1. Basic Concepts Electromagnetic compatibility (EMC) is a crucial quality indicator for electrical and electronic products. Product quality can be considered to consist of two main components: quality specifications and technical indicators. The former involves general specifications, namely international standards such as the IEC and domestic standards formulated by the state; the latter specifies the product's functions and technical requirements. EMC and safety requirements are fundamental standards. Currently, EMC has formed a complete system, from basic standards, general standards, family standards, to product standards. Furthermore, there is specific legislation for this internationally. For example, the European Union has enacted regulations stipulating that from January 1, 1996, electrical and electronic products must obtain Low Voltage Management (LV Directive) and Electromagnetic Compatibility Management (EMC Directive) certifications before they can be sold on the market. In recent years, new EMC standards have been continuously released in China. However, it should be noted that IEC EMC standards are constantly evolving from drafts or older versions to official versions, and national EMC standards are also continuously updated and released. EMC testing should be based on the latest version. EMC, as defined in GB/T4365-1996 "Electromagnetic Compatibility Terminology," refers to the ability of a device or system to function normally in its electromagnetic environment without causing unacceptable electromagnetic interference to anything in that environment. This definition encompasses three aspects: First, the limitation of electromagnetic interference. Electromagnetic interference is ubiquitous, but its harmfulness can be limited by quality specifications and technical means. This means that limits should be set on the intensity of electromagnetic interference emitted by products to ensure a qualified electromagnetic environment. Second, the exemption from electromagnetic interference. This means that products should function normally in an electromagnetic environment with specified electromagnetic interference intensity without compromising their performance. Third, the standardization and compatibility of the electromagnetic environment. This means that any measures taken against electromagnetic interference should not degrade the performance of the product itself or other products or systems in the same electromagnetic environment; they should only coexist peacefully. For example, to reduce conducted interference, a capacitor is connected in parallel between the power supply phase line and the ground line of a device. For the device, the capacitance of this capacitor must meet the leakage current limits required by safety standards; for the system, it must prevent it from becoming a source of interference coupling that affects system operation. Therefore, EMC testing of products should include two main aspects: (1) testing the intensity of electromagnetic interference emitted by the product to the outside world to confirm whether it meets the limit requirements specified in the relevant standards; (2) testing the product's sensitivity under electromagnetic environment conditions with specified electromagnetic interference intensity to confirm whether it meets the immunity requirements specified in the relevant standards. These two aspects are respectively electromagnetic interference or electromagnetic disturbance and immunity to a disturbance in EMC testing items; the latter was previously often referred to as sensitivity. Since the occurrence of electromagnetic interference depends on three factors: the intensity of the interference source, the coupling mode of the interference, and the sensitivity of the instrument and equipment to the interference. Therefore, in addition to classifying the performance requirements and testing methods of EMC according to different natures and types, the relevant standards also classify them into two categories according to the different propagation modes of electromagnetic interference: conducted interference and radiated interference. The former primarily measures the intensity and frequency range of power frequency harmonics and high-frequency noise transmitted outwards by the device under test (DUT) through power lines or signal lines. This falls under the near-field and induced field effects of electromagnetic interference (EMI). The latter measures the intensity and frequency range of radio frequency noise directly radiated outwards by the DUT, primarily addressing the far-field effects of EMI. It is noteworthy that in recent years, international attention has focused particularly on power harmonic interference and equipment immunity requirements. The former relates to environmental requirements for public power grids, while the latter ensures the reliability of equipment or systems. Therefore, many standards have specifically separated power harmonic current content requirements and immunity as two separate technical requirements from EMC projects, making them independent items. It must be pointed out that in the information society, losses due to malfunctions of information technology equipment are often difficult to quantify in monetary terms. Given the impossibility of completely avoiding EMI, improving product immunity under specified electromagnetic environment conditions is of particular importance. Specifically, routine EMC testing conducted by manufacturers not only determines whether the product's EMC performance meets national and industry standards (according to domestic and international literature, it is difficult to pass EMC tests without thorough EMC design and preliminary testing, and sufficient