电子工程师手册（英文版二）：ch040

Hemming, L.H., Ungvichian, V., Roman, J.M., Uman, M.A., Rubinstein, M. “Compatibility” The Electrical Engineering Handbook Ed. Richard C. Dorf Boca Raton: CRC Press LLC, 2000 40 Compatibility 40.1 Grounding, Shielding, and Filtering Grounding?Shielding?Filtering 40.2 Spectrum, Specifications, and Measurement Techniques Electromagnetic Spectrum?Specifications?Measurement Procedures 40.3 Lightning Terminology and Physics?Lightning Occurrence Statistics?Electric and Magnetic Fields?Modeling of the Return Stroke?Lightning-Overhead Wire Interactions 40.1 Grounding, Shielding, and Filtering Leland H. Hemming Electromagnetic interference (EMI) is defined to exist when undesirable voltages or currents are present to influence adversely the performance of an electronic circuit or system. Interference can be within the system (intrasystem), or it can be between systems (intersystem). The system is the equipment or circuit over which one exercises design or management control. The cause of an EMI problem is an unplanned coupling between a source and a receptor by means of a transmission path. Transmission paths may be conducted or radiated. Conducted interference occurs by means of metallic paths. Radiated interference occurs by means of near- and far- field coupling. These different paths are illustrated in Fig. 40.1. The control of EMI is best achieved by applying good interference control principles during the design process. These involve the selection of signal levels, impedance levels, frequencies, and circuit configurations that minimize conducted and radiated interference. In addition, signal levels should be selected to be as low as possible, while being consistent with the required signal-to-noise ratio. Impedance levels should be chosen to minimize undesirable capacitive and inductive coupling. The frequency spectral content should be designed for the specific needs of the circuit, minimizing interfer- ence by constraining signals to desired paths, eliminating undesired paths, and separating signals from inter- ference. Interference control is also achieved by physically separating leads carrying currents from different sources. For optimum control, the three major methods of EMI suppression—grounding, shielding, and filter- ing—should be incorporated early in the design process. The control of EMI is first achieved by proper grounding, then by good shielding design, and finally by filtering. Grounding is the process of electrically establishing a low impedance path between two or more points in a system. An ideal ground plane is a zero potential, zero impedance body that can be used as reference for all signals in the system. Associated with grounding is bonding, which is the establishment of a low impedance path between two metal surfaces. Shielding is the process of confining radiated energy to the bounds of a specific volume or preventing radiated energy from reaching a specific volume. Filtering is the process of eliminating conducted interference by controlling the spectral content of the conducted path. Filtering is the last step in the EMI design process. Leland H. Hemming McDonnell Douglas Helicopter Systems Vichate Ungvichian Florida Atlantic University John M. Roman Telematics Martin A. Uman University of Florida, Gainesville Marcos Rubinstein Swiss PTT ? 2000 by CRC Press LLC Grounding Grounding Principles The three fundamental grounding techniques—floating, single-point, and multiple-point—are illustrated in Fig. 40.2. Floating grounds are used to isolate circuits or equipment from a common ground plane. Static charges are a hazard with this type of ground. Dangerous voltages may develop or a noise-producing discharge might occur. Generally, bleeder resistors are used to control the static problem. Floating grounds are useful only at low frequencies where capacitive coupling paths are negligible. The single-point ground is a single physical point in a circuit. By connecting all grounds to a common point, no interference will be produced in the equipment because the configuration does not result in potential differences across the equipment. At high frequencies care must be taken to prevent capacitive coupling, which will result in interference. A multipoint ground system exists when each ground connection is made directly to the ground plane at the closest available point on it, thus minimizing ground lead lengths. A large conductive body is chosen for the ground. Care must be taken to avoid ground loops. Circuit grounding design is dependent on the function of each type of circuit. In unbalanced systems, care must be taken to reduce the potential of common mode noise. Differential devices are commonly used to suppress this form of noise. The use of high circuit impedances should be minimized. Where it cannot be avoided, all interconnecting leads should be shielded, with the shield well grounded. Power supply grounding must be done properly to minimize load inducted noise on a power supply bus. When electromechanical relays are used in a system, it is best that they be provided with their own power supplies. Cable shield grounding must be designed based upon the frequency range, impedance levels, (whether balanced or unbalanced) and operating voltage and/or current. Cross talk between cables is a major problem and must be carefully considered during the design process. Building facility grounds must be provided for electrical faults, signal, and lightning. The fault protection (green wire) subsystem is for the protection of personnel and equipment from the hazards of electrical power faults and static charge buildup. The lightning protection system consists of air terminals (lightning rods), heavy duty down-conductors, and ground rods. The signal reference subsystem provides a ground for signal circuits to control static charges and noise and to establish a common reference between signals and loads. FIGURE 40.1 Electromagnetic interference is caused by uncontrolled conductive paths and radiated near/far fields. FIGURE 40.2 The type of ground system used must be selected carefully. ? 2000 by CRC Press LLC Earth grounds may consist of vertical rods, horizontal grids or radials, plates, or incidental electrodes such as utility pipes or buried tanks. The latter must be constructed and tested to meet the design requirements of the facility. Grounding Design Guidelines The following design guidelines represent good practice but should be applied subject to the detailed design objectives of the system. Fundamental Concepts ? Use single-point grounding for circuit dimensions less than 0.03 l (wavelength) and multipoint ground- ing for dimensions greater than 0.15 l. ?The type of grounding for circuit dimensions between 0.03 and 0.15 l depends on the physical arrange- ment of the ground leads as well as the conducted emission and conducted susceptibility limits of the circuits to be grounded. Hybrid grounds may be needed for circuits that must handle a broad portion of the frequency spectrum. ?Apply floating ground isolation techniques (i.e., transformers) if ground loop problems occur. ?Keep all ground leads as short as possible. ?Design ground reference planes so that they have high electrical conductivity and can be maintained easily to retain good conductivity. Safety Considerations ?Connect test equipment grounds directly to the grounds of the equipment being tested. ?Make certain the ground connections can handle fault currents that might flow unexpectedly. Circuit Grounding ?Maintain separate circuit ground systems for signal returns, signal shield returns, power system returns, and chassis or case grounds. These returns then can be tied together at a single ground reference point. ?For circuits that produce large, abrupt current variations, provide a separate grounding system, or provide a separate return lead to the ground to reduce transient coupling into other circuits. ?Isolate the grounds of low-level circuits from all other grounds. ?Where signal and power leads must cross, make the crossing so that the wires are perpendicular to each other. ?Use balanced differential circuitry to minimize the effects of ground circuit interference. ?For circuits whose maximum dimension is significantly less than l/4, use tightly twisted wires (either shielded or unshielded, depending on the application) that are single-point grounded to minimize equipment susceptibility. Cable Grounding ?Avoid pigtails when terminating cable shields. ?When coaxial cable is needed for signal transmission, use the shield as the signal return and ground at the generator end for low-frequency circuits. Use multipoint grounding of the shield for high-frequency circuits. ?Provide multiple shields for low-level transmission lines. Single-point grounding of each shield is rec- ommended. Shielding The control of near- and far-field coupling (radiation) is accomplished using shielding techniques. The first step in the design of a shield is to determine what undesired field level may exist at a point with no shielding and what the tolerable field level is. The difference between the two then is the needed shielding effectiveness. This section discusses the shielding effectiveness of various solid and nonsolid materials and their application to various shielding situations. Penetrations and their design are discussed so that the required shielding effectiveness is maintained. Finally, common shielding effectiveness testing methods are reviewed. ? 2000 by CRC Press LLC Enclosure Theory The attenuation provided by a shield results from three loss mechanisms as illustrated in Fig. 40.3. 1.Incident energy is reflected (R) by the surface of the shield because of the impedance discontinuity of the air–metal boundary. This mechanism does not require a particular material thickness but simply an impedance discontinuity. 2.Energy that does cross the boundary (not reflected) is attenuated (A) in passing through the shield. 3.The energy that reaches the opposite face of the shield encounters another air–metal boundary and thus some of it is reflected (B) back into the shield. This term is only significant when A < 15 dB and is generally neglected because the barrier thickness is generally great enough to exceed the 15-dB loss rule of thumb. Thus: S = R + A + B dB (40.1) Absorption loss is independent of the type of wave (electric/magnetic) and is given by A = 1.314( f m r s r ) 1/ 2 d dB (40.2) where d is shield thickness in centimeters, m r is relative permeability, f is frequency in Hz, and s r is conductivity of metal relative to that of copper. Typical absorption loss is provided in Table 40.1. Reflection loss is a function of the intrinsic impedance of the metal boundary with respect to the wave impedance, and therefore, three conditions exist: near-field magnetic, near-field electric, and plane wave. The relationship for low-impedance (magnetic field) source is R = 20 log 10 {[1.173(m r /f s r ) 1/2 /D] + 0.0535 D( f s r /m r ) 1/2 + 0.354 dB} (40.3) where D is distance to source in meters. For a plane wave source the reflection loss is R = 168 – 10 log 10 ( f m r /s r ) dB (40.4) For a high-impedance (electric field) source the reflection loss R is R = 362 – 20 log 10 [(m r f 3 /s r ) 1/2 D] dB (40.5) FIGURE 40.3 Shielding effectiveness is the result of three loss mechanisms. ? 2000 by CRC Press LLC Figure 40.4 illustrates the shielding effectiveness of a variety of common materials versus various thicknesses for a source distance of 1 m. This is the shielding effectiveness of a six-sided enclosure. To be useful, the enclosure must be penetrated for various services or devices. This is illustrated in Fig. 40.5(a) for small enclosures and Fig. 40.5(b) for room-sized enclosures. TABLE 40.1 Absorption Loss Is a Function of Type of Material and Frequency (Loss Shown is at 150 kHz) Relative Relative Absorption Metal Conductivity Permeability Loss A, dB/mm Silver 1.05 1 52 Copper—annealed 1.00 1 51 Copper—hard drawn 0.97 1 50 Gold 0.70 1 42 Aluminum 0.61 1 40 Magnesium 0.38 1 31 Zinc 0.29 1 28 Brass 0.26 1 26 Cadmium 0.23 1 24 Nickel 0.20 1 23 Phosphor–bronze 0.18 1 22 Iron 0.17 1000 650 Tin 0.15 1 20 Steel, SAE1045 0.10 1000 500 Beryllium 0.10 1 16 Lead 0.08 1 14 Hypernik 0.06 80000 3500 a Monel 0.04 1 10 Mu-metal 0.03 80000 2500 a Permalloy 0.03 80000 2500 a Steel, stainless 0.02 1000 220 a a Assuming that material is not saturated. Source: MIL-HB-419A. FIGURE 40.4 The shielding effectiveness of common sheet metals, 1 m separation. (a) 26-gage steel; (b) 3-oz. copper foil; (c) 0.030-in. aluminum sheet; (d) 0.003-in. Permalloy; (e) is a common specification for shielded enclosures. 