Microwave and RF

Engineering
Volume 1
An Electronic Design Automation Approach

Ali A. Behagi
and
Stephen D. Turner

BT Microwave LLC
State College, PA 16803

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

Microwave and RF Engineering

ISBN 13: 978-0-9835460-1-6
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

Published in USA
BT Microwave LLC
State College, PA 16803

All rights reserved. Printed and bound in the United States of America. No
part of this book may be reproduced or transmitted in any form or by any
means without permission in writing from the authors.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

Table of
Contents
Foreword

xv

Preface

xvii

Chapter 1 RF and Microwave Concepts and Components

1.1

Introduction

1.2

Straight Wire, Flat Ribbon, and Skin Effects

1.2.1 Straight Wire Inductance
1.2.2 Simulating the Straight Wire Inductor in Genesys
1.2.3 Skin Effect in Conductors
1.2.4 Analytical Calculation of Flat Ribbon Inductance

3
3
5
8
11

1.3

Physical Resistors
1.3.1 Chip Resistors

12
14

1.4

Physical Inductors
1.4.1 Air Core inductors
1.4.2 Modeling the Air Core Inductor in Genesys
1.4.3 Inductor Q Factor
1.4.4 Chip Inductors
1.4.5 Chip Inductor Simulation in Genesys
1.4.6 Magnetic Core Inductors

16
18
22
27
28
29
32

1.5

Physical Capacitors
1.5.1 Single Layer Capacitor
1.5.2 Multilayer Capacitors
1.5.3 Capacitor Q Factor

40
41
43
44

References and Further Reading

Problems

49
49

Chapter 2 Transmission Lines

53

2.1

53

Introduction
v
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

2.2

Plane Waves
2.2.1 Plane Waves in a Lossless Medium
2.2.2 Plane Waves in a Good Conductor

53
53
55

2.3

Lumped Element Representation of Transmission Lines

56

2.4

Transmission Line Equations and Parameters

2.4.1 Definition of Attenuation and Phase Constant
2.4.2 Definition of Transmission Line Characteristic Impedance
2.4.3 Definition of Transmission Line Reflection Coefficient
2.4.4 Definition of Voltage Standing Wave Ratio, VSWR
2.4.5 Definition of Return Loss
2.4.6 Lossless Transmission Line Parameters
2.4.7 Lossless Transmission Line Terminations
2.4.8 Simulating Reflection Coefficient and VSWR in Genesys
2.4.9 Return Loss, VSWR, and Reflection Coefficient
Conversion

57
59
59
59
60
61
61
62
64

RF and Microwave Transmission Media

2.5.1 Free Space Characteristic Impedance and Velocity of
Propagation
2.5.2 Physical Transmission Lines

67
67
68

2.6

Coaxial Transmission Line

2.6.1 Coaxial Transmission Lines in Genesys
2.6.2 Using the RG8 Coaxial Cable Model in Genesys

70
73
74

2.7

Microstrip Transmission Lines

2.7.1 Microstrip Transmission Lines in Genesys

76
78

2.8

Stripline Transmission Lines

80

2.9

Waveguide Transmission Lines

2.9.1 Waveguide Transmission Lines in Genesys

82
86

2.10

Group Delay in Transmission Lines

2.10.1 Comparing Group Delay of Various Transmission lines

89
89

2.11

Transmission Line Components

2.11.1 Short-Circuited Transmission Line
2.11.2 Modeling Short-Circuited Microstrip Lines
2.11.3 Open-Circuited Transmission Line

91
91
93
94

2.5

vi
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

64

2.11.4
2.11.5
2.11.6
2.11.7
2.11.8
2.11.9
2.12

Modeling Open-Circuited Microstrip Lines

Distributed Inductive and Capacitive Elements
Distributed Microstrip Inductance and Capacitance
Step Discontinuities
Microstrip Bias Feed Networks
Distributed Bias Feed

95
96
97
98
99
100

Coupled Transmission Lines

2.12.1 Directional Coupler
2.12.2 Microstrip Directional Coupler Design

102
105
107

References and Further Reading

Problems

110
110

Chapter 3 Network Parameters and the Smith Chart

113

3.1

Introduction
3.1.1 Z Parameters
3.1.2 Y Parameters
3.1.3 h Parameters
3.1.4 ABCD Parameters

113
113
114
115
116

3.2

Development of Network S-Parameters

117

3.3

Using S Parameter Files in Genesys

3.3.1 Scalar Representation of the S Parameters

120
123

3.4

Development of the Smith Chart

3.4.1 Normalized Impedance on the Smith Chart
3.4.2 Admittance on the Smith Chart

124
126
128

3.5

Lumped Element Movements on the Smith Chart

3.5.1 Adding a Series Reactance to an Impedance
3.5.2 Adding a Shunt Reactance to an Impedance

130
130
132

3.6

VSWR Circles on the Smith Chart

134

3.7

Adding a Transmission Line in Series with an Impedance

137

vii
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

3.8

Adding a Transmission Line in Parallel with an Impedance

3.8.1 Short Circuit Transmission Lines
3.8.2 Open Circuit Transmission Lines

139
140
141

3.9

Open and Short Circuit Shunt Transmission Lines

141

References and Further Reading

Problems

144
144

Chapter 4 Resonant Circuits and Filters

147

4.1

Introduction

147

4.2

Resonant Circuits
4.2.1 Series Resonant Circuits
4.2.2 Parallel Resonant Circuits
4.2.3 Resonant Circuit Loss
4.2.4 Loaded Q and External Q

147
147
149
150
151

4.3

Lumped Element Parallel Resonator Design

4.3.1 Effect of Load Resistance on Bandwidth and QL

152
154

4.4

Lumped Element Resonator Decoupling

4.4.1 Tapped Capacitor Resonator
4.4.2 Tapped Inductor Resonator

155
156
157

4.5

Practical Microwave Resonators

4.5.1 Transmission Line Resonators
4.5.2 Microstrip Resonator Example
4.5.3 Genesys Model of the Microstrip Resonator

158
159
162
164

4.6

Resonator Series Reactance Coupling

166
4.6.1 One Port Microwave Resonator Analysis
167
4.6.2 Smith Chart Qo Measurement of the Microstrip Resonator 171

4.7

Filter Design at RF and Microwave Frequency

4.7.1 Filter Topology
4.7.2 Filter Order
4.7.3 Filter Type
4.7.4 Filter Return Loss and Passband Ripple
viii
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

175
176
177
178
180

4.8

Lumped Element Filter Design

4.8.1 Low Pass Filter Design Example
4.8.2 Physical Model of the Low Pass Filter in Genesys
4.8.3 High Pass Filter Design Example
4.8.4 Physical Model of the High Pass Filter in Genesys
4.8.5 Tuning the High Pass Filter Response
4.8.6 S Parameter File Tuning with VBScript

183
183
185
187
188
189
190

4.9

Distributed Filter Design

4.9.1 Microstrip Stepped Impedance Low Pass Filter Design
4.9.2 Lumped Element to Distributed Element Conversion
4.9.3 Electromagnetic Modeling of the Stepped Impedance
Filter
4.9.4 Reentrant Modes
4.9.5 Microstrip Coupled Line Filter Design
4.9.6 Electromagnetic Analysis of the Edge Coupled Filter
4.9.7 Enclosure Effects

195
195
196

References and Further Reading

Problems

212
213

200
204
205
207
210

Chapter 5 Power Transfer and Impedance Matching

217

5.1

Introduction

217

5.2

Power Transfer Basics

5.2.1 Maximum Power Transfer Conditions
5.2.2 Maximum Power Transfer with Purely Resistive Source
and Load Impedance
5.2.3 Maximum Power Transfer Validation in Genesys
5.2.4 Maximum Power Transfer with Complex Load Impedance

217
218

5.3

Analytical Design of Impedance Matching Networks

5.3.1 Matching a Complex Load to Complex Source Impedance
5.3.2 Matching a Complex Load to a Real Source Impedance
5.3.3 Matching a Real Load to a Real Source Impedance

225
227
234
242

5.4

Introduction to Broadband Matching Networks

247

ix
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

220
222
224

5.4.1 Analytical Design of Broadband Matching Networks

5.4.2 Broadband Impedance Matching Using N-Cascaded
L-Networks
5.4.3 Derivation of Equations for Q and the number of
L-Networks

247

5.5

Designing with Q-Curves on the Smith Chart

5.5.1 Q-Curve Matching Example

259
261

5.6

Limitations of Broadband Matching

5.6.1 Example of Fanos Limit Calculation

264
265

5.7

Matching Network Synthesis

5.7.1 Filter Characteristics of the L-networks
5.7.2 L-Network Impedance Matching Utility
5.7.3 Network Matching Synthesis Utility in Genesys
5.7.4 Effect of Finite Q on the Matching Networks

266
266
267
270
272

References and Further Reading

Problems

275
275

253
257

Chapter 6 Analysis and Design of Distributed

Matching Networks

277

6.1

Introduction

277

6.2

Quarter-Wave Matching Networks

6.2.1 Analysis of Quarter-Wave Matching Networks
6.2.2 Analytical Design of Quarter-Wave Matching Networks

277
278
281

6.3

Quarter-Wave Network Matching Bandwidth

6.3.1 Effect of Load Impedance on Matching Bandwidth
6.3.2 Quarter-Wave Network Matching Bandwidth and Power
Loss in Genesys

286
286

Single-Stub Matching Networks

6.4.1 Analytical Design of Series Transmission Line
6.4.2 Analytical Design of Shunt Transmission Line
6.4.3 Single-Stub Matching Design Example

