The Harder you Work,
the Luckier you get
By Phil Carter, aCEmark europe
When you are writing an article for an important anniversary issue of a magazine the tendency is to take a nostalgic view of the encompassing period. I would usually resist the temptation and concentrate on our industry’s most important issues such as problems with standards, interpretation or measurement uncertainty, (yes sad isn’t it) but 2003 is actually quite special, and a milestone year for a number of reasons, EMC Journal reaches its fiftieth edition, England won the Rugby world Cup and aCEmark reached ten years old! (no specific order of importance intended here). So I am afraid you are going to get a little bit of history as seen from these eyes, and you know, if I was given the same choices again, I suspect that the outcome would be much the same! My involvement in the EMC world has been a very exciting time for me, I have had the opportunity to travel widely and to make and meet so many friends, I have been very lucky, but the old adage is true ‘the harder you work the luckier you get’!
Ten years ago when I started aCEmark europe, I had already been involved in EMC for 9 years. I had joined Hewlett Packard (the part that is now Agilent) in 1984 as an RF and Microwave Specialist and very soon after was ‘volunteered’ to help with the EMC Seminars. I actually offered to do the Microwave Network Analyser Seminars, a subject which, at the time, I was much more familiar with. I remember being told at the time it would be ‘Character Building’ to learn the subject of EMC! Here I am still learning after nearly twenty years, funny thing is it’s still just as interesting!
A lot has happened in that time. Most of the work in EMC in the mid 80’s was still focused on Military testing, Def Stan 59/41 was in use merging into one standard to meet the requirements of the Army, Navy and Air, as did Mil Std 461 for the US. It was also true that most of the expertise in the EMC community at that time came from the military test area but changes were coming. Even then we had regular trade shows and EMC Exhibitions in the UK, organised by ERA at the Penta Hotel near Heathrow and alternating with an IEE conference at York. Internationally we had the conference in Zurich where during the early 80’s many papers on EMP and Military EMC were presented but which again progressed into commercial EMC; the IEEE in the US had long been having its annual conference on EMC and it still remains the largest gathering of EMC engineers under one roof anywhere in the world. (This year it will be in Santa Clara, California in August). I remember being Technical Program Chairman for a successful ARRMS (Automated RF and Microwave Measurement Society) conference at Herriot Watt University in Edinburgh about 1991. Sandown was in its infancy but had several successful years ahead of it until interest waned. We went through a few years with little in the way of major EMC conferences and exhibitions and in the vacuum, York re-established a few more successful conferences. This year however sees the eagerly awaited first EMCUK Conference and Exhibition at Newbury in October. Interest is building again with many emerging new technologies driving thoughts back to the EMC and RTTE Directives, and to the other new Directives which are imminent and will have a significant impact on Manufacturers of Electronics products.
Of course there have been many very successful conferences organised by the EMCTLA since its founding 12 years ago, with both Commercial and Military EMC Seminars, SAR, Measurement Uncertainty and several others addressing more specific issues and consideration of the standards which have impacted the EMC world, such as the still controversial Harmonics and Flicker standards, but more on them later.
Accreditation became more and more important as the eighties progressed and NAMAS (following the merger of NATLAS and BCS) was active in EMC accreditation with several Test Laboratories already accredited for EMC test. Although good work was being done by NATA in Australia and NVLAP in the US, much pioneering work in accreditation was done by NAMAS. The standards for Test and Calibration Laboratories ISO Guide 25 and EN45001 were not very detailed and did not cover many important issues. The Accreditation requirements of NAMAS were written into M10 and its supplement and provided a sound basis for accreditation, and throughout the 90’s NAMAS labs enjoyed widespread acceptance of their reports. It was not until 1999 when ISO/IEC 17025 was published that the world adopted a common standard,and any reader will note the strong similarities between ISO/IEC 17025 and M10! These much smaller differences in substance between the NAMAS requirements and ISO/IEC 17025 meant a lot less work for what had become the UKAS accredited laboratories. That Accreditation will eventually give worldwide acceptance of Test and Calibration reports is an essential aim of us all and indeed we now have a significant number of countries party to the ILAC agreement and this is growing but it all takes a long time. We have heard criticism that UKAS requires too high a standard and there is not a level playing field internationally. Whatever the real issues here, it is undeniable that a UKAS accreditation carries with it more credibility and acceptance than many others, and this is equally true in the EMC world. I know what I would want!
Commercial EMC was in the ascendance even in the mid eighties, emissions measurements were already well defined in CISPR and the ANSI C63 committee had long been supporting the FCC in setting the testing standards for the FCC rules where emissions requirements had long been required. Immunity testing was starting to be a driving force in the commercial world too, with the publication in 1984, of a series of Immunity test standards by IEC and given the numbering of 801 series, we had ESD in IEC 801.2, Radiated Immunity in IEC 801.3 and Electrical Fast Transient and Bursts Testing covered by IEC 801. These standards became very important and formed the basis of things to come with the approach of the New Approach Directive on EMC due for implementation in 1992. During the late eighties many Companies and Test Laboratories came into being specifically to address this rapidly growing area, and it wasn’t just in Europe that the excitement was building, many areas of the world starting to worry that these Directives were intended to be a barrier to trade. Fortress Europe was mused by our friends from across the North Atlantic who at the time were enjoying a massive trade surplus in the Electronics and Computer Industry with Europe, the EMC Directive was seen as potential barrier to this and the Manufacturers were looking for ways to ensure their continued shipments.
