Oscillators

Introduction to Feedback:

The phenomenon of feeding a portion of the output signal back to the

input circuit is known as feedback. The effect results in a dependence
between the output and the input and an effective control can be
obtained in the working of the circuit. Feedback is of two types.
• Negative Feedback
• Positive Feedback
Negative or Degenerate feedback:

• In negative feedback, the feedback energy (voltage or current), is out

of phase with the input signal and thus opposes it.
• Negative feedback reduces gain of the amplifier. It also reduce
distortion, noise and instability.
• This feedback increases bandwidth and improves input and output
impedances.
• Due to these advantages, the negative feedback is frequently used in
amplifiers.
Negative Feedback
Positive or regenerate feedback:
• In positive feedback, the feedback energy (voltage or currents), is in
phase with the input signal and thus aids it. Positive feedback
increases gain of the amplifier also increases distortion, noise and
instability.
• Because of these disadvantages, positive feedback is seldom employed
in amplifiers. But the positive feedback is used in oscillators.
Positive Feedback
In the above figure, the gain of the amplifier is represented as A. The
gain of the amplifier is the ratio of output voltage Vo to the input
voltage Vi. The feedback network extracts a voltage Vf = β Vo from
the output Vo of the amplifier.

This voltage is subtracted for negative feedback, from the signal

voltage Vs. Now,
Vi=Vs + Vf =Vs+βVo

The quantity β = Vf/Vo is called as feedback ratio or feedback

fraction.
The output Vo must be equal to the input voltage (Vs + βVo) multiplied
by the gain A of the amplifier.
Hence,
(Vs+βVo)A=Vo

AVs+AβVo=Vo

AVs=Vo(1-Aβ)

Vo/Vs=A/(1-Aβ)

Therefore, the gain of the amplifier with feedback is given by

Af =A/(1-Aβ)
Comparison Between Positive and Negative Feed Back:
S.No. Negative Feedback Positive Feedback

Feedback energy is out

Feedback energy is in
1. phase with their input signal
phase with the input signal.
Gain of the amplifier
2. Gain of the amplifier
decreases
increases
3. Gain stability increases Gain stability decreases
4. Noise and distortion Noise and distribution
5. decreases. increases.
Decreases bandwidth
6. Increase the band width
Used in Oscillators
Used in amplifiers
Barkhausen Criteria:
• The condition Aβ=1 is known as Barkhausen criteria. It implies
(1) Magnitude of the loop gain Aβ = 1
(2) Phase shift over the loop = 0 0r 360 degrees.

• Frequency of the noise in the amplifier for which this criteria are
satisfied, is the frequency of oscillations.

• By applying this criteria, we can even find the values of transistor

parameters, like gain, required for setting in oscillations.
Type of Oscillator:
There are many types of oscillators, but can broadly be classified into two
main categories – Harmonic Oscillators (also known as Linear Oscillators)
and Relaxation Oscillators.
In a harmonic oscillator, the energy flow is always from the active
components to the passive components and the frequency of oscillations is
decided by the feedback path.
Whereas in a relaxation oscillator, the energy is exchanged between the
active and the passive components and the frequency of oscillations is
determined by the charging and discharging time-constants involved in the
process. Further, harmonic oscillators produce low-distorted sine-wave
outputs while the relaxation oscillators generate non-sinusoidal (saw-
tooth, triangular or square) wave-forms.
The main types of Oscillators include:
RC Oscillators
•Wien Bridge Oscillator
•RC Phase Shift Oscillator
LC Oscillators
•Hartley Oscillator
•Colpitts Oscillator
•Clapp Oscillator
Crystal Oscillators
Wien Bridge Oscillator:

