CALIFORNIA STATE UNIVERSITY, NORTHRIDGE

“KU BAND MICROWAVE POWER AMPLIFIER”

A graduate project in partial fulfillment of the requirements

For the Degree of Masters of Science

In Electrical Engineering

By

Chinmay Joshi

May,2018
The thesis of Chinmay Joshi is approved by:

___________________________ ______________________

Dr. Xiaojun(Ashley) Geng Date

___________________________ _______________________

Dr. Jackson, Bradley R. Date

___________________________ _______________________

Dr. Matthew M. Radmanesh, Chair Date

California State University, Northridge

ii
 ACKNOWLEDGEMENT

It gives a great opportunity to express immense pleasure for the completion of my graduate
project, I would like to thank my family and friends who continuously kept me motivated
throughout my master’s studies and stood behind me as a backbone during my entire
journey.

With sincere regards and gratitude, I would like to thank Dr. Matthew Radmanesh, the
chair of my project committee whose guidance and productive criticism helped me in
completion of my graduate project. This was not limited up-to the thesis, but it was beyond.
Advices given during his lectures in classes helped me to improve the basics and to obtain
a thorough knowledge of RF/ Microwave, yet his method of teaching was a more practical
and analytical point of view. I am grateful for having his support during entire coursework.

With regards, I would also like to thank my committee members, Dr. Xiaojun Geng and
Dr. Jackson, Bradley R. for giving their valuable guidance and proving possible
corrections. I am also thankful to the Department of Electrical & Computer Engineering
for offering me necessary facilities and resources for completion of my project.

iii
 TABLE OF CONTENTS

SIGNATURE PAGE ……………………………………………………………………...ii

ACKNOWLEDGEMENT ................................................................................................. iii
TABLE OF CONTENTS ................................................................................................... iv
LIST OF TABLES ............................................................................................................. vi
LIST OF FIGURES .......................................................................................................... vii
ABSTRACT ....................................................................................................................... ix
CHAPTER 1: INTRODUCTION ...................................................................................... 1
CHAPTER 2: AMPLIFIER THEORY ............................................................................... 3
2.1 Amplifier Categories ................................................................................................. 4
2.1.1 Class A ................................................................................................................ 4
2.1.2 Class B ................................................................................................................ 4
2.1.3 Class C ................................................................................................................ 4
2.1.4 Class AB ............................................................................................................. 4
2.2 Class based on Signal-Level ..................................................................................... 5
2.2.1 Small Signal PAs ................................................................................................ 5
2.2.2 Large- Signal PAs ............................................................................................... 5
2.3 Key Parameters of PA Design ................................................................................... 5
2.3.1 S-parameter ......................................................................................................... 6
2.3.2 Stability............................................................................................................... 7
2.3.3 Gain of PA .......................................................................................................... 8
2.4 Design of PA circuit ................................................................................................ 11
2.4.1 DC Biasing Circuitry ........................................................................................ 11
2.4.2 RF/MW Circuit for Designing PAs .................................................................. 12
2.5 Design of Matching Networks with Smith Charts .................................................. 14
2.6 Power Coupler ......................................................................................................... 17
CHAPTER 3 DESIGN CONSIDERATION AND SIMULATION USING NI-AWR.... 18
3.1 Selection of Transistor for Amplifier ...................................................................... 18
3.2 Stability of TIM1314-8UL ...................................................................................... 20
3.3 Unilateral Figure of Merit ....................................................................................... 21
3.4 Power Saturation in Large Signal Amplifier ........................................................... 22

iv
 3.5 Biasing of Transistor ............................................................................................... 24
3.6 Design and Source Matching Networks for Load and Source ................................ 28
3.6.1 Matching Network for Source .......................................................................... 28
3.6.2 Matching Network for Load ............................................................................. 30
CHAPTER 4 MICROSTRIP LINE DESIGN AND CALCULATIONS ......................... 34
4.1 General Theory of Transmission Lines ................................................................... 34
4.2 Basics of Microstrip Lines ...................................................................................... 34
4.4 Gap Capacitance Calculations ................................................................................. 38
CHAPTER 5 POWER COMBINERS .............................................................................. 41
5.1 Theory of Power Combiners ................................................................................... 41
5.2 3-dB Power Coupler (Lange Coupler) .................................................................... 42
5.3 Designing of Lange Coupler ................................................................................... 44
CHAPTER 6 FINAL CIRCUIT FOR A TWO-STAGE POWER AMPLIFIER ............. 46
CHAPTER 7 CONCLUSIONS ........................................................................................ 53
REFRENCES .................................................................................................................... 54
APPENDIX A ................................................................................................................... 55
APPENDIX B ................................................................................................................... 58
APPENDIX C ................................................................................................................... 62
APPENDIX D ................................................................................................................... 68

v
 LIST OF TABLES

Table 1: Specifications of Desired Device.......................................................................... 2

Table 2: U vs 𝐺𝑇𝐺𝑇𝑈, 𝑚𝑎𝑥 [1]. ....................................................................................... 10
Table 3: S-parameters of Transistor [Appendix C]........................................................... 19
Table 4: Electrical Characteristics of Transistor [Appendix C]........................................ 19
Table 5: S-parameters for TIM1314-8UL......................................................................... 21
Table 6 Microstrip Lines Comparison .............................................................................. 40
Table 7 Parameters for 59.6Ω ........................................................................................... 40
Table 8 Parameters for 43.3Ω ........................................................................................... 40
Table 9 Results .................................................................................................................. 53

vi
 LIST OF FIGURES

Figure 1.1 Generalized Block Diagram of Power Amplifier .............................................. 1

Figure 2.1 Class A, B, AB and C Amplifiers [2] ................................................................ 5
Figure 2.2 Two-Port Network ............................................................................................. 6
Figure 2.3 PA in its Generalized Form with reflection Co-efficient .................................. 8
Figure 2.4 Characteristics graph of Transistor (𝐼𝐷-𝑉𝐷𝑆) [1] ........................................... 11
Figure 2.5 Amplifier with DC biasing Circuit [1] ............................................................ 12
Figure 2.6 Generalized Steps for designing RF/MW circuits ........................................... 13
Figure 2.7 Matching Networks Using Series and Shunt Stubs [2] ................................... 14
Figure 2.8 Stubbing using Smith Charts [2] ..................................................................... 15
Figure 2.9 (a & b) Open and Short Stub lengths [2] ......................................................... 16
Figure 2.10 General Block Diagram of N-way Amplifier ................................................ 17
Figure 3.1 Sub-circuit Model of Transistor TIM1314-8UL in AWR ............................... 20
Figure 3.2 S-parameters of Transistor & Simulation in AWR ......................................... 20
Figure 3.3 Pout vs Pin of TIM1314-8UL.......................................................................... 22
Figure 3.4 Different Techniques of Biasing [3] ................................................................ 24
Figure 3.5 Biasing of TIM1314-8UL [3] .......................................................................... 25
Figure 3.6 Biasing circuit in AWR ................................................................................... 27
Figure 3.7 Matching Network Source Side....................................................................... 28
Figure 3.8 Smith Chart of Matching Network(Source) .................................................... 29
Figure 3.9 Matching Network Load Side ......................................................................... 30
Figure 3.10 Smith Chart of (Load) Matching Network .................................................... 31
Figure 3.11 Single Stage Amplifier Using Distributed Networks .................................... 32
Figure 3.12 S-parameters of 1st Stage Amplifier .............................................................. 32
Figure 3.13 𝑃𝑖𝑛 vs 𝑃𝑜𝑢𝑡 of Amplifier at 14.5GHz .......................................................... 33
Figure 4.1 Field Configuration and Geometrics of Microstrip Lines [2] ......................... 35
Figure 4.2 Graph of 𝑍0 vs 𝑤ℎ at 𝜖𝑟 = 6.5[2] .................................................................... 35
Figure 4.2 Geometry for Microstrip Gap Capacitance [3] ................................................ 38
Figure 5.1 Power Couplers and Dividers [4] .................................................................... 41
Figure 5.2 Various types of power couplers/dividers [1] ................................................. 41
Figure 5.3 Layout of Lange Coupler [1], [4] .................................................................... 42

