Re: Selection of Directives
I read in !emc-pstc that am...@westin-emission.no wrote (in ) about 'Selection of Directives', on Sun, 6 Jan 2002: >Yes, is possible to make a magical Nothing magical involved. > matrix, but I think it would take much >much time to generate it. Maybe, if you want the matrix to eliminate the need ever to refer to a Directive itself. But your list gives, IMO, a wrong impression. In practice, you have a specific product, and you most likely have a very good idea whether the ATEX Directive applies or not, right from the beginning of the project. For example, if it is a food-mixer, the Directive is unlikely to apply. So you don't have to look all through that long list, only at the item (maybe two items) that could apply to your product. -- Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk After swimming across the Hellespont, I felt like a Hero. --- 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: Michael Garretson:pstc_ad...@garretson.org Dave Healddavehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: No longer online until our new server is brought online and the old messages are imported into the new server.
RE: SMPS Derating reqs for Altitude range of 5000-10000ft
While the question and discussion related to SMPS, and brought up spacing and cooling issues, you might also consider the rest of the system. Disk drives in the past have had altitude problems due to flying height of heads. I assume this is still an issue since technology is always pushing the limits of performance. Bob Johnson --- 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: Michael Garretson:pstc_ad...@garretson.org Dave Healddavehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: No longer online until our new server is brought online and the old messages are imported into the new server.
Re: EMC-related safety issues
I abstracted the physical process and did not attach numbers. 80 dBuV/m is orders of magnitude above RE limits. That is 10 mV/m.If you use the worst case assumptions of IEC 1000-4-6 and associate an open circuit drive potential numerically equivalent to the field intensity, but with a 150 Ohm source impedance, then you would have 10 mV driving current from a 150 Ohm source. If you assume the impedance (not transfer impedance) of the shield to be a dead short, then 10 mV will cause 67 uA to flow on the shield. If shield transfer impedance were 100 milliohms (pretty high, especially at frequencies where an op-amp might respond) you would get 6.7 uV induced common mode on the circuit. Assuming no CMR, this is the potential that is added to the intentional signal. That can affect a radio front-end, but not an op-amp. Once again, unintentional RE from ITE, even at unsuppressed levels, do not cause interference to other non-antenna connected electronics. on 1/6/02 10:28 AM, cherryclo...@aol.com at cherryclo...@aol.com wrote: Ken, can we take it that in the posting below you are agreeing that interference with non radio-receiving circuits from what you meant by "unintentional emitters" is a possibility, albeit a worst-case one? If my reading above is correct, how can you then go on to say that "...unintentional emissions from ITE can only upset a radio receiver tuned to the emission frequency." It seems to me that the words "can only" in the your quote above should be replaced by "are most likely to", which I would agree with. Regards, Keith Armstrong In a message dated 06/01/02 06:56:07 GMT Standard Time, ken.ja...@emccompliance.com writes: Subj:Re: EMC-related safety issues List-Post: emc-pstc@listserv.ieee.org Date:06/01/02 06:56:07 GMT Standard Time From:ken.ja...@emccompliance.com (Ken Javor) To:t...@tncokenias.org (Tom Cokenias), cherryclo...@aol.com, emc-p...@majordomo.ieee.org The analytical portion of this post is, as the author stated, worst case. A cable attached to a susceptible circuit picks up a common-mode potential, which most likely drives a current on a shield if the the circuit is sensitive. Then only the current multiplied by shield transfer impedance actually gets into the victim, assuming no CMR. Which just makes my original point - unintentional emissions from ITE can only upset a radio receiver tuned to the emission frequency. That is why, as another contributor posted, we use EMI receivers and spectrum analyzers with preamps to make OATS measurements. on 1/4/02 12:51 PM, Tom Cokenias at t...@tncokenias.org wrote: > At 8:34 AM -0500 1/4/2002, Keith Armstrong wrote: > >> Does anyone else think that ordinary semiconductors doesn't respond to RF? > > > > I agree that commonly used semiconductors have responses well into > the 100's of MHz. > > How much of a problem this is will depend on the nature and function > of the circuitry using these components. > > The EUT wires, cables, pcb traces etc. act like antennae, on which > the incident field voltages and currents. An antenna factor can be > thought of as ratio of the field strength to the voltage induced on > the terminated cable connected to the antenna. > > In an impedance matched system, > > > AF=9.734/lamda*(G)^0.5, lamda being wavelength in meters, G being > antenna gain over isotropic, > > or in dB > > AF dB = - G dBi -29.7 dB + 20logFMHz > > Assuming G is 1 (isotropic antenna), AF is 1 (= 0 dB) at about 30.8 > MHz, and AF get larger as frequency increases, to a factor of 32.7 > (= 30.3 dB) at 1 GHz . Since AF is field strength divided by > induced voltage, the voltage induced on the trace goes down as > frequency goes up for the same incident field strength. > > An effective receive antenna needs to be on the order of 1/2 > wavelength or so; for 30 MHz this is 15m, for 1000 MHz this is 15 cm. > > So if a victim EUT circuit has a pretty effective receive antenna, > and does not have any filtering and is equally sensitive across the > frequency range under consideration (all taken together, a worst case > scenario for susceptibility), > > (1) A 10 V/m field will theoretically induce a voltage 0.33V to > 10V, depending on frequency > > (2) A 5000 uV/m field (10x the FCC class B limit above 960 MHz) will > theoretically induce a voltage from 152 uV to 5 mV, depending on > frequency. > > (3) A 500 uV/m field will theoretically induce a voltage from 15 uV > to 500 uV depending on frequency. > > These are first order approximations, but they are useful in > determining the level of the potential EMI threat. For instance a > 4-30 mA sensor circuit using high gain operational amps will most > likely be interfered with in scenario (1), there may be some > susceptibility detected in scenario (2), and most likely no problem > encountered with scenario (3). > > A sensitive all - band AM communications receiver will have problems > with all three, a broadcast TV opera
Re: Selection of Directives
Only to tell how complex is would be to make a practical magical matrix, take a look below copied from the ATEX guide: EQUIPMENT SPECIFICALLY EXCLUDED FROM DIRECTIVE 94/9/EC These exclusions are based on Article 1.4, laid down in directive 94/9/EC: 1. medical devices intended for use in a medical environment; 2. equipment and protective systems where the explosion hazard results exclusively from the presence of explosive substances or unstable chemical substances; 3. equipment intended for use in domestic and non-commercial environments where potentially explosive atmospheres may only rarely be created, solely as a result of the accidental leakage of fuel gas; 4. personal protective equipment covered by directive 89/686/EEC 39 . There are occasions where personal protective equipment with its own potential sources of ignition is intended for use in potentially explosive atmospheres. This type of personal protective equipment should follow the procedures laid down in directive 94/9/EC to provide the necessary level of explosion safety (see as well chapter 6); 5. seagoing vessels and mobile offshore units together with equipment on board such vessels or units, because they are already covered by the IMO Convention. However, fixed offshore units together with equipment on board, and units and vessels, which are not considered to be seagoing (i.e. below 500 tonnes, not intended for high sea but intended for internal navigation on river, ship canal, lakes), are in the scope of directive 94/9/EC; 6. means of transport i.e. vehicles and their trailers intended solely for transporting passengers by air, road, rail or water networks, as well as means of transport in so far as such means are designed for transporting goods by air, by public road or rail networks or by water. Vehicles intended for use in a potentially explosive atmosphere shall not be excluded; 7. equipment covered by Article 296 (1)(b) of the EC Treaty, i.e. designed and manufactured specifically for use by the armed forces or in the maintenance of law and order. Dual-purpose equipment is not excluded. Yes, is possible to make a magical matrix, but I think it would take much much time to generate it. Amund -Opprinnelig melding- Fra: owner-emc-p...