The original message was received at Mon, 30 Dec 1996 17:51:40 -0500 from tvl1.tiac.net [206.119.76.124]
----- The following addresses had delivery problems ----- <[email protected]> (unrecoverable error) ----- Transcript of session follows ----- 550 <[email protected]>... Host unknown (Name server: .ieee.org: host not found) ----- Original message follows ----- Return-Path: [email protected] Received: from mike-hopkins (tvl1.tiac.net [206.119.76.124]) by stallion.crcmedia.com (8.6.12/8.6.12) with SMTP id RAA16064 for <[email protected]>; Mon, 30 Dec 1996 17:51:40 -0500 Message-Id: <[email protected]> X-Sender: [email protected] (Unverified) X-Mailer: Windows Eudora Light Version 1.5.4 (32) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" List-Post: [email protected] Date: Mon, 30 Dec 1996 17:52:38 -0500 To: <[email protected]> From: Michael Hopkins <[email protected]> Subject: ESD Simulators - Ben Bibb's reply to Paul A. I'd be very interested in obtaining the data and more details about the ESD tests you performed with the TEM Cell. As a manufacturer of ESD guns, KeyTek is very aware of the radiation from the simulators; however, I'm sure you are aware that for every ESD event, there IS radiated as well as conducted energy. The trick is to quantify that radiation -- not possible until recently -- and provide a simulator with a controlled radiation pattern. Unfortunately, no matter what one does, the test will be effected by the person doing the test. Unless the ESD simulator is completely shielded, as once proposed by Darren McCarthy of HP, radiation will exist as it does with real ESD events. B. The ESD Association is in the process of developing a document that will deal with the metrology of making ESD current measurements. This is being driven by HP's recognition that not all ESD simulators are alike, even though they may all meet the requirements of IEC 1000-4-2. In their study, several manufacturers guns were tested using instrumetation having a bandwidth of approximately 3GHz. The resulting current waveforms showed clearly that many (all but one) of the guns provided considerable energy at well above 1GHz. EUT's tested with such simulators showed a much higher failure level then those tested with the simulator that produced a "clean" waveform. C. David Pommerenke, formally of the Berlin Technical Institute and now with HP, has done a considerable amount of work to figure out how to measure and quantify the radiation from an ESD discharge. Some of his work was presented at the last ESD Association Symposium and at the last IEEE Symposium in Santa Clara. As his work progresses, we will gain an even better understanding of the radiation from ESD Simulators and work on ways to improve the situation, but right now, the metrology is in its infancy. We have tried to get NIST involed, and they seem to have some interest, but not enough to assist with developing the metrology. D. You say ESD simulators produce "substantial unnecessary RF fields". I for one, would very much like to know the details about how you measured these fields and with what kind of spectrum -- beyone a few GHz ?? We know the fields exist, and we know they exist in real world ESD events. Do you have enough data to allow us to quantify those fields ?? We know the events have energy in excess of 3GHz, which indicates a measurement system for a single shot event of between 100 and 300ns, must have a bandwidth of about 10GHz! As I said, NIST needs to get involved if we are to resolve the metrology issues. E. The next problem, of course, is to get the standards organizations to accept a new type simulator. World-wide standards today are being driven by the EU's EMC Directive, which means modifying IEC standards (IEC 1000-4-2), which then become mandatory EN's. This requires developing the data regarding fields (just beginning), showing that the new simulators will improve the test situation enought to justify everyone replaceing their existing simulators with new ones (not clear --- recent proposed changes to IEC 1000-4-4 were rejected because they would required many manufacturers to purchase new EFT simutors), modifying the standards, and then getting through the voting process (1 to 5 years). After all that, there will likely be 3-5 years before the new standard replaces the old one as being mandatory. Starting now, we can expect a new standard for ESD Simulators to be effective sometime around 2005. I'm about to submit a NWIP (New Work Item Proposal) to the US National Committee which will call for requireing a new type ESD Simulator in the next version of IEC 1000-4-2. This new simulator will deal with both the issues mentioned above: RF Fields, and high bandwidth measurements, but it is a very long time line. I don't know when the US National committe meets next, but it probably won't be in time for them to approve the NWIP, forward it to the IEC, and get it on the agenda for the next working group meeting in May. I expect the IEC will set up a task force in the fall of '97 to start work on a new document. Mike Hopkins Thermo Voltek [email protected]
