resistor. Resistor R_{E} develops the input signal and also acts as
the emitter swamping resistor. The tuned circuit consists of C1 and C2 in parallel with
the primary winding of transformer T1. The voltage developed across C2 is the feedback
voltage. Either or both capacitors may be adjusted to control the frequency. In the
common-base configuration there is no phase difference between the signal at the collector
and the emitter signal. Therefore, the phase of the feedback signal does not have to be
changed. When the emitter swings negative, the collector also swings negative and C2
charges negatively at the junction of C1 and C2. This negative charge across C2 is fed
back to the emitter. This increases the reverse bias on Q1. The collector of Q1 becomes
more negative and C2 charges to a negative potential. This feedback effect continues until
the collector of Q1 is unable to become any more negative. At that time the primary of T1
will act as a source because of normal tank circuit operation. As its field collapses, the
tank potential will reverse and C1 and C2 will begin to discharge. As C2 becomes less
negative, the reverse bias on Q1 decreases and its collector voltage swings in the
positive direction. C1 and C2 will continue to discharge and then charge in a positive
direction. This positive-going voltage across C2 will be fed back to the emitter as
regenerative feedback. This will continue until the field around the primary of T1
collapses. At that time the collector of Q1 will be at a maximum positive value. C1 and C2
will begin to discharge and the potential at their junction will become less positive.
This increases the reverse bias on Q1 and drives the collector negative, causing C1 and C2
to charge in a negative direction and to repeat the cycle.

Figure 2-16. - Common-base Colpitts oscillator.

Q.14 What is the identifying feature of a Colpitts oscillator?

RESISTIVE-CAPACITIVE (RC) FEEDBACK OSCILLATOR

As mentioned earlier, resistive-capacitive (RC) networks provide regenerative feedback
and determine the frequency of operation in RESISTIVE-CAPACITIVE (RC) OSCILLATORS.

The oscillators presented in this chapter have used resonant tank circuits (LC). You
should already know how the LC tank circuit stores energy alternately in the inductor and
capacitor.

The major difference between the LC and RC oscillator is that the frequency-determining
device in the RC oscillator is not a tank circuit. Remember, the LC oscillator can operate
with class A or C biasing because of the oscillator action of the resonant tank. The RC
oscillator, however, must use class A biasing because the RC frequency-determining device
doesn't have the oscillating ability of a tank circuit.

An RC FEEDBACK or PHASE-SHIFT oscillator is shown in figure 2-17. Components C1, R1,
C2, R2, C3, and R_{B} are the feedback and frequency-determining network. This RC
network also provides the needed phase shift between the collector and base.

Figure 2-17. - Phase-shift oscillator.

Phase-Shift Oscillators

The PHASE-SHIFT OSCILLATOR, shown in figure 2-17, is a sine-wave generator that uses a
resistive-capacitive (RC) network as its frequency-determining device.

As discussed earlier in the common-emitter amplifier configuration (figure 2-17), there
is a 180-degree phase difference between the base and the collector signal. To obtain the
regenerative feedback in the phase-shift oscillator, you need a phase shift of 180 degrees
between the output and the input signal. An RC network consisting of three RC sections
provides the proper feedback and phase inversion to provide this regenerative feedback.
Each section shifts the feedback signal 60 degrees in phase.

Since the impedance of an RC network is capacitive, the current flowing through it
leads the applied voltage by a specific phase angle. The phase angle is determined by the
amount of resistance and capacitance of the RC section.

If the capacitance is a fixed value, a change in the resistance value will change the
phase angle. If the resistance could be changed to zero, we could get a maximum phase
angle of 90 degrees. But since a voltage cannot be developed across zero resistance, a
90-degree phase shift is not possible.

With a small value of resistance, however, the phase angle or phase shift is less than
90 degrees. In the phase-shift oscillator, therefore, at least three RC sections are
needed to give the required 180-degree phase shift for regenerative feedback. The values
of resistance and capacitance are generally chosen so that each section provides about a
60-degree phase shift.

Resistors R_{B}, R_{F}, and R_{C} provide base and collector
bias. Capacitor C_{E} bypasses ac variations around the emitter resistor R_{E}.
Capacitors C1, C2, and C3 and resistors R1, R2, and R_{B} form the feedback and
phase-shifting network. Resistor R2 is variable for fine tuning to compensate for any
small changes in value of the other components of the phase-shifting network.

When power is applied to the circuit, oscillations are started by any random noise
(random electrical variations generated internally in electronic components). A change in
the flow of base current results in an amplified change in collector current which is
phase-shifted the 180 degrees. When the signal is returned to the base, it has been
shifted 180 degrees by the action of the RC network, making the circuit regenerative. View
(A) of figure 2-18 shows the amount of phase shift produced by C1 and R1. View (B) shows
the amount of phase shift produced by C2 and R2 (signal received from C1 and R1), and view
(C) shows the complete phase shift as the signal leaves the RC network. With the correct
amount of resistance and capacitance in the phase-shifting network, the 180-degree phase
shift occurs at only one frequency. At any other than the desired frequency, the
capacitive reactance increases or decreases and causes an incorrect phase relationship
(the feedback becomes degenerative). Thus, the oscillator works at only one frequency. To
find the resonant frequency (f_{r}) of an RC phase shift oscillator, use the
following formula:

where n is the number of RC sections.

Figure 2-18A. - Three-section, phase-shifting RC network. PHASE-SHIFT NETWORK C1 AND R1

Figure 2-18B. - Three-section, phase-shifting RC network. PHASE-SHIFT NETWORK C2 AND R2

Figure 2-18C. - Three-section, phase-shifting RC network. PHASE-SHIFT NETWORK C3 AND R_{B}

A high-gain transistor must be used with the three-section RC network because the
losses in the network are high. Using more than three RC sections actually reduces the
overall signal loss within the network. This is because additional RC sections reduce the
phase shift necessary for each section, and the loss for each section is lowered as the
phase shift is reduced. In addition, an oscillator that uses four or more RC networks has
more stability than one that uses three RC networks. In a four-part RC network, each part
shifts the phase of the feedback signal by approximately 45 degrees to give the total
required 180-degree phase shift.

Q.15 Which components provide the regenerative feedback signal in the phase-shift
oscillator?

Q.16 Why is a high-gain transistor used in the phase-shift oscillator?

Q.17 Which RC network provides better frequency stability, three-section or four-section?