18EC42 Analog Electronics Module-2

2.1 Basic MOSFET Amplifier

Configurations
FET as an amplifier, There , found that almost-
linear amplification can be obtained biasing the
MOSFET at an appropriate point in its saturation
region of operation and by keeping the signal vgs,
small. We then took a closer look at the small-
signal operation of the MOSFET in and developed
circuit models to represent the transistor, thus
facilitating the determination of amplifier
parameters such as voltage gain and input and shows the common-drain (CD) or grounded-drain
output resistances. amplifier. It is obtained by connecting the drain
terminal to ground, applying the input voltage
There are three basic configurations for connecting
signal vi between gate and ground, and taking the
the MOSFET as an amplifier. Each of these
output voltage signal between the source and
configurations is obtained by connecting one of the
ground, across a load resistance RD.
three MOSFET terminals to ground
In the circuit of Figure (a) 2.1.1 Characterizing Amplifiers

Above figure shows an amplifier fed with a signal

source having an open-circuit voltage vsig and an
the source terminal is connected to ground, the internal resistance Rsig.The amplifier is shown with
input voltage signal is applied between the gate and a load resistance RL connected to the output
ground, and the output voltage signal , is taken terminal. Here, RL, can be an actual load resistance.
between the drain and ground, across the resistance
RD. This configuration, therefore, is called the
grounded-source or common-source (CS)
amplifier. It is by far the most popular MOS
amplifier configuration
The common-gate (CG) or grounded-gate amplifier
is shown in Figure (b).

Above figure shows the amplifier circuit with the

amplifier block replaced by its equivalent-circuit
model. The input resistance Rin represents the
loading effect of the amplifier input on the signal
source. It is found from
𝑣𝑖
𝑅𝑖𝑛 =
𝑖𝑖
and together with the resistance Rsig forms a voltage
It is obtained by connecting the gate to ground, divider that reduces vsig to the value vi that appears
applying the input vi between the source and at the amplifier input,
ground, and taking the output across the resistance
RD, connected between the drain and ground 𝑅𝑖𝑛
𝑣𝑖 = 𝑣
𝑅𝑖𝑛 + 𝑅𝑠𝑖𝑔 𝑠𝑖𝑔
Figure (c)
18EC42 Analog Electronics Module-2

All the amplifier circuits studied in this section are to analyse this circuit to determine Rin AVO ,RO, and
unilateral. That is, they do not contain internal Gv. For this purpose we shall assume that RD, is part
feedback, and thus Rin will be independent of RL of the amplifier; thus if a load resistance RL, is
connected to the amplifier output, it appears in
The second parameter in characterizing amplifier
parallel with RD.
performance is the open-circuit voltage gain AVO
defined as Characteristic Parameters of the CS Amplifier
𝑣𝑜 Replacing the MOSFET with it's hybrid-π model,
𝐴𝑣𝑜 ≡ | 𝑅𝐿 =∞
𝑣𝑖 we obtain the CS amplifier equivalent circuit shown
in Figure
The third and final parameter is the output
resistance RO.It is the resistance seen looking back
into the amplifier output terminal with vi set to zero.
Thus RO, can be determined
𝑣𝑥
𝑅𝑜 =
𝑖𝑥

The controlled source AVO.vi and the output

resistance RO
We shall use this equivalent circuit to determine
𝑅𝐿
𝑣𝑜 = 𝐴 𝑣 the characteristic parameters Rin, AVO and RO, as
𝑅𝐿 + 𝑅𝑜 𝑣𝑂 𝑖 follows
Thus the voltage gain of the amplifier proper, AV, The input resistance Rin is obviously infinite
can be found as
𝑅𝑖𝑛 = ∞
𝑣𝑜 𝑅𝐿
𝐴𝑣 = = 𝐴
𝑣𝑖 𝑅𝐿 + 𝑅𝑜 𝑣𝑂 The output voltage vo, is found by multiplying the
current (gmvgs) by the total resistance between the
and the overall voltage gain G,
output node and ground,
𝑣𝑜
𝐺𝑣 = 𝑣𝑜 = −(𝑔𝑚 𝑣𝑔𝑠 )(𝑅𝐷 ||𝑟𝑜 )
𝑣𝑠𝑖𝑔
𝑅𝑖𝑛 𝑅𝐿 Since vgs = vi , the open-circuit voltage gain
𝐺𝑣 = . 𝐴
𝑅𝑖𝑛 + 𝑅𝑠𝑖𝑔 𝑅𝐿 + 𝑅𝑜 𝑣𝑂 𝐴𝑣𝑜 = 𝑣𝑜 ⁄𝑣𝑖 , can be obtained as
2.1.2 The Common source Amplifier 𝐴𝑣𝑜 = −𝑔𝑚 (𝑅𝐷 ||𝑟𝑜 )
without Source Resistance
Observe that the transistor output resistance ro,
Of the three basic MOS amplifier configurations, reduces the magnitude of the voltage gain. RD is
the common source is the most widely used. usually much lower than ro, and the effect of ro on
Typically, in an amplifier formed by cascading a reducing |AVO| is slight. Thus in many cases we can
number of stages, the bulk of the voltage gain is neglect ro and express AVO simply as
obtained by using one or more common-source
stages in the cascade. 𝐴𝑣𝑜 = −(𝑔𝑚 𝑅𝐷 )
The output resistance RO, is the resistance seen
looking back into the output terminal with vi, set to
zero (i.e. vgs =vi=0 , gmvgs=0)

So, 𝑅𝑂 = (𝑅𝐷 ||𝑟𝑜 )

Here, ro, has the effect of reducing the value of RD.

