RADIATED EMISSIONS OF ELECTRICALLY LONG PCB BASED ON...
Transcript of RADIATED EMISSIONS OF ELECTRICALLY LONG PCB BASED ON...
RADIATED EMISSIONS OF ELECTRICALLY
LONG PCB BASED ON IMBALANCE DIFFERENCE
AND DIPOLE ANTENNA MODELS
AHMED MOHAMMED YAHYA SAYEGH
UNIVERSITI TUN HUSSEIN ONN MALAYSIA
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
RADIATED EMISSIONS OF ELECTRICALLY LONG PCB BASED ON
IMBALANCE DIFFERENCE AND DIPOLE ANTENNA MODELS
AHMED MOHAMMED YAHYA SAYEGH
A thesis submitted in
fulfillment of the requirement for the award of the
Doctor of Philosophy
Faculty of Electrical and Electronic Engineering
Universiti Tun Hussein Onn Malaysia
JANUARY 2017
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
iii
To the memory of my father, who would have been glad to see me at this moment
To my beloved mother for her constant, unconditional love during all my life.
To my wife and beloved daughter, Malak, for their love and support.
To my brothers and my sisters for their support and encouragement
To all my family members and friends for their love and support
To science,
enlightening us.
DEDICATION
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
iv
ACKNOWLEDGEMENT
Alhamdulillah, I am so grateful to Allah for giving me enough strength,
inspiration and guidance throughout my Ph.D. study. Many people have redounded
directly or indirectly to the completion of this thesis and their assistances are highly
appreciated.
First and foremost, I would like to express my deepest gratitude to my
supervisor, Professor Dr. Mohd Zarar Mohd Jenu, for his invaluable guidance and
assistance during my Ph.D. journey. Without his patience and motivation, this thesis
would not have been completed successfully. He gave me the opportunity to start
with him a new constructive experience of research work. I have learned from him
many aspects not only in the academic life but also in my living attitude. I am also
thankful to him for spending many hours for reading and commenting to review my
research publication including this thesis.
Thanks to Prof. Dr. Frank Leferink from university of Twente and Prof. Dr.
Todd Hubing from Clemson University for their valuable technical advices. I am also
grateful to all my colleagues from the EMC centre for their generous assistance in
my research-related problems, namely Mr. Encik Mohd Rostam Bin Anuar, Madam
Miskiah bt. Muhamad Ihsan and Mr. Sharifunazri Johadi.
I would like to acknowledge Universiti Tun Hussein Onn Malaysia (UTHM)
for giving me the opportunity to undertake my doctorate program by bestowing upon
me university grant scholarship.
Finally, I would like to extend my deepest gratitude to my mother for her
never ending love, my wife for her kind support and encouragement. I also dedicate
this Ph.D. thesis to my lovely daughter, Malak Ahmed, who always enjoys my time.
At last, I want to thank all my family members and friends who supported me during
my Ph.D. journey.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
v
ABSTRACT
Radiated emission (RE) compliance is one of the most important requirements for
legal and global marketing of electronic devices. Conventionally, REs are quantified
using full wave numerical simulations or measurement in semi-anechoic chambers
(SAC). However, these methods are expensive and time-consuming. Thus, they are
not suitable for early prediction of RE. This thesis proposes novel analytical methods
for estimating RE of printed circuit board (PCB) attached with cables at the design
stage for time and cost saving. First, a novel analytical method has been introduced
to estimate the RE from electrically long PCB traces based on transmission-line (TL)
theory, imbalance difference model (IDM), travelling wave antenna (TWA) and
dipole antenna model. However, this proposed method is limited to PCB-traces with
Quasi-TEM (QTEM) current distribution. Therefore, artificial neural network (ANN)
model has been constructed and trained using backpropagation algorithm for
estimating the RE from electrically long traces beyond Quasi-TEM operation mode.
Secondly, a novel analytical method was developed to estimate the common mode
RE from cables attached to a PCB-trace by adopting the IDM in conjunction with
asymmetrical dipole antenna model. The effectiveness of the proposed methods was
verified by comparing the predicted results to both 3D HFSS simulation and
measurement results. The results obtained using the proposed methods were in a
good agreements with both HFSS simulation results and measurement results with
accuracy of ±3dB which is acceptable range from design point of view. Finally,
graphical user interface (GUI) was developed which would be useful for early
prediction of RE in the electronics industry. These GUIs can be enhanced in future to
characterize automatically the RE from the entire PCB.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
vi
ABSTRAK
Pancaran radiasi (RE) merupakan satu daripada keperluan penting dalam pemasaran
sah dan global peranti elektronik. Kebiasaannya, RE diukur menggunakan simulasi
berangka gelombang penuh atau pengukuran di dalam kebuk separa gema (SAC).
Walau bagaimanapun, kaedah-kaedah tersebut adalah mahal dan memakan masa.
Oleh itu, kaedah-kaedah tersebut tidak sesuai untuk perkiraan awal RE. Tesis ini
mencadangkan kaedah analisis untuk menganggarkan RE papan litar bercetak yang
dipasangkan kabel pada peringkat reka bentuk untuk menjimatkan masa dan kos.
Pertama, satu kaedah analisis baharu telah diperkenalkan untuk mengganggarkan RE
daripada laluan elektrik PCB yang panjang berdasarkan kepada teori talian
penghantaran (TL), model perbezaan ketakseimbangan (IDM), antena gelombang
kembara (TWA) dan model antena dwikutub. Walaubagaimanapun, kaedah yang
dicadangkan terhad kepada laluan elektrik PCB dengan pengagihan arus Quansi-
TEM (QTEM). Oleh yang demikian, model rangkaian neural buatan telah dibina dan
dilatih menggunakan algoritma rambatan belakang untuk menggangar RE daripada
laluan elektrik panjang melampaui mod operasi Quansi-TEM. Kedua, satu kaedah
analisis baharu telah dibangunkan untuk menganggarkan mod biasa RE daripada
kabel yang dipasang kepada laluan elektrik PCB dengan menggunakan IDM beserta
dengan model antenna dwikutub tak simetri. Keberkesanan kaedah yang
dicadangkan disahkan melalui perbandingan hasil anggaran dari kedua-dua hasil
ukuran dan simulasi 3D HFSS. Keputusan yang diperolehi menggunakan kaedah
yang dicadangkan adalah memuaskan dengan kedua-dua hasil ukuran dan simulasi
memperolehi keputusan ketepatan sekitar ±3dB iaitu julat yang diterima untuk
sesuatu rekabentuk. Akhirnya, antara muka pengguna bergrafik (GUI) dibangunkan,
yang bakal berguna untuk anggaran awal RE dalam industri elektronik. GUI ini
boleh ditambah baik pada masa hadapan untuk menentukan ciri RE secara automatik
daripada keseluruhan PCB.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
vii
CONTENTS
DECLARATION ii
DEDICATION iii
ACKNOWLEDGEMENT iv
ABSTRACT v
ABSTRAK vi
CONTENTS vii
LIST OF PUBLICATIONS xii
LIST OF TABLES xiv
LIST OF FIGURES xv
LIST OF SYMBOLS AND ABBREVIATIONS xxii
LIST OF APPENDICES xxiv
CHAPTER 1 INTRODUCTION 1
Research background 1
Overview of EMC 2
1.2.1 EMC standards 3
Characterization methods of radiated emissions 5
Problem statement 6
Objectives of the research 9
Scope of the research 9
Significance of the research 10
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
viii
Organisation of the thesis 10
CHAPTER 2 LITERATURE REVIEW 12
Fundamentals of the radiating systems 12
2.1.1 Infinitesimal dipole 12
2.1.2 Short dipoles 16
2.1.3 Long dipoles 17
Measurement and modelling of EM emissions of PCB 17
2.2.1 Measurement of EM emissions of PCBs 18
2.2.2 Modelling methods for EM emissions of PCB 22
Sources of EM emissions on PCBs 26
2.3.1 Radiated emissions of PCB-traces 27
2.3.2 Radiated emissions of PCB-attached cables 32
Introduction to artificial neural networks 37
2.4.1 Description of multi-layer perceptron network for
EMC Applications 39
Summary of the literature review 40
Research gap 44
Chapter summary 47
CHAPTER 3 RESEARCH METHODOLOGY 48
Overview 48
Estimation of RE from electrically long PCB-traces 50
3.2.1 Analytical model for estimating DM RE of
electrically-long traces 50
3.2.2 Analytical model for estimating CM RE of
electrically long traces 52
3.2.3 Estimation of RE from PCB traces using ANN 53
Estimation of RE from cables attached to PCB-traces 55
Verification process of the proposed models 56
3.4.1 Verification process using HFSS 57
Chapter summary 62
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
ix
CHAPTER 4 ANALYTICAL METHODS FOR RADIATED EMISSIONS
OF PCB-TRACES 63
Introduction 63
Novel solutions for estimating RE of electrically long
PCB-traces 65
4.2.1 Analytical modelling for DM-RE of electrically
long PCB-traces 65
4.2.2 Analytical modelling for CM RE of electrically
long traces 78
4.2.3 ANN-MLP model for predicting RE from PCB-
traces 84
Chapter summary 98
CHAPTER 5 ANALYTICAL METHODS FOR RADIATED EMISSIONS
OF PCB-ATTACHED CABLES 99
Impact of attaching cables to PCB-ground plane on
CM RE 99
Analytical method for CM RE of cables attached to
PCB-ground plane 109
5.2.1 Identification and quantification of CM voltage
on board-cable structure 110
5.2.2 Estimation of CM RE from PCB with one
attached cable 114
5.2.3 Novel model for CM RE from two cables
attached to PCB 119
5.2.4 Flow chart for maximum CM RE of cables
attached to PCB 122
Analytical method for CM RE from actual cables
attached to signal trace and ground plane of PCB 123
5.3.1 Calculation of CM RE using IDM and asymmetrical
dipole antenna 123
5.3.2 Application of the proposed model on practical
cables 126
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
x
Chapter summary 128
CHAPTER 6 RESULTS AND DISCUSSIONS 129
Validation results of the proposed models for RE from
PCB-traces 129
6.1.1 Results of the analytical modelling for RE based
on dipole antenna 131
6.1.2 Results of the analytical solution for DM RE using
TWA antenna 146
6.1.3 Results of the proposed ANN-MLP model for RE
of PCB-traces 155
Validation results of the proposed models for RE from
PCB-attached cables 160
6.2.1 Results of the proposed model for RE from cables
attached to PCB-ground plane 160
Results of the proposed model for CM RE from one cable
attached to PCB at load junction 161
Results of the proposed model for CM RE from two cables
attached to PCB at source/load junctions 169
6.4.1 Results of the proposed model for RE from real cables
attached to PCB- trace and PCB-ground plane 174
GUI of the proposed models 178
Summary 182
CHAPTER 7 CONCLUSIONS AND FUTURE WORKS 183
Conclusion 183
Recommendation for future works 184
REFERENCES 186
APPENDIX A- MEASUREMENT EQUIPMENT 200
APPENDIX B- HORN ANTENNA DATASHEET –
CALIBRATION CERTIFICATE 202
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xi
APPENDIX C- DIPOLE ANTENNA DATASHEET –
CALIBRATION CERTIFICATE 203
APPENDIX D- UNCERTAINTY EVALUATION FOR
RADIATED EMISSION (EMC MEASUREMENT) 204
APPENDIX E- MATLAB M-FILE OF THE MAIN GUI OF
THE SOFTWARE TOOL 205
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xii
LIST OF PUBLICATIONS
Journal Papers
[1] Ahmed Mohammed Sayegh, Mohd Zarar Mohd Jenu and Samsul Haimi Bin
Dahlan, “Analytical Solution for Maximum Differential-Mode Radiated
Emissions of Microstrip Trace”, IEEE Transaction on Electromagnetic
Compatibility, Vol. 58(5), OCT, 2016 (ISI indexed, Q1 with IF=1.3).