292
293
295
296

6.4

x
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

290

6.4.4 Automated Calculation of Line and Stub Lengths

6.4.5 Development of Single-Stub Matching Utility

298
299

6.5

Graphical Design of Single-Stub Matching Networks

6.5.1 Smith Chart Design Using an Open Circuit Stub
6.5.2 Smith Chart Design Using a Short Circuit Stub

301
301
303

6.6

Design of Cascaded Single-Stub Matching Networks

304

6.7

Broadband Quarter-Wave Matching Network Design

307

References and Further Reading

Problems

318
319

Chapter 7 Single Stage Amplifier Design

321

7.1

Introduction

321

7.2

Maximum Gain Amplifier Design

7.2.1 Transistor Stability Considerations
7.2 2 Stabilizing the Device in Genesys
7.2.3 Finding Simultaneous Match Reflection Coefficients
and Impedances

322
323
325

7.3

Analytical and Graphical Impedance Matching Techniques

7.3.1 Analytical Design of the Input Matching Networks
7.3.2 Synthesis Based Input Matching Networks
7.3.3 Synthesis Based Output Matching Networks
7.3.4 Ideal Model of the Maximum Gain Amplifier

328
329
331
333
334

7.4

Physical Model of the Amplifier

7.4.1 Transistor Artwork Replacement
7.4.2 Amplifier Physical Design and Layout
7.4.3 Optimization of the Amplifier Response
7.4.4 Optimization Setup Procedure

336
337
339
343
344

7.5

Specific Gain Amplifier Design

7.5.1 Specific Gain Match
7.5.2 Specific Gain Design Example
7.5.3 Graphical Impedance Matching Circuit Design

347
347
351
356

xi
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

328

7.5.4 Assembly and Simulation of the Specific Gain Amplifier

358

7.6

Low Noise Amplifier Design

7.6.1 Noise Circles
7.6.2 LNA Design Example
7.6.3 Analytical Design of the LNA Input Matching Network
7.6.4 Analytical Design of the LNA Output Matching Network
7.6.5 Linear Simulation of the Low Noise Amplifier
7.6.6 Amplifier Noise Temperature

360
362
365
367
368
371
373

7.7

Power Amplifier Design

7.7.1 Data Sheet Large Signal Impedance
7.7.2 Power Amplifier Matching Network Design
7.7.3 Input Matching Network Design
7.7.4 Output Matching Network Design

376
377
379
379
382

References and Further Reading

Problems

386
386

Chapter 8 Multi-Stage Amplifier Design and Yield Analysis 391

8.1

Introduction

391

8.2

Two-Stage Amplifier Design

8.2.1 First Stage Matching Network Design
8.2.2 Analytical Design of the Amplifier Input Matching
Network
8.2.3 Second Stage Matching Network Design
8.2.4 Inter-Stage Matching Network Design
8.2.5 Second Stage Output Matching Network

391
392

8.3

Two-Stage Amplifier Simulation

396

8.4

Parameter Sweeps

398

8.5

Monte Carlo and Sensitivity Analysis

400

8.6

Yield Analysis
8.6.1 Design Centering

405
407
xii

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

393
394
395
396

8.7

Low Noise Amplifier Cascade

8.7.1 Cascaded Gain and Noise Figure
8.7.2 Impedance Match and the Friis Formula
8.7.3 Reducing the Effect of Source Impedance Variation

408
408
410
412

8.8

Summary

413

References and Further Reading

Problems

414
414

Appendix
Appendix A
Appendix B.1
Appendix B.2
Appendix B.3
Appendix B.4
Appendix C
Appendix D

417
Straight Wire Parameters for Solid Copper Wire
i Line Generation
QL Lines on the Smith Chart
Ideal Q Circle on the Smith Chart
Q0 Measurement on the Smith Chart
VBScript file listing for the Matching Utility of
Chapter 5
VBScript file listing for the Line and Stub Matching
Utility of Chapter 6

417
418
420
422
424
425

Index

434
439

About the Authors

445

xiii
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

xiv
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

Foreword
Unlike many traditional books on RF and microwave engineering
written mainly for the classroom, this book adopts a practical, hands-on
approach to quickly introduce and familiarize engineers and students new to
this subject. The authors extensively include the use of electronic design
automation (EDA) tools to illustrate the foundation principles of RF and
microwave engineering. The use of EDA methodology in the book closely
parallels the latest tools and techniques used in the industry to accelerate the
design of RF and microwave systems and components to meet demanding
specifications and high yields.
This book introduces not only a solid understanding of RF and
Microwave concepts such as the Smith chart, S-parameters, transmission
lines, impedance matching, filters and amplifiers, but also more importantly
how to use EDA tools to synthesize, simulate, tune and optimize these
essential components in a design flow as practiced in the industry. The
authors made the judicious choice of an easy-to-use and full featured EDA
tool that is also very affordable so that the skills learnt from the book can be
put into practice immediately without the barriers of acquiring costly and
complex EDA tools.
Genesys from Agilent Technologies was chosen for its low cost and
ideal combination of capabilities in circuit synthesis, simulation and
optimization; Matlab equation handling; plus RF system, electromagnetic
and statistical analysis. It is proven by Agilent Technologies in the design of
state-of-the-art RF and microwave test instrumentation and time-tested by a
large following of users worldwide for over 20 years.
The investment in learning the RF and microwave foundation skills
with EDA techniques taught in this book results in knowledge that remains
relevant and sought-after for a long time to come.

xv
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

I wish such a book was available when I started my career as a

microwave component designer. It would have made gaining RF and
microwave insights much quicker than the countless hours of cut-and-try on
the bench.

How-Siang Yap
Agilent EEsof EDA Genesys Planning & Marketing
1400 Fountaingrove Parkway
Santa Rosa, CA 95403, USA

Xvi
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

C
o
p

Preface
Microwave Engineering can be a fascinating and fulfilling career path.
It is also an extremely vast subject with topics ranging from semiconductor
physics to electromagnetic theory. Unlike many texts on the subject this
book does not attempt to cover every aspect of Microwave Engineering in a
single volume. This text book is the first volume of a two-part series that
covers the subject from a computer aided design standpoint. The first
volume covers introductory topics which are appropriate to be addressed by
linear simulation methods. This includes topics such as lumped element
components, transmission lines, impedance matching, and basic linear
amplifier design. The second volume focuses on subject matter that is better
learned through non-linear computer simulation. This includes topics such
as oscillators, mixers, and power amplifier design.
Almost all subject matter covered in the text is accompanied by
examples that are solved using the Genesys linear simulation software by
Agilent. University students will find this a potent learning tool. Practicing
engineers will find the book very useful as a reference guide to quickly
setup designs using the Genesys software. The authors thoroughly cover the
basics as well as introducing CAD techniques that may not be familiar to
some engineers. This includes subjects such as the frequent use of the
Genesys equation editor and Visual Basic scripting capability. There are
also topics that are not usually covered such as techniques to evaluate the Q
factor of one port resonators and yield analysis of microwave circuits.
The organization of the book is as follows: Chapter 1 presents a
general explanation of RF and microwave concepts and components.
Engineering students will be surprised to find out that resistors, inductors,
and capacitors at high frequencies are no longer ideal elements but rather a
network of circuit elements. For example, a capacitor at one frequency may
in fact behave as an inductor at another frequency. In chapter 2 the
transmission line theory is developed and several important parameters are
defined. It is shown how to simulate and measure these parameters using
Genesys software. Popular types of transmission lines are introduced and
their parameters are examined. In Chapter 3 network parameters and the
Xvii
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

application of Smith chart as a graphical tool in dealing with impedance

behavior and reflection coefficient are discussed. Description of RF and
microwave networks in terms of their scattering parameters, known as SParameters, is introduced. The subject of lumped and distributed resonant
circuits and filters are discussed in Chapter 4. Using the Genesys software
a robust technique is developed for the evaluation of Q factor form the SParameters of a resonant circuit. An introduction to the vast subject of
filter synthesis and the electromagnetic simulation of distributed filters are
also treated in this chapter. In Chapter 5 the condition for maximum
power transfer and the lumped element impedance matching are
considered. The analytical equations for matching two complex
impedances with lossless two-element networks are derived. Both
analytical and graphical techniques are used to design narrowband and
broadband matching networks. The Genesys impedance matching
synthesis program is used to solve impedance matching problems. The
VBScript programming techniques developed in this chapter can be used
by students to generate their own synthesis applications within the
Genesys software. In Chapter 6 both narrowband and broadband
distributed matching networks are analytically and graphically analyzed.
In Chapter 7 single-stage amplifiers are designed by utilizing four
different impedance matching objectives. The first amplifier is designed
for maxim gain where the input and the output are conjugately matched to
the source and load impedance; the second amplifier is designed for
specific gain where the input or the output is mismatched to achieve a
specific gain less than its maximum; the third amplifier is a low noise
amplifier where the transistor is selectively mismatched to achieve a
specific Noise Figure; and the fourth amplifier is a power amplifier where
the transistor is selectively mismatched to achieve a specific amount of
output power. In Chapter 8 a two-stage amplifier is designed by utilizing a
direct interstage matching network. Monte Carlo and Yield analysis
techniques are also introduced in this chapter. Finally a brief introduction
to cascade analysis is presented.
Ali A. Behagi
Stephen D. Turner
August 2011
Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

1.1 Introduction
An electromagnetic wave is a propagating wave that consists of electric and
magnetic fields. The electric field is produced by stationary electric charges
while the magnetic field is produced by moving electric charges. A timevarying magnetic field produces an electric field and a time-varying electric
field produces a magnetic field. The characteristics of electromagnetic
waves are frequency, wavelength, phase, impedance, and power density. In
free space, the relationship between the wavelength and frequency is given
by Equation (1-1).
c
(1-1)

f
In the MKS system, is the wavelength of the signal in meters, c is the
velocity of light approximately equal to 300,000 kilometers per second, and
f is the frequency in cycles per second, or Hz.