300 280 260 240 220 200 180 160 140 120 100 80 60 40 20 0 Shielding Ef fectiveness, S dB , in dB Shielding Ef fectiveness, S dB , in dB 300 280 260 240 220 200 180 160 140 120 100 80 60 40 20 0 10Hz 30100Hz 300 1kHz 3 10kHz 30100kHz 3001MHz 310MHz 30100MHz 3001GHz 310GHz 30 Frequency H E EPW PW PW NSA-65-6 H (e)(e) (a) (b) (c) (c) (b) (d) (d) (a) (e) ? 2000 by CRC Press LLC Shielding Penetrations Total shielding effectiveness of an enclosure is a func- tion of the basic shield and all of the leakages asso- ciated with the penetrations in the enclosure. The latter includes seams, doors, vents, control shafts, piping, filters, windows, screens, and fasteners. The design of the seams is a function of the type of enclosure and the level and nature of the shielding effectiveness required. For small instruments, com- puters, and similar equipment, the typical shielding required is on the order of 60 dB for electric and plane wave shielding. EMI gaskets are commonly used to seal the openings in sheet metal construction. In some high-performance applications the shielding is achieved using very tight-fitting machined hous- ings. Examples are IF strips and large dynamic range log amplifier circuits. Various methods of sealing joints are illustrated in Fig. 40.6. EMI gasketing methods are shown in Fig. 40.7. For large room-sized enclosures, the performance requirements typically range from 60 to 120 dB. Conductive EMI shielding tape is used in the 60-dB realm, clamped seams for 80–100 dB, and continuous welded seams for 120-dB performance. These are illustrated in Fig. 40.8. A good electromagnetic shielded door design must meet a variety of physical and electrical requirements. Figure 40.9 illustrates a number of ways this is accomplished. For electronic equipment, a variety of penetrations must be made to make the shielded volume functional. These include control shafts, windows, lights, filters, and displays. Careful design is required to maintain the required shielding integrity. Shield Testing The most common specification used for shield evaluation is the procedure given in MIL-STD-285. This consists of establishing a reference level without the shield and then enclosing the receiver within the shield and determining the difference. The ratio is the shielding effectiveness. This applies regardless of materials used in the construction of the shield. Care must be taken in evaluating the results since the measured value is a function of a variety of factors, not all of which are definable. FIGURE 40.5 Penetrations in small (a) and large (b) enclosures. FIGURE 40.6Methods of sealing enclosure seams. ? 2000 by CRC Press LLC Summary of Good Shielding Practice Shielding Effectiveness ? Good conductors, such as copper and aluminum, should be used for electric field shields to obtain high reflection loss. A shielding material thick enough to support itself usually provides good electric shielding at all frequencies. ? Magnetic materials, such as iron and special high-permeability alloys, should be used for magnetic field shields to obtain high absorbtion loss. ? In the plane wave region, the sealing of all apertures is critical to good shielding practice. Multiple Shields ? Multiple shields are quite useful where high degrees of shielding effectiveness are required. Shield Seams ? All openings or discontinuities should be addressed in the design process to ensure achievement of the required shielding effectiveness. Shield material should be selected not only from a shielding requirement, but also from electrochemical corrosion and strength considerations. ? Whenever system design permits, use continuously overlapping welded seams. Obtain intimate contact between mating surfaces over as much of the seam as possible. FIGURE 40.7 Methods of constructing gasketed joints. FIGURE 40.8 Most common seams in large enclosures. (a) Foil and shielding tape; (b) clamped; (c) welded. ? 2000 by CRC Press LLC ?Surfaces to be mated must be clean and free from nonconducting finishes, unless the bonding process positively and effectively cuts through the finish. When electromagnetic compatibility (EMC) and finish specifications conflict, the finishing requirements must be modified. Case Construction ?Case material should have good shielding properties. ?Seams should be welded or overlapped. ?Panels and cover plates should be attached using conductive gasket material with closely spaced fasteners. ?Mating surfaces should be cleaned just before assembly to ensure good electrical contact and to minimize corrosion. ?A variety of special devices are available for sealing around doors, vents, and windows. ?Internal interference generating circuits must be isolated both electrically and physically. Electrical isolation is achieved by circuit design; physical isolation may be achieved by proper shielding. ?For components external to the case, use EMI boots on toggle switches, EMI rotary shaft seals on rotary shafts, and screening and shielding on meters and other indicator faces. Cable Shields ?Cabling that penetrates a case should be shielded and the shield should be terminated in a peripheral bond at the point of entry. This peripheral bond should be made to the connector or adaptor shell. Filtering An electrical filter is a combination of lumped or distributed circuit elements arranged so that it has a frequency characteristic that passes some frequencies and blocks others. FIGURE 40.9 Methods of sealing seams in RF enclosure small (a) and large (b) doors. ? 2000 by CRC Press LLC Filters provide an effective means for the reduction and suppression of electromagnetic interference as they control the spectral content of signal paths. The application of filtering requires careful consideration of an extensive list of factors including insertion loss, impedance, power handling capability, signal distortion, tun- ability, cost, weight, size, and rejection of undesired signals. Often they are used as stopgap measures, but if suppression techniques are used early in the design process, then the complexity and cost of interference fixes can be minimized. There are many textbooks on filtering, which should be used for specific applications. The types of filters are classified according to the band of frequencies to be transmitted or attenuated. The basic types illustrated in Fig. 40.10 include low-pass, high-pass, bandpass, and bandstop (reject). Filters can be composed of lumped, distributed, or dissipative elements; the type used is mainly a function of frequency. Filtering Guidance ?It is best to filter at the interference source. ?Suppress all spurious signals. ?Design nonsusceptible circuits. ?Ensure that all filter elements interface properly with other EMC elements, i.e., proper mounting of a filter in a shielded enclosure. Filter Design Filters using lumped and distributive elements generally are reflective, in that the various component combi- nations are designed for high series impedance and low shunt impedance in the stopband while providing low series impedance and high shunt impedance in the passband. The impedance mismatches associated with the use of reflective filters can result in an increase of interference. In such cases, the use of dissipative elements is found to be useful. A broad range of ferrite components are available in the form of beads, tubes, connector shells, and pins. A very effective method of low-pass filtering is to form the ferrite into a coaxial geometry, the properties of which are proportional to the length of the ferrite, as shown in Fig. 40.11. Application of filtering takes many forms. A common problem is transient suppression as illustrated in Fig. 40.12. All sources of transient interference should be treated at the source. Power line filtering is recommended to eliminate conducted interference from reaching the powerline and adjacent equipment. Active filtering is very useful in that it can be built in as part of the circuit design and can be effective in passing only the design signals. A variety of noise blankers, cancelers, and limiter circuits are available for active cancellation of interference. Special Filter Types A variety of special-purpose filters are used in the design of electronic equipment. Transmitters require a variety of filters to achieve a noise-free output. Receive preselectors play a useful role in interference rejection. Both distributed (cavity) and lumped element components are used. IF filters control the selectivity of a receiving system and use a variety of mechanical and electrical filtering components. Testing The general requirements for electromagnetic filters are detailed in MIL-F-15733, MIL-F-18327, and MIL-F- 25880. Insertion loss is measured in accordance with MIL-STD-220. Defining Terms Earth electrode system: A network of electrically interconnected rods, plates, mats, or grids, installed for the purpose of establishing a low-resistance contact with earth. The design objective for resistance to earth of this subsystem should not exceed 10 W. ? 2000 by CRC Press LLC FIGURE 40.10 Filters provide a variety of frequency characteristics. ? 2000 by CRC Press LLC Electromagnetic compatibility (EMC): The capability of equipment or systems to be operated in their intended operational environment at designed levels of efficiency without causing or receiving degrada- tion owing to unintentional electromagnetic interference. Electromagnetic compatibility is the result of an engineering planning process applied during the life cycle of the equipment. The process involves careful considerations of frequency allocation, design, procurement, production, site selection, installa- tion, operation, and maintenance. Electromagnetic pulse (EMP): A large impulsive-type electromagnetic wave generated by nuclear or chemical explosions. FIGURE 40.11 Ferrite provides a flexible means of achieving a low-pass filter with good high-frequency loss characteristics. FIGURE 40.12 Transient responses are controlled using simple filters at the source. (a) Resistance damping; (b) capacitance suppression; (c) RC suppression; (d) diode suppression; (e) back-to-back diode suppression. ? 2000 by CRC Press LLC Field strength: A general term that means the magnitude of the electric field vector (in volts per meter) or the magnitude of the magnetic field vector (in ampere-turns per meter). As used in the field of EMC/EMI, the term field strength shall be applied only to measurements made in the far field and shall be abbreviated as FS. For measurements made in the near field, the term electric field strength (EFS) or magnetic field strength (MFS) shall be used, according to whether the resultant electric or magnetic field, respectively, is measured. Penetration: The passage through a partition or wall of an equipment or enclosure by a wire, cable, pipe, or other conductive object. Radio frequency interference (RFI): Synonymous with electromagnetic interference. Shielding effectiveness: A measure of the reduction or attenuation in the electromagnetic field strength at a point in space caused by the insertion of a shield between the source and that point. Signal reference subsystem: This subsystem provides the reference points for all signal grounding to control static charges, noise, and interference. It may consist of any one or a combination of the lower frequency network, higher frequency network, or hybrid signal reference network. TEMPEST: A code word (not an acronym) which encompasses the government/industrial program for controlling the emissions from systems processing classified data. Individual equipment may be TEM- PESTed or commercial equipment may be placed in shielded enclosures. Related Topics 10.3 The Ideal Linear-Phase Low-Pass Filter ? 10.4 Ideal Linear-Phase Bandpass Filters ? 