[2] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu, “Prediction of
common mode radiation from cables attached to PCB using imbalance
difference and asymmetrical dipole antenna models”, IET Electronic Letters,
Vol. 52, No 8, April, 2016 (ISI indexed, Q2 with IF=0.93).
[3] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu "Prediction of
radiated emissions from high-speed printed circuit board traces using dipole
antenna and imbalance difference model", IET Science, Measurement &
Technology Journal, Vol. 10, No 1, Jan.,2016 (ISI indexed, Q3 with
IF=0.954).
[4] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu "Estimation of
common mode radiated emissions from cables attached to high speed PCB
using imbalance difference model", ARPN Journal of Engineering and
Applied Sciences, Vol. 10, No 19, Oct., 2015 (Scopus indexed journal).
Conferences Papers
[1] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu, “Estimation of
Radiated Emissions from Microstrip PCB using Neural Network Model”,
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xiii
2016 IEEE Asia-Pacific Conference on Applied Electromagnetics (APACE),
Langkawi, Kedah, Malaysia 11 - 13 Dec, 2016 .
[2] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu, “Estimation of EM
Emissions From Coaxial and Twin-Lead Cables Attached to A PCB Using
Imbalance Difference and Asymmetrical Dipole Models”, 2016 IEEE Asia-
Pacific Conference on Applied Electromagnetics (APACE), Langkawi,
Kedah, Malaysia 11 - 13 Dec, 2016 (Invited Paper).
[3] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu "Estimation of
common mode radiated emissions from cables attached to high speed PCB
using imbalance difference model", International Conference on Electrical
and Electronic Engineering (IC3E), Melaka, Malaysia, August 2015.
[4] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu, “Closed-form
expressions for estimating maximum radiated emissions from the traces of a
Printed Circuit Board”, 2015 Asia-Pacific Symposium on Electromagnetic
Compatibility (APEMC), Taipei, Taiwan,26 – 29 May 2015.
[5] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu, “Prediction of
radiated emissions from high speed PCB traces using travelling wave antenna
model”, 2014 IEEE Asia-Pacific Conference on Applied Electromagnetics
(APACE), Johor, 8 - 10 Dec, 2014.
[6] Ahmed Mohammed Sayegh , Mohd Zarar Mohd Jenu and Syarfa Zahirah
Sapuan “Neural network based model for radiated emissions prediction from
high speed PCB traces”, International Conference on Computer,
Communications, and Control Technology (I4CT), Langkawi, 2 - 4 Sep,2014
[7] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu, “High Power
Electromagnetics (HPEM) as Re-emerging Sources of NIR, Non Ionization
Radiation conference, Kuala Lumpur, Malaysia. 29-30 October, 2013.
[8] Ahmed Mohammed Sayegh and Mohd Zarar Mohd Jenu “Evaluation of the
Radiated Emission of a Printed Circuit Board Attached with Cables”, 3rd
International Conference on Electric and Electronics (EEIC 2013), Hong
Kong, Atlantis Press, pp 195-198, Nov., 2013.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xiv
LIST OF TABLES
2.1 Basic characteristics of the modelling methods of PCB-RE 26
2.2 Summary of the previous works for computing RE of PCB-traces 41
2.3 Summary of related works for computing RE from PCB-attached cables 43
2.4 Summary of literature review for RE from PCB using ANN 44
2.5 Research gap of this thesis for computing the RE from PCB-traces 45
2.6 Research gap for computing the RE from PCB-attached cables 46
2.7 Research gap for quantifying the RE from PCB using ANN 47
4.1 Samples of the PCB-parameters under study 86
4.2 Sample input data set for the proposed MLP model 94
4.3 RMSE of the proposed ANN with different hidden nodes 96
6.1 Accuracy and errors of the proposed model 143
6.2 Measurement and estimation of θmax in configuration A 153
6.3 Various configurations of PCBs to verify the proposed MLP using
HFSS 158
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xv
LIST OF FIGURES
1.1 EMC aspects and their interrelationship 2
1.2 RE limits for Class A and Class B digital device in the FCC standards 4
1.3 RE limits for Class A ITE equipment in CISPR 22 4
1.4 Radiated emissions of PCB attached with cables 6
1.5 Cost of EMC implementation on the electronic product 7
1.6 Basic idea of the proposed models 8
2.1 A z-directed electric point dipole 13
2.2 Definition of near-field and far-field regions 14
2.3 Wave impedance of electric and magnetic sources 15
2.4 A z-directed short dipole 16
2.5 Measurement set-up of EMC-RE using TEM cell 19
2.6 Measurement set-up of RE using NF scanning method 19
2.7 Measurement of RE using radiation pattern method 20
2.8 Measurement of RE using reverberation chamber method 21
2.9 Electromagnetic field components of one cell using FDTD 23
2.10 Sources of EM emissions on PCB 27
2.11 DM-RE of PCB-traces 28
2.12 Maximum electric field due to DM currents on PCB-traces 29
2.13 Maximum electric field due to CM currents 31
2.14 Voltage-driven coupling from the PCB-trace to the attached cable 34
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xvi
2.15 (a) 3D structure describing current driven mechanism (b) cross section
of PCB using current driven mechanism 36
2.16 (a) Structure of PCB with two attached cables (b) CM equivalent
structure using imbalance difference concept 37
2.17 3-layer MLP for EMC applications 40
3.1 Research methodology of the proposed models 49
3.2 Flow chart of the proposed model for DM RE of PCB-traces 51
3.3 Flow chart of the proposed model for CM RE from PCB-traces 52
3.4 Estimation of RE from PCB-traces using MLP 53
3.5 Flowchart of the proposed ANN model for RE from PCB-traces 54
3.6 Flowchart for estimating CM RE from cables attached to PCB 55
3.7 Flow chart for verification process of the proposed models 56
3.8 Process of HFSS configuration 58
3.9 The main interface of HFSS 58
3.10 Configuration of solution type in HFSS 59
3.11 Boundary assignment in HFSS 59
3.12 Excitation port assignment in HFSS 60
3.13 Configuration of frequency sweep in HFSS 61
3.14 Reporting results in HFSS 61
3.15 Configuration of far-field radiation setup in HFSS 61
4.1 Two PCBs under test for study the impact of trace length on RE 64
4.2 Impact of short and long trace on RE of PCB 64
4.3 (a) Simple single-sided PCB for illustrating trace segmentation method
(b) equivalent circuit of the PCB under study 66
4.4 Segmentation of long trace into small segments for RE calculations 67
4.5 Flow chart of trace segmentation method 69
4.6 Calculation of the far-fields of two-parallel long traces 71
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xvii
4.7 Simple microstrip PCB for maximum DM-RE formulation (b) PCB-
trace above ground plane equivalent to TWA 74
4.8 Simple microstrip PCB structure (b) equivalent CM model based on
asymmetrical dipole antenna 80
4.9 Equivalent asymmetrical dipole model for CM RE 82
4.10 Equivalent symmetrical dipole antenna 83
4.11 Structure of microstrip PCB used for data collection 85
4.12 Impact of board length on the maximum RE 86
4.13 Impact of board width on the maximum RE 87
4.14 Effect of trace length on the maximum RE 88
4.15 Effect of trace width on the maximum RE 89
4.16 Impact of source impedance on the maximum RE 90
4.17 Impact of load impedance on the maximum RE 90
4.18 Impact of signal voltage on the maximum RE 91
4.19 Impact of dielectric permittivity on the maximum RE 92
4.20 Impact of dielectric thickness on the maximum RE 93
4.21 Topology of the proposed ANN model. 95
4.22 Training process of the MLP model with 15 hidden nodes 97
4.23 Training, validation and testing of the MLP model 97
5.1 PCB board under test in SAC 100
5.2 (a) Structure of PCB-circuit with one attached wire (b) equivalent
circuit of PCB under study attached with one wire 101
5.3 (a) Structure of PCB-circuit with two attached wires (b) equivalent
circuit of PCB under study attached with two wires 102
5.4 RE of PCB under test without attached wires 103
5.5 Total emissions of PCB with one attached wire 103
5.6 Total emissions of PCB with two attached wires 104
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xviii
5.7 Total RE of PCB loaded with 50 ohm 105
5.8 Microstrip PCB under study with 21cm x 10cm geometry 106
5.9 PCB on wooden table inside SAC 106
5.10 PCB with 0.5 m attached cable in SAC 107
5.11 PCB on table attached with two cables in SAC 107
5.12 RE of PCB with/without attached cables 108
5.13 RE of PCB with one cable of various lengths 109
5.14 Simple microstrip PCB attached with two cables 110
5.15 Exaggerated structure of PCB with cables attached to ground plane 111
5.16 The equivalent CM structure based on IDM 112
5.17 Equivalent TL model of microstrip PCB 113
5.18 PCB attached with one cable 115
5.19 CM equivalent structure of PCB with one cable using IDM 115
5.20 Equivalent CM model based asymmetrical dipole antenna 116
5.21 Equivalent asymmetrical dipole antenna 117
5.22 Microstrip PCB under study attached with two cables 120
5.23 Microstrip PCB under study attached with two cables (a) The
equivalent CM structure using IDM (b) Decomposition process
of CM structure using the proposed method 121
5.24 Flow chart of computing CM RE from one/two cables attached to PCB 122
5.25 PCB with two-wire attached cable 123
5.26 Equivalent CM model of PCB with two-wire attached cable 124
5.27 The equivalent CM structure based on IDM and asymmetrical dipole 126
5.28 Twin-lead cable with 300 ohm characteristic impedance 127
5.29 Coaxial cable with 75 ohm 128
6.1 3D radiation pattern for three configurations (open, short and load) 130
6.2 Schematic layout of PCB#1 under study 131
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xix
6.3 Cross-sectional dimensions of PCB#1 under test 132
6.4 PCB#1 under study (a) top view (b) bottom view of DUT 133
6.5 (a) DUT placement in the shielding box (b) DUT within shielding
box in SAC (c) DUT Measurement set-up in SAC 134
6.6 (a) DM current versus CM current for short circuit and (b) DM current
versus CM current for open circuit (c) DM Current versus CM current
for 100-Ohm load 136
6.7 (a) DM radiations and (b) CM radiations both for short circuit
configuration, by proposed methods versus conventional method 138
6.8 (a) DM radiations and (b) CM radiations both for open circuit
configuration, by proposed methods versus conventional method 139
6.9 (a) DM radiations and (b) CM radiations both for 100-Ohm configuration,
by proposed methods versus conventional method 140
6.10 Predicted and measured total PCB radiated emissions for (a) short-
circuit (b) open circuit (c) 100 Ohm load 142
6.11 Simple microstrip PCB under test (PCB#2) 143
6.12 Estimated (a) DM & (b) CM RE of 150 mm long PCB trace with
75 ohm load 145
6.13 Prediction and measurement results of total RE from 150 mm long
PCB trace with 75 ohm load 145
6.14 Maximum DM- RE @ 3 meters of microstrip PCB with short-circuit
configuration 147
6.15 Maximum DM- RE @ 3 meters of microstrip PCB with open-circuit
configuration 148
