Throwing in my two cents: Keep in mind that Measurement Uncertainty goes both ways. For instance, EMC labs will typically have a Measurement Uncertainty between ±4 - ±5db for Radiated Emission. That is "Plus or Minus", not just minus.
Since no EMC lab is perfect it is very likely that a product can pass by 4db at one lab and fail by 4db at another lab and still be ok.?? I've told this story before, but it is worth repeating. Years ago in a life far far away, my employer at that time (not my current employer) did a buy/sell of a computer system built by another company in another land. Let's call is land Korea. By contract, this computer was to meet all safety and EMC requirements. Our EMC lab was not allowed to test this buy/sell product for ignorance is bliss (according to our corporate lawyers). One day, the good people of Sweden decided to verify the compliance of this system; to which they discovered that it failed Radiated Emissions by 2db. They sent us the test report which said that though it failed the limit, it was within their measurement uncertainty, so they could not declare the product as Non-Compliant. No action was taken or had to be taken. All parties were happy. Don't get me wrong. I've been doing this a long time and a huge supporter of margin and performing production audits. As an EMC engineer, we are responsible to our employer to produce a compliant and reliable product without over-burdening the design or the company we work for. It is a narrow path we walk. Sometimes decisions are left up to us and we must use our knowledge and familiarity with our own products to decide what margin we must have and what areas we can get away with less margin. We all do it. It's part of our jobs. Measurement Uncertainty makes the mathematicians happy and it gives us EMC Engineers something to go to our boss with to help support the argument for margin. But for me, we always design for 8db margin, and allow 5db margin during production audits. Any signal within 5db must be verified on several units and found to be stable. It would be irresponsible to go to production with test results with little margin on a single golden unit. We don't keep golden units. We are confident that we can pull any unit off the production line and it will pass all tests with good margin. It is the only way I can sleep at night and I am blessed to work for a company who supports the decisions our department makes. CISPR11 section 12 refers to the old 80/80 rule. If you do the statistical assessment, for instance, one unit can fail by 2db and be ok as long as you test 5 more units and they pass by 5db. One unit can fail by 4db if the next 7 units pass by 5db. Interesting, isn't it? How does this fit within the EU CE/DoC scheme where a manufacturer declares that every instrument they put on the market is compliant? So "compliant" doesn't necessarily mean "100% pass the limits", does it? I used to work for a man (again, in another life) who asked if he was going 55mph in a 55mph zone, if he was breaking the law? Of course the answer is, no. But the point he was making is would not another lab have to fail a product by more than their Measurement Uncertainty to be able to claim for sure that the product "Failed the Limit", and then, wouldn't the lab have to test between 3 and 12 units to perform a true Statistical Assessment to determine if the manufacturer is "Compliant" or not with the EMC Directive? Very Interesting. For this reason, I don't think many companies run into trouble with non-compliant instruments unless they fail by a fairly large margin or their products are causing very noticeable interference. Do you agree? I know it is difficult to share stories on this topic because this information can be considered confidential but I would love to hear if minor non-compliances are ever even found and when they are found what actions, if any, are taken and what penalties are imposed. We only seem to hear about the Big Non-compliances and even those stories are very rare in the area of EMC. Does anyone have first-hand experience dealing with EMC failures in the field? If you fail by 1db, are you dragged through the mud, fined, banned, prosecuted, black helicopters circle your house, masked men drag you out of bed in the middle of the night? Or is action taken only on severe non-compliances? What's the likely scenario? The Other Brian -----Original Message----- From: Pearson, John [mailto:[email protected]] Sent: Tuesday, October 13, 2015 1:34 PM To: [email protected] Subject: Re: [PSES] SV: [PSES] Stricter limits than legal (CISPR11, IEC, etc,) Where? Hello If you are selling into the EU your DoC declaring to the harmonized std (assuming you are taking this route) states that you are confirming that each and every item of product placed upon the market is compliant to the limit and not just the test sample. Does that not mean you will either have to test every item prior to shipment and sale, or apply some level of margin to feel comfortable in making this declaration? Of course testing every item is the total solution but is likely cost prohibitive unless you sell very high value items in single figures a year. Even if the stds makers have applied margin in writing the spec the DoC process does not recognize this and requires the above. -----Original Message----- From: Brian O'Connell [mailto:[email protected]] Sent: 13 October 2015 18:22 To: [email protected] Subject: Re: [PSES] SV: [PSES] Stricter limits than legal (CISPR11, IEC, etc,) Where? Bingo. Safety and EMC standards have 'built-in' margins per committee members that cared to converse with this plebian. A supplier's margin is in internal policy, or is per your customer's spec, or is per empirical numbers from the end-use installation. Measurement Uncertainty is not necessarily a 'margin', but does have statistical relevance. Control of manufacturing process and product construction cannot be reasonably specified in EMC standards, unless you want to write EMC standards to the content and format of safety standards. Have fun with that. Indeed, per Mr. Crane, why assume anything? Brian Sr Burrito and Ale Test Engineer -----Original Message----- From: Crane, Lauren [mailto:[email protected]] Sent: Tuesday, October 13, 2015 9:44 AM To: [email protected] Subject: Re: [PSES] SV: [PSES] Stricter limits than legal (CISPR11, IEC, etc,) Where? All the issues being raised regarding possible variability must be known to the members of various standards committees. Does anyone know that the issues are *not* taken into account when the committees set test levels? If standards are followed, including any instructions regarding EUT sampling and measurement uncertainty, why assume additional margins must be applied? Regards, Lauren Crane KLA-Tencor - ---------------------------------------------------------------- This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to <[email protected]> All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) 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]> - ---------------------------------------------------------------- This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to <[email protected]> All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) 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]> ________________________________ LECO Corporation Notice: This communication may contain confidential information intended for the named recipient(s) only. If you received this by mistake, please destroy it and notify us of the error. Thank you. - ---------------------------------------------------------------- This message is from the IEEE Product Safety Engineering Society emc-pstc discussion list. To post a message to the list, send your e-mail to <[email protected]> All emc-pstc postings are archived and searchable on the web at: http://www.ieee-pses.org/emc-pstc.html Attachments are not permitted but the IEEE PSES Online Communities site at http://product-compliance.oc.ieee.org/ can be used for graphics (in well-used formats), large files, etc. Website: http://www.ieee-pses.org/ Instructions: http://www.ieee-pses.org/list.html (including how to unsubscribe) 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]>
