Home | Glossary | Books | Links/Resources
|EMC Testing | Environmental
Control of EMC is an increasingly necessary discipline in these days of encroaching technology. Even where there are no laws or regulations on EMC, control of EMC is still needed for the financial benefit of both suppliers and users. This guide describes how EMC applies to systems and installations, and also describes proven best EMC practices in design, assembly, and installation. The correct application of this knowledge and these techniques will generally reduce project timescales, help ensure reliable operation, and reduce commercial risk, and meet regulations such as the EMC Directive where they apply. EMC techniques need to be employed from the very beginning of a project (even at the project specification and tendering stages) to achieve the greatest benefits and lowest risks.
The best EMC practices described here are taken from several IEC and European standards and moderated by many years of experience in their use, solving real-life interference problems, and EMC testing. EMC concerns all electromagnetic phenomena including power-frequency interference, so these best EMC practices are excellent at preventing "ground loop" problems and achieving good analogue and digital signal quality, improving signal integrity and signal-to-noise ratios and actually improving the functioning of all electrical/electronic/programmable apparatus and their systems and installations. Some modern best EMC practices contradict "traditional" techniques. Professional engineers have a duty to be up-to-date and to use modern best practices, and this often means revising or replacing traditional practices as technology advances.
Whether or not the EMC Directive applies, its "Protection Requirements" are nothing more than good engineering practice that all suppliers of systems and installations can follow. On emissions they should follow the relevant harmonized standard to protect radio reception in the world outside their control, unless there is very sensitive apparatus nearby which makes even lower levels of emissions necessary. On immunity they need to make sure that their products will function with adequate performance and reliability in their intended environment. This may not require following a harmonized immunity standard exactly: sometimes they can reduce a test level or skip some immunity phenomena altogether; sometimes they need to increase a test level or cover extra phenomena. However, for both emissions and immunity they still need to do the same EMC work in design and installation as if they were applying the CE mark, for reasons of good engineering and medium- to long-term financial success through the avoidance of commercial risk. Since many customers and users (and not just in the EU) expect, rightly or wrongly, to see a CE mark anyway it’s then a simple matter to complete the necessary documentation and apply the mark.
--- Intra- and inter-system EMC; EMC interactions \
The definition of electromagnetic compatibility
Before embarking on a discussion of systems EMC it would be as well to define the terms. Electromagnetic compatibility is defined in IEC publication 50, the International Electrotechnical Vocabulary (IEV), as "The ability of a device, equipment or system to function satisfactorily in its electromagnetic environment without introducing intolerable electromagnetic disturbance to anything in that environment." There are several things to note about this definition. The first is that it’s about the environment, which is itself defined in the IEV as "the totality of electromagnetic phenomena existing at a given location". Compatibility is defined in terms of the environment in which the apparatus is used. Implicit in this definition is the existence of a boundary between the apparatus and its environment. This is relatively straightforward to conceive for individual items of apparatus. But when we come to consider systems and installations, these both create their own environment within their boundary, and they also exist within a larger environment. So it’s necessary at the beginning to make the distinction between EMC internal to a system, and EMC complementary aspects of EMC: control of emissions, and the provision of adequate immunity. Historically, the prevention of interference to other equipment has been treated as largely separate from the correct operation of the equipment itself; but it’s now clear that "compatibility" means compatibility in both directions. As we shall see shortly, there are significant differences in the way that these two requirements are defined and met.
The third point is that the definition of EMC makes no mention or classification of what is meant by "disturbance", nor does it place any limits on any of the possible parameters or modes of coupling of such disturbances. It does, at least, make it clear that perfection is not expected. The equipment should not introduce intolerable disturbance and it should only need to perform satisfactorily in its environment. This implies that it’s necessary to know what is a "tolerable" disturbance and to decide what is "satisfactory" performance: more of this later. The IEV does give us a definition of electromagnetic disturbance:
"Any electromagnetic phenomenon which might degrade the performance of a device, equipment or system, or adversely affect living or inert matter. An electromagnetic disturbance might be electromagnetic noise, an unwanted signal or a change in the propagation medium itself" but this is not detailed enough for practical use. The generic standards, which explicitly define the essential protection requirements of the EMC Directive, do include the somewhat academic statement that: "Disturbances in the frequency range 0Hz to 400GHz are covered" even though they actually refer to a narrower and more practical frequency range. The most useful listing of disturbance phenomena is found in IEC61000-2-5, "Classification of electromagnetic environments". The right-hand columns labeled "coverage in EMC standards" are an attempt to list the extent to which each phenomenon is included in the various European generic and product standards. Note that military and aerospace standards do have a wider coverage of phenomena.