6.16 Maximum DM- RE @ 3 meters of microstrip PCB with matched load
configuration 148
6.17 (a) Placement of PCB on the wooden table in SAC (b) Conducted
measurement in SAC 150
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xx
6.18 Maximum RE @ 3 meters of microstrip PCB with (a) short-circuit
(b) open-circuit (c) 82 ohm load 152
6.19 (a) PCB under test installed on metallic box (b) DUT placement on
the wooden table in SAC 154
6.20 Measurement set-up of PCB shielded with metallic box in SAC 155
6.21 Maximum RE @ 3 meters of compact PCB shielded with box 155
6.22 RE of PCB#1 using HFSS simulator and MLP model 156
6.23 RE of PCB#2 using HFSS simulator and MLP model 157
6.24 RE of PCB#3 using HFSS simulator and MLP model 157
6.25 Envelope of the trapezoidal signal with 66 MHz, 6ns rise/fall time 159
6.26 RE of PCB under test using measurement in SAC 160
6.27 The estimated CM voltage at the (a) near-end to source (b) far-end near
to load 162
6.28 The estimated CM RE for PCB-cable structure using HFSS simulation
and analytical solution 163
6.29 Estimated and simulated results for RE in (a) 50 Ohm load configuration
(b) open-circuit configuration 164
6.30 RE of PCB under test (PCB#3) using HFSS and the proposed model 165
6.31 (a) Computation of the function )(f for the entire theta values
(b) determination of the angle, (θ), with maximum emission versus
frequency 166
6.32 PCB under test with one attached cable 166
6.33 (a) Measurement set-up of DUT in SAC (b) PCB inside metallic box
(c) placement of DUT above the wooden table 168
6.34 Estimated and measured RE of PCB under test 168
6.35 Simulated and estimated CM RE of PCB attached with two cables (a)
20cm x 4 cm PCB with two (0.5m, 0.5m) attached cables (b) 10cm x
4 cm PCB with two (0.5m, 0.5m) attached cables (c) 10cm x 4 cm
PCB with two (0.3m, 0.3m) attached cables 170
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xxi
6.36 Simulated and estimated CM RE of PCB attached with two cables (a)
10cm x 4 cm PCB with two (1m,1m) attached cables (b) 10cm x 4 cm
PCB with two (0.3m, 0.5m) attached cables 171
6.37 PCB under test with two attached cables 172
6.38 Estimation and measurement of CM RE of PCB attached with two cables
(a) Measurement setup of DUT in a SAC (b) Estimated and measured
results of CM RE of DUT 173
6.39 (a) PCB inside metallic box attached with twin-lead cable (b) Measure-
ment of CM of the twin-lead cable in SAC 174
6.40 CM voltage at the load junction between PCB and twin-lead cable 175
6.41 Estimated and measured CM RE of 1m twin-lead cable attached to PCB 176
6.42 Measurement set-up of CM RE from coaxial-cables attached to PCB 177
6.43 CM and DM voltages at the junction between the coaxial cable and PCB 177
6.44 Estimated and measured CM RE from coaxial cables attached to PCB 178
6.45 Main GUI of the developed software tool 179
6.46 GUI for choosing the source type to compute the RE from PCB-traces 179
6.47 Sub-GUI to input the PCB features 180
6.48 GUI for computing RE from one or two cables attached to PCB 181
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xxii
LIST OF SYMBOLS AND ABBREVIATIONS
E - Electric Field
H - Magnetic Field
dB - Decibels
V - Volt
θ - Antenna angle
f - Frequency
Wavelength - ߣ
Zin - Input Impedance
Z0 - Characteristic Impedance
I - Current
푍 - Load Impedance
휌 - Reflection Co-efficient
Ω - Ohm
β - Phase constant
푍 - Source Impedance
휂 - Intrinsic Impedance
휀 - Free Space Permittivity
휇 - Free Space Permeability
푘 - Wave Number
퐴 - Vector Potential
AF - Array Factor
ANN - Artificial Neural Network
CE - Conducted Emission
CISPR - Special International Committee on Radio
Perturbations Interference
CM - Common Mode
CS - Conducted Susceptibility
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xxiii
CST - Computer Simulation Technology
DM - Differential Model
DRC - Design Rule Checker
EMC - Electromagnetic Compatibility
EMI - Electromagnetic interference
EN - European Norm
EUT - Equipment under Test
FCC - Federal Communications Commission
FEM - Finite Element Method
FF - Far-Field
GUI - Graphical User Interface
HFSS - High Frequency Structural Simulator
ICs - Integrated Circuits
IEC - International Electro technical Commission
IEEE - Institute of Electrical and Electronics Engineers
MAPE - Mean Absolute Percentage Error
MoM - Method of Moment
NF - Near Field
OATS - Open Area Test Sites
PCB - Printed Circuit Board
PEC - Perfect Electric Conductor
QTEM - Quasi-Transverse Electromagnetic
RE - Radiated Emission
RF - Radio Frequency
RS - Radiated Susceptibility
SAC - Semi Anechoic Chamber
SIRIM - Standards and Industrial Research Institute of
Malaysia
TEM - Transverse Electromagnetic
TL - Transmission Line
TRP - Total Radiated Power
TSM - Trace Segmentation Method
US - United States
UTHM - Universiti Tun Hussein Onn Malaysia
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
xxiv
LIST OF APPENDICES
APPENDIX TITLE PAGE
A Measurement Equipment 200
B Horn Antenna Datasheet Calibration Certificate 202
C Dipole Antenna Datasheet Calibration Certificate 203
D Uncertainty for Radiated Emission Measurement 204
E M-file of the Main GUI for the Software Tool 205
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
1CHAPTER 1
INTRODUCTION
This chapter describes in details the research background, the focus of this research
study, the problem statement, research aim, research objectives, research scopes and
the significance of the study.
Research background
The worldwide proliferation of high-speed digital devices has imposed many
challenges to the circuit-designers of these modern devices. They must take into
account not only the product functionality, but also the electromagnetic compatibility
(EMC) regulatory requirements of their products. Thus, it is important to make those
electronic devices electromagnetically compatible with mandatory governmental
regulatory requirements before they can be sold legally and internationally [1], [2].
Radiated emission (RE) compliance is one of the EMC requirements to
ensure that the electronic product can function satisfactory without introducing any
electromagnetic interference (EMI) to the nearby electronic devices. In the past, the
electronic devices were operating in the low frequency range (less than 1 GHz)
resulting in lower emissions. Unfortunately, the modern devices are working in
higher frequencies (i.e. gigahertz range) where they become significant radiators of
electromagnetic (EM) energy. So the manufacturers of electronic devices must
control properly the RE of their products before they can sell them globally [3].
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
2
Overview of EMC
According to the glossary of the International Electrotechnical Commission (IEC)
[4], EMC is defined as the ability of any electronic device or equipment to function
satisfactorily in its electromagnetic environment, and at the same time, not contribute
excessive electromagnetic disturbance to other devices/equipment/systems. The
electronic product may interfere with itself or with another product in that certain
environment as illustrated in Figure 1.1. In other words, a system or product is
classified as electromagnetically compatible if successfully fulfilled the following
requirements [5]-[7]:
i. it does not cause interference with other systems;
ii. it can tolerate the radiation or emission from other systems;
iii. it does not cause interference with itself.
Figure 1.1: EMC aspects and their interrelationship [6]
Electromagnetic Compatibility (EMC) (Ability to function satisfactorily)
Electromagnetic Susceptibility (EMS) (Suffering interference)
Electromagnetic Emission (EME) (Introducing disturbances)
Inside system
To other systems
By other systems
By system itself
Intersystem compatibility
Intrasystem compatibility
RE as the main scope of this research
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
3
1.2.1 EMC standards
FCC [8] and CISPR [9] standards are the most widely adopted regulations for
commercial digital products around the world. FCC standard regulates the EM
emissions of digital devices marketed in US while the CISPR regulations focuses on
the EM emissions of digital devices sold in other countries of the world except U.S.
FCC standards classified the digital devices into two classes; Class A and
Class B [10]. The digital devices intended to be used in a business; industrial,
scientific and medical fields are classified to Class A, while the digital devices
applied in a residential environment belong to Class B.
Commonly, Class A limits are less strict than the Class B limits. This is due
to two reasons. The first reason is that the interference is more significant in the
residential environment since the electronic devices are working closer to each other,
so that it is not easy to minimize the interference effect. The second reason is to
consider the inability of the owners and users of Class B to protect their devices from
the electromagnetic interference. FCC standards specified the range for radiated
emissions from 30 MHz to 1 GHz and it can be extended further up to 40 GHz.
Figure 1.2 [6] shows the RE limits for Class A and Class B digital devices in the
FCC standards with 3 meters measurement distance.
In addition to FCC standards, CISPR is the European standard which had
been adopted in many countries around the world excepting U.S. One of most
important CISPR recommendations is CISPR 22 [9] which regulates the EM
emissions of information technology equipment (ITE). Similar to FCC, CISPR 22
also classified the digital devices to Class A and Class B. The RE is regulated in the
frequency range from 30 MHz to 1 GHz. Figure 1.3 [9] shows the radiated emission
limits for Class A and Class B ITE equipment in CISPR 22.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
4
Figure 1.2: RE limits for Class A and Class B digital device in the FCC standards [6]
In CISPR22, both Class A and Class B are always set at 10 meters distance
between the digital device under test and the receiving antenna. In contrast to that,
FCC regulations for Class A digital devices are set at 10 meters whereas Class B
regulations are set at 3 meters for the digital devices. Hence, the FCC regulations for
Class A for digital devices can be compared quite straight forward with CISPR 22
Class A whereas the CISPR 22 Class B for digital devices cannot be compared with
FCC Class B.
Figure 1.3: RE limits for Class A ITE equipment in CISPR 22 [9]
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
5
In this case, FCC limits for Class B digital devices must be scaled at -10.45 dB to be
compared with the CISPR22 Class B limits. Therefore, the emissions at 3 meters are
assumed to be reduced by a factor of 3/10 if the measurement distance is moved to a
farther distance of 10 meters and vice versa. However, most countries adopt the
CISPR recommendations. This sets limits for the radiated and conducted emissions
of ITE equipment. Although there are a large number of EMC standards, the primary
one is the European norm EN 55022.
