There is another implied concept of questionable validity in this latest posting. The way I read it, Mr. Wellington is talking about filtering signals emanating from the support equipment as it passes through a bulkhead between control and test chambers. Such signals usually require no filtering whatsoever, because if they have any bandwidth at all, they are run with dedicated returns (such as a twisted pair or a twisted shielded pair) and have no radiation efficiency to speak of. What requires filtering, and what is ameliorated by proper PCB layout, as he alluded to in an earlier post, is common mode emissions. These can be filtered to a very high degree with no impact on intentional signals. Common fixes are snap on ferrite beads and line-to-ground caps. Clearly the line-to-ground caps should not attenuate the desired signal, but in most if not all cases the undesired cm current is orders of magnitude higher in frequency than the intentional signal. In those cases where this is not the case, the military would run intentionally high frequency signals within shielded cables (think MIL-STD-1553 and fibre channel) which, if properly terminated, will provide all the protection necessary without resort to either filtering or over-braids. Specifically, if the support equipment or its environment resulted in high frequency cm currents conducted on the outside of a shielded cable, that cable should be terminated peripherally to a connector at the bulkhead as it passes into the test chamber. If the cable is not shielded and the source is the ambient, then shielding of the cable external to the test chamber is both proper and necessary, and has no effect on the validity of the test set-up within the chamber. Further, it requires no input from the customer, because it does not affect the delivered product configuration. If the test support equipment itself is the cm source, then any cm filtering necessary to attenuate those emissions before they enter the test chamber is again proper and necessary, external to the test chamber. It might be said that such filtering could reduce cm currents emanating from the test sample, but this is not a big problem for a couple of reasons. First it is easy to determine whether it is support equipment or the test sample which is driving the currents, by sequentially de-energizing suspected sources and noting the effect on the emissions. Secondly, the standard effectively requires at least 3 meters of cabling between test sample and bulkhead. Above 10 MHz the cable is electrically long and the effect of a filter at the bulkhead does not directly impact the level of cm current on the cable, but only indirectly as its impedance is transformed by the electrical length and distributed characteristic impedance of the cable in question. If the mil-std EMI test set-up were so well-controlled that every test chamber and every test bench were of precisely equal size and configuration and no matter where the test was conducted the entire test set-up including cable layout were identical within inches, then it might be productive to worry about changing a common mode impedance at the end of a three meter cable. In my experience, such is hardly the case.
on 2/20/03 4:53 PM, pwell...@csw.l-3com.com at pwell...@csw.l-3com.com wrote: > > Hi, > > I still disagree with the general use of filter connectors in an EMC test > set-up at the egress. > > If you look at MIL-STD-461C and the test methods document MIL-STD-462 (which > is what he is testing to - RE-02), you will see that you are required to > test in a simulation of the "actual installation and usage". > > This is located in MIL-STD-462 section 4.2.2.3 where it says that > "Interconnect cable assemblies and supporting structures shall simulate > actual installation and usage." > > MIL-STD-462 section 4.2.2.5 on Loads also requires you to use loads that are > "equivalent for which it is designed". It further states that "The loads > used shall simulate the resistance, inductance, and capacitance of the > load". Filter connectors at the chamber egress would be additional > capacitance and possibly inductance (type dependent) that modifies the > simulated "load" that is supposed to look like the application. Now, if the > application or user equipment has filters or filter connectors at their > egress, you would be simulating the proper "load" and be compliant to > MIL-STD-462. > > You are also required to meet a minimum of 6 dB margin below the limit line > with your EUT "deenergized" unless you have to test under section 4.2.8 > (Emission Measurements in the Presence of High Ambient Fields). The filter > connectors will help you meet the required ambient margin, but should not be > used as a general rule. > > You are correct about filtering power (AC and DC). It is usually expected to > have 100 dB isolation from 10 KHz to 1 GHz. Power is different in that you > are required to have capacitors (or LISNs for 461D/E) for RE-02. During > testing, we intentionally measure emissions from the EUT during Conducted > Emission tests and isolate emissions to the EUT not power cabling from the > chamber to the capacitors (or LISNs for 461D/E). The capacitors/LISNs are to > normalize the impedance for obtaining repeatable test data. This is > different from signal line filtering where the intended signals may be > modified (edge rates limited) causing