@majordomo.ieee.org [mailto:owner-emc-p...@majordomo.ieee.org]På vegne av John Woodgate Sendt: 5. januar 2002 21:07 Til: emc-p...@majordomo.ieee.org Emne: Re: SV: Selection of Directives I read in !emc-pstc that am...@westin-emission.no wrote (in ) about 'SV: Selection of Directives', on Sat, 5 Jan 2002: >I wish I had seen this "magical matrix", but unfortunately I have not. I >assume that a matrix like this would be very complex and comprehensive. No, a matrix is perhaps the best way of handling the subject > >Example: An ITE would have to qualify for EMCD and maybe RTTE, RTTE only in respect of any relevant interface. > maybe also >LVD if the voltages are within the scope Yes. Quite normal. >and if it is placed in an explosive >area then the ATEX directive might also apply. Well, it would apply. >A lot of maybe's and if's No, the matter seems quite clear. What uncertainty do you have in mind? >... >I think we have to gain knowledge about a lot of the directives in order to >know if a product falls within it. A matrix can not include that type of >experience I think. Header and one line of a matrix: EquipmentLVDEMCD RTTE ATEX ITEYESYESModem interfaceOnly if intended for Bluetooth explosive atmosphere > >I'm looking forward to hear other inputs from the list members. -- Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk After swimming across the Hellespont, I felt like a Hero. --- 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: Michael Garretson:pstc_ad...@garretson.org Dave Healddavehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: No longer online until our new server is brought online and the old messages are imported into the new server. <>
Re: EMC-related safety issues
I read in !emc-pstc that cherryclo...@aol.com wrote (in <14b.6d4a617.296 9c...@aol.com>) about 'EMC-related safety issues', on Sun, 6 Jan 2002: >Dear John >The incubator I described was already on the EU market in the latter half > of >the 1990s, when I helped to test and fix it. > >And I'm sorry to disappoint Inappropriate word; I'm not interested in scoring debating points but exploring the approaches to 'EMC and Safety', which I think need to be explored. >but I have already experienced several similar >examples I could quote, such as the electric blanket that would change its >heat settings randomly when a bedside light was switched on or off, or > from >other low-level mains transients. >This is a potentially fatal issue for certain kinds of invalid, or people >who are blind drunk (surely no person reading this would ever be in such a >state) and by the way, this is not me being emotive again, I agree; what you have written here is not emotive. >it was the >expressed concern of the manufacturer and one of the reasons why they > called >me in. They sacked their Technical Director over this incident. >They also didn't do a product recall despite having an estimated 100,000 >products with the problem already out in the field. Of course, as a >responsible engineer (and to cover my ass) I wrote them a letter >recommending that they did a product recall (while thinking of the > designers >of the Challenger Space Shuttle's infamous O-ring seals). > >I find that many independent EMC people have dozens of similar examples, >which they can't talk about very much because of commercial > confidentiality. >This is one reason why the EMC + Compliance Journal >(www.compliance-club.com) started its 'Banana Skins' column - to help >educate practising engineers about real EMC engineering problems they > almost >certainly weren't taught about at college and may not (yet) have > experienced >for themselves. But, by its nature, it tends to report very low-probability occurrences and/or anecdotes, which are probably not very effective as training examples. > >I also have personal experience of a UK company that in the late 90's was >selling a range of over 110 CE-marked products (such as incubators) > intended >for medical and chemical laboratories although less than 10% of their >products met both the EMCD and the LVD. The company in question had just >been purchased by another, which is why I was involved. > >Interestingly, the new owners continued to sell the non-compliant products >while they re-engineered them one at a time to be compliant (which took >several years). > >My simple investigations over a number of years into a number of > companies' >CE marked products have led me to be very cynical. As a rule of thumb I >guess that around 30% of CE marked products are non-compliant with EMC or >LVD, with another 30% being borderline cases. This seems to be borne out > by >recent enforcement surveys in Finland and in the UK and published articles >from some test labs. You experiences are certainly a great deal worse than mine. I do find quite a few 'compliance failures' i.e. things like incorrect labelling or items omitted from instruction books, but few real hazards. I have found many products that were submitted for pre-compliance EMC assessment that would never pass unless completely redesigned, but I haven't (yet) seen anything CE marked that obviously fails. > >Changing to another of your criticisms below... >If you think my proposed statement is fog-filled, what do you propose >instead? >Lets have constructive criticism instead of merely criticism. There are many possible statements that could be made on the subject, but here is one: 'Conducted and radiated emissions from equipment which does not include any one of: - switching of voltages above 10 V and currents above 100 mA; - generation of radio-frequency (150 kHz to 400 GHz) voltages above 100 mV; - power consumption greater than 75 W are extremely unlikely to cause malfunction of other equipment having the degree of immunity afforded by normal design practices.' This is the sort of statement/guideline that can be used by a designer or compliance engineer to determine what testing, if any, is necessary. > >In fact, in most scientific or engineering activities, one can only make >public statements using foggy words like 'generally'. See above. >Remember the UK government's teams of scientific advisors and their >pronouncements on BSE and the foot and mouth epidemic? Would you have >expected them to produce precise and accurate predictions? >I am of the opinion that the outbreak of foot and mouth disease in the UK >was better understood, had fewer variables, and could be better controlled >
Re: EMC-related safety issues
I read in !emc-pstc that cherryclo...@aol.com wrote (in <162.6b92ca5.296 9c...@aol.com>) about 'EMC-related safety issues', on Sun, 6 Jan 2002: >Yes, John, you are quite right in both your comments as far as you go: > >1) You are not the only person who can dramatise an issue so as to > encourage >people to debate it; I don't know what you are referring to. I have 146 articles already read in the thread: I don't see that the debate needs any encouragement. > >2) If you sold a single electronic safety-related circuit with a failure >probability of 10^ -9 to 100,000 customers the cumulative failure >probability is indeed 10^ -4. As you correctly said, Olber's Paradox does >not apply in this area. > >But nevertheless this does not mean we need to make electronic circuits > with >failure rates equal to or better than 10^ -9. As you have said (and I > agree) >this would be a very difficult task indeed and likely to be very > expensive, >especially for any product using software. > >So how do we square this particular circle? > >Those members who are familiar with safety engineering techniques will be >familiar with the idea of building very reliable systems up using a number >of independent systems or devices each with lower reliability. These have >various names, such as 'redundant channels' or 'duplicate channels' or >'safety back-ups' or 'fail-safe circuits' and many others. > I don't see how this applies to the reduction of emissions or, practicably, to the improvement of immunity. Do you envisage three separate systems in every product, with majority voting? I suspect that in terms of improving immunity, it would be ineffective, because a disturbance that compromised one system would be very likely to compromise at least one other. Consider you incubator, for example. -- Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk After swimming across the Hellespont, I felt like a Hero. --- 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: Michael Garretson:pstc_ad...@garretson.org Dave Healddavehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: No longer online until our new server is brought online and the old messages are imported into the new server.