Characterization methods of radiated emissions
Generally, the radiated emissions of electronic devices can be quantified using
measurement or modelling methods. Open area test site (OATS) is one of the
measurement methods preferred by FCC. However, it requires making the test site
free of the unwanted noise signals which cannot be obtained easily considering the
widespread of the electronic devices everywhere. Alternatively, semi-anechoic
chamber (SAC) is employed as test site for measurement of RE. SAC provides an
all-weather measurement capability as well as security.
A SAC consists of a shielded room lined with radio-frequency absorber
material on the sides and at the top of the room to prevent reflections and simulate
free space. The floor of the room constitutes a ground plane without an absorber, and
this causes reflections that must be accounted for when performing simulations by
models as described in details in Chapter 2.
Although these measurement methods can evaluate the RE accurately, they
are not proper option to avoid the EMC issues earlier before fabrication of first
prototype. Alternatively, the RE can be estimated earlier at the design stage of the
product using one of the modelling methods [3], [6], [10] such as the numerical and
analytical methods.
Full-wave numerical solver is one of the modelling methods [11]–[15] used
to compute numerically the RE of the electronic devices. It employs one of the most
popular numerical techniques such as Finite Difference Time Domain method
(FDTD), Finite Element Method (FEM), Method of Moments (MoM) and the Partial
Element Equivalent Circuit (PEEC) method. These numerical techniques are widely
employed in a lot of the commercial software such as Ansys High Frequency
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
6
Structural Simulator (HFSS) [16] which adopts the FEM method. Although those full
wave numerical solvers can compute the RE accurately, they require extensive
computational resources and time especially for today’s high speed devices.
Near Field (NF) to Far Field (FF) transformation is the second modelling
method for estimating the EMC-RE [17]–[23]. This method is relatively much more
efficient than the conventional full wave numerical methods. However, it is not easy
to obtain the equivalent model. In addition to that, it is difficult to adjust the
dipoles/current sources with the computation time. Furthermore, it still requires the
first prototype to be built for performing the NF scanning. Hence, it is not suitable
option for early estimation of RE from electronic devices.
In this research, novel analytical methods have been developed for RE
estimation. The analytical methods, however, require to identify the RE sources on
the electronic product. Two main sources of RE are identified in this study; namely
PCB-traces and the peripheral cables attached to PCB shown in Figure 1.4.
Figure 1.4: Radiated emissions of PCB attached with cables
Problem statement
The advanced technologies of integrated circuits as well as the ever increasing of
clock speed toward the Gigahertz range have enhanced the trends to smaller size
electronic devices. As a result, many electronic devices can work in less space which
increases the probability of EMI occurrence. So, the circuit designers must take into
the account the EMC-RE regulatory standards to avoid the excessive EMI above the
standard limit.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
7
Several solutions are available for compliance of the electronic products with
EMC-RE standards. However, solutions implemented at the end of the design cycle
resulted in product delays, as well as added cost [24]–[26]. Figure 1.5 shows that
implementations of EMC in the product phase are expensive and rather iterative
since it require fabricating the first prototype for measurement of RE. In contrast to
that, the implementation of EMC at the design stage is relatively cheaper than at the
production phase. In addition to that, many solutions and techniques are available at
the design stage of the electronic device [27]. Therefore, it is becoming critically
important to include EMC very early in the design phase of high speed systems.
Hence, properly designed PCB’s can offer a cost-effective approach for achieving
EMC compliance.
In the high frequencies, which may be existed due to the harmonics of the
digital signal outside the operating frequency, PCB has dimensions of the order of
several wavelengths and therefore it produces a significant amount of emissions.
Therefore, it is important to focus on the issue of EM emissions from PCBs to
provide a simple measure of the EMC based performance of PCB to the circuit
designers.
Figure 1.5: Cost of EMC implementation on the electronic product [27]
As described previously, the measurement of RE from PCB is not proper
option since it requires the first prototype to be built which may be involved in the
redesign process resulting in more delay in product marketing and increasing in the
unit cost. So, it cannot be employed for early prediction of RE from PCBs.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
8
Alternatively, 3D full wave numerical solvers can estimate the RE of PCB
accurately. Unfortunately, it still requires intensive computational time and powerful
resources considering the complexity of the today’s high speed PCBs. Therefore, it is
essential to develop new modelling approach which can estimate the RE from PCB
accurately and efficiently.
On PCB level, there are many sources of the unintentional RE located on
PCB such as the ICs, PCB-traces and PCB-attached cables. However, the two
predominant sources of RE on PCB are the PCB-traces and the PCB-attached cables
as shown in Figure 1.6. In high speed PCBs, the traces are electrically long and thus
become efficient radiators of EM energy. Therefore, it is important to model and
estimate the RE produced from these traces.
In addition to that, PCBs are mostly attached with cables to act as interfacing
and communication cables with other devices. These cables are well-known as a
significant source of the unintentional common-mode RE. Thus, it is important to
take into the account the emission produced from the cables attached to PCB to avoid
the product failure due to these cables.
Figure 1.6: Basic idea of the proposed models
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
9
This research clearly address the issue of RE from PCB-traces and PCB
attached cables. First, a simple circuit consists of source trace and load is analysed.
Once the RE of this PCB-trace is estimated, the same idea can be employed to
compute the RE of PCB fully populated by traces. Finally, the compliance of EMC-
RE of entire product can be checked based on these proposed models as shown in
Figure 1.6.
Objectives of the research
In this research, novel models are proposed for estimating the RE from electrically
long PCB-traces as well as from cables attached to PCB. Specifically, the major
objectives of this study are:
i. To investigate and apply novel analytical methods for estimating the RE from
electrically long PCB-traces.
ii. To develop novel analytical methods for computing the RE from cables
attached to electrically long PCB-traces
iii. To build neural network model for estimating RE of PCB-traces for non-
Quasi-TEM current distribution.
iv. To verify and validate the proposed models using HFSS simulation and
measurements in SAC.
Scope of the research
The scope of this research involved analysis, modelling, analytical study, simulation
and measurement. In order to achieve the objectives, more constraints are applied to
limit the scope as follows:
i. The PCB traces and the attached cables are electrically long (trace length >>
wavelength).
ii. PCBs under test are designed and fabricated with two configurations (single
sided and double-sided PCB) according to the available resources and
facilities.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
10
iii. The cables attached to PCB-traces are connected at the source / load junctions
only according to the practical design of PCB-cables structures
iv. The RE of PCB-traces is studied analytically within the frequency range from
30 MHz to 3GHz to satisfy the condition of electrically long PCB-traces and
the condition of QTEM operation mode (cross-section of PCB-trace is short
compared to wavelength).
v. The effects of PCB coupling to grounding are omitted.
Significance of the research
Any electronic product needs to be electromagnetically compatible with EMC
regulations before it can be sold to the consumer. If the product failed to satisfy the
RE standards, it will require second cycle of redesign and validation. This process is
rather iterative and thus it increases the unit cost and minimizes the market sharing.
Therefore, the estimation of PCB RE would give more options in the design
modification as well as to avoid many of RE issues.
This research work focuses on developing new analytical modelling methods
for estimating the RE of electrically long PCB-traces based on transmission-line
theory, dipole antenna and travelling-wave antenna. Additionally, novel methods are
developed for estimating common-mode RE from cables attached to PCB using
imbalance difference model and asymmetrically dipole antenna. These proposed
methods can be employed to overcome many realistic problems of PCB-cables
structure at reasonable computational time with high level of accuracy comparing to
the experimental results.
Organisation of the thesis
This thesis consists of 7 Chapters. Each chapter is described briefly as follows:
Chapter 1 presents the background of the study, the problem statement, the objective
of the study, scope and limitations of the research work, and the significance of the
research.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
11
Chapter 2 describes critically the latest literature review of this research and specifies
the research scope. Therefore, this research work is comparable with any other new
research around the world.
The research methodology is given in Chapter 3 which details the approaches taken
to introduce novel techniques including analyses, modelling, simulation and
measurement in the semi-anechoic chamber.
Chapter 4 describes the analytical method to determine the RE from PCB traces. It
presents in details the analytical method for computing both DM and CM RE. This
chapter describes also the ANN model for estimating the total RE.
The methods for computing the RE of cables attached to PCB are given in Chapter 5.
A detailed analytical solution has been presented in this chapter for computing RE
from one or two cables attached to PCB.
Chapter 6 discusses the analysis of the research results based on the works
introduced in Chapter 4 and Chapter 5. Detailed analysis of all the proposed models
are presented and verified by comparing with both 3D full wave HFSS simulation
results and the results taken from measurement in SAC.
Chapter 7 provides a conclusion and details on future work, on how to improve this
research, and gives suggestions for forthcoming improvement to the research.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
2CHAPTER 2
LITERATURE REVIEW
This chapter describes in details the literature related to EM emissions of PCB. First,
a brief description of the basic radiators such as dipole antenna has been presented.
Secondly, the main sources of EM emissions on PCB are identified; namely PCB
traces and PCB-attached cables. Third, the previous studies for estimating the
emissions produced from the PCB-traces and PCB-attached cables are presented. At
the end of this chapter, a quick review of the imbalance difference model and the
artificial neural network are introduced to address the gap of this research work.
Fundamentals of the radiating systems
Basically, EM emission is generated due to the time-varying current flowing through
the electronic devices. However, these emissions can be produced from intentional
radiators such as dipole antenna or from unintentional radiators such as the PCBs that
are designed for specific application. Therefore, it is important to give a brief
description of the basic intentional radiators for better understanding of the EM
emissions of PCBs. The most important radiating antennas such as infinitesimal
dipole, short dipole and long dipole antennas are described in this section.
2.1.1 Infinitesimal dipole
For a z-directed dipole located with current I and length dl at the origin of the
coordinate system, as shown in Figure 2.1, the vector potential 퐴(푟) can be
expressed as [28]:
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
13
퐴(푟) =휇4휋
퐽 푟 푒
푟 − 푟 푑푉 =
휇 퐼푑푙4휋
푒푟 푍
(2.1)
x
z
dlI
H
ErE
y
r
Figure 2.1: A z-directed electric point dipole [28]
where
k is the free space wave number (k=2π/λ)
λ is the wavelength,
푟, 푟 are the vector and scalar distance from the antenna respectively
휇 is the free space permeability
푍 is the characteristic impedance of the antenna
퐼 is the flowing current on the antenna
푑푙 is the antenna length
퐽 is the electric current density source.