A description of EMC phenomena
As shown, electromagnetic disturbances can be classified into four types of phenomena.
--- Conducted low frequency phenomena
Table of Disturbance phenomena:
Electromagnetic disturbance phenomena ; Coverage in EMC standards: Emissions | Immunity
Electromagnetic disturbance phenomena Conducted low frequency phenomena (< 9kHz)
Harmonics and inter-harmonics; Signaling voltages; Voltage fluctuations, dips and interruptions Voltage unbalance Power frequency variations Induced low frequency voltages DC in AC networks
Radiated low frequency phenomena (<9kHz)
Magnetic fields; Electric fields; Conducted high frequency phenomena (>9kHz) induced 0, voltages or currents; Unidirectional and oscillatory transients
Conducted low frequency phenomena
With respect to these phenomena, EMC can mostly be considered as "compatibility with the power supply network", which to all intents and purposes can be taken to mean the low-voltage AC mains supply. DC supplies, while important for many types of equipment, have been far less studied and are not nearly so well characterized as AC supplies; in any case, they tend to be specific to a particular system, whereas the AC supply is in a manner of speaking "public property", and therefore both acts as a carrier of disturbances and needs to be protected from them. Most of these disturbances appear between phases of the supply, though earth potentials can be more widely distributed.
Harmonics and inter-harmonics
Non-linear loads draw current at harmonics of the supply frequency (significant up to 2-3kHz), which when flowing in the supply network impedance cause harmonic voltage distortion. In the low voltage public supply, the main threat is from large numbers of products with electronic power supplies, notably personal computers and TV sets. Large industrial loads such as variable speed drives can also cause harmonic distortion in low, medium or high voltage supplies. Inter-harmonics are due to other loads which are modulated in the same frequency range but not related to the supply frequency.
A secondary use of power networks is for information transfer via "mains signaling". A specific standard (EN 50065-1) exists for such systems, which regulates the signaling voltages applied to the network in the frequency range from 3kHz to 148.5kHz (the bottom of the long wave broadcast band). Other users of the mains supply should expect these voltages to be present and be able to ignore them.
Power system voltage and frequency variations
Rapid fluctuations at low level in the supply voltage, occurring from once a minute to 25 per second, are known as "flicker" since they cause this visual effect in electric lighting, and are caused usually by fluctuating industrial loads such as arc furnaces, motor switching or welding equipment. Voltage dips and short interruptions also occur, due to fault clearance in the power network. Voltage unbalance is a feature of three phase supplies caused by large asymmetrical loads. Power frequency stability is normally better than 0.1Hz, but may depart from this by up to 3% during network disturbances. The general quality of the power supply in Europe, coveting all these features and others, is detailed in the European standard EN 50160, which gives the acceptable parameters for power supply quality.
Induced LF voltages
Power supply cables may induce interference voltages at the power frequency and its harmonics (possibly up to 20kHz) in adjacent signal cables. The effect is usually most significant in audio circuits such as are found in studio or telecomm applications. Cable layout and effective coupling length will affect the magnitude of the induced voltages, which may also appear in differential mode on some signal cables. Under power system fault conditions, the induced voltages can increase by several orders of magnitude.
