In message <[email protected]>, dated Wed, 4 Jul 2007, Gert Gremmen <[email protected]> writes:
>I have been reading this thread for a while, but missed an important >aspect. > >Engineers in our profession have been looking at limits as absolute >LIMITS, not to be exceeded by any device under any circumstances. Yes, unfortunately including people with enough authority to enforce that erroneous view. >To meet this in practical situations, many approaches have been >suggested, most suggesting extra severe limits, where the discussion is >about how much more severe. I don't really think you can describe 'margin' as 'an extra-sever limit'. Margin is there because some samples will inevitably have higher emissions/less immunity that the tested sample or samples. It seeks to ensure that those samples don't violate the real limit. > >Well in Europe, and this shows the strength of the EMC directive, the >bottom line is that equipment should not interfere. (simplified >essential requirement) > >The fact that a piece of equipment exceeds a limit does not mean that it >will interfere, just that chances are better that it will. The same for >equipment below the limit, chances are only slightly better that it >won't interfere. >A interference peak at 102 MHz, will not create interference until >someone within a certain radius will decide to listen to 102 MHz on his >portable radio. Quite right. That is why I say that the limits are set (whether intentionally or not) so that complaints of interference are kept at an acceptably-low level. > >(It's good to emphasize that the current limits will only protect radio >reception . An excess of a few dB will ONLY interfere with radio >reception. No other interference problems can be identified in my >opinion) Except for low-frequency conducted emissions (mains harmonics) of course. > >It's up to the manufacturer to show that a particular negative margin >will not cause problems in terms of interference. For this to happen he >may choose to create a technical construction file route, and from July >the 20th on he does not need a competent body approval anymore. An >approach showing that an emissions excess of i.e. 2.5 dB at a certain >frequency for a certain percentage of equipment could be well defended >with appropriate arguments, in terms of applied environment, statistics >(i.e. number of equipment in use) and arguments of technical nature. >(i.e. will not be used near in radio receiving equipment) . >Limits are not absolute. While this is true, I doubt that many manufacturers would do that, because senior management (even senior technical management) doesn't have enough confidence in staff who would make such a recommendation. We hear all the time here about such lack of confidence and a credibility gap, extending to denial of funds to buy the standards that the product should comply with. > > >Each of us knows that respecting a standard's limits will not guarantee >that no interference will happen, nor can this standard guarantee that >the particular equipment, if deviously used/installed, will not exceed >the limits anyway. Again, true, and there need be no 'devious' - too little separation between PC and AM radio is an innocent error. > > >Of course, maintaining the limits, following harmonized standards is an >easy (!!?!) legal way to obtain presumption of conformity. >Keeping your emission profile below this limit however, under all >circumstances in all operating modes and all configurations and at the >same time keeping track of manufacturing variations and measurement >uncertainty would lead to unacceptable low values of interference >in-house limits. It depends on the product, and all aspects of it; mechanical design can play a big part. Parallel to clock waveform asymmetry we have high-resistance joints between metal panels or between connector shells and panels. Some products are 'pussy-cats' in terms of production variations, some are 'tigers'. >This problem may be somewhat relaxed by applying the CISPR 80/80 >limits, but this approach requires multiple samples to be tested which >can be a problem for other reasons. This would lead to unacceptable >costs for EMC while not justifying a reduced interference profile. Use of the rule is practicable only for high-volume and high-margin products. > >Therefore, I suggest to just maintain the limits, do not add extra >margins, unless your measurement uncertainty exceeds the CISPR16 >suggested values. > >The radio protection limits were created with sufficient margins so as >that even 10 dB excess will not create excessive interference problems. >A margin was build-in. > >In terms of interference capabilities, the amplitude criterion is not >the holey grail. The applied QP detector will not also accurately >describe interference capabilities for FM, PM or digital communication >systems. > >In general, when speaking about the chance of a piece of equipment >causing interference , it also makes sense to qualify the type of >interference. >A data over power line solution, creating full continuous interference >all over the band while remaining below the limit will certainly do more >harm in terms of interference as one clock harmonics at 107.6 MHz. > >I believe that the a limit should be based on the integral of frequency >/ amplitude characteristics of a type of equipment. The more >interference components in the band, the lesser the amplitude may be. A >single component would be allowed to exceed the current limits, where a >large number of peaks need to remain substantially below the current >limits. This approach would do justice to the chances any interference >could be created. >Of course these new limits could be detailed out, creating stronger >protection in certain bands, while relaxing in other bands, but I wanted >to discuss the principle. It's all true, friends! > >Gentlemen and Ladies (yes there are) , > >Let me know your opinions... > >I know I opened (re-opened) a can of worms....... Yes, but they are particularly attractive worms. Now you must submit this to CISPR/A! -- OOO - Own Opinions Only. Try www.jmwa.demon.co.uk and www.isce.org.uk There are benefits from being irrational - just ask the square root of 2. John Woodgate, J M Woodgate and Associates, Rayleigh, Essex UK - This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. Website: http://www.ieee-pses.org/ To post a message to the list, send your e-mail to [email protected] Instructions: http://listserv.ieee.org/request/user-guide.html List rules: http://www.ieee-pses.org/listrules.html For help, send mail to the list administrators: Scott Douglas [email protected] Mike Cantwell [email protected] For policy questions, send mail to: Jim Bacher: [email protected] David Heald: [email protected] All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc ______________________________________________________________________ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email ______________________________________________________________________