Re: EMC-related safety issues
Ken, I believe other postings on this topic this weekend clearly show that electronic circuits which were not designed as radio receivers can possibly be interfered with by the emissions from products which meet FCC/CISPR 22 limits, for a number of possible reasons, especially when the product is closer than 10 metres to the source of the emissions. Certain kinds of transducers and their amplifiers appear to be particularly at risk because of the low levels of their transducer signals. Regards, Keith Armstrong In a message dated 06/01/02 06:56:28 GMT Standard Time, ken.ja...@emccompliance.com writes: > Subj:Re: EMC-related safety issues > Date:06/01/02 06:56:28 GMT Standard Time > From:ken.ja...@emccompliance.com (Ken Javor) > To:cherryclo...@aol.com > CC:emc-p...@majordomo.ieee.org > > My point was that only radios are sensitive to rf fields at the levels > controlled by FCC/CISPR22 and indeed, as Ing. Gremen pointed out, levels > well above the limits. Which means that the only rationale behind > FCC/CISPR22 is protection of radio broadcast reception. Period. > > on 1/5/02 12:10 PM, cherryclo...@aol.com at cherryclo...@aol.com wrote: > > >> Dear Ken >> I am truly sorry if I irritated you by misunderstanding your words, but I >> took your posting to imply that electronic circuits which are not designed >> as RF receivers would not respond very well to radio frequencies. >> >> My example was not intended to be a full answer to your example (there are >> other postings which are dealing with that) just to indicate that the >> frequency response of slow and commonplace ICs can be very high indeed. >> >> I am sensitive to this issue because I keep on running across electronics >> designers who say things like: "I don't need to worry about the RF >> immunity of my audio amplifier/motor >> controller/temperature/pressure/flow/weight/velocity measurement and >> control system (please delete where applicable) because the opamps I use >> have a GBW of under 1MHz so they won't see the RF" which is of >> course complete bollocks (a UK phrase that I hope translates well enough >> for all emc-pstc subscribers). >> >> And no, I still don't agree with you that only radio receivers are >> sensitive enough to RF to have a problem with what you are still calling >> 'unintentional emissions' (even though this term means very little in an >> international forum unless you define the relevant standards or laws). >> >> I think the problem you are concerned with is application dependant and we >> cannot make such broad assumptions. As I said earlier, most interference >> problems are caused by radio transmitters or radio receivers, but not all. >> >> Regards, Keith Armstrong >> >> In a message dated 05/01/02 01:20:27 GMT Standard Time, >> ken.ja...@emccompliance.com writes: >> >> >>> Subj:Re: EMC-related safety issues >>> Date:05/01/02 01:20:27 GMT Standard Time >>> From:ken.ja...@emccompliance.com (Ken Javor) >>> To:cherryclo...@aol.com, emc-p...@majordomo.ieee.org >>> >>> One sure way to REALLY irritate me is to twist my words and try to make >>> me look stupid (I do a fine job by myself on occasion and don't >>> appreciate any outside help). I did not say that pn junctions don't >>> detect and rectify rf, I said that the field intensities associated with >>> unintentional emissions from ITE are too low to cause susceptibility in >>> circuits other than radios. Your example here is 10 V/m, and you are >>> talking about an op-amp (gain unspecified) and that it was susceptible at >>> that level should be no surprise to anyone. >>> >>> on 1/4/02 7:34 AM, cherryclo...@aol.com at cherryclo...@aol.com wrote: >>> >>> Does anyone else think that ordinary semiconductors doesn't respond to RF? I have tested a product which was little more than an LM324 quad op-amp for RF immunity using IEC 61000-4-3. This op-amp has a slew rate of 1V/micro-second on a good day with the wind in its favour. It was housed in an unshielded plastic enclosure. Demodulated noise that exceeded the (not very tough) product specification were seen all the way up to 500MHz at a number of spot frequencies that appeared to be due to the natural resonances of the input and output cables. Above 500MHz this resonant behaviour vanished to be replaced by a steadily rising level of demodulated 1kHz tone as the frequency increased. I stopped testing at 1GHz, where the output error from the product was about 10% and still rising with increased frequency. OK, the field strength for the test was 10V/m (unmodulated) but the real surprise was how well this very cheap and very slow opamp demodulated the RF, and that it demodulated better at 1GHz than at 500MHz. I have done many many immunity tests using IEC 61000-4-3 on audio equipment and found muc
Re: EMC-related safety issues
Taking Toms' calculations a little further Typical thermocouple sensors have output voltages in the range 3 - 50 microvolts per degree C. So to create a 60C error in a thermocouple-based temperature control system (see my recent posting about the RF immunity of a blood sample incubator) all we need is an error voltage in the range 180 microvolts to 3 millivolts. According to Toms' calculations below: "(3) A 500 uV/m field will theoretically induce a voltage from 15 uV to 500 uV depending on frequency." Now, 500 microvolts per meter = 54dBmicrovolts/meter which is only 7dB above the CISPR 22 10 metre Class A limit and 17dB above the Class B limit, between 230 and 1000MHz. This implies a fully CISPR 22 compliant laptop PC is capable of generating significant errors (10s of degrees C) in certain kinds of thermocouple measuring and control systems even when the laptop is 10 metres away. If the laptop was closer to the thermocouple system than 10 metres the field strength of its RF emissions will obviously increase. But when it is in the laptop's near field we would expect the 10 metre measurements to be meaningless because there are components of near field emissions which fall off with the cube of the distance and are not detected by 10 metre tests. For example, many products seem to have very quite strong frequency magnetic fields nearby, at audio frequencies and lower, caused by variations in the loading on their DC power supplies. For this and several other reasons I can't agree with the idea that emissions from what are known under some US laws as "unintentional radiators" cannot possibly cause interference in circuits which are not intentional radio receivers. It seems to me that many types of transducer systems (maybe even the fluxgate magnetometers used in some types of compasses) can be vulnerable. Regards, Keith Armstrong In a message dated 04/01/02 19:03:15 GMT Standard Time, t...@tncokenias.org writes: > Subj:Re: EMC-related safety issues > Date:04/01/02 19:03:15 GMT Standard Time > From:t...@tncokenias.org (Tom Cokenias) > Sender:owner-emc-p...@majordomo.ieee.org > Reply-to: mailto:t...@tncokenias.org";>t...@tncokenias.org (Tom > Cokenias) > To:cherryclo...@aol.com, emc-p...@majordomo.ieee.org > > At 8:34 AM -0500 1/4/2002, Keith Armstrong wrote: > > >Does anyone else think that ordinary semiconductors doesn't respond to RF? > > > > I agree that commonly used semiconductors have responses well into > the 100's of MHz. > > How much of a problem this is will depend on the nature and function > of the circuitry using these components. > > The EUT wires, cables, pcb traces etc. act like antennae, on which > the incident field voltages and currents. An antenna factor can be > thought of as ratio of the field strength to the voltage induced on > the terminated cable connected to the antenna. > > In an impedance matched system, > > > AF=9.734/lamda*(G)^0.5, lamda being wavelength in meters, G being > antenna gain over isotropic, > > or in dB > > AF dB = - G dBi -29.7 dB + 20logFMHz > > Assuming G is 1 (isotropic antenna), AF is 1 (= 0 dB) at about 30.8 > MHz, and AF get larger as frequency increases, to a factor of 32.7 > (= 30.3 dB) at 1 GHz . Since AF is field strength divided by > induced voltage, the voltage induced on the trace goes down as > frequency goes up for the same incident field strength. > > An effective receive antenna needs to be on the order of 1/2 > wavelength or so; for 30 MHz this is 15m, for 1000 MHz this is 15 cm. > > So if a victim EUT circuit has a pretty effective receive antenna, > and does not have any filtering and is equally sensitive across the > frequency range under consideration (all taken together, a worst case > scenario for susceptibility), > > (1) A 10 V/m field will theoretically induce a voltage 0.33V to > 10V, depending on frequency > > (2) A 5000 uV/m field (10x the FCC class B limit above 960 MHz) will > theoretically induce a voltage from 152 uV to 5 mV, depending on > frequency. > > (3) A 500 uV/m field will theoretically induce a voltage from 15 uV > to 500 uV depending on frequency. > > These are first order approximations, but they are useful in > determining the level of the potential EMI threat. For instance a > 4-30 mA sensor circuit using high gain operational amps will most > likely be interfered with in scenario (1), there may be some > susceptibility detected in scenario (2), and most likely no problem > encountered with scenario (3). > > A sensitive all - band AM communications receiver will have problems > with all three, a broadcast TV operating in a strong signal area will > probably be OK with scenario 3 but not with 1 or 2. > > I guess what I'm really trying to say with all this is that EMC is a > systems thing, taking into account the nature of the culprit EMI > generator, the nature of the victim EMI receiver, and th
