Thanks for your comments, Kevin. I don't intend to make a big thing about this, but I'm not convinced by your arguments and in the hope that you (or someone else) will be able to make arguments which can show me the error of my ways, I'd make the following comments:
At 23:24 -0400 28/9/99, Kevin Richardson wrote: >Hi Nick, > <SNIP> >Some of the reasons are: >1. any group of people involved in the preparation of an international >standard are accepted as reasonably expert in their field. Typically >therefore they are not acknowledged as experts in terms of another >standard, developed by another group of experts (committee) and as such may >unwittingly include the wrong requirements or include the requirement in >the wrong context; I'm not arguing that the _requirements_ of the major 'general' standards should not be based closely on those of a 'specialist' standard - I'm just arguing that the test from the specialist standard should be included in its entirety in the general standard so as to avoid having to keep track of multiple documents. And I'd also point out that unless people responsible for the general standard know a good deal about the test they are specifying from the specialist standard, they should be very wary of using it. It's easy enough to get expert advice. >2. all standards, particularly standards involving complex requirements, >should be read in full regarding their requirements. It is unwise to >attempt to isolate one clause or requirement in a standard as other parts >of the standard may well impact that clause; A good point, but more an argument in favour of my argument than that of separating requirements out into numerous smaller standards. Multiple cross referencing leads to confusion, I would have said. >3. if requirements from another standard are included, the moment the >other standard changes or modifies that requirement in any way the first >standard which has included the requirement is immediately out of date and >it too now requires modification to come into line with the modified >requirement in the other standard. I don't agree with this. Firstly, all the main general standards (60950, 60335, 60204, 60598, 60730 etc.) seem to get updated or amended (sometimes for pretty trivial reasons) at least every eighteen months or so, so it is not necessary to wait all that long for an opportunity to make amendments to come round. Secondly, the life of the standards (i.e. time between amendments) is a lot shorter than that of the products they are concerned with. The changes we are talking about are usually minor modifications to test procedures, and as such they do not represent major advances in the safety of the products within the scope of any given standard. This is reflected in the timescales for introduction and withdrawal of newer and older versions of standards and products approved to them (several years, in all cases). Does it matter that the main standard is a bit out of date for a while? - as I said before, if people want to know what the latest version of a particular specialist test is, they can refer to the specialist standard, but most people responsible for the design and manufacture of specific products want a single set of requirements to work to which they know are going to be unchanged for a reasonable period. >Unnecessary duplication of effort and >expense (expenses which would need to be passed on to the purchaser of the >standard anyway). When a standard includes requirements from several other >standards it is an impossible and unrealistic task for the committee >responsible to keep a watching brief on all requirements in all standards >they have included requirements from; and This I strongly disagree with - it's a very simple cross referencing task which I would be horrified to discover is not already being done by the standards bodies responsible for producing the basic documents. If standards writers are relying on clauses from other standards in order to complete their own work, they damn well should be keeping an eye on what those others standards are saying. Surely this sort of liaison is one of the things the (paid) secretaries of the standards committees are there for? I repeat - I'm not arguing that each general standard should have its own individual test, I'm simply arguing that complete requirements from specialist standards should be duplicated in the general standards. It's not a difficult thing to conceive or put into practice with a minimum of skill at IT and administration. >4. lastly, the IEC have general rule that after a standard has had 4 >Amendments a new Edition of the entire standard must be published. This >again introduces unnecessary work and costs and would result in a marked >decrease in the period for which your standard that you purchased would be >current (ie the user would need to purchase the complete standard far more >frequently). I remain to be convinced of this. In practice amendments and re-issues take place pretty frequently anyway, but they are rarely doing more than tinkering at the margin of the requirements. It usually wouldn't hurt to save up the changes for a while so that larger scale revisions could take place but less frequently. In the vast majority of cases the requirements in the standards have remained the same for many years (like twenty or more). I think it's something of a myth that fast moving technology changes make it necessary to frequently update safety standards - the basics of safety don't change that much. If anything the complexity of current standards is a hangover from earlier times when it was necessary (for instance) to test non-synthetic insulations in three or four different ways to be confident of their performance. As I have already implied above, the fact that a new requirement is introduced does not automatically mean that everything made to the old version of the standard is unsafe. A slight digression, but there's a school of thought (in the UK at least) that says that now BSi are issuing amendments free of charge, the main document (re-issue of which _is_ chargeable) is done much more frequently than before when BSI could also make money on the amendments as well. Unless BSI have been lying to us for years when they say they need to have an enormous testing, inspection and certification business in order to support the costs of their standards writing activity, the cost of standards is set at a price it is believed the market will bear and does not not reflect the reality of the costs of production anyway. Regards Nick. --------- This message is coming from the emc-pstc discussion list. To cancel your subscription, send mail to [email protected] with the single line: "unsubscribe emc-pstc" (without the quotes). For help, send mail to [email protected], [email protected], [email protected], or [email protected] (the list administrators).