Figure 1-1 A time varying voltage waveform

The electromagnetic spectrum is the range of all possible frequencies of

electromagnetic radiation. They include radio waves, microwaves, infrared
radiation, visible light, ultraviolet radiation, X-rays and gamma rays. In the
field of RF and microwave engineering the term RF generally refers to

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

Radio Frequency signals with frequencies in the 3 KHz to 300 MHz range.
The term Microwave refers to signals with frequencies from 300 MHz to
300 GHz having wavelengths from 1 meter to 1 millimeter. The RF and
microwave frequencies form the spectrum of all radio, television, data, and
satellite communications. Figure 1-2 shows a spectrum chart highlighting
the RF and microwave frequencies up through the extremely high
frequency, EHF, range or 300 GHz. This text will focus on the RF and
microwave frequencies as the foundation for component design techniques.
The application of Agilents Genesys software will enhance the students
understanding of the underlying principles presented throughout the text.
The practicing engineer will find the text an invaluable reference to the RF
and microwave theory and techniques by using Genesys software. The
numerous Genesys examples enable the setup and design of many RF and
microwave circuit design problems.

Figure 1-2 Electromagnetic spectrums from VLF to EHF

The spectrum chart of Figure 1-2 is intended as a general guideline to the

commercial nomenclature for various sub bands. There is typically overlap
across each of the boundaries as there is no strict dividing line between the
categories. The RF frequencies typically begin in the very low frequency,
VLF, range through the very high frequency, VHF, range. Microwaves are
typically the ultra high frequency, UHF, super high frequency, SHF and
extremely high frequency, EHF, frequency ranges. During World War II
microwave engineers developed a further detailed classification of the
microwave frequencies into a band-letter designation. In 1984 the Institute
of Electrical and Electronics Engineers, IEEE, agreed to standardize the
letter designation of the microwave frequencies. These designators and their
frequency ranges are shown in Table 1-1.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

Band
Designator
Frequency
Range
GHz

L
Band

S
Band

C
Band

X
Band

Ku
Band

K
Band

Ka
Band

1 to 2

2 to 4

4 to 8

8 to 12

12 to 18

18 to 27

27 to 40

Table 1-1 Microwave band letter designators

Engineering students spend much of their formal education learning the

basics of inductors, capacitors, and resistors. Many are surprised to find that
as we enter the high frequency, HF, part of the electromagnetic spectrum
these components are no longer a singular (ideal) element but rather a
network of circuit elements. Components at RF and microwave frequencies
become a network of resistors, capacitors, and inductors. This leads to the
complication that the components characteristics become quite frequency
dependent. For example, we will see in this chapter that a capacitor at one
frequency may in fact be an inductor at another frequency.

1.2 Straight Wire, Flat Ribbon, and Skin Effects

In this section we will begin with a basic examination of the straight wire
inductance and move into more complete characterization of inductors.
Similarly we will look at resistor and capacitor design and their
implementation at RF and microwave frequencies. Discrete resistors,
capacitors, and inductors are often referred to as lumped elements. RF and
microwave engineers use the terminology to differentiate these elements
from those designed in distributed printed circuit traces. Distributed
component design is introduced in Chapter 2.

1.2.1 Straight Wire Inductance

A conducting wire carrying an AC current produces a changing magnetic
field around the wire. According to Faradays law the changing magnetic
field induces a voltage in the wire that opposes any change in the current
flow. This opposition to change is called self inductance. At high
frequencies even a short piece of straight wire possesses frequency
dependent resistance and inductance behaving as a circuit element.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

Example 1.2-1: Calculate the inductance of a three inch length of AWG

#28 copper wire in free space.
Solution: The straight wire inductance can be calculated from the empirical
Equation (1-2) [1].
4

L K ln
0.75 nH
D

(1-2)

where:

Length of the wire

D = Diameter of the wire (from Appendix A).
K = 2 for dimensions in cm and K=5.08 for dimensions in inches
Using Appendix A the diameter of the AWG#28 wire is found to be 0.0126
inches. Solving Equation (1-2) the inductance is calculated.
4 3

L 5.08 3 ln
0.75 93.1 nH
0.0126

It is interesting to examine the reactance of the wire. We know that the

reactance is a function of the frequency and is related to the inductance by
the following equation.
X L 2 f L
(1-3)
where: f is the frequency in Hz and L is the inductance in Henries
Calculating the reactance at 60Hz, 1MHz, and 1GHz we can see how the
reactive component of the wire increases dramatically with frequency. At
60Hz the reactance is well below 1 while at microwave frequencies the
reactance increases to several hundred ohms.

2 10 93.110 0.58
2 10 93.110 585

60 Hz:

X L 2 60 93.110 9 35

1 MHz:

XL

1 GHz:

XL

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

1.2.2 Simulating the Straight Wire Inductor in Genesys

To analyze the Example 1.2-1 in Genesys, create a schematic in Genesys
and add the straight wire model from the Parts Library. Attach a standard
input and output port as shown in Figure 1-3.

Figure 1-3 Genesys schematic of the straight wire

In Genesys, the input port represents a signal source in series with a source
impedance and the output port represents the load impedance connected to
ground. Set the wire diameter to 12.6 mils and the length to 3 inches.
Accept the value of 1 for Rho. Rho is not the actual resistivity of the wire
but rather the resistivity of the wire relative to copper. Because we are
modeling a copper wire the value should be set to one. It is a common
practice in most commercial microwave software programs to specify
resistivity in relative terms, compared to copper. Table 1-2 provides a
reference of common materials used in RF and microwave engineering in
terms of their actual resistivity and the relative resistivity to copper.
Material
Copper, annealed
Silver
Gold
Aluminum
Tungsten
Zinc
Nickel
Iron
Platinum
Tin
Nichrome
Carbon

Resistivity
Relative to
Copper
1.00
0.95
1.42
1.64
3.25
3.40
5.05
5.45
6.16
52.8
65.5
2083.3

Actual Resistivity
-meters
1.6810-8
1.5910-8
2.3510-8
2.6510-8
5.6010-8
5.9010-8
6.8410-8
1.0010-7
1.0610-7
1.0910-7
1.10106
3.50105

Actual
Resistivity inches
6.6110-7
6.2610-7
9.2510-7
1.0410-6
2.2010-6
2.3210-6
2.6910-6
3.9410-6
4.1710-6
4.2910-6
4.33105
1.38103

Table 1-2 Resistivity of common materials relative to copper

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

Create a Linear analysis to analyze the circuits impedance versus

frequency. The Linear Analysis Properties window is shown in Figure 1-4.

Figure 1-4 Linear analysis properties for simulation of wire impedance

Create a list of frequencies under the Type of Sweep setting. Enter the
frequencies that were used in the reactance calculation of the previous
section: 60 Hz, 1 MHz, 1 GHz. When an analysis is run in Genesys the
results are written to a Dataset. The results of a Dataset may then be sent to
a graph or tabular output for visualization. In this example an Equation
Editor is used to post process the solutions in the Dataset. Create an
Equation Editor, as shown in Figure 1-5, for the calculation of the wires
reactance and inductance. Add the Equation Editor to the design as follows.

Figure 1-5 Equation editor display for the calculation of the Inductance

More complex workspaces may contain multiple Datasets. It is a good

practice to specify which Dataset is used to collect data for post processing.
This is accomplished with the {using (Linear1_Data)} statement of line 1
in the Equation Editor. Genesys has a built-in function ZIN1 to calculate the

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

impedance of the circuit at each analysis frequency. We know that the

reactance is the imaginary part of the impedance. Line 2 of the Equation
Editor defines the reactance as the imaginary part of the impedance, ZIN1,
in the Linear1 Dataset. Line 3 calculates the inductance of the wire from the
reactance using Equation (1-3). The frequency, F, is the independent
variable created by the Linear Analysis. Note the use of the dot (.) notation
in the Equation Editor. There is a dot after Linear1_Data and reactance of
lines 2 and 3. The dot means that these variables are not singular quantities
but are arrays of values. There is a calculated array value for each
independent variable, F. The Equation Editor is an extremely powerful
feature of the Genesys software and is used frequently throughout this text.
It is an interactive mathematical processor similar to MATLAB by
MathWorks. There are two different syntaxes that may be used to define
equations and perform post processing operations using the Equation
Editor: the Engineering language and the Mathematics language. The
Engineering language is a simple structured format as shown in Figure 1-5.
The Mathematics language is compatible with the m-file syntax that is used
in MATLAB. It is very convenient for students and engineers that are
proficient in MATLAB. Both types will be used throughout this book to
demonstrate the use of both languages. The tab in the upper left corner of
Figure 1-5 displays the language syntax used in that particular Equation
Editor. Send the simulated value of reactance and the calculated value of
inductance to a tabular output. Figure 1-6 shows the tabular output and the
table properties. On the Table Properties tab, select the Equations as the
data source and then select both of the variables, reactance and inductance.

Figure 1-6 Output Table Properties window

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

Compare the results of the simulation of Figure 1-6 with the calculated
values of section 1.2.1. We can see that at 60 Hz and 1 MHz the reactance
and resulting inductance are very close to the calculated values. At 1 GHz
however, the values begin to diverge. The simulated reactance is 24 less
than the calculated reactance. Equation (1-2) was useful for calculating the
basic inductance at low frequency but as the frequency enters the
microwave region, the value begins to change. This is normal and due to the
skin effect of the conductor. The skin effect is a property of conductors
where, as the frequency increases, the current density concentrates on the
outer surface of the conductor.

1.2.3 Skin Effect in Conductors

At RF and microwave frequencies, due to the larger inductive reactance
caused by the increase in flux linkage toward the center of the conductor,
the current in the conductor is forced to flow near the conductor surface. As
a result the amplitude of the current density decays exponentially with the
depth of penetration from the surface. Figure 1-7 shows the cross section of
a cylindrical wire with the current density area shaded.