A Wien-Bridge Oscillator is a type of phase-shift oscillator which is

based upon a Wien-Bridge network comprising of four arms connected
in a bridge fashion. Here two arms are purely resistive while the other
two arms are a combination of resistors and capacitors. In particular,
one arm has resistor and capacitor connected in series (R1 and C1) while
the other has them in parallel (R2 and C2). This indicates that these two
arms of the network behave identical to that of high pass filter or low
pass filter, mimicking the behavior of the circuit shown by Figure.
In this circuit, at high frequencies, the reactance of the capacitors C1 and
C2 will be much less due to which the voltage V0 will become zero as R2
will be shorted. Next, at low frequencies, the reactance of the capacitors
C1 and C2 will become very high.
However even in this case, the output voltage V0 will remain at zero only,
as the capacitor C1 would be acting as an open circuit. This kind of
behavior exhibited by the Wien-Bridge network makes it a lead-lag
circuit in the case of low and high frequencies, respectively.
Wien Bridge Oscillator Frequency Calculation:
Nevertheless, amidst these two high and low frequencies, there exists a particular frequency
at which the values of the resistance and the capacitive reactance will become equal to each
other, producing the maximum output voltage. This frequency is referred to as resonant
frequency. The resonant frequency for a Wein Bridge Oscillator is calculated using the
following formula:
Further, at this frequency, the phase-shift between the input and the
output will become zero and the magnitude of the output voltage will
become equal to one-third of the input value. In addition, it is seen that
the Wien-Bridge will be balanced only at this particular frequency.
In the case of Wien-Bridge oscillator, the Wien-Bridge network of Figure 1 will be
used in the feedback path as shown in Figure 2. The circuit diagram for a Wein
Oscillator using a BJT (Bipolar Junction Transistor) is shown below:
In these oscillators, the amplifier section will comprise of two-stage
amplifier formed by the transistors, Q1 and Q2, wherein the output of Q2 is
back-fed as an input to Q1 via Wien-Bridge network (shown within the
blue enclosure in the figure). Here, the noise inherent in the circuit will
cause a change in the base current of Q1 which will appear at its collector
point after being amplified with a phase-shift of 180o.
This is fed as an input to Q2 via C4 and gets further amplified and appears
with an additional phase-shift of 180o. This makes the net phase-difference
of the signal fed back to the Wien-Bridge network to be 360o, satisfying
phase-shift criterion to obtain sustained oscillations.
However, this condition will be satisfied only in the case of resonant
frequency, due to which the Wien-Bridge oscillators will be highly
selective in terms of frequency, leading to a frequency-stabilized design.
Wien-bridge oscillators can even be designed using Op-Amps as a part of
their amplifier section, as shown by Figure 3. However it is to be noted
that, here, the Op-Amp is required to act as a non-inverting amplifier as the
Wien-Bridge network offers zero phase-shift. Further, from the circuit, it is
evident that the output voltage is fed back to both inverting and non-
inverting input terminals.
At resonant frequency, the voltages applied to the inverting and non-
inverting terminals will be equal and in-phase with each other. However,
even here, the voltage gain of the amplifier needs to be greater than 3 to
start oscillations and equal to 3 to sustain them. In general, these kind of
Op-Amp-based Wien Bridge Oscillators cannot operate above 1 MHz due
to the limitations imposed on them by their open-loop gain.
Wien-Bridge networks are low frequency oscillators which are used to
generate audio and sub-audio frequencies ranging between 20 Hz to 20 KHz.
Further, they provide stabilized, low distorted sinusoidal output over a wide
range of frequency which can be selected using decade resistance boxes. In
addition, the oscillation frequency in this kind of circuit can be varied quite
easily as it just needs variation of the capacitors C1 and C2. However these
oscillators require large number of circuit components and can be operated up
to a certain maximum frequency only.
RC Phase Shift Oscillator:

RC phase-shift oscillators use resistor-capacitor (RC)

network (Figure 1) to provide the phase-shift required by
the feedback signal. They have excellent frequency
stability and can yield a pure sine wave for a wide range
of loads.
Ideally a simple RC network is expected to have an output which leads the input by 90o.
However, in reality, the phase-difference will be less than this as the capacitor used in the
circuit cannot be ideal. Mathematically the phase angle of the RC network is expressed as

Where, XC = 1/(2πfC) is the reactance of the capacitor C and R is the resistor. In

oscillators, these kind of RC phase-shift networks, each offering a definite phase-shift can
be cascaded so as to satisfy the phase-shift condition led by the Barkhausen Criterion.
One such example is the case in which RC phase-shift oscillator is formed by
cascading three RC phase-shift networks, each offering a phase-shift of 60o, as
shown by Figure 2.
Here the collector resistor Rc limits the collector current of the transistor, resistors R1
and R (nearest to the transistor) form the voltage divider network while the emitter
resistor RE improves the stability. Next, the capacitors CE and Co are the emitter by-
pass capacitor and the output DC decoupling capacitor, respectively. Further, the
circuit also shows three RC networks employed in the feedback path.

