You are on page 1of 49

10 - RF Oscillators

The information in this work has been obtained from sources believed to be reliable. The author does not guarantee the accuracy or completeness of any information presented herein, and shall not be responsible for any errors, omissions or damages as a result of the use of this information.

April 2012

2006 by Fabian Kung Wai Lee

Main References
[1]* D.M. Pozar, Microwave engineering, 2nd Edition, 1998 John-Wiley & Sons. [2] J. Millman, C. C. Halkias, Integrated electronics, 1972, McGraw-Hill. [3] R. Ludwig, P. Bretchko, RF circuit design - theory and applications, 2000 Prentice-Hall. [4] B. Razavi, RF microelectronics, 1998 Prentice-Hall, TK6560. [5] J. R. Smith,Modern communication circuits,1998 McGraw-Hill. [6] P. H. Young, Electronics communication techniques, 5th edition, 2004 Prentice-Hall. [7] Gilmore R., Besser L.,Practical RF circuit design for modern wireless systems, Vol. 1 & 2, 2003, Artech House. [8] Ogata K., Modern control engineering, 4th edition, 2005, Prentice-Hall.

April 2012

2006 by Fabian Kung Wai Lee

Agenda
Positive feedback oscillator concepts. Negative resistance oscillator concepts (typically employed for RF oscillator). Equivalence between positive feedback and negative resistance oscillator theory. Oscillator start-up requirement and transient. Oscillator design - Making an amplifier circuit unstable. Constant |1| circle. Fixed frequency oscillator design. Voltage-controlled oscillator design.

April 2012

2006 by Fabian Kung Wai Lee

1.0 Oscillation Concepts

April 2012

2006 by Fabian Kung Wai Lee

Introduction
Oscillators are a class of circuits with 1 terminal or port, which produce a periodic electrical output upon power up. Most of us would have encountered oscillator circuits while studying for our basic electronics classes. Oscillators can be classified into two types: (A) Relaxation and (B) Harmonic oscillators. Relaxation oscillators (also called astable multivibrator), is a class of circuits with two unstable states. The circuit switches back-and-forth between these states. The output is generally square waves. Harmonic oscillators are capable of producing near sinusoidal output, and is based on positive feedback approach. Here we will focus on Harmonic Oscillators for RF systems. Harmonic oscillators are used as this class of circuits are capable of producing stable sinusoidal waveform with low phase noise.
April 2012 2006 by Fabian Kung Wai Lee 5

2.0 Overview of Feedback Oscillators

April 2012

2006 by Fabian Kung Wai Lee

Classical Positive Feedback Perspective on Oscillator (1)


Consider the classical feedback system with non-inverting amplifier, Assuming the feedback network and amplifier do not load each other, we can write the closed-loop transfer function as:
Non-inverting amplifier

Si(s)

+ +

E(s) A(s)
High impedance

So(s)

So ) (s ) = 1 AAs(sF (s ) (2.1a) () Si

Positive Feedback

Feedback network

High impedance

T (s ) = A(s )F (s ) (2.1b)
Loop gain (the gain of the system around the feedback loop)

F(s)

) Writing (2.1a) as: S o (s ) = 1 AAs()sF (s ) S i (s ) ( We see that we could get non-zero output at So, with Si = 0, provided 1-A(s)F(s) = 0. Thus the system oscillates!
April 2012 2006 by Fabian Kung Wai Lee 7

Classical Positive Feedback Perspective on Oscillator (1)


The condition for sustained oscillation, and for oscillation to startup from positive feedback perspective can be summarized as:
For sustained oscillation For oscillation to startup

1 A(s )F (s ) = 0

Barkhausen Criterion

(2.2a) (2.2b)

A(s )F (s ) > 1

arg( A(s )F (s )) = 0

Take note that the oscillator is a non-linear circuit, initially upon power up, the condition of (2.2b) will prevail. As the magnitudes of voltages and currents in the circuit increase, the amplifier in the oscillator begins to saturate, reducing the gain, until the loop gain A(s)F(s) becomes one. A steady-state condition is reached when A(s)F(s) = 1.
Note that this is a very simplistic view of oscillators. In reality oscillators are non-linear systems. The steady-state oscillatory condition corresponds to what is called a Limit Cycle. See texts on non-linear dynamical systems.
April 2012 2006 by Fabian Kung Wai Lee 8

Classical Positive Feedback Perspective on Oscillator (2)


Positive feedback system can also be achieved with inverting amplifier:
Inverting amplifier + -

Si(s)

E(s) -A(s) Inversion F(s)

So(s)

So ) (s ) = 1 AAs(sF (s ) () Si

To prevent multiple simultaneous oscillation, the Barkhausen criterion (2.2a) should only be fulfilled at one frequency. Usually the amplifier A is wideband, and it is the function of the feedback network F(s) to select the oscillation frequency, thus the feedback network is usually made of reactive components, such as inductors and capacitors.
April 2012 2006 by Fabian Kung Wai Lee 9

Classical Positive Feedback Perspective on Oscillator (3)


In general the feedback network F(s) can be implemented as a Pi or T network, in the form of a transformer, or a hybrid of these. Consider the Pi network with all reactive elements. A simple analysis in [2] and [3] shows that to fulfill (2.2a), the reactance X1, X2 and X3 need to meet the following condition:
+ -

E(s)

-A(s)

So(s)

X 3 = ( X 1 + X 2 ) (2.3)

X3

If X3 represents inductor, then X1 and X2 should be capacitors. X2

X1

April 2012

2006 by Fabian Kung Wai Lee

10

Classical Feedback Oscillators

+ -

The following are examples of oscillators, based on the original circuit using vacuum tubes.
+ + -

Colpitt oscillator
+

Armstrong oscillator

Hartley oscillator Clapp oscillator

April 2012

2006 by Fabian Kung Wai Lee

11

Example of Tuned Feedback Oscillator (1)


A 48 MHz Transistor Common -Emitter Colpitt Oscillator
R RC R=330 Ohm

2.0 1.5 1.0

VB, V VL, V
VL R RL R=220 Ohm

V_DC SRC1 Vdc=3.3 V R RB1 R=10 kOhm

C CD1 C=0.1 uF

0.5 0.0 -0.5

VC C Cc2 C=0.01 uF pb_mot_2N3904_19921211 Q1 R RB2 R=10 kOhm R RE R=220 Ohm C CE C=0.01 uF

-1.0 -1.5

VB C Cc1 C=0.01 uF

A( )F ( )
1 0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

time, usec

t E(s) -A(s) So(s)

Si(s)
L C L1 C1 L=2.2 uH C=22.0 pF R= C C2 C=22.0 pF

+ -

F(s)
April 2012 2006 by Fabian Kung Wai Lee 12

Example of Tuned Feedback Oscillator (2)


A 27 MHz Transistor Common-Base Colpitt Oscilator
600 400 200
V_DC SRC1 Vdc=3.3 V R RB1 R=10 kOhm C CD1 C=0.1 uF

VE, mV VL, mV
L L1 L=1.0 uH R= C C3 C=4.7 pF R R1 R=1000 Ohm

R RC R=470 Ohm

0 -200 -400 -600 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

VC C Cc2 C=0.1 uF pb_m ot_2N3904_19921211 Q1 VE R RE R=100 Ohm

VL C C1 C=100.0 pF

VB C Cc1 C=0.1 uF

R RB2 R=4.7 kOhm

C C2 C=100.0 pF

time, usec

Si(s)

+ +

E(s) A(s)

So(s)

April 2012

F(s)

2006 by Fabian Kung Wai Lee

13

Example of Tuned Feedback Oscillator (3)


