FILTERS AND OSCILLATORS

IT-4110

1
 TABLE OF CONTENTS

Second Order Filter .........................................................................................................3

Objectives....................................................................................................................3
Theory .........................................................................................................................3
Second Order Low-Pass Filter .........................................................................................5
Experiment No. 1.........................................................................................................9
Second Order High-Pass Filter ......................................................................................13
Experiment No. 2.......................................................................................................16
Second Order Band-Pass Filter ......................................................................................20
Experiment No. 3.......................................................................................................23
Second Order Band Reject Filter ...................................................................................27
Experiment No. 4.......................................................................................................30
Rf Oscillators ................................................................................................................34
Objectives..................................................................................................................34
Theory .......................................................................................................................34
The Operation Theory Of Oscillators .........................................................................34
Colpitts And Hartley Oscillators ................................................................................35
Crystal Oscillator.......................................................................................................41
Voltage Controlled Oscillator.....................................................................................44
Measured Results.......................................................................................................47

2
 SECOND ORDER FILTER

Objectives
1. To understand the basic theory of filters.
2. To design and implement the second order low-pass filters.
3. To design and implement the second order high-pass filters.
4. To design and implement the second order band-pass filters.
5. To design and implement the second order band reject filters.

Theory
Filters are ubiquitous in telecommunication systems. The function of filters is to remove
the unwanted signal and reserve the wanted signal at frequency domain. Normally filters
can be classified into three different classifications: If discriminated by filtering range,
then there are four types which are low-pass filter (LPF), high-pass filter (HPF), band-
pass filter (BPF) and band reject filter (BRF). Figure 1-1 shows the frequency responses
of low-pass, high-pass, band-pass and band-reject filters.

If discriminated by the frequency response of pass-band area, then there are two popular
types, which are Butterworth filter and Chebyshev filter, as shown in figure 1-2.
Butterworth filter is also known as maximally flat filter since no ripple is permitted in its
pass-band. However, ripple is appeared at the inner part of stop-band and the attenuation
of transition band is not sharp enough. Chebyshev filter is also known as equal ripple
filter since the ripple is equally in its pass-band. There are no ripples in the stop-band and
the attenuation of transition band is sharper than Butterworth filter.

(a) Low-pass filter (b) High-pass filter

3
 (c) Band-pass filter (d) Band-reject filter

Figure 1-1 Frequency responses of ideal filters

(a) Butterworth filter (b) Chebyshev filter

Figure 1-2 Frequency responses of Butterworth and Chebyshev filters

If discriminated by components, then there are two types, which are active filter and
passive filter. Early days, filter is formed by resistors, capacitors and inductors, so this
type of filter is called passive filter. Recent years, active components such as operation
amplifier (OP) are generally used; therefore we called type of filter as active filter. The
advantages active filter are as below:

1. After special circuit arrangement, the transfer function provides property of

inductor, therefore it can replace inductor components.
2. Since operation amplifier (OP) has high input impedance and low output
impedance, so the isolation is very good and can easily be used for series
application.
3. Active components have function of amplifying, hence active filter also provides
gain.

In the experiments, we will introduce the implementation of second order low-pass filter,
high-pass filter, band-pass filter and band-stop filter. Normally the cut-off frequency of a
filter is determined by the frequency with an attenuation of 3 dB.

4
 SECOND ORDER LOW-PASS FILTER

The purpose of low-pass filter is o enable low frequency signal pass through and high
frequency signal attenuate. Figure 1-3 shows the inverting integrator or Miller integrator
circuit, its transfer function is:

1 1
Vout (S )  SC  RC 
   0 (1-1)
Vin (S ) R S S
Where
1
o 
RC

From equation (1-1), we know that the Miller integrator circuit is the first order low-pass
filter. If we want to design the second order active low-pass filter, we need two Miller
integrator circuits and an inverting amplifier.

Figure 1-4 shows the block diagram of the second order active low-ass filter. It is
comprised by two Miller integrator circuits, an unit gain inverting amplifier and an adder.
Assume that the output of the adder is A(s), then the transfer function is

o 1 
A( S )  KVin (S )  ( ) A( S )  (( o ) 2 ) A( S )
S Q S

 1   
A( S ) 1  ( o )  ( o )2   KVin ( S )
 Q S S 

o 2  KVin (S )
Vout ( S )  ( ) A(S )  ( o )2
S S 1  ( o )  (o ) 2
1
Q S S

Figure 1-3 Miller integrated circuit

5
 Figure 1-4 Block diagram of second order active low-pass filter

Simplify the above equation, we get

Vout (S ) K o2
 (1-2)
Vin (S )  
S 2   0  S  02
Q

Equation (1-2) is the standard equation of the second order active low-pass filter, if we
follow the block diagram to design the circuit, we can obtain the second order low-pass
filter.

