Electromagnetic Compatibility Engineering

PREFACE

Electromagnetic Compatibility Engineering started out being a third edition to my previous book Noise Reduction Techniques in Electronic Systems, but it turned out to be much more than that, hence, the title change. Nine of the original twelve chapters were completely rewritten. In addition, there are six new chapters, plus two new appendices, with over 600 pages of new and revised material (including 342 new figures). Most of the new material relates to the practical application of the theory of electromagnetic compatibility (EMC) engineering, and it is based on experience gained from my EMC consulting work, and teaching of EMC training seminars over the last 20 plus years.

Some of the more difficult and frustrating problems faced by design engineers concerns electromagnetic compatibility and regulatory compliance issues. Most engineers are not well equipped to handle these problems because the subject is not normally taught in engineering schools. Solutions to EMC problems are often found by trial and error with little or no understanding of the theory involved. Such efforts are very time consuming, and the solutions are often unsatisfactory. This situation is unfortunate, because most of the principles involved are simple and can be explained by elementary physics. This book is intended to remedy that situation.

This book is intended primarily for the practicing engineer who is involved in the design of electronic equipment or systems and is faced with EMC and regulatory compliance issues. It addresses the practical aspects of electromagnetic compatibility engineering, covering both emission and immunity. The concepts presented in this book are applicable to both analog and digital circuits operating from below audio frequencies up to the GHz range. Emphasis is on cost-effective EMC designs, with the amount and complexity of the mathematics kept to a minimum. The reader should obtain the knowledge necessary to design electronic equipment that is compatible with the electromagnetic environment and compliant with national and international EMC regulations.

The book is written in such a way that it can easily be used as a textbook for teaching a senior level or continuing education course in electromagnetic compatibility. To this end, the book contains 251 problems for the student to work out, the answers to which are included in Appendix F.

The book is divided into two parts: Part 1, EMC Theory and includes Chapters 1 to 10. Part 2, EMC Applications, includes Chapters 11 to 18. In addition, the book contains six appendices with supplemental information.

The organization of the material is as follows. Chapter 1 is an introduction to electromagnetic compatibility and covers national and international EMC regulations, including the European Union, FCC, and U.S. Military. Chapter 2 covers both electric and magnetic field cable coupling and crosstalk, as well as cable shielding and grounding. Chapter 3 covers safety, power, signal, and hardware/systems grounding.

Chapter 4 discusses balancing and filtering as well as differential amplifiers, and low-frequency analog circuit decoupling. Chapter 5 is on passive components and covers the nonideal characteristics of components that affect their performance. In addition to resistors, capacitors, and inductors—ferrite beads, conductors and transmission lines are also included. Chapter 6 is a detailed analysis of the shielding effectiveness of metallic sheets as well as conductive coatings on plastic, and the effect of apertures on the shielding effectiveness.

Chapter 7 covers contact protection for relays and switches. Chapters 8 and 9 discuss internal noise sources in components and active devices. Chapter 8 covers intrinsic noise sources, such as thermal and shot noise. Chapter 9 covers noise sources in active devices.

Chapters 10, 11, and 12 cover electromagnetic compatibility issues associated with digital circuits. Chapter 10 examines digital circuit grounding, including ground plane impedance and a discussion on how digital logic currents flow. Chapter 11 is on digital circuit power distribution and decoupling, and Chapter 12 covers digital circuit radiation mechanisms, both common mode and differential mode.

Chapter 13 covers conducted emissions on alternating current (ac) and direct current (dc) power lines, as well as EMC issues associated with switching power supplies and variable-speed motor drives. Chapter 14 covers radio frequency(rf) and transient immunity, as well as a discussion of the electromagnetic environment. Chapter 15 covers electrostatic discharge protection in the design of electronic products. It focuses on the importance of a three-prong approach, which includes mechanical, electrical, and software design.

Chapter 16 covers printed circuit board layout and stackup, a subject not often discussed. Chapter 17 addresses the difficult problem of partitioning, grounding, and layout of mixed-signal printed circuit boards.

The final chapter (Chapter 18) is on precompliance EMC measurements, that is, measurements that can be performed in the product development laboratory, using simple and inexpensive test equipment, which relate to the EMC performance of the product.

At the end of each chapter, there is a summary of the most important points discussed as well as many problems for the reader to work out. For those desiring additional information on the subjects covered, each chapter has an extensive reference, and further reading section.

Supplemental information is provided in six appendices. Appendix A is on the decibel. Appendix B covers the 10 best ways to maximize the emission from your product. Appendix C derives the equations for multiple reflections of magnetic fields in thin shields.

Appendix D, Dipoles for Dummies, is a simple, insightful, and intuitive discussion of how a dipole antenna works. If a product picks up or radiates electromagnetic energy, then it is an antenna, therefore, an understanding of some basic antenna theory would be helpful for all engineers, especially EMC engineers.

Appendix E explains the important, and not well understood, theory of partial inductance, and Appendix F provides answers to the problems contained at the end of each chapter.

I would like to express my gratitude and appreciation to all those who took the time to comment on Noise Reduction Techniques in Electronic Systems and to all those who encouraged me to write Electromagnetic Compatibility Engineering. In particular, I would especially like to thank John Celli, Bob German, Dr. Clayton Paul, Mark Steffka, and Jim Brown for their insightful review of major portions of the manuscript, as well as for their encouragement and the many fruitful discussions we had on the subject of EMC. Electromagnetic Compatibility Engineering is a better book because of them.

Portions of the manuscript were also used for an electromagnetic compatibility class taught by Mark Steffka at the University of Michigan–Dearborn, during the 2007 and 2008 semesters. My heartfelt thanks go out to the students in those two classes for the large number of comments and suggestions that I received (many of which have been incorporated into this book), in particular their suggestions for additional problems to be included in the book. I would also like to express my appreciation to James Styles who, Mark Steffka and I both agreed, submitted the most useful comments.

Finally, I would like to thank all my colleagues who took the time to review various portions of this manuscript and make useful comments and suggestions.

Additional technical information, updated information on EMC regulations, as well as an errata sheet for this book are on the Henry Ott Consultants website at www.hottconsultants.com.