55.3 Dielectric Breakdown References AFSC Design Handbook, DH1-4, Electromagnetic Compatibility, 4th ed., U.S. Air Force, Wright-Patterson Air Force Base, Ohio, January 1991. R. F. Ficchi, Ed., Practical Design for Electromagnetic Compatibility, Hayden, 1971. E. R. Freeman, Electromagnetic Compatibility Design Guide for Avionics and Related Ground Support Equipment, Norwood, Mass.: Artech House, 1982. L. H. Hemming, Architectural Electromagnetic Shielding Handbook, New York: IEEE Press, 1991. B. Keiser, Principles of Electromagnetic Compatibility, 3rd ed., Norwood, Mass.: Artech House, 1987. Y.J. Lubkin, Filter Systems and Design: Electrical, Microwave, and Digital, Reading, Mass.: Addison-Wesley, 1970. MIL-HDBK-419A, Grounding, Bonding, and Shielding of Electronic Equipment and Facilities, U.S. Department of Defense, Washington, D.C., 1990. R. Morrison, Grounding and Shielding Techniques in Instrumentation, New York: John Wiley, 1986. R. Morrison and W. H. Lewis, Grounding and Shielding Techniques in Facilities, New York: John Wiley, 1990. T. Rikitake, Magnetic and Electromagnetic Shielding, Amsterdam: D. Reidel, 1987. N. O. N. Violetto, Electromagnetic Compatibility Handbook, New York: Van Nostrand Reinhold, 1987. D. R. J. White, Shielding Design, Methodology and Procedures, Springfield, Va.: Interference Control Technologies, 1986. D. R. J. White, A Handbook on Electromagnetic Shielding Materials and Performance, Springfield Va.: Interference Control Technologies, 1975. Further Information The annual publication Interference Technology Engineers’ Master (Item), published by R&B Enterprises, West Conshohocken, Pennsylvania, covers all aspects of EMI including an extensive product directory. The periodical IEEE Transactions on Electromagnetic Compatibility, which is published by The Institute of Electrical and Electronics Engineers, Inc., provides theory and practice in the EMI field. The periodical EMC Test & Design, published by the Cardiff Publishing Company, is a good source for practical EMI design information. ? 2000 by CRC Press LLC ? 2000 by CRC Press LLC The periodical “emf-emi control” published bimonthly by EEC Press, Gainesville, VA, is an excellent source of practical EMI information. The periodical Compliance Engineering, published quarterly by Compliance Engineering, Inc., is a good source for information on EMC regulations and rules. 40.2 Spectrum, Specifications, and Measurement Techniques Vichate Ungvichian and John M. Roman Electromagnetic radiation is a form of energy at a particular frequency that can propagate through a medium. This intentionally or unintentionally generated electromagnetic energy is considered as electromagnetic inter- ference (EMI) if it degrades the performance of electronic systems. The purposeful generation of electromag- netic energy for communications can be defined as intentionally generated EMI; unintentionally generated EMI can be created, for example, by the electrical signals in a computer and may be radiated into space by way of the interconnecting cables and/or by openings in the device enclosures. All electrical devices create some form of electromagnetic energy that may potentially interfere with the operation of other electrical devices outside the system (inter-system) or within the system (intra-system). Due to the increasing man-made EMI generated around the globe, allowable limits as well as measurement tech- niques on RF noise/interference have been set at national and international levels. The Federal Communications Commission and the Military are the two governing bodies in the United States setting standards on EMI, whereas the International Electrotechnical Commission is the ruling body in Europe. These ruling bodies are concerned with only a fraction of the total electromagnetic spectrum. Electromagnetic Spectrum The frequency spectrum of electromagnetic energy can span from dc to gamma ray (10 21 Hz) and beyond. Figure 40.13 shows the typical frequency spectrum chart over a fraction of hertz to 6 ′ 10 22 Hz. The spectrum for use in electromagnetic compatibility (EMC) purposes covers only from a few hertz (extreme low frequency, ELF) to 40 GHz (microwave bands). ELF has been in use mostly in the area of biological research and ELF communications. On the other side of the spectrum, the electronic devices must function in a hostile environment, in military applications, over the gigahertz frequency range. FIGURE 40.13The frequency spectrum chart. (Contributed by Luther Monell, North America Rockwell Corp.) Specifications In the United States, the Federal Communications Commission (FCC) and the military (MIL) are the two regulating bodies governing the EMC standards for commercial and military-based electronic devices and systems. In Europe, each country has its own EMC governing body as well as its own standards. The Verband Deutscher Elektrotechniker (VDE), British Standards Institute (BSI), European Telecommunications Standards Institute (ETSI), and International Special Committee on Radio Interference (CISPR) standards are very few examples of the acceptable standards in European countries such as Germany and Great Britain. Not until January 1, 1996, more than 15 European countries have adopted the IEC 1000 Part 4 Sections 2 to 4 as common EMI/EMC standards with several additional sections being in the final stages of approval. The United States of America Federal Communications Commission. The FCC sets limits on the amount of electromagnetic radiation allowed to be emitted from commercial electronic equipment. Any electronic device capable of emitting radio frequency energy by radiation or conduction is defined by the Commission as a radio frequency device and is subject to comply with the standards. The relevant limits and some general measurement techniques, along with equipment authorization procedures, are given in the Code of Federal Regulations (CFR), Title 47 (Tele- communications). Listed in the CFR, Title 47 are five different types of equipment authorization procedures, namely, type acceptance, type approval, notification, certification, and verification. Restrictions are placed on the marketing and sale of radio frequency devices until the appropriate equipment authorization criteria are met. Devices and systems that require allocation of the frequency spectrum fall under either type acceptance or type approval equipment authorization procedures. These radio frequency devices usually radiate high powers as in radio or television broadcast transmitters. Radio frequency devices not within the allocated part of the RF spectrum would require either certification or verification equipment authorizations. Some receivers, such as pagers, require notification equipment authorization as well. The requirements for these types of radio frequency devices are listed in Part 15 of the CFR, Title 47. Part 15 contains three categories of equipment. Incidental radiators (such as dc motors, mechanical light switches, etc.) are not subject to FCC Part 15 emission control requirements. Unintentional radiators are radio frequency devices that intentionally generate radio frequency energy for use within the device, but which are not intended to emit RF energy by radiation or induction. Intentional radiators are radio frequency devices that intentionally radiate radio frequency energy by radiation or by induction. Unintentional Radiators. There are two different classifications of digital devices listed under unintentional radiators. Class A digital devices are defined as devices that are intended for use in the commercial, industrial, or business environment. Class B digital devices are defined as devices that are intended to be used in a residential TABLE 40.2 Unintentional Radiator Equipment Authorizations Type of Device Equipment Authorization Required a TV broadcast receiver Verification FM broadcast receiver Do CB receiver Certification Superregenerative receiver Do Scanning receiver Do All other receivers subject to Part 15 Notification TV interface device Certification Cable system terminal device Notification Stand-alone cable input selector switch Verification Class B personal computers and peripherals Certification Other Class B digital devices and peripherals Verification Class A digital devices and peripherals Do External switching power supplies Do All other devices Do a See additional provisions in CFR Part 15.101 and Part 15.103. ? 2000 by CRC Press LLC environment. Table 40.2 lists the different types of unintentional radiators and their corresponding equipment authorization procedures. Unintentional Radiator Exempted Devices.There are some unintentional radiators that are listed as exempt devices. These devices are exempted from the technical requirements but are subject to the general requirements of Part 15, Sections 15.5 and 15.29, which state that if the devices cause harmful interference their operation must be ceased until such time that the interference is corrected, and that the device must be made available for inspection upon request by the Commission. It is also recommended (although not required) that these devices meet the technical specifications in Part 15. The exempted devices are as follows: a.A digital device utilized exclusively in any transportation vehicle, including motor vehicles and aircraft. b.A digital device used exclusively as an electronic control or power system utilized by a public utility or in an industrial plant. The term public utility includes equipment only to the extent that it is in a dedicated building or in a large room owned or leased by the utility and does not extend to equipment installed in a subscriber facility. c.A digital device used exclusively as industrial, commercial, or medical test equipment. d.A digital device utilized exclusively in an appliance, e.g., microwave oven, dishwasher, clothes dryer, air conditioner, etc. e.Specialized medical digital devices (generally used at the direction of or under the supervision of a licensed health care practitioner) whether used in a patient’s home or a health care facility. Nonspecialized medical devices, i.e., devices marketed through retail channels for use by the general public, are not exempted. This exemption also does not apply to digital devices used for recordkeeping or any purpose not directly connected with medical treatment. f.Digital devices having a power consumption not exceeding 6 nW. g.Joystick controllers, or similar devices such as a mouse, used with digital devices but which contain only nondigital circuitry or a simple circuit to convert the signal to the format required are viewed as passive add-on devices and are not directly subject to the technical standards or the equipment authorization requirements. h.Digital devices in which the highest frequency generated and the highest frequency used are less than 1.705 MHz and which do not operate from the ac power lines or contain provisions to operate while connected to the ac power lines. i.It should be noted that equipment containing more than one device is not exempt from the technical standards in Part 15 unless all of the devices meet the criteria for exemption. FIGURE 40.14Class A and Class B conducted emission limits. ? 2000 by CRC Press LLC Unintentional Radiator Conducted Emission Limits.The limits on the interference conducted back into the ac power distribution system are given in Fig. 40.14. Unintentional Radiator Radiated Emission Limits.Unintentional radiator radiated emission limits for Classes A and B devices are given in Figs. 40.15(a) and (b). It should be noted that a 10-m distance is required for Class A limit whereas 3 m is needed for Class B limit. Intentional Radiators.There are different limits on devices which intentionally radiate radio frequency energy. The authorization procedure required by the Commission is the same as a certification. In addition to the radiated and conducted emission limits, there are restricted bands of operation in which the devices may not intentionally radiate (spurious emissions are permitted in these bands); these bands are given in Table 40.3. FIGURE 40.15(a)Unintentional (10 m) Class A radiated emission limit. FIGURE 40.15(b)Unintentional (3 m) Class B radiated emission limits. ? 