technical measures), but also evaluates the impact of external electromagnetic interference on the product and the effectiveness of related suppression measures. It also identifies the specific causes (sources and pathways) of EMC test damage to the product so that appropriate measures can be taken. Therefore, conducting EMC testing in the early stages of product design finalization is an essential step before a product enters commercialization. EMC testing for power supply products has its own unique requirements, determined by the product's function. Firstly, as the power interface between the power supply (usually AC mains power) and the load it serves (typically information technology equipment sensitive to electromagnetic interference), the basic function of a power supply is to ensure that the connected load is not affected by power supply factors, thus preventing malfunction or damage. Therefore, the EMC requirements for power supply products are naturally higher than for other products. The most typical example is the requirement in EMC standards for power supply products to conduct conducted interference tests on both the input source and output load ends separately. Furthermore, if the power supply product is an indispensable part of the system's operation (such as a UPS), and it is sold as a general-purpose product, then the product may require secondary EMC testing. The first test tests the EMC performance characteristics specified in the product's own standards; the second test is conducted jointly with the system based on user feedback. Numerous studies have shown that electromagnetic interference from mains power is one of the most prevalent and severe types of electromagnetic interference. Solving this type of interference essentially solves the immunity problem. Therefore, some describe the technological characteristics of today's information society as "one machine, three components": computer hardware, software, and power supply. Thus, AC regulated power supplies, as the power interface between mains power and electronic equipment, especially information technology equipment, should have an effective power filter function, at least significantly attenuating and suppressing electromagnetic interference. This should be an essential function of AC regulated power supplies. Naturally, for AC voltage regulators with anti-interference capabilities, it's not enough to simply improve their immunity; they should also provide a significant EMC safety margin for EMI-sensitive electronic products connected to their output terminals. This is a truly essential functional requirement for an anti-interference AC voltage regulator that effectively purifies noise. This is one of the bases for the compilation of SJ/T10541-94, "General Technical Conditions for Anti-interference AC Voltage Regulators." On the other hand, some requirements similar to EMC are already reflected in the performance indicators of power supply products. For example, the requirements for source voltage effect and total relative harmonic content of the output voltage of AC voltage regulators. Furthermore, some EMC items that are only sensitive to weak-frequency electronic equipment, such as resistance to power frequency magnetic field interference, electrostatic discharge, and radiated electromagnetic field interference, may not have a significant impact on high-power electrical equipment, and therefore are not listed as mandatory test items in SJ/T10541-94. Thus, the EMC requirements for AC voltage regulators differ from those for other electrical and electronic products. [b]2 EMC Testing Items and Requirements[/b] EMC testing requirements are divided into three main categories based on product application: military, industrial/commercial, and civilian/residential environments. The testing items, requirements, and methods for the latter two are largely consistent, differing mainly in the specific indicator requirements. Military applications, due to their unique usage, differ significantly from the latter two categories. Furthermore, aviation and marine equipment, due to their specific usage, have the same high requirements as military equipment, as well as internationally recognized standards and specifications. Based on the usage conditions of AC regulated power supplies sold in the market, this article focuses on the latter two categories. Given the increasing public concern about EMC issues, involving numerous professions and products, the IEC has treated EMC requirements as a fundamental standard. This is the well-known IEC 61000 series of standards. This standard is internationally recognized as a universal standard of equal importance to safety standards. IEC 61000-4, "Test Techniques," is one of the fundamental standards guiding EMC testing. Because EMC technology is a complex, multidisciplinary, and constantly evolving technology, the relevant EMC testing items, requirements, and methods are constantly being revised and improved. Therefore, many items in IEC 61000-4 are still in draft form and have not yet been officially released. To facilitate readers' understanding of this knowledge, we will introduce the items related to AC regulated power supplies, focusing on the IEC items adopted by relevant national standards. IEC 61000-4, short for "Electromagnetic Compatibility for Electrical and Electronic Equipment, Part 4: Testing and Measurement Techniques," lists 11 testing items. The items and requirements related to the AC regulated power supply standard SJ/T10541-94 are introduced below: IEC 61000-4-4: Electrical Fast Transients (Burst) Immunity Test, adopted by SJ/T10541-94 and SJ/T10542-94. The purpose of this project is to verify the equipment's immunity to various transient