IEEE Circuits and Systems Magazine - Q4 2020 - 10

Today, the common definition of the Colpitts oscillator is an oscillator where
the tank in the feedback is made of an inductor in parallel with a splitted
capacitor, forming a r shaped circuit.

very high frequency application, largely due to better
high frequency properties as stated in [13]. In the beginning of the 20th century researches realized the potential of using piezoelectric crystals in oscillator designs
[14], [15]. In 1924 George W. Pierce [16] introduced his
single vacuum tube crystal oscillator circuits [17] simply
by adding a crystal to the feedback loop of a vacuum
tube. In his publication he mentions only the Hartley oscillator used in a calibration setup and not the Colpitts
despite the fact that his circuit reminds of a Colpitts.
The Colpitts is more suited for crystal oscillators than
Hartley due to the arrangement of the tank where the
single inductor in the feedback is normally replaced by
a Crystal, something researchers and circuit designers
realized early on [18]. Various literature today, place
the Pierce oscillator in the same family as the Colpitts
[19]. Pierce later patented his design [20] which was approved in 1934.
In the beginning of 1940s, the vacuum tube cathodefollower circuits were first proposed. [21] and [22] describe some of the first Colpitts oscillator based on the
cathode-follower circuits, which can be compared to a
modern common drain (collector) design, and regarded
as the first circuits based on this configuration. A major improvement in the cathode-follower oscillators was
proposed by Clapp [23] in 1948. In his design, he added
a capacitor in series with the inductor in the feedback.
He explains this modification as a way of improving the
stability of the oscillator.
Early crystal Colpitts oscillators suffered from low
precision at high frequencies since they were operated
at their inductive region, so Butler proposed a series of
oscillators to drive the crystals at their series resonance

Vout
L

C1
(a)

C2 C2

Vout

C1
(b)

Vout
L

(c)

Figure 5. Colpitts Topologies, (a) the inductor is connected
to the input, (b) the inductor is connected to the output and
(c) the inductor is connected between the input and the output and the amplifier needs to be inverting.
10â

[24], [25]. In his publication, he uses the Hartley oscillator to show his new circuits, even though his circuits
can easily be implemented using Colpitts oscillators.
After the introduction of the bipolar transistor by
Bell Telephone Laboratories in 1948 [26], Colpitts oscillators were among the first circuits that were adopted
to the new technology [27]-[30]. Since the transistors
were similar to vacuum tubes in operation, this task became relatively easy. Bipolar gain stages can be realized
in three different ways depending on the location of the
common node for the input and the output, akin to the
implementation using vacuum tubes. These are Common base, Common collector and Common emitter and
the Colpitts oscillator was adopted to all of these configurations [31]. It is worth mentioning that the original
design by Hartley and Colpitts has similarity with the
common emitter (source) as shown in Fig. 2.
Today, the common definition of the Colpitts oscillator is an oscillator where the tank in the feedback is
made of an inductor in parallel with a splitted capacitor,
forming a r shaped circuit, independent of the amplifier
configuration. We follow this definition of Colpitts oscillator in this paper. This definition is also affirmed based
on our review of the papers presented on the topic in the
last hundred years. We recognize that a classification of
r type feedback oscillators presented in [32] from 1963,
has more limitations than our definition.
Fig. 5 shows the three possible ways of connecting
the feedback in a Colpitts configuration. The oscillation
frequency of the Colpitts oscillators in Fig. 5, if we ignore
the parasitics in the circuit, is equal to the resonance
frequency of the LC tank and given by:

ft =

L C1 C2
C1 + C2

(1)

In circuits a) and b) the inductor is connected between ground and either the input or the output. In
these two configurations the amplifier is non-inverting,
while in the case of circuit in c) the amplifier is inverting.
Frerking, in his 1978 book on oscillators refers to the
Fig. 5 a) as Colpitts, b) as Clapp and c) as Pierce. However,
this classification is not coherent with the original circuit
proposed by Colpitts, which is similar of the one in Fig. 5(c),
while the circuit proposed by Pierce is a crystal version of the same topology. All the amplifiers presented
in Fig. 5 can be realized by vacuum tubes, MOSFETs or

IEEE CIRCUITS AND SYSTEMS MAGAZINE

FOURTH QUARTER 2020

IEEE Circuits and Systems Magazine - Q4 2020

Table of Contents for the Digital Edition of IEEE Circuits and Systems Magazine - Q4 2020