2000 by CRC Press LLC There is also a requirement that the antenna on the device be attached such that it may not be replaced by a different antenna. Intentional Radiator Conducted Emissions.The conducted emission limits on intentional radiators are the same as for a Class B digital device (see Fig. 40.14). Intentional Radiator Radiated Emissions.There are general provisions for the amount of radio frequency energy the intentional radiators are allowed to emit. Table 40.4 shows the general requirements for the radiated emission limits. There are also some additional provisions for operation in specific frequency bands. Listed below are the frequency bands (and their corresponding part in the CFR) that contain additional provisions on the limits of radiated emissions. It is recommended that the CFR, Title 47 be consulted for the limits if the device in question intentionally radiates in the listed bands. 1.Part 15.217: Operation in the band 160–190 kHz. 2.Part 15.219: Operation in the band 510–1705 kHz. 3.Part 15.221: Operation in the band 525–1705 kHz (carrier current systems, am broadcast stations on a college or university). TABLE 40.3Restricted Bands of Operation MHz MHz MHz GHz 0.090–0.110 162.0125–167.17 2310–2390 9.3–9.5 0.49–0.51 167.72–173.2 2483.5–2500 10.6–12.7 2.1735–2.1905 240–285 2655–2900 13.25–13.4 8.362–8.366 322–335.4 3260–3267 14.47–14.5 13.36–13.41 399.9–410 3332–3339 15.35–16.2 25.5–25.67 608–614 3345.8–3358 17.7–21.4 37.5–38.25 960–1240 3600–4400 22.01–23.12 73–75.4 1300–1427 4500–5250 23.6–24.0 108–121.94 1435–1626.5 5350–5460 31.2–31.8 123–138 1660–1710 7250–7750 36.43–36.5 149.9–150.05 1718.8–1722.2 8025–8500 Above 38.6 156.7–156.9 2200–2300 9000–9200 TABLE 40.4General Requirement Radiated Emission Limits Field Strength Measurement Frequency (MHz) (mV/m) Distance (m) 0.009–0.490 2400/F (kHz) 300 0.490–1.705 24000/F (kHz) 30 1.705–30.0 30 30 30–88 100* 3 88–216 150* 3 216–960 200* 3 Above 960 500 3 *Except as provided in paragraph (g), fundamental emissions from intentional radiators operating under this Section shall not be located in the frequency bands 54–72 MHz, 76–88 MHz, 174–216 MHz, or 470–806 MHz. However, operation within these frequency bands is permitted under other sections of this part, e.g., §§ 15.231 and 15.241. Note: All of the limits specified above are measured using a CISPR Quasi-Peak adapter except for the frequency bands 9–90 kHz, 110–490 kHz, and above 1000 MHz; these are specified for an average measurement. ? 2000 by CRC Press LLC 4.Part 15.223: Operation in the band 1.705–10 MHz. 5.Part 15.225: Operation in the band 13.553–13.567 MHz. 6.Part 15.227: Operation in the band 26.96–27.28 MHz. 7.Part 15.229: Operation in the band 40.66–40.70 MHz. 8.Part 15.231: Periodic operation in the band 40.55–40.70 MHz and above 70 MHz (alarm systems, door openers, remote switches, etc.). 9.Part 15.233: Operations within the bands 43.71–44.49 MHz, 46.60–46.98 MHz, 48.75–49.51 MHz, and 49.66–50.0 MHz (cordless telephones). 10.Part 15.235: Operation within the band 49.82–49.90 MHz. 11.Part 15.237: Operation within the bands 72.0–73.0 MHz, 74.6–74.8 MHz, and 75.2–76.0 MHz. 12.Part 15.239: Operation in the band 88–108 MHz. 13.Part 15.241: Operation in the band 174–216 MHz (biomedical telemetry devices only). 14.Part 15.243: Operation in the band 890–940 MHz (devices that use radio frequency energy to measure the characteristics of materials only). 15.Part 15.245: Operation in the bands 902–928 MHz, 2435–2465 MHz, 5785–5815 MHz, 10500–10550 MHz, and 24075–24175 MHz (field disturbance sensors only, excluding perimeter protection systems). 16.Part 15.247: Operation in the bands 902–928 MHz, 2400–2483.5 MHz, and 5725–5850 MHz (certain frequency hopping and direct sequence spread spectrum intentional radiators). 17.Part 15.249: Operation within the bands 902–928 MHz, 2400–2483.5 MHz, 5725–5875 MHZ, and 24.0–24.25 GHz. 18.Part 15.251: Operation within the bands 2.9–3.26 GHz, 3.267–3.332 GHz, 3.339–3.3458 GHz, and 3.358–3.6 GHz. Military Standards.The standards and requirements related to the electromagnetic interference and suscep- tibility for the Military in the United States are described in the MIL-STD-461C document. Electromagnetic interference is defined as the radiated and conducted energy emitted from the device. Electromagnetic suscep- tibility is defined as the amount of radiated or conducted energy that the device can withstand without degrading its performance. The standards are broken down into 17 segments defined by the two-letter suffix code and followed by three numbers ranging from 101–999 in the requirement name. The letter codes are conducted emissions (CE), conducted susceptibility (CS), radiated emissions (RE), radiated susceptibility (RS). Table 40.5 is a list and descriptions of the different emission and susceptibility requirements for a particular branch or type of application. TABLE 40.5Appropriated Requirements for Different Platforms and Installations Requirement Description CE101 Conducted emissions, power leads, 30 Hz to 10 kHz CE102 Conducted emissions, power leads, 10 kHz to 10 MHz CE106 Conducted emissions, antenna terminal, 10 kHz to 40 GHz CS101 Conducted susceptibility, power leads, 30 Hz to 50 kHz CS103 Conducted susceptibility, antenna port, intermodulation, 15 kHz to 10 GHz CS104 Conducted susceptibility, antenna port, rejection of undesired signals, 30 Hz to 20 GHz CS105 Conducted susceptibility, antenna port, cross-modulation, 30 Hz to 20 GHz CS109 Conducted susceptibility, structure current, 60 Hz to 100 kHz CS114 Conducted susceptibility, bulk cable injection, 10 kHz to 400 MHz CS115 Conducted susceptibility, bulk cable injection, impulse excitation CS116 Conducted susceptibility, damped sinusoidal transients, cables and power leads, 10 kHz to 100 MHz RE101 Radiated emissions, magnetic field, 30 Hz to 100 kHz RE102 Radiated emissions, electric field, 10 kHz to 18 GHz RE103 Radiated emissions, antenna spurious and harmonic outputs, 10 kHz to 40 GHz RS101 Radiated susceptibility, magnetic field, 30 Hz to 100 kHz RS103 Radiated susceptibility, electric field, 10 kHz to 40 GHz RS105 Radiated susceptibility, transient electromagnetic field ? 2000 by CRC Press LLC There are different equipment and subsystem classes defined for the various environments in which they are to be installed into. Table 40.6 gives the descriptions of the different classes and applicapability in MIL-STD- 461D. If the requirement is applicable, three letters (A, L, S) are assigned in the matrix entry. The letter A means the equipment must meet the particular requirement. An L means consulting more detail of the requirement which may have different limits due to type of equipment and installation environment. An S means depending on the procurement requirement. From Table 40.6 RE102, RS103, CE102, CS101, and CS114, segments are required for all types of equipment and platforms. In this Handbook only RE102 and CE102 limits will be described. The RE102 encompasses electric field limits as shown in Figure 40.16 a, b, and c. The frequency bands of the testing requirements typically cover from 10 kHz to 18 GHz depending on the clock frequency of the device or types of platforms which is described in Table 40.7. Up to 30 MHz, only the vertical polarization of the electric field will be measured and compared with the limits. Above 30 MHz, both horizontal and vertical field components must be measured and again compared with the limits. The device is in conformance with the RE102 if the electric field intensity in the appropriated frequency band is less than the prescribed limit. The CE102 requirement is applicable to all power cords (either AC or DC source) including returns utilizing five nominal voltage levels. The limit based on a nominal 28 V or less and the limit for 200 V are depicted in Fig. 40.16 d. For the other operating voltage levels (115, 270, and 440 V), the limit is modified by adding a relaxation factor corresponding to the intended line voltage to the 28-V limit. The 220-V limit is 9 dB less stringent than the 28-V limit. The European Union (EU) During the early 1980s, a plan to harmonize the electromagnetic compatibility (EMC) requirements for the European Nations into one sovereign community was introduced. This effort, still growing today, attempts to align commercial, judicial, and financial objectives. There are currently more than 15 countries that participate TABLE 40.6MIL-STD-461D Emission and Susceptibility Requirements Requirement Applicability Equipment and Subsystems Installed In, On, or Launched From the Following Platforms or Installations CE101 CE102 CE106 CS101 CS103 CS104 CS105 CS109 CS114 CS115 CS116 RE101 RE102 RE103 RS101 RS103 RS105 Surface ships A A L A S S S A A A A L A A L Submarines A A L A S S S L A A A A L A A L Aircraft, Army, including flight line A A L A S S S A A L A A L A A L Aircraft, Navy LALASSS AAALALLAL Aircraft, Air Force A L A S S S A A A A L A Space systems, including launch vehicles A L A S S S A A A A L A Ground, Army A L A S S S A L L A L L A Ground, Navy A L A S S S A A A L L A L Ground, Air Force A L A S S S A L A A L A TABLE 40.7RE102 Applicability and Frequency Band of Testing Applied for: Frequency Band Ground 2 MHz to 18 GHz a Ships, surface 10 kHz to 18 GHz a Submarines 10 kHz to 1 GHz Aircraft (Army) 10 kHz to 18 GHz a Aircraft (Air Force and Navy) 2 MHz to 18 GHz a a If the highest clock frequency of the device is less than 1.8 GHz, replace the upper frequency limit to 1 GHz or 10 times the clock frequency, whichever is greater. ? 2000 by CRC Press LLC in the EU requirement for EMI/EMC. Those countries have agreed to the common regulatory requirements placed on commercial products. The agreements, called directives, are listed in the Official Journal of the European Council (OJEC). Different directives for each type of equipment (Telecommunications, Information Technology Equipment [ITE], etc.) spell out the regulatory rules (EMC, Low Voltage, Telecommunications, etc.) required for approval to ship the product into the member countries, which is signified by the application of the conformity mark (CE). The EU has approved a set of standards called the European Norms (EN). One of the EMC directives contains the International Electrotechnical Commission (IEC) test requirements for individual equipment approval. The IEC series of standards currently called upon by the EMC directive for generic equipment are EN 50081-1 (Generic Emission Standard; Part 1, residential, commercial and light industry) and EN 50082-1 (Generic Immunity Standard; Part 1, residential, commercial, and light industry equipment). These standards reference the IEC documents which list the requirements for measurement of emission and immunity characteristics. FIGURE 40.16(a) RE102 electric field limit for surface ship and submarine. FIGURE 40.16(b) RE102 electric field limit for aircraft and space system. ? 2000 by CRC Press LLC The 1000-4-2 is the Electrostatic Discharge (ESD) test standard; 100-4-3 is the immunity standard; and 100- 4-4 is the Electric Fast Transient (EFT) standard. The EN 55022 document contains the limits and test methods for both radiated and conducted emissions of ITE. EN 55022 and IEC 1000 Series (Part 4) EN 55022, also known as CISPR Publication 22, involves the limits and methods of measurement of radio interference characteristics of ITE. Currently, the ITE must meet three Parts (standards) of the IEC 1000-4 series. However, the future revision of the IEC 1000-4 series may mandate an additional eight requirements (presently int he process of review). Consult the latest versions of the applicable generic standard for the current requirements. EN 55022 Document. The EN 55022 procedure describes for measurement of spurious signal strength, in the frequency range 0.15 MHz to 1 GHz, generated by pulsed electrical waveforms either through the power FIGURE 40.16(c) RE102 electric field limit for ground application. FIGURE 40.16(d) The CE102 limits for 28 V and 220 V nominal operating voltages. ? 