Radiated LF phenomena
Local magnetic fields exist around components of the power network as a result of current flow. Overhead and buried lines will exhibit magnetic fields, typically up to 40~tT depending on proximity, and the nature and configuration of the line. Fault conditions may significantly exceed this figure; for instance, a common problem in offices is VDU screen wobble due to earth faults in the building supply wiring, which allow return currents (and hence high magnetic fields) to flow outside of the normal wiring routes. Power transformers will have stray fields around them, as will many operating appliances, especially those containing motors. VDUs and TV sets will suffer perceptible effects on-screen at a threshold of 0.5-2~tT, and themselves generate magnetic fields at the screen scan frequencies.
High-level electric fields of 10kV/m or more occur underneath overhead lines and in substations, but the magnitude is reduced by a factor of 10-20 within buildings due to the shielding effect of the structure (no such effect occurs for magnetic fields). Electric fields due to internal building wiring or operating appliances are generally low unless there is a fault in the wiring, typically due to a broken earth connection. Again, VDUs and TV sets may generate electric fields perpendicular to the screen if no design precautions are taken.
--- Radiated LF phenomena
--- Conducted HF phenomena
Conducted HF phenomena
High frequency disturbances can be either transient or continuous. They are denoted as "conducted" if the principal coupling occurs in either differential mode or common mode t via the cables, either power or signal, which are connected to the affected or disturbing equipment. For evaluation purposes in the EMC test standards the distinction between conducted and radiated is normally made on the basis of frequency, but in real life both conducted and radiated coupling can occur over a wide and overlapping frequency range.
Induced continuous wave voltages or currents
The fields from nearby radio transmitters induce currents on any conductor exposed to them. This current when appearing at the interface to the affected equipment then induces a voltage at that interface whose amplitude depends on the common mode source and port impedances. The amplitude of the induced current depends on the conductor length, its separation from the ground reference, loops formed by stray capacitance and any resonance effects. Emissions from the equipment are coupled out onto conductors and radiated from them, and hence into the aerials of victim receivers, by the reverse process.
These are classified into a number of subsets depending on their source and nature.
Surges can be classified as oscillatory or unidirectional and have relatively high energy, i.e. are capable of inducing damage in coupled equipment. Their rise time and duration is generally in the order of microseconds to milliseconds, and their spectral distribution is limited to the lower radio frequencies. Typical sources are lightning, power system fault clearance or capacitor switching, and collapse of stored energy in inductive loads such as large motors.
Fast transient bursts are due to switching of local inductive loads, normally by electromechanical means (switches or relays). The resultant arcing causes a burst of short pulses each of a few nanoseconds duration, with little energy, but capable of causing severe upset when they couple into electronic circuits.
Coupling in each of these cases can be either differential mode between conductors or common mode with respect to a ground reference. Emissions of fast transient bursts by equipment which includes switching functions, though widespread, are generally not considered worthy of limitation, although the radio frequency content of automatically-produced, repetitive burst-type noise is regulated in the provisions for discontinuous disturbance of EN 55 014.
Radiated HF phenomena
Continuous radiated interference
Due to nearby radio transmitters and other radio-frequency generating equipment (RF heating being the most common), this will impinge upon the affected equipment in different ways depending on the source impedance of the field. Three cases can be distinguished:
The first two cases occur in the near field of the transmitter while the last occurs in the far field. The distinction between the two is a function of wavelength. At 30MHz, the wavelength is 10m and the near field/far field transition occurs at around 1.6m. As a very broad assumption, the separation distance between source and victim may be taken to be of this order of magnitude and so the 30MHz breakpoint is often used to divide field-related phenomena into electromagnetic (above 30MHz) and electric or magnetic (below 30MHz). Conveniently, 27MHz is the lowest frequency at which portable high power transmitters (for CB radio) are likely to be encountered in typical environments. Below this frequency, the spectrum is mostly allocated to broadcasting and fixed communications links, using fixed transmitters; exceptions occur for marine and military purposes.
--- Radiated HF phenomena
Transient radiated field interference
This tends to arise from lightning strikes and from high power switching events in equipment it’s less relevant than conducted transients. Equipment in close proximity to sources of fast transient bursts may see such bursts coupled by radiation but such proximity is unusual by comparison to the more typical conducted coupling path.