However, this effect is slight and we can make the
Figure shows a common-source amplifier (with the approximation
biasing arrangement omitted) fed with a signal 𝑅𝑂 ≃ 𝑅𝐷
source vsig having a source resistance Rsig. We wish
18EC42 Analog Electronics Module-2

We can now make the following observations.

1. The input resistance is ideally infinite.
2. The output resistance is moderate to high
3. The open-circuit voltage gain AOV, can be high,
4. The bandwidth of the CS amplifier is severely
limited.
Overall Voltage Gain
To determine the overall voltage gain Gv, we first
note that the infinite input resistance will make the From above figure we see that the input resistance
entire signal vsig appear at the amplifier input, Rin is infinite and thus vi = vsig Unlike the CS
amplifier, however, here only a fraction of vi,
vi= vsig ( obviously an ideal situation) appears between gate and source as vgs. It can be
At this point we should remind the reader that to determined from the voltage divider composed of
maintain vsig should be kept much smaller than 2 1/gm and RS, that appears across the amplifier input,
VOV. as follows:

If a load resistance RL, is connected to the output 1⁄ 𝑣𝑖

𝑔𝑚
terminal of the amplifier, this resistance will appear 𝑣𝑔𝑠 = 𝑣𝑖 =
1⁄ 1 + 𝑔𝑚 𝑅𝑠
𝑔𝑚 + 𝑅𝑠
in parallel with RD. It follows that the voltage gain
AV Thus we can use the value of RS. to control the
𝐴𝑣 = −𝑔𝑚 (𝑅𝐷 ||𝑟𝑜 ||𝑅𝐿 ) magnitude of the signal vgs and thereby ensure that
vgs does not become too large and cause
2.1.3 The Common-Source Amplifier unacceptably high nonlinear distortion. This is the
with a Source Resistance first benefit of including resistor RS
It is often insert a resistance RS, in the source lead RS introduces negative feedback, if while keeping
of the common-source amplifier as shown in Figure vi constant, for any reason the drain current
increases, the source current also will increase,
resulting in an increased voltage drop across RS.
Thus the source voltage rises, and the gate-to-
source voltage decreases. Then the drain current
decrease, this the indication of the presence of
negative feedback.

The output voltage vo, is obtained by multiplying

The corresponding small-signal equivalent circuit the controlled-source current i by RD
is shown in Figure, where we note that the
𝑣𝑜 = −𝑖 𝑅𝐷
MOSFET has been replaced with its T equivalent-
circuit model. The - model is used in preference to The current i in the source lead can be found by
the π model because it makes the analysis in this dividing vi, by the total resistance in the source.
case somewhat simpler. In general, whenever a
𝑣𝑖 𝑔𝑚
resistance is connected in the source lead, the T 𝑖 = = 𝑣𝑖
1⁄
model is preferred. The source resistance then 𝑔𝑚 + 𝑅𝑠 1 + 𝑔𝑚 𝑅𝑠
simply appears in series with the resistance 1/gm
Thus, the voltage gain AVO, can be found as
and can be added to it
𝑣𝑜 𝑅𝐷
𝐴𝑣𝑜 = =−
𝑣𝑖 1⁄ + 𝑅
𝑔 𝑚 𝑠

The above equation is same as

18EC42 Analog Electronics Module-2

𝑉𝑜𝑙𝑡𝑎𝑔𝑒 𝑔𝑎𝑖𝑛 𝑓𝑟𝑜𝑚 𝐺𝑎𝑡𝑒 𝑡𝑜 𝐷𝑟𝑎𝑖𝑛 Note that we have included ro simply because it is
𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑖𝑛 𝐷𝑟𝑎𝑖𝑛 very easy to do so. However, since rom effect
=
𝑇𝑜𝑡𝑎𝑙 𝑅𝑒𝑠𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑖𝑛 𝑆𝑜𝑢𝑟𝑐𝑒 appears in parallel with RL, and since ro>> RL, we
Which can be expressed as can neglect ro, and obtain the simplified equivalent
circuit shown in Figure (c).
𝑔𝑚 𝑅𝐷
𝐴𝑣𝑜 = −
1 + 𝑔𝑚 𝑅𝑠
Finally, we consider the situation of a load
resistance RL connected at the output. We can
obtain the gain AV, using the open-circuit voltage
gain AVO, together with the output resistance RO =
RD
𝑔𝑚 (𝑅𝐷 ||𝑅𝐿 )
𝐴𝑣 = −
1 + 𝑔𝑚 𝑅𝑠

2.1.4 The Common-Drain Amplifier or

Source Follower
The common drain amplifier is more commonly
known as the source follower.