HENRY W. OTT

Livingston, New Jersey

January 2009

PART I

EMC Theory

1

Electromagnetic Compatibility

1.1 INTRODUCTION

The widespread use of electronic circuits for communication, computation, automation, and other purposes makes it necessary for diverse circuits to operate in close proximity to each other. All too often, these circuits affect each other adversely. Electromagnetic interference (EMI) has become a major problem for circuit designers, and it is likely to become even more severe in the future. The large number of electronic devices in common use is partly responsible for this trend. In addition, the use of integrated circuits and large-scale integration has reduced the size of electronic equipment. As circuitry has become smaller and more sophisticated, more circuits are being crowded into less space, which increases the probability of interference. In addition, clock frequencies have increased dramatically over the years—in many cases to over a gigahertz. It is not uncommon today for personal computers used in the home to have clock speeds in excess of 1 GHz.

Today’s equipment designers need to do more than just make their systems operate under ideal conditions in the laboratory. Besides that obvious task, products must be designed to work in the real world, with other equipment nearby, and to comply with government electromagnetic compatibility (EMC) regulations. This means that the equipment should not be affected by external electromagnetic sources and should not itself be a source of electromagnetic noise that can pollute the environment. Electromagnetic compatibility should be a major design objective.

1.2 NOISE AND INTERFERENCE

Noise is any electrical signal present in a circuit other than the desired signal. This definition excludes the distortion products produced in a circuit due to nonlinearities. Although these distortion products may be undesirable, they are not considered noise unless they are coupled into another part of the circuit. It follows that a desired signal in one part of a circuit can be considered to be noise when coupled to some other part of the circuit.

Noise sources can be grouped into the following three categories: (1) intrinsic noise sources that arise from random fluctuations within physical systems, such as thermal and shot noise; (2) man-made noise sources, such as motors, switches, computers, digital electronics, and radio transmitters; and (3) noise caused by natural disturbances, such as lightning and sunspots.

Interference is the undesirable effect of noise. If a noise voltage causes improper operation of a circuit, it is interference. Noise cannot be eliminated, but interference can. Noise can only be reduced in magnitude, until it no longer causes interference.

1.3 DESIGNING FOR ELECTROMAGNETIC COMPATIBILITY

Electromagnetic compatibility (EMC) is the ability of an electronic system to (1) function properly in its intended electromagnetic environment and (2) not be a source of pollution to that electromagnetic environment. The electromagnetic environment is composed of both radiated and conducted energy. EMC therefore has two aspects, emission and susceptibility.

Susceptibility is the capability of a device or circuit to respond to unwanted electromagnetic energy (i.e., noise). The opposite of susceptibility is immunity. The immunity level of a circuit or device is the electromagnetic environment in which the equipment can operate satisfactorily, without degradation, and with a defined margin of safety. One difficulty in determining immunity (or susceptibility) levels is defining what constitutes performance degradation.

Emission pertains to the interference-causing potential of a product. The purpose of controlling emissions is to limit the electromagnetic energy emitted and thereby to control the electromagnetic environment in which other products must operate. Controlling the emission from one product may eliminate an interference problem for many other products. Therefore, it is desirable to control emission in an attempt to produce an electromagnetically compatible environment.

To some extent, susceptibility is self-regulating. If a product is susceptible to the electromagnetic environment, the user will become aware of it and may not continue to purchase that product. Emission, however, tends not to be self-regulating. A product that is the source of emission may not itself be affected by that emission. To guarantee that EMC is a consideration in the design of all electronic products, various government agencies and regulatory bodies have imposed EMC regulations that a product must meet before it can be marketed. These regulations control allowable emissions and in some cases define the degree of immunity required.

EMC engineering can be approached in either of two ways: one is the crisis approach, and the other is the systems approach. In the crisis approach, the designer proceeds with a total disregard of EMC until the functional design is finished, and testing—or worse yet—field experience suggests that a problem exists. Solutions implemented at this late stage are usually expensive and consist of undesirable add ons. This is often referred to as the Band Aid approach.

As equipment development progresses from design to testing to production, the variety of noise mitigation techniques available to the designer decreases steadily. Concurrently, cost goes up. These trends are shown in Fig. 1-1. Early solutions to interference problems, therefore, are usually the best and least expensive.

The systems approach considers EMC throughout the design; the designer anticipates EMC problems at the beginning of the design process, finds the remaining problems in the breadboard and early prototype stages, and tests the final prototypes for EMC as thoroughly as possible. This way, EMC becomes an integral part of the electrical, mechanical, and in some cases, software/firmware design of the product. As a result, EMC is designed into— and not added onto—the product. This approach is the most desirable and cost effective.

If EMC and noise suppression are considered for one stage or subsystem at a time, when the equipment is initially being designed, the required mitigation techniques are usually simple and straightforward. Experience has shown that when EMC is handled this way, the designer should be able to produce equipment with 90% or more of the potential problems eliminated prior to initial testing.

A system designed with complete disregard for EMC will almost always have problems when testing begins. Analysis at that time, to find which of the many possible noise path combinations are contributing to the problem, may not be simple or obvious. Solutions at this late stage usually involve the addition of extra components that are not integral parts of the circuit. Penalties paid include the added engineering and testing costs, as well as the cost of the mitigation components and their installation. There also may be size, weight, and power dissipation penalties.

FIGURE 1-1. As equipment development proceeds, the number of available noise-reduction techniques goes down. At the same time, the cost of noise reduction goes up.

1.4 ENGINEERING DOCUMENTATION AND EMC

As the reader will discover, much of the information that is important for electromagnetic compatibility is not conveyed conveniently by the standard methods of engineering documentation, such as schematics, and so on. For example, a ground symbol on a schematic is far from adequate to describe where and how that point should be connected. Many EMC problems involve parasitics, which are not shown on our drawings. Also, the components shown on our engineering drawings have remarkably ideal characteristics.

The transmission of the standard engineering documentation alone is therefore insufficient. Good EMC design requires cooperation and discussion among the complete design team, the systems engineer, the electrical engineer, the mechanical engineer, the EMC engineer, the software/firmware designer, and the printed circuit board designer.