2000 by CRC Press LLC main cable or through direct radiation. There are two classes of ITE:Class A and Class B. Class A equipment is usually intended for use in commercial establishments whereas Class B equipment is for domestic use. In order to conform with EN 55022, the measured voltages or field strengths shall meet the following limits. Tables 40.8 and 40.9 are the limits of mains terminal interference voltage in the frequency range 0.15 to 30 MHz for Class A and Class B, respectively. Tables 40.10 and 40.11 are the limits of radiated interference field strength in the frequency range 30 MHz to 1 GHz for Class A and Class B equipment. IEC 1000-4 Section 2.The IEC 1000-4-2 document describes the static electricity discharge (ESD) require- ments. At the time of publication, the current limits tabulated herein are utilized; however, the limits are subject for a revision. Since contact discharge method and air discharge method are acceptable for ESD tests, two limits have been specified and tabulated in Table 40.12(a) and (b). IEC 1000-4 Section 3.The IEC 1000-4-3 document involves the suscep- tibility requirements for equipment under test (EUT). The primary concern is the degradation of EUT under the influence of the hand-held transceiver or other sources of radiation in the frequency range 80 to 1000 MHz. Table 40.13 gives the tabulated limits of the susceptibility requirements. TABLE 40.8 Limits of Mains Terminal Interference Voltage in the Frequency Range 0.15 to 30 MHz for Class A Equipment TABLE 40.9Limits of Mains Terminal Interference Voltage in the Frequency Range 0.15 to 30 MHz for Class B Equipment Frequency Range Limits [dB (mV)] Frequency Range Limits [dB (mV)] (MHz) Quasi-peak Average (MHz) Quasi-peak Average 0.15 to 0.50 79 66 0.15 to 0.50 66 to 56 56 to 46 0.50 to 30 73 60 0.50 to 5 56 46 5 to 30 60 50 The lower limit shall apply at the transition frequencies. Note: In Table 40.9 the limit decreases linearly with the logarithm of the frequency in the range 0.15 to 0.50 MHz. TABLE 40.10 Limits of Radiated Interference Field Strength in the Frequency Range 30 MHz to 1 GHz at a Test Distance of 30 m for Class A Equipment TABLE 40.11 Limits of Radiated Interference Field Strength in the Frequency Range 30 MHz to 1 GHz at a Test Distance of 10 m for Class B Equipment Frequency Range Quasi-peak Limits Frequency Range Quasi-peak Limits (MHz) [dB (mV/m)] (MHz) [dB (mV/m)] 30 to 230 30 30 to 230 30 230 to 1000 37 230 to 1000 37 The lower limit shall apply at the transition frequencies. TABLE 40.12(a)Contact Discharge Severity Levels TABLE 40.12(b)Air Discharge Severity Levels Level Test Voltage Contact Discharge (kV) Level Test Voltage Air Discharge (kV) 121 2 242 4 363 8 484 15 x 1 Special x 1 Special 1 x is an open level. The level is subject to negotiations and has to be specified in the dedicated equipment specification. If higher voltages than those shown are specified, special test equipment may be needed. TABLE 40.13Severity Levels for Susceptibility Requirements Level Test Field Strength, (V/m) 11 23 310 x 1 Special 1 x is an open class. ? 2000 by CRC Press LLC There are four classes of severity levels. Class 1 is for low-level electromagnetic radiation environments, such as those typical of local radio/television stations located more than 1 km away and levels typical of low-power transceivers. Class 2 is for moderate electromagnetic radiation environments, such as portable transceivers that are close to the EUT but not closer than 1 m. Class 3 is for severe electromagnetic radiation environments, such as levels typical of high-power transceivers in close proximity to the EUT. Class 4 is an open class for situations involving very severe electromagnetic radiation environments. The manufacturer sets the level relative to the environmental conditions into which the EUT would be installed. IEC 1000-4 Section 4.The IEC 1000-4-4 document involves electrical fast transient/burst requirements. The purpose of this requirement is to evaluate the performance of the EUT when exposed to switching transients with high repetition frequency which may couple into the power main supply and external communication lines. Table 40.14 gives the test severity levels recommended at the time of publication. The repetition rate of the impulses is 5 kHz with a tolerance of ±20% for Levels 1 through 4, with an exception of 2.5 kHz for Level-4 power supply port. Measurement Procedures To determine the emission or susceptibility levels, measurement procedures were established. There are many protocols existing around the world. Each country may adopt its own measurement guidelines. In this section, only the FCC procedure will be described in detail. A list of some other procedures is given below. FCC The procedures used by the Commission to determine compliance are as follows: l.FCC/OET MP-l: FCC Measurements for Determining Compliance of Radio Control and Security Alarm Devices and Associated Receivers. 2.FCC/OET MP-2: Measurement of UHF Noise Figures of TV Receivers. 3.FCC/OET MP-3: FCC Methods of Measurement of Output Signal Level, Output Terminal Conducted Spurious Emissions, Transfer Switch Characteristics, and Radio Noise Emissions from TV Interface Devices. 4.ANSI C63.4, 1992: FCC Procedure for Measuring RF Emissions from Computing Devices. 5.FCC/OET MP-9: FCC Procedure for Measuring Cable Television Switch Isolation. In addition to the documents listed above, the FCC has outlined some generic measurement characteristics in the CFR, Title 47, Part 15, Sections 15.31 through 15.35. Listed in the American National Standards Institute (ANSI) Specification C63.4-1992 measurement proce- dure document are the configuration setups for Class A and B computing devices and peripheral equipment. Some of the pertinent highlights for Class B computing devices are as follows: ?The equipment must be set up in a system configuration which includes the computer controller, a monitor, keyboard, serial device, parallel device, and any other device which may typically be connected to the system. TABLE 40.14Severity Levels for the Fast Transient/Burst Requirements Open Circuit Output Test Voltage ±10% (kV) On I/O (Input/Output) Level On Power Supply Signal, Data and Control Lines 1 0.5 0.25 2 1 0.5 32 1 44 2 x 1 Special Special 1 x is an open level. The level is subject to negotiation between the user and the manufacturer or is specified by the manufacturer. ? 2000 by CRC Press LLC ?The computer controller must be configured with the peripheral cards needed for typical operation (serial card, parallel card, video card, disk controller, memory), along with any other specialized cards defined in the typical setup to be marketed. ?A program to display, print, store, and/or send capital H characters to all of the pertinent devices (inclusive of the drives, CRT, printer, and any other data receiving devices) must be run for the duration of the evaluation. ?Typical cables and power cords are required for the test; the cables are bundled serpentine fashion at the center of the cable in 30- to 40-cm bundle lengths, to an overall length of 1 meter. Conducted Emission Testing.Measurements are recommended to be performed inside of an RF shielded room in order to eliminate interference from ambient electromagnetic fields. The system units are placed on a nonconducting table 0.8 m high and 10 cm from the rear wall of the RF shielded enclosure. The measurements are performed with a line impedance stabilization network (LISN). This type of network is specifically designed to present a known impedance to the device under test, filter the noise present on the power line, and to match impedances with the measurement receiver. Figure 40.17 shows the typical FCC LISN circuit, applicable for monitoring the conducted noise present on either the phase or neutral line. Data is collected across phase and neutral to ground over the 450-kHz to 30-MHz frequency range and is compared with the aforementioned limits. FIGURE 40.17LISN circuit diagram. FIGURE 40.18Minimum obstruction-free area for open field test site. ? 2000 by CRC Press LLC Radiated Emission Testing.The radiated emissions are measured at an FCC listed site (either semi-anechoic or open field), which requires a metal ground plane over the floor (typically hardware cloth). The site must satisfy a certain minimum size criteria, depending on the prescribed measurement distance. The accepted criteria is based on the Fresnel ellipse, which is presented in Fig. 40.18. The procedure for listing a site with the FCC includes the submittal of the site attenuation measurements, the site description, and a list of mea- surement equipment. The qualified site should meet the ±4-dB variation from theoretical values. The EUT is placed on a nonconducting table 0.8 m above the ground plane floor. The receiving antenna is placed at the prescribed measurement distance (R) from the system (3 m for Class B and 10 m for Class A) and is scanned from 1 to 4 m in height while the EUT is rotated 360 degrees. The maximum emission data (per azimuth, elevation, and antenna orientation) is collected over the appropriate frequency range. Abbreviations ANSI: American National Standards Institute BSI: The British Standards Institute CFR: Code of Federal Regulations CISPR:International Special Committee on Radio Interference EMC: Electromagnetic compatibility EMI: Electromagnetic interference EN: European Norms ESD: Electrostatic discharge EU: European Union EUT: Equipment under test FCC: The Federal Communications Commission IEC: International Electrotechnical Commission IEEE: The Institute of Electrical and Electronics Engineers ITE: Information technology equipment LISN: Line impedance stabilization network MIL: The U.S. Military Defining Terms Conducted emission: An RF current propagated through an electrical conductor. Electromagnetic interference: An unwanted electromagnetic signal which may degrade the performance of an electronic device. Far field:The region where the ratio of the electric to magnetic field is approximately equal to 377 W. Field strength: An amount of electric or magnetic field measured in far-field region and expressed in volts/meter or amps/meter. Line impedance stabilization network (LISN):A network designed to present a defined impedance at high frequency to a device under test, to filter any existing noise on the power mains, and to provide a 50-W impedance to the noise receiver. Radiated emission:An electromagnetic field propagated through space. Related Topics 16.1 Spectral Analysis?73.2 Noise References Air Force Systems Command Electromagnetic Compatibility Handbook, 3rd ed., January 5, 1975. CEI International Standard, IEC 801-1 to 801-4, 2nd ed., 1991. Code of Federal Regulations, Title 47, Telecommunications, Part 15, October, 1995. ? 2000 by CRC Press LLC Electromagnetic Interference and Compatibility Handbook, vol. 1, Donald White Consultants, Inc., 1971. Military Standard 461C, Requirements for the Control of Electromagnetic Interference Emissions and Suscep- tibility, January, 1993. C.R. Paul, Introduction to Electromagnetic Compatibility, New York: John Wiley & Sons, 1992. Further Information The aforementioned measurement procedures used by the FCC are available from the Government Printing Office, Washington, D.C., 20402. The ANSI C63.4 document is available from the IEEE, 345 East 47th Street, New York, NY, 10017-2394. The procedures utilized for the measurements performed to military EMC specifications are given in the following documents which are available from the Naval Publications and Forms Center, NPODS, 700 Robbins Ave., Philadelphia, Pa., 19111-2394. MIL-STD-285 Attenuation Measurements for Enclosure, EM Shielding, for Electronic Test Purposes, Method of MIL-STD-462D Electromagnetic Emission and Susceptibility, Measurement of Electromagnetic Interference Characteristics. MIL-STD-463 Definitions and System of Units, Electromagnetic Interference and Electromagnetic Compat- ibility MIL-STD-1377 Effectiveness of Cable, Connector and Weapon Enclosure Shielding and Filters in Precluding Hazards of Electromagnetic Radiation to Ordnance, Measurement of The EC procedures and standards listed below are available from the Bureau Central de la Commission Electrotechnique Internationale 3, rue de Varembe, Geneve, Suisse. EN 50081-1 Electromagnetic Compatibility—Generic Emission Standard; Part 1: Residential, Commercial and Light Industry. EN 50082-1 Electromagnetic Compatibility—Generic Immunity Standard; Part 1: Residential, Commercial and Light Industry EN55022 Limits and Methods of Measurement of Radio Interference Characteristics of Information Tech- nology Equipment IEC 1000-4-2 Electrostatic Discharge Immunity Test IEC 1000-4-3 Radiated, Radio-Frequency, Electromagnetic Field Immunity Test IEC 1000-4-4 Electrical Fast Transient/Burst Immunity Test 40.3 Lightning Martin A. Uman and Marcos Rubinstein An understanding of lightning and of the electric and magnetic fields produced by lightning is critical to an understanding of lightning-induced effects on electronic and electric power systems. This section begins with an overview of the terminology and physics of lightning. Then, statistics on lightning occurrence are given. Next the characteristics of the electric and the magnetic fields resulting from lightning charges and currents are examined, and the models used to describe that relationship are discussed. The section ends with a discussion of the coupling of the electric and magnetic fields from lightning to overhead wires. Terminology and Physics Lightning is a transient, high-current electric spark whose length is measured in kilometers. Lightning discharges can occur within a cloud, between clouds, from cloud to air, and from cloud to ground. All discharges except the latter are known as cloud discharges. The usual cloud-to-ground lightning is initiated in the cloud, lasts about half a second, and lowers to ground some 20 to 30 Coulombs of negative cloud charge. A less frequent type of cloud-to-ground discharge, accounting for less than 10% of all cloud-to-ground lightning, also begins ? 