Cross Section Area

of Wire at DC

Cross Section Area of Wire

at High Frequency

Figure 1-7 Cross section of current flow in the conductor showing effect of
skin depth

At low frequencies the entire cross sectional area is carrying the current. As
the frequency increases to the RF and microwave region, the current flows
much closer to the outside of the conductor. At the higher end of microwave
frequency range, the current is essentially carried near the surface with

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

almost no current at the central region of the conductor. The skin depth, d,
is the distance from the surface where the charge carrier density falls to
37% of its value at the surface. Therefore 63% of the RF current flows
within the skin depth region. The skin depth is a function of the frequency
and the properties of the conductor as defined by Equation (1-4). As the
cross sectional area of the conductor effectively decreases the resistance of
the conductor will increase [7].
d

where:

(1-4)

d = skin depth
resistivity of the conductor
f = frequency
permeability of the conductor

Use caution when solving Equation (1-4) to keep the units of and
consistent. Table 1-2 contains values of resistivity in units of -meters and
-inches. The permeability is the permeability of the conductor. It is a
property of a material to support a magnetic flux. Some reference tables will
show relative permeability. In this case the relative permeability is
normalized to the permeability of free space which is: 4Henries per
meter. The relationship between relative permeability to the actual
permeability is given in Equation (1-5). Most conductors have a relative
permeability r very close to one. Therefore conductor permeability is
often given the same value as o.

r o

where:
= actual permeability of the material
r = relative permeability of material
o = permeability of free space
Example 1.2-2: Calculate the skin depth of copper wire at a frequency of
25 MHz.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

10

Solution: From Table 1-2, 6.6110-7 -inches. Using Equation (1-4),

and converting the permeability from H/m to H/inch, the skin depth is:
d

6.61 10 7

3.19 10 25 10
8

5.14 10 4 inches

Although we have been considering the skin depth in a circular wire, skin
depth is present in all shapes of conductors. A thick conductor is affected
more by skin effect at lower frequencies than a thinner conductor. One of
the reasons that engineers are concerned about skin effect in conductors is
the fact that as the resistance of the conductor increases, so does the thermal
heating in the wire. Heat can be a destructive force in high power RF
circuits causing burn out of conductors and potentially hazardous conditions
to personnel. Also the frequency dependence of the skin effect may make it
difficult to maintain the impedance of a transmission line structure. This
effect will be examined in Chapter 2 with the study of transmission lines.
As the frequency increases, the current is primarily flowing in the region of
the skin depth. It can be visualized from Figure 1-7 that a wire would have
greater resistance at higher frequencies due to the skin effect. The resistance
of a length of wire is determined by the resistivity and the geometry of the
wire as defined by Equation (1-6).
R

where:

(1-6)

= Resistivity of the wire

Length of the wire
A = Cross sectional area

Example 1.2-3: Calculate the resistance of a 12 inch length of AWG #24

copper wire at DC and at 25 MHz.
Solution: The radius of the wire can be found in Appendix A. The DC
resistance is then calculated using Equation (1-6).

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

11

RF and Microwave Concepts and Components

6.6110 12 0.0021
7

RDC

0.0201

To calculate the resistance at 25 MHz the cross sectional area of the

conduction region must be redefined by the skin depth of Figure 1-7. We
can refer to this as the effective area, Aeff.
Aeff ( R 2 r 2 )

(1-7)

where: r = R-d
For the AWG#24 wire at 25 MHz the Aeff is calculated as:
2

0.0201
0.0201
4
5
2
Aeff

5.14 10 3.14 10 in
2
2

Then apply Equation (1-6) to calculate the resistance of the 12 inch wire at
25 MHz.
R25MHz

7
12 6.61 10

3.14 10

0.253

We can see that the resistance at 25 MHz is more than 10 times greater than
the resistance at DC.

1.2.4 Analytical Calculation of Flat Ribbon Inductance

Flat ribbon style conductors are very common in RF and microwave
engineering. Flat ribbon conductors are encountered in RF systems in the
form of low inductance ground straps. Flat ribbon conductors can also be
encountered in microwave integrated circuits (MIC) as gold bonding straps.
When a very low inductance is required the flat ribbon or copper strap is a
good choice. The flat ribbon inductance can be calculated from the
empirical Equation (1-8).
LK

W T
ln
0.223
0.5

W T

nH

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

(1-8)

RF and Microwave Concepts and Components

where:

12

The length of the wire

K= 2 for dimensions in cm and K=5.08 for dimensions in inches
W = the width of the conductor
T = the thickness of the conductor

Example 1.2-4: Calculate the inductance of the 3 inch Ribbon at 60 Hz, 1

MHz, and 1 GHz. Make the ribbon 100 mils wide and 2 mils thick.
Solution: The schematic of the Ribbon is shown in Figure 1-8 along with
the table with the resulting reactance and inductance. Note that the same
length of ribbon as the AWG#28 wire has almost half of the inductance
value. Also note that the inductance value, 49.69 nH does not change over
the 60 Hz to 1 GHz frequency range. It is why the ribbon is desirable as a
low inductance conductor for use in RF and microwave applications.

Figure 1-8 Flat ribbon schematic and tabular output of reactance and
inductance

1.3 Physical Resistors

The resistance of a material determines the rate at which electrical energy is
converted to heat. In Table 1-2 we have seen that the resistivity of materials
is specified in -meters rather than /meter. This facilitates the calculation
of resistance using Equation (1-6). When working with low frequency or
logic circuits we are used to treating resistors as ideal resistive components.
If we examine the impedance of an ideal resistor in Genesys over a
frequency range of 1 MHz to 3 GHz we obtain a plot as shown in Figure 19. This plot shows a constant resistance at all frequencies. At RF and
microwave frequencies however, resistors also possess inductive and
capacitive elements. The stray inductance and capacitance associated with a
resistor are often called parasitic elements. Consider the leaded resistor as
shown in Figure 1-10.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

Figure 1-9 Ideal 50 resistor impedance versus frequency

Figure 1-10 Leaded resistor impedance versus frequency for various

resistor values

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

13

14

RF and Microwave Concepts and Components

For such a 1/8 watt leaded resistor it is not uncommon for each lead to have
about 10 nH of inductance. The body of the resistor may exhibit 0.5 pF
capacitance between the leads. Designing the network in Genesys reveals an
interesting result of the impedance versus frequency response. The
impedance plotted in Figure 1-10 shows the impedance for a 10 , 50 ,
500 , and 1 k resistors swept from 0 to 2.0 GHz. By tuning the parasitic
elements in Genesys we find that the low value resistors are influenced
more by the lead inductance. The high value resistors are influenced more
by the parasitic capacitance.

1.3.1 Chip Resistors

Thick film resistors are used in most contemporary electronic equipment.
The thick film resistor, often called chip resistor, comes close to eliminating
much of the inductance that plagues the leaded resistor. The chip resistor
works well with popular surface mount assembly techniques preferred in
modern electronic manufacturing. Figure 1-11 shows a typical thick film
chip resistor along with a cross section of its design.

Figure 1-11 Thick film chip resistors and dimensions (courtesy of KOA
Speer Electronics)

There are many types of chip resistors designed for specific applications.
Common sizes and power ratings are shown in Table 1-3.
Size
0201
0402

Length x Width
20mils x 10mils
40mils x 20mils

Power Rating
50mW
62mW

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

15

RF and Microwave Concepts and Components

0603
0805
1206
2010
2512

60mils x 30mils
80mils x 50mils
120mils x 60mils
200mils x 100mils
250mils x 120mils

100mW
125mW
250mW
500mW
1W

Table 1-3 Standard thick film resistor size and approximate power rating

The thick film resistor is comprised of a carbon based film that is deposited
onto the substrate. Contrasted with a thin film resistor that is typically
etched onto a substrate or printed circuit board, the thick film resistor can
usually handle higher power dissipation. The ends of the chip have
metalized wraps that are used to attach the resistor to a circuit board.
Genesys does not include native models for thick film resistors. Some
manufacturers may provide models that can be incorporated into Genesys
There are also companies that specialize in developing CAD models of
components such as Modelithics, Inc. Modelithics has a wide variety of
component model libraries that can be incorporated into Genesys.
Example 1.3-1: Plot the impedance of 1 k, 0603 size chip resistor,
manufactured by KOA, from 0 to 3 GHz. This model is available in the
Modelithics evaluation model kit. Select the Modelithics Library in the
Genesys Part Selector as shown in Figure 1-12.

Figure 1-12 Modelithics chip component library in Genesys

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

16

Solution: Create a schematic with the resistor and sweep the impedance
from 1MHz to 3 GHz. The resulting schematic and response are shown in
Figure 1-13.

Figure 1-13 Modelithics chip resistor model schematic and impedance

versus frequency

Note the roll off of the impedance with increasing frequency. This suggests
that the chip resistor does have a parasitic capacitance that is in parallel with
the resistor similar to the discrete model of Figure 1-10.

1.4 Physical Inductors

In section 1.2 we introduced the topic of inductance. The inductance of
straight cylindrical and flat rectangular conductors was examined. The
primary method of increasing inductance is not to simply keep increasing
the length of a straight conductor but rather form a coil of wire. Forming a
coil of wire increases the magnetic flux linkage and greatly increases the
overall inductance. Because of the greater surrounding magnetic flux,
inductors store energy in the magnetic field. Lumped element inductors are
used in bias circuits, impedance matching networks, filters, and resonators.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

17

RF and Microwave Concepts and Components

As we will see throughout this section inductors are realized in many forms
including: air-core, toroidal and very small chip inductors. The concept of Q
factor is introduced and will come up frequently in RF and microwave
circuit design. It is a unit-less figure of merit that is used in circuits in which
both reactive and resistive elements coexist. Because we know that
individual air wound inductors, shown in Figure 1-14, are actually networks
that are made up of resistors, inductors, and capacitors, each individual
component is also characterized by a Q factor.