The input resistance Rin is obviously infinite

𝑅𝑖𝑛 = ∞
and obtain AV, from the voltage divider formed by
1/gm and RL, as
𝑣𝑜 𝑅𝐿
𝐴𝑣 = =
𝑣𝑖 1⁄
𝑔𝑚 + 𝑅𝐿
Figure (a) shows a source follower with the bias Setting RL=∞
circuit omitted. The source follower is fed with a
signal generator (vsig, Rsig) and has a load resistance 𝐴𝑣 = 1
RL. Since the MOSFET has a resistance RL,
The output resistance RO, is found by setting vi = 0
connected in its source terminal, it is most
convenient to use the T model, as shown in Figure (i.e., by grounding the gate). Now looking back into
(b). the output terminal, excluding RL, we simply see
1/gm .
1
𝑅𝑜 =
𝑔𝑚

Finally, because of the infinite Rin, vi, = vsig , and

the overall voltage gain is
𝑅𝐿
𝐴𝑣 =
1⁄
𝑔𝑚 + 𝑅𝐿
In conclusion
The source follower features a very high input
resistance (ideally, infinite), a relatively low output
resistance, and an open-circuit voltage gain that is
near unity (ideally, unity), Thus the source follower
is ideally suited for implementing the unity-gain
18EC42 Analog Electronics Module-2

voltage buffer . The source follower is also used as capacitance WLCox to the gate-body model
the output stage in a multistage amplifier, where its capacitance; thus,
function is to equip the overall amplifier with a low
output resistance, thus enabling it to supply 𝐶𝑔𝑏 = 𝑊𝐿𝐶𝑜𝑥
relatively large load currents without loss of gain. 𝐶𝑔𝑠 = 𝐶𝑔𝑑 = 0
2.2.The MOSFET Internal Capacitance 4.There is an additional small capacitive
And High Frequency Model component that should be added to 𝐶𝑔𝑠 & 𝐶𝑔𝑑 in all
The device has internal capacitances. . There are the preceding formulas. This is the capacitance that
basically two types of internal capacitances in the results from the fact that the source and drain
MOSFET: diffusions extend slightly under the gate oxide. If
the overlap length is denoted we see that the overlap
1. The gate capacitive effect: The gate electrode capacitance component is
forms a parallel-plate capacitor with the channel,
with the oxide layer serving as the capacitor 𝐶𝑜𝑣 = 𝑊𝐿𝑜𝑣 𝐶𝑜𝑥
dielectric and denoted its value per unit area as Cox Typically Lov =0.05 to 0.1L
2. The source-body and drain-body depletion-layer 2.2.2. Junction Capacitance
capacitances: These are the capacitances of the
reverse-bi0ased p-n junctions formed by the n+ The depletion-layer capacitances of the two
source region (also called the source diffusion) and reverse-biased p-n junctions formed between each
the p-type substrate and by the n+ drain region (the of the source and the drain diffusions and the body
drain diffusion) and the substrate. can be determined using the formula

2.2.1. The Gate Capacitive Effect Ɛ𝑠 𝑞 𝑁𝐴 𝑁𝐷 1

𝐶𝑗 = 𝐴√( )( )( )
The gate capacitive effect can be modelled by the 2 𝑁𝐴 + 𝑁𝐷 𝑉𝑂
three capacitances Cgs, Cgd, and Cgb. The values
of these capacitances can be determined as Thus, for the source diffusion, we have the source
follows: body capacitance, Csb

1. When the MOSFET is operating in the triode 𝐶𝑠𝑏0

𝐶𝑠𝑏 =
region at small vDS, the channel will be of uniform 𝑉𝑆𝐵
√1 +
depth. The gate-channel capacitance will be WL Cox 𝑉𝑂
and can be modelled by dividing it equally between where Csb0 is the value of Csb at zero body-source
the source and drain ends; thus, bias, VSB is the magnitude of the reverse bias
1 voltage, and VO is the junction built-in voltage (0.6
𝐶𝑔𝑠 = 𝐶𝑔𝑑 = 𝑊𝐿𝐶𝑜𝑥 V to 0.8 V).
2
2. When the MOSFET operates in saturation, the Similarly, for the drain diffusion, we have the
channel has a tapered shape and is pinched off at drain-body capacitance Cdb
the drain end. It can be shown that the gate-to-
𝐶𝑑𝑏0
channel capacitance in this case is approximately 𝐶𝑑𝑏 =
2 𝑉𝐷𝐵
𝑊𝐿𝐶𝑜𝑥 and can be modelled by assigning this √1 +
3 𝑉𝑂
entire amount to Cgs, and a zero amount to Cgd
(because the channel is pinched off at the drain); where Cdb0 is the capacitance value at zero reverse-
thus, bias voltage and VDB is the magnitude of this
reverse-bias voltage.
2
𝐶𝑔𝑠 = 𝑊𝐿𝐶𝑜𝑥
3 2.2.3 The High-Frequency MOSFET Model
𝐶𝑔𝑑 = 0 Figure shows the small-signal model of the
MOSFET, including the four capacitances Cgs Cgd,
3. When the MOSFET is cut off, the channel
disappears, and thus Cgs= Cgd = 0. However, we Csb, and Cdb. This model can be used to predict the
can model the gate capacitive effect by assigning a high-frequency response of MOSFET amplifiers.
18EC42 Analog Electronics Module-2

In this case when the source is connected to the

body, the model simplifies as shown below

In this model, Cgd, although small, plays a

significant role in determining the high-frequency
response of amplifiers and thus must be kept in the
model. Capacitance Cdb can be neglected, resulting
in significant simplification of manual analysis.
The resulting circuit is shown in figure below