This is a specific phenomenon which is hard to classify in the same way as others treated here. It occurs as the result of equalization of charge between two objects carrying different levels of charge, one of which is often a person. The ESD victim sees first the approaching electric field associated with the charge and then, when air breakdown occurs, there is a transient current with associated magnetic and electric fields. Usually the most disturbing effect of the discharge event is the very high rate of rise of current or field. The duration of the event is no more than a few nanoseconds but the current di/dt may approach 10^9 or 10^10 amps/second, and the field rate of change may be greater than 1kV/m per nanosecond or 10 A/m per nanosecond. The ESD threat in a given environment is greatly affected by the nature of the insulating materials contained in the environment - that is, whether they are effective at supporting and generating electrostatic charge- and by the ambient relative humidity, since high relative humidity allows charge levels to dissipate more readily.
[...] substations or near other parts of the electrical supply infrastructure. For most types of equipment it’s less relevant than conducted transients. Equipment 'in close proximity to sources of fast transient bursts may see such bursts coupled by radiation but such proximity is unusual by comparison to the more typical conducted coupling path.
This is a specific phenomenon which is hard to classify in the same way as others treated here. It occurs as the result of equalization of charge between two objects carrying different levels of charge, one of which is often a person. The ESD victim sees first the approaching electric field associated with the charge and then, when air breakdown occurs, there is a transient current with associated magnetic and electric fields. Usually the most disturbing effect of the discharge event is the very high rate of rise of current or field. The duration of the event is no more than a few nanoseconds but the current di/dt may approach 10 9 or 1010 amps per second, and the field rate of change may be greater than 1kV/m per nanosecond or 10 A/m per nanosecond. The ESD threat in a given environment is greatly affected by the nature of the insulating materials contained in the environment -- that is, whether they are effective at supporting and generating electrostatic charge- and by the ambient relative humidity, since high relative humidity allows charge levels to dissipate more readily.
The need for EMC
As we have seen, EMC has the complementary aspects of emissions and immunity. But the roots of these two aspects are different, as is discussed in the following sections.
Control of emissions
In most countries of the world, the radio spectrum is heavily used for many kinds of traffic. Broadcasting and telecommunications are the most obvious uses, but telemetry, radar, radio navigation and space research are some other purposes. Spectrum users pay a license for the privilege of being allowed to transmit and receive and in return they expect this privilege to be unaffected by interfering sources. Services which rely on radio communication for the economics of their operation are normally able to put a price on its continued reliability. In addition, safety related services demand assured reliability of spectrum access.
For these reasons governments have found it necessary to regulate the spread of types of apparatus which, though not licensed or intended as radio transmitters ("unintentional radiators" in U.S. jargon) have the potential to disrupt such services.
Historically, such apparatus has been dominated by broadband sources, notably motor- driven equipment, fluorescent lights and pulsed ignition (from petrol engines). For decades emission limits have been placed on these kinds of equipment and have been enforced by many countries. This has been a well established aspect of national and international EMC control.
Electronic equipment also has the potential to generate radio frequency interference as a by-product of its operation, unless it’s specifically designed to avoid doing so. The main culprits are digital electronics incorporating microprocessors, and power switching circuits using fast electronic switches -- MOSFETs, IGBTs, transistors, triacs and so on. Even though they are not intended deliberately to transmit RF energy, sufficient is produced in sensitive parts of the spectrum to make it necessary to apply the same types of legislative control as has been applied to other types of unintended source in the past.
With the expansion of the concept of EMC to include compatibility with the supply network, control of other types of emissions are becoming necessary. These are principally power frequency harmonic currents, and flicker. Because the mains power supply at a given node may be shared among several users, and has a finite source impedance, any load current disturbances attributed to one user will cause voltage distortion which is presented to all other users at that node. The amplitude of such distortion has to be limited, and since the source impedance cannot economically be made arbitrarily low, this implies some limit on the allowable current disturbances.
Control of immunity
The immunity aspect of EMC is a rather more complex and contentious issue. It has several strands which need to be separated and examined:
Safety is addressed separately.