V_DC SRC1 Vdc=3.3 V R RB1 R=10 kOhm C CD1 C=0.1 uF

A 16 MHz Transistor Common-Emitter Crystal Oscillator

R RC R=330 Ohm

VC C Cc2 C=0.1 uF pb_mot_2N3904_19921211 Q1 R RB2 R=10 kOhm R RE R=220 Ohm C CE C=0.1 uF

VL R RL R=220 Ohm

VB C Cc1 C=0.1 uF

C C1 C=22.0 pF

sx_stk_CX-1HG-SM_A_19930601 XTL1 Fres=16 MHz

C C2 C=22.0 pF

April 2012

2006 by Fabian Kung Wai Lee

14

Limitation of Feedback Oscillator


At high frequency, the assumption that the amplifier and feedback network do not load each other is not valid. In general the amplifiers input impedance decreases with frequency, and its output impedance is not zero. Thus the actual loop gain is not A(s)F(s) and equation (2.2) breakdowns. Determining the loop gain of the feedback oscillator is cumbersome at high frequency. Moreover there could be multiple feedback paths due to parasitic inductance and capacitance. It can be difficult to distinguish between the amplifier and the feedback paths, owing to the coupling between components and conductive structures on the printed circuit board (PCB) or substrate. Generally it is difficult to physically implement a feedback oscillator once the operating frequency is higher than 500MHz.

April 2012

2006 by Fabian Kung Wai Lee

15

3.0 Negative Resistance Oscillators

April 2012

2006 by Fabian Kung Wai Lee

16

Introduction (1)
An alternative approach is needed to get a circuit to oscillate reliably. We can view an oscillator as an amplifier that produces an output when there is no input. Thus it is an unstable amplifier that becomes an oscillator! For example lets consider a conditionally stable amplifier. Here instead of choosing load or source impedance in the stable regions of the Smith Chart, we purposely choose the load or source impedance in the unstable impedance regions. This will result in either |1 | > 1 or |2 | > 1. The resulting amplifier circuit will be called the Destabilized Amplifier. As seen in Chapter 7, having a reflection coefficient magnitude for 1 or 2 greater than one implies the corresponding port resistance R1 or R2 is negative, hence the name for this type of oscillator.
April 2012 2006 by Fabian Kung Wai Lee 17

Introduction (2)
For instance by choosing the load impedance ZL at the unstable region, we could ensure that |1 | > 1. We then choose the source impedance properly so that |1 s | > 1 and oscillation will start up (refer back to Chapter 7 on stability theory). Once oscillation starts, an oscillating voltage will appear at both the input and output ports of a 2-port network. So it does not matter whether we enforce |1 s | > 1 or |2 L | > 1, enforcing either one will cause oscillation to occur (It can be shown later that when |1 s | > 1 at the input port, |2 L | > 1 at the output port and vice versa). The key to fixed frequency oscillator design is ensuring that the criteria |1 s | > 1 only happens at one frequency (or a range of intended frequencies), so that no simultaneous oscillations occur at other frequencies.

April 2012

2006 by Fabian Kung Wai Lee

18

Recap - Wave Propagation Stability Perspective (1)


From our discussion of stability from wave propagation in Chapter 7
Zs or s Source b1
Port 1 Port 2

2-port Network

a1 = bs + bs 1s + bs 12s 2 + ... bs a1 = 1 1s
b1 = bs 1 + bs 12s + bs 13s 2 + ... b1 = bs 1 1 1s 1 1 1s

Z1 or 1
bs

a1

bs1

bss 1
bss 12

b 1= bs

bss 212
bss 213

Compare with equation (2.1a)

bss 313
bss 314
April 2012

2006 by Fabian Kung Wai Lee

So ) (s ) = 1 AAs(sF (s ) () Si

Similar mathematical form


19

Recap - Wave Propagation Stability Perspective (2)


We see that the infinite series that constitute the steady-state incident (a1) and reflected (b1) waves at Port 1 will only converge provided | s1| < 1. These sinusoidal waves correspond to the voltage and current at the Port 1. If the waves are unbounded it means the corresponding sinusoidal voltage and current at the Port 1 will grow larger as time progresses, indicating oscillation start-up condition. Therefore oscillation will occur when | s1 | > 1. Similar argument can be applied to port 2 since the signals at Port 1 and 2 are related to each other in a two-port network, and we see that the condition for oscillation at Port 2 is |L2 | > 1.

April 2012

2006 by Fabian Kung Wai Lee

20

10

Oscillation from Negative Resistance Perspective (1)


Generally it is more useful to work with impedance (or admittance) when designing actual circuit. Furthermore for practical purpose the transmission lines connecting ZL and Zs to the destabilized amplifier are considered very short (length 0). In this case the impedance Zo is ambiguous (since there is no transmission line). To avoid this ambiguity, let us ignore the transmission line and examine the condition for oscillation phenomena in terms of terminal impedance.
Very short Tline

Zs

Zo

Z1

Destabilized Amp. and Load

April 2012

Z Z s Z Z1

2006 by Fabian Kung Wai Lee

21

Oscillation from Negative Resistance Perspective (2)


We consider Port 1 as shown, with the source network and input of the amplifier being modeled by impedance or series networks.
Zs jXs Source Rs Network V R1 Port 1 Z1 Amplifier with load ZL

jX1 Vamp

Z2 ZL

Port 2

Using circuit theory the voltage at Port 1 can be written as:


V=
April 2012

R1 + jX 1 Z1 Vs = Vs R1 + Rs + j ( X 1 + X s ) Z s + Z1
2006 by Fabian Kung Wai Lee

(3.1)
22

11

Oscillation from Negative Resistance Perspective (3)


Furthermore we assume the source network Zs is a series RC network and the equivalent circuit looking into the amplifier Port 1 is a series RL network.
Zs Z1 Cs Rs V Vs R1 L1
Vamp

Z2 ZL

Using Laplace Transform, (3.1) is written as: R1 + sL1 (3.2a) V (s ) = Vs (s ) 1 R1 + Rs + sL1 + sC s (3.2b) where s = + j
April 2012 2006 by Fabian Kung Wai Lee 23

Oscillation from Negative Resistance Perspective (4)


The expression for V(s) can be written in the standard form according to Control Theory [8]: 2 V 1 (s ) = 1 2 s(RR +RsL1 ) 1 = sCs (R1 + sL1 )n (3.3a) + 2 2 Vs L1 s + s ( L ) + L C s + 2ns + n
1 s 1 1 s

where

R1 + Rs 2 L1 Cs

= Damping Factor

n =

1 L1C s

= Natural Frequency

(3.3b)

The transfer function V(s)/Vs(s) is thus a 2nd order system with two poles p1, p2 given by: 2

p1, 2 = n n 1

(3.4)

Observe that if (R1 + Rs) < 0 the damping factor is negative. This is true if R1 is negative, and |R1| > Rs. R1 can be made negative by modifying the amplifier circuit (e.g. adding local positive feedback), producing the sum R1 + Rs < 0.
April 2012 2006 by Fabian Kung Wai Lee 24

12

Oscillation from Negative Resistance Perspective (5)


Assuming ||<1 (under-damped), the poles as in (3.4) will be complex and exist at the right-hand side of the complex plane. From Control Theory such a system is unstable. Any small perturbation will result in a oscillating signal with frequency n 2 1 that grows exponentially. A small disturbance Im
v(t)

Rs + R1 | o < 0
0

Complex pole pair

or impulse starts the exponentially growing sinusoid

Re

t
Time Domain

Complex Plane

Usually a transient or noise signal from the environment will contain a small component at the oscillation frequency. This forms the seed in which the oscillation builts up.
April 2012 2006 by Fabian Kung Wai Lee 25

Oscillation from Negative Resistance Perspective (6)