vii
Figure 5.4 Ratio Graph for Lange Coupler ....................................................................... 43
Figure 5.5 Lange Coupler in AWR ................................................................................... 44
Figure 5.6 3-dB Coupling Efficiency................................................................................ 45
Figure 5.7 Phase Difference in Lange Coupler ................................................................. 45
Figure 6.1 Block Diagram for Two Stage Design ............................................................ 46
Figure 6.2 S-parameter Simulation for Two-Stage Design .............................................. 47
Figure 6.4 One Tone Test Bench for the Final Circuit ..................................................... 48
Figure 6.5 𝑃𝑜𝑢𝑡 vs 𝑃𝑖𝑛for Final Schematics .................................................................... 49
Figure 6.6 Evaluation of the Reflection Co-efficient........................................................ 49
Figure 6.7 Two Stage Implementation in AWR ............................................................... 50
Figure 6.8 Layout of the Final Circuit .............................................................................. 51

viii
 ABSTRACT

Ku-Band Microwave Power Amplifier

By

Chinmay Joshi

Master of Science in Electrical Engineering

The aim of the graduate project is to design a Ku-band microwave power amplifier that
operates at 14.5 GHz frequency. A high-power GaAs FET, TIM1314-8UL BY TOSHIBA
ELECTRIC has been used to design a power amplifier. Distributed elements are utilized
to realize the input and output matching networks. The device is designed to achieve a
linear power gain of 7.50dB and an output power up to 7.30dBm. As the design is two-
stage, Lange Coupler is used to couple both stages. Lange Coupler is designed by using
microstrip-line technology based on RO3006C substrate. NI-AWR is used for circuit
simulations. It is one of the leading softwares used in RF & Microwave industries.
Analytical calculations and results are verified using softwares like MATLAB,
RF/Microwave E-Book by Dr. Matthew Radmanesh.

ix
 CHAPTER 1: INTRODUCTION

An amplifier is a device which is accountable for strengthening of a signal; the name for
the device originated from this solely purpose for amplifying the signal strength. The
original concept was use of ‘Triode Vacuum Tubes’ which was based on negative
resistance latterly replaced by transistors. Amplifiers made from transistors were more
robust and dependable. Power amplifiers are widely used in transmitters for the sole
purpose of amplifying the modulated signal from modulators along with sufficient power
gain for transmission through antenna.
Basically, amplifiers are classified based on ‘Point of Operations’ and ‘Signal Level’, Class
A, AB, C are categorized under a point of operations while Small Signal and Large Signals
falls under the Signal Level category.
PAs play an important role in radio and microwave industry which are also called ‘Large
Signal Amplifiers’. General purpose applications of PAs are in surveillance and radar used
in military operations, and communications where there is need of the highest efficiency
and linearity in the system.

Figure 1.1 Generalized Block Diagram of Power Amplifier

Power delivered at the output and gain of amplifiers are the key aspects while designing
them, it is always good to have the highest possible output power for a linear gain of a
power device. While designing of PAs the first step is proper selection of an ‘active device’
i.e. transistor, for this design many datasheets of various manufacturers were analyzed

1
before selecting the correct one. The device stability is checked at operating frequency
which we want to work; it is done by using s-parameters of the device, which is followed
by biasing of the device within its electrical characteristics. Designing of matching
networks is done i.e. input and output matching networks is done by both using lumped as
well as distributed networks with the help of smith charts. The design is “single stage power
amplifier”, by observing the results of and to make the work more challenging we decided
to go for the second stage. The 2-stage design is coupling of two single stages, to do so
Lange Coupler is used to achieve 3dB coupling ratios and 90˚ phase difference between
output port. After completion of the two-stage the design is simulated and verified using
‘NI- AWR Microwave Office’.
The report is divided into chapters to get good idea of the design. The first chapter is an
introduction giving the glimpse about how the report proceeds. The second chapter talks
about design theory of amplifier, the third consist of design procedures, mathematical
calculations to make the device workable moreover it also talks about designing of
matching networks using smith charts.
Fourth and fifth chapters, includes the design of coupler using microstrip-lines and
simulations results of software. The project concludes with what is achieved along with
comparison tables of hand calculations and software results.

PARAMETERS SPECIFIED GOALS

Frequency 14.5GHz
Gain (dB) 7.50dB
Output Power 7.30dBm
Table 1: Specifications of Desired Device

2
 CHAPTER 2: AMPLIFIER THEORY

This chapter is about understanding the theory behind amplifiers, and its only principle of
increasing amplitude and power. Thus, it can be said that the amplification is a process of
increasing output power related to its input. Characteristics of the amplifier includes
constrains like gain, linearity, efficiency, stability and power. Linearity is expressed in
terms of gain of amplifier that is given by the equation (2.1).

𝑉0 = G 𝑉𝐼 (2.1)
𝑉0 = Output Signal
𝑉𝐼 = Input Signal
G = Gain of amplifier
So, gain can be given by;

𝑆𝑖𝑔𝑛𝑎𝑙 𝑜𝑢𝑡𝑝𝑢𝑡 𝑃𝑜𝑤𝑒𝑟

G= (2.2)
𝑆𝑖𝑔𝑛𝑎𝑙 𝐼𝑛𝑝𝑢𝑡 𝑃𝑜𝑤𝑒𝑟

Quality of the signal throughout the operation defines linearity, any non- linearity leads to
harmonic distortion which leads to a reduction in power available at output and gain.
Another parameter that plays important role along with gain is noise, i.e. any unwanted
signal inside the system which also gets amplified along with the input signal. So, to
maintain stability it is must to reduce the noise thus to maintain its stability.
PA’s efficiency can be expressed by equation (2.3).

𝑃𝑜𝑢𝑡
η= (2.3)
𝑃𝑑𝑐

Further in this chapter we discuss various types of amplifiers along with their classification;
calculations of its gain, its biasing techniques, various power couplers and microstrip lines
getting knowledge about these will lead to efficient design of PA.

