Nick Rouse wrote: >........Most of the standard levels are set after >consideration of statistical surveys .......
also >..... Unless you have >reason to believe that the environment is markedly >worse that that envisioned in the standard or the >results of a failure are particularly catastrophic I >would not inflate the test levels above the standards >and the most severe of the European product and >gereric standards call 8kV air discharge. I would not put too much credence on the values from the immunity standards. >From what I have heard, unofficially, the immunity levels in the standard were not set through consideration of statistical surveys. My understanding of the way that the limits came about was that the power companies had pushed through the immunity standards to near adoption before the ITE industry knew anything about it. When the ITE industry finally got involved, they were too late to do anything about the implementation of immunity testing, but were able to influence the limits, so they made the limits low enough so that equipment would pass without problem. At the time that this came about, the computer companies with which I was affiliated tested their equipment to about the following values and acceptance criteria: 10 to 12 KV with no degradation of performance whatsoever. 15 KV with degradation of performance which was recoverable, but no damage to the equipment. 20 KV could have damage to components. I say about those values because we did not follow set industry standards, but followed company policy, and the values varied somewhat from company to company. The company limits were set to provide equipment which would work reliably for our customers. Then the immunity standard came out and it specified 6 KV, so we had no trouble passing the tests, but we still subjected the equipment to our own company standard of performance. Let's face it, 6 or 8 KV is just slightly above the discharge from your hand when you touch a doorknob after walking across a carpet when the air is dry. My personal opinion is that equipment should be protected to considerably higher levels, an opinion which was supported by internal company standards. We did not do radiated immunity as part of company testing for performance before the immunity standards came out. After the immunity standards came out, we had no trouble passing. The 3 V/m limit is so low that any equipment passing FCC/CISPR class A should have no trouble passing 3 V/m. Work out the math backwards from the allowable 40 dBmv at 10 meters to the field generated by the EUT at 0 meters and you'll see what I mean. Gabriel Roy Hughes Network System Maryland ....Disclaimer.... The opinions mentioned above are those of a sane man employed by Hughes Network Systems (if those two statements can be made together) and do not reflect the opinions of Hughes Network Systems, or anybody else necessarily..... but probably should. ------ snip --------------------- snip ------------------------- While it is admirable to strive for excellance in ones products it must be recognised that perfection is not achievable. Very few threats to the performance and reliability of a product have a hard cut-off value, Nearly all show a statistical drop in probability as the intensity of the threat increases. If this is so then all test levels will only provide a statistical level of protection. Most of the standard levels are set after consideration of statistical surveys I do not think corona will provide an absolute limit at 25kV. I remember physics experiments at school where a combination of frictionally induced charge and voltage amplification by conductor separation produced sparks of an inch and a half, implying a voltage of about 40kV. If you are going to apply ESD test limits way above those those judged by the standards bodies to give adequate protection in the environment concerned there is no logical reason not to do so to all the other aspects of EMC. There is absolute limit to the size of RF field that may be encountered in rare circumstances and you may occasionally encounter extremely sensitive equipment that may be interferred with by emissions way below those of the standards. Unless you have reason to believe that the environment is markedly worse that that envisioned in the standard or the results of a failure are particularly catastrophic I would not inflate the test levels above the standards and the most severe of the European product and gereric standards call 8kV air discharge. Nick Rouse
