The ‘Amended’ EMC Directive

 

By John Woodgate B.Sc.(Eng), E.Eng., M.I.E.E., F.A.E.S., F.Inst.S.C.E.

 

Two years ago, the EMC Directive was subjected by the European Commission to the SLIM (Simplified Legislation for the Internal Market) process. The review team reported that no fundamental changes should be made, but that a considerable number of details needed attention. An EMC Working Party was set up in order to produce an ‘amended’ Directive. In fact, they have really made a new Directive, keeping the commas and full stops and putting new words between them.

 

At the end of last year, a sixth draft of the revision was produced for comments from interested parties. It included a number of choices on the wording of critical provisions. In February 2001, a seventh draft was issued, in which one of each set of choices had been selected. It is not at all clear how this selection was carried out! The seventh, and allegedly final draft was supposed to be discussed at a meeting of the working party early in March 2001, but it appears that much of the discussion was about the text of the sixth draft! Nevertheless, the discussion disclosed several matters in the seventh draft which need further attention, so it cannot really be ‘final’.

 

The EMC Working Party

 

The Working Party consists of representatives of the Commission, of industry (read ‘big business’), of regulatory authorities, of national standards committees and of test houses (which have a vested interest in making the provisions of the Directive such that most manufacturers have no option but to use a test house to demonstrate compliance, of course) and includes some EMC experts. It does not seem to include any legal draftsmen, maybe due to experiences some years ago, when legal draftsmen tried to write technical guidance documents on the Directive but misapplied many technical terms, such as ‘modular’, ‘component’, ‘active’ and ‘passive’. Unfortunately, the result of this is that the draft may very well not mean what it undoubtedly appears to mean, in several places.

 

The seventh draft

 

The seventh draft was put on the Commission web site, as:

 

http://europa.eu.int/comm/enterprise/electr_equipment/emc/slim/review.htm

 

with an invitation for all and sundry to comment! The web page actually asks for responses from Trade Associations directly, but any response should presumably also go to the appropriate department of the national government.

 

In addition, the Commission, having spent a lot of money on a report by a French consultancy on the Directive as a whole (which report was greeted with some amusement in many circles), has now instigated a cost/benefit analysis by a British firm of environmental consultants.

 

The seventh (eighth? ninth?) draft will be submitted to the European Parliament at the end of 2001, and will go to the Council of Ministers early in 2002. It should be published in the Official Journal later in 2002, and will be implemented by legislation in the Member States during 2003. However, it is not entirely clear when that legislation will come into effect, but probably in 2005.

 

WP meeting, March 2001

 

Many comments were received for discussion at the March meeting of the Working Party, both from national and organizational delegates. Several of these indicated the delegates’ selections from the choices offered in the sixth edition, concerning:

 

·     precise definition of ‘fixed installation’;

 

·     involvement of Notified Bodies (re-named from ‘Competent Bodies’) in determining compliance;

 

·     precise wording for certain products that require separately-supplied external accessories for conformity, and for the competence of installers of such accessories;

 

·     information mandatorily to be supplied by the manufacturer to the user or installer;

 

·     restrictions on appeal to standards to demonstrate compliance.

 

This last subject is highly controversial, because, for all but the largest manufacturers who can justify their own fully-equipped test facility, it affects the second bullet point as well.

It is naturally very important to have precise definitions for key words like ‘apparatus’ and ‘installation’, but it is equally essential to ensure that, wherever such a defined word is used, the definition and the context are totally consistent.

 

Problems have been reported with low-cost cables that are inadequately screened, and cause r.f emissions above the limits. Most electronic products are physically too small to radiate significantly below 30 MHz, but cables can act as antennas. So there is a new provision for ‘ready-made connecting devices’ (which may or may not include objects other than cables, maybe such as dongles) to be subject to the Directive. However, the subject is by no means closed yet.

 

Apparently remaining problems

 

Much of the wording of the Directive is in four Annexes, where ‘foot in mouth’ problems seem rife. Article 3 of the draft very simply states the need for equipment to conform to the ‘essential requirements’ given in Annex I, of which the ‘protection requirements’ are substantially unchanged from the present Directive. However, there are now two additional pages of ‘specific requirements’, several of which are certainly in need of further consideration. For example, for apparatus that can take several configurations, ‘the ... assessment  must confirm (note, not imply) that the apparatus shall meet the protection requirements ... in all possible configurations. Now, a quite simple modular product ,such as an audio mixer/amplifier for professional sound reinforcement, may have ten possible input and control modules, all physically interchangeable. Not all need be installed in one product. Suppose the minimum number of modules to make a working product is three. Then, since the sequence in which the modules are arranged may matter, we have six ways for three modules, 24 for four modules, 120 for 5 modules, 720 for 6 modules, ....... 3 628 800 for 10 modules. That’s rather a lot of testing, by any measure! At present, it is permitted to check only likely worst-case configurations: it appears that the practical impossibility of checking every one has been forgotten!

 

Annex II deals with documentation, the Declaration of Conformity and CE marking, of which the latter is substantially unchanged. However, major inconsistencies seem to be present between Annex II and Annex IV.

 

Clause 1 of Annex II requires a ‘report of compliance with the harmonised standards, if any, applied in full or in part. But clause 2 of Annex IV says ‘Compliance with a harmonised standard means conformity with its provisions (e.g. limits) and demonstration thereof by the methods the harmonised standard describes or refers to.’