errors in measured data from the lab > versus the customer application. If you use filter connectors at the chamber > egress in your lab, the cables appear to have the same RF impedance. This is > very likely *not* the same as the intended application. Since all test labs > do not use filter connectors at the egress, there may be wide measurement > variation between labs. This would be the case from lab data with filter > connectors to the customer application of the product. > > > What this all means is that you need clean exercise equipment, cabling, and > well bonded cable shielding to meet EMC test set-up requirements. It also > means that you need a clean Electromagnetic environment like an ante room > for the exercise equipment. You may also need a room for the measurement > equipment to shield it and the antenna cabling from the ambient > Electromagnetic environment. > > The best way to control these set-up problems is to use a chamber with an > Ante room for the exercise equipment. This is a shielded enclosure (room) > that has an egress panel between the EMC test chamber and the ante room. The > door may be closed to prevent outside emissions from being conducted onto > exercise cables and re-radiated on cables in the EMC test chamber. At our > facility, we have a 40' X 40' X 28' shielded chamber with Anechoic material > and ferrite tiles, 12' X 16' Ante room chamber, and 10' X 12' measurement > equipment room, plus a 12' X 16' development chamber. We have never used > filter connectors through the chamber egress, we don't think it is proper > set-up. But, others may not have the facility to do this. > > While every situation may not be able to be tested this way, in my opinion, > using filter connectors as a general rule for a MIL-STD-461 EMI test set-up > should be discouraged. > > Meeting MIL-STD-461 levels is difficult, but can be done with careful > attention to lower level source suppression design attributes, Signal > Integrity design, well thought out containment methods, and well > designed/built cabling. These are always less expensive and less painful at > the beginning of the design cycle rather than at the end. > > > One last thing is that if a military customer agreed that filter connectors > are acceptable for the test configuration, they have full authority to > accept the test configuration. In that case, I would expect to have this > well documented in my proposals, EMC Test Plan, and EMC Test Report. > > Sorry for the long response. > > > Philip Ross Wellington > Mgr. Signal Integrity & EMI > L-3 Communications CSW > > > -----Original Message----- > From: Cortland Richmond [mailto:72146....@compuserve.com] > Sent: Wednesday, February 19, 2003 7:33 PM > To: ieee pstc list > Subject: RE: RE02 cabling problem > > > > > Philip Ross Wellington wrote: > >>> I do not recommend that you use heavy filtering at the chamber egress > wall > to control emissions for a couple of reasons... << > > Valid statements. However, it's necessary to control chamber ingress, > otherwise the test would be contaminated by the outside EM environment and > no good test could be done. An external signal filter does that. > > We heavily filter POWER, after all, and for the same reasons. > > Cortland > > ------------------------------------------- > This message is from the IEEE EMC Society Product Safety > Technical Committee emc-pstc discussion list. > > Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ > > To cancel your subscription, send mail to: > majord...@ieee.org > with the single line: > unsubscribe emc-pstc > > For help, send mail to the list administrators: > Ron Pickard: emc-p...@hypercom.com > Dave Heald: davehe...@attbi.com > > For policy questions, send mail to: > Richard Nute: ri...@ieee.org > Jim Bacher: j.bac...@ieee.org > > Archive is being moved, we will announce when it is back on-line. > All emc-pstc postings are archived and searchable on the web at: > http://www.ieeecommunities.org/emc-pstc > > ------------------------------------------- > This message is from the IEEE EMC Society Product Safety > Technical Committee emc-pstc discussion list. > > Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ > > To cancel your subscription, send mail to: > majord...@ieee.org > with the single line: > unsubscribe emc-pstc > > For help, send mail to the list administrators: > Ron Pickard: emc-p...@hypercom.com > Dave Heald: davehe...@attbi.com > > For policy questions, send mail to: > Richard Nute: ri...@ieee.org > Jim Bacher: j.bac...@ieee.org > > Archive is being moved, we will announce when it is back on-line. > All emc-pstc postings are archived and searchable on the web at: > http://www.ieeecommunities.org/emc-pstc > -- Ken Javor EMC Compliance Huntsville, Alabama 256/650-5261 This message is from the IEEE EMC Society Product Safety Technical Committee emc-pstc discussion list. Visit our web site at: http://www.ewh.ieee.org/soc/emcs/pstc/ To cancel your subscription, send mail to: majord...@ieee.org with the single line: unsubscribe emc-pstc For help, send mail to the list administrators: Ron Pickard: emc-p...@hypercom.com Dave Heald: davehe...@attbi.com For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org Archive is being moved, we will announce when it is back on-line. All emc-pstc postings are archived and searchable on the web at: http://www.ieeecommunities.org/emc-pstc