Figure 1-5 shows the circuit diagram of the second order active low-pass filter, which is
based on the block diagram in figure 1-4. We combine the adder and the first Miller
integrator circuit b replacing an OP (U1). So we can save an OP, but the calculation of the
circuit parameters will be more complicated. If we assume

C1 = C2 = C

R6 = R5 = R4

Then the transfer function can be expressed as

R3 1
Vout ( S ) R1 R3 R4C 2

Vin (S ) S 2  1 S  1
CR2 R3 R4C 2

6
 R3 1
R1 R3 R4C 2
 (1-3)
R3 R4 1 1
S2 S
R2 C R3 R4 R3 R4C 2

Figure 1-5 Circuit diagram of second order active low-pass filter

Compare equations (1-2) and (1-3), we obtain

R3
K (1-4)
R1

1
0  (1-5)
C R3 R4

R2
Q (1-6)
R3 R4

From the circuit diagram at figure 1-5, R1, R2, R3, C1 and U1 : A not only comprise a
Miller integrator circuit, but also have the function of weighted summer. The objective is
to multiply an individual weight for input signal and the output signal of U1 : C. After that
make a summation on input signal and the output signal of U1 : C . R4, C2 and U1 : B
comprise a Miller integrator circuit; R5, R6 and U1 : C comprise an inverting amplifier.
According to the theory of network composition, this circuits satisfies the conditions of

7
Butterworth, therefore the curve for frequency response at pass-band area is very smooth
with no ripple.

8
 EXPERIMENT 1

Objective
To Study Second Order Low Pass Filter

Requirements
1. IT-4110 Trainer Board
2. 2mm Patch Cords
3. Oscilloscope

Experimental Setup
Refer to the following diagram to configure setup for the present experiment

Procedure:
1. Short circuit point 1 with 2 and 3 with 4 so that C1 = C2 =10nF.
2. At the input of Low pass filter, provide 500mV, 10 Hz sine wave frequency. Then
by using oscilloscope observe the LPF output signal and record the measured
results in table 1-1.
3. Now without changing the signal amplitude, change the frequency to 30 Hz, 50
Hz, 70 Hz, 100 Hz, 300 Hz, 500 Hz, 700 Hz, 1 kHz, 3 kHz, 5 kHz, 7 kHz and 10

9
 kHz. Then by using the oscilloscope observe the output signals at the LPF output
and record the measured results in table 1-1.
4. Find the voltage gain of each frequency and record the measured results in table 1-
1.
5. From the data in table 1-1, sketch and label the voltage gain in Bode plots in
figure 1-14.
6. Now remove the patch cord from point 1 and 4 and short circuit 2 with 3 and 5
with 6 so that C3 = C4 = 3.3nF.

7. Repeat step 2 to 5 and record the measured results in table 1-2 and figure 1-15.

10
 Input
Signal
10 30 50 70 100 300 500 700 1k 3k 5k 7k 10k
Frequency
(Hz)
Output
Signal
Amplitude
(mV)

Voltage
Gain (dB)

Table 1-1 Measured results of second order low-pass filter. (C1 = C2 = 10 nF)

Figure 1-14 Bode plots of voltage gain versus frequency

11
 Input
Signal
10 30 50 70 100 300 500 700 1k 3k 5k 7k 10k
Frequency
(Hz)
Output
Signal
Amplitude
(mV)

Voltage
Gain (dB)

Table 1-2 Measured results of second order low-pass filter. (C3 = C4 = 3.3 nF)

Figure 1-15 Bode plots of voltage gain versus frequency

12
 SECOND ORDER HIGH-PASS FILTER

The frequency response of second order high-pass filter is just the reverse of second order
low-pass filter. The objective is to enable high frequency signal pass through and
attenuate low frequency signal. Figure 1-6 is the block diagram of high-pass filter. It is
constructed by two Miller integrator circuits, an inverting amplifier and two adders. We
can obtain the transfer function as

Vout (S ) KS 2
 (1-7)
Vin (S )  
S 2   0  S  02
Q

The above function is the standard function of high-pass filter. We can design and obtain
the circuit of second order high-pass filter by following the function from the block
diagram.