Figure 1-14 Air wound inductor showing the wire resistance and interwinding capacitance

Basically, the higher the Q factor, the less loss or resistance exists in the
energy storage property. The inductor quality factor Q is defined as:

X
RS

(1-9)

where:
X is the reactance of the inductor
Rs is the resistance in the inductor
At low RF frequencies the resistance comes primarily from the resistivity of
the wire and as such is quite low. At higher frequencies the skin effect and
inter-winding capacitance begin to influence the resistance and reactance
thus causing the Q factor to decrease. In most applications we want as high
a component Q factor as possible. We can increase the Q factor of inductors
by using larger diameter wire or by silver plating the wire. In a multi-turn
coil, the windings can be separated to reduce the inter-winding capacitance

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

18

RF and Microwave Concepts and Components

which in turn will increase the Q factor. Winding the coil on a magnetic
core can increase the Q factor.

1.4.1 Air Core Inductors

Forming a wire on a removable cylinder is the basic realization of the air
core inductor. When designing an air-core inductor, use the largest wire size
and close spaced windings to result in the lowest series resistance and high
Q. The basic empirical equation to calculate the inductance of an air core
inductor is given by Equation (1-10) [2].

17 N D D1
L
D1 S

1.7

1.3

0.7

(1-10)

where:
N = Number of turns of wire
D = Core
form diameter in inches
D1 = Wire diameter in inches
L = Coil inductance in nH
S = Spacing between turns in inches
Example 1.4-1: As an interesting comparison with Example 1.2.1 calculate
the amount of inductance that we can realize in that same three inches of
wire if we wind it around a core to form an inductor. Choose a core form of
0.095 inches diameter as a convenient form to wrap the wire around.
Solution: First we need to calculate the approximate number of turns that
we can expect to have with the three inch length of wire. We know that the
circumference of a circle is related to the diameter by the following
equation.
Circumference Diameter 0.095 0.2985 inches

With a circumference of 0.2985 inches we can calculate the approximate

number of turns that we can wrap around the 0.095 inch core with three
inches of wire.
N

3
approximately 10 turns
0.2985

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

19

From Equation (1-10) we can see that the spacing between the turns has a
strong effect on the value of the inductance that we can expect from the
coil. When hand winding the coil, it may be difficult to maintain an exact
spacing of zero inches between the turns. Therefore it is useful to solve
Equation (1-10) in terms of a variety of coil spacing so that we can see the
effect on the inductance. The Equation Editor in Genesys can be used to
solve Equation (1-10) for a variety of coil spacing. Add an Equation Editor
to the Workspace as shown in Figure 1-15.

Figure 1-15 Adding an Equation Editor to the Genesys workspace

Create an Equation set similar to that shown in Figure 1-16.

Figure 1-16 Equation Editor to calculate the air core coil inductance

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

20

Note that the coil spacing variable, Spacing, has been defined as an array
variable. Placing a semicolon between the values makes the array organized
in a column format. This is handy for viewing the results in tabular format.
Placing a comma between the values would organize the array in a row
format. The array variables are displayed under the variable column as Real
[6], indicating that these values are a six element array containing scalar
values. As an aid in forming the coil, the overall coil length is also
calculated with the Equation Editor. The coil length is simply the
summation of the overall wire thickness times the number of turns and the
spacing between the turns. A quick way to add the results to a table is
shown in Figure 1-17.

Figure 1-17 Send the calculated variable to a tabular output directly from
the Equation Editor

Right click on the variable and add it to an existing or new table. Add the
calculated inductance, spacing, and coil length to the table as shown in
Figure 1-18.

Figure 1-18 Coil Inductance versus coil spacing and length

The table shows that the coil inductance with no spacing between the turns
is 163.78 nH and it is 0.126 inches long. Contrast this to the 93.1 nH
inductance with the same three inches of wire in a straight length. We can

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

21

clearly see the dramatic impact of the magnetic flux linkage in increasing
the inductance by forming the wire into a coil. The Figure also shows the
strong influence of the inter-winding capacitance in influencing the
inductance of the coil. Just 10 mils spacing between the turns reduces the
coils inductance from 163.78 nH to 108.8 nH. It is clearly important to
consider the turn spacing when analyzing the inductors performance. In
practice this is an effective means to tune the inductors value in circuit.
When designing and building the inductor it is necessary to solve Equation
(1-10) for the number of turns given a desired value of inductance. Figure
1-19 shows an Equation Editor setup to solve for the number of turns, N.

Figure 1-19 Equation Editor to calculate the air core number of turns

Note that there is one subtle difference in line 12, the calculation of the coil
length. In this case both variables, Spacing and N, are array variables. When
multiplying array variables use a period in front of the multiplication sign to
signify that this is an operation on arrays. This makes sure that the correct
array index is maintained between the variables. In the previous Equation
Editor of Figure 1-16 this notation was not necessary because N was a
constant.

Figure 1-20 Number of turns versus coil spacing and length

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

22

1.4.2 Modeling the Air Core Inductor in Genesys

The air core inductor is modeled in Genesys using the AIRIND1 model as
shown in Figure 1-21.

Figure 1-21 Genesys model of the air core inductor

Note the required parameters: number of turns, wire diameter, core

diameter, and coil length. The coil spacing cannot be entered directly but is
accounted for by the overall coil length for a given number of turns. Make
the coil length a variable so that we can analyze the inductance as a function
of the coil length. Simulate the value of inductance at a fixed frequency of 1
MHz. Set up a fixed frequency Linear Analysis as shown in Figure 1-23.
Simulate the value of inductance at a fixed frequency of 1 MHz.

Figure 1-22 Fixed frequency linear analysis at 1MHz

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

23

Figure 1-23 Adding the Parameter Sweep to the workspace

To vary the length of the inductor we will use the Parameter Sweep
capability in Genesys. Add a Parameter Sweep to the Workspace as shown
in Figure 1-23.

Figure 1-24 Equation Editor to work with parameter sweep dataset

On the Parameter-to-Sweep entry make sure that the coil length is selected
from the drop down box. Under the Type-of-Sweep, select list and enter the
six coil lengths that were calculated in the table of Figure 1-18. Finally
create an Equation Editor as shown in Figure 1-24 to calculate the coils
inductance from the simulated reactance and plot it as a function of the coil

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

24

length as shown in Figure 1-25. Note the use of the Setindep statement in
line 5. This keyword is used to set the dependent and independent variables.
The first variable in the statement is the dependent variable while the
second is the independent variable. In this example the inductance is the
dependent variable while the coil length is the independent variable. As the
plot of Figure 1-25 shows the model has very close correlation with the
inductance calculated with Equation (1-10).

Figure 1-25 Plot of inductance versus coil length for the air inductor model

Change the Linear Analysis to a linear frequency sweep with 401 points
over a range of 1 MHz to 1300 MHz. Plot the impedance of the inductor
across the frequency range as shown in Figure 1-26. Note the interesting
spike, or increase in impedance that occurs around 1098.6 MHz. This is the
parallel, self resonant, frequency of the inductor.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

25

Figure 1-26 Impedance of the air core inductor as a function of frequency

The inductor is not an ideal component or a pure inductance but rather a

network that includes parasitic capacitance and resistance. As an example,
create a simple RLC network that gives an equivalent impedance response
to Figure 1-26. One such circuit is shown in Figure 1-27.

Figure 1-27 Equivalent ideal element network of the air core inductor

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

26

Resonant circuits are covered in detail in chapter 4 but it is important to

understand that each individual component such as the air core inductor has
its own resonant frequency. The resonant frequency is the frequency at
which the inductive reactance and capacitive reactance are equal and cancel
one another. When this condition occurs in the inductor it is a parallel
resonant circuit which results in a very high real impedance. If we plot the
reactance along with the impedance a very interesting response is obtained.
This response is shown in Figure 1-28.

Figure 1-28 Impedance and reactance of the air core inductor

The reactance, up to the resonant frequency, is positive but beyond

resonance the reactance becomes negative. From basic circuit theory we
know that a negative reactance is associated with a capacitor. Therefore
above the Self Resonant Frequency, SRF, the inductor actually becomes a
capacitor. In practice we want to make sure that our inductor really behaves
like an inductor. A good design practice is to keep this self resonant
frequency about four times higher than the frequency of operation. However
using the inductor near its resonant frequency might make a good choke. A
choke is a high reactance inductor often used to feed voltage to a circuit in

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

27

which all RF energy is blocked from the DC side of the circuit. The
inductor manufacturer will typically specify SRF of the inductor. It is
important to remember that the inductors SRF is the parallel resonant
frequency; not the series resonant frequency.

1.4.3 Inductor Q Factor

Example 1.4-2: Setup an Equation Editor to calculate the Q factor of the
inductor based on Equation (1-9).
Solution: The Equation Editor is shown in Figure 1-29.

Figure 1-29 Equation Editor to calculate the Q factor of the inductor

The plot of inductor Q factor versus frequency is shown in Figure 1-30.

Figure 1-30 Air core inductor model Q factor versus frequency

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

28

It is interesting to note that the Q factor peaks at a frequency well below the
self resonant frequency of the inductor. The actual frequency at which the Q
factor peaks will vary among inductor designs but is usually ranges from 2
to 5 times less than the SRF. Close winding spacing results in inter-winding
capacitance, which lowers the self-resonant frequency of the inductor. Thus
there is a tradeoff between maximum Q factor and high self resonant
frequency. It is also noteworthy that the Q factor goes to zero at the self
resonant frequency. Figure 1-30 shows the Q factor monotonously
increasing beyond the self resonant frequency. This is erroneous and is due
to the fact that the model used to simulate the inductors performance is
invalid beyond the self resonance. The Air Core inductor model uses a
simplified network similar to the one shown in Figure 1-27. Beyond self
resonance the complexity and number of ideal elements need to increase in
order to accurately model the inductor. A resistor needs to be added in
series with the capacitor to begin to model the Q factor because the inductor
is becoming a capacitor above self resonance. For most practical work
however the native model will work fine because we should be using the
inductor well below the self resonant frequency. A technique commonly
used by microwave engineers to increase the Q factor of an inductor is to
silver plate the wire. This can be modeled by setting Rho = 0.95 in the
inductor model.