The "virtual tradeoff" between emissions and immunity
There is a common misconception that there is some relationship between emission and immunity requirements, so that in order fairly to distribute the compatibility burden, the requirements for emissions limits and immunity limits must be shared. This idea totally ignores the reality that emissions and immunity limits apply to two quite different phenomena. RF emissions limits are placed on unintentional emitters in order to protect radio reception, they are specified in values of microvolts per meter or millivolts, and one cannot expect a radio receiver to be immune from the very signals it’s expected to receive. On the other hand, immunity requirements are necessary because of the inherent nature of the operation of radio transmitters and other man-made and natural phenomena, and are specified in values of volts per meter or kilovolts (a million times greater). In other words, these two aspects of EMC address quite different issues and cannot be traded off one against the other.
One area where there could be a sharing of the burden is in the phenomenon of transients due to switching operations. In this case, control of the amplitudes of emitted transients could, with time and with the eventual turnover of the installed base of equipment, lead to a reduced immunity requirement against such transients; but there are no serious moves in the standardization committees to address the problem of regulating such transients.
Fitness for purpose
As explained earlier, EMC is an environmental issue. If a particular product is intended to be used in a given environment, the user ought reasonably to be able to assume that it will cope with any disturbances that might be expected in that environment. If it has a particular function, then that function should be unaffected by typical levels of interference. In other words, a product should be fit for its intended purpose, and of course this purpose includes being used in a particular electromagnetic environment.
Clearly there must be a match between the actual environment of use and the intended environment. It would be unreasonable, for instance, to expect components of a fire alarm system that had been designed for a maximum RF field strength of 10V/m to work correctly when exposed to 50V/m because they had been placed within a few inches of a communications base station transmitter. But if it’s in fact reasonable to expect such transmitters to be located in close proximity to such components, then what is wrong is the characterization of the environment at 10V/m. For this reason EMC standards for various types of product are written around an environment classification, as are the generic EMC standards. In fact, because there is a strong statistical component to the make-up of any real environment- sources of a given type, strength and frequency are only present at a particular location for a proportion of the time, often a small proportion-- it’s accepted that occasional interference problems will occur which have to be dealt with on an ad hoc basis. This does not invalidate the expectation of fitness for purpose in the general case.
A further difficulty lies in defining what is adequate "fitness" — that is, if the product suffers a degradation in performance due to interference, to what degree is this acceptable. The question of performance criteria is fundamental to immunity compliance and will appear at various points in this guide.
Mandatory product quality
The effect of regulating the electromagnetic immunity of products is to set a baseline for product quality in EMC terms. If a product has to meet certain minimum standards for immunity before it can be marketed, this effectively removes the consumer' s right to trade off price against quality - even though a manufacturer can specify his own performance criteria in many cases. As the EMC Directive has taken hold, there have been many cases of customers discovering that the price of something that they have been specifying for years has gone up sharply, or worse, that the product is suddenly unavailable. On inquiring, they are told that the supplier can no longer supply that product, or has had to redesign it, "because of the new European regulations". The natural reaction on hearing this news is to rail against the idiocy of the European regulations, and to ask the supplier if he will accept a waiver against the regulations in this case. If the supplier wishes to continue trading legally, he has regretfully to decline.
The fact that more often than not, the reason the supplier can't supply as before is because the product was never designed for EMC and failed the newly-required tests miserably, is of no concern to the customer. It always worked in his application, and he sees no reason why the imposition of a new law should disadvantage him.
This is a political question rather than a technical one. There has certainly been heated debate over the past few years as to whether the European Union has a right to insist on minimum levels of product quality in its marketplace; but the weight of opinion so far seems to be that it has, or at least that those who would argue the opposite case are unable to muster much support. There is of course a valid argument- made throughout this guide- that good EMC design and installation practices will save many problems and costs in commissioning and operational reliability. The question is whether these costs should be made transparent to the customer.
For any system which could pose a hazard to people or property if it was incorrectly controlled, safety questions must invariably be addressed, and these questions must include the issue of safety in the face of external interference.