Figure 1-6 Block diagram of second order high-pass filter

Figure 1-7 Circuit diagram of second order high-pass filter

13
Figure 1-7 is the circuit diagram of second order high-pass filter, which is based on the
block diagram of figure 1-6. The first adder and the first Miller integrator circuit are
combined and replaced by U1: A. The second adder and inverting amplifier is replaced by
U1: B. Then we can save two operation amplifiers, but for the circuit parameters, the
calculation will become more complicated. We assume

C1 = C2 = C

R7 = R6 = R5

Then the transfer function is

R5 2 R5  1 R2 
 S    
Vout (S ) R2 CR2  R3 R1 R4 
 (1-8)
Vin (S ) 1 1
S2  S
R3C R4 R5C 2

If assume R1R4 = R2R3,

Then

R5 2
 S
Vout ( S ) R2

Vin (S ) S 2  1 S  1
R3C R4 R5C 2

R5 2
 S
R2
 (1-9)
R4 R5
2 S 1
S    2
R3 C R4 R5 R4 R5C

Compare equations (1-7) and (1-9), we obtain

R5
K (1-10)
R2

1
o  (1-11)
C R4 R5

R2
Q (1-12)
R4 R5

14
From figure 1-7, R1, R3, R7, C1 and U1: A not only comprise a Miller integrator circuit,
but also have the function of weighted summer. The objective is to multiply an individual
weight for input signal and the output signal of U1: C. After that make a summation on
input signal and the output signal of U1: C. R2, R4, R5 and U1: B comprise a weighted
summer circuit, which can multiply an individual weight for input signal and the output
signal of U1: A. After that make a summation on input signal and the output signal of U1:
A. R6, C2 and U1: C comprise a miller integrator circuit. This circuit also satisfies the
conditions of Butterworth, because of its pole overlapping, therefore the curve for
frequency response at pass-band area is very smooth with no ripple.

15
 EXPERIMENT 2

Objective
To Study Second Order High Pass Filter

Requirements
1. IT-4110 Trainer Board
2. 2mm Patch Cords
3. Oscilloscope

Experimental Setup
Refer to the following diagram to configure setup for the present experiment

Procedure:
1. Short circuit point 7 with 8 and 10 with 11 so that C1 = C2 =2.2nF.
2. At the input of High pass filter, provide 500mV, 70 Hz sine wave frequency. Then
by using oscilloscope observe the HPF output signal and record the measured
results in table 1-3.
3. Now without changing the signal amplitude, change the frequency to 100 Hz, 300
Hz, 500 Hz, 700 Hz, 1 kHz, 3 kHz, 5 kHz, 7 kHz, 10 kHz, 30 kHz, 50 kHz and

16
 100 kHz. Then by using the oscilloscope observe the output signals at the LPF
output and record the measured results in table 1-3.
4. Find the voltage gain of each frequency and record the measured results in table 1-
3.
5. From the data in table 1-3, sketch and label the voltage gain in Bode plots in
figure 1-16.
6. Now remove the patch cord from point 7 and 10 and short circuit 8 with 9 and 11
with 12 so that C3 = C4 = 1nF.

7. Repeat step 2 to 5 and record the measured results in table 1-4 and figure 1-17.

17
 Input
Signal
70 100 300 500 700 1k 3k 5k 7k 10k 30k 50k 100k
Frequency
(Hz)
Output
Signal
Amplitude
(mV)

Voltage
Gain (dB)

Table 1-3 Measured results of second order high-pass filter. (C1 = C2 = 2.2 nF)

Figure 1-16 Bode plots of voltage gain versus frequency

18
 Input
Signal
70 100 300 500 700 1k 3k 5k 7k 10k 30k 50k 100k
Frequency
(Hz)
Output
Signal
Amplitude
(mV)

Voltage
Gain (dB)

Table 1-4 Measured results of second order high-pass filter. (C3 = C4 = 1 nF)

Figure 1-17 Bode plots of voltage gain versus frequency

19
 SECOND ORDER BAND-PASS FILTER

The objective of band-pass filter is to make a certain low and high frequency signal
attenuate, but enable a certain frequency band signal pass through. From Figure 1-8, we
can see that the characteristic curve is a combination of low-pass and high-pass filters.
The band-pass filter consists of a f3dB = f2 low-pass filter. Moreover, the center frequency
is f o  f1  f 2 and 3 dB frequency band (BW3dB) is f2 – f1.

Figure 1-8 Frequency response of band-pass filter

Figure 1-9 Block diagram of second order active band-pass filter

20
Figure 1-9 shows the block diagram of the second order active band-pass filter, which
consists of two Miller integrator circuits, a unit gain inverting amplifier and an adder. The
transfer function is expressed as

Vout ( S ) KSo
 (1-13)
Vin (S )  
S 2  0  S  02
Q

Equation (1-13) is the standard equation of the second order active band-pass filter, if we
follow the block diagram to design the circuit, we can obtain the second order hand-pass
filter.