This arrangement causes the output waveform to shift by 180o during its
course of travel from output terminal to the base of the transistor. Next, this signal
will be shifted again by 180o by the transistor in the circuit due to the fact that the
phase-difference between the input and the output will be 180o in the case of common
emitter configuration. This makes the net phase-difference to be 360o, satisfying the
phase-difference condition.
One more way of satisfying the phase-difference condition is to use four RC
networks, each offering a phase-shift of 45o. Hence it can be concluded that
the RC phase-shift oscillators can be designed in many ways as the number
of RC networks in them is not fixed. However it is to be noted that, although
an increase in the number of stages increases the frequency stability of the
circuit, it also adversely affects the output frequency of the oscillator due to
the loading effect.
The generalized expression for the frequency of oscillations produced by a
RC phase-shift oscillator is given by

Where, N is the number of RC stages formed by the resistors R and the

capacitors C.
Further, as is the case for most type of oscillators, even the RC phase-shift
oscillators can be designed using an OpAmp as its part of the amplifier
section (Figure 3). Nevertheless, the mode of working remains the same
while it is to be noted that, here, the required phase-shift of 360o is offered
collectively by the RC phase-shift networks and the Op-Amp working in
inverted configuration.
Further, it is to be noted that the frequency of the RC phase-shift
oscillators can be varied by changing either the resistors or the
capacitors. However, in general, the resistors are kept constant while the
capacitors are gang-tuned. Next, by comparing the RC phase-shift
oscillators with LC oscillators, one can note that, the former uses more
number of circuit components than the latter one. Thus, the output
frequency produced from the RC oscillators can deviate much from
the calculated value rather than in the case of LC oscillators.
Nevertheless, they are used as local oscillators for synchronous
receivers, musical instruments and as low and/or audio-frequency
generators.
LC- Oscillators

LC Oscillator Z1 Z2 Z3

Hartley L L C
Oscillator

Colpitts C C L
Oscillator

Clapp C C L-C
Oscillator
Hartley Oscillator:
Hartley Oscillator is a type of harmonic oscillator which was invented by Ralph Hartley in
1915. These are the Tuned Circuit Oscillators which are used to produce the waves in the
range of radio frequency and hence are also referred to as RF Oscillators. Its frequency of
oscillation is decided by its tank circuit which has a capacitor connected in parallel with the
two serially connected inductors, as shown by Figure 1.
Here the RC is the collector resistor while the emitter resistor RE forms the
stabilizing network. Further the resistors R1 and R2 form the voltage
divider bias network for the transistor in common-emitter CE
configuration. Next, the capacitors Ci and Co are the input and output
decoupling capacitors while the emitter capacitor CE is the bypass
capacitor used to bypass the amplified AC signals. All these components
are identical to those present in the case of a common-emitter amplifier
which is biased using a voltage divider network. However, Figure 1 also
shows one more set of components viz., the inductors L1 and L2 and the
capacitor C which form the tank circuit (shown in red enclosure).
 On switching ON the power supply, the transistor starts to conduct, leading to an
increase in the collector current, IC which charges the capacitor C. On acquiring the
maximum charge feasible, C starts to discharge via the inductors L1 and L2. This
charging and discharging cycles result in the damped oscillations in the tank circuit.
The oscillation current in the tank circuit produces an AC voltage across the inductors
L1 and L2 which are out of phase by 180o as their point of contact is grounded.