Because the inherent susceptibility to electromagnetic disturbances of electronic control devices has been understood for many years, it’s usual for true safety-critical functions to be controlled only by electro-mechanical means. For instance, the emergency stop on a large machine will be a simple switch which physically disconnects the power to the machine when it’s activated. On more complex systems, relay logic might be used for the same function. The great merit of such an approach is that electromechanical devices are immune to normal electromagnetic disturbance sources such as radio transmitters, welders and switch arcing; such disturbances simply don't have enough energy to cause a change of state.
However, the more complex control systems become, the harder it becomes to implement safety critical functions only in such simple form, and the question of the immunity of the electronic control devices has to be addressed.
Documents linking EMC and safety
IEC publication 61000-1-2 on "Methodology for the achievement of functional safety of electrical and electronic equipment", in draft at the time of this writing, offers guidance on the safety of equipment which is exposed to electromagnetic disturbances.
It points out that as well as testing, an understanding of the relevant electromagnetic environment and its relationship to safety requirements is needed, and there should be a "dependability analysis" to identify the hazards which can cause safety risks due to electromagnetic disturbances. It stresses that the levels of electromagnetic disturbances indicated in various standards must be considered very cautiously as regards safety implications, in particular:
++ the disturbance levels vary according to a statistical distribution and the given values can be significantly exceeded in some circumstances, i.e. infrequently or on particular sites
++ the standard levels and performance criteria are generally related to functional requirements and not safety
++ equipment immunity characteristics may worsen with age.
With regard to immunity performance criteria, degradation or loss of function can have safety implications which must be analyzed; fail-to-safe behavior in the face of interference might be acceptable. Equipment failure or degradation may not be equivalent to system failure or degradation, for instance if redundancy is used, but t his means that parallel channels should not be affected in the same way by a disturbance.
The document stresses that higher than "functional" immunity test levels, and/or additional EMC testing, may be needed to cover safety implications. This is especially true if the only available immunity standard is a generic standard that does not anticipate any special operating conditions. With respect to safety requirements, functional immunity levels should be increased by some appropriate margin. It may also be that a type of disturbance that was neglected in the normal standards could have a safety implication and should be included in the test regime. Due to the complexities of system interaction, a system should be tested wherever possible at the highest degree of integration.
The Institue of Electronic and Electrical Engineers (IEEE) has a working group on EMC and Functional Safety, which produces a reports periodically.
The concept of multiple immunity test series, one for system parts not relevant for safety and one for safety-related parts with more severe requirements, is becoming common in some industries. This approach is typically taken by specifiers of safety- critical systems such as aircraft, cars or military systems. To take an automotive example, the engine management and anti-lock braking systems must have an assured immunity at the highest possible level, but the operation of electric windows and sunroof can have a relaxed immunity, and the immunity of the dashboard clock is purely a matter of convenience. In the context of safety certification of civil airliners, two levels are defined: a "critical" function is one whose failure would prevent the continued safe flight and landing of the aircraft, while an "essential" function is one whose failure would reduce the capability of the aircraft or the ability of the crew to cope with adverse operating conditions. A critical function must be maintained in the presence of a "severe" level of external RF interference while an essential function is only required to function correctly within a "normal" environment of RF interference.
Another example is provided by the Navy's specification:
For Naval applications an additional type of equipment to those defined in this Standard has been added. This is to cover equipment that is not essential to the float, move or fight function of the ship or submarine. Equipment in this category includes those for the function of the galley, laundry and crew entertainment.
This flexibility of approach allows for a reasonable matching of protection measures against the electromagnetic threat- provided, of course, that the definition of the functions and of the environment is done adequately. For instance, the critical functions need to consider foreseeable environmental extremes (lightning, proximity to high- power radars), misuse (ignoring the "turn off your cell-phone" sign) and faults (a blown fuse).
Prev: Article Index
Home | Glossary | Books | Links/Resources
|EMC Testing | Environmental
Updated: Sunday, 2012-10-28 5:37 PST