讨论：为什么有这么多的EMC标准

Steve Hayes, TRaC Global, Worcestershire, United Kingdom;
JACK McFADDEN, Wyle Laboratories, Huntsville, Alabama, USA;
STEVE O’STEEN, Advanced Compliance Solutions, Atlanta, Georgia, USA;
Kenneth Wyatt, Wyatt Technical Services, Woodland Park, Colorado, USA;
David Zimmerman, Spectrum EMC Consulting, Eagan, Minnesota.
October 2011 sees the end of the transitional period from previous versions of EN55022 to the latest version, which now requires testing above 1 GHz for the first time. At the same time, beginning of October 2011, the Official Journal of the European Union listing the harmonized standards for the EMC Directive has also been updated. Notable (and predicted) is the inclusion of both generic standards for emissions (EN61000-6-3 and -4). Both these standards now include emission requirements above 1 GHz in the same way that EN55022 has. While the transitional period for EN55022 has just ended, it has just begun for the generic standards, making them mandatory only from January 2014. (See box on Page 10 for more on EMC testing above 1 GHz).
This issue recently inspired Steve Hayes, CEng MIET, managing director for EMC and safety at TRaC Global Ltd., to pose a question to his colleagues on LinkedIn, asking: Is this transitional period, where a manufacturer can choose to use either standard to demonstrate compliance to the EMC Directive, too long? After all, the product churn in today’s world is largely much faster than this and, hence, manufacturers may see at least another product launch before above-1 GHz test requirements are required. Equally, the protection of the RF spectrum will be under pressure while millions of new products enter the market without any control of their electromagnetic emissions for another two years.
Interference Technology invited Jack McFadden, senior project engineer at Wyle Laboratories; Steve O’Steen, EMC director at Advanced Compliance Solutions, Inc.; Ken Wyatt of Wyatt Technical Services; and David Zimmerman, president of Spectrum EMC Consulting, to expand on this and other questions posed by Hayes.
HAYES: Is the transitional period from old to new standards too long? ZIMMERMAN: This is an excellent point, and a shortcoming of the European Union EMC compliance system, in my opinion. In fact, there are other product family standards that will lag behind in the emission requirement above 1 GHz, and some may not change in the next 10 years. In answer to the question, in many cases, the time between the need for testing, and the date that a standard exists that requires testing, is too long to ensure that EMC conflicts are avoided.
O’STEEN: The “product churn” cycle is a significant variable which will vary widely across all industries so I would not dismiss it so easily. There needs to be balance between what’s economically prudent for the manufacturer and technically responsible in the interests of EMC. Equipment manufacturers are still held responsible for any interference caused by their equipment and are forced to resolve field complaints whether the interference is above 1 GHz or not.
I think the current two-year transition is adequate to allow the manufacturer to complete the redesign and address any compliance concerns during preliminary reviews. Large manufacturers who have their own compliance labs and permanent compliance staff are aware of the coming requirements and will typically get a jump on any issues which would be necessary due to their broad product line. Smaller manufacturers with fewer internal compliance resources will have to resort to alternate means via their local independent compliance lab, which would likely introduce some additional delay during the transition. From the perspective of the small independent compliance test lab, this particular example (EN61000-3 and -4) is not a concern, considering most labs have already completed the site validation above 1 GHz for EN55022:2006 + A1:2007. Any new revision of a standard that requires radical changes in the way a test is performed or in the way equipment is being used or introduces the need for a new category of test equipment could certainly require a significant amount of the transition time.
WYATT: From the manufacturer’s and test lab’s point of view, a two to three year transition period is good in that it provides a defined length of time to procure new equipment (typically, long lead times to develop budgets, evaluate equipment, etc.), and update test procedures. More importantly for the manufacturers, because the EU does not allow “grandfathering” of existing products, this provides sufficient time to re-qualify existing products with longer lifetimes. Not all product lives are measured in months.
HAYES: Do you think the limits in EMC standards are too onerous? You can put a mobile phone (acknowledged as generating a high RF field) next to a PC or mobile electronic gadget without electrical interference occurring. Based on this, could the limits be increased?
WYATT: The example specified may be true in some cases, but not necessarily for all products. For example, products with sensitive analog circuitry, such as measuring equipment, medical products and sensors, are very likely to be affected by mobile phones and two-way radios or other ambient signals.
ZIMMERMAN: I would have to agree with the general consensus that the limits are properly set. A lot of work goes into determining what the limits should be. In fact, there are standards committees that are hard at work to make testing to these limits more thorough. Things like testing all surfaces of an EUT, and bore sighting of the antenna above 1 GHz for example.