Re: EMC-related safety issues
Yes, John, you are quite right in both your comments as far as you go: 1) You are not the only person who can dramatise an issue so as to encourage people to debate it; 2) If you sold a single electronic safety-related circuit with a failure probability of 10^ -9 to 100,000 customers the cumulative failure probability is indeed 10^ -4. As you correctly said, Olber's Paradox does not apply in this area. But nevertheless this does not mean we need to make electronic circuits with failure rates equal to or better than 10^ -9. As you have said (and I agree) this would be a very difficult task indeed and likely to be very expensive, especially for any product using software. So how do we square this particular circle? Those members who are familiar with safety engineering techniques will be familiar with the idea of building very reliable systems up using a number of independent systems or devices each with lower reliability. These have various names, such as 'redundant channels' or 'duplicate channels' or 'safety back-ups' or 'fail-safe circuits' and many others. Some examples... I understand that car braking systems have (by law in Europe and North America at least) an independent hydraulic back-up system in case the primary system fails - because it is practically impossible to make the primary system reliable enough at a cost anyone would want to pay. The electronic flight-control systems in modern aircraft have two or three independent hardware 'channels'. Where software is involved they sometimes use three sets of independently-coded software each using architecturally-different operating systems and each running on an architecturally different hardware processor voting 2 out of 3 on every decision/output). I understand that the Space Shuttle launch control system uses 5 independent computers voting on each decision/output. The pressure relief valve on most pressure systems does not have a very high reliability, but when combined with the statistical probability of the system pressure going out of control the whole system is considered to be reliable enough. (Of course, pressure system designers must remember to site the pressure relief valve so that if it operates it doesn't cause a hazard of its own.) Three cheap and cheerful independent circuits, each achieving merely 10^ -3 reliability, can easily be combined together to create a system with 10^ -9 reliability – achieving very high levels of safety at low cost without any heartache in design or heart attacks from management. This is the way that high reliability is normally achieved at reasonable cost in practice (and has been achieved for many many years). IEC 61508 describes (or refers to) the necessary techniques. (PS: My statistical maths is rusty, so don't rely on the above simple calculation for any designs. Refer to IEC 61508 for more detail). Regards, Keith Armstrong In a message dated 05/01/02 21:01:18 GMT Standard Time, j...@jmwa.demon.co.uk writes: > Subj:Re: EMC-related safety issues > Date:05/01/02 21:01:18 GMT Standard Time > From:j...@jmwa.demon.co.uk (John Woodgate) > Sender:owner-emc-p...@majordomo.ieee.org > Reply-to: mailto:j...@jmwa.demon.co.uk";>j...@jmwa.demon.co.uk > (John Woodgate) > To:emc-p...@majordomo.ieee.org > > I read in !emc-pstc that cherryclo...@aol.com wrote (in > <43.47bb025.29689...@aol.com>) about 'EMC-related safety issues', on > Sat, 5 Jan 2002: > >The "one in a billion" John refers to sounds very dramatic and > difficult. > > More dramatic than you 'infant daughter' and '40 mph past a school'? > > I explained in VERY GREAT DETAIL the effect of cumulative probability in > requiring very low probability events to be taken into account. In > principle, as the probability goes down, the number of risk scenarios > increases *combinatorially*. There is no Olber's Paradox in this area, > the 'night sky is infinitely brighter than the Sun'! > > > >So it may be helpful to refer to IEC 61508 which is a > recently-published > >'basic safety publication' covering "The functional safety of > electrical / > >electronic / programmable safety-related systems" > > > >IEC 61508 uses the concept of the Safety Integrity Level (or SIL) to > help > >design safety-related systems which have quantified failure > probabilities. > > > >The SILs for average probability of failure to perform design > function on > >demand are: > >SIL level 1: up to 10^ -2 > >SIL level 2: 10^ -2 to 10^ -3 > >SIL level 3: 10^ -3 to 10^ -4 > >SIL level 4: 10^ -4 to 10^ -5 or even lower levels > > > >The SILs for average probability of dangerous failure per hour of > operation > >are: > >SIL level 1: up to 10^ -6 > >SIL level 2: 10^ -6 to 10^ -7 > >SIL level 3: 10^ -7 to 10^ -8 > >SIL level 4: 10^ -8 to 10^ -9 or even lower levels > > > >The standard describes how to select the SIL level
Re: EMC-related safety issues
Ken, replies below. Regards, Keith Armstrong In a message dated 06/01/02 06:56:46 GMT Standard Time, ken.ja...@emccompliance.com writes: > Subj:Re: EMC-related safety issues > Date:06/01/02 06:56:46 GMT Standard Time > From:ken.ja...@emccompliance.com (Ken Javor) > To:cherryclo...@aol.com, cortland.richm...@alcatel.com > CC:emc-p...@majordomo.ieee.org > > What an EMC engineer who understands the physics of field-to-wire coupling > would say is that the operation of non-antenna connected electronics > associated with one subsystem will not be degraded by close proximity with > the non-antenna connected electronics of another subsystem. Forget 10 > meters. Are the PCs in your office separated by 10 m? Would you expect > two PCs stacked side-by-side or one on top of the other to interact in any > manner? These are rhetorical questions. > I don't see the relevance of this paragraph to my example. This might be because I am a little slow. I would appreciate more explanation. > About the blood pressure monitor example. Not enough info here to back out > what is wrong, but basic logic theory says when the conclusion is > impossible, you must re-examine your assumptions. If 92 dBuV/m were enough > to make the device malfunction, it would malfunction a lot and there would > have been enough trouble reports to get it fixed or withdrawn. And 92 > dBuV/m at 10 meters is SCREAMING!!! I am on location and don't have FCC > regs easily available, but the limits stair-step around 40 dBuV/m at 3 m, > per my recollection. This is exactly why I emphasised that the front-panel display of the blood sample incubator (not a blood pressure monitor) continued to read the set-point temperature (37.1C) even though the actual temperature could be way off. I am confident that in actual operation the incubator temperature was quite often at least a few degrees C different from what was displayed on its front panel, probably affecting the performance of the reagents. And I don't think that 92dBuV/m is a high field strength to be emitted by a PC placed nearby, or for a non-compliant laptop at 10 metres. > Would I feel comfortable placing a CISPR compliant PC next to a medical > device qualified to 1 V/m? There is an inherent (not planned) margin of > safety here that is many orders of magnitude. The answer is absolutely > yes. If there were a problem, I would expect it more to occur below 30 > MHz, at the power supply switching frequency, IF the medical device > processed extremely low levels of electrical signals and was poorly > shielded. But I believe there are separate immunity requirements which > cover this eventuality as well. > Maybe when you wrote the above you weren't thinking of the previous correspondence in this thread about the proximity of the low-energy lamp to a bedside radio. Yes, I know, this concerned a radio receiver, what I mean to draw your attention to is the discussion about the intention and validity of the EMC standards – they simply do not cover situations where devices are placed close to each other – so they cannot be relied upon to provide compatibility in such situations. Military EMC standards are more thorough in this respect. And as I have already said, commercial EMC standards were not written with safety issues in mind, and most safety standards have not been written with EMC-related issues in mind (see my IEEE 2001 EMC Symposium paper and my longer article in ITEM UPDATE 2001 for details). So I cannot see that there is any 'inherent' margin of safety in the above situation, as you claim there "absolutely is". (And may I suggest that anyone who thinks that the statement in all (or most) IEC EMC immunity standards: "Products shall not become unsafe as a result of these tests." means that that products which pass those immunity tests are necessarily free from EMC-related safety problems, needs to think a little bit harder about the subject?) Regards again, Keith Armstrong > on 1/5/02 12:23 PM, cherryclo...