3
2.1 Amplifier Categories
They are classified by:
I. Based on Operating Point
II. Based on Signal Level
Under Operating Point amplifier operates among the classes given below:
(1) Class A (3) Class C (5) Class D
(2) Class B (4) Class AB
In this project, the device operates in class A region, brief regarding each of them is
discussed below.
2.1.1 Class A
In this class amplifier conducts for an entire span of input cycle, it can be explained as the
period in which amplifier remains in active region, i.e. remains ON for that cycle. [1] The
main advantage is its linearity at the output and distortion which is lowest for this class,
one more advantage is this make design simpler. The maximum efficiency achieved in this
class is 25%.
2.1.2 Class B
In this class device remains active i.e. ON only 50% of its input signal, this means
amplification is carried out during that period. Efficiency of the amplifiers in this class is
higher than the previous one i.e. upto 78.5% efficiency.
2.1.3 Class C
Conduction in this class takes place less than 50 percentage than that of the input cycle; it
can also be said the other way around that amplifier shows its activeness for less than 50%.
In this case efficiency is achieved high as compared to other discussed classes, but
distortion is high.
2.1.4 Class AB
This case is little interesting it works as a Class A for small signals and for large signals it
acts as a Class B. By looking at its name we get an idea about the device which falls under
this class operates in both the classes A and B. In addition to this Class D, E, F, G and H
are established.

4
 Figure 2.1 Class A, B, AB and C Amplifiers [2]

2.2 Class based on Signal-Level

In this, devices are categorized relative to their signal strength, it’s divided into sub-
categories as follows:

2.2.1 Small Signal PAs

In this type signal divergence is very much less from its stable region, due to property of
less deviation it covers a very small part of its operating characteristics, results its operation
in the linear region.
2.2.2 Large- Signal PAs
The signal divergence is very high from its stable region compared to its previous case this
leads to the non-linear operation of the device. Further in this chapter we will discuss about
key constraints required while designing an amplifier.

2.3 Key Parameters of PA Design

S-parameters, stability, gain are the important parameters while designing amplifiers. Each
of them are discuss as follows:

5
2.3.1 S-parameter
The scattering parameters are set of complex numbers with magnitude and phase which
describes the electrical characteristics of any network. This parameter is taken into
consideration while designing any network at higher frequency, thus its function of
frequency for any system. Given below are scattering parameters for single, two and three
port networks.
(𝑆11 ) For single port network

𝑆 𝑆12
( 11 ) For two port networks
𝑆21 𝑆22
Scattering parameters for single, two and three port networks are given as above.
Figure 2.2 indicates two-port network.

Figure 2.2 Two-Port Network

𝑉− 𝑆 𝑆12 𝑉1+
( 1− )= ( 11 )( ) (2.4)
𝑉2 𝑆21 𝑆22 𝑉2+
Equation (2.4), shows the relationship between reflected and incident waves that leads to
description of Two port parameters as shown below.
𝑉− 𝑉−
𝑆11= 𝑉1+ | 𝑉2+ = 0, 𝑆21 = 𝑉2+ | 𝑉2+ = 0 (2.5)
1 1

𝑉2− 𝑉1−
𝑆22 = | 𝑉1+ = 0, 𝑆12= | 𝑉1+ = 0 (2.6)
𝑉2+ 𝑉2+

Where,
𝑆11 is Reflection Coefficient at port1 i.e. input.
𝑆22 is Reflection Coefficient at port 2 i.e. output.
𝑆21 is Forward Gain.
𝑆12 is Reverse Gain.

6
After understanding S-parameters let us now try to get how stability plays an important
role in designing microwave amplifiers.
2.3.2 Stability
Stability is the effectiveness of the device to operate in stable condition despite of changes
or variations in source or load impedance values. Stability of the transistor can be checked
analytically by K-Δ test and µ- test parameters.
K-Δ stability check can be done by following equation for two port networks.
Please refer Book [1] for equations.
Δ = 𝑆11 𝑆22 - 𝑆21 𝑆12 (2.7)

(2.8)
If K > 1, 𝑆22 & 𝑆11 < 1 , this condition achieved for a device then the device is said to
“Unconditionally Stable”. The condition discussed is only for two port networks only, if
the condition is not achieved than another parameter is checked i.e. 𝐵1must be greater than
0, where;
𝐵1 = 1+ |𝑆11 |2 − |𝑆22 |2 − |𝛥|2 (2.9)
For K < 1, this indicates that the entire smith chart is not stable to find the stable region
stability circles are plotted.
To check the stability K-Δ test is checked where K should be greater than one and |∆|
should be less than one.
One more test is used to verify the stability of the transistor which is µ-Test in this we
compare two devices and try to check which transistor is more stable and vice-a-versa.

1−|𝑆11 |2
µ= |𝑆 ∗ (2.10)
22 −𝑆11 ∆|+|𝑆12 𝑆21 |

For a device to be unconditionally stable µ > 1. [1]If this criterion is not satisfied than there
is need to draw stability circle to find stable region on smith chart.

7
2.3.3 Gain of PA
Figure 2.3 shows the generalized diagram of power amplifier including matching networks.

Figure 2.3 PA in its Generalized Form with reflection Co-efficient

Gain is termed as the amount of “amplification” of a device, this constraint plays a major
role while designing PA. Let’s try to understand this term in brief. Equations reference [1].
Consider,
𝑃𝑖𝑛 = Transistor Input Power.
𝑃𝑎𝑣𝑠 = Power available from source in perfectly matched condition from source,
i.e. Γ𝐼𝑁 = Γ𝑆∗ .
∗
𝑃𝑎𝑣𝑛 = Power from transistor at perfectly matched condition i.e. Γ𝐿 = Γ𝑜𝑢𝑡 .
𝑃𝑙 = Power at load.
From these different power levels, we can define power gain of the amplifier.
𝐺𝑇 is transducer gain given by;
𝑃𝑙
𝐺𝑇 = (2.11)
𝑃𝑎𝑣𝑠
1−|Γ𝑆 |2 1−|Γ𝐿 |2
𝐺𝑇 = |1−Γ 2
|𝑆21 |2 (2.12)
𝑖𝑛 Γ𝑆 | |1−S22 Γ𝐿 |2

The above equation is deduced as below;

𝐺𝑇 = 𝐺𝑆 𝐺𝑂 𝐺𝐿 (2.13)
1−|Γ𝑆 |2
𝐺𝑆 = |1−Γ 2
(2.14)
𝑖𝑛 Γ𝑆 |

𝐺𝑂 = |𝑆21 |2 (2.15)

8
 1−|Γ𝐿 |2
𝐺𝐿 = |1−S 2
(2.16)
22 Γ𝐿 |

Γ𝑖𝑛 and Γ𝑜𝑢𝑡 are reflection co-efficient of input and output network can be given as shown
below.