When the signal amplitude builds up, nonlinear effects such as transistor saturation and cut-off will occur, this limits the of the transistor and finally limits the amplitude of the oscillating signal. The effect of decreasing of the transistor is a reduction in the magnitude of R1 (remember R1 is negative). Thus the damping factor will approach 0, since Rs+ R1 0. Steady-state sinusoidal oscillation is achieved when =0, or equivalently the poles become

p1, 2 =0 = jn

The steady-state oscillation frequency o corresponds to n,

n 2 = L11 s n L1 = n1Cs X 1 = X s C
X1 + X s = 0
o
April 2012 2006 by Fabian Kung Wai Lee 26

13

Oscillation from Negative Resistance Perspective (7)


From (3.3b), we observe that the steady-state oscillation frequency is determined by L1 and Cs, in other words, X1 and Xs respectively. Since the voltages at Port 1 and Port 2 are related, if oscillation occur at Port 1, then oscillation will also occur at Port 2. From this brief discussion, we use RC and RL networks for the source and amplifier input respectively, however we can distill the more general requirements for oscillation to start-up and achieve steadystate operation for series representation in terms of resistance and reactance:

Rs + R1 |o < 0
X s + X 1 | o = 0
Start-up

(3.5a) (3.5b)

Rs + R1 |o = 0
X s + X 1 | o = 0
Steady-state

(3.6a) (3.6b)

April 2012

2006 by Fabian Kung Wai Lee

27

Illustration of Oscillation Start-Up and Steady-State


The oscillation start-up process and steady-state are illustrated.
1.4 1.2 1.0

0.8

0.6

0.4

0.2

0.0

-0. 2

-0. 4

-0. 6

-0. 8 0 10 20 30 40 50 60 70 80 90 10 0 110 120

R1+Rs
0

Oscillation start-up

Steady-state

Zs Z1 Zs t Destabilized Amplifier

ZL

We need to note that this is a very simplistic view of oscillators. Oscillators are autonomous non-linear dynamical systems, and the steady-state condition is a form of Limit Cycles.
April 2012 2006 by Fabian Kung Wai Lee 28

14

Summary of Oscillation Requirements Using Series Network


By expressing Zs and Z1 in terms of resistance and reactance, we conclude that the requirement for oscillation are.
Zs jXs Source Rs Network V R1 Port 1 Z1 jX1 Vamp

Z2 ZL

Port 2

Rs + R1 |o = 0
X s + X 1 | o = 0
Steady-state

(3.6a) (3.6b)

Rs + R1 |o < 0
X s + X 1 | o = 0
Start-up

(3.5a) (3.5b)

A similar expression for Z2 and ZL can also be obtained, but we shall not be concerned with these here. by Fabian Kung Wai Lee April 2012 2006 29

The Resonator
The source network Zs is usually called the Resonator, as it is clear that equations (3.5b) and (3.6b) represent the resonance condition between the source network and the amplifier input. The design of the resonator is extremely important. We shall see later that an important parameter of the oscillator, the Phase Noise is dependent on the quality of the resonator.

April 2012

2006 by Fabian Kung Wai Lee

30

15

Summary of Oscillation Requirements Using Parallel Network


If we model the source network and input to the amplifier as parallel networks, the following dual of equations (3.5) and (3.6) are obtained.
Port 1 Gs jBs V Z2 G1 jB1 Vamp ZL

The start-up and steady-state conditions are:

Gs + G1 |o = 0
Bs + B1 |o = 0
Steady-state
April 2012

(3.7a) (3.7b)

Gs + G1 |o < 0

(3.8a) (3.8b)
31

Bs + B1 |o = 0
Start-up

2006 by Fabian Kung Wai Lee

Series or Parallel Representation? (1)


The question is which to use? Series or parallel network representation? This is not an easy question to answer as the destabilized amplifier is operating in nonlinear region as oscillator. Concept of impedance is not valid and our discussion is only an approximation at best. We can assume series representation, and worked out the corresponding resonator impedance. If after computer simulation we discover that the actual oscillating frequency is far from our prediction (if theres any oscillation at all!), then it probably means that the series representation is incorrect, and we should try the parallel representation. Another clue to whether series or parallel representation is more accurate is to observe the current and voltage in the resonator. For series circuit the current is near sinusoidal, where as for parallel circuit it is the voltage that is sinusoidal.
April 2012 2006 by Fabian Kung Wai Lee 32

16

Series or Parallel Representation? (2)


Reference [7] illustrates another effective alternative, by computing the large-signal S11 of Port 1 (with respect to Zo) using CAD software. 1/S11 is then plotted on a Smith Chart as a function of input signal magnitude at the operating frequency. By comparing the locus of 1/S11 as input signal magnitude is gradually increased with the coordinate of constant X or constant B circles on the Smith Chart, we can decide whether series or parallel form approximates Port 1 best. We will adopt this approach, but plot S11 instead of 1/S11. This will be illustrated in the examples in next section. Do note that there are other reasons that can cause the actual oscillation frequency to deviate a lot from prediction, such as frequency stability issue (see [1] and [7]).

April 2012

2006 by Fabian Kung Wai Lee

33

4.0 Fixed Frequency Negative Resistance Oscillator Design

April 2012

2006 by Fabian Kung Wai Lee

34

17

Procedures of Designing Fixed Frequency Oscillator (1)


Step 1 - Design a transistor/FET amplifier circuit. Step 2 - Make the circuit unstable by adding positive feedback at radio frequency, for instance, adding series inductor at the base for commonbase configuration. Step 3 - Determine the frequency of oscillation o and extract Sparameters at that frequency. Step 4 With the aid of Smith Chart and Load Stability Circle, make R1 < 0 by selecting L in the unstable region. Step 5 (Optional) Perform a large-signal analysis (e.g. Harmonic Balance analysis) and plot large-signal S11 versus input magnitude on Smith Chart. Decide whether series or parallel form to use. Step 6 - Find Z1 = R1 + jX1 (Assuming series form).

April 2012

2006 by Fabian Kung Wai Lee

35

Procedures of Designing Fixed Frequency Oscillator (2)


Step 7 Find Rs and Xs so that R1 + Rs<0, X1 + Xs=0 at o. We can use the rule of thumb Rs=(1/3)|R1| to control the harmonics content at steady-state. Step 8 - Design the impedance transformation network for Zs and ZL. Step 9 - Built the circuit or run a computer simulation to verify that the circuit can indeed starts oscillating when power is connected. Note: Alternatively we may begin Step 4 using Source Stability Circle, select s in the unstable region so that R2 or G2 is negative at o .

April 2012

2006 by Fabian Kung Wai Lee

36

18

Making an Amplifier Unstable (1)


An amplifier can be made unstable by providing some kind of local positive feedback. Two favorite transistor amplifier configurations used for oscillator design are the Common-Base configuration with Base feedback and Common-Emitter configuration with Emitter degeneration.

April 2012

2006 by Fabian Kung Wai Lee

37

Making an Amplifier Unstable (2)


This is a practical model of an inductor Base bypass capacitor
V_DC SRC1 Vdc=4.5 V

At 410MHz
DC
DC DC1

S-PARAMETERS
S_Param SP1 Start=410.0 MHz Stop=410.0 MHz Step=2.0 MHz

SStabCircle

S_StabCircle S_StabCircle1 SSC=s_stab_circle(S,51)

StabFact

Common Base Configuration

R Rb1 R=10 kOhm

L LC L=330.0 nH R=

StabFact StabFact1 K=stab_fact(S)

LStabCircle

L_StabCircle L_StabCircle1 LSC=l_stab_circle(S,51) Vout

C Cc2 C=10.0 nF L LB L=22 nH R= R RLB R=0.77 Ohm C Cb C=10.0 nF

Term Term2 Num=2 Z=50 Ohm

C CLB C=0.17 pF

pb_phl_BFR92A_19921214 Q1 R Rb2 L R=4.7 kOhm LE L=330.0 nH R=

Vin C Cc1 C=10.0 nF

Term Term1 Num=1 Z=50 Ohm

An inductor is added in series with the bypass capacitor on the base terminal of the BJT. This is a form of positive series feedback.