disturbances caused by switching transients (inductive load interruption, relay contact tripping, etc.). The test severity levels (the open-circuit output test voltage of the interference generator (50Ω internal resistance) at the power input) are: Level 1, 0.5kV; Level 2, 1kV; Level 3, 2kV; and Level 4, 4kV. This falls under IEC 61000-4-5 Surge Transients Immunity Test. This section is currently in draft form. Related references include the well-known IEC 801-5 and the American standard IEEE std 587-1980, "IEEE Guide for Surge Voltages in Low-Voltage AC Power Circuits." Electronic industry standards SJ/T10541-94 and SJ/T10542-94 reference parts of this content. The purpose of this project is to verify the equipment's immunity to interference caused by power switching and high-energy surges from lightning. The severity levels are classified as above, but the output impedance of the interference generator is 2Ω, while the former is 50Ω. The interference generator output is divided into two categories: a 1.2×50μs short-circuit voltage waveform (for high-resistance loads) and an 8×20μs short-circuit (discharge) current waveform (for low-resistance loads). Similar test requirements are listed in many electrical and electronic product standards. In addition, internationally, high-frequency spike noise sensitivity testing is used. This test is particularly common in Japan. The US military standards MILSTD461 and 462 use similar items, but require a much higher power noise simulation generator. Similar standards exist domestically, such as the 50kHz～100MHz power line conducted sensitivity test in GB4859-84; and the CS06 item in GJB151-86 and GJB152-86, where the noise is a rectangular pulse wave. Furthermore, GB6162-85 (referencing IEC255-4) uses a damped oscillating wave. SJ/T10541-94 and SJ/T10542-94 suggest choosing either of these two methods. The purpose of this item is the same as IEC61000-4-4. Rectangular pulse waves are characterized by rapid rise and low repetition rate, while damped oscillating waves are characterized by high amplitude and low energy. For products containing digital circuits, rectangular pulse waves are better able to illustrate the sensitivity to interference. Electromagnetic interference items include power frequency harmonic limits, conducted interference, and radiated interference limits. SJ/Z9029.2-87 (equivalent to IEC555-2-1982) specifies the harmonic current limit requirements generated by equipment in low-voltage power supply systems. For high-power semiconductor converters, GB/T3859.2-93 limits current harmonics by specifying the maximum power capacity (the ratio of the power system short-circuit capacity to the converter's fundamental apparent power) for different converter devices (based on pulse number). It must be pointed out that the problem of current harmonics conducted by electrical and electronic products to the mains power supply through power lines has been recognized as electrical pollution, and as an object of "environmental protection" for the mains power system, the requirements in this regard will be increasingly strengthened. The determination of EMC test results uses completely different methods for anti-interference testing and electromagnetic interference testing. The latter uses quantitatively specified limit values ​​as the threshold for acceptance; the former generally uses a qualitative method, classifying products according to their performance in the test (taking GB/T13926-92 as an example): Class a: Normal performance within the product performance specifications (tolerances); Class b: Temporary reduction or loss of function or performance, but can recover on its own; Class c: Temporary reduction or loss of function or performance, but requires operator intervention or system reset; Class d: Irreversible reduction or loss of function due to damage. Of these four categories, category A is undoubtedly considered合格 (qualified) and category D不合格 (unqualified). For categories B and C, the qualification criteria are determined through negotiation between the manufacturer and the user based on specific circumstances. Naturally, the technical measures adopted for these two categories differ. While the immunity test criteria may seem too lenient, this actually reflects the principle of maximum flexibility in the standard. Because the tested equipment is diverse and varies greatly, it is difficult to provide a universal quantitative standard for qualification. Of course, a specific evaluation standard should be given for a particular product category. SJ/T10541-94 embodies this requirement. GB/T3859.1-93 similarly specifies the disturbance categories of converters as the basis for qualification determination. This standard defines three levels: F-level: Performance level, referring to the combination of all electrical disturbance limits that the converter can withstand without degrading its performance; T-level: Tripping level, referring to the combination of all electrical disturbance limits that the converter can withstand without interrupting operation due to the action of protective devices; this can be further divided into two cases: automatic reclosing after the disturbance passes and non-automatic reclosing (manual methods are required); D-level: Damage level, referring to the combination of all electrical disturbance limits that the converter can withstand without causing permanent damage. Clearly, F-level is equivalent to class a, D-level to class d, and T-level includes classes b and c. For AC regulated power supplies with anti-interference requirements, SJ/T10541 stipulates that, in addition to ensuring normal operation, a suitable sensitivity threshold should be provided to the load at the output terminal; the peak