In addition, many computer-assisted design (CAD) tools do not include sufficient, if any, EMC considerations. EMC considerations therefore must often be applied manually by overriding the CAD system. Also, you and your printed circuit designer often have different objectives. Your objective is, or should be, to design a system that works properly and meets EMC requirements. Your printed circuit board (PCB) designer has the objective of doing what ever has to be done to fit all the components and traces on the board regardless of the EMC implications.

1.5 UNITED STATES’ EMC REGULATIONS

Added insight into the problem of interference, as well as the obligations of equipment designers, manufacturers, and users of electronic products, can be gained from a review of some of the more important commercial and military EMC regulations and specifications.

The most important fact to remember about EMC regulations is that they are living documents and are constantly being changed. Therefore, a 1-year-old version of a standard or regulation may no longer be applicable. When working on a new design project, always be sure to have copies of the most recent versions of the applicable regulations. These standards may actually even change during the time it takes to design the product.

1.5.1 FCC Regulations

In the United States, the Federal Communications Commission (FCC) regulates the use of radio and wire communications. Part of its responsibility concerns the control of interference. Three sections of the FCC Rules and Regulations* have requirements that are applicable to nonlicensed electronic equipment. These requirements are contained in Part 15 for radio frequency devices; Part 18 for industrial, scientific, and medical (ISM) equipment; and Part 68 for terminal equipment connected to the telephone network.

Part 15 of the FCC Rules and Regulations sets forth technical standards and operational requirements for radio frequency devices. A radio-frequency device is any device that in its operation is capable of emitting radio-frequency energy by radiation, conduction, or other means (§ 2.801). The radio-frequency energy may be emitted intentionally or unintentionally. Radio-frequency (rf) energy is defined by the FCC as any electromagnetic energy in the frequency range of 9 kHz to 3000 GHz (§15.3(u)). The Part 15 regulations have a twofold purpose: (1) to provide for the operation of low-power transmitters without a radio station license and (2) to control interference to authorized radio communications services that may be caused by equipment that emits radio-frequency energy or noise as a by-product to its operation. Digital electronics fall into the latter category.

Part 15 is organized into six parts. Subpart A—General, Subpart B—Unintentional Radiators, Subpart C—Intentional Radiators, Subpart D—Unlicensed Personal Communications Devices, Subpart E—Unlicensed National Information Infrastructure Devices, and Subpart F—Ultra-Wide-band Operation. Subpart B contains the EMC Regulations for electronic devices that are not intentional radiators.

Part 18 of the FCC Rules and Regulations sets forth technical standards and operational conditions for ISM equipment. ISM equipment is defined as any device that uses radio waves for industrial, scientific, medical, or other purposes (including the transfer of energy by radio) and that is neither used nor intended to be used for radio communications. Included are medical diathermy equipment, industrial heating equipment, rf welders, rf lighting devices, devices that use radio waves to produce physical changes in matter, and other similar non-communications devices.

Part 68 of the FCC Rules and Regulations provides uniform standards for the protection of the telephone network from harm caused by connection of terminal equipment [including private branch exchange (PBX) systems] and its wiring, and for the compatibility of hearing aids and telephones to ensure that persons with hearing aids have reasonable access to the telephone network. Harm to the telephone network includes electrical hazards to telephone company workers, damage to telephone company equipment, malfunction of telephone company billing equipment, and degradation of service to persons other than the user of the terminal equipment, his calling or called party.

In December 2002, the FCC released a Report and Order (Docket 99-216) privatizing most of Part 68, with the exception of the requirements on hearing aid compatibility. Section 68.602 of the FCC rules authorized the Telecommunications Industry Association (TIA) to establish the Administrative Council for Terminal Attachments (ACTA) with the responsibility of defining and publishing technical criteria for terminal equipment connected to the U.S. public telephone network. These requirements are now defined in TIA-968. The legal requirement for all terminal equipment to comply with the technical standards, however, remains within Part 68 of the FCC rules. Part 68 requires that terminal equipment connected directly to the public switched telephone network meet both the criteria of Part 68 and the technical criteria published by ACTA.

Two approval processes are available to the manufacturer of telecommunications terminal equipment, as follows: (1) The manufacturer can provide a Declaration of Conformity (§68.320) and submit it to ACTA, or (2) the manufacturer can have the equipment certified by a Telecommunications Certifying Body (TCB) designated by the Commission (§68.160). The TCB must be accredited by the National Institute of Standards and Technology (NIST).

1.5.2 FCC Part 15, Subpart B

The FCC rule with the most general applicability is Part 15, Subpart B because it applies to virtually all digital electronics. In September 1979, the FCC adopted regulations to control the interference potential of digital electronics (at that time called computing devices). These regulations, Technical Standards for Computing Equipment (Docket 20780); amended Part 15 of the FCC rules relating to restricted radiation devices. The regulations are now contained in Part 15, Subpart B of Title 47 of the Code of Federal Regulations. Under these rules, limits were placed on the maximum allowable radiated emission and on the maximum allowable conducted emission on the alternating current (ac) power line. These regulations were the result of increasing complaints to the FCC about interference to radio and television reception where digital electronics were identified as the source of the interference. In this ruling the FCC stated the following:

Computers have been reported to cause interference to almost all radio services, particularly those services below 200 MHz,* including police, aeronautical, and broadcast services. Several factors contributing to this include: (1) digital equipment has become more prolific throughout our society and are now being sold for use in the home; (2) technology has increased the speed of computers to the point where the computer designer is now working with radio frequency and electromagnetic interference (EMI) problems—something he didn’t have to contend with 15 years ago; (3) modern production economics has replaced the steel cabinets which shield or reduce radiated emanations with plastic cabinets which provide little or no shielding.

In the ruling, the FCC defined a digital device (previously called a computing device) as follows:

An unintentional radiator (device or system) that generates and uses timing signals or pulses at a rate in excess of 9000 pulses (cycles) per second and uses digital techniques; inclusive of telephone equipment that uses digital techniques or any device or system that generates and uses radio frequency energy for the purpose of performing data processing functions, such as electronic computations, operations, transformations, recording, filing, sorting, storage, retrieval or transfer (§ 15.3(k)).

Computer terminals and peripherals, which are intended to be connected to a computer, are also considered to be digital devices.

This definition was intentionally broad to include as many products as possible. Thus, if a product uses digital circuitry and has a clock greater than 9 kHz, then it is a digital device under the FCC definition. This definition covers most digital electronics in existence today.