Further from the figure, it is evident that the output of the amplifier is applied
across the inductor L1 while the feedback voltage drawn across L2 is applied to the
base of the transistor. Thus one can conclude that the output of the amplifier is in-
phase with the tank circuit’s voltage and supplies back the energy lost by it while the
energy fed back to amplifier circuit will be out-of-phase by 180o. The feedback voltage
which is already 180o out-of-phase with the transistor is provided by an additional
180o phase-shift due to the transistor action. Hence the signal which appears at the
transistor’s output will be amplified and will have a net phase-shift of 360o.
At this state, if one makes the gain of the circuit to be slightly greater than the feedback ratio given by

(if the coils are wound on the same core with M indicating the mutual inductance)

then the circuit generates the oscillations which can be sustained by maintaining the gain of the circuit to be equal
to that of the feedback ratio. This causes the circuit in Figure 1 to act as an oscillator as it would then satisfy both
the conditions of the Barkhausen criteria.
The frequency of such an oscillator is given as

Where,
Hartley oscillators are available in many different configurations including
series-or shunt-fed, common-emitter or common-base configured,
and BJT (Bipolar Junction Transistor) or FET (Field Effect Transistor)
amplifier based. Further it is to be noted that the transistor-based amplifier
section of Figure 1 can even be replaced by an amplifier of any other kind like
that of an inverting amplifier formed by an Op-Amp as shown by Figure 2. The
working of this kind of oscillator is similar to that of the one shown earlier.
However, here, the gain of the oscillator can be individually adjusted using the
feedback resistor Rf due to the fact that the gain of the inverting amplifier is
given as -Rf / R1. From this, it can be noted that, in this case, the gain of the
circuit is less dependent on the circuit elements of the tank circuit. This
increases the stability of the oscillator in terms of its frequency.
Hartley Oscillators are advantageous as they are easy-tunable circuits with a very
few components including a capacitor and either two inductors or a tapped coil. This
results in a constant amplitude output throughout its wide operational frequency range
which typically ranges from 20 KHz to 30 MHz. However, this kind of oscillator is
not suitable for low frequency as it would result in a large-sized inductor which
makes the circuit bulky. Further, the output of Hartley Oscillator has high content
of harmonics in it and hence does not suit for the applications which require pure sine
wave.
Colpitts Oscillator:

Colpitts Oscillator is a type of LC oscillator which falls

under the category of Harmonic Oscillator and was invented
by Edwin Colpitts in 1918. Figure 1 shows a typical Colpitts
oscillator with a tank circuit in which an inductor L is
connected in parallel to the serial combination of capacitors C1
and C2 (shown by the red enclosure).
 Other components in the circuit are the same as that found in the case of
common-emitter CE which is biased using a voltage divider network i.e. RC is the
collector resistor, RE is the emitter resistor which is used to stabilize the circuit
and the resistors R1 and R2 form the voltage divider bias network. Further, the
capacitors Ci and Co are the input and output decoupling capacitors while the
emitter capacitor CE is the bypass capacitor used to bypass the amplified AC
signals.

Here, as the power supply is switched ON, the transistor starts to

conduct, increasing the collector current IC due to which the capacitors C1 and C2
get charged. On acquiring the maximum charge feasible, they start to discharge
via the inductor L. During this process, the electrostatic energy stored in the
capacitor gets converted into magnetic flux which in turn is stored within the
inductor in the form of electromagnetic energy. Next, the inductor starts to
discharge which charges the capacitors once again. Likewise, the cycle continues
which gives rise to the oscillations in the tank circuit.
 Further the figure shows that the output of the amplifier appears across
C1 and thus is in-phase with the tank circuit’s voltage and makes-up for the
energy lost by re-supplying it. On the other hand, the voltage feedback to the
transistor is the one obtained across the capacitor C2, which means the
feedback signal is out-of-phase with the voltage at the transistor by 180o. This is
due to the fact that the voltages developed across the capacitors C1 and C2 are
opposite in polarity as the point where they join is grounded. Further, this
signal is provided with an additional phase-shift of 180o by the transistor
which results in a net phase-shift of 360o around the loop, satisfying the phase-
shift criterion of Barkhausen principle.
At this state, the circuit can effectively act as an oscillator
producing sustained oscillations by carefully monitoring the
feedback ratio given by (C1 / C2). The frequency of such a
Colpitts Oscillator depends on the components in its tank
circuit and is given by