Figure 1-10 Circuit diagram of second order active band-pass filter

Figure 1-10 is the circuit diagram of the second order active band-pass filter, which is
based on the block diagram in figure 1-9. We combine the adder and the first Miller
integrator circuit by replacing an OP (U1: A). So we can save an OP, but the calculation of
the circuit parameters will be more complicated. If we assume

C1 = C2 = C

R6 = R5 = R4

Then the transfer function can be expressed as

21
 R3 R4 1
 s
Vout (S ) R1 R3 R4 C
 (1-14)
Vin (S ) 2 R3 R4 1 1
S  S
R2 C R3 R4 R3 R4C 2

Comparing equations (1-13) and (1-14), we get

R3 R4
K (1-15)
R1

1
o  (1-16)
C R3 R4

R2
Q (1-17)
R3 R4

In figure 1-10, R1, R2, R3, C1 and U1: A not only comprise a Miller integrator circuit, but
also provide the function of weighted summer. The objective is to multiply a weighted to
the input signal and output signal of U1: C, respectively, after that sum the input signal
and output signal of U1: C. R4, C2 and U1: B comprise a Miller circuit R5, R6 and U1: C
comprise an unit gain inverting amplifier. This circuit satisfies the conditions of
Butterworth filter, therefore, there is no ripple permitted in the pass-band and the
frequency response is very flat.

22
 EXPERIMENT 3

Objective
To Study Second Order Band Pass Filter

Requirements
1. IT-4110 Trainer Board
2. 2mm Patch Cords
3. Oscilloscope

Experimental Setup
Refer to the following diagram to configure setup for the present experiment

Procedure:
1. Short circuit point 13 with 14 and 16 with 17 so that C1 = C2 =10nF.
2. At the input of Band pass filter, provide 500mV, 10 Hz sine wave frequency. Then
by using oscilloscope observe the BPF output signal and record the measured
results in table 1-5.
3. Now without changing the signal amplitude, change the frequency to 30 Hz, 50
Hz, 70 Hz, 100 Hz, 300 Hz, 500 Hz, 700 Hz, 1 kHz, 3 kHz, 5 kHz, 7 kHz and 10

23
 kHz. Then by using the oscilloscope observe the output signals at the BPF output
and record the measured results in table 1-5.
4. Find the voltage gain of each frequency and record the measured results in table 1-
5.
5. From the data in table 1-1, sketch and label the voltage gain in Bode plots in
figure 1-18.
6. Now remove the patch cord from point 13 and 16 and short circuit 14 with 15 and
17 with 18 so that C3 = C4 = 3.3nF.

7. Repeat step 2 to 5 and record the measured results in table 1-6 and figure 1-19.

24
 Input
Signal
10 30 50 70 100 300 500 700 1k 3k 5k 7k 10k
Frequency
(Hz)
Output
Signal
Amplitude
(mV)

Voltage
Gain (dB)

Table 1-5 Measured results of second order band-pass filter. (C1 = C4 = 10 nF)

Figure 1-18 Bode plots of voltage gain versus frequency

25
 Input
Signal
10 30 50 70 100 300 500 700 1k 3k 5k 7k 10k
Frequency
(Hz)
Output
Signal
Amplitude
(mV)

Voltage
Gain (dB)

Table 1-6 Measured results of second order band-pass filter. (C3 = C4 = 3.3 nF)

Figure 1-19 Bode plots of voltage gain versus frequency

26
 SECOND ORDER BAND REJECT FILTER

The objective of band stop filter is to make a certain low and high frequency signal pass
through, but disable a certain frequency band signal pass through. On the other words, if a
circuit rejects a finite frequency band that does not include zero (DC) and infinite
frequency, then it is known as band reject filter. Thus band reject filter is specified by two
stop band frequencies to set the frequency band as shown in figure 1-11. Form figure 1-11,
we can see that the characteristic curve is a combination of low-pass filter and a f3dB = f2
high-pass filter. Moreover, the center frequency is f o  f1  f 2 and 3 dB frequency band
(BW3dB) is f2-f1.

Figure 1-12 shows the block diagram of the second order active band reject filter, which
consists of a low-pass filter, high-pass filter and an adder. Let the output the low-pass
filter be FLPF(S) and the output of the high-pass filter be VHPF(S). Then transfer function is
expressed as

Vout(S) = - (K1 · VLPF + K2 · VHPF) (1-18)

= - Vin(S)(K1 · TLPF +K2 · THPF)

where

TLPF(S) : The standard transfer function of low-pass filter

THPF(S) : The standard transfer function of high-pass filter
K1 and K2 : The weight of the linear adder.