value of the residual voltage of the interference superimposed on the output voltage should not exceed 20% of the nominal value of the output voltage. This reflects the standard's adherence to the principle of standard purpose, fully considering the fundamental differences in functional requirements between AC regulated power supplies and other electrical and electronic products. In other words, the former serves the latter, acting as a power filter for the latter's power supply EMI, aiming to provide sufficient EMI safety margin for EMI-sensitive equipment and improve its immunity level. According to GB6833.4, electronic instruments must withstand transient voltage surges of 20% of the nominal source voltage change without malfunctioning. Therefore, SJ/T10541 specifies this value as the maximum permissible transient voltage value (sensitivity threshold) for AC regulated power supplies during immunity tests. Furthermore, considering that AC regulated power supplies must provide suitable AC voltage conditions for electronic equipment, SJ/T10541 also stipulates that during immunity tests, the relative deviation of the AC regulated power supply's output voltage (i.e., output effect) should be within its reference conditions (tolerance G), serving as the basis for determining performance degradation. This combination of methods scientifically, rationally, practically, and easily operablely solves the problem of lacking quantitative indicators for specific assessment and qualification evaluation of AC regulated power supply immunity performance. General standards often use vague criteria like malfunctions, performance degradation, or downgrading as the basis for assessment and qualification, which is clearly difficult to implement qualitatively. [b]3 EMC Testing Conditions and Methods[/b] Testing relies on three factors: method, technology, and equipment. The method is determined by both the measurement principle and the usage of the testing equipment. Technology refers to all testing means employed to obtain accurate test results (high accuracy). Equipment refers to all technical devices that embody the above two factors in serving the testing process. All of these must be standardized to ensure reproducibility and authenticity of the test. EMC testing conditions are determined by the testing method. Specific testing methods are divided into bench testing under laboratory conditions and field testing under actual usage conditions. It is impossible to simulate all possible interference phenomena encountered in the field, especially since the field method has insurmountable limitations. However, standardized testing can provide a more comprehensive understanding of the EMC performance of the device under test. Therefore, internationally, bench testing is recommended as the first choice, and field testing is generally avoided unless it is impossible to conduct in a laboratory. The main method of immunity testing is to select an appropriate severity level based on the electromagnetic environment conditions of the equipment, combined with the measures taken by the user, and conduct tests according to relevant testing methods. Finally, the test results are evaluated according to the acceptance criteria proposed in the product standard. This is the main difference between immunity testing and other tests. The electromagnetic interference sources in the electromagnetic environment, the coupling mode of the electromagnetic interference sources to the equipment, the sensitivity of the equipment to electromagnetic interference, and the user's protective measures at the work site are directly related to the severity level. That is, the environment determines the form of interference, and the installation protection conditions determine the severity level of the interference. GB/T13926.4 specifically specifies the electrical environment conditions for equipment operation under the corresponding severity levels in the electromagnetic environment: Level 1, well-protected environment, such as computer rooms; Level 2, protected environment, such as control rooms or terminal rooms in factories and power plants; Level 3, typical industrial environment, such as relay rooms in industrial process plants, power plants, and open-air high-voltage substations; Level 4, severe industrial environment, such as power plants, industrial process equipment without special installation measures, and outdoor areas. IEC 801-5 classifies surge conditions and protective measures for equipment based on power switching transients or indirect lightning strikes as follows (applicable to surges): Class 0: Well-protected electrical environments with primary and secondary overvoltage protection, typically located in special rooms, where surge voltage does not exceed 25V; Class 1: Electrical environments with partial protection and primary overvoltage protection, where surge voltage does not exceed 500V; Class 2: Electrical environments where power lines are separated from other lines and cables are well isolated, where surge voltage does not exceed 1kV; Class 3: Electrical environments where power cables and signal cables are laid in parallel, where surge voltage does not exceed 2kV; Class 4: Electrical environments where interconnecting lines are laid along power cables as if outdoors, and both electronic and electrical circuits use cables, where surge voltage does not exceed 4kV; Class 5: Electrical environments in non-densely populated areas where electronic devices connect to telecommunications cables and overhead power lines. Surge testing is not performed for Class 0. For general power products operating in Class 1 or 2 electrical environments, severity levels 1 or 2 can be selected. It must be pointed out that