Neven, While I understand your point of view, I object against your arguments.. I preferred to read some sound proof instead of suggestions and statements like
Ethernet devices, as well as other high-speed devices that utilize balanced differential pairs, are very sensitive to CM-DM conversion I am not at all convinced that this is the case. I have the followoing arguments: - Most high speed receiver chips do have a decent common mode rejection. A few percent of asymmetry will not have any impact on the receiption quality. This can be proved easily by adding a grounding resistor at one input terminal of say 1kOhm introducing tens of percents of asymmetry. I also noted multiple times that adding serial impedances such as 20-30 ohms resistor or cm- coils (used to suppress EMI = common mode current) do not impact error rates (there was no error rate: 0 faults -Most (all) high speed cards do have a transformer with a considerable amount of CM to DM suppression. 20 dB is the minimum one may expect. Before reaching the receivers the CM component introduced on purpose by a ISN will be substantially attenuated by it. -The decoupling part of the ISN/CDN generally consists of a Common mode coil with extreme low leakage field (read substantial symmetry) for the frequencies the ISN is designed for. This CM coil recovers (by design) the major part of the introduced asymmetry. As long as the Diff Mode properties as a 1-2-4 times 2-port of the ISN can be proved to cover more then the spectrum of the signal that needs to be transmitted through it, i am not afraid of using one. Most transmission problems in ethernet connections relate to frequency dependent attenuation, (aging and humidity) high voltage CM (50-60 Hz) voltage introduced by mains cabling close to the UTP, TIP: do you have a large ethernet building installation which underperforms heavily ? Add a grounding resistor of 100 K to chassis ground on one wire of each of the pairs. Good chance that you have a capacitively coupled mains voltage in Common Mode on your UTP pairs. Waiting your reaction WITH arguments. Gert Gremmen ce-test, qualified testing _____ Van: [email protected] namens [email protected] Verzonden: do 10-5-2007 19:39 Aan: [email protected] Onderwerp: Re: ISN for Conducted Emission in accordance to CISPR 22 Ed 5.2:2006 Thanks to everyone who replied off- and online. I am trying to look into the TECHNICAL aspects of these standards, and unfortunately my conclusion is that they still haven't been thoroughly examined by anyone who has much experience with the details and issues related to differential high-speed multi-pair interfaces, such as balance, conversion, and how to measure these parameters. I spent a number of years in that area, and I am still not sure what measurement method is better and more appropriate (it can depend on what you are trying to see), and consequently what measured levels of balance and conversion are "true" - as they depend on how they are measured. One thing that appears evident is that a single-pair measurement is not appropriate as e.g. 4-pair interface exhibits coupling between all 4 pairs (8 lines) and relation between the common-mode and differential mode (i.e. conversion) when only one pair is driven is not the same as when the whole cable is driven by the ! common mode (or all 4 pairs are driven by differential mode). The lack of the appropriate measurement method for balance is the major reason why the levels of LCL for SLCE ISN and the required balance for CDN cannot be properly (realistically) defined, which in turn allows manufacturers of these devices to build couplers with various characteristics that can easily cause apparent failures. That's one of the reasons why I find the currently available standards inappropriate. I do not question the intent of the standards, as I believe it is good, but as Ken Hall pointed out it appears there is positive voting going on, without fully understanding the technical (and economic) implications. The methods in the standards are plausible, i.e., they seem to make sense to someone with general technical background in EMC, but at the detailed level of the specific application, e.g. Ethernet, they do not take into account the related issues, which make some methods (IMO) completely unsuitable. That’s why I currently strongly prefer clamp-methods – they don’t affect the cable characteristics. I know many have observed considerable differences in results (SLCE and Immunity) depending on the test method, as e.g. in the mentioned paper as well as in numerous conversations I had with test-engineers as well as in my first-hand observations, but I wish to draw attention of the EMC community to things that I see as the root-cause of these effects and that I believe very few people actually looked at the same level. Before they are thoroughly understood I don’t think we are ready for the standards in question in their current form and for CDN/ISN devices currently on the market. I do not sit on these standards and I am in no way associated with the committees. All my observations are coming from reading the standards and papers, my personal experience with issues observed