Digital devices covered by this definition are divided into the following two classes:

Class A: A digital device that is marketed for use in a commercial, industrial, or business environment (§ 15.3(h)).

Class B: A digital device that is marketed for use in a residential environment, notwithstanding use in commercial, business, and industrial environments (§ 15.3(i)).

Because Class B digital devices are more likely to be located in closer proximity to radio and television receivers, the emission limits for these devices are about 10 dB more restrictive than those for Class A devices.

Meeting the technical standards contained in the regulations is the obligation of the manufacturer or importer of a product. To guarantee compliance, the FCC requires the manufacturer to test the product for compliance before the product can be marketed in the United States. The FCC defines marketing as shipping, selling, leasing, offering for sale, importing, and so on (§ 15.803(a)). Until a product complies with the rules, it cannot legally be advertised or displayed at a trade show, because this would be considered an offer for sale. To advertise or display a product legally prior to compliance, the advertisement or display must contain a statement worded as follows:

This device has not been authorized as required by the rules of the Federal Communications Commission. This device is not, and may not be, offered for sale or lease, or sold or leased, until authorization is obtained (§ 2.803(c)).

For personal computers and their peripherals (a subcategory of Class B), the manufacturer can demonstrate compliance with the rules by a Declaration of Conformity. A Declaration of Conformity is a procedure where the manufacturer makes measurements or takes other steps to ensure that the equipment complies with the applicable technical standards (§ 2.1071 to 2.1077). Submission of a sample unit or representative test data to the FCC is not required unless specifically requested.

For all other products (Class A and Class B—other than personal computers and their peripherals), the manufacturer must verify compliance by testing the product before marketing. Verification is a self-certification procedure where nothing is submitted to the FCC unless specifically requested by the Commission, which is similar to a declaration of conformity (§ 2.951 to 2.956). Compliance is by random sampling of products by the FCC. The time required to do the compliance tests (and to fix the product, and redo the test if the product fails) should be scheduled into the product’s development timetable. Precompliance EMC measurements (see Chapter 18) can help shorten this time considerably.

Testing must be performed on a sample that is representative of production units. This usually means an early production or preproduction model. Final compliance testing must therefore be one of the last items in the product development timetable. This is no time for unexpected surprises! If a product fails the compliance test, then changes at this point are difficult, time consuming, and expensive. Therefore, it is desirable to approach the final compliance test with a high degree of confidence that the product will pass. This can be done if (1) proper EMC design principles (as described in this book) have been used throughout the design and (2) preliminary pre-compliance EMC testing as described in Chapter 18 was performed on early models and subassemblies.

It should be noted that the limits and the measurement procedures are inter-related. The derived limits were based on specified test procedures. Therefore, compliance measurements must be made following the procedure outlined by the regulations (§ 15.31). The FCC specifies that for digital devices, measurements to show compliance with Part 15, must be performed following the procedures described in measurement standard ANSI C63.4–1992 titled Methods of Measurement of Radio-Noise Emissions from Low-Voltage Electrical and Electronic Equipment in the Range of 9 kHz to 40 GHz, excluding Section 5.7, Section 9, and Section 14 (§ 15.31(a)(6)).*

The test must be made on a complete system, with all cables connected and configured in a reasonable way that tends to maximize the emission (§ 15.31(i)). Special authorization procedures are provided in the case of central processor unit (CPU) boards and power supplies that are used in personal computers and sold separately (§ 15.32).

1.5.3 Emissions

The FCC Part 15 EMC Regulations limit the maximum allowable conducted emission, on the ac power line in the range of 0.150 to 30 MHz, and the maximum radiated emission in the frequency range of 30 MHz to 40 GHz.

1.5.3.1 Radiated Emissions.    For radiated emissions, the measurement procedure specifies an open area test site (OATS) or equivalent measurement made over a ground plane with a tuned dipole or other correlatable, linearly polarized antenna. This setup is shown in Fig. 1-2. ANSI C63.4 allows for the use of an alternative test site, such as an absorber-lined room, provided it meets specified site attenuation requirements. However, a shielded enclosure without absorber lining may not be used for radiated emission measurements.

The specified receive antenna in the 30- to- 1000-MHz range is a tuned dipole, although other linearly polarized broadband antennas may also be used. However, in case of a dispute, data taken with the tuned dipole will take precedence. Above 1000 MHz, a linearly polarized horn antenna shall be used.

Table 1-1 lists the FCC radiated emission limits (§ 15.109) for a Class A product when measured at a distance of 10 m. Table 1-2 lists the limits for a Class B product when measured at a distance of 3 m.

FIGURE 1-2. Open area test site (OATS) for FCC radiated emission test. The equipment under test (EUT) is on the turntable.

TABLE 1-1. FCC Class A Radiated Emission Limits Measured at 10 m.

TABLE 1-2. FCC Class B Radiated Emission Limits Measured at 3 m.

TABLE 1-3. FCC Class A and Class B Radiated Emission Limits Measured at 10 m.

A comparison between the Class A and Class B limits must be done at the same measuring distance. Therefore, if the Class B limits are extrapolated to a 10-m measuring distance (using a 1/d extrapolation), the two sets of limits can be compared as shown in Table 1-3. As can be observed, the Class B limits are more restrictive by about 10 dB below 960 MHz and 5 dB above 960 MHz. A plot of both FCC Class A and Class B radiated emission limits over the frequency range of 30 MHz to 1000 MHz (at a measuring distance of 10 m) is shown in Fig. 1-5.

The frequency range over which radiated emission tests must be performed is from 30 MHz up to the frequency listed in Table 1-4, which is based on the highest frequency that the equipment under test (EUT) generates or uses.

1.5.3.2 Conducted Emissions.    Conducted emission regulations limit the voltage that is conducted back onto the ac power line in the frequency range of 150 kHz to 30 MHz. Conducted emission limits exist because regulators believes that at frequencies below 30 MHz, the primary cause of interference with radio communications occurs by conducting radio-frequency energy onto the ac power line and subsequently radiating it from the power line. Therefore, conducted emission limits are really radiated emission limits in disguise.