The electric (퐸) and magnetic field (퐻) can be obtained based on the following
expressions [28]
퐻 =1휇 ∇ × 퐴
(2.2)
퐸 =1
푗휔휀 ∇ × 퐻 (2.3)
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
14
D
Based on (2.1) and (2.2), the electric field components of the point dipole are written
as [28]
퐸 = −퐼푑푙푘 푒
4휋 푍 sin휃 1푗푘푟 +
1(푗푘푟) +
1(푗푘푟)
(2.4)
퐸 = −퐼푑푙푘 푒
4휋 푍 2 cos 휃 1
(푗푘푟) +1
(푗푘푟)
(2.5)
퐻 = −퐼푑푙푘 푒
4휋 sin휃 1푗푘푟 +
1(푗푘푟)
(2.6)
where the free space impedance 푍 = 휇휀 =377 Ohm. The other components
퐸 ,퐻 and 퐻 are zeros.
According to (2.4), (2.5) and (2.6), it is observed that the radiated field
depends on the distance from the source to the observation point. Considering
radiation source with the largest dimension D, the emissions can be divided into
three regions, namely the reactive near-field ( 362.0 Dr ), radiating near field
( 23 262.0 DrD ) and far field ( 22Dr ) starting from closest point to
source to the farthest point [28], as shown in Figure 2.2. In fact the boundaries
between the regions are only vaguely defined and changes between them are gradual.
Figure 2.2: Definition of near-field and far-field regions [28]
The wave impedance, which is the ratio of the electric and magnetic field,
also depends on the distance from the source. In the far field ( 3r ) , only the
r
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
15
radiating term ( r1 term) of the field is significant so the electric and magnetic field
is related by the free space wave impedance as [28]
푍| ≫ =퐸퐻
≫= 푍
(2.7)
But in the near field, the wave impedance varies widely and depends on the
characteristics of the source. It can be expressed as [28]
푍 =퐸퐻 = 푍
1 − 푗 1(푘푟)
1 + 푗 1(푘푟)
(2.8)
Figure 2.3: Wave impedance of electric and magnetic sources [28]
It is well-known that the wave impedance near the electric source is very high
whereas the wave impedance near the magnetic source is very small as shown in
Figure 2.3. In the NF situation, the characteristics of the source are reflected in the
EM wave properties. But in the far-field there is nothing in the field properties to
identify the characteristics of its source. Thus, the electric/magnetic field components
in the far-field region ( 1kr ) can be simplified and approximated by [28]
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
16
퐸 ≅ 푗휂푘퐼푑푙푒
4휋푟 sin휃
(2.9)
퐻 ≅ 푗푘퐼푑푙푒
4휋푟 푠푖푛 휃
(2.10)
where η is intrinsic impedance (377 ohm for free space). The other components
퐸 ≅ 퐸 = 퐻 = 퐻 are approximately zero.
2.1.2 Short dipoles
Commonly, the dipoles are classified based on the ratio between wavelength and the
physical length of the dipole. While the infinitesimal dipole has a length 50dl ,
the short dipoles are that dipoles with length 5010 dl .
x
z
y
r
2l
2l
R),,( rP
zd
Figure 2.4 : A z-directed short dipole [28]
Following the same procedure in Section 2.1.1, the far-field of both the
electric and magnetic fields can be written as [28]
퐸 ≅ 푗휂푘퐼푙푒
8휋푟 sin휃
(2.11)
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
17
퐻 ≅ 푗푘퐼푑푙푒
4휋푟 sin휃
(2.12)
The other components 퐸 ≅ 퐸 = 퐻 = 퐻 are approximately zero.
2.1.3 Long dipoles
The long dipoles are those dipoles with length greater than one tenth wavelength and
no longer than one wavelength. Although the near field and far fields of the dipoles
can be expressed analytically, it is very difficult to be obtained for the near fields. In
addition to that, the most important in EMC is the maximum emission at 3 or 10 m
far from the source which is in the far-field region. Therefore, we focused on the
electric field in the far-field region. Following the same procedure in Section 2.1.1,
the electric field can be given as [28]
퐸 ≅ 푗휂퐼푒
2휋푟 cos 푘푙
2 cos휃 − cos 푘푙2
sin휃 (2.13)
The length is varied with maximum one wavelength. However, a special case
can be extracted by replacing the dipole length by half-wave length 2l . This
antenna is widely known as half-wavelength dipole antenna which its electric field in
the far-field region is expressed as [28]
퐸 ≅ 푗휂퐼푒
2휋푟 cos 휋
2 cos휃sin휃 (2.14)
Measurement and modelling of EM emissions of PCB
In Section 1.3, a brief introduction has been presented for the measurement and
modelling methods of RE. In this section, a detailed description of several
measurement and modelling methods are introduced. In fact, the modern PCBs
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
18
compose of ICs, active and passive components, connecting traces, as well as
input/output ports and attached cables.
Although the PCB may function properly as it is designed to do, it may
produce unintentional emissions which may disturb the functional operation of the
other nearby devices [6]. So, the electronic systems must satisfy all the design
requirements not only the functional operation but also controlled emissions. Thus
emissions from electronic equipment have to be compliant with the limits defined in
national and international EMC standards [29]. Hence, those emissions can be
characterized using measurement or modelling methods [30] as illustrated in the next
sections.
2.2.1 Measurement of EM emissions of PCBs
Several measurement methods of RE from PCBs are available in the literature such
as TEM cell method [31], [32] and reverberation chamber method [33] and SAC
method. However, each method has its own measurement set-up as described in the
next sub-sections.
2.2.1.1 TEM cell method
The emissions of a device under test (DUT) can be measured inside TEM cell, as
shown in Figure 2.5 in the frequency range from 150 KHz to 1 GHz, and can be
extended beyond 1 GHz with gigahertz TEM (GTEM) cell [31], [32].
In this method, the DUT is positioned in the top or bottom of the TEM cell.
The RF voltage detected by TEM cell is then passed to the spectrum analyser via pre-
amplifier. The EM emissions of the DUT can be then computed based on the
detected RF voltage. However, this method requires performing the measurement
with at least two orientations of DUT to capture the total emission which is time
consuming process.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
19
Figure 2.5: Measurement set-up of EMC-RE using TEM cell [31]
2.2.1.2 Near-field scanning method
The other option for measurement of RE is to perform near-field scanning [34] of the
DUT at the distance less than one sixth of the wavelength. During the scanning
process, the magnetic probe is positioned closer to the DUT to detect the radiated
signals as illustrated in in Figure 2.6.
Figure 2.6: Measurement set-up of RE using NF scanning method [34]
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
20
The outputs of the probes are then converted to a 2D field map showing the
field strength distribution. Using NF scanning results, the far-field electric field can
be computed using one of the transformation techniques [35]–[37].
2.2.1.3 Radiation pattern measurement method
This method is widely used for evaluating EMC-RE because the standard test EN-
55022 [9] adopted this method for measuring the RE of DUT. Based on FCC and
CISPR 22, the DUT is placed at 10m, or 3m in case of high ambient noise level far
away from the receiving antenna. A biconical antenna is used as a receiving antenna
for frequencies below 1 GHz. For frequencies above 1 GHz, a log-periodical antenna
or a horn antenna should be used. The DUT is mounted on a turntable that can rotate
through 360° to find the maximum emission. The receiving antenna is scanned in
height from 1m to 4m to find the maximum level of RE as shown in Figure 2.7. The
radiation pattern of DUT is then obtained by varying the antenna azimuth and
polarization through 360° during the measurement.
Figure 2.7: Measurement of RE using radiation pattern method
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
21
2.2.1.4 Reverberation chamber method
In this method, the maximum REs are computed by measurement of the total
radiated power (TRP) of a DUT instead of direct measurement of the electric field.
For the purpose of measurement of TRP, an electrically large reverberation chamber
is used. This chamber is basically highly conductive overmoded enclosed cavity
equipped with one or more metallic tuners/stirrers [33] to achieve a statistically
uniform and isotropic electromagnetic environment as shown in Figure 2.8.
The stirrers are adjusted to rotate very slowly compared to the sweep time of
the EMI receiver to obtain sufficient number of samples. During a cycle period of the
stirrers, the maximum received power or averaged received power can be measured
and recorded. Those recorded signals are then converted to the TRP and the free
space field strength generated from the DUT.
Although all the measurement methods provide accurate results, the main
disadvantage is that they require fabricating the first prototype to be used as DUT.
Thus, they are not suitable option for early prediction of RE from PCBs. So, it is
crucially important to employ another modelling approach for estimating the
emissions of PCBs based on the PCB design specifications before the product is
fabricated to ensure time and cost savings.
Figure 2.8: Measurement of RE using reverberation chamber method [33]
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
22
2.2.2 Modelling methods for EM emissions of PCB
Today, PCBs in high speed devices have a very complex design. Therefore, it is
difficult and expensive to modify the PCB-layout at the production phase of the
electronic devices. Thus, it is important to consider EMC-RE at the design process.
Modelling methods are preferable option since they can estimate the RE of the
electronic products earlier in the design phase. This section describes briefly several
modelling methods for computing the RE of PCB such as design-rule checking,
equivalent modelling and analytical modelling.
2.2.2.1 Full-wave numerical modelling
In this method, the entire structure of PCBs must be modelled by discretizing the
space in terms of grid or mesh and then solving numerically Maxwell equations in
free space, dielectrics, and conductors. Actually, the physical geometries of all the
elements (dielectrics, traces, excitations, loads, etc.) are modelled and specific
attention is given for the material properties and frequency.
The models are then solved numerically using one of the computational
electromagnetics (CEM) methods [38]. Many numerical methods have been
developed over the past decades for solving the Maxwell equations either in
differential forms such FDTD [14], FEM [39], and transmission line matrix (TLM)
[40] or integral forms such as MoM [41] and PEEC [42]. Many commercial tools
have adopted these numerical techniques such as FEKO which adopted MoM [43],
FEM based HFSS [16] and CST which employed hybrid methods [44].
For better illustration of these numerical techniques, detailed descriptions are
introduced for each technique as follows:
i. FDTD technique
In this technique, the structure is chunked into small cells each of which do not
exceed one tenth of the wavelength of the maximum operating frequency. Figure 2.9
shows the discretized cell, which is defined as Yee cell [45]. The EM field
components ZYXZYX E,E,E,H,H,H are given as below.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
23
⎝
⎜⎜⎜⎛휕퐻휕푦 −
휕퐻휕푧
휕퐻휕푧 − 휕퐻
휕푥휕퐻휕푥 − 휕퐻
휕푦 ⎠
⎟⎟⎟⎞
=
⎝
⎜⎜⎛휀
휕퐸휕푡
휀휕퐸휕푡
휀 휕퐸휕푡 ⎠
⎟⎟⎞
+퐽퐽퐽
(2.15)
⎝
⎜⎜⎜⎜⎛휕퐸푌휕푧 − 휕퐸푍
휕푦휕퐸푍휕푥 − 휕퐸푋
휕푧휕퐸푋휕푦 − 휕퐸푌
휕푥 ⎠
⎟⎟⎟⎟⎞
=
⎝
⎜⎜⎛휇
휕퐻휕푡
휇 휕퐻휕푡휇 휕퐻휕푡 ⎠
⎟⎟⎞
(2.16)
where J is the electric current density, is the material permittivity and is the
material permeability.