Positive feedback here


April 2012

R Re R=100 Ohm

2006 by Fabian Kung Wai Lee

38

19

Making an Amplifier Unstable (3)


s22 and s11 have magnitude > 1
S(1,2) 0.162 / 166.9... S(2,1) 2.068 / -12.723 S(2,2) 1.154 / -3.535

freq 410.0MHz freq 410.0MHz

K -0.987 S(1,1) 1.118 / 165.6...

L Plane

Unstable Regions

s Plane

April 2012

2006 by Fabian Kung Wai Lee

39

Making an Amplifier Unstable (4)


DC
DC DC1

S-PARAMETERS
S_Param SP1 Start=410.0 MHz Stop=410.0 MHz Step=2.0 MHz

SStabCircle

S_StabCircle S_StabCircle1 SSC=s_stab_circle(S,51)

V_DC SRC1 Vdc=4.5 V

StabFact

R Rb1 R=10 kOhm

L LC L=330.0 nH R=

StabFact StabFact1 K=stab_fact(S)

LStabCircle

L_StabCircle L_StabCircle1 LSC=l_stab_circle(S,51) Vout

C Cc1 C=1.0 nF Term Term1 Num=1 Z=50 Ohm

C Cc2 C=1.0 nF pb_phl_BFR92A_19921214 Q1 C Ce1 C=15.0 pF

Term Term2 Num=2 Z=50 Ohm

R Rb2 R=4.7 kOhm

Feedback
R Re R=100 Ohm

C Ce2 C=10.0 pF

Common Emitter Configuration Positive feedback here


40

April 2012

2006 by Fabian Kung Wai Lee

20

Making an Amplifier Unstable (5)


S22 and S11 have magnitude > 1
freq 410.0MHz freq 410.0MHz K -0.516 S(1,1) 3.067 / -47.641 S(1,2) 0.251 / 62.636 S(2,1) 6.149 / 176.803 S(2,2) 1.157 / -21.427

L Plane

s Plane

Unstable Regions
April 2012 2006 by Fabian Kung Wai Lee 41

Precautions
The requirement Rs= (1/3)|R1| is a rule of thumb to provide the excess gain to start up oscillation. Rs that is too large (near |R1| ) runs the risk of oscillator fails to start up due to component characteristic deviation. While Rs that is too small (smaller than (1/3)|R1|) causes too much nonlinearity in the circuit, this will result in large harmonic distortion of the output waveform.
V2
Clipping, a sign of too much nonlinearity t

V2 Rs too small For more discussion about the Rs = (1/3)|R1| rule, and on the sufficient condition for oscillation, see [6], which list further requirements.
April 2012 2006 by Fabian Kung Wai Lee

Rs too large
42

21

Aid for Oscillator Design - Constant |1| Circle (1)


In choosing a suitable L to make |L | > 1, we would like to know the range of L that would result in a specific |1 |. It turns out that if we fix |1 |, the range of load reflection coefficient that result in this value falls on a circle in the Smith chart for L . The radius and center of this circle can be derived from:

S D 1 = 11 L 1 S 22L
Assuming = |1 |:
By fixing |1 | and changing L .

Tcenter =

2 S 22* + D*S11 D S 22
2 2 2

(4.1a)

Radius =

S12 S 21 D 2 S 22
2 2

(4.1b)

April 2012

2006 by Fabian Kung Wai Lee

43

Aid for Oscillator Design - Constant |1| Circle (2)


The Constant |1 | Circle is extremely useful in helping us to choose a suitable load reflection coefficient. Usually we would choose L that would result in |1 | = 1.5 or larger. Similarly Constant |2 | Circle can also be plotted for the source reflection coefficient. The expressions for center and radius is similar to the case for Constant |1 | Circle except we interchange s11 and s22, L and s . See Ref [1] and [2] for details of derivation.

April 2012

2006 by Fabian Kung Wai Lee

44

22

Example 4.1 CB Fixed Frequency Oscillator Design


In this example, the design of a fixed frequency oscillator operating at 410MHz will be demonstrated using BFR92A transistor in SOT23 package. The transistor will be biased in Common-Base configuration. It is assumed that a 50 load will be connected to the output of the oscillator. The schematic of the basic amplifier circuit is as shown in the following slide. The design is performed using Agilents ADS software, but the author would like to stress that virtually any RF CAD package is suitable for this exercise.

April 2012

2006 by Fabian Kung Wai Lee

45

Example 4.1 Cont...


Step 1 and 2 - DC biasing circuit design and S-parameter extraction.
DC
DC D C1

S-PAR AME T ER S
S t ab F ac t

S_Param SP1 Start=410. 0 MH z Stop=410.0 MHz Step=2. 0 MH z

StabFact StabFact 1 K=st ab_f act (S )

V_DC SR C1 Vdc =4.5 V

R Rb1 R=10 kO hm

L LC L=330.0 nH R=

Port 2 - Output
Term Term 2 Num=2 Z=50 Ohm

LSt abCircle

L_St abCircle L_St abCircle1 load_s tabcir=l_s tab_c irc le(S, 51)

C Cc2 C =1. 0 nF L LB L=12. 0 nH R= C Cb C =1.0 nF R Rb2 R=4.7 kOhm

SStabCircle

pb_phl_BF R 92A_19921214 Q1 L LE L=220.0 nH R=

S_StabCircle S_StabCircle1 source_s tabc ir=s_st ab_c irc le(S ,51)

LB is chosen carefully so that the unstable regions in both L and s planes are large April enough. 2012

C Cc1 C =1. 0 nF

Term Term 1 Num=1 Z=50 O hm

Port 1

Amplifier

Port 2

R Re R =100 Ohm

Port 1 - Input
2006 by Fabian Kung Wai Lee 46

23

Example 4.1 Cont...


freq 410.0MHz freq 410.0MHz K -0.987 S(1,1) 1.118 / 165.6... S(1,2) 0.162 / 166.9... S(2,1) 2.068 / -12.723 S(2,2) 1.154 / -3.535

Unstable Regions

Load impedance here will result in |1| > 1


April 2012

Source impedance here will result in |2| > 1


47

2006 by Fabian Kung Wai Lee

Example 4.1 Cont...


Step 3 and 4 - Choosing suitable L that cause |1 | > 1 at 410MHz. We plot a few constant |1 | circles on the L plane to assist us in choosing a suitable load reflection coefficient.
This point is chosen because it is on real line and easily matched.

LSC |1 |=1.5 |1 |=2.0

L = 0.5<0
ZL = 150+j0

L Plane
April 2012 2006 by Fabian Kung Wai Lee

|1 |=2.5 Note: More difficult to implement load impedance near edges of Smith Chart
48

24

Example 4.1 Cont...


Step 5 To check whether the input of the destabilized amplifier is closer to series or parallel form. We perform large-signal analysis and observe the S11 at the input of the destabilized amplifier.
LSSP
Var Eqn

V_DC SRC1 Vdc=4.5 V R RB1 R=10 kOhm L LC L=330.0 nH R=

LSSP HB1 Freq[1]=410.0 MHz Order[1]=5 LSSP_FreqAtPort[1]= SweepVar="Poutv" Start=-20 Stop=-5 Step=0.2

VAR VAR1 Poutv=-10.0

Large-signal S-parameter Analysis control in ADS software.

