Joe, Yes we are in agreement, and you make several valid points regarding gas discharge tubes. As is always the case in any design, there are trade-offs.
I have found it beneficial to use a combination of the very good ideas being discussed on this thread. For example, solid insulation barriers in combination with EMI filtering and surge suppression. The solid insulation an be judicially placed with thickness sufficient to prevent punch-thru and sufficient creepage distance to prevent flashover. An EMI filter adds come level of series impedance to high frequencies and surge suppression devices behind this are less likely to fail. Gas discharge tubes have another problem in that a poorly selected voltage breakdown may cause the gas to glow under normal operating conditions and these devices are nothing like the old neon lamps, they will fail as a result. In addition, when they do fire, they are unlike MOVs in that they clamp to near zero volts and the only way to extinguish the plasma within the tube is a zero crossing of the line voltage. MOVs are always suspect because of the leakage current heating problem and catastrophic failure mode where they sputter metal on nearby surfaces. There are companies who make air gaps and these are viable so long as they do not have a problem with corona when the electrodes erode (due to arcing) into ragged edges. Sharp points can lower the breakdown voltage (e-fields) just as the ice pick experiment did back in high school. You would also need a way to replenish the air supply within the gaps since corona can build up and eventually arc over without requiring a surge event. In any design it is useful to test your mitigation attempts in the actual product design. thanks, –doug Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01 From: Joe Randolph Sent: Friday, September 06, 2013 11:34 AM To: Doug Powell ; EMC-PSTC@LISTSERV.IEEE.ORG Subject: Re: [PSES] Spark Gap PCB Layout on AC Mains Hi Doug: I think we are in agreement about some of the pitfalls of spark gaps that are simply added in the copper layout. Regarding "spark gap" components, the devices you cite (also known as gas tubes) typically fire at just a few hundred volts. Adding these to enhance the surge tolerance of an air gap can have unintended consequences. My philosophy for lightning protection is that the designer has only two options: 1) Block the surge current using adequate insulation 2) Direct the surge current to a known, safe path My preference is to use Option 1 whenever possible. As I noted in my earlier posting, the probability of experiencing a given surge drops off dramatically with increasing voltage. So, if a product is designed with spacings that break down at 10 KV, the probability of a surge exceeding that threshold is fairly low. If you then add a 400 volt gas tube across the barrier, the probability of a surge exceeding that threshold is far higher. The result is that the gas tube "invites" surge current that otherwise would not have flowed. This may be okay *if* the gas tube is reliable and *if* the surge path that it creates is also reliable and robust. The key thing to be aware of is that with 400 volt gas tubes installed across the barrier, surge currents will flow across the barrier on a fairly regular basis. If the system has a 10,000 volt breakdown without the gas tubes installed, surge currents flowing across the barrier will be far less common without the gas tubes (probably by a factor of 1000 or more). Another unintended consequence of using conventional gas tubes is that they create a very steep current rise time when they trigger. This generates an electromagnetic pulse that can propagate through the system and upset sensitive electronics. In the IEEE PSES Telecom TAC we have been discussing this problem for a while now. Interestingly, some Ethernet ports have higher failure rates with gas tubes installed than without the gas tubes installed. For the above reasons, I prefer to use Option 1 (block the surge current) whenever possible. If I had a system that could withstand 10 KV but I wanted to add a spark gap for the (rare) cases where the surge exceeds 10 KV, I would try to set the spark gap breakdown as high as possible, such as 9 KV. I am not aware a conventional gas tube with a 9 KV threshold, but perhaps such devices exist. One alternative that I have often wished someone would make is a simple air gap component with tungsten electrodes that could withstand multiple surges. Such a device could be made quite inexpensively and would be more robust than a simple gap in a copper layout. If the nominal trigger voltage was very high (say, 9 KV), the device could be expected to trigger very rarely, if ever, in actual use. Joe Randolph Telecom Design Consultant Randolph Telecom, Inc. 781-721-2848 (USA) j...@randolph-telecom.com http://www.randolph-telecom.com Brian, I am personally opposed to such an idea. The first time I saw this was circa 1982 on an electronic security system; designed to tie into the public telephone network. The manufacturer wanted additional protection above the outdoor surge protector. The design involved two parallel zig-zag traces, one ground and the other telephone line with alternating points in close proximity. The problem with this is if the gaps you created ever activate, they always leave a permanent carbon track in the surface of the PCB. Given adequate time and humidity exposure, these tracks become conductive and leakage current can begin to increase dramatically. Eventually, this circuits will become permanently shorted. This the main reason for evaluating comparative tracking index (CTI) of insulating materials along with creepage evaluation. Best to simply use a glass or ceramic spark gaps which are inorganic and cannot produce carbon when arced (http://www.globalsources.com/manufacturers/Glass-Switching-Spark-Gap.html). thanks, doug Douglas E Powell doug...@gmail.com http://www.linkedin.com/in/dougp01 From: Kunde, Brian Sent: Friday, September 06, 2013 9:56 AM To: EMC-PSTC@LISTSERV.IEEE.ORG Subject: [PSES] Spark Gap PCB Layout on AC Mains Our engineers are working on an AC Mains Distribution PCB. Like most electronic devices, we have seen the damage caused by lightning strikes. So we are increasing our creepage and clearance distances as wide as we can and still meet other requirements. But no matter what spacing you design to, there is a lightning bolt out there that will exceed the design and it will arc somewhere. So the question came up to whether it makes sense to deliberately make a weak spot, or an area where the clearance is slightly smaller to control where a lightning/surge pulse will arc and/or discharge, like a Spark-Gap. I have seen spark-gap lay outs on PC boards on I/O connectors; usually for ESD protection, but not on AC Mains. Is this a bad bad idea or something worth doing? Pros and Cons? Other suggestions?? Thanks to all for your help. The Other Brian ------------------------------------------------------------------------------ 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 <emc-p...@ieee.org> 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://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 <emcp...@radiusnorth.net> Mike Cantwell <mcantw...@ieee.org> For policy questions, send mail to: Jim Bacher <j.bac...@ieee.org> David Heald <dhe...@gmail.com> - ---------------------------------------------------------------- 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 emc-p...@ieee.org 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://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 emcp...@radiusnorth.net Mike Cantwell mcantw...@ieee.org For policy questions, send mail to: Jim Bacher j.bac...@ieee.org David Heald dhe...@gmail.com Joe Randolph Telecom Design Consultant Randolph Telecom, Inc. 781-721-2848 (USA) j...@randolph-telecom.com http://www.randolph-telecom.com - ---------------------------------------------------------------- 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 <emc-p...@ieee.org> 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://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 <emcp...@radiusnorth.net> Mike Cantwell <mcantw...@ieee.org> For policy questions, send mail to: Jim Bacher: <j.bac...@ieee.org> David Heald: <dhe...@gmail.com>