considering the environment as a relevant condition for immunity testing is a crucial characteristic of immunity testing. Ignoring these relevant conditions and disregarding the operating environment of the device, assuming that the device should be "independent" and suitable for insertion into any combination of devices (or systems), leads to the erroneous conclusion that all devices under test must undergo all interference tests to the highest severity level. This not only imposes excessively strict limitations on the devices used but also incurs a significant economic burden due to the need for numerous tests. Furthermore, immunity testing involves high-voltage signals; in addition to strictly adhering to relevant safety regulations, it is necessary to conduct safety tests on the equipment after the immunity test. For high-power electrical products such as AC regulated power supplies, it is essential to select immunity tests characterized by high frequency and high energy input from the mains power supply, and to choose a severity level higher than that for other electrical and electronic products. Another important characteristic of immunity testing is the strict and clear specification of the technical parameters of the test generator. To compare the immunity performance of devices, a test device capable of producing relatively consistent and reproducible results is needed; this is the interference simulation generator. Obviously, it is necessary to specify the generator's output internal resistance, output waveform requirements, open-circuit voltage amplitude, and error to ensure consistent and repeatable test results. Otherwise, due to different source impedances of different tested devices, the impedance matching of the generator will vary, making it impossible to achieve the same output waveform or amplitude under load. In fact, impedance mismatch is an effective means of suppressing electromagnetic interference. Electromagnetic interference tests for AC regulated power supplies to the outside world (through the mains network) include: harmonic conducted interference test and high-frequency conducted interference test. The harmonic conducted interference test measures the power frequency current harmonics at the power input terminal of the equipment; it measures the maximum value of each current harmonic below the 40th order. For three-phase power supplies, the neutral current harmonics should also be tested. In the performance evaluation of AC regulated power supplies, this item is assessed based on the relative harmonic content of the source current. Conducted interference testing of AC regulated power supplies is similar to that of other electronic products and can be conducted using GB6833-86 Electromagnetic Compatibility Test Specification for Electronic Measuring Instruments (referencing HP standards or GB9254-88 Radio Interference Limits and Measurement Methods for Information Technology Equipment, equivalent to CISPR-22-1985). An important testing device in high-frequency conducted interference testing is an artificial main network, which is called a line impedance stabilization network in US standards. (LISN). This is because the high-frequency impedance of the mains power varies at the power input terminals of different devices under different power conditions. To ensure that the test results reflect the actual situation, a suitable network must be connected between the device under test and its power terminals. This network can provide both radio frequency isolation between the device and the power grid and a stable high-frequency impedance for the device. The number of branches in the artificial power network is the same as the number of lines in the power supply system. The connection between the network and the interference measuring instrument should ensure impedance matching (50Ω/50μH). Each power line is tested separately, and the interference voltage value is measured. GJB152-86 recommends using the current probe method to measure conducted interference current; in this method, a 10μF feedthrough capacitor is connected in parallel between the power line and ground, serving the same purpose as LISN. The current probe method is simple to use, fast to measure, convenient for field testing, and more closely approximates the actual situation, and may be the primary method for future measurements. In addition, the military standard uses a peak detector, while GB9254 uses a quasi-peak detector. Radio frequency (RF) radiated interference (RF) testing is complex, involving issues such as test site, antenna, and test circuit connections. The test site should be an open outdoor area with a background electromagnetic noise level at least 6 dB lower than the permissible limit. This requirement is difficult to achieve, and the standard recommends using an electromagnetically shielded room (or an anechoic chamber) as an alternative. When testing radiated field strength, the device under test (DUT) should be wired strictly according to its actual operating method; power and signal cables should not be intentionally bent or retracted to reflect realism. In short, based on the practical value requirements of AC regulated power supplies, their EMC performance should, in addition to achieving high severity levels of immunity, meeting electromagnetic interference limits, and providing suitable AC voltage conditions, more importantly, provide sufficient EMC safety margins for their loads (EMI-sensitive electronic instruments and equipment, especially information technology equipment) operating under severe electromagnetic environments. This is not only a basic function of AC regulated power supplies but also the basis for their EMC requirements and EMC testing.