C Cc2 C=1.0 nF L LB L=12.0 nH R= C CB C=1.0 nF R RB2 R=4.7 kOhm

R RL R=150 Ohm

pb_phl_BFR92A_19921214 Q1 L LE L=220.0 nH R= C Cc1 C=1.0 nF

We are measuring large-signal S11 looking towards here P_1Tone


PORT1 Num=1 Z=50 Ohm P=polar(dbmtow(Poutv),0) Freq=410 MHz

R RE R=100 Ohm

April 2012

2006 by Fabian Kung Wai Lee

49

Example 4.1 Cont...


Compare the locus of S11 and the constant X and constant B circles on the Smith Chart, it is clear the locus is more parallel to the constant X circle. Also the direction of S11 is moving from negative R to positive R as input power level is increased. We conclude the Series form is more appropriate. Compare Region where R or G is negative
1 1

Direction of S11 as magnitude of P_1tone source is increased


S(1,1)

Boundary of Normal Smith Chart

Locus of S11 versus P_1tone power at 410MHz (from -20 to -5 dBm)


April 2012

Region where R1 or G1 is positive


Poutv (-20.000 to -5.000)
2006 by Fabian Kung Wai Lee 50

25

Example 4.1 Cont...


Step 6 Using the series form, we find the small-signal input impedance Z1 at 410MHz. So the resonator would also be a series network. For ZL = 150 or L = 0.5<0: S DL 1 = 11 = 1.422 + j 0.479 1 S 22 L

Z1 = Z o

1 + 1 = 10.257 + j 7.851 1 1
R1

X1

Step 7 - Finding the suitable source impedance to fulfill R1 + Rs<0, X1 + Xs=0: 1

Rs =

R1 3.42 3 X s = X1 7.851

April 2012

2006 by Fabian Kung Wai Lee

51

Example 4.1 Cont...


The system block diagram:
Port 1 Zs = 3.42-j7.851 Port 2

Common-Base (CB) Amplifier with feedback

ZL = 150

April 2012

2006 by Fabian Kung Wai Lee

52

26

Example 4.1 Cont...


Step 5 - Realization of the source and load impedance at 410MHz.
Zs= 3.42-j7.851 49.44pF ZL=150 27.27nH

3.42

CB Amplifier @ 410MHz
3.49pF

50

1 C 1 C= = 49.44 pF 7.851 7.851 =


April 2012 2006 by Fabian Kung Wai Lee

Impedance transformation network


53

Example 4.1 Cont... - Verification Thru Simulation

Vpp

BFR92A

Vpp = 0.9V V = 0.45V Power dissipated in the load:

PL = = 0.5
April 2012 2006 by Fabian Kung Wai Lee

1V 2 RL

0.452 = 2.025mW 50 54

27

Example 4.1 Cont... - Verification Thru Simulation


Performing Fourier Analysis on the steady state wave form:
The waveform is very clean with little harmonic distortion. Although we may have to tune the capacitor Cs to obtain oscillation at 410 MHz.

484 MHz
April 2012 2006 by Fabian Kung Wai Lee 55

Example 4.1 Cont... The Prototype


Voltage at the base terminal and 50 Ohms load resistor of the fixed frequency oscillator:

1.4

1.2

1.0

Vbb

0.8

0.6

0.4

0.2

0.0

-0.2

Vout

-0.4

Output port
20 30 40 50 60 70 80 90 100 110 120

-0.6

-0.8 0 10

Startup transient ns
April 2012 2006 by Fabian Kung Wai Lee 56

28

Example 4.2 450 MHz CE Fixed Frequency Oscillator Design


Small-signal AC or S-parameter analysis, to show that R1 or G1 is negative at the intended oscillation frequency of 450 MHz.
0
S-PARAMETERS
V_DC SRC1 Vdc=3.0 V R RB R=47 kOhm S_Param SP1 Start=100.0 MHz Stop=800.0 MHz Step=10.0 MHz

-100 -500

real(Z(1,1))

L LC L=220.0 nH R=

Selection of load resistor as in Example 4.1.

imag(Z(1,1))

-200 -300 -400 -1500 -500 -1000

C Cc2 C=330.0 pF DC_Block DC_Block1 Term Term1 Num=1 Z=50 Ohm

R RL R=150 Ohm

-600 100 200 300 400 500 600 700 800

-2000

freq, MHz
C C1 C=2.2 pF pb_phl_BFR92A_19921214 Q1

C C2 C=4.7 pF

There are simplified expressions to find C1 and C2, see reference [5]. R Here we just trial and RE R=220 Ohm error to get some reasonable values.

0.000

0.020

0.015

imag(Y(1,1))

real(Y(1,1))

-0.005 0.010 -0.010 0.005

Destabilized amplifier
April 2012

-0.015 100 200 300 400 500 600 700 800

0.000

freq, MHz

2006 by Fabian Kung Wai Lee

57

Example 4.2 Cont


The large-signal analysis to check for suitable representation.
LSSP
LSSP HB1 Freq[1]=450.0 MHz Order[1]=7 LSSP_FreqAtPort[1]= Sw eepVar="Poutv" Start=-5 Stop=15 Step=0.2
Var Eqn

VAR VAR1 Poutv=-10.0

V_DC SRC1 Vdc=3.0 V R RB R=47 kOhm L LC L=220.0 nH R=

Since the locus of S11 is close in shape to constant X circles, and it indicates R1 goes from negative value to positive values as input power is increased, we use series form to represent the input network looking towards the Base of the amplifier.

S11
DC_Block DC_Block1 P_1Tone PORT1 Num=1 Z=50 Ohm P=polar(dbmtow (Poutv),0) Freq=450 MHz

C Cc2 C=330.0 pF

R RL R=150 Ohm

S(1,1)

C C1 C=2.2 pF

pb_phl_BFR92A_19921214 Q1

Compare
C C2 C=4.7 pF R RE R=220 Ohm

Boundary of Normal Smith Chart


Poutv (-5.000 to 15.000)

April 2012

2006 by Fabian Kung Wai Lee

Direction of S11 as magnitude of P_1tone source is increased 58 from -5 to +15 dBm

29

Example 4.2 Cont


Using a series RL for the resonator, and performing time-domain simulation to verify that the circuit will oscillate.
VtPWL SRC2 V_Tran=pw l(time, 0ns,0V, 2ns,0.1V, 4ns,0V)
t

1.0
TRANSIENT

V_DC SRC1 Vdc=3.0 V

R RB R=47 kOhm

Tran Tran1 StopTime=100.0 nsec L MaxTimeStep=1.0 nsec LC L=220.0 nH R= VC C Cc2 C=330.0 pF VL R RL R=150 Ohm

0.5

VL, V

0.0 -0.5 -1.0 -1.5 0 20

vL(t)

VB C Cc1 L C=1.0 nF L1 L=39.0 nH R=10

40

60

80

100

C C1 C=2.2 pF

pb_phl_BFR92A_19921214 Q1

Eqn VfL=fs(VL)
0.8

m1 freq= 450.0MHz mag(VfL)=0.733

time, nsec

m1

C C2 C=4.7 pF

R RE R=220 Ohm

0.6

Large coupling capacitor


April 2012 2006 by Fabian Kung Wai Lee

mag(VfL)

0.4

|VL(f)|

0.2

0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

59

freq, GHz

Example 4.3 Parallel Representation


An example where the network looking into the Base of the destabilized amplifier is more appropriate as parallel RC network.
Var VAR Eqn LSSP VAR5 Poutv=1.0 LSSP fo=2300 HB1 Freq[1]=fo MHz Order[1]=8 LSSP_FreqAtPort[1]=fo MHz SweepVar="Poutv" Start=-7 Stop=12 Step=0.2