𝑆12 𝑆21 Γ𝐿
Γ𝑖𝑛 = 𝑆11+ (2.17)
1−𝑆22 Γ𝐿

𝑆12 𝑆21 Γ𝑆
Γ𝑜𝑢𝑡 = 𝑆22 + (2.18)
1−𝑆22 Γ𝑆

Thus, transducer gain is in dB, given in equation (2.19).

𝐺𝑇 (dB)= 𝐺𝑆 (𝑑𝐵) + 𝐺𝑂 (𝑑𝐵) + 𝐺𝐿 (𝑑𝐵) (2.19)

Now if 𝑆12of the transistor is equal to zero, thus this condition is the necessary condition
for a transistor to be unilateral.
So, equation (2.13) becomes as shown below;

𝐺𝑇𝑈,𝑚𝑎𝑥 = 𝐺𝑆,𝑚𝑎𝑥 𝐺𝑂 𝐺𝐿,𝑚𝑎𝑥 (2.20)

Where,
∗
Γ𝑆 = 𝑆11 (2.21)
∗
Γ𝐿 = 𝑆22 (2.22)
&
1
𝐺𝑆,𝑚𝑎𝑥 = 1−|𝑆 2
11 |

1
𝐺𝐿,𝑚𝑎𝑥 = 1−|𝑆 2
22 |

𝐺𝑂 = |𝑆21 2 |
Thus 𝐺𝑇𝑈,𝑚𝑎𝑥 obtained is the maximum gain obtained for the transistor, this is only
possible when it’s conjugately matched condition. i.e. only when equation (2.21) and (2.22)
is possible.

9
However, the case of unilateral is not possible for every transistor, so under that situation
error analysis is introduced which is calculated by “Unilateral Merit”, given by;

1 𝐺𝑇 1
<𝐺 < (1−𝑈)2 (2.23)
(1+𝑈)2 𝑇𝑈,𝑚𝑎𝑥

Where U a Merit for Unilateral is given by,

|𝑆 ||𝑆21 ||𝑆11 ||𝑆22 |

12
U = (1−|𝑆 2 2 (2.24)
11 | ) (1−|𝑆22 | )

The value of U varies with frequency as it depends on S-parameters. By solving equation

(2.24) value of U can be calculated and error range is checked at a desired frequency. The
error range can be checked from the table below;

𝐺𝑇
Table 2: U vs 𝐺 [1].
𝑇𝑈,𝑚𝑎𝑥

If the error is within the range then the transistor is acceptable as unilateral but if it does
not fall within the range the transistor is not unilateral.

10
2.4 Design of PA circuit
After understanding the key constrains of amplifier, now we will discuss about two distinct
circuits the entire amplifier design is divided.
1. Biasing Circuit i.e. DC circuit
2. RF/Microwave Circuit
These two circuits are combined and integrated to become one single unit.

2.4.1 DC Biasing Circuitry

For linear operation of amplifier, it is necessary to operate it in class A mode. In such
condition Q-point must be in the middle of 𝐼𝐷 -𝑉𝐷𝑆 curve or better known as characteristics
graph. As we shift up the Q-point the transistor gets into saturation mode even if it is in
active region i.e. the peak input current is chopped of by saturation current. If we shift Q-
point more down it approaches towards cut off region i.e. the peak gets chopped to zero.

Figure 2.4 Characteristics graph of Transistor (𝐼𝐷 -𝑉𝐷𝑆 ) [1]

As RF signals tend to take the lowest impedance path, therefore it becomes necessary to
block RF going to DC source. This can be achieved by several schemes, which is described
below; Reference [1]

11
 ➢ The use of RF choke between the two circuitries so there is a high impedance path
to RF signals. Thus, preventing signals to leak in DC circuit.
➢ Another method to provide high impedance path is to connect a quarter-wave
transformer between two circuits such that its characteristics impedance remains
high i.e.(𝑍𝑂 >>1).
➢ In addition to the quarter-wave transformer a large capacitor is attached to it as a
load. The use of high value capacitor would appear as a open circuit for RF signals
at input where it is connected to RF circuit and block it entering the D.C side.
Combining of all schemes in an amplifier design provide high amount of isolation between
the two circuities. The above discussed scheme is necessary for the proper functioning of
amplifier as shown in figure 2.5, after that we will discuss about RF/Microwave circuits.

Figure 2.5 Amplifier with DC biasing Circuit [1]

2.4.2 RF/MW Circuit for Designing PAs

Steps for designing RF/Microwave circuit are as follows:

➢ Selection of proper device i.e. transistors as per specification of amplifier’s design.
e.g. check for gain such that |𝑆21 ⁄𝑆12| > G at the desire frequency at which
amplifier must be designed similarly noise figure i.e. 𝐹𝑂 should be greater than 𝐹𝑚𝑖𝑛
(𝐹𝑂 > 𝐹𝑚𝑖𝑛 ).
➢ Biasing of transistor should be in mid-range of 𝐼𝐷 vs 𝑉𝐷𝑆 curves.
➢ S-parameters are measured at selected Quiescent point for desired frequency.

12
 ➢ Stability of the device is checked i.e. K-∆ test/ 𝜇- test at the desired frequency. If
stability is not achieved then there is need for finding stable regions by drawing
input and output stability circles.
➢ Now check for 𝑆21 of the transistor if it is zero then its unilateral case, but if 𝑆21 ≠
0 then check for Unilateral figure of merit, if it’s within the error limit then it is a
Unilateral case otherwise check of bilateral design.
The last step is to design matching networks using Smith charts. Fig 2.6 explains about the
generalized process of designing a amplifier.

Figure 2.6 Generalized Steps for designing RF/MW circuits

13
2.5 Design of Matching Networks with Smith Charts
Design of matching network is necessary for maximum power. In this design, distributed
elements are used for impedance matching, because at higher frequencies the use of lumped
elements becomes more comparable with wavelength which leads to mismatch in the
circuit and eventually affects the power transfer. Four possible schematics for distributed
matching networks are shown in Figure 2.7.

Figure 2.7 Matching Networks Using Series and Shunt Stubs [2]

14
Figure 2.7 shows matching network using microstrip-line technology, which is single
stubbing where there are two variables “d” distance from load to stub and “l” length of stub
providing a value of its susceptance or reactance.

➢ In a shunt stub selection of “d” in shunt is done in such a way that 𝑌𝑃 = 𝑌𝑂 + jB

where – jB is stub susceptance giving matching condition.
➢ For series stubs “d” is selected in a way that 𝑍𝑠 = 𝑍𝑂 + jX where -jX is a stub
reactance giving matching condition.
➢ Open or Short stub depends upon type of transmission line depend on application
level. There is difference of 𝜆⁄4 between two which provides needed value of
reactance or susceptance for design. In the design quarter-wave is used for matching
network.