TABLE 1-4. Upper Frequency Limit for Radiated Emission Testing.

TABLE 1-5. FCC/CISPR Class A Conducted Emission Limits.

TABLE 1-6. FCC/CISPR Class B Conducted Emission Limits.

aLimit decreases linearly with log of frequency.

The FCC conducted emission limits (§ 15.107) are now the same as the International Special Committee on Radio Interference (CISPR, from its title in French) limits, used by the European Union. This is the result of the Commission amending its conducted emission rules in July 2002 to make them consistent with the international CISPR requirements.

Tables 1-5 and 1-6 show the Class A and Class B conducted emission limits, respectively. These voltages are measured common-mode (hot to ground and neutral to ground) on the ac power line using a 50-Ω/50-μH line impedance stabilization network (LISN) as specified in the measurement procedures.* Figure 1-3 shows a typical FCC conducted emission test setup.

FIGURE 1-3. Test setup for FCC conducted emission measurements.

A comparison between Tables 1-5 and 1-6 shows that the Class B quasi-peak conducted emission limits are from 13 dB to 23 dB more stringent than the Class A limits. Note also that both peak and average measurements are required. The peak measurements are representative of noise from narrowband sources such as clocks, whereas the average measurements are representative of broadband noise sources. The Class B average conducted emission limits are from 10 to 20 dB more restrictive than the Class A average limits.

Figure 1-4 shows a plot of both the average and the quasi-peak FCC/CISPR conducted emission limits.

1.5.4 Administrative Procedures

The FCC rules not only specify the technical standards (limits) that a product must satisfy but also the administrative procedures that must be followed and the measuring methods that must be used to determine compliance. Most administrative procedures are contained in Part 2, Subpart I (Marketing of Radio Frequency Devices), Subpart J (Equipment Authorization Procedures), and Subpart K (Importation of Devices Capable of Causing Harmful Interference) of the FCC Rules and Regulations.

Not only must a product be tested for compliance with the technical standards contained in the regulations, but also it must be labeled as compliant (§ 15.19), and information must be provided to the user (§ 15.105) on its interference potential.

FIGURE 1-4. FCC/CISPR conducted emission limits.

In addition to the technical standards mentioned above, the rules also contain a noninterference requirement, which states that if use of the product causes harmful interference, the user may be required to cease operation of the device (§ 15.5). Note the difference in responsibility between the technical standards and the noninterference requirement. Although meeting the technical standards (limits) is the responsibility of the manufacturer or importer of the product, satisfying the noninterference requirement is the responsibility of the user of the product.

In addition to the initial testing to determine compliance of a product, the rules also specify that the manufacturer or importer is responsible for the continued, or ongoing, compliance of subsequently manufactured units (§ 2.953, 2.955, 2.1073, 2.1075).

If a change is made to a compliant product, the manufacturer has the responsibility to determine whether that change has an effect on the compliance of the product. The FCC has cautioned manufacturers (Public Notice 3281, April 7, 1982) to note that:

Many changes, which on their face seem insignificant, are in fact very significant. Thus a change in the layout of a circuit board, or the addition or removal or even rerouting of a wire, or even a change in the logic will almost surely change the emission characteristics of the device. Whether this change in characteristics is enough to throw the product out of compliance can best be determined by retesting.

As of this writing (September 2008), the FCC has exempted eight subclasses of digital devices (§ 15.103) from meeting the technical standards of the rules. These are as follows:

1. Digital devices used exclusively in a transportation vehicle such as a car, plane, or boat.

2. Industrial control systems used in an industrial plant, factory, or public utility.

3. Industrial, commercial, or medical test equipment.

4. Digital devices exclusively used in an appliance such as a microwave oven, dishwasher, clothes dryer, air conditioner, and so on.

5. Specialized medical devices generally used at the direction or under the supervision of a licensed health care practitioner, whether used in a patient’s home or a health care facility. Note, medical devices marketed through retail channels for use by the general public, are not exempted.

6. Devices with power consumption not exceeding 6 nW, for example, a digital watch.

7. Joystick controllers or similar devices (such as a mouse) that contain no digital circuitry. Note, a simple analog to digital converter integrated circuit (IC) is allowed in the device.

8. Devices in which the highest frequency is below 1.705 MHz and that does not operate from the ac power line, or contain provisions for operation while connected to the ac power line.

Each of the above exempted devices is, however, still subject to the noninterference requirement of the rules. If any of these devices actually cause harmful interference in use, the user must stop operating the device or in some way remedy the interference problem. The FCC also states, although not mandatory, it is strongly recommended that the manufacturer of an exempted device endeavor to have that device meet the applicable technical standards of Part 15 of the rules.

Because the FCC has purview over many types of electronic products, including digital electronics, design and development organizations should have a complete and current set of the FCC rules applicable to the types of products they produce. These rules should be referenced during the design to avoid subsequent embarrassment when compliance demonstration is required.

The complete set of the FCC rules is contained in the Code of Federal Regulations, Title 47 (Telecommunications)—Parts 0 to 300. They consist of five volumes and are available from the Superintendent of Documents, U.S. Government Printing Office. The FCC rules are in the first volume that contains Parts 0 to 19 of the Code of Federal Regulations. A new edition is published in the spring of each year and contains all current regulations codified as of October 1 of the previous year. The Regulations are also available online at the FCC’s website, www.fcc.gov.

When changes are made to the FCC regulations, there is a transition period before they become official. This transition period is usually stated as x-number of days after the regulation is published in the Federal Register.

1.5.5 Susceptibility

In August 1982, the U.S. Congress amended the Communications Act of 1934 (House Bill #3239) to give the FCC authority to regulate the susceptibility of home electronics equipment and systems. Examples of home electronics equipment are radio and television sets, home burglar alarm and security systems, automatic garage door openers, electronic organs, and stereo/high-fidelity systems. Although this legislation is aimed primarily at home entertainment equipment and systems, it is not intended to prevent the FCC from adopting susceptibility standards for devices that are also used outside the home. To date, however, the FCC has not acted on this authority. Although it published an inquiry into the problem of Radio Frequency Interference to Electronic Equipment in 1978 (General Docket No. 78-369), the FCC relies on self-regulation by industry. Should industry become lax in this respect, the FCC may move to exercise its jurisdiction.