V_DC VCC Vdc=3.3 V R RB1 R=1000 Ohm

S11
P_1Tone PORT1 Num=1 Z=50 Ohm P=polar(dbmtow(Poutv),0) R Freq=fo MHz RB2 R=1000 Ohm C Cc1 C=1.2 pF C C1 C=0.6 pF {t}

C Cc2 C=1.0 pF pb_phl_BFR92A_19921214 Q1

R RL R=50 Ohm

Direction of S11 as magnitude of P_1tone source is increased from -7 to +12 dBm

S(1,1)

C Cdec1 C=100.0 pF L LC L=2 nH R=0.2

Compare

R C RE C2 R=100 Ohm C=0.7 pF {t}

Poutv (-7.000 to 12.000)

S11 versus Input power


60

April 2012

2006 by Fabian Kung Wai Lee

30

Frequency Stability
The process of oscillation depends on the non-linear behavior of the negative-resistance network. The conditions discussed, e.g. equations (3.1), (3.8), (3.9), (3.10) and (3.11) are not enough to guarantee a stable state of oscillation. In particular, stability requires that any perturbation in current, voltage and frequency will be damped out, allowing the oscillator to return to its initial state. The stability of oscillation can be expressed in terms of the partial derivative of the sum Zin + Zs or Yin + Ys of the input port (or output port). The discussion is beyond the scope of this chapter for now, and the reader should refer to [1] and [7] for the concepts.

April 2012

2006 by Fabian Kung Wai Lee

61

Some Steps to Improve Oscillator Performance


To improve the frequency stability of the oscillator, the following steps can be taken. Use components with known temperature coefficients, especially capacitors. Neutralize, or swamp-out with resistors, the effects of active device variations due to temperature, power supply and circuit load changes. Operate the oscillator on lower power. Reduce noise, use shielding, AGC (automatic gain control) and biasline filtering. Use an oven or temperature compensating circuitry (such as thermistor). Use differential oscillator architecture (see [4] and [7]).

April 2012

2006 by Fabian Kung Wai Lee

62

31

Extra References for This Section


Some recommended journal papers on frequency stability of oscillator: Kurokawa K., Some basic characteristics of broadband negative resistance oscillator circuits, Bell System Technical Journal, pp. 19371955, 1969. Nguyen N.M., Meyer R.G., Start-up and frequency stability in highfrequency oscillators,IEEE journal of Solid-State Circuits, vol 27, no. 5 pp.810-819, 1992. Grebennikov A. V., Stability of negative resistance oscillator circuits, International journal of Electronic Engineering Education, Vol. 36, pp. 242-254, 1999.

April 2012

2006 by Fabian Kung Wai Lee

63

Reconciliation Between Feedback and Negative Resistance Oscillator Perspectives


It must be emphasized that the circuit we obtained using negative resistance approach can be cast into the familiar feedback form. For instance an oscillator circuit similar to Example 4.2 can be redrawn as:
V_DC VCC Vdc=3.0 V

Negative Resistance Oscillator


L LC L=2.2 nH {t} R=0.2 R RB1 R=10000 Ohm {t} C Cc2 C=1.0 pF R RL R=50 Ohm

V_DC VCC Vdc=3.0 V L LC L=2.2 nH {t} R=0.2 R RB1 R=10000 Ohm {t}

Amplifier
VL R RL R=50 Ohm

VL

C Cc2 C=1.0 pF pb_phl_BFR92A_19921214 Q1

L L1 L=15.0 nH {t} R=0.1

pb_phl_BFR92A_19921214 Q1 C Cc1 C=4.7 pF C C1 C=1.0 pF {t}


C Cc1 C=4.7 pF C C1 C=1.0 pF {t}

R RE R=100 Ohm {t}

C C2 C=0.8 pF {t}

R RE R=100 Ohm {t}

L L1 L=15.0 nH {t} R=0.1

C C2 C=0.8 pF {t}

Feedback Network

April 2012

2006 by Fabian Kung Wai Lee

64

32

5.0 Voltage Controlled Oscillator

April 2012

2006 by Fabian Kung Wai Lee

65

About the Voltage Controlled Oscillator (VCO) (1)


A simple transistor VCO using Clapp-Gouriet or CE configuration will be designed to illustrate the principles of VCO. The transistor chosen for the job is BFR92A, a wide-band NPN transistor which comes in SOT-23 package. Similar concepts as in the design of fixed-frequency oscillators are employed. Where we design the biasing of the transistor, destabilize the network and carefully choose a load so that from the input port (Port 1), the oscillator circuit has an impedance (assuming series representation is valid):

Z1 ( ) = R1 ( ) + jX 1 ( )

Of which R1 is negative, for a range of frequencies from 1 to 2.


Lower
April 2012 2006 by Fabian Kung Wai Lee

Upper
66

33

About the Voltage Controlled Oscillator (VCO) (2)

Zs

Clapp-Gouriet Oscillator Circuit with Load

ZL

Z1 = R1 + jX1

April 2012

2006 by Fabian Kung Wai Lee

67

About the Voltage Controlled Oscillator (VCO) (3)


If we can connect a source impedance Zs to the input port, such that within a range of frequencies from 1 to 2:

Z s ( ) = Rs ( ) + jX s ( ) Rs ( ) < R1 ( ) R1 ( ) < 0

X s ( ) = X 1 ( )

The circuit will oscillate within this range of frequencies. By changing the value of Xs, one can change the oscillation frequency.
The rationale is that only the initial spectral of the noise signal fulfilling Xs = X1 will start the oscillation.

For example, if X1 is positive, then Xs must be negative, and it can be generated by a series capacitor. By changing the capacitance, one can change the oscillation frequency of the circuit. If X1 is negative, Xs must be positive. A variable capacitor in series with a suitable inductor will allow us to adjust the value of Xs.
April 2012 2006 by Fabian Kung Wai Lee 68

34

Schematic of the VCO


Initial noise source to start the oscillation
DC
DC D C1

T R ANS IE NT
Tran Tran1 StopTim e=100. 0 ns ec MaxTimeS tep=1.2 nsec

P ARAM ET ER SWEEP
ParamSweep Sweep1 SweepVar="R load" SimI ns tanc eNam e[1] ="Tran1" SimI ns tanc eNam e[2] = SimI ns tanc eNam e[3] = SimI ns tanc eNam e[4] = SimI ns tanc eNam e[5] = SimI ns tanc eNam e[6] = St art=100 St op=700 V ar VAR E qn St ep=100 VAR 1 X=1. 0 R load=100

VtP WL Vtrig V_Tran=pwl(t ime, 0ns , 0V, 1ns,0. 01V, 2ns ,0V) R Rb R=47 k Ohm V_DC Vcc Vdc =3.0 V

L Lc L=220.0 nH R=

Variable capacitance tuning network

L L2 L=47. 0 nH R=

R C Rout C c2 C =330. 0 pF R=50 O hm R RL R=Rload

C Cb1 C=2. 2 pF

pb_phl_BF R92A_19921214 Q1

C Cb3 C=4. 7 pF C Cb2 C=10. 0 pF

R Re R=220 O hm

R V _D C R1 R=4700 Ohm S RC1 V dc=-1.5 V

di_s ms _bas 40_19930908 D1 C C b4 C =4.7 pF

2-port network

April 2012

2006 by Fabian Kung Wai Lee

69

More on the Schematic


L2 together with Cb3, Cb4 and the junction capacitance of D1 can produce a range of reactance value, from negative to positive. Together these components form the frequency determining network. Cb4 is optional, it is used to introduce a capacitive offset to the junction capacitance of D1. R1 is used to isolate the control voltage Vdc from the frequency determining network. It must be a high quality SMD resistor. The effectiveness of isolation can be improved by adding a RF choke in series with R1 and a shunt capacitor at the control voltage. Notice that the frequency determining network has no actual resistance to counter the effect of |R1()|. This is provided by the loss resistance of L2 and the junction resistance of D1.