The value of the midband gain Am corresponds to

the overall voltage gain Gv
𝑅𝐺
𝐴𝑚 = − 𝑔 (𝑅 ||𝑟 ||𝑅 )
𝑅𝐺 + 𝑅𝑠𝑖𝑔 𝑚 𝐷 𝑜 𝐿

Figure (b) shows that the gain falls off at signal

frequencies below and above the midband. The
2.3. FREQUENCY RESPONSE OF gain fall off in the low-frequency band is due to the
THE CS AMPLIFIER fact that CC1, CC2, and CS are large capacitors (in the
uF range), as the signal frequency is reduced, their
In this section we study the dependence of the gain
impedances increase, and they no longer behave as
of the MOSFET common-source amplifier of short circuits. On the other hand, the gain falls off
Figure (a) on the frequency of the input signal in the high-frequency band as a result of Cgs and Cgd
2.3.1. The Three Frequency Bands which though very small (in the pF ), their
impedances at high frequencies decrease and thus
When we learned CS amplifier before we have can no longer be considered as open circuit. These
neglected all the capacitor effects, As a result of two sets of capacitances affect the amplifier gain in
ignoring all capacitive effects, the gain expressions the low-frequency and the high-frequency bands. In
derived were independent of frequency. In reality, this way, determine the frequencies fL and fH, which
this situation applies over only a limited band of define the extent of the midband
frequencies. This is illustrated in Figure (b), which
shows a sketch of the magnitude of the overall Usually, fL and fH are the frequencies at which the
voltage gain, |Gv|, of the CS amplifier versus gain drops by 3 dB below its value at midband. The
frequency. We observe that the gain is almost amplifier bandwidth or 3-dB bandwidth is defined
constant over a wide frequency band, called the as the difference between the lower (fL) and the
midband. upper or higher (fH) 3-dB frequencies
18EC42 Analog Electronics Module-2

𝐵𝑊 = 𝑓𝐻 − 𝑓𝐿
A figure-of-merit for the amplifier is its gain-
bandwidth product (GB), which is defined as
𝐺𝐵 = |𝐴𝑀 |. 𝐵𝑊

2.3.2.The High-Frequency Response

To determine the gain of the amplifier as shown
below For further simplifying, consider the bridging
capacitor Cgd that connects the output node to the
input side. Consider first the output node. It can be
seen that the load current is (gmVgs - Igd), where
(gmVgs) is the output current of the transistor and Igd
is the current supplied through the very small
capacitance Cgd. At frequencies in the vicinity of
fH, Igd is much smaller than (gmVgs), with the result
that VO, can be given approximately by

𝑉𝑜 = −(𝑔𝑚 𝑉𝑔𝑠 )𝑅𝐿′

Where 𝑅𝐿′ = 𝑟𝑜 ||𝑅𝐷 ||𝑅𝐿

at high frequencies replace the MOSFET with its The current Igd can now be found as
high-frequency model as shown below Figure.
𝐼𝑔𝑑 = 𝑠𝐶𝑔𝑑 (𝑉𝑔𝑠 − 𝑉𝑂 )

𝐼𝑔𝑑 = 𝑠𝐶𝑔𝑑 (𝑉𝑔𝑠 + (𝑔𝑚 𝑉𝑔𝑠 )𝑅𝐿′ )

𝐼𝑔𝑑 = 𝑠𝐶𝑔𝑑 (1 + 𝑔𝑚 𝑅𝐿′ )𝑉𝑔𝑠

Where
𝐶𝑒𝑞 = 𝐶𝑔𝑑 (1 + 𝑔𝑚 𝑅𝐿′ )

Using Ceq use to simplify the equivalent circuit at

the input side to that shown in Figure below

At these frequencies, CC1, CC2, and Cs will be

behaving as perfect short circuits. The result is the
high-frequency amplifier equivalent circuit shown
in figure below

Which is single-time-constant (STC) circuit of the

low-pass type. To express the output voltage Vgs of
the STC circuit in the form

The above equivalent circuit can be simplified by 𝑅𝐺 1

𝑉𝑔𝑠 = ( 𝑉𝑠𝑖𝑔 ) 𝑠
utilizing the Thevenin theorem at the input side and 𝑅𝐺 + 𝑅𝑠𝑖𝑔 1+𝜔
by combining the three parallel resistances at the 0