Surveys of the electromagnetic environment (Heirman 1976, Janes 1977) have shown that a field strength greater than 2 V/m occurs about 1% of the time. Because no legal susceptibility requirements exist for commercial equipment in the United States, a reasonable minimum immunity level objective might be 2 to 3 V/m. Clearly products with susceptibility levels of less than 1 V/m are not well designed and are very likely to experience interference from rf fields during their life span.

In 1982, the government of Canada released an Electromagnetic Compatibility Advisory Bulletin (EMCAB-1) that defined three levels, or grades, of immunity for electronic equipment, and stated the following:

1. Products that meet GRADE 1 (1 V/m) are likely to experience performance degradation.

2. Products that meet GRADE 2 (3 V/m) are unlikely to experience degradation.

3. Products that meet GRADE 3 (10 V/m) should experience performance degradation only under very arduous circumstances.

In June 1990, an updated version of EMCAB-1 was issued by Industry Canada. This updated version concludes that products located in populated areas can be exposed to field strengths that range from 1 V/m to 20 V/m over most of the frequency band.

1.5.6 Medical Equipment

Most medical equipment (other than what comes under the Part 18 Rules) is exempt from the FCC Rules. The Food and Drug Administration (FDA), not the FCC, regulates medical equipment. Although the FDA developed EMC standards, as early as 1979 (MDS-201-0004, 1979), they have never officially adopted them as mandatory. Rather, they depend on their inspectors’ guideline document to assure that medical devices are properly designed to be immune to electromagnetic interference (EMI). This document, Guide to Inspections of Electromagnetic Compatibility Aspects of Medical Devices Quality Systems, states the following:

At this time the FDA does not require conformance to any EMC standards. However, EMC should be addressed during the design of new devices, or redesign of existing devices.

However, the FDA is becoming increasingly concerned about the EMC aspects of medical devices. Inspectors are now requiring assurance from manufacturers that they have addressed EMC concerns during the design process, and that the device will operate properly in its intended electromagnetic environment. The above-mentioned Guide encourages manufacturers to use IEC 60601-1-2 Medical Equipment, Electromagnetic Compatibility Requirements and Tests as their EMC standard. IEC 60601-1-2 provides limits for both emission and immunity, including transient immunity such as electrostatic discharge (ESD).

As a result, in most cases, IEC 60601-1-2 has effectively become the unofficial, de facto, EMC standard that has to be met for medical equipment in the United States.

1.5.7 Telecom

In the United States, telecommunications central office (network) equipment is exempt from the FCC Part 15 Rules and Regulations as long as it is installed in a dedicated building or large room owned or leased by the telephone company. If it is installed in a subscriber’s facility, such as an office or commercial building, the exemption does not apply and the FCC Part 15 Rules are applicable.

Telecordia’s (previously Bellcore’s) GR-1089 is the standard that usually applies to telecommunications network equipment in the United States. GR-1089 covers both emission and susceptibility, and it is somewhat similar to the European Union’s EMC requirements. The standard is often referred to as the NEBS requirements. NEBS stands for New Equipment Building Standard. The standard is derived from the original AT&T Bell System internal NEBS standard.

These standards are not mandatory legal requirements but are contractual between the buyer and the seller. As such, the requirements can be waived or not applied in some cases.

1.5.8 Automotive

As stated, much (although not all) of the electronics built into transportation vehicles are exempt from EMC regulation, such as the FCC Part 15 Rules, in the United States (§ 15.103). This does not mean that vehicle systems do not have legal EMC requirements. In many regions of the world, there are legislated requirements for vehicle electromagnetic emissions and immunity. The legislated requirements are typically based on many internationally recognized standards, including CISPR, International Organization for Standardization (ISO), and the Society of Automotive Engineers (SAE). Each of these organizations has published several EMC standards applicable to the automotive industry. Although these standards are voluntary, the automotive manufacturers either rigorously apply them or use these standards as a reference in the development of their own corporate requirements. These developed corporate requirements may include both component and vehicle level items and are often based upon the customer satisfaction goals of the manufacturer—therefore, they almost have the effect of mandatory standards.

For example, SAE J551 is a vehicle-level EMC standard, and SAE J1113 is a component-level EMC standard applicable to individual electronic modules. Both standards cover emissions and immunity and are somewhat similar to the military EMC standards.

The resulting vehicle EMC standards cover both emissions and immunity and are some of the toughest EMC standards in the world, partly because of the combination of types of systems on vehicles and their proximity to each other. These systems include high-voltage discharges (such as spark ignition systems) located near sensitive entertainment radio receiver systems, wiring for inductive devices such as motors and solenoids in the same wiring harness as data communication lines, and with the newer hybrid vehicles high-current motor drive systems that operate at fast switching speeds. The radiated emission standards are typically 40 dB more stringent than the FCC Class B limits. Radiated immunity tests are specified up to an electric field strength of 200 V/m (or in some cases higher) as compared with 3 or 10 V/m for most non-automotive commercial immunity standards.

In the European Union, vehicles and electronic equipment intended for use in these vehicles are exempt from the EMC Directive (204/108/EC), but they do fall within the scope of the automotive directive (95/54/EC) that contains EMC requirements.

1.6 CANADIAN EMC REQUIREMENTS

The Canadian EMC regulations are similar to those of the United States. The Canadian regulations are controlled by Industry Canada. Table 1-7 lists the Canadian EMC standards applicable to various types of products. These standards can be accessed from the Industry Canada web page (www.ic.gc.ca).

TABLE 1-7. Canadian EMC Test Standards.

aDigital Equipment.

The ITE and ISM standards can be accessed from the Industry Canada home page by following the following links: A-Z Index/Spectrum Management and Telecommunications/Official Publications/Standards/Interference-Causing Equipment Standards (ICES). The telecom standard can be accessed from the Industry Canada home page by following the following links: A-Z Index/ Spectrum Management and Telecommunications/Official Publications/Standards/Terminal Equipment-Technical Specifications List.

The methods of measurement and actual limits for ITE are contained in CAN/CSA-CEI/IEC CISPR 22:02, Limits and Methods of Measurement of Radio Disturbance Characteristics of Information Technology Equipment.