2000 by CRC Press LLC in the cloud but lowers positive cloud charge. An even less frequent type of cloud-to-ground lightning is initiated in an upward direction from tall man-made structures such as TV towers or tall geographical features such as mountaintops. A complete lightning discharge of any type is called a flash. The usual negative cloud-to-ground lightning flash starts in the cloud when a so-called preliminary breakdown, a particular type of electric discharge in the cloud, occurs. This process is followed by a discharge, termed the stepped leader, that propagates towards the ground in a series of luminous steps tens of meters in length. In progressing toward the ground, the negatively charged stepped leader branches in a downward direction. When one or more leader branches approach within a hundred meters or so of the ground, after 10 to 20 ms of stepped leader travel at an average speed of 10 5 to 10 6 m/s, the electric field at the ground (or at objects on the ground) increases above the critical breakdown field of the surrounding air and one or more upward-going discharges is initiated, starting the attachment process. After traveling a few tens of meters, one of the upward-going discharges, which is essentially at ground potential, contacts the tip of one branch of the stepped leader, which is at a high negative potential, probably some tens of megavolts. From that point, ground potential propagates upward, discharging to ground some or all of the negative charge previously deposited along the channel by the stepped leader. This upward propagating potential discontinuity is called the return stroke. Its front is a region of high electric field that causes increased ionization, current, temperature, and pressure as it travels the 5-km or more length of the leader channel. That trip is made in about 100 ms at an initial speed of the order of one third to one half the speed of light, the speed decreasing with height. The current at ground associated with the negative first return stroke has a peak of typically 35 kA achieved in a few microseconds, has a maximum current derivative of about 10 11 A/s and falls to half of peak value in some tens of microseconds. The cessation of the first return stroke current may or may not end the flash. If more cloud charge is made available to the first stroke channel by in-cloud discharges, another leader-return stroke sequence may ensue, typically after tens of milliseconds. Preceding and initiating a subsequent return stroke is a continuous leader lowering negative charge, called a dart leader. The dart leader typically propagates down the residual channel of the previous stroke, generally ignoring the first stroke branches, although in about 50% of cloud-to-ground flashes there is at least one dart leader which transforms to a stepped leader on the downward trip, creating a new path to ground. There are typically three or four leader-return stroke sequences per negative cloud-to-ground flash, but ten or more is not uncommon. Of the many different processes that occur during the various phases of a negative cloud-to-ground lightning (e.g., the in-cloud K processes, in-cloud J processes, and cloud to ground M components that occur between strokes and after the final stroke and are not discussed here), the electric and magnetic fields associated with the return stroke described above generally are the largest and hence the most significant in inducing unwanted voltages in electronic and electric power systems. This is the case because the currents in all other lightning processes are generally smaller than return stroke currents and the ground strike point of the return stroke can be much closer to objects on the ground than are in-cloud discharges. Cloud discharges exhibit currents similar to those of the in-cloud processes occurring in ground discharges and hence produce similar relatively small fields at or near ground level. Positive flashes to ground, those initiated in the cloud and lowering positive charge to earth, generally contain only one return stroke, which is preceded by a “pulsating” leader rather than the stepped leader characteristically preceding negative first strokes and is generally followed by a period of continuous current flow. Positive flashes contain a greater percentage of very large return stroke currents, in the 100- to 300-kA range, than do negative flashes. Positive flashes may represent half of all flashes to ground in winter storms, which produce few total flashes, and typically represent 1 to 20% of the overall flashes in summer storms, that percentage increasing with increasing latitude. Lightning Occurrence Statistics Lightning flash density is defined as the number of lightning flashes per unit time per unit area and is usually measured in units of lightning flashes, either cloud or cloud-to-ground or both, per square kilometer per year. The two most common techniques for directly measuring flash density are (1) the use of so-called flash counters, relatively crude devices which trigger on electric fields above a value of the order of 1 kV/m in a frequency band centered in the hundreds of hertz to kilohertz range, of which two models are extensively used, the CIGRE ? 2000 by CRC Press LLC 10-kHz and the CIGRE 500-Hz and (2) the use of networks of wideband magnetic direction finders, networks of wideband time-of-arrival detectors, and networks combining the two technologies, such networks now covering the U.S., Canada, Japan, Korea, Taiwan, most of Europe, and parts of many other countries. The average flash density varies considerably with geographical location, generally increasing with decreasing lati- tude. Typical ground flash densities are 1 to 5 km –2 yr –1 , with the world’s highest being 30 to 50 km –2 yr –1 . Significant variations in flash density are observed with changes in local meteorological conditions within distances of the order of 10 km, for example, perpendicular to and inland from the Florida coastline. A ground flash density map of the U.S. for 1989, obtained from the U.S. National Lightning Detection Network of 114 wideband magnetic direction finders, is given by Orville [1991]. Flash densities in the U.S. are maximum in Florida with 10 to 15 km –2 yr –1 and minimum along portions of the Pacific coast which has essentially no lightning. An extensively measured parameter used to describe lightning activity worldwide is the thunderday or isokeraunic level, T D, the number of days per year that thunder is heard at a given location. This parameter has been recorded by weather station observers worldwide for many decades, whereas the accurate direct measure- ment of flash density has been possible only recently. Commonly used relations to convert thunderday level to ground flash density N g are of the form N g = aT b D km –2 yr –1 (40.6) where the value of a is near, and usually less than, 0.1 and the value of b is near, and usually greater than, 1. It should be noted that Eq. (40.6) is relatively inaccurate in that the data to which it is a fit is highly variable. The literature contains more than ten different values of a and b determined by different investigators. Finally, from both worldwide thunderday and earth-orbiting satellite measurements, it has been estimated that there are about 100 total flashes, cloud and cloud-to-ground, per second over the whole earth. This number corresponds to an average global total flash density of 6 km -2 yr -1 . Electric and Magnetic Fields For the usual negative return stroke, measurements of the vertical component of the electric field and the two horizontal components of the magnetic field at ground level using wide- band systems with upper frequency 3-dB points in the 1- to 20-MHz range are well documented in the literature. Mea- sured vertical electric field and horizontal magnetic field waveshapes are shown in Fig. 40.19. Sketches of typical elec- tric and magnetic fields are given in Fig. 40.20 for lightning in the 1- to 5-km range and in Fig. 40.21 for lightning at 10, 15, 50, and 200 km. Measured vertical and measured hori- zontal electric fields near ground are shown in Fig. 40.22. The mean value of the initial peak vertical electric field, normal- ized to 100 km by assuming an inverse distance dependence, is about 7 V/m for negative first strokes and about 4 V/m for negative subsequent strokes. The return stroke vertical electric field rise to peak is com- prised of two distinguishable parts, evident in Fig. 40.19: a slow front immediately followed by a fast transition to peak. For first strokes the slow front has a duration of a few micro- seconds and rises to typically half the peak amplitude, while for subsequent strokes the same slow front lasts less than 1 ms and rises only to typically 20% of the peak. The mean 10–90% fast transition time is about 200 ns regardless of FIGURE 40.19 Simultaneously measured return stroke vertical electric field (E) and two horizontal magnetic flux densities (B EW and B NS ) as observed about 2 and 50 km from a two-stroke flash, the first stroke being designated “1”, the second “2”. (Source: Adapted from Y.T. Lin et al., J. Geophys. Res., vol. 84, pp. 6307–6314, 1979. With permission.) ? 2000 by CRC Press LLC FIGURE 40.20 Drawings of typical return stroke electric fields and magnetic flux densities at 1, 2, and 5 km with definition of pertinent characteristic features. Solid lines represent first strokes, dotted subsequent strokes. (Source: Adapted from Y.T. Lin et al., J. Geophys. Res., vol. 84, pp. 6307–6314, 1979. With permission.) FIGURE 40.21 Drawings of typical return stroke electric fields and magnetic flux densities at 10, 15, 50, and 200 km; a continuation of Fig. 40.24. (Source: Adapted from Y.T. Lin et al., J. Geophys. Res., vol. 84, pp. 6307–6314, 1979. With permission.) ? 2000 by CRC Press LLC stroke order for strokes observed over saltwater where there is minimal distortion of the waveform due to propagation. The waveforms in Fig. 40.19 have suffered distortion in propagating over land. After the initial field peak, the waveshapes of the vertical electric field and the horizontal magnetic field for close lightning exhibit a valley followed by a hump in the case of the magnetic field and by a ramp in the case of the electric field, as is evident from Figs. 40.19 through 40.22. Relative to the amplitude of the initial peak, the hump and the ramp decrease with increasing distance of the return stroke. For distances of 25 km or greater, the ramp in the electric field is no longer significant, and for distances of 50 km or more and for times of the FIGURE 40.22 Measured horizontal electric field components (EN and EE) and vertical electric field (EV) one meter above ground for a first stroke (a) and a subsequent stroke (b) at a distance of 7 km presented on two time scales. (Source: Adapted from E.M. Thomson et al., J. Geophys. Res., vol. 93, pp. 2429–2441, 1988. With permission.) ? 