Substitute TLPF(S) and THPF(S) by using equation (1-2) and equation (1-7), then equation
(1-18) can be rewritten as

Vout (S ) K 2 S 2  K1 o2
 (1-19)
Vin (S )  
S  S  0   02
2

Q

27
 Figure 1-11 Frequency response of band reject filter

Equation (1-19) is standard equation of the second order active band reject filter, if we
follow the block diagram to design the circuit, we can obtain the second order band-pass
filter. Therefore, practically, we just need to sum the outputs of figure 1-5 and figure 1-7
by using the adder in figure 1-13, where │K1 │= │K2│ = 1, then equation (1-19) can be
rewritten as

Vout (S ) S 2   o2
K (1-20)
Vin (S )  
S  S  0   02
2

Q

28
 Figure 1-12 Block diagram of second order active band reject filter

Figure 1-13 Circuit diagram of second order active band reject filter

All the above-mentioned circuits are second order filters. By comparing their transfer
functions, we know that the order of the filters is determined by the denominator. If the
order of the denominator is higher than the order of the numerator, then the filter is a low-
pass filter. On the other hand, if the order of the denominator is smaller than the order of
the numerator, then the filter is a high-pass filter. For higher levels filter, we can series the
mentioned circuits to obtain the required efficiency (Only the components’ values need to
redesign for satisfying the coefficient of Butterworth or Chebyshev filter). In this
experiment, the IC that we use is LM348 package with four operation amplifier (µA741).
The unit gain bandwidth is about 1 MHz, therefore same as µA741, it is not desirable at
high frequency response. Then for experiment of second order high-pass filter, we change
the IC to LM318. The unit gain bandwidth is about 15 MHz, so we can improve the high
frequency response.

29
 EXPERIMENT 4

Objective
To Study Second Order Band Reject Filter

Requirements
1. IT-4110 Trainer Board
2. 2mm Patch Cords
3. Oscilloscope

Experimental Setup
Refer to the following diagram to configure setup for the present experiment

Procedure:
1. Short circuit LPF point 1with 2 and 4 with 5 so that C1 = C2 =10nF.
2. Now short circuit HPF point 7 with 8 and 10 with 11 so that C1 = C2 =2.2nF.
3. Refer to the block diagram in figure 1-12, connect the output of LPF and HPF to
the LPF and HPF inputs of Band Reject Filter respectively.
4. At the input of Band Reject filter, provide 500mV, 30 Hz sine wave frequency.
Then by using oscilloscope observe the BRF output signal and record the
measured results in table 1-7.

30
5. Now without changing the signal amplitude, change the frequency to 50 Hz, 70
Hz, 100 Hz, 300 Hz, 500 Hz, 700 Hz, 1 kHz, 3 kHz, 5 kHz, 7 kHz, 10 kHz, 30
kHz, 50 kHz and 100 kHz. Then by using the oscilloscope observe the output
signals at the BRF output and record the measured results in table 1-7.
6. Find the voltage gain of each frequency and record the measured results in table 1-
7.
7. From the data in table 1-7, sketch and label the voltage gain in Bode plots in
figure 1-20.
8. Now remove the patch cord from point 1 and 4 and short circuit 2 with 3 and 5
with 6 of LPF so that C3 = C4 = 3.3nF. Also remove patch cords from 7 and 10 and
short circuit 8 with 9 and 11 with 12 of HPF so that C3 = C4 = 1nF.

9. Repeat step 3 to 7 and record the measured results in table 1-8 and figure 1-21.

31
 (C1 = C2 = 10 nF for LPF; C1 = C2 = 2.2nF for HPF)
Input Signal
Frequency 30 50 70 100 300 500 700 1k
(Hz)
Output Signal
Amplitude
(mV)

Voltage Gain
(dB)

Input Signal
Frequency 3k 5k 7k 10k 30k 50k 70k 100k
(Hz)
Output Signal
Amplitude
(mV)

Voltage Gain
(dB)

Table 1-7. Measured results of Second Order Band Reject filter.

Figure 1-20 Bode plots of voltage gain versus frequency

32
 (C1 = C2 = 3.3 nF for LPF; C1 = C2 = 1nF for HPF)
Input Signal
Frequency 30 50 70 100 300 500 700 1k
(Hz)
Output Signal
Amplitude
(mV)

Voltage Gain
(dB)

Input Signal
Frequency 3k 5k 7k 10k 30k 50k 70k 100k
(Hz)
Output Signal
Amplitude
(mV)

Voltage Gain
(dB)

Table 1-8. Measured results of Second Order Band Reject filter.

Figure 1-21 Bode plots of voltage gain versus frequency

33
 RF OSCILLATORS

Objectives

1. To understand the basic theory of oscillators.

2. To design and implement the Colpitts and Hartley oscillators.
3. To design and implement the crystal and voltage controlled oscillators.
4. To understand the measurement and calculation of the output frequency.