on many projects across many companies worldwide, and from my lab work trying to measure and understand these issues, especially conversion mechanisms and balance in the Ethernet and other high-speed differential-signaling world. Regards, Neven -------------- Original message ---------------------- From: "Grace Lin" <[email protected]> > Hello Ken, > > Thank you for pointing out the paper. Sorry, I missed your presentation. I > remember my friends discussed the topic after attending your presentation. > I believe the concern they had were measuement instrument and necessity. > > I have a difficulty to understand the technical implications of telcom port > requirement. I am not challenging the CISPR committee. I just need some > help to understand the requirement/necessity. Telcom port measurement is > through the conducted method. A telecom port does not directly connect to > (interfere) public utility/power network. If it radiates, it should be > covered by the radiated emission measurement. > > From a manufacturer point of view, telecom port measurement is not a big > deal if products pass. Labs don't add additional fee for this test. For > in-house testing, there is not much work to setup a test site (using the > conducted emission site plus an ISN and adaptors). > > Best regards, > Grace > > On 5/10/07, Hall, Ken <[email protected]> wrote: > > > > Hello Grace, > > > > This topic has been argued and discussed, multiple papers, in CISPR/I AND > > CISPR/G since the early 80's. The problem may be/is political. However, with > > a few countries voting against the LCL requirements they ended up in the > > standard because > 75% of the voting countries approved. > > > > We need to influence every countries voting membership to only vote > > positive if they understand the technical implications of a standard. > > Attached is a paper from IEEE EMC 2001. > > > > Regards, > > > > KEn > > > > ------------------------------ > > *From:* [email protected] [mailto:[email protected]] *On Behalf Of *Grace > > Lin > > *Sent:* Thursday, May 10, 2007 4:00 AM > > *To:* [email protected] > > *Subject:* Re: ISN for Conducted Emission in accordance to CISPR 22 Ed 5.2 > > :2006 > > > > > > Hi Neven, > > > > Thank you for the detail comments. I suggest you write a paper to the > > IEEE International Symposium on EMC. There are many EMC experts attending > > this Symposium, including many equipment manufacturers, rule makers, and > > government agencies. The Symposium record is distributed worldwide and kept > > forever. This is a good way to get attention and make friends. > > > > Grace > > > > > > On 5/9/07, [email protected] <[email protected]> wrote: > > > > > > Be careful if you want to use ISN for Cinducted Emission or Immunity > > > testing of Ethernet or any other high speed differential-pair interface. I > > > just could not resist to reply, because of so poor specifications which > > > allow the ISN to cause APPARENT failures, and can cost the industry millions > > > to "fix" problems that do not exist. I tried to talk with a major vendor of > > > these devices, offering help to make them actually appropriate for use with > > > high-speed diff-pair interfaces ( e.g. Ethernet), but there appears to > > > be absolutely no interest. It appears there is too much politics and > > > marketing, as opposed to engineering, that dictates how these devices are > > > constructed (and specified). The cost is paid by the industry who is then > > > forced to follow the standards. I have worked on many cases where after a > > > reported emission failure, measured using ISN, the clamp-method ( e.g. > > > VOltage and Current Clamp method) resulted in very different and passing > > > result. > > > > > > So here are some of my thoughts on this topic.They are mostly based on > > > the use of CDN/ISN for Immunity, but they really equally apply to both SLCE > > > and Immunity. > > > > > > • The most recent 6100-4-6 standard specifies the common-mode (CM) > > > impedance Zcm (Table 3) at the EUT-port of the coupling device > > > • It also specifies that the Common-to-Differential mode > > > conversion (defined by Longitudinal Conversion Loss or LCL) of the CDN > > > should be better than the conversion of a cable or the equipment connected > > > to the cable, whichever conversion is lower, i.e. whichever LCL value is > > > larger (Note in 6.2.1.2) > > > • The standard recognizes that a given method and coupling device > > > may not be appropriate for all types of devices under test, so it allows for > > > alternate test methods. Specifically, in 6.2 it says: > > > > > > "If CDNs are not suitable or available, other injection methods can be > > > used". > > > > > > • The note in 6.2.1.2, regarding common to differential mode > > > conversion, recognizes that the CDNs for multi-pair balanced lines (e.g. > > > UTP) may not be suitable: > > > > > > "Often, clamp injection needs to be applied to multi-pair balanced > > > cables, > > > because suitable CDNs might not be available." * > > > > > > *What defines "suitable" can be the major limiting factor > > > > > > > > > • Rules for selecting appropriate method are given in 7.1. The > > > main