@aol.com at cherryclo...@aol.com wrote: > > >> Dear Cortland >> People can't simply say: "ordinary semiconductors won't demodulate RF >> levels produced by an unintentional radiator" even the smallest >> amount of RF can be demodulated there are no hysteresis or threshold >> effects in a PN semiconductor junction or FET that is biased into its >> conduction region (at least not until you get below signal levels >> equivalent to less than a single electron). >> >> What I am sure most engineers would really mean to say is: >> "ordinary semiconductors exposed to RF levels from an information >> technology product which is fully compliant with all relevant EMC >> emissions standards and is at 10 metres distance will generally not >> demodulate a sufficient level of interference to make an appreciable >> difference to most electronic systems." >> >> Now we have a statement which has some scientific ri
Re: EMC-related safety issues
A) I don't agree with Ken that: "Emissions from a laptop are naturally (without suppression) on the order of 10 uV/m to 100s of uV/m." Maybe IBM PC clone laptops use similar enough architecture and chipsets and design techniques to be this consistent (I don't know) but I have seen the following emissions in portable computing products that would come under the same FCC/EMCD emissions standards as a laptop... A.1) A portable computing device with a large additional RAM array for use by people who had lost the power of speech. They would type the word they wanted to say and it would speak it for them using loudspeakers. This was a product that was sold in the UK without a CE mark at around the time the EMC directive came into force (1996). I only measured its with a 1cm diameter unshielded shorted-turn probe. I am used to seeing levels of around 60 to 100 dBmicrovolts from this probe when it is closer than one inch to typical laptop PCBs, with less than 20 dBmicrovolts at a distance of more than one foot. But with this product I saw over 50 dBmicrovolts at frequencies from 30 to 300MHz when my 1cm loop probe was 4 feet away! Maybe some less mathematically-challenged member of the group would like to work out what the electric field strength could have been measured at 10 metres. I've never seen anything quite as bad as that before or since, but it shows what poor EMC design can achieve in the hands of some designers of computing-related products. (Incidentally, they sold it with a clamp for fixing to wheelchairs - some models of which have already been mentioned in this thread as being known to run amok under certain conditions of RF field.) A.2) A portable computing device used in an automatic change machine on board transport was tested to be fully compliant with EN 55022 (approx = CISPR 22). I helped the manufacturer investigate complaints of interference and discovered that sub-fitted variant, which had not been tested for EMC compliance, left an HCMOS inverter IC with an unterminated inverter - which promptly decided to self-oscillate at 200MHz. (Many manufacturers of products with a number of build variants only test the fully-loaded one for EMCD compliance and assume the others are at least as good.) The very interesting thing about this example is that the power-ground structure of the PCB made a beautifully tuned antenna and resonant circuit at 200MHz, so although the inverter was hard-switching and did not run hot, the only emissions were at the 200MHz fundamental - no harmonics were emitted at all. Another very interesting thing is that some of the complainants had measured the equivalent radiated RF power from these devices as 2W. Who would have expected an unterminated HCMOS gate to be able to emit 2W at 200MHz? Well, now you know that it is possible (if not probable) it becomes something you should think about when designing safety-related systems. A.3) Laptops which are fully compliant with CISPR 22 or the equivalent US regs will have higher field strengths when they are less than 10 metres away. How many of us can guarantee that there is always at least 10 metres between each item of electronic equipment? Another thread to this discussion has discussed the problems of low-energy lamps interfering with bedside radios, and this shows up the difficulties of confusing compliance with standards intended for legal market entry and protection of the radio spectrum with the actual EMC engineering to prevent interference in real applications. Another problem is that when we are in the near field of a product its 10 metre emissions measurements are meaningless. There are components of near field emissions which fall off with the cube of the distance and are not usually detected at all by 10 metre tests. For example I understand that many products have very quite strong frequency magnetic fields nearby, at audio frequencies and lower, caused by variations in the loading on their DC power supplies. A.4) The standard commercial and industrial tests for emissions do not test the full range of possible emissions, since they were only ever intended to protect the radio communication and broadcasting spectrum. Actual electronic devices, especially certain kinds of transducers and their amplifiers, can be very sensitive to frequencies outside said spectrum. Military EMC standards recognise this real-life problem and typically test for emissions down to 100 or even 20Hz. B) I don't agree with Ken that "1000 uV/m would represent at least a 20 dB outage at frequencies that could possibly interfere with sensor electronics." See my other posting today adding to Tom Cokenias's calculations to show that RF field strengths of 500uV/m can easily cause severe errors (tens of degrees C) in some thermocouple temperature measuring systems. Also refer to my recent posting about the significant RF immunity problems experienced with a blood sample in
Re: EMC-related safety issues
Dear John In previous postings from Ken Javor and myself, I believe that Ken (who I was replying to in the fragment below) has made it clear that what he is really concerned with is "the kinds of emissions controlled by CISPR 22 and Title 47, part 15B of the US Code of Federal Regulations" (I hope I have got this right, Ken!). In earlier postings I believe that Ken complained that I was widening the definition to include such things as the unintentional emissions from welding apparatus (I didn't think of your electric fence example). I think John's comments emphasise a point I made earlier in this correspondence, which is that we need to be very careful in an international forum when using terms like "unintentional emitter". Such terms can have specific definitions in some EMC standards or Regulations in some countries, but they can also have a wider EMC engineering usage, such as that mentioned by John below. Confusion is possible unless we are more precise. Regards, Keith Armstrong In a message dated 05/01/02 21:01:28 GMT Standard Time, j...@jmwa.demon.co.uk writes: > >And no, I still don't agree with you that only radio receivers are > sensitive > >enough to RF to have a problem with what you are still calling > >'unintentional emissions' (even though this term means very little in > an > >international forum unless you define the relevant standards or laws). > > I think this term is quite legitimate and well-understood. If the > equipment requires to emit in order to perform its intended function, it > is an 'intentional emitter'. If it does not need to do so, but emits > anyway, it is an 'unintentional emitter'. It is difficult to see how > there could be any confusion or ambiguity about this.