Theory
Nowadays, wireless communication is widely used and expanded rapidly. Therefore, RF
oscillator becomes one of the important members in wireless communications. The
characteristic of oscillator is that it can produce sinusoidal wave or square wave at output
terminal without any input signal. So oscillator becomes an important role no matter for
modulated signals or carrier signals. In this chapter, we will focus on the theory of
feedback oscillators and the design and implementation of different kinds of oscillators.
Besides, we can also learn to measure and calculate the output of oscillators in this
chapter.
The Operation theory of Oscillators
Figure 2-1 shows the basic block diagram of the oscillator circuit. It includes an amplifier
and a resonator, which comprise the positive feedback network. When we switch on the
power, the circuit will produce noise. The noise will be amplified by the amplifier, and
pass through a resonator circuit which has filter function. At last what’s left is the signal
in the pass-band. The unwanted signal is filtered by the resonator. So the pass through
signal will then send to the input port of amplifier and combine to the original signal,
which their phases are same and will be amplified and combine to the original signal,
which their phases are same and will be amplified again. In figure 2-1, the transfer
function can be expressed as

Vo ( j ) A( j )
Af ( j )   (2-1)
Vi ( j ) 1  A( j )  ( j )

The definition of open loop gain is

L(jω) = A(jω)β(jω)

Using Barkhausen principle, we know the oscillation condition is

L(jωo) = A(jωo)β(jωo) = 1

Therefore, we can obtain a specific corner frequency ωo to ensure that the open loop gain
L(ωo) is equal to 1, and the phase must be 0o, that is

34
 │A(jωo)β(jωo)│ = 1 (2-3)

arg[A(jωo)β(jωo)] = 0o (2-4)

Figure 2-1 Basic block diagram of oscillator circuit

From the above mention, in order to satisfy equations (2-3) and (2-4), we should make
sure that the product of the feedback factor and the amplifier gain is 1. Meanwhile, the
total summation of the phases is zero after feedback. Therefore, figure 2-1 can be changed
to figure 2-2 for different structures of amplifier.

Figure 2-2 Oscillator circuits comprised by non-inverting and inverting amplifier

Colpitts and Hartley Oscillators

Figure 2-3 shows the basic structure of LC feedback oscillator which Z1, Z2 and Z3
represent inductance or capacitance components. Figure 2-4 is a small signal equivalent
circuit for LC feedback oscillator. From figure 2-4, we get

35
 (Z 2  Z 3 ) Z1 Z2
A    ( A)  
ro  ( Z 2  Z3 ) Z1 Z2  Z3
(2-5)
Z1  Z 2
 ( A) 
ro  ( Z1  Z 2  Z 3 )  Z1 (Z 2  Z 3 )

Let Zi = jXi , where ZL = jXL = jωL ; ZC = jXC = J(-1/ωC)

substitute into equation (1-5), we get

( jX 1 )  ( jX 2 )
A    ( A) 
ro  ( jX 1  jX 2  jX 3 )  jX 1 ( jX 2  jX 3 )
(2-6)
A  X1  X 2

ro  j ( X 1  X 2  X 3 )  X 1 ( X 2  X 3 )

From equation (2-4), we know that the Aβ is real number, therefore, the first condition for
LC feedback oscillator to oscillator to oscillate is

X1 + X2 + X3 = 0 (2-7)

Since equation (2-3) │A(jωo) ·β(jωo)│ = 1, then

A  X1  X 2 X
A( jo )   ( jo )  A 2 1
 X1  ( X 2  X 3 ) X1

So, the second condition is

X1
A (2-8)
X2

36
 Figure 2-3 Feedback oscillator diagram

Figure 2-4 Small signal equivalent circuit for LC feedback oscillator

From the above-mentioned terms, we can make a conclusion: The basic diagram of an
oscillator includes an amplifier and a resonator to form a feedback network. When we
switch on the power, the circuit will produce noise. The noise will be amplified by the
amplifier, and pass through a resonator circuit which has filter function. At last what’s left
is the signal in the pass-band. The unwanted signal is filtered by the resonator. So the pass
through signal will then send to the input port of amplifier and combine with the original
signal, which their phases are same and be amplified again. This is how the oscillation
been formed. On the other hand, base on Barkhausen oscillation principle, the first and
second conditions inform us:
1. Since the voltage gain of the amplifier is real number, therefore Z1 and Z2 are
same components with same reactance and Z3 is another component with different
reactance.

37
 2. The voltage gain, A of the amplifier must be greater than the ratio of Z1 and Z2.

Figure 2-5 shows three common types of oscillators, which are Colpitts, Hartley and
Clapp. If we combine the oscillators with transistor by utilizing either common gate mode,
common drain mode or common source mode, then there are many types of oscillators
mode for selection.

Figure 2-6 is the AC equivalent circuit of Colpitts oscillator. The parallel LC resonant
circuit links between the base and collector of transistor. So part of the voltage come from
the voltage divider formed by C1 and C2, and feedback to the base of the transistor. R
represents the total summation of output resistor, load resistor together with the
equivalent resistor of the inductor and capacitor of a transistor.