Figure 2.8 Stubbing using Smith Charts [2]

15
Design steps of matching network in case of unilateral transistor is discussed below:

∗ ∗
➢ For matching network design, we use Y-Z smith charts to plot 𝑆11 & 𝑆22 .
➢ Draw VSWR circles, each passing through two points which were plotted in first
step.
➢ Starting from the center of the Smith chart moving on a constant unity
(conductance) circle we intersect at two points giving two possible solutions i.e.
1+j𝑏1 & 1- j𝑏1 .
∗ ∗
➢ From points 𝐴1 and 𝐴2 moving towards generator side to arrive at points 𝑆11 & 𝑆22
as shown in fig 2.8.
➢ Length “l” is measured from point Z=0 for short stub, the value of -j𝑏1 or +j𝑏1 ,
distance “d” is measured on radial scale.

The same procedure can be followed moving on a constant reactance circle to get a series
stub. Open and short stub can be determined from starting from Z = ∞ for open stubs and
Z = 0 for short stubs such that we need to travel towards the generator to arrive at -j𝑏1 or
+j𝑏1 stubs. Lengths are read off from the radial scale discussed above, shown in Figure 2.9
(a & b).

Figure 2.9 (a & b) Open and Short Stub lengths [2]

16
2.6 Power Coupler
There are various couplers available such as Quadrature Couplers, Wilkinson, branch line
couplers. As per design requirements Lange couplers are used as they gives 3dB coupling
ratios. In a Lange coupler several coupled lines are used between edge coupled line such
that fringing at the end is contributed to coupling. Figure 2.10 is generalized N-way power
divider/coupler. Couplers are explained in detail in further chapters.

Figure 2.10 General Block Diagram of N-way Amplifier

17
CHAPTER 3 DESIGN CONSIDERATION AND SIMULATION USING NI-AWR

In this chapter we will discuss in detail about a design of an amplifier. How the single stage
is done and then second stage is achieved using Lange Coupler. The whole process is
described below:
➢ Device (i.e. transistor) selection was made at the desired frequency (14.5 GHz) in
our project simultaneously its s-parameters at that frequency was noted.
➢ Stability of device is checked by K-∆ Test, and further it is doubled checked by 𝜇-
test.
➢ Unilateral figure of merit and its error is checked and noted whether it in
permissible limit or not, depending on the results it can be known whether it is
unilateral or bilateral design.
➢ Matching networks for a source and load are designed with help of Smith charts.
➢ DC Biasing for a single stage is done for the device and proper isolation scheme is
implemented between RF and DC circuits as discussed before.
➢ Single stage PAs design is done and verified using NI-AWR simulation software.
➢ Coupler is designed and its results were verified using the same software.
➢ Final stage is to combine the two-single stage with use of coupler, i.e. lange coupler
and final results were obtained.
➢ At last all the results of simulations, mathematical calculations and MATLAB
results are compared and checked.

3.1 Selection of Transistor for Amplifier

TOSHIBA’s TIM1314-8UL High power FET is used in this project. This is high power
GaAs device mainly use for Ku band amplifiers due to its high-power output. Details
regarding transistor is stated in Appendix C.

18
Device s-parameters and its electrical characteristics are in Table 3 & 4:

Table 3: S-parameters of Transistor [Appendix C]

Table 4: Electrical Characteristics of Transistor [Appendix C]

For verification purpose we simulated the transistor’s s-parameters in NI-AWR software

and results were obtained and compared to datasheet of the device given in Appendix C.

19
 Figure 3.1 Sub-circuit Model of Transistor TIM1314-8UL in AWR

Simulation of S-parameters of transistor and its results are in Figure 3.2 at the desired
frequency 14.5GHz i.e. operating frequency. The purpose is just to compare and check the
results are right.

Figure 3.2 S-parameters of Transistor & Simulation in AWR

3.2 Stability of TIM1314-8UL

Stability of the device is checked with the help of K-Δ test and it is doubled checked by µ-
test at 14.5 GHz.

20
S-parameters of TIM1314-8UL at 14.5 GHz are as below in table 5.

S-parameters Magnitude Angle

𝑆11 0.145 157.3
𝑆12 0.104 34.5
𝑆21 2.340 47.3
𝑆22 0.202 -23.5
Table 5: S-parameters for TIM1314-8UL
|∆|= 𝑆11 𝑆22 - 𝑆21 𝑆12
|∆|= (0.145∠ 157.3 * 0.202 ∠ -23.5) – (2.340 ∠ 47.3 * 0.104 ∠34.5)
|∆| = 0.323 123.66

K = 2.03

So, from the results K >1 and |∆| <1 this states that the device is stable unconditionally.
These results are also verified using MATLAB and E- Book Microwave software as
mentioned in Appendix.

1−|𝑆11 |2
µ= |𝑆 ∗
22 −𝑆11 ∆|+|𝑆12 𝑆21 |

µ = 2.39

Since µ > 1 thus this verifies the device is unconditionally stable at 14.5GHz.

3.3 Unilateral Figure of Merit

From table 5, 𝑆12 ≠ 0 this indicates that transistor is not unilateral and needs to check the
error range. U is given by;

|𝑆 ||𝑆21 ||𝑆11 ||𝑆22 |

12
U = (1−|𝑆 2 2
11 | ) (1−|𝑆22 | )

U = 0.00759 (ratio)

21
Error range is found out and given by equation;

1 𝐺𝑇 1
<𝐺 < (1−𝑈)2
(1+𝑈)2 𝑇𝑈,𝑚𝑎𝑥

1
= 0.992 dB
(1+𝑈)2

1
= 1.0076 dB
(1−𝑈)2

Error range is acceptable limits; the design is unilateral.

3.4 Power Saturation in Large Signal Amplifier

To operate an amplifier in class A under large signals all the S-parameters remains same
except 𝑆21 . The 𝑆21 parameter gets reduce as power level goes up with increase in
frequency; reduction of forward gain results in power saturation and gain role-off as
frequency is high. This takes design to a unique level in which small signal parameters
remain same except 𝑆21 .[1]

Figure 3.3 Pout vs Pin of TIM1314-8UL

22
The fig 3.3 is the graph of Pout vs Pin provided in datasheet of manufacturer, 𝐺1𝑑𝐵 is
obtained by graph. Gain equation used are same as in chapter 2.

At 14.5 GHz,

From 𝑃𝑜𝑢𝑡 − 𝑃𝑖𝑛

𝐺1𝑑𝐵 = 6.85 dB

𝐺1𝑑𝐵 = 10 log|𝑆21 |2 (3.1)

6.85 = 10 log|𝑆21 |2

𝑆21 =2.200

The parameters for 𝑆21 we found by the graph plot and in datasheet by manufacturer are
approximate same.

𝐺1𝑑𝐵 = 𝐺0 − 1dB (3.2)

∴ 𝐺0 = 7.85 dB

For maximum gain calculations;

𝐺𝑇𝑈,𝑚𝑎𝑥 = 𝐺𝑆,𝑚𝑎𝑥 𝐺𝑂 𝐺𝐿,𝑚𝑎𝑥

1
𝐺𝑆,𝑚𝑎𝑥 = 1−|𝑆 2
= 0.0922 dB
11 |

1
𝐺𝐿,𝑚𝑎𝑥 = 1−|𝑆 2
= 0.1809 dB
22 |

𝐺𝑂 = |𝑆21 2 | = 7.659 dB
𝐺𝑇𝑈,𝑚𝑎𝑥 = 7.93 dB

23
3.5 Biasing of Transistor
There are 5 different types of biasing techniques as discussed in Dr. Matthew Radmanesh’s
book. The figure 3.4 below shows all the five approaches.

