PREFACE

PREFACE

The study of electromagnetic compatibility (EMC) is a young, old science. It is relatively old because the problem of radio frequency interference (RFI) arose nearly 100 years ago with the first use of radio waves as a communication medium. However, it is only in the last 20 to 25 years that the progress in numerical computation has allowed the scientist and engineer not only to propose models for the physical phenomena underlying this interference, but also to use these models to better understand and visualize these phenomena, and to mitigate the effects of the interference.

The development and use of models has been the focus of much human activity. To quote the late Prof. Peter Johns, developer of the TLM model,

"Throughout history men have been making models of the physical events they observe. From cave pictures to surrealist art, from models of the atom to models of the universe, all have sought to focus attention on particular ideas for analysis or for communication to other people. As engineers we must be particularly skilled in the art of making or choosing the mathematical models of our engineering concepts. Desirable properties of a model, such are the enhancement of areas for which analysis is required and rejection of areas of no interest, seem obvious. However the computational power at the fingertips of the professional engineer is increasing enormously through the proliferation of calculators and computers Thus the methods for analyzing models are changing, and this means that the models we have chosen in the past may not always be the best for the present and future." Although numerical models can be very useful in our understanding of electrical phenomena, they are inherently limited in their ability to predict as much as we would like. As pointed out by Prof. Johns, sometimes the fundamental assumptions going into a mathematical foundation of a model are themselves approximations to reality. As a consequence, the model based on these approximations may have a limited range of validity. In other instances, the situation being modeled (i.e., the "real world") can be so complex that accurate modeling becomes difficult, if not impossible, to carry out. In such cases, one often resorts to measurement or experimental tests to understand the phenomenon.

Is it then a lack of modesty to present to the scientific and technical community a book on EMC modeling?

We think not, for several reasons. The first is that an incredible progress in electromagnetic field modeling has already been achieved, in particular, over the last 15 years. We think that the present state of the art in modeling represents a good basis for further progress in this field, and that such progress is likely to be very rapid in the future.

The second reason is that, while present models may be far from perfect, they are very useful in understanding of the fundamental principles of EM interference control. Much can be learned from an imperfect model if the user takes the effort to understand why the model does not work as it should . This understanding leads naturally to new, and hopefully improved, models.

In writing this book, we believe that graduate students, post doctoral researchers and senior researchers, as well as electrical engineers engaged in the research and development work for practical applications in the EMI area, will benefit in having at their disposal material which presently is only found dispersed throughout journals, technical reports or conference papers. Much of the material in this book is in constant evolution, due to research work in progress at universities and laboratories throughout the world. During the development of this book, however, we were obliged to present the status of this subject of EMC modeling at certain fixed point in time, which was when this book (or even when each chapter) was written. A typical example is the model of the transfer impedance of braided cables. This model, far from being perfect, does not yet reflect all of the complexity of the real braid shield. It does represent, however, a big step in our understanding of what happens in a braided shield, and our effort has been aimed at giving the reader an adequate picture of the important work conducted in this area during the last few years.

The material in this book is divided into five parts: Part 1 - Preliminaries, Part 2 - Low Frequency Circuit Models, Part 3 - High Frequency and Broad Band Coupling Models, Part 4 - Transmission Line Models and Part 5 - Shielding Models. Part 1 presents the customary introduction to the subject matter treated in this book: EMC analysis methods and computational models. Chapter 1 reviews the overall concept of model development and discusses the impact that modeling has had in the area of EMC. Some of the different types of signals, both transient and continuous wave (CW) are reviewed, and the link between the transient domain and the frequency domain is explained by the Fourier transform.

A key aspect of model development is how to take an electrically complex system and break it up into smaller, more manageable pieces for which an EM model can be developed. This is done through the concept of electromagnetic topology, which is the topic of Chapter 2. As discussed in this chapter, the word topology as used in this book is not cast a rigorous mathematical context, but rather, it provides a conceptual tool in trying to view the system as if the observer were an electromagnetic wave impinging on an electrical system — such as an aircraft. Where are the global shielding surfaces that keep you from entering into the interior of the system? Where are the points of entry into the system that allow you to pass? How do you move from one region to another within the system? What effect do you have on components within the system? The answers to each of these questions are related to viewing the system in a topological manner.

Part 2 of this book consists of a Chapter 3 on lumped-parameter circuit models. At sufficiently low frequencies, interference between two electrical circuits can be adequately described by low frequency circuit models. In this chapter, conducted interference models are first introduced with a discussion of Thévenin and Norton equivalent circuits. This concept is then generalized to the case of active and passive two-port networks. In addition to the direct wire connections between circuits, capacitive and inductive field coupling can also be important in low frequency models. These coupling mechanisms are discussed in this chapter, as is the case of galvanic coupling between two circuits connected to a common, lossy conductor.