O’STEEN: No. In the example above, you give one possible interference source and one receptor for which general rules should not be based. Compliance levels and performance criteria should be based on historical data and sound engineering judgment and should apply to new standards as well as revised standards. Again, a balance must be achieved between the manufacturer’s desire to market their products and their compliance responsibilities.
HAYES: Why are the automotive EMC standards so different from commercial ones given that the environment is the same? The basic premise that in a vehicle electrical noise is generated by the spark ignition system and the limits are set based on interference to only the FM radio band seem somewhat dated.
MCFADDEN: I need to start with the question’s premises that “given that the environment is the same”. I do not agree with this premise. The environment is different. It would be easier if everything was black and white, if one size could really fit all. The world exists with various shapes, sizes and colors. There are vast environmental differences between the automotive and most commercial industries. One example, an engine control unit (ECU) operating temperatures are minimum of -40°C to a maximum of +150°C, reference SAE J1211, Table 1. The typical commercial products have a much more benign temperature operating range. I will not take the time to go into detail to discuss the commercial vs. automotive industry differences within the power bus, grounding schemes and etc. Just keep in mind, temperature itself alters the behavior of electronic components.
Just as there are differences within the thermal environment, there are also differences within the electromagnetic spectrum. The electromagnetic spectrum varies from one location to another location. As an experiment, take a look at analog compass as you pass over a bridge or go through a toll booth. The needle in the analog compass will lose magnetic north for a brief period then return to normal after the magnetic disturbance has passed. Electronic compasses may also be affected but it depends on their update rate, time within the disturbance area and the embedded software used within the electronic compass. The premise that the vehicle noise is limited to just concerns from the ignition firing (spark) effect on the radio is no longer true. It has not been true since the dawn of the electronic ignition. The electrical/electronic sub-components, sensors and modules are placed all over the vehicle. Windows open through the activation of a switch. The vehicle doors lock or open with press of a button. Airbags deploy when the vehicle system senses a crash. Some vehicles can actually speak to you, give you directions and make phone calls. The vehicle technology has drastically increased over the past few decades. As the technology has increased, so has the complexity of the vehicle, causing the electromagnetic compatibility concerns to grow. For example, airbags deployment must be during a crash and the airbags should not deploy as you drop off someone at the airport. Yet the automotive technology successes have been so pervasive that the vehicles’ quality and reliability are often taken for granted.
Even if we were to visit “Alice in Wonderland” and the “environment” was truly the same you need to answer this question. What is the risk if there is a susceptibility condition? Going back to the case of the analog compass; you will see a momentary deviation from normal operation with the compass (product) returning to normal operation after the disturbance has passed. It may be inconvenient but it is not life threatening. Now what if the automobile’s operation was affected (susceptible) to its environment? What is the worst possible outcome? The vehicle accelerates without driver action and crashes into another vehicle. So the risk between most commercial products and automotive industry is different.The automotive standards need to be set higher to account for this risk (and they are). If we were to prescribe to “one size fits all” then we would need to raise the universal standard to the represent the highest (worse case) risk. We would be, in effect, adding cost to the commercial products without value. The value the individual standards have is that they permit the industry to address their individual needs. You tailor your product requirements to meet needs of its intended environment in order to reduce the risk. Or in other words, the individual standards are designed for safe and proper operation of the product within its intended environment.
WYATT: I would disagree the environments are the same for office versus field. More and more automotive products today use “fly by wire” technology with a myriad of microprocessor-based subsystems. Hey, even the transmission on my truck has its own processor！ Because rapidly moving vehicles can quickly turn into deadly weapons, safety standards (and corresponding EMC standards) are necessarily more stringent. There have been numerous incidents where poorly shielded vehicles have developed operational issues (braking, cruise control, etc.) due to aftermarket two-way radios being installed.
ZIMMERMAN: This appears to be another case where standards take a path of their own based on the committee that is writing them. These standards are driven in large part by the manufacturers and not the consumers. From my limited experience in this area, it seems that as long as the equipment installed in vehicle does not interfere with the other installed equipment, there is no need for concern about interference beyond that which is known.
HAYES: Why are there so many EMC standards? Surely the environment is the same whether you use a toaster or PC in the home, yet the test standards (and in some cases, limits) are different. Could a ‘one standard fits all’ based on environment be produced? Imagine the savings that could be made not having to investigate which standard is the most appropriate.