Figure 3.4 Different Techniques of Biasing [3]

24
Discussing all the approaches is beyond this reports scope. Technique “d” is selected for
Biasing purpose as it is used for delivering high power and gain addition to that gain can
be changed by varying the value of 𝑅5 , Fig 3.5 shows the approach in detail.

Figure 3.5 Biasing of TIM1314-8UL [3]

Analytical Approach for DC Biasing

𝐼𝐷𝑆𝑆 = 4 A, 𝐼𝐷𝑆 = 2 A; [Appendix C]

𝑉𝐷𝑆𝑆 = 3 V, 𝑉𝑡 = -0.7 V, 𝑉𝐷 = 0, 𝑉𝐷𝑆 = 10;

By using the equation;

𝐼𝐷𝑆 = K [𝑉𝐺𝑆 − 𝑉𝑡 ]2 (3.3)

𝐼
K = 𝑉𝑡𝐷2

4
K = (0.7)2 = 8.163

Putting the value of K in (3.3);

2 = 8.163 [𝑉𝐺𝑆 − (− 0.7)]2

25
So, we get two values of 𝑉𝐺𝑆 ;

1. 𝑉𝐺𝑆 = − 0.205 V > 𝑉𝑡 (acceptable)

2. 𝑉𝐺𝑆 = − 1.195 which less than threshold voltage 𝑉𝑡 = -0.7 V;(cut off mode)

Now, 𝑉𝐷𝑆 = 𝑉𝐷 − 𝑉𝑆 ; But 𝑉𝐷 = 0;

𝑉𝐷𝑆 = − 𝑉𝑆 ;

∴ 𝑉𝑆 = − 10V;

𝑉𝐺𝑆 = 𝑉𝐺 − 𝑉𝑆 ;

𝑉𝐺 = -10.205 V ~ -10 V;

Taking, 𝑉𝐺𝐺 = − 15 V.

Assuming values of 𝑅1 and 𝑅2 in formula, such that;

𝑅2 1
=3
𝑅1 + 𝑅2

𝑅1 = 10 MΩ

𝑅2 = 20 MΩ

Calculating value of 𝑅𝑠 ;

𝑅𝑠 = 𝑉⁄𝐼 =10.1025 Ω

For RF isolation necessary conditions should be met 𝑋𝐿 ≫ 50 Ω and 𝑋𝐶 ≪ 50 Ω. Assuming

𝑋𝐿 = 1000 Ω and 𝑋𝐶 = 5 Ω from this calculate the value of inductance and conductance.

j𝑋𝐿 = j𝜔L

1000
L = 2𝜋 ×14.5 ×109 = 11nH

𝑗
j𝑋𝐶 = 𝑋
𝐶

26
 1
C = 2𝜋 ×14.5 × 109×5 = 2.2 pF

Biasing circuit is shown in Figure (3.6)

Figure 3.6 Biasing circuit in AWR

27
3.6 Design and Source Matching Networks for Load and Source
We already discussed regarding designing of matching networks in chapter 2.

3.6.1 Matching Network for Source

∗
We are designing on 50 Ω system. Plot 𝑆11 on Smith Chart for Input matching network.

∗
𝑆11 = 0.145 ∠ −157.3

Start from the center of the Smith Chart and move on the real axis i.e. (horizontal axis)
∗
towards conductance circle passing through 𝑆11 giving us a quarter wave of length 𝜆⁄4 and
its characteristics impedance is calculated to be 43.3Ω, a shorted shunt stub of length
0.274𝜆 of 50 Ω, where 𝜆= 9.62 mm.

Figure 3.7 Matching Network Source Side

28
SMITH CHART:

Figure 3.8 Smith Chart of Matching Network(Source)

29
3.6.2 Matching Network for Load
∗
We used the same quarter-wave transformer for matching purpose, plot 𝑆22 on the graph.
Start from the center similar as we designed matching network for source move on a real
∗
axis at point A where it intersects conductance circle passing through 𝑆22 gives us a shunt
stub.

∗
𝑆22 = 0.202 ∠ 23.5

Figure 3.9 Matching Network Load Side

Value of a quarter wave of length 𝜆⁄4 is 59.6 Ω and we get a 50Ω shunt stub of length of
0.22𝜆, where 𝜆 = 9.62mm.

30
SMITH CHART

Figure 3.10 Smith Chart of (Load) Matching Network

31
 Figure 3.11 Single Stage Amplifier Using Distributed Networks
Figure 3.11 shows schematics of a single stage PA (power amplifier) in AWR. This
simulation software helps to understand matching networks for input and output sides. The
power gain obtained by this circuit is shown in Figure 3.12 i.e. 7.584 dB. The design is
properly matched as the reflection at port 1 is -25.8dB this indicates that the power loss is
very low.

Figure 3.12 S-parameters of 1st Stage Amplifier

32
 Figure 3.13 𝑃𝑖𝑛 vs 𝑃𝑜𝑢𝑡 of Amplifier at 14.5GHz
Figure 3.13 shows 𝑃𝑖𝑛 vs 𝑃𝑜𝑢𝑡 at 14.5 GHz is 7.3dBm at 0 dBm.

33
 CHAPTER 4 MICROSTRIP LINE DESIGN AND CALCULATIONS

4.1 General Theory of Transmission Lines

Transmission lines are the medium in which information or electrical signals are
transferred from one point to another point. In RF and Microwaves, the frequency range is
very high so it requires a path or say a medium in which it can propagate. They are
considered as a bridge between lumped and distributed elements which helps to carry
information from one part of circuit to the another.

There are six major transmission lines as listed below;

i. Conventional 2 wire transmission line

ii. Waveguides

iii. Microstrip lines (used in the project)

iv. Strip lines

v. Parallel-Plate Waveguides

vi. Coaxial Cables

Implementation of microstrip-lines is done in this project discussing all of them is beyond

scope.

4.2 Basics of Microstrip Lines

Use of microstrip-lines is most popular in transmission line technologies as it is very easily
integrated and miniaturized with active and passive microwave devices. It can be stated
that it’s a conductor having thickness ‘t’, width ‘d’ separated by a dielectric medium at a
distance ‘h’. In microstrip-lines the mode of propagation is Quasi-TEM not TEM mode as
it is open conduit, this fact shows that not all electric and magnetic fields are confined
inside the structure. Cross-segment and its field is shown in Figure 4.1.

34
 Figure 4.1 Field Configuration and Geometrics of Microstrip Lines [2]
4.3 Calculation for Designing Microstrip Lines
Design procedure starts by Graphical method as shown in Figure 4.2, it’s a graph of 𝑍0 vs
𝑤 ⁄ℎ. In our project we use R30006 substrate by Rogers Co-corporation with 𝜖𝑟 = 6.5 and
𝑍0 = 50Ω.