1.4.4 Chip Inductors

The inductor core does not have to be air. Other materials may be used as
the core of an inductor. Similar in size to the chip resistor there is a large
assortment of chip inductors that are popular in surface mount designs. The
chip inductor is a form of dielectric core inductor. There are a variety of
modeling techniques used for chip inductors. One of the more popular
modeling techniques is with the use of S parameter files. The subject of S
parameters is covered in chapter 3. At this point consider the S parameter
file as an external data file that contains an extremely accurate network
model of the component. Most component manufacturers provide S
parameter data files for their products. It is a good practice to always check
the manufacturers website for current S parameter data files. Coilcraft, Inc

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

29

RF and Microwave Concepts and Components

is one manufacturer of chip inductors. A typical chip inductor is built with

extremely small wire formed on a ceramic form as shown in Figure 1-31.

Figure 1-31 chip inductors (courtesy of Coilcraft, Inc)

Chip inductors are manufactured in many standard chip sizes as shown in

Table 1-4.
Size
0201
0302
0402
0603
0805
1008
1206
1812

Length x Width
20mils x 18mils
34mils x 15mils
44mils x 20mils
69mils x 30mils
90mils x 50mils
105mils x 80mils
140mils x 56mils
195mils x 100mils

Table 1-4 Standard chip Inductor size and approximate power rating

The characteristic differences among the various sizes are more difficult to
quantify than the chip resistors. A careful study of the data sheets is
required for the proper selection of a chip inductor. In general the larger
chip inductors will have higher inductance values. Often the smaller chip
inductors will have higher Q factor. The impedance and self resonant
frequency can vary significantly across the sizes as well as the current
handling capability.

1.4.5 Chip Inductor Simulation in Genesys

The Genesys library has a collection of S parameter files for the Coilcraft
chip inductors. Use the Part Selector to navigate to the Coilcraft chip

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

30

inductors and select the 180 nH 1008 series inductor. The Part Selector and
inductor schematic are shown in Figure 1-32.

Figure 1-32 Part selector to locate chip inductor and the circuit schematic

Example 1.4-3: Setup a Linear Analysis and plot the impedance of the
inductor from 1 MHz to 1040 MHz.
Solution: Create an Equation Editor to calculate and plot the inductance
from the impedance as shown in Figure 1-33. For reference the
manufacturers specifications for the inductance is overlaid on the plot.

Figure 1-33 Plot of the 180 nH chip inductor inductance versus frequency

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

31

The manufacturer specifies the inductance as 180 nH at a frequency of 25

MHz. From the marker on the plot we see that the simulation of the S
parameter file has very close correlation measuring 180.1 nH. From the plot
we can see that the nominal inductance value remains close to specification
up to about 500 MHz. The self resonant frequency (parallel resonant) is
specified as a minimum of 750 MHz. The actual SRF could be higher but it
is guaranteed not be be less than 750 MHz. We can see from the plot in
Figure 1-33 that the actural SRF is closer to 1 GHz. From Figure 1-34 we
see that the manufacturer specifies a minimum Q factor of 45 at a frequency
of 100 MHz.

Figure 1-34 180 nH chip inductor Q factor versus frequency

The plot of the Q factor derived from the S parameter file shows that the
Q factor is 51.21 near 100 MHz. Manufacturers will often plot the Q vs
frequency on a logarithmic scale. It is very easy to change the x-axis to a
logarithmic scale on the rectangular graph properties window in
Genesys. This allows us to have a visual comparison to the
manufacturers catalog plot.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

32

1.4.6 Magnetic Core Inductors

We have seen that the inductance of a length of wire can be increased by
forming the wire into a coil. We can make an even greater increase in the
inductance by replacing the air core with a magnetic material such as ferrite
or powdered iron. Two popular types of magnetic core inductors are the rod
core and toroidal core inductors shown in Figure 1-35.

Figure 1-35 Rod and toroidal magnetic core inductors

The magnetic field around an inductor is characterized by the magnetic

force H, and the magnetic flux B. They are related by the level of the
applied signal and the permeability, , of the core material. This
relationship is given by Equation (1-11).
B H

(1-11)

where:
B = Flux density in Gauss
H = Magnetization intensity in Oersteds
= Permeability in Webers/Ampere-turn
This relationship is nonlinear in that as H increases, the amount of flux
density will eventually level off or saturate. We will consider the linear
region of this relationship throughout the discussion of this text. In an iron
core inductor the permeability of the magnetic core is much higher than an
air core and produces a high flux density. This magnetic flux density for
each type of inductor is shown in Figure 1-36.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

33

RF and Microwave Concepts and Components

I
B

B
I

I
Rod Inductor

I
Toroidal Inductor

Figure 1-36 Magnetic flux densities for rod and toroidal magnetic core
inductors

The rod inductor has magnetic flux outside of the core as well as inside the
core. Rod inductors that are used in tuned circuits generally require a metal
shield around the inductor to contain this magnetic flux so that it does not
interfere or couple to adjacent circuits and other inductors. The toroidal
inductor flux remains primarily inside the core material. This suggests that
the toroidal inductor experiences less loss and should have higher Q factor.
This also gives the toroidal inductor a self shielding characteristic and does
not require a metallic enclosure. Because of its self-shielding properties and
high Q factor the toroidal inductor is one of the most popular of all
magnetic core inductors. However one advantage of the rod inductor is that
it is much easier to tune. The coil can be wound on a hollow plastic
cylindrical form in which the magnetic rod can be placed inside. The rod is
then free to move longitudinally which can tune the inductance value.
Core materials are characterized by their permeability. Core permeability
can vary quite a bit with frequency and temperature and can be confusing to
specify for a given application. The stability of the permeability can change
with the magnetic field due to DC current or RF drive through the inductor.
As the frequency increases the permeability eventually reduces to the same
value as air. Therefore iron core inductors are used only up to about 200
MHz. In general powdered iron can handle higher RF power without
saturation and permanent damage. Ferrite cores have much higher
permeability. The higher permeability of ferrite results in higher inductance
values but lower Q factors. This characteristic can be advantageous in the
design of RF chokes and broad band transformers. For inductors used in

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

34

RF and Microwave Concepts and Components

tuned circuits and filters, however, the higher Q factor of powdered iron is
preferred. The powdered iron cores are manufactured in a variety of mixes
to achieve different characteristics. The iron powders are made of hydrogen
reduced iron and have greater permeability and lower Q factor. These cores
are often used in RF chokes, electromagnetic interference (EMI) filters, and
switched mode power supplies. Carbonyl iron tends to have better
temperature stability and more constant permeability over a wide range of
power. At the same time the Carbonyl iron maintains very good Q factor
making them very popular in RF circuits. These characteristics lead to the
popularity of toroidal inductors of Carbonyl iron for the manufacture of RF
inductors. There is a wide variety of sizes and mixtures of Carbonyl iron
that are used in the design of toroidal inductors. A few of the popular sizes
that are manufactured by Micrometals Inc. are shown in Table 1-5.
Core
Designator

OD,
inches

ID,
inches

Height,
inches

T30

0.307

0.151

0.128

T37

0.375

0.205

0.128

T44

0.440

0.229

0.159

T50

0.500

0.303

0.190

T68

0.690

0.370

0.190

T80

0.795

0.495

0.250

T94

0.942

0.560

0.312

T106

1.060

0.570

0.437

T130

1.300

0.780

0.437

T157

1.570

0.950

0.570

T200

2.000

1.250

0.550

T300

3.040

1.930

0.500

T400

4.000

2.250

0.650

Height

Inner
Diameter,
ID

Outer
Diameter,
OD

Table 1-5 Partial listing of popular toroidal cores with designators

The inductance per turn of a toroidal inductor is directly related to its

permeability and the ratio of its cross section to flux path length as given by
Equation (1-12) [4].

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

35

RF and Microwave Concepts and Components

4 N2 A
length

nH

(1-12)

where:
LnH = inductance
= permeability
A = cross sectional area
length = flux path length
N = number of turns
As Equation (1-12) shows the inductance is proportional to the square of the
turns. A standard specification used by toroid manufacturers for the
calculation of inductance is the inductive index, AL. The inductive index is
typically given in units of nH/turn. The inductance can then be defined by
Equation (1-13).
(1-13)
L N 2 AL
nH
Some manufacturers specify the AL in terms of uH or mH. To convert
among the three quantities use the following guideline.
1 nH
10 uH
1 mH

turn 100 turns 1000 turns

(1-14)

The various powdered iron mixes are optimized for good Q factor and
temperature stability over certain frequency bands. A partial summary of
some popular mixtures is shown in Table 1-6. Powdered Iron cores have a
standard color code and material sub-type designator. The toroid is painted
with the appropriate color so that the mixture can be identified. A given A L
is dependent on both the size of the toroid and the material mix.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

36

RF and Microwave Concepts and Components

Material
Mix
Designator

Material
Permeability

Magnetic
Material

Color Code

Frequency Range

Temperature
Stability
(ppm/oC)

-17

4.0

Carbonyl

Blue/Yellow

20 200 MHz

50

-10

6.0

Carbonyl W

Black

10 100 MHz

150

-6

8.5

Carbonyl SF

Yellow

2.0 30 MHz

35

-7

9.0

Carbonyl TH

White

1.0 20 MHz

30

-2

10.0

Carbonyl E

Red

0.25 10 MHz

95

-1

20.0

Carbonyl C

Blue

0.15 2.0 MHz

280

-3

35.0

Carbonyl HP

Grey

0.02 1.0 MHz

370

Table 1-6 Partial listing of powdered iron core mixes and suggested
frequency range

Because of the complex properties of the core material, the determination of

the Q factor of a toroidal inductor can be difficult. It is not simply the
magnetic material properties alone, but also the wire winding loss as well
that determines the overall Q factor. These losses can then vary greatly with
frequency, flux density, and the toroid and wire size. The optimal Q factor
occurs when the winding losses are equal to the core losses [4]. In general
for a given inductance value and core mix a larger toroid will produce larger
Q factors. Conversely for a given toroid size higher Q factor is achieved at
higher frequency as the permeability decreases. The skin effect of the wire
in the windings can have a significant impact on the achievable Q factor.
Just as we have seen with the air core inductor, the inter-turn capacitance
will also have a limiting effect on the resulting Q factor as well as the self
resonant frequency. Manufacturers often provide a set of Q curves that the
designer can use as a design guide for determining the toroidal inductor Q
factor. Figure 1-37 shows a typical set of optimal Q curves for Carbonyl W
core material at various toroid sizes.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

37

RF and Microwave Concepts and Components

Figure 1-37 Optimal Q factor versus toroid size for core mix -10 (courtesy
Micrometals Inc.)