output side. The resulting simplified circuit is 1

Where 𝜔0 = 𝐶 ′
shown in Figure below 𝑖𝑛 𝑅𝑠𝑖𝑔
18EC42 Analog Electronics Module-2

𝐶𝑖𝑛 = 𝐶𝑔𝑠 + 𝐶𝑔𝑑 (1 + 𝑔𝑚 𝑅𝐿′ ) 3. The multiplication effect that Cgd

′
undergoes comes about because it is
𝑅𝑠𝑖𝑔 = 𝑅𝑠𝑖𝑔 ||𝑅𝐺 connected between two nodes whose
We Know that voltages are related by a large negative
gain 𝑔𝑚 𝑅𝐿′ This effect is known as the
𝑉𝑜 = −(𝑔𝑚 𝑉𝑔𝑠 )𝑅𝐿′ Miller effect, an(1 + 𝑔𝑚 𝑅𝐿′ ), is known as
the Miller multiplier. It is the Miller effect
Substituting 𝑉𝑔𝑠 expression on above expression that causes the CS amplifier to have a large
total input capacitance Cin and hence a low
𝑅𝐺 1
𝑉𝑜 = − (𝑔𝑚 ( 𝑉𝑠𝑖𝑔 ) ′
𝑠 ) 𝑅𝐿 fH.
𝑅𝐺 + 𝑅𝑠𝑖𝑔 1+
𝜔0 4. To extend the high-frequency response of a
MOSFET amplifier, we have to find
𝑉𝑜 𝑅𝐺 1 configurations in which the Miller effect is
= − (𝑔𝑚 𝑅𝐿′ ( ) )
𝑉𝑠𝑖𝑔 𝑅𝐺 + 𝑅𝑠𝑖𝑔 1 + 𝑠 absent or at least reduced.
𝜔0 5. The above analysis, resulting in an STC or
a single-pole response, is a simplified one.
ie,
Specifically, it is based on neglecting Igd
𝑉𝑜 𝐴𝑚 relative to gmVgs, an assumption that
=
𝑉𝑠𝑖𝑔 1 + 𝑠 applies well at frequencies not too much
𝜔 𝐻
higher than fH.
Where
𝑅𝐺
𝐴𝑚 = − 𝑔 (𝑅 ||𝑟 ||𝑅 )
𝑅𝐺 + 𝑅𝑠𝑖𝑔 𝑚 𝐷 𝑜 𝐿

And
𝑅𝐿′ = (𝑅𝐷 ||𝑟𝑜 ||𝑅𝐿 )
1
𝜔𝐻 = 𝜔0 = ′
𝐶𝑖𝑛 𝑅𝑠𝑖𝑔

And 2.3.3. The Low-Frequency Response

1
𝑓𝐻 = To determine the low-frequency gain or transfer
′
2𝜋𝐶𝑖𝑛 𝑅𝑠𝑖𝑔 function of the common-source amplifier, we show
in figure the circuit with the dc sources eliminated.
1. The upper 3-dB frequency is determined by
′ We shall perform the small-signal analysis directly
the interaction of 𝑅𝑠𝑖𝑔 = 𝑅𝑠𝑖𝑔 ||𝑅𝐺 and
on this circuit. However, we will ignore ro. This is
𝐶𝑖𝑛 = 𝐶𝑔𝑠 + 𝐶𝑔𝑑 (1 + 𝑔𝑚 𝑅𝐿′ ). Since the done in order to keep the analysis simple and thus
bias resistance RG , is usually very large, it focus attention on significant issues.
′
can be neglected, resulting in 𝑅𝑠𝑖𝑔 = 𝑅𝑠𝑖𝑔
the resistance of the signal source. It
follows that a large value of 𝑅𝑠𝑖𝑔 will cause
𝑓𝐻 to be lowered.
2. The total input capacitance Cin is usually
dominated by Ceq which in turn is made
large by the multiplication effect that Cgd
undergoes. Thus, although Cgd is usually a
very small capacitance, its effect on the
amplifier frequency response can be very
significant as a result of its multiplication
by the factor (1 + 𝑔𝑚 𝑅𝐿′ ), which is
approximately equal to the midband gain of
the amplifier.
18EC42 Analog Electronics Module-2

The effect of ro , on the low-frequency operation of From which see that CC2 introduces a third STC
this amplifier is The analysis begins at the signal high-pass factor, giving the amplifier a third break
generator by finding the fraction of Vsig that appears frequency at
at the transistor gate 1
𝜔𝑃3 =
𝑅𝐺 𝐶𝐶2 (𝑅𝐷 + 𝑅𝐿 )
𝑉𝑔 = 𝑉𝑠𝑖𝑔
1
𝑅𝐺 + 𝑠𝐶 + 𝑅𝑠𝑖𝑔 Substituting 𝐼𝑑 and 𝑉𝑔 on 𝑉𝑜 equation
𝐶1
𝑽𝑶 𝑹𝑮 𝑺 𝑺 𝑺
= −( ) [𝒈𝒎 (𝑹𝑫 ||𝑹𝑳 )] ( )( )( )
𝑽𝒔𝒊𝒈 𝑹𝑮 +𝑹𝒔𝒊𝒈 𝑺 +𝝎𝑷𝟏 𝑺 +𝝎𝑷𝟐 𝑺 +𝝎𝑷𝟑

This can be re written as

𝑅𝐺 𝑆
𝑉𝑔 = 𝑉𝑠𝑖𝑔
𝑅𝐺 + 𝑅𝑠𝑖𝑔 𝑆 + 1
𝐶𝐶1 (𝑅𝐺 + 𝑅𝑠𝑖𝑔 )

The effect of the coupling capacitor CC1 is to

introduce a high-pass STC response with a break
frequency that we shall denote wP1
1
𝜔𝑃1 = 𝜔0 =
𝐶𝐶1 (𝑅𝐺 + 𝑅𝑠𝑖𝑔 )