Figure 4.2 Graph of 𝑍0 vs 𝑤 ⁄ℎ at 𝜖𝑟 = 6.5[2]

35
Figure 4.2 shows approximate value of 𝑤 ⁄ℎ = 1.5. Analytical method is used to get more
specific results. Design specifications are as given below;

𝑍0 = 50Ω

f = 50GHz

𝜖𝑟 = 6.5 (R3006 substrate)

h = 0.13mm

c = 3× 108 𝑚⁄𝑠

By using the value of graphical method, various values can be calculated by using formulas
as discussed. Reference [2] for equation of microstrip-lines.

For 𝑤 ⁄ℎ ≤ 2:

𝑤 8eA
= e2A −2
ℎ

𝑤 ⁄ℎ ≥ 2:

𝑤 2 𝜀 −1 0.61
= 𝜋 [B−1 ln(2B-1) + 2𝜀𝑟 +1 {ln(B−1) + 0.39 − }]
ℎ 𝑟 𝜀𝑟

But 𝑤 ⁄ℎ is less than 2, so use the first equation.

𝑍 𝜀𝑟 +1 𝜀 −1 0.11
A = 600 √ + 𝜀𝑟 +1(0.23 + )
2 𝑟 𝜀𝑟

By solving the equation, we get A= 1.8. Inserting the value in an equation of 𝑤 ⁄ℎ.
𝑤
∴ = 1.4
ℎ

Now as per the datasheet of R3006 substrate h = 0.13mm, the value of ‘w’ is calculated.

w = 0.182mm

36
Now calculate the value of effective dielectric constant 𝜖𝑓𝑓 , we use equations given below;

For 𝑤 ⁄ℎ ≤ 1;

𝜀𝑟 +1 𝜀𝑟 −1 ℎ 𝑤
𝜀𝑓𝑓 = + [(1+12 𝑤)−1⁄2 + 0.04(1+12 ℎ )2]
2 2

𝑤 ⁄ℎ ≥ 1;

𝜀𝑟 +1 𝜀𝑟 −1 ℎ
𝜀𝑓𝑓 = + [(1+12 𝑤)−1⁄2]
2 2

As we know that 𝑤 ⁄ℎ ≥ 1, we use the first equation and calculate the value of 𝜖𝑓𝑓 .

𝜖𝑓𝑓 = 4.64

Calculation of 𝑍0 characteristics impedance of microstrip-line.

For 𝑤 ⁄ℎ ≤ 1;

60 8ℎ 𝑤
𝑍0 = ln ( 𝑤 + 4ℎ)
√𝜀𝑓𝑓

120𝜋
𝑍0 =
√𝜀𝑓𝑓 [𝑤⁄ℎ +1.393+0.667 ln(𝑤⁄ℎ +1.444)]

𝑍0 = 50.16Ω

Phase velocity is given by;

𝑐
𝑉𝑝 =
√𝜖𝑓𝑓

𝑉𝑝 = 1.39 × 1010 𝑐𝑚⁄𝑠

𝜆 𝑇𝐸𝑀 is a wavelength of TEM wave propagating in dielectric medium which is given by;

𝜆0
𝜆 𝑇𝐸𝑀 = ; where 𝜆0 is wavelength in free space.
√ 𝜖𝑟

𝑐
𝜆0 = 𝑓 = 2.1cm; 𝜆 𝑇𝐸𝑀 = 8.23mm.

37
So, formula of wavelength in microstrip line is given by;

𝑤 ⁄ℎ ≤ 0.6;

𝜆0 𝜀𝑟
𝜆= [ ]1⁄2
√𝜀𝑟 1+0.63 ( 𝜀𝑟 −1)(𝑤⁄ℎ)0.027

𝑤 ⁄ℎ ≥ 0.6;

𝜆0 𝜀𝑟
𝜆= [ ]1⁄2
√𝜀𝑟 1+0.63 ( 𝜀𝑟 −1)(𝑤⁄ℎ)0.1255

By calculating the value of 𝜆,we get 𝜆 = 9.62mm.

Cut-off frequency be calculated as

𝑍0
𝑓0 (GHz) = 0.3 √ℎ
√𝜀𝑟 −1

𝑓0 =12.15 GHz

As cut-off frequency is lower than our center frequency thus waves propagating inside
microstrip-line will be non-dispersive.

4.4 Gap Capacitance Calculations

Figure 4.2 Geometry for Microstrip Gap Capacitance [3]

Consider 𝑤1= 𝑤2 = 0.182mm

𝜖𝑟 = 6.5

h = 0.13mm

38
Formula for Gap Capacitance is given by;

𝑠 ℎ 𝑤2
𝐶𝑠 (pF) = 500 h exp (−1.86 ℎ) 𝑄1 (1+ 4.19(1− exp (− 0.785√𝑤 )))
1 𝑤1

Where 𝑄1,
𝑤
𝑄1= 0.04598 (0.03 + ( ℎ1 )𝑄5 )(0.272 + 0.07𝜀𝑟 )

1.23
𝑄5 = 1+0.12 (𝑤 0.9
2 ⁄𝑤1 −1)

So, 𝑄5 = 1.23

By solving equation for 𝑄1 we get;

𝑄1= 0.05183

Ideally, value of spacing ‘S’ is equal to one third of the width.

1
S = 3 × w; where w = 0.182mm

∴ S = 0.0606 mm.

Substituting the value of ‘S’ in the equation of gap capacitance.

𝐶𝑠 = 4.29 pF.

So, Impedance of Gap capacitance ,

1
𝑍𝑐 = 𝜔𝐶 = 2.55 Ω.
𝑠

39
All mathematical calculations for 𝑍0 =50Ω, 59.6Ω and 43.3Ω are verified by MATLAB
simulations and its comparison table is below.

For 50Ω,

Parameters MATLAB calculations Hand calculations

𝑤 ⁄ℎ 1.406 1.4
w 0.1820mm 0.182mm
𝑍0 50.16Ω 50.16Ω
𝜖𝑓𝑓 4.6389 4.64
Table 6 Microstrip Lines Comparison

For 59.6Ω,
Parameters MATLAB Calculations Hand Calculations
𝑤 ⁄ℎ 1.175 1.17

w 0.1528 0.152
l 0.2486 0.2480
Table 7 Parameters for 59.6Ω
For 43.3Ω,

Parameters MATLAB Calculations Hand Calculations

𝑤 ⁄ℎ 1.955 1.95
w 0.2345 0.2445
l 0.2378 0.2403
Table 8 Parameters for 43.3Ω

40
 CHAPTER 5 POWER COMBINERS

5.1 Theory of Power Combiners

Power couplers are used to couple two or more stages of amplifiers. The passive
components used for dividing or combining power are said to be power couplers or
dividers. This can be further stated that in dividers, input signal is divided into two or more
output signals while in couplers its reverse action takes place.

Figure 5.1 Power Couplers and Dividers [4]

Classification of power dividers and combiners as shown in figure 5.2.