Example 1.4-4: Design a 550 nH inductor using the Carbonyl W core of

size T30. Determine the number of turns and model the inductor in
Genesys.
Solution: From the manufacturers data sheet the AL value is 2.5 for a T3010 toroidal core. Rearranging Equation (1-13) to solve for the number of
turns we find that 14.8 turns are required.
N

AL

550 nH
14.8
2.5

To reduce the winding loss we want to use the largest diameter of wire that
will result in a single layer winding around the toroid. Equation (1-15) will
give us the wire diameter.
d

ID
N

where:
d = Diameter of the
wire in inches
ID = Inner diameter of the core in inches (from Table 1-5)
N = Number of turns

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

(1-15)

38

RF and Microwave Concepts and Components

therefore,
d

(0.151)
0.4744

0.0238 inches
14.8
19.942

From Appendix A, AWG#23 wire is the largest diameter wire that can be
used to wind
a single layer around the T30 toroid. Normally AWG#24 is
chosen because this is a more readily available standard wire size. The
toroidal inductor model in Genesys requires a few more pieces of
information. The Genesys model requires that we enter the total winding
resistance, core Q factor, and the frequency for the Q factor, Fq. As an
approximation, set Fq to about six times the frequency of operation. In this
case set Fq to (6). 25 MHz = 150 MHz. Then tune the value of Q to get the
best curve fit to the manufacturers Q curve. We know that we have 14.8
turns on the toroid but we need to calculate the length of wire that these
turns represent. The approximate wire length around one turn of the toroid
is calculated from the following equation.
Length 2 Height OD ID #turns

(1-16)

Using the dimensions for the T30 toroid from Table 1-6 we can calculate
the total length of the wire as 6.10 inches.

20.128 0.307 0.15114.8 6.10

inches

Then use the techniques covered in section 1.2.3 to calculate the resistance
of the 6.10 inches of wire taking into account the skin effect. To get a better
estimate of the actual inductor Q, use the Fq frequency rather than the
operating frequency for the skin effect calculation. The resistance at 150
MHz for the AWG#24 wire is calculated as 0.30 . Use an Equation Editor
to calculate the Q and inductance. Figure 1-38 shows the schematic of the
toroidal inductor and the simulated Q.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

39

Figure 1-38 Genesys toroidal inductor model and simulated Q factor

The Qc parameter of the model has been tuned to give a reasonably good fit
with the manufacturers (T30) curve of Figure 1-37. Figure 1-39 shows that
the simulated self resonant frequency of the inductor model is near 150
MHz.

Figure 1-39 Impedance of toroidal inductor model

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

40

Figure 1-40 shows the inductance that was calculated from the impedance.
The inductance is exactly 550 nH at 10 MHz and begins to increase slightly
to 563 nH at the design frequency of 25 MHz. Given the myriad of
variables associated with the toroidal inductor, this simple model gives a
good first order model of the actual inductor and will enable accurate
simulation of filter or resonator circuits.

Figure 1-40 Inductance value of the toroidal inductor model

1.5 Physical Capacitors

The capacitor is an electrical energy storage component. The amount of
energy that can be stored is dependent on the type and thickness of the
dielectric material and the area of the electrodes or plates. Capacitors take
on many physical forms throughout electrical circuit designs. These range
from leaded bypass capacitors in low frequency applications to monolithic
forms in millimeter wave applications. Table 1-7 summarizes many of the
applications in which capacitors are found. The table also shows some of
the types of materials in which the capacitors are manufactured.
Application

Dielectric Type

Notes

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

41

RF and Microwave Concepts and Components

Audio Frequency
Coupling
Power Supply Filtering
RF Coupling

Tuned Circuits,
Resonators

Aluminum Electrolytic
Tantalum
Polyester/ Polycarbonate
Aluminum Electrolytic
Ceramic NPO (COG)
Ceramic X7R
Polystyrene
Silver Mica
Ceramic NPO (COG)

Very High Capacitance

High Capacitance for given size
Medium capacitance, low cost
High Capacitance, high ripple current
Small, low loss, low cost
Small, low cost, higher loss than COG
Very low loss in RF range, larger than
ceramic
Low loss, low tolerance for RF applications
Low loss, low tolerance, not as good as
silver mica.

Table 1-7 Applications for various types of capacitors

As Table 1-7 shows, the ceramic capacitors dominate the higher RF and
microwave frequency applications. Two of the most popular of the ceramic
capacitors are the single layer and multilayer ceramic capacitors, as shown
in Figure 1-41. These capacitors are available with metalized terminations
so that they are compatible with a variety of surface mount assembly
techniques from hand soldering to wire bonding and epoxy attachment.
L
T

L
W

Single Layer Capacitor

Multi Layer Chip Capacitor

Figure 1-41 Single layer and multi layer chip capacitor dimensions

1.5.1 Single Layer Capacitor

The single layer capacitor is one of the simplest and most versatile of the
surface mount capacitors. It is formed with two plates that are separated by
a single dielectric layer as shown in Figure 1-42. Most of the electric field
(E) is contained within the dielectric however there is a fraction of the E
field that exists outside of the plates. This is known as the fringing field.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

42

RF and Microwave Concepts and Components

A
r

Figure 1-42 Single layer parallel plate capacitor

The capacitance formed by a dielectric material between two parallel plate

conductors is given by Equation 1-17 [1].

KA r
C N 1
FF
t

pF

(1-17)

where, A = plate area

r = relative dielectric constant
t = separation
K = unit conversion factor; 0.885 for cm and 0.225 for inches
FF = fringing factor; 1.2 when mounted on microstrip
N = number of parallel plates.
Example 1.5-1: Consider the design of a single layer capacitor from a
dielectric that is 0.010 inches thick and has a dielectric constant of three.
Each plate is cut to 0.040 inches square.
Solution: When the capacitor is mounted with at least one plate on a large
printed circuit board track, a value of 1.2 is typically used in calculation.
The Equation Editor in Genesys can be used to solve Equations (1-17) as
shown in Figure 1-43.

Figure 1-43 Single layer capacitance calculation

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

43

The single layer capacitor can be modeled in Genesys using the Thin Film
Capacitor.
.2250.04 0.043
C 2 1
1.2 0.13 pF
0.010

The ceramic dielectrics used in capacitors are divided into two major
classifications. Class 1 dielectrics have the most stable characteristics in
terms of temperature stability. Class 2 dielectrics use higher dielectric
constants which result in higher capacitance values but have greater
variation over temperature. The temperature coefficient is specified in either
percentage of nominal value or parts per million per degree Celsius
(ppm/oC). Ceramic materials with a high dielectric constant tend to
dominate RF applications with a few exceptions. NPO (negative-positivezero) is a popular ceramic that has extremely good stability of the nominal
capacitance versus temperature.
Dielectric Material
Vacuum
Air
Mylar
Paper
Mica
Glass
Alumina
Ceramic (low r)
Ceramic (high r)

Dielectric Constant
1.0
1.004
3
4-6
4-8
3.7 - 19
9.9
10
100 10,000

Table 1-8 Dielectric constants of materials

1.5.2 Multilayer Capacitors

Multilayer capacitors, shown in Figure 1-44, are very popular in surface
mount designs. They are physically larger than single layer capacitors and
can be attached by hand or by automatic pick-and-place machines. As
Figure 1-44 shows, the multilayer chip capacitor is a parallel array of
capacitor plates in a single package. Due to this type of construction the
chip capacitor can handle higher voltages than the single layer capacitor.
The insulation resistance of the capacitor is its ability to oppose the flow of

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

44

RF and Microwave Concepts and Components

electricity and is a function of the dielectric material and voltage. The

insulation resistance is typically specified as a minimum resistance value in
M at a specified working voltage. The working voltage rating, WVDC, is
the maximum DC voltage at which the capacitor can operate over the
lifetime of the capacitor. The AC voltage rating is approximately one half of
the WVDC value. The dielectric withstand voltage (DWV) is the electrical
strength of the dielectric at 2.5 times the rated voltage. This is a maximum
short term over-voltage rating and is usually specified as a length of time
that the dielectric can withstand the 2.5 times the WVDC value without
arcing through [6].

Figure 1-44 Multilayer chip capacitor construction

There is no physical model for the multilayer chip capacitor in Genesys.

The designer must rely on S parameter files or Modelithic models as we
have used for the chip resistor in section 1.3.1. Two of the major
manufacturers of multilayer chip capacitors are American Technical
Ceramics, ATC, and Dielectric Laboratories Inc., DLI. These manufacturers
provide S parameter files for their capacitors that are readily available on
the company websites. They also provide helpful software applications that
can aid the designer in making decisions for the proper selection of chip
capacitors in specific applications.