2000 by CRC Press LLC order of 100 ms, the waveshapes of the electric and magnetic fields are nearly identical, exhibiting a zero crossing and polarity reversal at some tens of microseconds. For positive return strokes, there are more very large peak currents at the channel base, in the 100-kA range, than for return strokes lowering negative charge to ground, although the median value for both positives and negatives is not much different [Berger et al., 1975]. This observation is supported by measurements of the initial peak magnetic field from positive and negative return strokes made with magnetic direction-finding networks worldwide, where various investigations have found the mean peak positive field to be about twice the mean peak negative. The horizontal component of the electric field has not been as well studied or characterized as the vertical. For the case of a finite-conducting earth and lightning beyond a few kilometers, Thomson et al. [1988] give wideband measurements of the three perpendicular components of the electric field about 1 m above ground level. An example is shown in Fig. 40.22. The horizontal field waveshapes are more impulsive and vary on a faster time scale than their associated vertical electric field waveshapes. In fact, the horizontal field appears to be a crude derivative of the vertical. The peak amplitudes of the horizontal electric fields are on the order of 30 times smaller than those of the vertical fields for ground conductivities of the order of 10 -2 mho/m, this ratio being roughly proportional to the square root of the ground conductivity. The horizontal field, although considerably smaller for distant lightning, can be as important as the vertical electric field in inducing voltages on an overhead horizontal wire because of the greater horizontal extent of the wire relative to its height above ground, a fact well established by recent research, whereas in the earlier literature on power line coupling, for example, only the vertical field was considered to be important. The so-called wavetilt formula, given in Eq. (40.7), models the ratio, in the frequency domain, of the horizontal to vertical electric field of a plane wave at the surface of an earth of conductivity s and permittivity e r e 0 for the case of grazing incidence and is certainly applicable to lightning return strokes occurring beyond a few kilometers, probably beyond a few hundred meters. (40.7) To the best of our knowledge, no horizontal or vertical electric fields very close to natural lightning, at distances from tens to hundreds of meters, necessary to the understanding of the voltages induced by very close lightning, have been published, although such close fields have been calculated by Diendorfer [1990] and by Rubinstein et al. [1990] using different return stroke models. These two sets of calculated fields are to be considered model-dependent estimates. Although there is disagreement between the two studies as to the waveshape of fields and in how appreciable is the influence of a finite ground conductivity at small distances, both studies yield horizontal field amplitudes at the height of a typical power distribution line comparable to the amplitude of the vertical field. Note that no return stroke model used to date (see next section) takes proper account of the attachment process referred to earlier and hence probably none accurately models the fields at very early times. Further, the leader fields preceding the return stroke field change are not taken into account in the existing models, although such fields at very close range are clearly important since it is the leader charge near ground that the return stroke discharges to ground, and hence the leader and return stroke electrostatic field changes should be of equivalent magnitude very close to the ground strike point (Rubinstein et al., 1995). Modeling of the Return Stroke General A number of return stroke current models are found in the literature from which, if the current at the channel base is specified (e.g., from measurement) along with the model parameters, the channel current can be calculated as a function of height and time: the Bruce–Golde (BG) model, the transmission line (TL) model, the modified transmission line (MTL) model, the traveling current source (TCS) model, the Lin–Uman–Stan- dler (LUS) model, the Diendorfer–Uman (DU) model, and the modified DU model. Two assumptions are E E j H V r () () w w s w = + ? è ? ? ? ÷ 1 e e o ? 2000 by CRC Press LLC common to all of these models: that the lightning channel is perfectly straight and vertical and that the ground is a perfect conductor. Once the channel currents are determined as a function of height and time, the remote electric and magnetic fields can be calculated from Eqs. (40.8) through (40.14) (40.8) (40.9) (40.10) (40.11) (40.12) (40.13) (40.14) where i(z￠,t) is the current along the channel obtained from one of the return stroke current models mentioned above, and the geom- etry by which the above equations are to be interpreted is shown in Fig. 40.23. Note that the spatial integral includes the image current below the perfectly conducting ground plane so as to take account of reflections from the earth’s surface. The three terms on the right- hand side of Eq. (40.8) [expanded in Eqs. (40.9) through (40.11)] are called, from left to right, the electrostatic, induction, and radia- tion terms. Similarly, the two terms on the right-hand side of Eq. (40.12) [expanded in Eqs. (40.13) and (40.14)] are termed the induc- tion and radiation terms. For large distances to the lightning channel, the radiation part of the electric and magnetic fields is dominant due to its 1/R depen- dence (as compared to the 1/R 2 and 1/R 3 dependencies of the induc- tion and electrostatic terms, respectively). By a similar argument, for close distances, the dominant terms will be the electrostatic term in the case of the electric field and the induction term for the magnetic field. It can be readily shown from Eqs. (40.8) through (40.14) and the preceding discussion that for any individual lightning return stroke model, the waveshapes of the vertical electric field and the horizontal magnetic field are almost identical for great distances, and this fact is also evident in the experimental data (see Fig. 40.21). Moreover, it can be shown that for great distances, the ratio of the electric field intensity E to the magnetic flux density B is the speed of light c. EEEE=++ ele ind rad E aa R iz R c dtdz R t h h ele = ￠ + ￠ ￠ ? è ? ? ? ÷ ì í ? ? ? ü y ? t ? ￠ òò 1 4 2 0 3 0p qq t q e cos ? sin ? ,– – uu E aa cR izt R c dz R h h ind = ￠ + ￠ ￠ ? è ? ? ? ÷ ￠ ￠ ò 1 4 2 0 2 p qq q e cos ? sin ? ,– – uu E cR izt R c t adz h h rad = ￠ ? è ? ? ? ÷ ￠ ￠ ò 1 4 1 0 2 p ? ? q e uu,– ? – BBB=+ ind rad B R izt R c adz h h ind = ￠ ￠ ? è ? ? ? ÷ ￠ - ￠ ò m p q f 0 2 4 sin ,– ?uu B cR izt R c t adz h h rad = ￠ ￠ ? è ? ? ? ÷ ￠ - ￠ ò m p q ? ? f 0 4 sin ,– ? uu FIGURE 40.23The geometry for the cal- culation of the fields using Eqs. (40.8) through (40.14). ? 2000 by CRC Press LLC A brief examination of return stroke current models follows. We discuss here only the transmission line (TL) model and the model of Diendorfer and Uman [1990] along with its modified version. More details of all models are found in Nucci et al. [1990], Diendorfer and Uman [1990], and Thottappillil et al. [1991], including fields calculated from the various models. The Transmission Line Model In the transmission line model it is assumed that the current waveform at the ground travels undistorted up the lightning channel at a constant speed v. Mathematically, this current is represented by i(z￠,t) = i(0,t – z￠/v) z￠ < vt (40.15) i(z￠,t) = 0 z￠ > vt No charge is removed by the transmission line return stroke current along the channel since the charge entering the bottom of any section of the channel leaves the top when the current reaches it. All the charge is therefore transferred from the bottom to the top of the channel, an unrealistic situation given our knowledge of lightning physics. Willett et al. [1989] have presented return stroke current, field, and speed data from artificially initiated (by firing small rockets trailing grounded wires) lightning in an attempt to validate the TL model. Using these data, Rakov et al. [1992] have shown, at least for subsequent strokes in artifically initiated lightning, that return stroke peak current can be derived from return stroke peak field by the expression I = 1.5 – 0.037DE where the peak current I is in kA and is negative, the distance D is in km, and the peak electric field E is in V/m and is positive. Several investigators have published lightning peak current statistics derived from the magnetic radiation fields recorded by networks of magnetic direction finders by making use of the transmission line model. These studies are discussed by Rakov et al. [1992]. The Diendorfer–Uman (DU) Model and a Modification of It The DU model [Diendorfer and Uman, 1990] is a physically reasonable model that can predict the salient features of the measured lightning electric and magnetic fields. Given the return stroke current at ground level, the channel current above ground is assumed to discharge the leader by way of two independent processes: (1) the discharge of the highly ionized core of the leader channel, termed the breakdown discharge process, with a time constant of 1 ms or less, and (2) the discharge of the corona envelope with a larger time constant. In both cases, the discharge at a height z￠ starts when the return stroke front, assumed to travel up at a constant speed v, arrives at z￠. The liberated currents are assumed to flow to the ground at the speed of light. For a current at ground i(0,t), Diendorfer and Uman [1990] show that the current as a function of height and time is i(z￠,t) = i(0,t m ) – i(0,z￠/v*) exp(–t e /t) (40.16) where t m = (t + z￠/v), t e = (t – z￠/v), v* = v (1 + v/c) and t is the discharge time constant. The Diendorfer–Uman model described above assumes that the return stroke propagates up the lightning channel at a constant speed and that the current from activated sections of the channel travels to ground at the speed of light. An analytical generalization of the DU model which allows for the return stroke speed and the downward current speed to be arbitrary functions of height has been presented by Thottappillil et al.[1991]. Lightning-Overhead Wire Interactions General Lightning interactions with overhead wires such as power distribution lines are a major source of electromag- netic compatibility problems, resulting in inferior power quality, power outages, and damaged electronics. Only a small fraction of all the cloud-to-ground lightning flashes directly strike overhead lines, making induced overvoltages a significant source of power disturbances. This section begins with a discussion of the appropriate transmission line equations. Then, examples of measured lightning-induced voltages on overhead lines as well as calculated voltages are presented. ? 2000 by CRC Press LLC Transmission Line Equations The transmission line equations for a nonuniform electromagnetic field impinging on a system of horizontal wires have been derived in the time domain by Agrawal et al. [1980], who adapted the theory to the case of wires above an imperfectly conducting ground. The main advantage of a time domain model over an equivalent frequency domain model is its applicability to cases of time varying and nonlinear loads and its ability to account for multiple reflections on a line with two or more discontinuities. On the other hand, with a frequency domain model it is intrinsically easier to handle frequency-dependent parameters such as the ground impedance. The derivation of the time domain coupling equations is conceptually simple: Maxwell’s equations are first integrated over closed cylindrical surfaces and along closed rectangular paths. The resulting integral equations, which are in terms of electric and magnetic fields, are then recast in terms of voltages and currents. One version of the transmission line equations, due to Agrawal et al. [1980], follows: (40.17) (40.18) (40.19) where the superscript s identifies the “scattered” quantities, the superscript i identifies the “incident” quantities, the superscript t identifies the total, measurable quantities, and the asterisk is the convolution operator. In these equations, the only source along the horizontal portion of the line is the horizontal component of the incident electric field. At the line terminations, the boundary condition and the termination current, I, are used to determine the end voltage. At those vertically oriented terminations, the vertical electric fields drive currents through the terminations into the line. The total voltage, V t (x,t), at the line terminations must equal I T *Z T at all times, where Z T is the termination impedance. Equations (40.17) through (40.19) can be represented by the circuit model in Fig. 40.24. Two basic assumptions are used to arrive at Eqs. (40.17) through (40.19): (1) The response of the power line (scattered voltages and currents) to the impinging EM wave (incident field) is quasi-TEM (i.e., the scattered fields can be approximated as transverse electromagnetic). This allows us to define a “static” voltage along the line and to relate the line current and the scattered magnetic flux by an inductance, as well as the line scattered voltage and charge by a capacitance. (2) The transverse dimensions of the line system are small compared to the minimum wavelength, l min , of the excitation wave, and the height of the line is much larger than the diameter of the wire. FIGURE 40.24Equivalent circuit model obtained from Eqs. (40.17) through (40.19). ? ? ? ? Vxt t ZIxtL Ixt t Exzht s gx i (,) *(,) (,) (, ,)++== ? ? ? ? Ixt x C Vxt t s (,) (,) +=0 VVV ExztdzV tis z i h s =+= + ò –(,,) 0 ? 