Figure 2.9: Electromagnetic field components of one cell using FDTD [26]
It is necessary to compute the electric and magnetic field components of the
entire structure for solving the equations (2.15) and (2.16). FDTD is a good option
for computing the maximum RE of PCBs. However, it is not proper option since it
needs to be involved in post processing to obtain the far-field electric field.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
24
i. FEM technique
FEM is the second technique for finding the maximum RE of PCBs. The entire
structure is modelled by large number of finite elements each of which has its
modelling equation. Generally, FEM method can solve more complex geometries
and more complex loading conditions comparing to the other methods such as
FDTD. But, it is time-consuming process especially for complex structures of PCBs.
ii. MoM technique
In contrast to the FDTD and FEM, this technique does not desire more computational
process of the structure. Instead, it desires to calculate the boundary values only. As a
result, it is efficient for solving problems with small surfaces. However, it is also not
preferred for estimating the RE from PCBs since it still requires intensive
computational time especially for large structures.
Although the full wave numerical solvers can provide accurate estimation of
EM emissions, we can conclude that these numerical solvers are not well-suited
option due to the huge requirement for computer speed and memory to estimate the
RE of PCBs.
2.2.2.2 Design rule checking
Design rule checker is software that can check whether the rules of design are
violated or not in the same manner as professional EMC expert does. It reads the
board layout information from automated board layout tools (such as Allegro, Protel,
Board Station, etc.) and checks if certain EMC design guidelines have been
breached. This method does not provide quantitative estimation of EM emissions
from the PCB. However, it gives indication of the overall correctness of the design.
DRC can also help to identify and locate the potential source of emissions on the
designs which is very difficult to point it out. In the recent days, several DRC
softwares are available such as EMI Stream [46], EMSAT [47], and Zuken CR-5000
Lightning EMC [48].
Although these DRC softwares are easy to use even for non-expert peoples,
they can do not provide any quantitative estimation of RE from the device. In
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
186
REFERENCES
1. Montrose, M. I.. Printed Circuit Board Design Techniques for EMC
Compliance — A Handbook for Designers. 2nd edition. New York: IEEE
press. 2000.
2. Hockanson, D. M., Drewniak, J. L., Hubing, T. H., Van Doren, T. P., Sha, F.,
&Wilhelm, M. J.. Investigation of fundamental EMI source mechanisms
driving common-mode radiation from printed circuit boards with attached
cables. IEEE Transaction on Electromagnetic Compatibility. 1996. 38(4):
557–566.
3. Leroux, E., Caniggia, S., Canavero, F., & Demoulin, B.. Evaluation of
radiated emissions from printed circuit boards and cables at post-layout level.
International Symposium on Electromagnetic Compatibility. September 17-
20 . Rome, Italy: EMC’96 Roma. 1996. 662–667.
4. British Standard Institution. BS 4727-1:Group09. London, IEC 60050-161.
1990.
5. Kodali, V. P.. Engineering Electromagnetic Compability-Principles,
Measurements, and Technologies, 2nd edition. New Jersey: IEEE Press,
2001.
6. Paul, C. R.. Introduction to Electromagnetic Compatibility, 2nd edition. New
Jersey: Wiley intersceince, 2006.
7. Knott, A. Improvement of out-of-band Behaviour in Switch-Mode Amplifiers
and Power Supplies by their Modulation Topology. Ph.D. Thesis.Technical
University of Denmark; 2010.
8. US Government Publishing Office. FCC-Part 15. U.S., no 5. 2010.
9. International Special Committee on Radio Interference (CISPR). EN55022.
Brussels, CISPR22. 2006.
10. Caniggia, S., & Maradei, F.. Signal Integrity and Radiated Emission of High-
Speed Digital Systems. Chichester: John Wiley & Sons. 2008.
11. Yeung, L. K. & Wu, K.-L.. Generalized partial element equivalent circuit
(PEEC) modeling with radiation effect. IEEE Transaction on Microwave
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
187
Theory and Techniques. 2011. 59(10): 2377–2384.
12. Yeung, L. K. & Wu, K.-L.. PEEC modeling of radiation problems for
microstrip structures. IEEE Transaction on Antennas and Propagation.
2013. 61(7): 3648–3655.
13. Laroussi, R. & Costache, G. I.. Finite-element method applied to EMC
problems. IEEE Transaction on Electromagnetic Compatibility.1993. 35(2):
178–184.
14. Fornberg, P. E., Byers, A. & Piket-May, M.. FDTD modeling of printed
circuit board signal integrity and radiation. IEEE International Symposium
on Electromagnetic Compatibility. August 21-25. Washington, DC: IEEE
press. 2000. 307–312.
15. Cangellaris, A. C.. Frequency-domain finite element methods for
electromagnetic field simulation: fundamentals, state of the art, and
applications to EMI/EMC analysis. IEEE International Symposium on
Electromagnetic Compatibility. August 19-23. Santa Clara, CA: IEEE press.
1996. 107–116.
16. Ansys Corporation. Ansoft High Frequency Structure Simulator. Pittsburgh:
User’s Guide v9. 2013.
17. Fan, H., Schlagenhaufer, F.. Near fieldfar field conversion based on genetic
algorithm for predicting radiation from PCBs. IEEE International Symposium
on Electromagnetic Compatibility. July 9-13. Honolulu, HI: IEEE press.
2007. 1–6.
18. Las-Heras, F., Galocha, B. & Besada, J. L.. Far-field performance of linear
antennas determined from near-field data. IEEE Transaction on Antennas &
Propagation. 2002. 50(3): 408–410.
19. Li, P. & Jiang, L. J.. Source reconstruction method-based radiated emission
characterization for PCBs. IEEE Transaction on Electromagnetic
Compatibility. 2013. 55(5): 933–940.
20. Petre, P. & Sarkar, T. K.. Planar near-field to far-field transformation using
an array of dipole probes,” IEEE Transaction on Antennas & Propaation.
1994. 42(4): 534–537.
21. Ramanujan, A., Riah, Z., Louis, A. & Mazari, B.. Modeling the
electromagnetic radiation of passive microwave components using a near-
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
188
field scanning method. IEEE Transaction on Electromagnetic Compatibility.
2010. 52(4): 1056–1059.
22. Ravelo, B., Liu, Y. & Jastrzebski, A. K.. PCB near-field transient emission
time-domain model. IEEE Transaction on Electromagnetic Compatibility.
2015. 57(6): 1320–1328.
23. Sarkar, T. K. & Taaghol, A.. Near-field to near/far-field transformation for
arbitrary near-field geometry utilizing an equivalent electric current and
MoM. IEEE Transaction on Antennas & Propagation. 1999. 47(3): 566–
573.
24. Hubing, T., Kashyap, N., Drewniak, J., Van Doren, T. & DuBroff, R.. Expert
system algorithms for EMC analysis. 14th Annual Review of Progress in
Applied Computational Electromagnetics. March 16-20. Monterey, CA:
Naval postgraduate School. 1998. 905–910.
25. Shim, H., Hubing, T., Van Doren, T., DuBroff, R., Drewniak, J.,
Pommerenke, D. & Kaires, R.. Expert system algorithms for identifying
radiated emission problems in printed circuit boards. International
Symposium on Electromagnetic Compatibility. August 9-13. Boston, MA:
IEEE press. 2004. 57–62.
26. Archambeault, B., Connor, S., Halligan, M. S., Drewniak, J. L. & Ruehli, A.
E.. Electromagnetic radiation resulting from PCB/high density connector
interfaces. IEEE Transaction on Electromagnetic Compatibility. 2013. 55(4):
614–623.
27. Ott, H. W.. Electromagnetic Compatibility Engineering. New Jersey: John
Wiley & Sons. 2009.
28. Balanis, C. A.. Antenna Theory: Analysis and Design. 4th edition. New
Jersey: John Wiley & Sons. 2012.
29. Christopoulos, C.. Principles and Techniques of Electromagnetic
Compatibility. 2nd edition. Boca Raton: CRC press. 2007.
30. Fu, Y. & Hubing, T.. Analysis of radiated emissions from a printed circuit
board using expert system algorithms. IEEE Transaction on Electromagnetic
Compatibility. 2007. 49(1): 68–75.
31. British Standard Institute. Electromagnetic Compatibility - Testing and
Measurement TechniquesEmission And Immunity Testing in Transverse
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
189
Electromagnetic (TEM) Waveguides. London, EN61000-4-20. 2003.
32. Kaiser, Kenneth L.. Electromagnetic Compatibility Handbook. Boca Raton:
CRC press. 2004.
33. British Standard Institute. Electromagnetic Compatibility - Testing And
Measurement Techniques –Reverberation Chamber Test Methods. London,
EN61000-4-21. 2003.
34. British Standard Institute. Integrated Circuits-Measurement of
Electromagnetic Emissions From 150 kHz to 1 GHz. London, IEC61967-3.
2005.
35. Ren, X., Maheshwari, P., Zhang, Y., Khilkevich, V., Fan, J., Zhou, Y., Bai,
Y. & Yu, X.. The impact of near-field scanning size on the accuracy of far-
field estimation. IEEE International Symposium on Electromagnetic
Compatibility. August 4-8. Raleigh, NC: IEEE press. 2014. 582–587.
36. Weng, H., Beetner, D. G. & DuBroff, R. E.. Predicting TEM cell
measurements from near field scan data. IEEE International Symposium on
Electromagnetic Compatibility. August 14-18. Portland, OR: IEEE
press.2006. 560–564.
37. Vives-Gilabert, Y., Arcambal, C., Louis, A., de Daran, F., Eudeline, P. &
Mazari, B.. Modeling magnetic radiations of electronic circuits using near-
field scanning method. IEEE Transaction on Electromagnetic Compatibility.
2007. 49(2): 391–400.
38. Hubing, T. H.. Survey of Numerical Electromagnetic Modeling Techniques. University of Missouri-Rolla. TR91-1-001.3, 1991.
39. Jin, jian-Ming.. The Finite Element method in Electromagnetics. 3rd edition.
New Jersey: Wiley-IEEE Press. 2014.
40. Russer, P., & Russer, J. A.. Application of the transmission line matrix
(TLM) method to EMC problems. Asia-Pacific Symposium on
Electromagnetic Compatibility. May 21-24. Singapore: IEEE press. 2012.
141–144.
41. Harrington, R. F.. Field Computation by Moment Methods. New Jersey:
Wiley-IEEE Press, 1993.
42. Ekman, J.. Electromagnetic Modeling Using the Partial Element Equivalent
Circuit Method. Ph.D. Thesis. Lule°a University of Technology; 2003.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
190
43. FEKO Altair HyperWorks Development S.A. Ltd. FEKO 3D Numerical
Simulator. Hampton, VA: User Manual. 2016.
44. CST Corporation. EMC Simulation for Electronics Products using CST.
Darmstadt, Germany: CST Brochure. 2015.
45. Yee, K. S.. Numerical solution of initial boundary value problems involving
maxwell’s equations in isotropic media. IEEE Transaction on Antennas and
Propagation. 1966.14(3): 302–307.