To reduce the burden on U.S. and Canadian manufacturers, the United States and Canada have a mutual recognition agreement whereby each country agrees to accept test reports from the other country for equipment authorization purposes (FCC Public Notice 54795, July 12, 1995).

1.7 EUROPEAN UNION’S EMC REQUIREMENTS

In May 1989, the European Union (EU) published a directive (89/336/EEC) relating to electromagnetic compatibility, which was to be effective January 1, 1992. However, the European Commission underestimated the task of implementing the directive. As a result, the European Commission amended the directive in 1992 allowing for a 4-year transition period and requiring full implementation of the EMC directive by January 1, 1996.

The European EMC directive differs from the FCC regulations by including immunity requirements in addition to emission requirements. Another difference is that the directive, without exception, covers all electrical/electronic equipment. There are no exemptions—the EMC directive even covers a light bulb. The directive does, however, exclude equipment that is covered by another directive with EMC provisions, such as the automotive directive. Another example would be medical equipment, which comes under the medical directive (93/42/EEC) not the EMC directive.

1.7.1 Emission Requirements

As stated, the EU’s conducted emission requirements are now the same as the FCC’s (see Tables 1-5 and 1-6 as well as Fig. 1-4). The radiated emission standards are similar but not exactly the same. Table 1-8 shows the European Union’s Class A and Class B radiated emission limits when measured at 10 m.

TABLE 1-8. CISPR Radiated Emission Limits at 10 m.

FIGURE 1-5. Comparison of FCC and CISPR radiated emission limits, measured at a distance of 10 m.

Figure 1-5 compares the EU’s radiated emission standard with the current FCC standard over the frequency range of 30 MHz to 1000 MHz. The FCC Class B limits have been extrapolated to a 10-m measuring distance for this comparison. As can be observed the European (CISPR) limits are more restrictive in the frequency range from 88 to 230 MHz. Below 88 MHz and above 230 MHz the CISPR and FCC limits are virtually the same (within 0.5 dB of each other). However, the EU has no radiated emission limit above 1 GHz, whereas the FCC limits, under some circumstances (see Table 1-4), go up to 40 GHz.

Table 1-9 is a composite worst-case combination of the FCC and CISPR radiated emission limits when measured at 10 m.

TABLE 1-9. Composite Worst-Case Radiated Emission Limits for Commercial Products, Measured at a Distance of 10 m.

1.7.2 Harmonics and Flicker

The EU has two additional emission requirements that relate to power quality issues—harmonics and flicker. These regulations apply to products that draw an input current of 16 A per phase or less and are intended to be connected to the public ac power distribution system. The FCC has no similar requirement.

The harmonic requirement (EN 61000-3-2) limits the harmonic content of the current drawn by the product from the ac power line, (see Table 18-3). The generation of harmonics is the result of the nonlinear behavior of the loads connected to the ac power line. Common nonlinear loads include switched-mode power supplies, variable-speed motor drives, and electronic ballasts for fluorescent lamps.

A major source of harmonics is a full-wave rectifier connected directly to the ac power line and followed by a large-value capacitor input filter. Under these circumstances, current is only drawn from the power line when the input voltage exceeds that on the filter capacitor. As a result, current is drawn from the power line only on the peaks of the ac voltage waveform (see Fig. 13-4). The resultant current waveshape is rich in odd harmonics (third, fifth, seventh, etc.). Total harmonic distortion (THD) values of 70% to 150% are not uncommon under these circumstances.

The number of harmonics present is determined by the rise and fall time of the current pulse, and their magnitude by the current wave shape. Most switching power supplies (the exception is very low-power supplies) and variable-speed motor drives cannot meet this requirement without some kind of passive or active power factor correction circuitry.

To alleviate this problem, the ac input current pulse must be spread out over a larger portion of a cycle to reduce the harmonic content. Normally the THD of the current pulse must be reduced to 25% or less to be compliant with the EU regulations.

The flicker requirements (EN 61000-3-3) limit the transient ac power line current drawn by the product; see Table 18-4. The purpose of this requirement is to prevent lights from flickering, because it is perceived as being disturbing to people. The regulations are based on not providing a noticeable change in the illumination of a 60-W incandescent lamp powered off the same ac power supply as the equipment under test.

Because of the finite source impedance of the power line, the changing current requirements of equipment connected to the line produces corresponding voltage fluctuations on the ac power line. If the voltage variation is large enough, it will produce a perceptible change in lighting illumination. If the load changes are of sufficient magnitude and repetition rate, the resulting flickering of lights can be irritating and disturbing.

To determine an applicable limit, many people were subjected to light flicker to determine the irritability threshold. When the flicker rate is low (< 1 per minute), the threshold of irritability is when the ac line voltage changes by 3%. People are most sensitive to light flicker when the rate is around 1000 times per minute. At a rate of 1000 times per minute, a 0.3% voltage change is just as irritating as a 3% change at less than one change per minute. Above about 1800 changes per minute, light flicker is no longer perceived.

Most EMC emission requirements are based on the magnitude of a measured parameter not exceeding a specified amount (the limit). However, flicker tests are different in that they require many measurements to be made and then a statistical analysis to be performed on the measured data to determine whether the limit is exceeded.

For most equipment, this requirement is not a problem because they naturally do not draw large transient currents off the ac power line. However, the requirement can be a problem for products that suddenly switch on heaters that draw large currents, or motors under a heavy load. An example would be when an air conditioner compressor or a large heater in a copy machine is suddenly switched on.

1.7.3 Immunity Requirements

The EU’s immunity requirements cover radiated and conducted immunity, as well as transient immunity that include ESD, electrical fast transient (EFT), and surge.

The EFT requirement simulates noise generated by inductively switched loads on the ac power line. As a contactor is opened to an inductive load, an arc is formed that extinguishes and restarts many times. The surge requirement is intended to simulate the effect of a nearby lightning pulse.

In addition, the EU has susceptibility requirements that cover ac voltage dips, sags, and interruptions.

For additional information on these transient immunity and power line disturbance requirements, see Sections 14.3 and 14.4.

1.7.4 Directives and Standards

The European regulations consist of directives and standards. The directives are very general and are the legal requirements. The standards provide one way, but not the only way, to comply with the directive.