Figure 5.2 Various types of power couplers/dividers [1]

41
Lange coupler is used in our project to couple two single stage amplifiers, so let’s know
about it in detail.

5.2 3-dB Power Coupler (Lange Coupler)

Lange coupler consist of four microstrip-lines integrated to each other, such that it’s
alternate are connected to each other making it as a four-port device. Figure 5.3 shows
microstrip Lange Coupler.

Figure 5.3 Layout of Lange Coupler [1], [4]

As you can see from the Figure 5.3 it consists of four different port are as given below;

➢ Input port

➢ Through port

➢ Isolated port

➢ Coupled port

Coupler is designed in a way that the input signal enters from port 1, equal power signal is
sent to port 2 and 3, but they are 90° out of phase with each other. Coupling efficiency of
Lange coupler is 3dB. Further we will discuss about odd mode and even mode impedance
used for designing coupler.

42
Even mode impedance is given by;

Where C is the coupling efficiency of the coupler i.e. 3dB

C = 0.707 (ratio)

By solving the equation

𝑍0𝑒 = 176.4Ω

Odd mode impedance is given by;

We get, 𝑍𝑂0 = 52.54Ω

By using the ratio graph in fig.5.4 value of ‘w’ width and ‘s’ spacing can be calculated.

Figure 5.4 Ratio Graph for Lange Coupler

43
5.3 Designing of Lange Coupler
Important parameters to design Lange couplers are its spacings, ‘s’ and width ‘w’ which is
found out using Figure 5.4.

𝑠⁄ℎ = 0.085 and 𝑤 ⁄ℎ= 0.15

h = 0.13mm as per the datasheet provided by Roger’s corporation.

s = 0.011mm, w = 0.0195mm.

Implementing the values in AWR and simulating it, the results were obtained shown in
Figure 5.5.

Figure 5.5 Lange Coupler in AWR

Figure 5.6 shows the results obtained by simulation of s-parameters of Lange coupler,
which shows coupling efficiency of -3.1 dB at center frequency of 14.5GHz.

44
 Figure 5.6 3-dB Coupling Efficiency
There is 90° phase shift between ‘through port’ and ‘coupled port’ results are shown in
Figure 5.7

Figure 5.7 Phase Difference in Lange Coupler

45
 CHAPTER 6 FINAL CIRCUIT FOR A TWO-STAGE POWER AMPLIFIER

Final chapter discuss about the two-stage design formed by coupling of two individual
single stages with help of lange coupler. Figure 6.1 shows block diagram for two-stage
design where at in 1st block lange coupler is used as power divider followed by two
amplifiers, outputs of amplifiers are combined to lange coupler at the end.

Figure 6.1 Block Diagram for Two Stage Design

Linear Gain 𝐺0 = 7.85 dB

𝐺1𝑑𝐵 = 6.85 dB

Considering, Insertion Loss = 0.25 dB

At point A:

𝑃𝐴 = Output Power + Insertion Loss ₋ 3dB

= 7.3 + 0.25 −3

= 4.55 dBm

46
At point B:

𝑃𝐵 = Output Power− 𝐺1𝑑𝐵

= 4.55 − 6.85

= − 2.3 dBm

At point C:

𝑃𝐶 = Output Power + Insertion Loss + 3dB

= − 2.3 + 0.25 +3

= 0.95 dBm

Figure 6.2 shows simulation of s-parameters for two-stage design, the total power gain
obtained is of 7.539dB. 𝑆11 at 14.5 GHz is -29.64dB which shows that the design is
matched.

Figure 6.2 S-parameter Simulation for Two-Stage Design

47
A comparison of VSWR for single stage and two-stage design in Figure 6.3 . In the two-
stage design the reflections from both the single stages are 90˚ phase with each other, this
causes significant improvement of VSWR is noted over a range of frequency.

Figure 6.3 Comparison of VSWR

One tone test is carried out of final circuit from which plot for 𝑃𝑜𝑢𝑡 vs 𝑃𝑖𝑛 and output power
is 7.32dBm at 14.5GHz.

Figure 6.4 One Tone Test Bench for the Final Circuit

48
 Figure 6.5 𝑃𝑜𝑢𝑡 vs 𝑃𝑖𝑛 for Final Schematics
Finally, there is comparison of 𝑆11for single stage, two-stage design and using lumped
elements in Figure 6.6. There significant improvement in reflection co-efficient for two-
stage design as compared to others.

Figure 6.6 Evaluation of the Reflection Co-efficient

49
Schematics and layout of Final Circuits are as shown in Figures 6.7 and 6.8.

Figure 6.7 Two Stage Implementation in AWR

50
Figure 6.8 Layout of the Final Circuit

51
Effect of Parasitic

Elements used in the project are without considering parasitic, and the simulation results
are obtained as per the ideal case. At higher frequencies use of lumped elements such as
capacitors, inductors and resistors becomes limited as the size of the lumped elements in
the circuits whose size becomes comparable with signal wavelength. If length of the
component is less than 0.1𝜆 then (MIC) technology has pushed the use of lumped elements
at microwave frequency up to 60 GHz.[1] Spiral inductors with low parasitic and high Q
at higher frequencies can be designed. Reference [7] gives a detail about modeling of each
lumped elements at higher frequencies. Ferrite beads can be used in case of RF chokes in
biasing circuit to block RF signals entering DC circuit.

52
 CHAPTER 7 CONCLUSIONS

A balanced configuration for two-stage power amplifier is designed at an operating

frequency of 14. 5 GHz. The design is simulated and its results are verified by using NI-
AWR software, MATLAB and RF/Microwave Ebook.

TIM1314-8UL GaAs transistor from TOSHIBA Electric is used in this project. Device’s
stability is checked by using K-Δ and µ test at center frequency. Microstrip line technology
is used to design the distributed elements for matching networks. RO3006C dielectric
substrate is used for matching networks from Roger’s Corporation having a dielectric
constant of 6.5 and the height of 0.130mm. Biasing of a device is done using the electrical
characteristics provided by manufacturer’s datasheet.

3-dB coupling efficiency is achieved by the use of Lange coupler which is designed by
using microstrip-line technology. The final circuit is simulated in AWR Microwave Office
giving linear gain of 7.53dB and output power of 7.32dBm at 0dBm of input power. In
addition to that reflection coefficients of the final design are improved as compared to that
of single stage, i.e. -25.82dB for a single stage and -29.63 for two-stage design.
Comparison between AWR results and hand calculations are shown in a Table 9 below:

Parameters Hand Calculations AWR Results RF/MW Book

Power Gain of Two 7.85dB 7.53dB 7.5dB
Stage
Output Power 7.5dBm 7.32dBm 7.33dBm
Table 9 Results

53
REFRENCES

54
APPENDIX A

55
For 50 Ω,

For 59.6 Ω,

56
For 43.3Ω,

57
 APPENDIX B

RF/Microwave Software Calculations:

For 50Ω line,

58
For 59.6Ω line,

59
For 43.3Ω line,

60
K-Δ test,

Gain Calculations,

61
APPENDIX C

62
63
64
65
66
67
APPENDIX D

68
69
70
71
72