46. NEC Corporation. EMIStream v4.9-002. Morrison, CO: EMI suppression
Tool. 2016.
47. EMSAT Corporation. (2016). EMSAT. Retrieved on July 25, 2016, from
http://www.mossbayeda.com/products/emsat/emsat-overview
48. Zuken Corporation, (2016). CR-5000 Lighting EMC. Retrieved on July 25,
2016, from http://www.zuken.com/en/products/pcb-design/cr-5000.aspx
49. Pan, J., Gao, X. & Fan, J.. Far-field prediction by only magnetic near fields
on a simplified huygens’s surface. IEEE Transaction on Electromagnetic
Compatibility. 2015. 57(4): 693–701.
50. Wang, J., Yan, Z., Zhang, W., Kang, T. & Zhang, M.. Effective method of
near to far field transformation based on dipoles model. Asia-Pacific
Microwave Conference (APMC). December 6-9. Nanjing: IEEE press. 2015.
1–3.
51. Farouq, M., Serhir, M. & Picard, D.. Far-field antenna calculation based on
randomly distributed near-field measurement Data. 9th European Conference
on Antennas and Propagation (EuCAP). May 13-17. Lisbon: IEEE press.
2015. 1–5.
52. Hongmei, F. & Schlagenhaufer, F.. Far-field approximation of electrically
moderate-sized structures by infinitesimal electric and magnetic dipoles.
IEEE Transaction on Electromagnetic Compatibility. 2010. 52(1): 82–88.
53. Beghou, L., Liu, B., Pichon, L., & Costa, F.. Synthesis of equivalent 3-D
models from near field measurements - Application to the EMC of power
printed circuit boards. IEEE Transaction on Magnetics. 2009. 45(3): 1650–
1653.
54. Tong, X., Thomas, D. W. P., Nothofer, A., Sewell, P. & Christopoulos, C..
Modeling electromagnetic emissions from printed circuit boards in closed
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
191
environments using equivalent dipoles. IEEE Transaction on
Electromagnetic Compatibility. 2010. 52(2): 462–470.
55. Yuri, Á., Fernando, L.-H. & Rodriguez, P. M.. Reconstruction of equivalent
currents distribution over arbitrary three-dimensional surfaces based on
integral equation algorithms. IEEE Transaction on Antennas and
Propagation. 2007. 55(12): 3460–3468.
56. Quijano, J. L. A. & Vecchi, G.. Improved-accuracy source reconstruction on
arbitrary 3-D surfaces. IEEE Antennas and Wireless Propagation Letters.
2009. 8: 1046–1049.
57. Serhir, M., Besnier, P. & Drissi, M.. An accurate equivalent behavioral model
of antenna radiation using a mode-matching technique based on spherical
near field measurements. IEEE Transaction on Antennas and Propagation.
2008. 56(1): 48–57.
58. Leone, M.. Closed-form expressions for the electromagnetic radiation of
microstrip signal traces. IEEE Transaction on Electromagnetic Compatibility.
2007. 49(2): 322–328, 2007.
59. Leferink, F. B. J. & Van Dooren, M. J. C. M. “Inductance of printed circuit
board ground planes. IEEE International Symposium on Electromagnetic
Compatibility. August 9-13. Dallas, TX: IEEE press. 1993. 327– 329.
60. Hall, S. H., Hall, G. W. & McCall, J. A.. High-Speed Digital System Design
— A Handbook of Interconnect Theory and Design Practices. New York:
Wiey intersceince. 2000.
61. Paul, C. R.. A Comparison of the contributions of common-mode and
differential-mode currents in radiated emissions. IEEE Transaction on
Electromagnetic Compatibility. 1989. 31(2): 189–193.
62. Wu, T., Buesink, F. & Canavero, F.. Overview of signal integrity and EMC
design technologies on PCB: fundamentals and latest progress. IEEE
Transaction on Electromagnetic Compatibility. 2013. 55(4): 624–638.
63. Archambeault, B., Brench, C. & Connor, S.. Review of printed-circuit-board
Level EMI / EMC Issues and Tools. IEEE Transaction on Electromagnetic
Compatibility. 2010. 52(2): 455–461.
64. Caniggia, S., Catello, L., Costa, V., Marano, G., O’Riordan, N. & Vitucci, L..
Controlling radiated emission from a physically large telecommunication
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
192
system. IEEE Transaction on Electromagnetic Compatibility. 1996. 38(3):
290–299.
65. Hill, D. A., Camell, D. G., Cavcey, K. H. & Koepke, G. H.. Radiated
emissions and immunity of microstrip transmission lines: theory and
reverberation chamber measurements. IEEE Transaction on Electromagnetic
Compatibility. 1996. 38(2): 165–172.
66. Costa, V., Caniggia, S. & Catello, L.. On the computation of electromagnetic
Field components from a transmission line: theory and application. IEEE
International Symposium on Electromagnetic Compatibility. August 21-23.
Washington, DC: IEEE press. 1990. 651– 656.
67. Shi, R. S., Darcherif, A. & Sabonnadikre, J. C.. Computation of transient
electromagnetic fields radiated by a transmission line: An exact model. IEEE
Transaction on Magnetics. 1995. 31(4): 2423-2431.
68. Wong, B. W. -j. & Cantoni, A.. Modeling and analysis of radiated emissions
and signal integrity of capacitively loaded printed circuit board
interconnections. IEEE Transaction on Electromagnetic Compaibility. 2012.
54(5): 1087–1096.
69. Wong, B., & Cantoni, A.. Radiated emissions and signal integrity of printed
circuit board microstrips. Electromagnetic Compatibility Symposium-
Melbourne (EMC Melbourne). September 8-10. Melbourne, VIC: IEEE
press. 2010. 1– 6.
70. Hsieh, H.-C., Chiu, C.-N., Wang, C.-H. & Chen, C. H.. A New approach for
fast analysis of spurious emissions from RF/microwave Circuits. IEEE
Transaction on Electromagnetic Compatibility. 2009. 51(3): 631–638.
71. Li, P. & Jiang, L. J.. A rigorous approach for the radiated emission
characterization based on the spherical magnetic field scanning. IEEE
Transaction on Electromagnetic Compatibility. 2014. 56(3): 683–690.
72. Kuo, N., Chiu, C., Hsieh, H.-C. & Chen, J.-S.. Radiation minimization of
single-stub microstrip impedance transformers. IEEE Transaction on
Microwave theory and Techniques. 2013. 61(3): 1018–1029.
73. Duan, X., Archambeault, B., Bruens, H.-D. & Schuster, C.. EM emission of
differential signals across connected printed circuit boards in the GHz range.
2009 IEEE International Symposium on Electromagnetic Compatibility.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
193
August 17-21. Austin, TX: IEEE press. 2009. 50–55.
74. Xiao, F., Murano, K. & Kami, Y.. Analytical solution of the electromagnetic
radiation from coupled differential microstrip pairs. Asia-Pacific
International Symposium on Electromagnetic Compatibility. May 26-29.
Taipei: IEEE press. 2015. 708–711.
75. Rogard, E., Azanowsky, B. & Ney, M. M.. Comparison of radiation modeling
techniques up to 10 GHz — application on a microstrip PCB trace. IEEE
Transaction on Electromagnetic Compaibility. 2010. 52(2): 479–486.
76. Meng, J., Teo, Y. X., Thomas, D. W. P. & Christopoulos, C.. Fast prediction
of transmission line radiated emissions using the Hertzian dipole method and
line-end discontinuity models. IEEE Transaction on Electromagnetic
Compaibility. 2014. 56(6): 1295–1303.
77. Tseng, B.-C., Liao, L.-C., Wu, L.-K. & Lung, H.-T.. Analytical solutions for
the radiated emission of parallel microstrip traces. IEEE Transaction on
Electromagnetic Compaibility.2011. 53(3): 842–845.
78. Chiariello, A. G. & Miano, G.. An hybrid model for the evaluation of the
full-wave far-field radiated emission from PCB traces. Progress in
Electromagnetic Research (PIER). 2010. 101: 125–138.
79. Li, J., Y.-Zhang, Liu, D., Bhobe, A., Drewniak, J. L. & Fan, J.. Radiation
Physics from Two-Wire Transmission Lines. IEEE Symposium on
Electromagnetic Compatibility and Signal Integrity. March 15-21. Santa
Clara, CA: IEEE press. 2015. 160–164.
80. Cozza, A., Canavero, F. & Demoulin, B.. A Closed-form formulation for the
total power radiated by a single-wire overhead Line. 16th International
Symposium on Electromagnetic Compatibility. February 13-18. Zurich,
Switzerland: Politecnico Di Torino. 2005. 529-534.
81. Ho, C., Chen, K., Horng, T., Wu, J. & Huang, C.. Method of measuring
common-mode current conversion coefficient for estimating variation in
radiated emission from printed circuit board components. Radioengineering.
2014. 23(2): 709–716.
82. Sasabe, K., Bullivant, A., Yoshida, K. & Fujiwara, O.. Prediction of electric
far-field strength from printed circuit boards by measuring the common-mode
current. IEEE International Symposium on Electromagnetic Compatibility.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
194
August 21-25. Washington, DC: IEEE press. 2000. 379–384.
83. Wang, J., Sasabe, K. & Fujiwara, O.. A simple method for predicting
common-mode radiation from a cable attached to a conducting enclosure.
IEICE Transaction on Communication. 2002. E85-B(7): 1360–1367.
84. Park, H. H., Park, H. & Lee, H. S.. A Simple method of estimating the
radiated emission from a cable attached to a mobile device. IEEE Transaction
on Electromagnetic Compatibility. 2013. 55(2): 257–264.
85. Shim, H.-W. & Hubing, T. H.. Model for estimating radiated emissions from
a printed circuit board with attached cables driven by voltage driven sources.
IEEE Transaction on Electromagnetic Compatibility. 2005. 47(4): 899–907.
86. Deng, S., Hubing, T. & Beetner, D.. Estimating maximum radiated emissions
from printed circuit boards with an attached cable. IEEE Transaction on
Electromagnetic Compatibility. 2008. 50(1): 215–218.
87. Wang, J., Fujiwara, O. & Sasabe, K.. A simple method for predicting
common-mode radiation from a cable attached to a conducting enclosure.
Asia-Pacific International Symposium on Electromagnetic Compatibility.
December 3-6. Taipei: IEEE press. 2001. 1119–1122.
88. Su, C. & Hubing, T. H.. Improvements to a method for estimating the
maximum radiated emissions from PCBs with cables. IEEE Transaction on
Electromagnetic Compatibility. 2011. 53(4): 1087–1091.
89. Zhang, N., Kim, J., Ryu, S. & Nah, W.. Prediction of common-mode radiated
emission of PCB with an attached cable using imbalance difference model.
IEICE Transaction on Communication. 2015. E98-B(4): 638–645.
90. Drewniak, J. L., Hubing, T. H. & Van Doren, T. P.. Investigation of
fundamental mechanisms of common-mode radiation from printed circuit
boards with attached cables. IEEE International Symposium on
Electromagnetic Compatibility. August 22-26. Chicago, IL: IEEE press.