Re: EMC-related safety issues
Dear John The incubator I described was already on the EU market in the latter half of the 1990s, when I helped to test and fix it. And I'm sorry to disappoint but I have already experienced several similar examples I could quote, such as the electric blanket that would change its heat settings randomly when a bedside light was switched on or off, or from other low-level mains transients. This is a potentially fatal issue for certain kinds of invalid, or people who are blind drunk (surely no person reading this would ever be in such a state) – and by the way, this is not me being emotive again, it was the expressed concern of the manufacturer and one of the reasons why they called me in. They sacked their Technical Director over this incident. They also didn't do a product recall despite having an estimated 100,000 products with the problem already out in the field. Of course, as a responsible engineer (and to cover my ass) I wrote them a letter recommending that they did a product recall (while thinking of the designers of the Challenger Space Shuttle's infamous O-ring seals). I find that many independent EMC people have dozens of similar examples, which they can't talk about very much because of commercial confidentiality. This is one reason why the EMC + Compliance Journal (www.compliance-club.com) started its 'Banana Skins' column - to help educate practising engineers about real EMC engineering problems they almost certainly weren't taught about at college and may not (yet) have experienced for themselves. I also have personal experience of a UK company that in the late 90's was selling a range of over 110 CE-marked products (such as incubators) intended for medical and chemical laboratories although less than 10% of their products met both the EMCD and the LVD. The company in question had just been purchased by another, which is why I was involved. Interestingly, the new owners continued to sell the non-compliant products while they re-engineered them one at a time to be compliant (which took several years). My simple investigations over a number of years into a number of companies' CE marked products have led me to be very cynical. As a rule of thumb I guess that around 30% of CE marked products are non-compliant with EMC or LVD, with another 30% being borderline cases. This seems to be borne out by recent enforcement surveys in Finland and in the UK and published articles from some test labs. Changing to another of your criticisms below... If you think my proposed statement is fog-filled, what do you propose instead? Lets have constructive criticism instead of merely criticism. In fact, in most scientific or engineering activities, one can only make public statements using foggy words like 'generally'. Remember the UK government's teams of scientific advisors and their pronouncements on BSE and the foot and mouth epidemic? Would you have expected them to produce precise and accurate predictions? I am of the opinion that the outbreak of foot and mouth disease in the UK was better understood, had fewer variables, and could be better controlled than many real-life EMC-related safety engineering problems. I believe the debate in question (whether "unintentional emitters" can interfere with electronic circuits which are not intentional radio receivers) cannot be answered with a definitive yes or no. I believe that each safety-related application needs to be investigated and firm engineering conclusions drawn. Even then, when one actually does such exercises in real life (and I have) one still finds statements concerning personal estimates of probability are necessary. You can deride these as being 'foggy' if you like but I don't think even you could be more precise in such circumstances. Absolute certainty just does not exist in the real engineering world of interactions between complex systems and I am sure you understand this well. As for the rest of your comments, I plead guilty to raising the emotional stakes. I deliberately used emotive arguments because I find that most designers (and test lab engineers) prefer to keep their heads down doing the engineering work they believe they are paid to do. Where people could be injured or killed by their products I generally find that designers are uncomfortable even thinking about this. Maybe this is because it would mean them fighting with their management to get more resources allocated. I also find that most designers (and their managers) - if they think about their potential 'victims' at all - also tend to think of them as 'other people'. They don't seem to think of their customers ands third parties as if they were members of their own family (as if other people's families were less important). So this is an emotive litmus test I often use to test designers' and managers ethics. Yes, ETHICS. Now that the word has been mentioned no doubt there will be a new thread begun, f
Re: EMC-related safety issues
Ken, can we take it that in the posting below you are agreeing that interference with non radio-receiving circuits from what you meant by "unintentional emitters" is a possibility, albeit a worst-case one? If my reading above is correct, how can you then go on to say that "...unintentional emissions from ITE can only upset a radio receiver tuned to the emission frequency." It seems to me that the words "can only" in the your quote above should be replaced by "are most likely to", which I would agree with. Regards, Keith Armstrong In a message dated 06/01/02 06:56:07 GMT Standard Time, ken.ja...@emccompliance.com writes: > Subj:Re: EMC-related safety issues > Date:06/01/02 06:56:07 GMT Standard Time > From:ken.ja...@emccompliance.com (Ken Javor) > To:t...@tncokenias.org (Tom Cokenias), cherryclo...@aol.com, > emc-p...@majordomo.ieee.org > > The analytical portion of this post is, as the author stated, worst case. A > cable attached to a susceptible circuit picks up a common-mode potential, > which most likely drives a current on a shield if the the circuit is > sensitive. Then only the current multiplied by shield transfer impedance > actually gets into the victim, assuming no CMR. Which just makes my > original point - unintentional emissions from ITE can only upset a radio > receiver tuned to the emission frequency. That is why, as another > contributor posted, we use EMI receivers and spectrum analyzers with preamps > to make OATS measurements. > > on 1/4/02 12:51 PM, Tom Cokenias at t...@tncokenias.org wrote: > > > At 8:34 AM -0500 1/4/2002, Keith Armstrong wrote: > > > >> Does anyone else think that ordinary semiconductors doesn't respond to > RF? > > > > > > > > I agree that commonly used semiconductors have responses well into > > the 100's of MHz. > > > > How much of a problem this is will depend on the nature and function > > of the circuitry using these components. > > > > The EUT wires, cables, pcb traces etc. act like antennae, on which > > the incident field voltages and currents. An antenna factor can be > > thought of as ratio of the field strength to the voltage induced on > > the terminated cable connected to the antenna. > > > > In an impedance matched system, > > > > > > AF=9.734/lamda*(G)^0.5, lamda being wavelength in meters, G being > > antenna gain over isotropic, > > > > or in dB > > > > AF dB = - G dBi -29.7 dB + 20logFMHz > > > > Assuming G is 1 (isotropic antenna), AF is 1 (= 0 dB) at about 30.8 > > MHz, and AF get larger as frequency increases, to a factor of 32.7 > > (= 30.3 dB) at 1 GHz . Since AF is field strength divided by > > induced voltage, the voltage induced on the trace goes down as > > frequency goes up for the same incident field strength. > > > > An effective receive antenna needs to be on the order of 1/2 > > wavelength or so; for 30 MHz this is 15m, for 1000 MHz this is 15 cm. > > > > So if a victim EUT circuit has a pretty effective receive antenna, > > and does not have any filtering and is equally sensitive across the > > frequency range under consideration (all taken together, a worst case > > scenario for susceptibility), > > > > (1) A 10 V/m field will theoretically induce a voltage 0.33V to > > 10V, depending on frequency > > > > (2) A 5000 uV/m field (10x the FCC class B limit above 960 MHz) will > > theoretically induce a voltage from 152 uV to 5 mV, depending on > > frequency. > > > > (3) A 500 uV/m field will theoretically induce a voltage from 15 uV > > to 500 uV depending on frequency. > > > > These are first order approximations, but they are useful in > > determining the level of the potential EMI threat. For instance a > > 4-30 mA sensor circuit using high gain operational amps will most > > likely be interfered with in scenario (1), there may be some > > susceptibility detected in scenario (2), and most likely no problem > > encountered with scenario (3). > > > > A sensitive all - band AM communications receiver will have problems > > with all three, a broadcast TV operating in a strong signal area will > > probably be OK with scenario 3 but not with 1 or 2. > > > > I guess what I'm really trying to say with all this is that EMC is a > > systems thing, taking into account the nature of the culprit EMI > > generator, the nature of the victim EMI receiver, and the path > > between them. Then we have the economics of operating different > > devices in the same vicinity, the politics of who gets how much of > > what kind of protection, etc., etc. All things considered, we should > > have jobs for life! > > > > best regards and a Happy New Year to all. > > > > Tom Cokenias > > > > T.N. Cokenias Consulting > > P.O. Box 1086 > > El Granada CA 94018 > > > > tel 650 726 1263 > > cell 650 302 0887 > > fax 650 726 1252 >
Re: EMC-related safety issues
Thanks for the correction. Medicine is not something I know much about.