 

This is a provision that in effect FORCES every product to be tested by a test house, by insisting that ONLY the test methods specified in standards, involving extremely costly special- purpose equipment, can be used. It means that anyone **self-certifying** must either buy all the special-purpose test equipment (think £100 k or so!) or use a test house for all measurements. Note also that it rules out the application of a standard ‘in part’ - it’s all or nothing!

 

There is no measurement in any EMC standard that I can recall at present that cannot be performed with at least equal ‘stringency’ by a method other than that specified in the standard.

 

I need to explain ‘stringency’ and maybe it is best to use an example.

 

Suppose the ‘official’ method requires the use of a spectrum analyser to determine that no emission is above a certain level. If, instead, I do a wide-band measurement over the specified frequency range, and the TOTAL emission is less than the limit, then obviously the emission in any narrow band of frequencies within the range is below the limit.

 

The wide band method is not ‘more accurate’ or ‘more precise’, but it IS ‘more stringent’ or, in the degenerate case of only one narrow-band emission within the frequency range, equally stringent.

 

All that the law requires at present is that the Declaration of Conformity be true. Up to now, it has been understood, except by people with their own agenda, that, provided the equipment conforms to all relevant standards, the ‘standards route’ can be used, even though the way the manufacturer assures himself that the equipment does conform to a standard is NOT necessarily the method of measurement specified in the standard, but another that gives sufficient assurance.

 

We are now faced with the situation that it will not be possible to use the standards route at all, unless every relevant standard is applied to the letter, irrespective of whether this is necessary or not.

 

For example, the requirements that limit flicker are based on the flicker characteristics of a 60 W mains lamp. So if your product doesn’t cause a 60 W lamp on the same mains distribution circuit to flicker, it meets EN61000-3-3, doesn’t it? Oh no! You must buy an IEC61000-4-7 flicker meter to test it or go to a test house that has one. This make a nonsense of the statement in the standard that ‘if it doesn’t cause flicker, you don’t have to test it’!

 

That is by no means the only problem, although it is quite enough by itself! Annex 2 (2) says ‘The EC declaration of conformity must contain, at least, the following: .... - dated reference to the specifications under which conformity is declared to ensure the conformity of the apparatus with the provisions of the Directive;

 

What ‘dated ......specifications’ ? This MUST mean either harmonized or national standards, surely?

 

If the manufacturer has not ‘applied’ the relevant standards completely, as defined in Annex 4 (2), he cannot fulfil this requirement of Annex 2! Effectively, he is FORCED to apply the standards and CANNOT choose not to!

 

Annex 4, Clause 3 says, in effect, is that if an essential requirement is not dealt with in, for example, the relevant product- family standard, you have to trawl through all the other EMC standards notified in the OJEC looking for one that *does* cover that essential requirement. For instance, it appears to mean that though there are no supply surge tests (only fast transients) called up in EN55020, you have to do such tests on a TV set, presumably by applying another standard, one that doesn’t include TV sets in its scope!

 

The clause was almost certainly written with EN61000-3-2 or -12 and -3 or -11 in mind: they have to be applied in addition to the relevant EN55xxx product-family standards. But by people not seeing all the possibilities raised by the text as written, a big new problem has been created.

 

You may not agree that words that I have cited above mean what I suggest they do. But I think you would find it difficult to say that they could not reasonably be considered to mean what I say. And there will always be someone who will prefer the most stringent possible interpretation, wherever there is the slightest doubt, especially if they stand to gain by it.

 

All bad news?

 

Well, it dos not entirely appear so. In the Preamble, which is full of formal ‘whereas’ phrases, that ‘set the scene’ for the Directive and may be boring but are very important, clause 19 says ‘ Whereas it is not pertinent to carry out the conformity assessment of apparatus placed on the market for incorporation into a given fixed installation, and otherwise not commercially available, in isolation from the fixed installation into which it will be incorporated; whereas this justifies that such apparatus are exempted from the conformity assessment procedures applicable to apparatus; whereas it should be prevented that such apparatus compromise the conformity of the fixed installation into which they are incorporated;’

 

Apart from the last ‘whereas’, all that sequipedalian verbiage means ‘One-offs are exempt’! But before you dance a jig of delight, what about that last ‘whereas’? That seems to throw the previous ‘exempted’ back into doubt. But there is no resolution of that anywhere in the draft, as far as I can see.

 

What to do?

 

Send your comments to the Commission’s URL! TODAY! Send a copy to the DTI, through your Trade Association or directly. Write to this journal, agreeing or disagreeing. Maybe you will find a problem that I didn’t find.

 

John became an independent consultant in 1984. He is a Chartered Engineer, a Member of the Institution of Electrical Engineers, a Fellow of the Audio Engineering Society and a Fellow of the Institute of Sound and Communication Engineers.

 

He is a member of several other BSI committees, Chairman of technical drafting panels, a past Chairman of the Audio Engineering Society British Section, vice-chair of AES Standards Sub-committee SC-05 and Honorary Secretary of the Institute of Sound and Communication Engineers. He regularly leads the UK delegations to committees of the International Electrotechnical Commission, and has made major contributions to several important international standards. He is a member of six international Working Groups and a former member of ITU UK Study Groups, having attended Study Group Plenary meetings as a UK delegate.

 

John can be contacted on

tel: +44 (0)1268 747839 fax: +44 (0)1268 777124

Email: webmaster@jmwa.demon.co.uk  Web: www.jmwa.demon.co.uk