Where, the Ceff is the effective capacitance of the

capacitors expressed as
As a result, these oscillators can be tuned either by varying their
inductance or the capacitance. However the variation of L does
not yield a smooth variation. Hence they are usually tuned by
varying the capacitances which are generally ganged, due to
which a change in any one of them changes both of them.
Nevertheless, the process is tedious and requires special large-
valued capacitor. Thus, the Colpitts oscillators are seldom
preferred in the applications where in the frequency varies but are
more popular as fixed frequency oscillators due to their simple
design. Further they offer better stability in comparison with the
Hartley Oscillators as they are exempted from the mutual
inductance effect present in-between the two inductors of the
latter case.
 Apart from the BJT-based Colpitts Oscillator shown, they are also
realizable using valves or FET (Field Effect Transistor) or Op-Amp. Figure 2
shows such a Colpitts oscillator which uses an Op-Amp in inverting
configuration in its amplifier section while the tank circuit remains similar to that
in the case of Figure 1. This kind of circuit functions almost analogous to that of
the one explained earlier. However, here the gain of the oscillator can be adjusted
individually just by using the feedback resistor Rf, as the gain of the inverting
amplifier is given as -Rf / R1. From this, it can be noted that, in this case, the gain
of the circuit is less dependent on the circuit elements of the tank circuit.

Typically, the operating frequency of the Colpitts oscillators ranges from

20 KHz to 300 MHz. However they can even be used for microwave applications
as their capacitors provide low reactance path for the high-frequency signals. This
results in better frequency stability as well as a better sinusoidal output waveform.
Moreover, they are also extensively used as surface acoustical wave (SAW)
resonators, sensors and in mobile and communication systems.
Clapp Oscillator:
Clapp oscillator is a variation of Colpitts oscillator in which an additional
capacitor (C3) is added into the tank circuit to be in series with the inductor
in it, as shown by Figure 1.
Apart from the presence of an extra capacitor, all other components and
their connections remain similar to that in the case of Colpitts oscillator.

Hence, the working of this circuit is almost identical to that of the

Colpitts, where the feedback ratio governs the generation and
sustainability of the oscillations. However the frequency of oscillation in
the case of Clapp oscillator is given by
Usually the value of C3 is chosen to be much smaller than the other two
capacitors. This is because, at higher frequencies, smaller the C3, larger will be
the inductor, which eases the implementation as well as reduces the influence of
stray inductance. Nevertheless, the value of C3 is to be chosen with utmost care.
This is because, if it is chosen to be very small, then the oscillations will not be
generated as the L-C branch will fail to have a net inductive reactance.
However, here it is to be noted that when C3 is chosen to be smaller in
comparison with C1 and C2, the net capacitance governing the circuit will be
more dependent on it.
Thus the equation for the frequency can be approximated as
 Further, the presence of this extra capacitance will make the Clapp
oscillator preferable over Colpitts when there is a need to vary the frequency as is
the case with Variable Frequency Oscillator (VCO). The reason behind this can
be explained as follows.

In the case of Colpitts oscillator, the capacitors C1 and C2 need to be

varied inorder to vary its frequency of operation. However during this process,
even the feedback ratio of the oscillator changes which inturn affects its output
waveform. One solution to this problem is to make both C1 and C2 to be fixed in
nature while achieve the variation in frequency using a separate variable
capacitor. As could be guessed, this is what the C3 does in the case of Clapp
oscillator, which inturn makes it more stable over Colpitts interms of frequency.
The frequency stability of the circuit can be even more increased by enclosing the
entire circuit in a chamber with constant temperature and by using a Zener diode
to ensure constant supply voltage.
 In addition, it is to be noted that the values of the capacitors C1 and
C2 are prone to the effect of stray capacitances unlike that of C3. This means
that the resonant frequency of the circuit would be affected by the stray
capacitances if one had a circuit with just C1 and C2, as in the case of Colpitts
oscillator. However if there is C3 in the circuit, then the changes in the values
of C1 and C2 would not vary the resonant frequency much, as the dominant
term would then be C3.