(a) Colpitts Oscillator (b) Hartley Oscillator (c) Clapp Oscillator

Figure 2-5. Three common types of oscillators

Figure 2-6 AC equivalent circuit of Colpitts oscillator

If the operation frequency is low, then we can ignore the internal junction capacitance of
the transistor. Hence, from equation (2-7), the oscillation frequency of Colpitts oscillator
is

1 1
X1  X 2  X 3    o L  0
oC1 oC2

38
 1
fo  ( Hz ) (2-9)
CC
2 L 1 2
C1  C2

We need to consider the conditions of Colpitts oscillator. The voltage gain of A of the
amplifier is gmR. Then, from equation (2-8), we know that condition of oscillation is

1

X oC1 C2
A 1  
X2 oC2 C1

C2
gm R  (2-10)
C1

Figure 2-7 is the circuit diagram of Colpitts oscillator. R1, R2, R3 and R4 provide operation
bias to transistor, C1 is coupling capacitor, C2 is bypass capacitor, C3, C4 and L1 comprise
a resonant circuit for selecting suitable operation frequency.

Figure 2-7 Circuit diagram of Colpitts oscillator

39
 Figure 2-8 AC equivalent circuit of Hartley oscillator

Figure 2-8 is the AC equivalent circuit of Hartley oscillator. Same as Colpitts oscillator,
the parallel LC resonant circuit connects between the base and collector of the transistor,
the difference is part of the voltage come from the voltage divider formed by L1 and L2,
and feedback to the base of the transistor. R represents the total summation of output
resistor, load resistor together with the equivalent resistor of the inductor and capacitor of
a transistor.

If the operation frequency is low, then we can ignore the internal junction capacitance of
the transistor. Therefore from equation (2-19), the oscillation frequency can be obtained
as
1
X 1  X 2  X 3  o L1  o L2  0
o C

1
fo  ( Hz ) (2-11)
2 ( L1  L2 )C

Similarly, from equation (2-20), we can obtain the condition of oscillation as

X 1 o L1 L1
A  
X 2 o L2 L2

L1
gm R  (2-12)
L2

Figure 2-9 is circuit diagram of Hartley oscillator. R1, R2 and R3 provide the operation
bias to transistor, C1 is coupling capacitor, C2 is bypass capacitor, C3, L1 and L2 comprise
a resonant circuit for selecting suitable operation frequency.

40
 Figure 2-9 Circuit diagram of Hartley oscillator
Crystal Oscillator
In order to get better frequency stability, it is obvious that we have to choose a high Q
circuit when designing the oscillator circuits, such as transistors with piezoelectric effect,
for example, quartz, ceramic and so on. These transistors are usually used to design the
oscillator circuits with high stability due to the reason that the loss of the transistors is
very low and the Q value of the transistors is very high and stable.

Crystals are tri-dimensional structure. It is mechanical oscillator, which has various types
of oscillation. A crystal is a device that is usually made by cutting a pure quartz crystal in
a very thin slice and then plating the faces with a conductor in order to make an electrical
connection. The property that makes the crystal useful in designing the oscillator is the
piezoelectric effect. When the crystal is excited by the voltage, it will cause a deformation
of the quartz material and produce various types of oscillation. In addition, we can choose
specific oscillation type and high order harmonic via different product process of crystals.
Figure 2-10 shows the equivalent circuit and the impedance characteristic of crystal. In
figure 2-10, the parallel capacitor Cp is the static capacitor in the range about 7 ~ 10 pF.
The series capacitor Cs and inductor L correspond to the disposal sequence and mass of
the crystal. Generally, the value of C8 is about 0.05 pF, and L is about 10 H. The internal
loss is represented by resistor r, which mainly comes from plating, brace of the crystal
and the impedance caused by inner-friction or leads etc. since the Q value of crystal is
very high, therefore, r seems to be very small, only few ohms. Besides, we also can get
the series or parallel resonant frequency, respectively. In figure 2-10, we have

1
fs  (2-13)
2 LCs

41
 1
fp  (2-14)
Cs C p
2 L
Cs  C p

(a) Equivalent circuit of crystal (b) Characteristic curve of impedance

Figure 2-10 Equivalent circuit of crystal and the characteristic curve of impedance

Since Cp ≈ 140 Cs, then the difference between fs and fp is around 0.36%.

C8
f p  (1  ) fs (2-15)
2Cp

Crystal always plays a role as the parallel or series resonant circuit in oscillator circuit.
Due to the high Q value of crystals, the stability of oscillation frequency can be higher
than using the general inductors and capacitors. If crystal is used in a parallel resonant
circuit, then it is called as the parallel mode crystal oscillator, as shown in figure 2-11(a).
In the oscillation circuit with parallel mode, the crystal can be seen as an inductor. On the
other hand, if the crystal is operated in series resonant circuit, then it is called as the series
mode crystal oscillators as shown in figure 2-11(b). In the oscillation circuit with series
mode, the crystal can be seen as a capacitor. Besides, the design of crystal oscillator is
similar to the design methods without using crystal. However, we should pay more
attention to design the bias circuit for the reason that the DC signal may not pass through
the crystal.