Competent Bodies were being appointed in Europe and several saw an opportunity in the US and elsewhere in the world, in assisting the local test laboratories to work with Manufacturers in compliance with a rapidly expanding set of standards covering many aspects of EMC and EMC related phenomena. As in the UK and most other countries, most of the EMC experience for immunity testing was spawned in Military project. It was due to the growth in testing for the EMC Directive that I first started visiting test laboratories, and by the end of 1991 I had audited about 15 laboratories in the US and also completed the NAMAS (now UKAS) assessor training and started working with UKAS as an EMC assessor. On behalf of Competent Bodies, I visited well over 30 test laboratories around the world, mostly in the US, but also in Japan, Poland and the Czech Republic, and it is interesting to note, how interpretation of the standards was such an issue then and still remains so today.
“Since then I have made well over a thousand assessment visits on behalf of UKAS, the Irish Accreditation Board, Norwegian Accreditation and others.”
Thanks for indulging me in the nostalgia that is generated from these milestones, I hope you will see that a good part of my working life has now been spent in EMC and assessment/accreditation and the desire to establish confidence in testing. I will come on to make a few points that come from all this ‘experience’ and see how we can draw some conclusions on the route map for the next ten or twenty years.
Confidence in Testing
I have long been associated with the consideration of measurement uncertainty in EMC testing, it is a subject from which many recoil from, after all who wants to explain to their customer they are unsure of the results?
This dilemma is mainly the result of the standards themselves for not indicating how much tolerance for measurement uncertainty has been allowed for in establishing the limits. I have heard the arguments about product compliance uncertainty being a much bigger problem than can be addressed just with the uncertainty of the measurement, but if we continue to hide behind these arguments we will never consider and address the things that are within our control. We do not find many solutions with our heads buried in the sand and we certainly do get the real benefit of analysing and estimating our uncertainties if we don’t do it. The real impact of a proper assessment of the uncertainties in a measurement is simply a better understanding of the measurement! If for no other reason this shows a requirement to do it.
Standards committees must tackle these issues and properly include estimation for the allowed measurement uncertainty or we lose the need for metrology and it goes straight backwards. Product Compliance uncertainty certainly needs to be addressed but completely separately, just as CISPR has established the CISPR Limit and recommends a statistical sample to be measured for emissions to establish compliance. In the recently introduced CISPR 16-4 estimated values have been evaluated of the uncertainties associated with the measurements and although I disagree with some aspects of the evaluation itself, these can be addressed later; the approach is sound. Recent documents on their way to amendments for the RF immunity standards do indeed bury their heads in the sand by making the statement that measurement uncertainty is included in the limits, but failing to say how much. This just opens the floodgates for abuse and allows direct competition between a laboratory who has spent very little time and effort on equipment and methodology and a laboratory who has taken the time and money necessary to reduce their uncertainties to acceptable levels. Soon there will be no tangible difference! Might as well toss a coin to establish compliance!
If a Manufacturer is happy to ship his product when it does not meet the required EMC standards he is taking a Risk. Just as in all areas of life, all risks should be analysed and understood before they are taken. Industry now has many areas where we are becoming familiar with Risk assessment; Health and Safety at work is a common example of everyday use of Risk Assessment! But what does it mean for EMC? We have seen recent examples in both Germany and France where products have been found to fail the limits and their place on the market is now in question, what is the risk here?
While I am having a bash at standards, no article would be complete without some reference to the ludicrously overcomplicated Harmonics and Flicker standards. If the manufacturers of test equipment find it hard to make something that complies with all the requirements, what hope is there for the user? Our National Physical Laboratory has spent a couple of years and significant expense in trying to establish a traceable calibration to demonstrate the various complex parameters involved in these Analysers. It is not clear if they have found a single one that fully complies in all areas yet! Many years after the standards were first published and they are still almost unusable in some cases! This is the outcome when the research interests are not tempered with real world issues!
“I am convinced the Harmonics standard is the biggest own goal in the history of standards.”
Think how much pollution could have been avoided in the intervening years if a sensible and simple standard had been implemented instead of the nightmare we have ended up with, bred from a mixture of justifying research funding and paranoia. Awareness and some control over power factor or even the first few harmonics would have converged KW and KVA, and even just a little improvement would still have made the atmosphere a better place.
Phil can be contacted on +44 (0)1793 771488, or email: email@example.com.
Phil is also part of the EMC UK consortium who are organising the associated conference at EMCUK2004.