deciding factor for selection of the method is (again): "are the CDNs > > > suitable?", which could result in a subjective decision > > > • Conversion can be a problem with the currently available CDN and > > > can make it not-suitable for Ethernet lines > > > • Relatively well designed RJ45/Ethernet devices can have between > > > -40 dB and -46 dB conversion (2%-1% impedance balance) > > > • Ethernet devices, as well as other high-speed devices that > > > utilize balanced differential pairs, are very sensitive to CM-DM conversion > > > • It is critical that the conversion characteristics of any > > > coupling devices be at least the same if not better than the conversion by > > > the well-designed signal lines used for the application > > > • In practice, on multi-pair balanced cables where all wires are > > > coupled (e.g. 8 wires of a CAT5 UTP), it is not easy to: > > > – specify how to measure conversion > > > – measure conversion > > > – specify required levels of conversion > > > • Lack of clear and appropriate method for measuring conversion > > > can also be a problem with currently pending draft CISPR22 emission > > > standard. > > > • The problem lies in the fact that the CDN/ISN is inserted into > > > an otherwise well-balanced line. The design of the CDN/ISN is not balanced > > > (the standards lack realistic requirement and appropriate verification > > > method) and it creates mode-conversion, which can create apparent CM > > > emission and susceptibility. This condition does not resemble the realistic > > > situation were the cable is fairly balanced and uninterrupted > > > • Conversion by the CDN/ISN can be well beyond the conversion > > > characteristics of the cable. The currently specified test method for > > > measuring conversion (or LCL) is not appropriate for multi-pair interfaces. > > > • Test equipment should not increase the conversion by introducing > > > imbalance considerably higher than that of the cable. Conversion by the test > > > equipment artificially increases the test levels beyond the characteristics > > > of the measured device and intended cabling, causing apparent failures > > > > > > > > > • Therefore, unless better CDN/ISN devices are constructed (and/or > > > better-defined in the standards), I recommend methods using clamps (e.g. > > > Voltage and Current Clamp method), which preserve integrity of the used > > > cabling and allow testing in more realistic environment. The CM-impedance > > > requirement that can be of some concern with the clamp-methods is not hard > > > to meet, at least for 4-pair differential interfaces, but that is another > > > topic. > > > > > > I hope it helps make your decision. > > > > > > > > > > > > Regards, Neven > > > > > > > > > > > > > > > > > > > > > -------------- Original message ---------------------- > > > From: < [email protected]> > > > > Hi, > > > > this may be easier than you think (or wish). > > > > To my knowledge there are only two manufacturers at the moment: Teseq > > > > and Fischer. > > > > > > > > Regards, > > > > Ari Honkala > > > > > > > > > > > > On 5/8/07, wendy <[email protected] > wrote: > > > > > > > > Dear All, > > > > > > > > > > > > > > > > My company intends to purchase the above mentioned. Will > > > > someone recommend any brand/model that compiles to the latest > > > > requirements? > > > > > > > > > > > > > > > > Regards, > > > > > > > > Wendy Nya > > > > > > > > - > > > > ---------------------------------------------------------------- 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 > > > > 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 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 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 > > > > > > > > > ---------- Forwarded message ---------- > > > From: <[email protected]> > > > To: < [email protected]> > > > Date: Tue, 8 May 2007 11:17:17 +0000 > > > Subject: RE: ISN for Conducted Emission in accordance to CISPR 22 Ed 5.2 > > > :2006 > > > Hi, > > > this may be easier than you think (or wish). > > > To my knowledge there are only two manufacturers at the moment: Teseq > > > and Fischer. > > > > > > Regards, > > > Ari Honkala > > > > > > > > > > > > On 5/8/07, wendy <[email protected] > wrote: > > > > > > > > Dear All, > > > > > > > > > > > > > > > > My company intends to purchase the above mentioned. Will someone > > > > recommend any brand/model that compiles to the latest requirements? > > > > > > > > > > > > > > > > Regards, > > > > > > > > Wendy Nya > > > > - ---------------------------------------------------------------- > > > > 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 > > > 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 > > > 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 > > > 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 > > 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 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 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 ______________________________________________________________________ - ---------------------------------------------------------------- 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