The point remains that a huge benefit was derived from these vaccines which
were discontinued due to a very small fraction of bad reactions. Forum
members: when it turned out that a small number of people were killed or
hurt by airbags, there was no general hue and cry to remove them or sue the
automobile manufacturers. Why?
Diphtheria and whooping cough are not mere viruses. They are life
threatening illnesses that in former centuries contributed greatly to infant
morality. And tetanus is potentially deadly at any age.
on 1/4/02 11:31 PM, Jim Freeman at free...@chelsio.com wrote:
Hi Ken,
The reason that those companies stopped was because it was found that
there was mercury in the formulation of the vaccine. The mercury had no
other use other than stabilization. The mercury is known to cause brain
damage. Prior to around 1980, DPT was not given to infants. The rise in
autism has correlated with the increased use of the infant vaccines.Those
companies were also found to have poor process control that allowed too much
of a live virus in their vaccines causing a so-called 'hot batch'. The
company that is left doesn't have mercury in their formulation and has
superior process control. I would much rather see my child suffer through a
virus than be permanently brain damaged(usually undetectably)
BTW, whooping cough and pertusis are the same thing. the D stands for
Diptheria.
Jim Freeman
Ken Javor wrote:
My take on it is that rather than appease ridiculous demands, a company
ought to look at the profit vs. risk vs. cost to consumer and decide, heck,
it ain't worth it. Case in point on the news today I heard that DPT shots
are in short supply, because two companies quit making it. They quit making
it because there were a very small number of bad reactions to it and there
were lawsuits or gov't action. Well, my kids are beyond that stage but I
sure feel sorry for the people out there whose infants are at risk for
whooping cough, diphtheria and pertussis. The only thing worse than
watching your child become seriously ill is knowing it was easily
preventable.
S on 1/4/02 7:37 AM, cherryclo...@aol.com at cherryclo...@aol.com wrote:
Hey, Ken, let's try to be realistic here!
Sure - we should try to get laws we don't like changed, but that isn't going
to happen overnight and in the meantime we have to operate within the law as
it stands.
Or are you suggesting immediate insurrection by product manufacturers?
(Outlaw manufacturers roaming the wild wild west - an interesting concept!)
The IEE's guide on EMC and Functional Safety is concerned with such legal
aspects, but is also concerned with saving lives in a world where electronic
control of safety-related functions is proliferating madly.
As my paper at the IEEE's EMC Symposium in Montreal and my recent article in
ITEM UPDATE 2001 show - at present EMC standards don't address safety
issues, and most safety standards don't address EMC-related functional
safety issues.
Regards, Keith Armstrong
In a message dated 03/01/02 17:24:42 GMT Standard Time,
ken.ja...@emccompliance.com writes:
Subj:Re: EMC-related safety issues
List-Post: emc-pstc@listserv.ieee.org
Date:03/01/02 17:24:42 GMT Standard Time
From:ken.ja...@emccompliance.com (Ken Javor)
Sender:owner-emc-p...@majordomo.ieee.org
Reply-to: ken.ja...@emccompliance.com (Ken Javor)
To:c...@dolby.co.uk (James, Chris), acar...@uk.xyratex.com
('acar...@uk.xyratex.com'), emc-p...@majordomo.ieee.org
There is an inherent contradiction in this anti-profit, anti-technology
point-of-view that I cannot and will not defend. All I am saying is that
people who feel this is wrong should stand up and say so, not write guides
for how to go along with it.
Re: EMC-related safety issues
What an EMC engineer who understands the physics of field-to-wire coupling would say is that the operation of non-antenna connected electronics associated with one subsystem will not be degraded by close proximity with the non-antenna connected electronics of another subsystem. Forget 10 meters. Are the PCs in your office separated by 10 m? Would you expect two PCs stacked side-by-side or one on top of the other to interact in any manner? These are rhetorical questions. About the blood pressure monitor example. Not enough info here to back out what is wrong, but basic logic theory says when the conclusion is impossible, you must re-examine your assumptions. If 92 dBuV/m were enough to make the device malfunction, it would malfunction a lot and there would have been enough trouble reports to get it fixed or withdrawn. And 92 dBuV/m at 10 meters is SCREAMING!!! I am on location and don't have FCC regs easily available, but the limits stair-step around 40 dBuV/m at 3 m, per my recollection. Would I feel comfortable placing a CISPR compliant PC next to a medical device qualified to 1 V/m? There is an inherent (not planned) margin of safety here that is many orders of magnitude. The answer is absolutely yes. If there were a problem, I would expect it more to occur below 30 MHz, at the power supply switching frequency, IF the medical device processed extremely low levels of electrical signals and was poorly shielded. But I believe there are separate immunity requirements which cover this eventuality as well. on 1/5/02 12:23 PM, cherryclo...@aol.com at cherryclo...@aol.com wrote: Dear Cortland People can't simply say: "ordinary semiconductors won't demodulate RF levels produced by an unintentional radiator" even the smallest amount of RF can be demodulated there are no hysteresis or threshold effects in a PN semiconductor junction or FET that is biased into its conduction region (at least not until you get below signal levels equivalent to less than a single electron). What I am sure most engineers would really mean to say is: "ordinary semiconductors exposed to RF levels from an information technology product which is fully compliant with all relevant EMC emissions standards and is at 10 metres distance will generally not demodulate a sufficient level of interference to make an appreciable difference to most electronic systems." Now we have a statement which has some scientific rigor and some engineering validity to it. (Although I do worry that in Europe our harmonised EMC standards only test emissions up to 1GHz, so what does that say about the possible emitted fields strengths from a PC with a 1.2GHz clock frequency?) Let's see if we can put some meat into this discussion with a real-life example... I once tested a blood sample incubator for RF field immunity. The incubator was used during screening programs (for cancer and other diseases) and kept about 100 test tubes at 37.1C (normal blood temperature), while the reagents in the test tubes changed colour. After 24 hours of incubation medical staff would inspect the test tubes and write letters to people telling them they were sick, or that they were clear of the disease. I don't know what temperature tolerance the reagents had to give an accurate medical diagnosis, so assume ±0.1C. On the front panel of the incubator was a display of its temperature, which was of course 37.1C. We found that field strengths as low as 1V/m would cause the incubation temperature to range over full scale, from heaters fully off (in which case the temperature would decline to ambient) to maximum (in which case the water used to incubate the test tubes would boil). We could use the RF test frequency to control the temperature between plus and minus full scale over the frequency range 80 to 1000MHz at 1V/m (and did not test beyond 1GHz). Most worryingly, the front panel display would only show temporary variations from its 37.1C when the RF field was turned off or on, and would continue to show 37.1C even when the water in the incubator was stone cold or actually boiling. Most demodulation effects in bipolar and FET devices approximate to a square law - for example a 1dB fall in the field strength (keeping everything else constant) would typically result in a 2dB fall in the demodulated 'interference' error signal, as John Woodgate has recently pointed out. If we assume that the 1V/m field strength was causing a 60C temperature error, how low would we need to make the RF field to get down to the 0.1C accuracy of the front panel display? Assuming square-law characteristics for the device doing the demodulation I calculate a field strength of around 40mV/m or 92dBmicrovolts/metre. You will notice that I have been generous to the incubator and assumed that the 1V/m field just about caused its temperature error to increase by 60C to boil the water, whereas it could have been overdriving the internal circuits by a considerable margin and still suf