WYATT: I believe there has been a trend in harmonizing many standards. I know this was moderately high on our agenda when I was serving on a standards working group. There will always be differences in certain products, I suppose. From a manufacturer’s point of view, it’s sometimes difficult to figure out the appropriate standard to use for a given product type, however, I’ve found that most test labs or consultants can assist with this. When all else fails, there ARE the generic standards.
O’STEEN: In my opinion, this would be very difficult to accomplish based on the engineering time and expertise devoted to tailoring the Product Specific Standards to address potential issues and performance criteria for each product category. Getting all of the industry experts to agree on common limits, performance criteria and methods would not be feasible. All of these Product Specific Standards are typically not written exclusive to one environment but are focused on a category of equipment but more importantly, it’s the performance criteria that separates these categories. Could you imagine a Product Specific Standard that would apply to all devices employed in a residence, taking into account for the variety of test levels, variety of test methods and performance criteria would result in an unnecessarily complex and unmanageable standard.
In addition, standard writing committees are constantly addressing the need to revise current Product Specific Standards in an effort to address compliance concerns for new technologies that happen to fall within the scope of that standard. These new environment-based standards would be in constant revision and release and would continue to lag behind the technology curve. I can certainly see the appeal of the environment-based standard system, but I don’t think each of these standards could effectively address the specific requirements and concerns associated with each product type found in that environment.
MCFADDEN: The environment is not the same. The environment of a toaster and personal computer (PC) in the home has some similarities but they are truly different upon closer inspection. The similarities are the utility power and the general location (home). Most toasters do not have intentional frequency generators within the circuitry. The typical toaster has one input power cable connecting it to the power bus. They are more electrical in design rather than electronic. The PC typically has several intentional frequency generators within its circuitry. It is a digital (electronic) device. The PC has multiple cable connections bringing it to printers, Ethernet, monitors and etc. It can generate interference over a larger spectrum than the typical interference measured from a toaster. The PC can also be affected by interference from a larger spectrum than a typical toaster. The toaster and PC may share the same home, but its reaction and its impact to its environment is completely different.
If the environment is not the same and the products’ function/operation are not the same is it possible to make one universal standard? I believe all things are possible, but many are improbable. It is possible to generate one universal standard. The question that should be answered is what would be the cost of the universal standard? Would the universal standard be a value added or will the universal standard generate additional cost without benefit? It comes down to determining acceptable risks. The regulating bodies and industry have determined that individual standards that tailor products to specific categories are the most effective way to keep the desired product quality while keeping the cost aligned. This leads to a great deal of confusion when you are searching for the appropriate standard. Many times the governing standard will reference another common standard to require the product to be tested with these specific tests at these specific levels (limits) while using this other standard’s methodology. This is true for the EN 55011, Industrial, scientific and medical (ISM) radio-frequency equipment – Electromagnetic disturbance characteristics – Limits and methods of measurements; and EN55024, Information technology equipment – Immunity characteristics – Limits and methods of measurement and a host of others.
If you wanted to develop one universal standard then you would require acceptance and input from all of the regulating bodies as well as all of the industries. The industries’ concerns regarding their products are highly exclusive. Their concern is only for their particular product(s). Now consider the logistics of bringing all the industries and all the regulating bodies, then have them willing to be inclusive. You are going against human nature and it is going to be painful. If you have had any experience in attempting to achieve consensus with a small body of people, imagine the challenge of getting a global industries and their regulating bodies to agree. I am not saying it is impossible. I believe it could be done. I just believe it is improbable.
ZIMMERMAN: The problem is that there are many differing opinions about what tests and limits are needed for a given environment. As long as there are two or more groups writing a standard for a given environment you will most certainly get at least as many different standards as there are groups writing them.
HAYES: Will ISO and IEC ever align their test methods, limits and procedures? Surely we don’t need multiple ways of assessing the same issue – electrical interference?
WYATT: Well, I sure hope so, for the sake of everyone’s sanity.
ZIMMERMAN: The likelihood of this happening is not great. Peace in the Middle East has a similar chance of happening in our lifetime. Standards writing bodies have a lot of pride and defend their positions with great zeal. If you check the “About ISO” Web page, you will see that it starts by stating “ISO is the world’s largest developer and publisher of International Standards. Size matters. If you go to the “About IEC” Web page, you will find this statement: “The International Electrotechnical Commission (IEC) is the world’s leading organization that prepares and publishes International Standards for all electrical, electronic and related technologies.” So the IEC is the world leader. You can see where these two groups will not want to concede their ranking. Would it make sense to align these test methods? The general consensus would be a resounding “yes”, but this does not mean that it will happen anytime soon.

以上是老外的说法，大家怎么看。