Continuing with the analysis, next determine the

drain current Id, by dividing Vg, by the total 2.4. OSCILLATOR
impedance in the source circuit which is
[(1/gm)+(1/sCS)1 to obtain 2.4.1 The Phase-Shift Oscillator
𝑉𝑔 The basic structure of the phase-shift oscillator is
𝐼𝑑 =
1 1 shown in Figure.
𝑔𝑚 + 𝑠𝐶𝑠
Which can be written in the alternate form
𝑆
𝐼𝑑 = 𝑔𝑚 𝑉𝑔 𝑔𝑚
𝑆+
𝐶𝑆
This observe that CS introduces a frequency-
dependent factor, which is also of the STC high
pass type. Thus the amplifier acquires another It consists of a negative-gain amplifier (-K) with a
break frequency three-section RC ladder network in the feedback.
𝑔𝑚 The circuit will oscillate at the frequency for which
𝜔𝑃2 = the phase shift of the RC network is 180°. Only at
𝐶𝑆
this frequency will the total phase shift around the
To complete the analysis, we find VO, by first loop be 0° or 360° since the BJT or MOSFET
using the current-divider rule to determine the provides rest 180o phase shift. Here we should note
fraction of Id that flows through RL that the reason for using a three-section RC network
𝑅𝐷 is that three is the minimum number of sections that
𝐼𝑂 = −𝐼𝑑 is capable of producing a 180° phase shift (60o
1
𝑅𝐷 + 𝑠𝐶 + 𝑅𝐿 phase shift per RC network)
𝐶2

And then multiplying IO by RL to obtain For oscillations to be sustained, the value of K

should be equal to the inverse of the magnitude of
𝑅𝐷 𝑅𝐿 𝑆 the RC net work transfer function at the frequency
𝑉𝑜 = 𝐼𝑜 𝑅𝐿 = −𝐼𝑑
𝑅𝐷 + 𝑅𝐿 𝑆 + 1 of oscillation. However, to ensure that oscillations
𝐶𝐶2 (𝑅𝐷 + 𝑅𝐿 ) start, the value of K has to be chosen slightly higher
18EC42 Analog Electronics Module-2

than the value that satisfies the unity-loop-gain

condition.

FET -Based-Phase-Shift-Oscillator
Let us first consider an FET-based phase-shift
oscillator as shown in Figure.

Here, the current source is replaced by the

equivalent voltage source 𝑔𝑚 𝑉𝑖 𝑅𝐷′ ; , where 𝑅𝐷′ =
𝑟𝐷 ||𝑅𝐷 .In the above figure Applying KVL, we get
1
𝐼1 (𝑅 + ) − 𝐼2 𝑅 = −𝑔𝑚 𝑅𝐷′ 𝑉𝑖
𝑆𝐶
1
−𝐼1 𝑅 + 𝐼2 (2𝑅 + ) − 𝐼3 𝑅 = 0
𝑆𝐶
1
−𝐼2 𝑅 + 𝐼3 (2𝑅 + ) =0
𝑆𝐶
Also
The circuit consists of an amplifier and a feedback 𝑉𝑓 = 𝐼3 𝑅
network. The FET amplifier is self-biased with a
capacitor-bypassed source resistor and a drain Hence the loop gain
resistor. The feedback network consists of three 𝑉𝑓 𝑔 𝑅′
cascaded RC sections. 𝐴𝛽 = = (1−5𝛼2𝑚 𝐷
)+(𝛼3
𝑉𝑖 −6𝛼)
As shown, the feedback voltage Vf, available at the Where
output of the last RC section of the feedback
1
network is fed to the gate as input. As the FET 𝛼 = 𝜔𝑅𝐶
amplifier provides a phase shift of 180°, the
remaining 180° phase shift has to be provided by As loop gain is real quantity
the feedback network to obtain a total phase shift of
𝛼 3 − 6𝛼 = 0
360° around the loop. Thus, each RC section is
designed so as to provide a phase shift of 60° at the 𝛼2 = 6
desired frequency of oscillation. We will now
compute the loop gain 𝐴𝛽 from the ac equivalent 1
𝜔2 𝑅 2 𝐶 2 =
circuit given in Figure. For this, we will assume that 6
the input impedance of the FET amplifier stage is The frequency oscillation becomes
infinite.
1
𝑓𝑜 =
2𝜋𝑅𝐶√6

′
𝑔𝑚 𝑅𝐷
|𝐴𝛽 | =
29

And know 𝐴𝛽 > 1 for sustained oscillation

𝑔𝑚 𝑅𝐷′ > 29
The voltage gain of the FET amplifier is given as
The above figure can be further simplified as shown
in Figure. |𝐴| > 𝑔𝑚 𝑅𝐷′
18EC42 Analog Electronics Module-2

|𝐴| > 29
1
And 𝛽=
29

FET -Based-Phase-Shift-Oscillator
If a BJT is used as the active element of the
amplifier stage instead of an FET, for the circuit
given in Figure.

where the current source is replaced by a voltage

source.
1
𝐼1 (𝑅𝐶 + 𝑅 + ) − 𝐼2 𝑅 = −ℎ𝑓𝑒 𝑅𝐶 𝐼𝑏
𝑆𝐶
1
−𝐼1 𝑅 + 𝐼2 (2𝑅 + ) − 𝐼3 𝑅 = 0
𝑆𝐶
1
−𝐼2 𝑅 + 𝐼3 (2𝑅 + ) =0
𝑆𝐶
Let
1 𝑅𝐶
𝛼 = 𝜔𝑅𝐶 & 𝐾= 𝑅