42
 (a) Parallel mode crystal oscillator circuit (b) Series mode crystal oscillator circuit

Figure 2-11 Circuit structures of crystal oscillator

Figure 2-12 Circuit diagram of the Colpitts crystal oscillator

Figure 2-12 is the circuit diagram of the Colpitts crystal oscillator. The operating bias of
the transistor is provided by R1, R2 and R3. Moreover, C1 and C2 are the external parallel
capacitors added on the crystal. The values that we choose should be higher until the
parasitic capacitor can be neglected. The bypass capacitor and coupled capacitor are
denoted as C3 and C4, respectively. The oscillation frequency of this circuit is decided by
the frequency of the crystal oscillator we used.

43
Voltage Controlled Oscillator
Voltage controlled oscillator is an oscillator circuit that the output frequency can be varied
by voltage. The main design concepts and methods are similar to the LC feedback
oscillator as mentioned before. However, the only difference is that we use varactor diode,
which the capacitance can be varied by the voltage to replace the original capacitor.
Therefore, we may not discuss the theory of the oscillator but we will focus on the theory
of the varactor diode.

Varactor diode or tuning diode is mainly used for changing the capacitance value of
oscillator. The objective is to let the output frequency of oscillator can be adjusted or
tunable, therefore varactor diode dominates the tunable range of the whole voltage
controlled oscillator. Varactor diode is a diode, which its capacitance can be varied by
adding a reverse bias voltage to PN junction. When reverse bias voltage increases, the
depletion region becomes wide, this will cause the capacitance value decreases;
nevertheless when reverse bias voltage decreases, the depletion region will be reduced,
this will cause the capacitance value increases. Varactor diode also can be varied by the
amplitude of AC signal.

Figure 2-13 is the capacitance analog diagram of varactor diode. When a varactor diode
without bias voltage, the concentration will be differed from minor carriers at pn junction.
Then these carriers will diffuse and become depletion region. The p type depletion region
carries negative ions. We can use parallel plate capacitor to obtain the expression as
shown as follow

A
C (2-16)
d

where

ε = 11.8εo (dielectric constant of silicon)

εo = 8.85 × 10-12
A: the cross section area of capacitor.
d: the width of depletion region

When reverse bias voltage increases, the width of depletion region d will increase but the
cross section area A remains same, therefore the capacitance value would be reduced. On
the other hand, the capacitance value will increase voltage when reverse bias decreases.

Varactor diode can be equivalent to a capacitor series a resistor (Rs) and an inductor (Ls)
as shown in figure 2-14. From figure 2-14, Cj is junction capacitor of semiconductor,
which only exits in PN junction. Rs is the sum bulk resistance and contact resistance of
semiconductor material, which is related to the quality of varactor diode (generally below
a few ohm). Ls is the equivalent inductor of bounding wire and semiconductor material.

44
Figure 2-20 is the circuit diagram of voltage controlled oscillator by using the structure of
Clapp oscillator in figure 2-10(c). R1, R2 and R3 provide the operating bias voltage of the
transistor. C2, C3, L1, CV1 and CV2 comprise the resonant circuit to select a proper
operation frequency. Finally, C1 is the bypass capacitor and C4 is the coupled capacitor.

Figure 2-13 Capacitance analog diagram of varactor diode

(a) Circuit symbol (b) Equivalent circuit

Figure 2-14 Circuit symbol and equivalent circuit diagram of varactor diode

45
Figure 2-15 Circuit diagram of voltage controlled oscillator

46
Measured Results

Table 2-1 Measured results of Colpitts and Hartley oscillator

Components Values of Output Signal Waveform

Colpitts Oscillator

L1 :_________________

C3 :_________________ O/P

C4 :_________________

Theoretical value fo = __________

Measured value fo = ___________

Components Values of Output Signal Waveform

Hartley Oscillator

L2 :_________________

C3 :_________________ O/P

C5 :_________________

Theoretical value fo = __________

Measured value fo = ___________

47
Table 2-2 Measured results of Crystal Oscillator

Output
Signal Components Values of Hartley Oscillator
Waveforms

C2 :________; C3 :__________; X’tall :_________

O/P

Theoretical value fo = __________

Measured value fo = ___________

Table 2-3 Measured results of voltage controlled oscillator

Input DC
3 4 5 6 7 8 9 10 11 12
Bias (Vt)

Output
Signal
Frequency
(MHz)

48
Figure 2-16 Characteristic curve of frequency versus voltage

49