The EMC Directive 2004/108/EC (which superceded the original EMC Directive 89/336/EEC) defines the essential requirements for a product to be marketed in the EU. They are as follows:

1. The equipment must be constructed to ensure that any electromagnetic disturbance it generates allows radio and telecommunication equipment and other apparatus to function as intended.

2. The equipment must be constructed with an inherent level of immunity to externally generated electromagnetic disturbances.

These are the only legal requirements with respect to EMC and the requirements are vague. The directive provides for two methods of demonstrating compliance with its requirements. The most commonly used is by a declaration of conformity; the other option is the use of a technical construction file.

If a product is tested to and complies with the applicable EMC standards it is presumed to meet the requirements of the directive, and the manufacturer can produce a declaration of conformity attesting to that fact.

A declaration of conformity is a self-certification process in which the responsible party, manufacturer or importer, must first determine the applicable standards for the product, test the product to the standards, and issue a declaration declaring compliance with those standards and the EMC directive. The declaration of conformity can be a single-page document but must contain the following:

• Application of which council directives (all applicable directives)

• Standards used (including date of standard) to determine conformity

• Product name and model number, also serial numbers if applicable

• Manufacturer’s name and address

• A dated declaration that the product conforms to the directives

• A signature by a person empowered to legally bind the manufacturer

The technical construction file approach to demonstrating conformity is unique to the European Union. The technical construction file is often used where no harmonized standards exist for the product and the manufacturer does not think that the generic standards are appropriate. In this case, the manufacturer produces a technical file to describe the procedures and tests used to ensure compliance with the EMC directive. The manufacturer can develop its own EMC specifications and test procedures. The manufacturer can decide how, where, when, or if, the product is tested for EMC. An independent competent body, however, must approve the technical construction file. The competent bodies are appointed by the individual states of the European Union, and the European Commission publishes a list of them in the Official Journal of the European Union. The competent body must agree that, using the manufacturer’s procedures and tests, the product satisfies the essential requirements of the EMC directive. This approach is acceptable, because in the European Union, the EMC directive is the legal document that must be satisfied, not the standards. In most other jurisdictions, the standards are the legal documents that must be complied with.

Products whose compliance with the EMC directive has been demonstrated by one of the above procedures shall be labeled with the CE mark. The CE mark consists of the lower case letters ce in a specified, distinctive font. Affixing the CE mark to a product indicates conformity to all applicable directives, not just the EMC directive. Other applicable directives might be, the safety directive, the toy directive, the machinery directive, and so on.

Two types of standards exist in the European Union: product specific and generic.* Product-specific standards always take precedence over generic standards. However, if no applicable product-specific standard exists for a product, the generic standards are then applicable. Emission and immunity requirements for a product are usually covered by different standards. Currently, over 50 different standards are associated with the EMC directive. Table 1-10 lists some of the more commonly applicable product-specific standards, as well as the four generic EMC standards. If a product-specific standard does not exist in a category, then the requirement defaults to the appropriate generic standard.

The EU’s standards writing organization CENELEC (the European Committee for Electro-Technical Standardization) has been given the task of drawing up the corresponding technical specifications meeting the essential requirements of the EMC directive, compliance with which will provide a presumption of conformity with the essential requirements of the EMC directive. Such specifications are referred to as harmonized standards. Most CENELEC standards are derived from International Electro-Technical Committee (ITC) or CISPR standards—ITC for immunity standards and CISPR for emission standards. The CENELEC standards, or European Norms (EN), are not official until a reference to them is published in the Official Journal of the European Union.

TABLE 1-10. European Union’s EMC Test Standards.

aCovered by the Medical Directive (93/42/EEC), not the EMC Directive

As new standards come into existence and existing standards are modified, as regularly happens, a transition period, usually of 2 years is specified in the standard. During the transition period, either the old standard or the new standard can be used to demonstrate compliance with the EMC directive.

The latest information on the EMC Directive 2004/108/EC and the harmonized standards can be obtained on the following website: http://europa.eu.int/comm/enterprise/newapproach/standardization/harmstds/reflist/emc.xhtml.

In light of the large breadth and scope of the EMC Directive and the variety of products covered, the European Commission in 1997 felt it necessary to publish a 124-page guideline to the interpretation of the EMC directive to be used by manufacturers, test laboratories, and other parties affected by the directive (European Commission, 1997). This guideline was intended to clarify matters and procedures relating to the interpretation of the EMC Directive. It also clarified the application of the Directive to components, subassemblies, apparatus, systems, and installations, as well as the application of the Directive to spare parts, used, and repaired apparatus.

1.8 INTERNATIONAL HARMONIZATION

It would be desirable to have one international EMC standard for allowable emission and immunity of electronic products, instead of many different national standards. This would allow a manufacturer to design and test a product to one standard that would be acceptable worldwide. Figure 1-6 depicts a typical commercial product and shows the different types of EMC requirements, both emission and immunity, that it might have to meet in a harmonized world market.

Even more important than a single uniform EMC standard is a single uniform EMC test procedure. If the test procedure is the same, then an EMC test could be performed once and the results compared against many different standards (limits) to determine compliance with each regulation. When the test procedures are different, however, the product must be retested for each standard, which is a costly and time-consuming task.

The most likely vehicle for accomplishing harmonization is the European Union’s EMC standards, which are based on the CISPR standards. CISPR was formed in 1934 to determine measurement methods and limits for radio-frequency interference to facilitate international trade. CISPR has no regulatory authority, but its standards, when adopted by governments, become national standards. In 1985 CISPR adopted a new set of emission standards (Publication 22) for Information Technology Equipment (computer and digital electronics). The European Union has adopted the CISPR standard as the basis for their emission requirements. As a voting member of CISPR, the United States voted in favor of the new standard. This action puts considerable pressure on the FCC to adopt the same standards.

FIGURE 1-6. Typical composite worldwide commercial EMC requirements.

In 1996, the FCC modified its Part 15 Rules to allow manufacturers to use a Declaration of Conformity as a compliance procedure for personal computers and their peripherals, which is similar to that used by the EU’s EMC regulations. As stated, the FCC also has adopted the CISPR limits for conducted emission.

1.9 MILITARY STANDARDS

Another important group of EMC standards are those issued by the U.S. Department of