The resistance R of the feedback network is now in 𝐼2 = 𝐼3 (2 − 𝑗𝛼)

parallel with the low input resistance hie of the
𝐼1 = 𝐼3 (3 − 𝛼 2 − 𝑗4𝛼)
transistor. This reduces the effective value of R in
the last section of the feedback. the feedback signal Thus, simplifying the equation
is coupled through the feedback resistor R3 in series
𝐼3 −ℎ𝑓𝑒 𝐾
with the amplifier stage input resistor. =
𝐼𝑏 1 + 3𝐾 − (5 + 𝐾)𝛼 2 − 𝑗(6 + 4𝐾)𝛼 − 𝛼 3
The small-signal ac equivalent model is shown in
Figure. As the lop gain is real quantity
(6 + 4𝐾)𝛼 − 𝛼 3 = 0

𝛼 2 = 6 + 4𝐾
Then, the Frequency
1
𝑓𝑜 =
2𝜋𝑅𝐶 √6 + 4𝐾
At this frequency
𝐼3 −ℎ𝑓𝑒 𝐾
The value of R, is chosen such that = 2
𝐼𝑏 4𝐾 + 23𝐾 + 29
𝑅3 + 𝑅𝑖 = 𝑅 𝐼3
Also for sustained oscillation 𝐼𝑏
>1
𝑅3 + ℎ𝑖𝑒 = 𝑅
29
ℎ𝑓𝑒 > 4𝐾 + 23 +
𝐾
which makes the three RC sections identical.
The value of 𝐾 such that ℎ𝑓𝑒 is minimum can be
Further simplifying the circuit in the above Figure
determine as follows
we get following Figure.
𝑑ℎ𝑓𝑒 29
=4− 2 =0
𝑑𝐾 𝐾
18EC42 Analog Electronics Module-2

29 1/2 They are known as the Colpitts oscillator and the

𝐾=( ) = 2.7 Hartley oscillator. Both utilize a parallel LC circuit
4
connected between collector and base (or between
29 drain and gate if a FET is used) with a fraction of
(ℎ𝑓𝑒 )𝑚𝑖𝑛 = 4(2.7) + 23 + = 44.5
2.7 the tuned-circuit voltage fed to the emitter (the
Hence can say that the value of ℎ𝑓𝑒 for a transistor source in a FET). This feedback is achieved by way
must be at least 45 for the circuit oscillate. of a capacitive divider in the Colpitts oscillator and
by way of an inductive divider in the Hartley
Ring Oscillator circuit. (The bias details are not shown) .In both
circuits, the resistor R models the combination of
The simplest example of the phase-shift oscillator
the losses of the inductors, the load resistance of the
the ring oscillator, which is extensively used in
oscillator, and the output resistance of the transistor
digital integrated circuits for clock generation. The
ring oscillator consists of odd number of inverters If the frequency of operation is sufficiently low that
connected in a ring as shown in Figure we can neglect the transistor capacitances, the
frequency of oscillation will be determined by the
resonance frequency of the parallel-tuned circuit
(also known as a tank circuit because it behaves as
a reservoir for energy storage).
Thus for the Colpitts oscillator, we have
The load of each inverter is a capacitor. It is known
𝜔0 = 1
that every inverter provides a 180° phase shift ⁄√𝐿 ( 𝐶1 𝐶2 )
between input and output. If the inverters and the 𝐶1 + 𝐶2
associated load capacitors generate a net phase at
Thus for the Hartley oscillator, we have
the last inverter output in multiples of 360°, then the
feedback signal is in phase with the input and the 𝜔0 = 1⁄
oscillations can be sustained. Since the output of an √(𝐿1 + 𝐿2 )𝐶
inverter is a pulse, the ring oscillator generates
pulses. The frequency of this pulse generator is
decided by the propagation delay per inverter stage
2nd the number of such (odd) stages.

2.5 LC AND CRYSTAL OSCILLATOR

Oscillators utilizing transistors (FETs or BJTs),
with LC-tuned circuits or crystals as feedback
elements, are used in the frequency range of 100
kHz to hundreds of megahertz. However, LC
oscillators are difficult to tune over wide ranges,
and crystal oscillators operate at a single frequency A node equation at the transistor collector (node
2.5.1 LC-Turned Oscillator C) in the circuit of Figure yields
1
Figure shows two commonly used configurations 𝑠𝐶2 𝑉𝜋 + 𝑔𝑚 𝑉𝜋 + (𝑠𝐶1 + ) (1 + 𝑆 2 𝐿𝐶2 )𝑉𝜋 = 0
of LC-tuned oscillators. 𝑅
Since 𝑉𝜋 ≠ 0
𝐿𝐶2 1
𝑠 3 𝐿𝐶1 𝐶2 + 𝑠 2 ( ) + 𝑠(𝐶1 + 𝐶2 ) + (𝑔𝑚 + ) = 0
𝑅 𝑅
Substituting s=jω
1 𝜔2 𝐿𝐶2
(𝑔𝑚 + 𝑅 + 𝑅
)+ 𝑗[𝜔(𝐶1 + 𝐶2 ) − 𝜔3 𝐿𝐶1 𝐶2 ]=0
18EC42 Analog Electronics Module-2