2000 by CRC Press LLC For other formulations of the overhead wire coupling equations, written in terms of field variables different from those used by Agrawal et al. [1980], see Rachidi [1993], Cooray [1994], and Nucci and Rachidi [1995]. Measured and Calculated Lightning-Induced Voltages on Overhead Wires Several experiments have been carried out to test the coupling theory [e.g., Georgiadis et al., 1992; Rubinstein et al., 1994; Barker et al. 1996]. The basic strategy is the same in each experiment: to measure the lightning electric and magnetic fields in the vicinity of an instrumented overhead line while simultaneously measuring the voltages induced on the line, the measured fields then being used as inputs to a computer program written to solve Eqs. (40.17) through (40.19) and the computer-calculated voltage waveforms being compared with the measured voltage waveforms. The following discussion illustrates the types of voltage waveforms induced on overhead wires by lightning beyond a few kilometers and the degree of agreement that has been obtained in the coupling-model calculations. Examples of voltages induced on a 450-m overhead line about 10 m above the ground are shown in Fig. 40.25. Each line end was either terminated in its characteristic impedance or FIGURE 40.25Examples of simultaneously measured lightning-induced voltages at the east end (E) and west end (W) of a 450-m line. Each line end is either open or terminated in its characteristic impedance, as noted. Directions to the lightning are determined from the ratio of the horizontal magnetic flux densities (B’s) or from a commercial lightning location system (LLP). (Source: Adapted from N. Georgiadis et al., “Lightning-induced voltages at both ends of a 450-meter distribution line,” IEEE Trans. EMC, vol. 34, pp. 451–460, 1992. ó1992 IEEE. With permission.) ? 2000 by CRC Press LLC open-circuited (four different cases), and voltages were measured simultaneously at each end. Figures 40.26 and 40.27 contain specific examples of measured and calculated voltage waveforms at each line end as well as the measured vertical electric field and calculated horizontal electric field via Eq. (40.7). It is clear from Figs. 40.25 through 40.27 that the induced voltage polarities and waveshapes are strongly dependent on the angle to the lightning and on the line end terminations. It is apparent also from Georgiadis et al. [1992] that while measured and calculated voltage waveshapes are in good agreement, the measured voltage amplitudes are, on average, a factor of three smaller than calculated voltages. This amplitude discrepancy remains unex- plained but is probably due to the fact that the fields reaching the power line were shielded by trees along the line whereas the fields measured were in an open area and hence were unshielded. FIGURE 40.26Measured and calculated voltages at the east and west ends of a 450-m line for both line ends open. Although the direction to the lightning as determined from the ratio of the two magnetic flux density (B) components was 40°, as shown in Fig. 40.29(a), the best calculated fit to the data was found for 65° as shown, the angular error apparently being caused by variation in the magnitudes of the magnetic flux density components due to nearby conductors as determined from comparing azimuths computed from the B’s and from a commercial lightning location system (LLP). (Source: Adapted from N. Georgiadis et al., “Lightning-induced voltages at both ends of a 450-meter distribution line,” IEEE Trans. EMC, vol. 34, pp. 451–460, 1992. ó1992 IEEE. With permission.) ? 2000 by CRC Press LLC Defining Terms Attachment process: A process that occurs when one or more stepped leader branches approach within a hundred meters or so of the ground and the electric field at the ground increases above the critical breakdown field of the surrounding air. At that time one or more upward-going discharges is initiated. After traveling a few tens of meters, one of the upward discharges, which is essentially at ground potential, contracts the tip of one branch of the stepped leader, which is at a high potential, completing the leader path to ground. Dart leader: A continuously moving leader lowering charge preceding a return stroke subsequent to the first. A dart leader typically propagates down the residual channel of the previous stroke. Flash: A complete lightning discharge of any type. Preliminary breakdown: An electrical discharge in the cloud that initiates a cloud-to-ground flash. Return stroke: The upward propagating high-current, bright, potential discontinuity following the leader that discharges to the ground some or all of the charge previously deposited along the channel by the leader. FIGURE 40.27 Measured and calculated voltages at the east and west ends of a 450-m line with the west end terminated and the east end open. The azimuth was determined from LLP data and is shown in Fig. 40.29(b). (Source: Adapted from N. Georgiadis et al., “Lightning-induced voltages at both ends of a 450-meter distribution line,” IEEE Trans. EMC, vol. 34, pp. 451–460, 1992. ó1992 IEEE. With permission.) ? 2000 by CRC Press LLC Stepped leader: A discharge following the preliminary breakdown that propagates from cloud towards the ground in a series of intermittant luminous steps with an average speed of 10 5 to 10 6 m/s. Negatively charged leaders clearly step, while positively charged leaders are more pulsating than stepped. Related Topic 33.1 Maxwell Equations References A. K. Agrawal, H. J. Price, and S. H. Gurbaxani, “Transient response of multiconductor transmission lines excited by a non-uniform electromagnetic field,” IEEE Trans. EMC, vol. EMC-22, pp. 119–129, 1980. P. Barker, T. Short, A. Eybert-Berard, and J. Berlandis, “Induced voltage measurements on an experimental distribution line during nearby rocket triggered lightning flashes,” IEEE Trans. Pow. Delivery, Vol. 11, pp. 980–995, 1996. K. Berger, R. B. Anderson, and H. Kroninger, “Parameters of lightning flashes,” Electra, vol. 80, pp. 23–37, 1975. V. Cooray, “Calculating lightning-induced voltages in power lines: a comparison of two coupling models,” IEEE Trans. EMC, vol. 36, pp. 170–182, 1994. G. Diendorfer, “Induced voltage on an overhead line due to nearby lightning,” IEEE Trans. Electromag. Comp., vol. 32, pp. 292–299, 1990. G. Diendorfer and M. A. Uman, “An improved return stroke model with specified channel-base current,” J. Geophys. Res., vol. 95, pp. 13,621–13,644, 1990. N. Georgiadis, M. Rubinstein, M. A. Uman, P. J. Medelius, and E. M. Thomson, “Lightning-induced voltages at both ends of a 450-meter distribution line,” IEEE Trans. EMC, vol. 34, pp. 451–460, 1992. Y. T. Lin, M. A. Uman, J. A. Tiller, R. D. Brantley, W. H. Beasley, E. P. Krider, and C. D. Weidman, “Character- ization of lightning return stroke electric and magnetic fields from simultaneous two-station measure- ments,” J. Geophys. Res., vol. 84, pp. 6307–6314, 1979. C. A. Nucci, G. Diendorfer, M. A. Uman, F. Rachidi, M. Ianoz, and C. Mazzetti, “Lightning return stroke current models with specified channel-base current: A review and comparison,” J. Geophys. Res., vol. 95, pp. 20,395–20,408, 1990. C. A. Nucci and F. Rachidi, “On the contribution of the electromagnetic field components in field-to-trans- mission line interaction,” IEEE Trans. EMC, vol. 37, pp. 505–508, 1995. R. E. Orville, “Annual summary—Lightning ground flash density in the contiguous United States—1989,” Monthly Weather Review, vol. 119, pp. 573–577, 1991. F. Rachidi, “Formulation of the field-to-transmission line coupling equations in terms of magnetic excitation field,” IEEE Trans. EMC, col. 35, pp. 404–407, 1993. V. A. Rakov, R. Thottappillil, and M. A. Uman, “On the empirical formula of Willett, et al., relating lightning return stroke peak current and peak electric field,” J. Geophys. Res., vol. 97, pp. 11,527–11,533, 1992. M. Rubinstein, M. A. Uman, E. M. Thomson, and P. J. Medelius, “Voltages induced on a test distribution line by artificially initiated lightning at close range: Measurement and theory,” in Proceedings of the 20th International Conference on Lightning Protection, Interlaken, Switzerland, September 24–28, 1990. M. Rubinstein, M. A. Uman, P. J. Medelius, and E. M. Thomson, “Measurements of the voltage induced on an overhead power line 20 m from triggered lightning,” IEEE Trans. EMC, vol. 36, pp. 134–140, 1994. M. Rubinstein, F. Rachidi, M. A. Uman, R. Thottappillil, V. A. Rakov, and C. A. Nucci, “Characterization of vertical electric fields 500 m and 30 m from triggered lightning, J. Geophys. Res., vol. 100, pp. 8863–8872, 1995. E. M. Thomson, P. Medelius, M. Rubinstein, M. A. Uman, J. Johnson, and J. Stone, “Horizontal electric fields from lightning return strokes,” J. Geophys. Res., vol. 93, pp. 2429–2441, 1988. R. Thottappillil, D. K. McLain, G. Diendorfer, and M. A. Uman, “Extension of the Diendorfer–Uman lightning return stroke model to the case of a variable upward return stroke speed and a variable downward discharge current speed,” J. Geophys. Res., vol. 96, pp. 17,143–17,150, 1991. J. E. Willett, J. C. Bailey, V. P. Idone, A. Eybert-Berard, and L. Barret, “Submicrosecond intercomparison of radiation fields and currents in triggered lightning return strokes based on the transmission-line model,” J. Geophys. Res., vol. 94, pp. 13,275–13,286, 1989. ? 2000 by CRC Press LLC Further Information For more details on the material presented here, see The Lightning Discharge (Academic Press, San Diego, 1987) by M. A. Uman and the review article “Natural and Artificially Initiated Lightning” (Science, vol. 246, 457–464, 1989) by M. A. Uman and E. P. Krider. For the most recent information on return stroke properties and references to previous work, see “Some Properties of Negative Cloud to Ground Lightning vs. Stroke Order” (J. Geophys. Res., vol. 95, 5447–5453, 1990), by V. A. Rakov and M. A. Uman, and “Lightning Subsequent Stroke Electric Field Peak Greater than the First Stroke Peak and Multiple Ground Terminations” (J. Geophys. Res., vol. 97, 7503–7509, 1992), by R. Thottappillil, V. A. Rakov, M. A. Uman, W. H. Beasley, M. J. Master, and D. V. Shelukhin. For more information on lightning properties derived from networks of wideband magnetic direction finders, see “Cloud to Ground Lightning Flash Characteristics from June 1984 through May 1985” (J. Geophys. Res., vol. 92, 5640–5644, 1992), by R. E. Orville, R. A. Weisman, R. B. Pyle, R. W. Henderson, and R. E. Orville, Jr., and “Calibration of a Magnetic Direction Finding Network Using Measured Triggered Lightning Return Stroke Peak Currents” (J. Geophys. Res., vol. 96, 17,135–17,142, 1991), by R. E. Orville. TRIGGERED LIGHTNING he photograph shows lightning that was artificially initiated or “triggered” from a natural thun- derstorm at the University of Florida’s International Center for Lightning Research and Testing at Camp Blanding Army National Guard Base in Florida. Triggered lightning is presently being used to study the close electromagnetic environment of lightning as well as to determine lightning’s effects on communication and power systems. (Photo courtesy of the University of Florida, Department of Electrical and Computer Engineering.) T ? 2000 by CRC Press LLC