1.5.3 Capacitor Q Factor

RF losses in the dielectric material of a capacitor are characterized by the
dissipation factor. The dissipation factor is also referred to as the loss
tangent and is the ratio of energy dissipated to the energy stored over a
period of time. It is essentially the capacitors efficiency rating. The

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

45

dissipation factor and other ohmic losses lead to a parameter known as the
Equivalent Series Resistance, ESR. The dissipation factor is the reciprocal
of the Q factor. Just as we have seen with resistors and inductors, the
physical model of a capacitor is a network of R, L, and C components.
Example 1.5-2: Calculate the Q factor versus frequency for the physical
model of an 8.2 pF multilayer chip capacitor shown in Fig. 1-45.

Physical Model

Capacitive
SRF

Inductive

Figure 1-45 Physical model of the 8.2 pF chip capacitor and impedance

The capacitor has a series inductance and resistance component along with
a resistance in parallel with the capacitance. The parallel resistor sets the
losses in the dielectric material. The series resistance and inductance
represent any residual lead inductance and ohmic resistances.
Solution: The values entered for the physical model and the Q factor can be
obtained from the capacitor manufacturer. The plot of Figure 1-45 shows
the impedance of the capacitor versus frequency. Note that the impedance
decreases as would be expected until the self resonant frequency is reached.
Above the self resonant frequency the impedance begins to increase
suggesting that the capacitor is behaving as an inductor. The self resonant
frequency is due to the series inductance resonating with the capacitance. At

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

46

resonance the reactance cancels leaving only the resistances R1 and R2. The
parallel resistance, R2, can be converted to an equivalent series resistance
by Equation (1-18).
R 2

1
R2
1 Q2

(1-18)

These two series resistances can then be added to find the equivalent series
resistance, ESR, as defined by Equation (1-19).
ESR R1 R2

(1-19)

The capacitor Q factor is then calculated by Equation (1-20). XT is the total

series reactance of the inductive and capacitive reactance.
Q

XT
ESR

(1-20)

As Figure 1-45 shows, the 8.2 pF multilayer chip capacitor has a series
resonant frequency of 3313 MHz. A marker is placed on the trace indicating
the impedance at the SRF as 0.149 . Because the reactance is cancelled at
the SRF, this impedance essentially becomes the ESR of the capacitor. R2
of Figure 1-45 is extremely frequency dependent. This means that the
capacitors Q factor is also extremely frequency dependent. An improved
model for analyzing the capacitors characteristics over a wide frequency
range is to use the Genesys model for capacitor with Q. Using this model
the Q factor of the capacitor can be made proportional to the square root of
the applied frequency. The capacitor Q factor can be calculated from the
impedance using the Equation Editor as shown in Figure 1-46. The resistive
and inductive components are used to define the Q factor.

Figure 1-46 Equation Editor used to calculate Q factor and capacitance

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

47

Figure 1-47 shows the large dependence on the capacitor Q with frequency.
At 1006 MHz the Q factor is 120 while at 100 MHz the Q factor is greater
than 1000. The Q factor goes to zero at the self resonant frequency. Above
the self resonant frequency the Q is undefined.

Figure 1-47 Calculated Q factor of 8.2 pF chip capacitor and equivalent

model

The Equation Editor of Figure 1-46 also calculates the effective capacitance
from the total reactance of the model using Equation (1-21).
C

1
2 F X T

(1-21)

The plot of Figure 1-48 reveals some interesting characteristics about the
chip capacitor. From 100 MHz to 300 MHz the capacitance value is fairly
constant at 8.217 pF. As the frequency increases above 300 MHz we see
that the capacitance actually increases. The parasitic inductive reactance of
the capacitor package actually makes the effective capacitance greater than
its nominal value.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

48

Figure 1-48 Effective capacitance of the 8.2 pF chip capacitor

This is a property of the capacitor that is not always intuitive. As the

frequency approaches the self resonant frequency the capacitance rapidly
approaches infinity. The capacitor actually becomes nearly a short circuit to
RF at the self resonant frequency. This is an important property of the
capacitor that is used frequently in RF and microwave design. In RF
coupling or bypass capacitor applications, capacitors are very often used at
or near their self resonant frequency. The capacitors self resonant
frequency is due to the series resonant circuit. In applications requiring
bypassing over a wide range of frequencies it is often necessary to use
several capacitors each selected to have a uniquely spaced self resonant
frequency. In filter and other tuned circuit applications where we want the
chip to appear as an 8.2 pF capacitor, we clearly must stay well below the
self resonant frequency of the capacitor. A typical rule-of-thumb is to use
the capacitor over a frequency range up to 35% of the self resonant
frequency. Therefore the 8.2 pF chip capacitor with a self resonant
frequency of 3313 MHz would be used as a capacitor in tuned circuits up to
a frequency of about 1160 MHz. Figure 1-48 shows that at frequencies
above 1160 MHz the capacitance is extremely nonlinear. However the
capacitor could be used as a DC blocking, RF coupling capacitor that would
efficiently pass microwave energy near its SRF of 3300 MHz.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

49

RF and Microwave Concepts and Components

References and Further Reading

[1]

Besser Associates Applied RF Techniques Course, 480 San Antonio

Road, Mountain View, CA. 94040

[2]

Paul Lorrain, Dale P. Corson, and Francois Lorrain,

Electromagnetic Fields and Waves, W.H. Freeman and Company,
New York, 1988

[3]

Design Guide, Microwave Components Inc., P.O. Box 4132, South

Chelmsford, MA 01824

[4]

Iron Powder Cores for High Q Inductors, Micrometals, Inc.

[5]

Agilent Technologies, Genesys 2010.05 Documentation Set, EEsof

EDA Division, Santa Rosa, California [Link]

[6]

Capacitors for RF Applications, Dielectric Laboratories, Inc., 2777

Rt.20 East, Cazenovia, NY. 13035

[7]

R. Ludwig and P. Bretchko, RF Circuit Design -Theory and

Applications, Prentice Hall, New Jersey, 2000

[8]

William
Sinnema
and
Robert
McPherson,
Communication, Prentice Hall Canada, 1991

[9]

M.F. Doug DeMaw, Ferromagnetic Core Design & Application

Handbook, MFJ Publishing Co., Inc. Starkville, MS. 39759, 1996

[10]

The RF Capacitor Handbook, American Technical Ceramics, One

Norden Lane, Huntington, New York 11746

Electronic

Problems
1-1.

Calculate the wavelength of an electromagnetic wave operating at a

frequency of 428MHz.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

50

1-2.

Calculate the inductance of a 5 inch length of AWG #30 straight

copper wire.

1-3.

Using the Equation Editor, calculate the reactance of the wire from
Problem 2 at 10 Hz, 10 MHz, and 10 GHz. Create a Linear analysis
in Genesys and display the wire impedance vs. frequency.

1-4.

Calculate the resistance of a 12 inch length of AWG #24 copper

wire at DC and at 25MHz

1-5.

Find the skin depth and the resistance of a 2 meter length of copper
coaxial line at 2 GHz. The inner conductor radius is 1 mm and the
outer conductor is 4 mm.

1-6.

Calculate the inductance of a 5 inch length of copper flat ribbon

conductor. The dimensions of the ribbon are 0.100 inches in width
and 0.002 inches thick.

1-7.

Model a chip resistor (size 0603) with a resistance of 50 in

Genesys. Consider an application in which 50 impedance must be
maintained with +10%. Create a Linear Analysis and determine the
maximum usable frequency of the chip resistor.

1-8.

Design an air core inductor with an inductance value of 84nH. Use a

copper wire of 0.050 inch diameter wound on a core diameter or
0.100 inch. Determine the number of turns required assuming a tight
spaced winding.

1-9.

Using the inductor from Problem 8 and the techniques developed in

Section 1.4.1 examine the change in the coil inductance as the turn
spacing is increased from zero to 0.10 inch in 0.002 inch increments.

1-10. Using the inductor from Problem 8, determine the self resonant
frequency of the inductor and comment on the maximum frequency
in which the inductor may be used in a tuned circuit application.

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

RF and Microwave Concepts and Components

51

1-11. Using the inductor from Problem 8, determine the maximum Q

factor of the inductor and the frequency at which the maximum Q
factor is obtained.
1-12. In Genesys, select a chip inductor from the CoilCraft library with an
inductance value of 80nH. Determine the maximum Q factor of the
inductor and comment on the maximum usable frequency of the
inductor in a filter application.
1-13. Design a 1mH toroidal inductor on a Carbonyl W core size T30.
Determine the maximum wire size that could be used to realize a
single layer winding.
1-14. Using the inductor from Problem 13, model the inductor in Genesys
and determine the approximate self resonant frequency. Comment
on the maximum usable frequency of the inductor in the front end of
a radio receiver.
1-15. A 0.05pF capacitor is required to couple a transistor to the resonator
of a microwave oscillator. Design a single layer capacitor using a
0.020 inch thick dielectric with er=2.2. Determine the dimensions of
the capacitor assuming square footprint is desired.
1-16. For the single layer capacitor of Problem 15, determine the
dimensions of the capacitor with a dielectric constant er=10.2.
1-17. In Genesys model a 47pF chip capacitor using the ATC 0603 model
from the Modelithics library. Determine the Q factor of the capacitor
at a frequency of 1000MHz.
1-18. Determine the self resonant frequency of the capacitor in Problem
17 and comment on the maximum frequency that this capacitor
could be used in a tuned circuit. At what frequency does the
capacitor have the lowest amount of energy loss?

Copyright 2011 by Ali A. Behagi and Stephen D. Turner

Copyright 2011 by Ali A. Behagi and Stephen D. Turner