For oscillation to start ,both real and imaginary Parallel resonance at 𝜔𝑃

parts must be zero. Equating imaginary part to zero,
Then frequency of oscillation 𝜔𝑃 = 1
⁄√𝐿 𝐶𝑆 𝐶𝑃
𝐶𝑆 + 𝐶𝑃
𝜔0 = 1
⁄√𝐿 ( 𝐶1 𝐶2 )
𝐶1 + 𝐶2
𝐶2 1 𝜔2 − 𝜔𝑠2
𝑔𝑚 𝑅 = 𝑍(𝑗𝜔) = −𝑗 ( )
𝐶1 𝜔𝐶𝑃 𝜔 2 − 𝜔𝑃2

2.5.2 Crystal Oscillator the crystal reactance is inductive over the very
narrow frequency band between w, and w, For a
A piezoelectric crystal, such as quartz, exhibits given crystal, this frequency band is well defined.
electromechanical-resonance characteristics that Thus we may use the crystal to replace the inductor
are very stable and highly selective. The circuit of the Colpitts oscillator. The resulting circuit will
symbol of a crystal is shown in Figure (a), and its oscillate at the resonance frequency of the crystal
equivalent circuit model is given in Figure (b).
In addition to the basic Colpitts oscillator, a variety
of configurations exist for crystal oscillators.
Figure

The resonance properties are characterized by a

large inductance L, a very small series capacitance
CS , a series resistance r representing a Q factor
𝜔0 𝐿⁄
𝑟 that can be as high as a few hundred
thousand, and a parallel capacitance Cp. Capacitor
Cp represents the electrostatic capacitance between
the two parallel plates of the crystal. Note that shows a popular configuration (called the Pierce
Cp>>Cs oscillator)utilizing a CMOS inverter as amplifier.
Resistor Rf, determines a dc operating point in the
𝑍(𝑠) = 1 high-gain region of the VTC of the CMOS inverter.
⁄ 1 Resistor R together with capacitor C provides a
[𝑠𝐶𝑃 + 1 ] low-pass filter that discourages the circuit from
𝑠𝐿 + 𝑠𝐶
𝑠 oscillating at a higher harmonic of the crystal
frequency. Note that this circuit also is based on the
Colpitts configuration.

1 𝑠 2 + (1⁄𝐿𝐶 ) 2.5.3 Miller Oscillator

𝑠
𝑍(𝑠) =
𝑠𝐶𝑃 𝑠 2 + [(𝐶𝑃 + 𝐶𝑠 )⁄
𝐿𝐶𝑠 𝐶𝑃 ] Miller oscillators are generally used in
electromechanical applications. The basic structure
of a Miller oscillator is shown in Figure.
Two resonance frequencies:
Series resonance at 𝜔𝑆

𝜔𝑆 = 1⁄
√𝐿𝐶𝑠
18EC42 Analog Electronics Module-2

oscillator (VCXO), which is used in PLL circuits,

where the frequency is controlled by the external
voltage. Oven-controlled crystal oscillators
(OCXO) are known to be most precise and highly
stable oscillators. These are used in many wireless
systems and, in particular, at their base stations.

Video Links :
Further, a typical Miller oscillator using a JFET is Basic configurations, characterizing amplifiers
shown in Figure.
https://www.youtube.com/watch?v=TzLF9Lw
P2_o
CS amplifier with and without source resistance
RS

https://www.youtube.com/watch?v=HcTMpyD
yb8s
https://www.youtube.com/watch?v=gINP5gsls
c4
Source follower

https://www.youtube.com/watch?v=2XZGx6rx
30A
High frequency MOS transistor model, frequency
response of CS amplifier
It shows that the input of this oscillator is a signal (Detail explanation available in this video . Refer
generated by a crystal and the output is obtained text book/PDF note/Syllabus)
from a tank-circuit load. The inductance L1 and
capacitor C1 of the tank circuit are adjusted to the https://www.youtube.com/watch?v=_6htztp92t
desired resonant frequency. In general, the inductor U
is chosen to have a fixed value L1, and the
capacitance C1 is varied (by using a gang https://www.youtube.com/watch?v=A_ZY7y3
condenser). The output is taken at the drain end of XPFM
the JFET itself or bypassed through a capacitor C2,
https://www.youtube.com/watch?v=tsZRUkkIe
as shown in the figure. The feedback in this circuit
is provided by the drain-source capacitance. 0U

However, this itself is disadvantageous, as different https://www.youtube.com/watch?v=_z68WmG

devices will have small variations in the values of Psfs
Cgd and this may lead to variation in the frequency. RC Phase Shift Oscillator (Using BJT. Similar way
Further, the output may be subject to amplitude you can proceed the derivation for MOSFET)
variations due to load conditions. An improved
version of the Miller oscillator uses the dual gate https://www.youtube.com/watch?v=GkW7gB-
MOSFET and is normally used as a fundamental mjJw
mode oscillator, with no overtones possible.
Crystal Oscillator
The most important characteristic of the crystal
oscillator is its stability. It is possible to have a https://www.youtube.com/watch?v=yrcOwQB
temperature-compensated crystal oscillator knSw
(TCXO), which finds use in many digital ICs.
Another type is the voltage-controlled crystal
18EC42 Analog Electronics Module-2

Hartley Oscillator (You proceed Similar derivation

by inter changing L and C, become Colpitts
oscillator)

https://www.youtube.com/watch?v=Nu5vgH-
3AmU
Colpitts Oscillator

https://www.youtube.com/watch?v=1fgw-
ONlAcc