April 2012

2006 by Fabian Kung Wai Lee

70

35

Time Domain Result

1.0

0.5

0.0

-0.5

-1.0

-1.5 0 10 20 30 40 50 60 70 80 90 100

Vout when Vdc = -1.5V

April 2012

2006 by Fabian Kung Wai Lee

71

Load-Pull Experiment
Peak-to-peak output voltage versus Rload for Vdc = -1.5V.

Vout(pp)

1 100 200 300 400 500 600 700 800

RLoad

April 2012

2006 by Fabian Kung Wai Lee

72

36

Controlling Harmonic Distortion (1)


Since the resistance in the frequency determining network is too small, large amount of non-linearity is needed to limit the output voltage waveform, as shown below there is a lot of distortion.

Vout

April 2012

2006 by Fabian Kung Wai Lee

73

Controlling Harmonic Distortion (2)


The distortion generates substantial amount of higher harmonics. This can be reduced by decreasing the positive feedback, by adding a small capacitance across the collector and base of transistor Q1. This is shown in the next slide.

April 2012

2006 by Fabian Kung Wai Lee

74

37

Controlling Harmonic Distortion (3)


The observant person would probably notice that we can also reduce the harmonic distortion by introducing a series resistance in the tuning network. However this is not advisable as the phase noise at the oscillators output will increase ( more about this later). Control voltage Vcontrol

Capacitor to control positive feedback


DC DC1

DC

TRANSIENT
Tran Tran1 StopTime=280.0 nsec MaxTimeStep=1.2 nsec

VtPWL L Vtrig Lc V_Tran=pwl(time, 0ns, 0V, 1ns,0.01V, 2ns,0V) L=220.0 nH R R= Rb R=47 kOhm I_Probe V_DC I_Probe Iload Vcc IC Vdc=3.0 V C Ccb C=1.0 pF L L2 L=47.0 nH C R= Cb1 pb_phl_BFR92A_19921214 C=6.8 pF Q1 C Cb3 C=4.7 pF C Cb2 C=10.0 pF R Re R=220 Ohm

C Cc2 C=330.0 pF

R Rout R=50 Ohm

R RL R=50 Ohm

R R1 V_DC R=4700 Ohm SRC1 Vdc=0.5 V

di_sms_bas40_19930908 D1 C Cb4 C=0.7 pF

April 2012

2006 by Fabian Kung Wai Lee

75

Controlling Harmonic Distortion (4)


The output waveform Vout after this modification is shown below:

Vout

April 2012

2006 by Fabian Kung Wai Lee

76

38

Controlling Harmonic Distortion (5)


Finally, it should be noted that we should also add a low-pass filter (LPF) at the output of the oscillator to suppress the higher harmonic components. Such LPF is usually called Harmonic Filter. Since the oscillator is operating in nonlinear mode, care must be taken in designing the LPF. Another practical design example will illustrate this approach.

April 2012

2006 by Fabian Kung Wai Lee

77

The Tuning Range


Actual measurement is carried out, with the frequency measured using a high bandwidth digital storage oscilloscope.
410

D1 is BB149A, a varactor manufactured by Phillips Semiconductor (Now NXP).

405 f

MHz

400

395

0.5

1 Vdc

1.5

2.5

April 2012

2006 by Fabian Kung Wai Lee

Volts

78

39

Phase Noise in Oscillator (1)


Since the oscillator output is periodic. In frequency domain we would expect a series of harmonics. In a practical oscillation system, the instantaneous frequency and magnitude of oscillation are not constant. These will fluctuate as a function of time. vosc (t ) = (Vo + mnoise (t )) cos(t + + noise (t )) These random fluctuations are noise, and in frequency domain the effect of the spectra will smear out.
t

Ideal oscillator output


t

fo

2fo

3fo

Smearing

Real oscillator output


April 2012 2006 by Fabian Kung Wai Lee

fo

2fo

3fo

79

Phase Noise in Oscillator (2)


Mathematically, we can say that the instantaneous frequency and magnitude of oscillation are not constant. These will fluctuate as a function of time. As a result, the output in the frequency domain is smeared out.
v(t) T = 1/fo
Leesons expression

LPM 10 log

FkT A

1 8 QL

( )]
fo 2 f offset

t fo v(t) f Contains both phase and amplitude modulation of the sinusoidal waveform at frequency fo f

Large phase noise

t fo

Small phase noise


April 2012 2006 by Fabian Kung Wai Lee 80

40

Phase Noise in Oscillator (3)


Typically the magnitude fluctuation is small (or can be minimized) due to the oscillator nonlinear limiting process under steady-state. Thus the smearing is largely attributed to phase variation and is known as Phase Noise. Phase noise is measured with respect to the signal level at various offset frequencies. Signal level vosc (t ) Vo cos(t + + noise (t )) Phase noise is measured in
v(t) - 90dBc/Hz t 100kHz f fo dBc/Hz @ foffset. dBc/Hz stands for dB down from the carrier (the c) in 1 Hz bandwidth. For example -90dBc/Hz @ 100kHz offset from a CW sine wave at 2.4GHz.
81

Assume amplitude limiting effect Of the oscillator reduces amplitude fluctuation


April 2012 2006 by Fabian Kung Wai Lee

Reducing Phase Noise (1)


Requirement 1: The resonator network of an oscillator must have a high Q factor. This is an indication of low dissipation loss in the tuning network (See Chapter 3a impedance transformation network on Q factor).
Xtune Variation in Xtune due to environment causes small change in instantaneous frequency. Xtune

X1

Tuning Network with High Q f


f

X1 f

Tuning Network with Low Q


f

-X1

2|X1|

-X1 Ztune = Rtune +jXtune


2006 by Fabian Kung Wai Lee

2|X1|

April 2012

82

41

Reducing Phase Noise (2)


A Q factor in the tuning network of at least 20 is needed for medium performance oscillator circuits at UHF. For highly stable oscillator, Q factor of the tuning network must be in excess or 1000. We have looked at LC tuning networks, which can give Q factor of up to 40. Ceramic resonator can provide Q factor greater than 500, while piezoelectric crystal can provide Q factor > 10000. At microwave frequency, the LC tuning networks can be substituted with transmission line sections. See R. W. Rhea, Oscillator design & computer simulation, 2nd edition 1995, McGraw-Hill, or the book by R.E. Collin for more discussions on Q factor. Requirement 2: The power supply to the oscillator circuit should also be very stable to prevent unwanted amplitude modulation at the oscillators output.
April 2012 2006 by Fabian Kung Wai Lee 83

Reducing Phase Noise (3)


Requirement 3: The voltage level of Vcontrol should be stable. Requirement 4: The circuit has to be properly shielded from electromagnetic interference from other modules. Requirement 5: Use low noise components in the construction of the oscillator, e.g. small resistance values, low-loss capacitors and inductors, low-loss PCB dielectric, use discrete components instead of integrated circuits.

April 2012

2006 by Fabian Kung Wai Lee

84

42

Example of Phase Noise from VCOs


Comparison of two VCO outputs on a spectrum analyzer*.

VCO output with high phase noise

VCO output with low phase noise

*The spectrum analyzer internal oscillator must of course has a phase noise of an order of magnitude lower than our VCO under test.

April 2012

2006 by Fabian Kung Wai Lee

85

More Materials
This short discussion cannot do justice to the material on phase noise. For instance the mathematical model of phase noise in oscillator and the famous Leesons equation is not shown here. You can find further discussion in [4], and some material for further readings on this topic: D. Schere, The art of phase noise measurement, Hewlett Packard RF & Microwave Measurement Symposium, 1985. T. Lee, A. Hajimiri, The design of low noise oscillators, Kluwer, 1999.