Next, it is seen that the Clapp oscillators are comparatively compact

as they employ a relatively small capacitor to tune the oscillator over a wide
frequency band. This is because, here, even a slight change in the value of the
capacitance varies the frequency of the circuit upto a great extent. Further they
exhibit high Q-factor with a high L/C ratio and lesser circulating current in
comparison with Colpitts oscillators. Lastly it is to be noted that these
oscillators are highly reliable and are hence preferred inspite of having a
limited range of frequency of operation.
Crystal Oscillators:
Crystal Oscillators can be designed by connecting the crystal into the circuit
such that it offers low impedance when operated in series-resonant mode
(Figure 2a) and high impedance when operated in anti-resonant or parallel
resonant mode (Figure 2b).
 In the circuits shown, the resistors R1 and R2 form the voltage divider network
while the emitter resistor RE stabilizes the circuit. Further, CE (Figure 2a) acts as an AC
bypass capacitor while the coupling capacitor CC (Figure 2a) is used to block DC
signal propagation between the collector and the base terminals. Next, the capacitors
C1 and C2 form the capacitive voltage divider network in the case of Figure 2b. In
addition, there is also a Radio Frequency Coil (RFC) in the circuits (both in Figure 2a
and 2b) which offers dual advantage as it provides even the DC bias as well as frees
the circuit-output from being affected by the AC signal on the power lines.

On supplying the power to the oscillator, the amplitude of the oscillations in

the circuit increases until a point is reached wherein the nonlinearities in the
amplifier reduce the loop gain to unity. Next, on reaching the steady-state, the
crystal in the feedback loop highly influences the frequency of the operating
circuit. Further, here, the frequency will self-adjust so as to facilitate the crystal to
present a reactance to the circuit such that the Barkhausen phase requirement is
fulfilled.
In general, the frequency of the crystal oscillators will be fixed to be
the crystal’s fundamental or characteristic frequency which will be
decided by the physical size and shape of the crystal. However, if the
crystal is non-parallel or of non-uniform thickness, then it might resonate
at multiple frequencies, resulting in harmonics. Further, the crystal
oscillators can be tuned to either even or odd harmonic of the
fundamental frequency, which are called Harmonic and Overtone
Oscillators, respectively. An example for this is the case where the
parallel resonance frequency of the crystal is decreased or increased by
adding a capacitor or an inductor across the crystal, respectively.
The typical operating range of the crystal oscillators is from 40 KHz to 100
MHz wherein the low frequency oscillators are designed using OpAmps
while the high frequency-ones are designed using the transistors (BJTs or
FETs). The frequency of oscillations generated by the circuit is decided
by the series resonant frequency of the crystal and will be unaffected by
the variations in supply voltage, transistor parameters, etc. As a result,
crystal oscillators exhibit high Q-factor with excellent frequency
stability, making them most suitable for high-frequency applications.
However care should be taken so as to drive the crystal with optimum power
only. This is because, if too much of power is delivered to the crystal, then
the parasitic resonances might be excited in the crystal which leads to
unstable resonant frequency. Further even its output waveform might be
distorted due to the degradation in its phase noise performance. Moreover it
can even result in the destruction of the device (crystal) due to overheat.
Applications:
Crystal oscillators are compact in size and are of low cost due to which
they are extensively used in electronic warfare systems, communication
systems, guidance systems, microprocessors, microcontrollers, space
tracking systems, measuring instruments, medical devices, computers, digital
systems, instrumentation, phase locked loop systems, modems, sensors, disk
drives, marine systems, telecommunications, engine controlling systems,
clocks, Global Positioning Systems (GPS), cable television systems, video
cameras, toys, video games, radio systems, cellular phones, timers, etc.
GATE Questions with Solutions
10. BJT wien bridge oscillator has angular frequency of oscillation in radians/sec equal to
𝟏 𝟔 𝟐 𝟏
1. 2. 3. 4.
𝟐𝑹𝟏 𝑪𝟏 𝑹𝟐 𝑪𝟐 𝑹𝟏 𝑪𝟏 𝑹𝟐 𝑪𝟐 𝑹𝟏 𝑪𝟏 𝑹𝟐 𝑪𝟐 𝑹𝟏 𝑪𝟏 𝑹𝟐 𝑪𝟐

Sol:

Q1 and Q2 provide 360° phase shift

Output is feedback to input through wein bridge. The bridge is balanced when z1 × z4 = z2 × z3
Here z1 = R1 + 1/sC1, z2 is R2 parallel with C2, z3 is R3, Z4 is R4.
1
On solving and equating imaginary part ω= rad/sec
𝑅1 𝐶1 𝑅2 𝐶2