In Part 3 of the text, models suitable for higher frequencies are discussed. As the dimensions of coupled circuits begin to approach the size of the wavelength of the EM fields, the circuit models are no longer adequate and the models must take into account the wave nature of the fields. Using Maxwell’s equations, these models lead to the concept of radiation. Chapter 4 presents an overview of the radiation process - first from elementary electric and magnetic dipole sources, and later from extended line sources (i.e., from wire antennas). Several examples of electric dipole radiation in the presence of perturbing bodies (a ground plane, a sphere, within parallel plates, and in a cavity) are given. A key aspect in the analysis of wire antennas is the knowledge of the current distribution on the structure. This distribution can be estimated, approximated from the integral equation describing the current, or computed numerically using the method of moments. One topic of current interest is the singularity expansion method (SEM) which provides a link between the resonant behavior of R-L-C circuits and antennas.

Chapter 5 concludes Part 3 with a discussion of radiation, diffraction and scattering models for apertures, which cannot be analyzed using the relatively simple one dimensional wire models of the previous chapter. In this chapter, the scalar diffraction theory for apertures is first discussed, followed by the more rigorous vector field diffraction. This theory is also applied to radiating antennas. A case of special importance, especially in braid shield modeling, is when an aperture is small compared to the wavelength. In this case, the fields penetrating thorough the aperture can be modeled as arising from equivalent electric and magnetic dipole moments located at the aperture, with the strengths of the dipoles being related to the size and shape of the aperture through aperture polarizabilities.

Part 4 of the book, dealing with transmission line models, contains three chapters. Chapter 6, entitled Transmission Line Theory lays the foundation for the discussions to come. The concept of distributed (per-unit-length) line parameters are introduced, and their use in the Telegrapher’s equations for the line current and voltage are discussed. Using the two conductor line as an example, the solution to these equations is developed, first for simple traveling waves on the line, and later, for lumped voltage or current sources. A particularly simple representation for the load responses of a transmission line is the so-called BLT equation, which is examined in this chapter. In addition, the solution of transmission line problems directly in the time domain is discussed. This chapter concludes with a detailed discussion of the calculation of the per-unit-length line inductance and capacitances.

Chapter 7 continues the discussion of transmission line modeling by examining the issues of EM field excitation of lines. It is pointed out that transmission line coupling models provide only part of the complete solution – the differential mode response, with the common mode response (i.e., the antenna mode) being neglected. Fortunately, for many practical cases, including lines over a conducting ground plane, the transmission line model provides an accurate way of computing the induced responses. Also in this chapter, models for highly resonant transmission lines, radiation from lines, transmission line networks and lines with nonlinear loads are formulated and discussed.

The important case of transmission lines in the presence of a lossy earth is developed in Chapter 8. The derivation of the Telegrapher’s equations for this case is presented and the determination of the per-unit-length line parameters for above-ground and buried lines is presented. Furthermore, the behavior of the incident, reflected and transmitted EM fields from the lossy earth is discussed. Several examples of cable responses due to distributed EM field excitation are provided.

The last part of the book, Part 5, discusses models for shielding. Continuing with transmission line issues, Chapter 9 presents information on the protection of cables through the use of solid or braided coaxial shields. The classical shielding formula for a solid shield is reviewed, and the more recent work for modeling the behavior of braided cables is outlined. In addition, the issue of discrete breaks in a shield (due to connectors or pig-tail connections) are discusses and appropriate computational models are suggested.

Chapter 10 investigates the more general aspects of EM shielding for enclosures, such as screen rooms. The usual "infinite slab" shielding model is briefly discussed, and it is pointed out that this provides an unrealistic view of the shielding provided by real shields. More appropriate models are those which have a finite shielded volume, bounded by a finite surface area – the so called volumetric shield. In this chapter a number of useful formulas for shielding are presented and illustrated.

This book concludes with appendices which present information on appropriate physical constants useful for model development, data for various types of transmission lines and pertinent vector identities. Also included is general information about four computer programs which accompany this book. These programs are designed based on the models developed in the text, and may be used to calculate various transmission line responses.

At the end of each chapter, several exercises are suggested to further the understanding of the basic theory behind the models, as well as to indicate how some of the models can be applied to practical situations of EMI prediction and control. Several of these exercises utilize the computer programs provided with the text.

Writing this book has been an interesting adventure for us. The material presented here results from many years of collective work, not only by ourselves, but by many of our colleagues. It is truly impressive to consider the past efforts on the part of researchers throughout the world that have gone into the formulation, development and use of computational models for electromagnetics. We see that much has been accomplished; however, even more remains to be done. In this book, we have only scratched the surface of this immense technical area.

January, 1996

F. M. Tesche, Dallas, TX USA

M. Ianoz, Lausanne, Switzerland

T. Karlsson, Linköping, Sweden