April 2012

2006 by Fabian Kung Wai Lee

86

43

More on Varactor
The varactor diode is basically a PN junction optimized for its linear junction capacitance. It is always operated in the reverse-biased mode to prevent nonlinearity, which generate harmonics.
As we increase the negative biasing voltage Vj , Cj decreases, hence the oscillation frequency increases. Cj The abrupt junction varactor has high Q, but low sensitivity (e.g. Cj varies little over large voltage change). The hyperabrupt junction varactor Cjo Forward biased has low Q, but higher sensitivity.
0
April 2012

Vj

Reverse biased Linear region

Vj
2006 by Fabian Kung Wai Lee 87

A Better Variable Capacitor Network


The back-to-back varactors are commonly employed in a VCO circuit, so that at low Vcontrol, when one of the diode is being affected by the AC voltage, the other is still being reverse biased. When a diode is forward biased, the PN junction capacitance becomes nonlinear. The reverse biased diode has smaller junction capacitance, and this dominates the overall capacitance of the back-to-back varactor network. This configuration helps to decrease the harmonic distortion. To suppress RF signals Vcontrol
Vcontrol Symbol for Varactor
88

To negative resistance amplifier

At any one time, at least one of the diode will be reverse biased. The junction capacitance of the reverse biased diode will dominate the overall capacitance of the network.
April 2012

Vcontrol

2006 by Fabian Kung Wai Lee

44

Example 5.1 VCO Design for Frequency Synthesizer


To design a low power VCO that works from 810 MHz to 910 MHz. Power supply = 3.0V. Output power (into 50 load) minimum -3.0 dBm.

April 2012

2006 by Fabian Kung Wai Lee

89

Example 5.1 Cont


Checking the d.c. biasing and AC simulation.
DC
DC DC1

S-PARAMETERS
S_Param SP1 Start=0.7 GHz Stop=1.0 GHz Step=1.0 MHz

V_DC SRC1 Vdc=3.3 V R RB R=33 kOhm

b82496c3120j000 LC param=SIMID 0603-C (12 nH +-5%)

100pF_NPO_0603 Cc2 pb_phl_BFR92A_19921214 Q1 4_7pF_NPO_0603 Term Cc1 Term1 Num=1 Z=50 Ohm 2_2pF_NPO_0603 C1 R RL R=100 Ohm

Z11
April 2012

3_3pF_NPO_0603 C2 R RE R=100 Ohm

2006 by Fabian Kung Wai Lee

90

45

Example 5.1 Cont


Checking the results real and imaginary portion of Z1 when output is terminated with ZL = 100 .
m1 freq=775.0MHz m1=-89.579
-40 -50 -60

m2 freq=809.0MHz m2=-84.412

imag(Z(1,1)) real(Z(1,1))

-70 -80 -90 -100 -110 -120 0.70 0.72 0.74 0.76 0.78 0.80 0.82 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00

m2 m1

freq, GHz

April 2012

2006 by Fabian Kung Wai Lee

91

Example 5.1 Cont


The resonator design.
PARAMETER SWEEP
ParamSweep Sweep1 SweepVar="Vcontrol" SimInstanceName[1]="SP1" SimInstanceName[2]= SimInstanceName[3]= SimInstanceName[4]= SimInstanceName[5]= SimInstanceName[6]= Start=0.0 Stop=3 Step=0.5

S-PARAMETERS
S_Param SP1 Start=0.7 GHz Stop=1.0 GHz Step=1.0 MHz

Var Eqn

VAR VAR1 Vcontrol=0.2

L L1 L=10.0 nH R= C C3 C=0.68 pF Term Term1 Num=1 Z=50 Ohm

Vvar V_DC SRC1 Vdc=Vcontrol V L L2 L=33.0 nH R= 100pF_NPO_0603 C2 BB833_SOD323 D1

April 2012

2006 by Fabian Kung Wai Lee

92

46

Example 5.1 Cont


The resonator reactance.
120

-X1 of the destabilized amplifier


m1 freq=882.0MHz m1=64.725 Vcontrol=0.000000 m1

100

-imag(VCO_ac..Z(1,1)) imag(Z(1,1))

80

60

40

Resonator reactance as a function of control voltage


April 2012

20

The theoretical tuning range

0 0.70 0.75 0.80 0.85 0.90 0.95 1.00

freq, GHz
2006 by Fabian Kung Wai Lee 93

Example 5.1 Cont


The complete schematic with the harmonic suppression filter.
TRANSIENT
Tran Tran1 StopT ime=1000.0 nsec MaxTimeStep=1.0 nsec DC DC1

DC

Low-pass filter
b82496c3120j000 L3 param=SIMID 0603-C (12 nH +-5%) 100pF_NPO_0603 Cc2 b82496c3150j000 L4 param=SIMID 0603-C (15 nH +-5%)

VtPWL Src_trigger V_T ran=pwl(time, 0ns,0V, 1ns,0.1V, 2ns,0V) V_DC SRC1 Vdc=3.3 V R RB R=33 kOhm

b82496c3100j000 L1 param=SIMID 0603-C (10 nH +-5%) 4_7pF_NPO_0603 C Cc1 C6 C=2.2 pb_phl_BFR92A_19921214 Q1 pF 2_7pF_NPO_0603 C8 0_47pF_NPO_0603 C9 R RL R=100 Ohm

b82496c3330j000 L2 param=SIMID 0603-C (33 nH +-5%) Vvar V_DC SRC2 Vdc=1.2 V 100pF_NPO_0603 C4 R R1 R=100 Ohm BB833_SOD323 D1

C C5 C=0.68 pF

C C7 C=3.3 pF

R RE R=100 Ohm

April 2012

2006 by Fabian Kung Wai Lee

94

47

Example 5.1 Cont


The prototype and the result captured from a spectrum analyzer (9 kHz to 3 GHz).

Fundamental -1.5 dBm

Harmonic VCO suppression filter


- 30 dBm

April 2012

2006 by Fabian Kung Wai Lee

95

Example 5.1 Cont


Examining the phase noise of the oscillator (of course the accuracy is limited by the stability of the spectrum analyzer used).

-0.42 dBm

Span = 500 kHz RBW = 300 Hz VBW = 300 Hz

300Hz April 2012 2006 by Fabian Kung Wai Lee 96

48

Example 5.1 Cont


VCO gain (ko) measurement setup:

Variable power supply

V_DC SRC1 Vdc=3.3 V R RB R=33 kOhm

b82496c3120j000 L3 param=SIMID 0603-C (12 nH +-5%) 100pF_NPO_0603 Cc2 b82496c3150j000 L4 param=SIMID 0603-C (15 nH +-5%) R Rattn R=50 Ohm Port Vout Num=2

b82496c3100j000 L1 param=SIMID 0603-C (10 nH +-5%) 4_7pF_NPO_0603 Cc1 C C5 C=0.68 pF C C6 C=2.2 pF pb_phl_BFR92A_19921214 Q1 2_7pF_NPO_0603 C8

Spectrum Analyzer

0_47pF_NPO_0603 C9

Vvar Port Vcontrol Num=1 R Rcontrol R=1000 Ohm BB833_SOD323 D1

C C7 C=3.3 pF

R RE R=100 Ohm

April 2012

2006 by Fabian Kung Wai Lee

97

Example 5.1 Cont


Measured results:
fVCO / MHz

950

900

850

800

ko 55 MHz = 40.74 MHz/Volt


1.35 Volt

750 0.0
April 2012

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Vcontrol/Volts
98

2006 by Fabian Kung Wai Lee

49

You might also like