Re: EMC-related safety issues
The analytical portion of this post is, as the author stated, worst case. A cable attached to a susceptible circuit picks up a common-mode potential, which most likely drives a current on a shield if the the circuit is sensitive. Then only the current multiplied by shield transfer impedance actually gets into the victim, assuming no CMR. Which just makes my original point - unintentional emissions from ITE can only upset a radio receiver tuned to the emission frequency. That is why, as another contributor posted, we use EMI receivers and spectrum analyzers with preamps to make OATS measurements. on 1/4/02 12:51 PM, Tom Cokenias at t...@tncokenias.org wrote: > At 8:34 AM -0500 1/4/2002, Keith Armstrong wrote: > >> Does anyone else think that ordinary semiconductors doesn't respond to RF? > > > > I agree that commonly used semiconductors have responses well into > the 100's of MHz. > > How much of a problem this is will depend on the nature and function > of the circuitry using these components. > > The EUT wires, cables, pcb traces etc. act like antennae, on which > the incident field voltages and currents. An antenna factor can be > thought of as ratio of the field strength to the voltage induced on > the terminated cable connected to the antenna. > > In an impedance matched system, > > > AF=9.734/lamda*(G)^0.5, lamda being wavelength in meters, G being > antenna gain over isotropic, > > or in dB > > AF dB = - G dBi -29.7 dB + 20logFMHz > > Assuming G is 1 (isotropic antenna), AF is 1 (= 0 dB) at about 30.8 > MHz, and AF get larger as frequency increases, to a factor of 32.7 > (= 30.3 dB) at 1 GHz . Since AF is field strength divided by > induced voltage, the voltage induced on the trace goes down as > frequency goes up for the same incident field strength. > > An effective receive antenna needs to be on the order of 1/2 > wavelength or so; for 30 MHz this is 15m, for 1000 MHz this is 15 cm. > > So if a victim EUT circuit has a pretty effective receive antenna, > and does not have any filtering and is equally sensitive across the > frequency range under consideration (all taken together, a worst case > scenario for susceptibility), > > (1) A 10 V/m field will theoretically induce a voltage 0.33V to > 10V, depending on frequency > > (2) A 5000 uV/m field (10x the FCC class B limit above 960 MHz) will > theoretically induce a voltage from 152 uV to 5 mV, depending on > frequency. > > (3) A 500 uV/m field will theoretically induce a voltage from 15 uV > to 500 uV depending on frequency. > > These are first order approximations, but they are useful in > determining the level of the potential EMI threat. For instance a > 4-30 mA sensor circuit using high gain operational amps will most > likely be interfered with in scenario (1), there may be some > susceptibility detected in scenario (2), and most likely no problem > encountered with scenario (3). > > A sensitive all - band AM communications receiver will have problems > with all three, a broadcast TV operating in a strong signal area will > probably be OK with scenario 3 but not with 1 or 2. > > I guess what I'm really trying to say with all this is that EMC is a > systems thing, taking into account the nature of the culprit EMI > generator, the nature of the victim EMI receiver, and the path > between them. Then we have the economics of operating different > devices in the same vicinity, the politics of who gets how much of > what kind of protection, etc., etc. All things considered, we should > have jobs for life! > > best regards and a Happy New Year to all. > > Tom Cokenias > > T.N. Cokenias Consulting > P.O. Box 1086 > El Granada CA 94018 > > tel 650 726 1263 > cell 650 302 0887 > fax 650 726 1252 > > > > > > > > --- 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: Michael Garretson:pstc_ad...@garretson.org Dave Healddavehe...@mediaone.net For policy questions, send mail to: Richard Nute: ri...@ieee.org Jim Bacher: j.bac...@ieee.org All emc-pstc postings are archived and searchable on the web at: No longer online until our new server is brought online and the old messages are imported into the new server.
Re: EMC-related safety issues
My point was that only radios are sensitive to rf fields at the levels controlled by FCC/CISPR22 and indeed, as Ing. Gremen pointed out, levels well above the limits. Which means that the only rationale behind FCC/CISPR22 is protection of radio broadcast reception. Period. on 1/5/02 12:10 PM, cherryclo...@aol.com at cherryclo...@aol.com wrote: Dear Ken I am truly sorry if I irritated you by misunderstanding your words, but I took your posting to imply that electronic circuits which are not designed as RF receivers would not respond very well to radio frequencies. My example was not intended to be a full answer to your example (there are other postings which are dealing with that) just to indicate that the frequency response of slow and commonplace ICs can be very high indeed. I am sensitive to this issue because I keep on running across electronics designers who say things like: "I don't need to worry about the RF immunity of my audio amplifier/motor controller/temperature/pressure/flow/weight/velocity measurement and control system (please delete where applicable) because the opamps I use have a GBW of under 1MHz so they won't see the RF" which is of course complete bollocks (a UK phrase that I hope translates well enough for all emc-pstc subscribers). And no, I still don't agree with you that only radio receivers are sensitive enough to RF to have a problem with what you are still calling 'unintentional emissions' (even though this term means very little in an international forum unless you define the relevant standards or laws). I think the problem you are concerned with is application dependant and we cannot make such broad assumptions. As I said earlier, most interference problems are caused by radio transmitters or radio receivers, but not all. Regards, Keith Armstrong In a message dated 05/01/02 01:20:27 GMT Standard Time, ken.ja...@emccompliance.com writes: Subj:Re: EMC-related safety issues List-Post: emc-pstc@listserv.ieee.org Date:05/01/02 01:20:27 GMT Standard Time From:ken.ja...@emccompliance.com (Ken Javor) To:cherryclo...@aol.com, emc-p...@majordomo.ieee.org One sure way to REALLY irritate me is to twist my words and try to make me look stupid (I do a fine job by myself on occasion and don't appreciate any outside help). I did not say that pn junctions don't detect and rectify rf, I said that the field intensities associated with unintentional emissions from ITE are too low to cause susceptibility in circuits other than radios. Your example here is 10 V/m, and you are talking about an op-amp (gain unspecified) and that it was susceptible at that level should be no surprise to anyone. on 1/4/02 7:34 AM, cherryclo...@aol.com at cherryclo...@aol.com wrote: Does anyone else think that ordinary semiconductors doesn't respond to RF? I have tested a product which was little more than an LM324 quad op-amp for RF immunity using IEC 61000-4-3. This op-amp has a slew rate of 1V/micro-second on a good day with the wind in its favour. It was housed in an unshielded plastic enclosure. Demodulated noise that exceeded the (not very tough) product specification were seen all the way up to 500MHz at a number of spot frequencies that appeared to be due to the natural resonances of the input and output cables. Above 500MHz this resonant behaviour vanished to be replaced by a steadily rising level of demodulated 1kHz tone as the frequency increased. I stopped testing at 1GHz, where the output error from the product was about 10% and still rising with increased frequency. OK, the field strength for the test was 10V/m (unmodulated) but the real surprise was how well this very cheap and very slow opamp demodulated the RF, and that it demodulated better at 1GHz than at 500MHz. I have done many many immunity tests using IEC 61000-4-3 on audio equipment and found much the same effects with every product I've ever tested. With most larger products there is usually a roll-off in the demodulation above 500MHz - not because the semiconductors in the ICs can't respond (they can) but apparently because larger products have higher losses above 500MHz or so between the cable ports and the semiconductors, plus a denser structure that might provide more self-screening. The transistors and diodes in all modern ICs (analog or digital) are so tiny that they make excellent detectors at UHF and beyond. As they get smaller (and they are) their frequency response increases (and their vulnerability to upset and damage decreases). Regards, Keith Armstrong