1994. 110–115.
91. Shim, H. W. & Hubing, T. H.. Derivation of a closed-form approximate
expression for the self-capacitance of a printed circuit board trace. IEEE
Transaction on Electromagnetic Compatibility. 2005. 47(4): 1004–1008.
92. Watanabe, T., Wada, O., Miyashita, T. & Koga, R.. Common mode current
generation caused by difference of unbalance of transmission lines on a
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
195
printed circuit board with narrow ground pattern. IEICE Transaction on
Communication. 2000. E83-B(3): 593–599.
93. Watanabe, T., Wada, O.,Yoshitaka, T. & Koga, R.. Estimation of common-
mode EMI caused by a signal line in the vicinity of ground edge on a PCB.
IEEE International Symposium on Electromagnetic Compatibility. August
19-23. Minneapolis: IEEE press. 2002. 113–118.
94. Holloway, C. L., Kuester, E. F., Ruehli, A. E. & Antonini, G.. Partial and
internal inductance : two of Clayton R . Paul’s many passions. IEEE
Transaction on Electromagnetic Compatibility.2013. 55(4): 600–613.
95. Holloway, C. L. & Kuester, E. F.. Net and partial inductance of a microstrip
ground plane. IEEE Transaction on Electromagnetic Compatibility. 1998.
40(1): 33–46.
96. Hockanson, D. M., Drewniak, J. L., Hubing, T. H.,Van Doren, T. P., Sha, F.
Lam, C. & Rubin, L.. Quantifying EMI resulting from finite-impedance
reference planes. IEEE Transaction on Electromagnetic Compatibility.1997.
39(4): 286–297.
97. Watanabe, T., Kishimoto, M., Matsunaga, S., Tanimoto, T., Koga, R., Wada,
O. & Namba, A.. Equivalence of two calculation methods for common-mode
excitation on a printed circuit board with narrow ground plane. IEEE
Symposium on Electromagnetic Compatibility. August 18-22. Istanbul. IEEE
press: 2003. 22–27.
98. Sayegh, A. & Mohd Jenu, M. Z.. Evaluation of the radiated emission of a
printed circuit board attached with cables. 3rd International Conference on
Electric and Electronics. December 24-25. Hong Kong. Atlantis press: 2013.
195–198.
99. German, R. F., Ott, H. W. & Paul, C. R.. Effect of an image plane on printed
circuit board radiation. IEEE International Symposium on Electromagnetic
Compatibility. August 21-23. Washington, DC. IEEE press: 1990. 284–291.
8100. Zhang, Z. & Gong, F.. Prediction and analyze of PCB common-mode
radiation based on current-driven mode. Workshop on Power Electronics and
Intelligent Transportation System. August 2-3. Guangzhou. IEEE press:
2008. 320-323.
101. Leone, M.. Design expressions for the trace-to-edge common-mode
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
196
inductance of a printed circuit board. IEEE Transaction on Electromagnetic
Compatibility. 2001. 43(4): 667–671.
102. Su, C. & Hubing, T. H.. Imbalance difference model for common-mode
radiation from printed circuit boards. IEEE Transaction on Electromagnetic
Compatibility. 2011. 53(1): 150–156.
103. Freeman, J. A. & Skapura, D. M.. Neural Networks Algorithms, Applications,
and Programming Techniques. Boston: Addison-Wesley Pub. 1991.
104. Dong, X., Weng, H., Beetner, D. G., Hubing, T. H., Wunsch, D. C., Noll, M.
Göksu, H. & Moss, B.. Detection and identification of vehicles based on their
unintended electromagnetic emissions. IEEE Transaction on Electromagnetic
Compatibility. 2006. 48(4): 752–758.
105. Tsai, C. Y., Rothwell, E. J. & Chen, K. M.. Target discrimination using
neural networks with time domain or spectrum magnitude response. Journal
of Electromagnetic Waves and Applications. 1996. 10(3): 341–382.
106. Koroglu, S., Umurkan, N., Kilic, O. & Attar, F.. An approach to the
calculation of multilayer magnetic shielding using artificial neural network.
Simulation Modelling Practice and Theory. 2009. 17(7): 1267–1275.
107. McCormick, G., Khan, Z. A., Devabhaktuni, V., Alam, M. & Wood, A..
Estimating radiated emissions from printed circuit boards and cables inside
EMC chambers. IEEE International Symposium on Electromagnetic
Compatibility. July 25-30. Fort Lauderdale. IEEE press: 2010. 36–39.
108. Aunchaleevarapan, K., Paithoonwatanakij, K., Khan-Ngern, W. & Nitta, S..
Novel method for predicting PCB configurations for near-field and far-field
radiated EMI using a neural network. IEICE Transaction on Communication.
2003. E86-B(4): 1364–1375.
109. Chahine, I., Kadi, M., Gaboriaud, E., Louis, A., & Mazari, B..
Characterization and modeling of the susceptibility of integrated circuits to
conducted electromagnetic disturbances up to 1 GHz. IEEE Transaction on
Electromagnetic Compatibility. 2008. 50(2): 285–293.
110. Dangkham, P., Sujintanarat, P., Chaichana, S., Aunchaleevarapan, K. &
Teekaput, P.. Recognition and identification of radiated EMI for shielding
aperture using neural network. PIERS Online. 2007. 3(4): 225–229.
111. Kozyrev, S. V.. Classification by ensembles of neural networks. P-Adic
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
197
Numbers Ultrametric Analysis and Applications. 2012. 4(1): 27-33.
112. Kvale, D. T. Artificial Neural Networks Based Approaches for modelling
Radiated Emisssions from Printed Circuit Board Structures and Shields.
Master Thesis. University of Toledo; 2010.
113. Firmino, L. , Raizer, A. & Marechal, Y.. Estimating far-field emissions from
simulated near-field data with artificial neural networks. IEEE Transaction
on Magnetics. 2014. 50(2): 7–8.
114. Y. Diao, W. Sun, S. W. Leung, K. H. Chan, and Y. M. Siu, “Prediction of
magnetic field emissions by current source reconstruction using radial basis
function network,” Electron. Lett., vol. 51, no. 16, pp. 1243–1245, 2015.
115. Regué, J. R., Ribó, M., Garrell, J. M. & Martín, A.. A genetic algorithm
based method for source identification and far-field radiated emissions
prediction from near-field measurements for PCB characterization. IEEE
Transaction on Electromagnetic Compatibility. 2001. 43(4): 520–530.
116. Wilson, P. F., Hill, D. A. & Holloway, C. L.. On determining the maximum
emissions from electrically large sources. IEEE Transaction on
Electromagnetic Compatibility. 2002. 44(1): 79–86.
117. Chen, I.-F., Peng, C.-M. & Hsue, C.-W.. Circuit-concept approach to
radiated emissions of printed circuit boards. IEE Proceedings - Science,
Measurement and Technology. 2004. 151(3): 205–210.
118. Yuancheng, Ji. Modeling and Measurement of Electromagnetic radiated
Emission from High Speed Interconnects Digital Circuits. Ph.D. Thesis.
National University of Singapore; 2014.
119. Drewniak, J. L., Sha, F., Van Doren, T. P., Hubing, T. H. & Shaw, J..
Diagnosing and modeling common-mode radiation from printed circuit
boards with attached cables. IEEE International Symposium on
Electromagnetic Compatibility. August 14-18. Atlanta, GA. IEEE press:
1995. 465–470.
120. Al-rubaye, H., Kuganesan, P. & Kane, M.. Estimating radiated emissions
from microstrip transmission lines based on the imbalance model. IEEE
Electrical Design of Advanced Packaging and Systems Symposium.
December 9-11. Taipei. IEEE press: 2012. 93–96.
121. Takahashi, T., Niu, L. & Hubing, T.. Estimation of common mode current on
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
198
coaxial cable with twisted wire pair. International Symposium on
Electromagnetic Compatibility (EMC'14/Toky). May 12-16. Tokyo. IEEE
press: 2014. 553–556.
122. Luo, M. & Huang, K.. Prediction of the electromagnetic field in metallic
Enclosures using artificial neural networks. Progress of Electromagnetic
Research. 2011. 116: 171–184.
123. Zhao, W. J., Park, H. B., Tan, M., Park, H. H., Liu, E.-X., Song, E. & Li, E.-
P.. Far-field prediction from amplitude-only near-field measurements using
equivalent electric currents. IEEE International Symposium on
Electromagnetic Compatibility. August 6-10. Pittsburgh. IEEE press: 2012.
590–593.
124. Gao, X., Fan, J., Zhang, Y., Kajbaf, H. & Pommerenke, D.. Far-field
prediction using only magnetic near-field scanning for EMI test. IEEE
Transaction on Electromagnetic Compatibility. 2014. 56(6): 1335–1343.
125. Thomas, D. W. P., Obiekezie, C. & Tong, X.. Equivalent dipole models of
electromagnetic emissions from near-field scanning. IEEE Electromagnetic
Compatibility Magazine. 2015. 4(3): 74–78.
126. Ramo, S., Whinnery, J. & Van Duzer, T.. Fields and Waves in
Communication Electronics. 3rd edition. New York. John Wiley & Sons.
1994.
127. Hoffmann, R. K.. Handbook of Microwave Integrated Circuits. Norwood.
Artech House. 1987.
128. King, R. & Harrison, C. W.. The distribution of current along a symmetrical
center-driven Antenna. Proceedings of the IRE Journal. 1943. 31(10): 548–
56.
129. King, R.. Asymmetrically driven antennas and sleeve dipole. Proceedings of
the IRE Journal. 1950. 38(10): 1154–1164.
130. Schelkunoff, S. A. & Friis, H. T.. Antennas: Theory and Practice. New
York: John Wiley & Sons. 1951.
131. Wefky, A. M., Espinosa, F., De Santiago, L., Gardel, A., Revenga, P. &
Martinez, M.. Modeling radiated electromagnetic emissions of electric
motorcycles in terms of driving profile using MLP neural networks. Progress
of Electromagnetic Research. 2012. 135: 231–244.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH
199
132. Hagan, M., Demuth, H. B., Beale, M. H. & De Jesús, O.. Neural Network
Design. 2nd edition. Boston: Martin Hagan. 2014.
133. Bishop, C. M.. Neural Networks for Pattern Recognition. Oxford,UK:
Clarendon. 1996.
134. Haykin, S.. Neural Networks: a comprehensive foundation. 2nd edition.
Hamilton, Ontario: Pearson Education. 1999.
135. Sengupta, D. L. & Liepa, V. V.. Applied Electromagnetics and
Electromagnetic Compatibility. New Jersey: Wiley intersceince. 2006.
136. Kwak, H. & Hubing, T. H.. Investigation of the imbalance difference model
and its application to various circuit board and cable geometries. IEEE
International Symposium on Electromagnetic Compatibility. August 6-10.
Pittsburgh. IEEE press: 2012. 273–278.
PTTAPERP
USTAKAAN TUNKU T
UN AMINAH