By J. M. Woodgate B.Sc.(Eng.) C.Eng. MIET SMIEEE FAES Hon FInstSCE MIOA
A number of significant topics have generated documents during the last few months at the highest level in IEC EMC, i.e. in the CISPR committee itself. This committee is the parent of all the CISPR subcommittees and in addition it has a Steering Committee, CISPR-S.
Increasing number of devices
The International Amateur Radio Union (IARU) asked whether the increasing number of electronic devices per household might call into question the limits for individual products. CISPR/S investigated and found that the subject is not addressed, even implicitly, in current CISPR standards [but it is in some SC77A standards]. It commented that the number of devices in use fluctuates and can be very difficult to estimate. This probably applies particularly to portable computers and the like and mobile phones.
The initial conclusion was the traditional 'More research is necessary', and a call was made for National Committee comments. 5 MB of comments were received. Some National Committees submitted extensive comments, ranging from 'We know of no problems', through a proposal to reduce all limits by 10 dB to 20 dB, to a requirement for filtration at the mains input for every product that rectifies the mains supply, irrespective of rated power – at present some very low-power products meet the emission limits without a filter. This is because the rectifier produces intermodulation products of all the HF emission components present on the mains supply, thus adding to the spectrum of conducted emissions, most of which can be radiated.
Some technical contributions were based on simulations, not actual measurements. One set of actual measurements, comparing the conducted emissions (assumed to correlate well with radiated emissions below 30 MHz) from five identical products to those of one product, showed the 14 dB increase that is expected from highly-correlated sources. But a similar test with five far 'noisier' products showed an increase of 20 dB at low frequencies, which is difficult to explain. A direct measurement of radiated emissions showed a more explicable difference of 18 dB at 30 MHz falling to about 5 dB at 150 MHz. A third type of product showed a trend-line difference between 5 dB and 8 dB between 30 MHz and 150 MHz, comparable with the 7 dB expected for five wholly uncorrelated sources, but with some peaks up to 20 dB, perhaps indicating strong correlation in certain frequency bands but there could be other explanations.
Additional comments included a request that no requirements for collections of devices be considered, a statement that better market surveillance and enforcement would produce more useful results. Some National Committees and the IARU already submitted technical reports, although no information appears to have been collected about the trend in complaints of interference, which is the only justifiable criterion for setting limits.
Radio-enabled products in CISPR standards
CISPR 32 is the only current CISPR standard that addresses radio-enabled products (REP), i.e. those that can receive, and possibly transmit, information-bearing radio waves. But other CISPR standards do not explicitly exclude such products, so a clarification is necessary. In principle, this would involve amending the Scope clauses and change test conditions to allow testing with and without the radio function in operation, excluding testing the radio functions (which are the responsibility of ITU and/or ETSI).
Thirteen National Committee responses with supporting comments were received. Some go into details of what changes would be required, which is a bit premature. Changing the Scope clauses is a purely editorial task, and the changes to test procedures could also be editorial in nature, if 'generic' statements are sufficient, such as 'Emission tests with the radio function operating shall include filters applied to measuring equipment that sufficiently exclude signals from the radio part so as not to affect the results'.
These responses prompted a statement from the Chair of CISPR, clarifying that CISPR will not set requirements for the radio function, but will require testing with the radio function in operation. All CISPR subcommittees will study the implementation of these changes in their standards.
Surprisingly, 'protection distance' has never been officially defined, but the variable r in CISPR 16-4-4 (not familiar to too many people) is described as 'minimum operational distance', but is also what is commonly meant by 'protection distance', i.e. the minimum distance between emitter and victim at which the emission and immunity requirements result in satisfactory operation of the victim. Text in CISPR 23 differs, and is not appropriate (e.g. the victim may not have an antenna).
For intentional receivers, the maximum permitted interference field strength can be calculated, as a first step, from the minimum field strength of the wanted signal minus the protection ratio, and both of these figures can be obtained from ITU documents. Some correction factors may be necessary, as described in CISPR 16-4-4. In all cases, a protection distance has to be assumed in order for an emission limit to be set.
In early editions of CISPR standards, some measurement distances (not necessarily equated to protection distances) were as large as 100 m, even for what we would now identify as 'Class B limits'. However, these standards are silent on the assumed value of protection distance, and in any case the assumed distance may not be valid now. There is a proposal for a definition of 'protection distance' but it is not applicable to victims that are not intentional receivers, so it may not (should not) be accepted. The overall conclusion is that a new CISPR publication is needed, documenting the values of the variables (described as 'parameters') used for the calculation of each limit.
National Committee comments were uniformly supportive but included numerous detailed amendments and cautions concerning the original document. We will probably not see the proposed new CISPR publication in the short term.
Maritime LED luminaires
ITU-R has complained about interference with navigation and communication equipment from maritime LED luminaires, even if compliant with IEC 60945, produced by IEC TC80. Other standards for emissions from lighting equipment are CISPR 15 and CISPR 25. The Scope of CISPR 15 states exclusions but does not include equipment whose emissions limits are set in other IEC (as opposed to CISPR) standards. This is likely an accident of history; that wording was probably adopted before TC80 produced IEC 60945. A simple amendment will fix that. CISPR/F works closely with IEC TC34 but may well have had less contact with TC80 in the past. CISPR 25 applies to small craft, and may thus include lighting products for them. A review of IEC 60945 and CISPR 25 is proposed, to check that the provisions are adequate and, where necessary, consistent.
The circulation of some of the CISPR documents mentioned above has created sufficient anxiety in some quarters that there is an intention to tighten limits 'across the board'. This has prompted the Chair of CISPR to issue a reassurance.
The investigations are necessary because of demands from industry for the justification of some limits and other provisions. In very many cases, data vital for support has been lost or was never archived, so a retrospective investigation, based on probabilities is often the best that can be done. In some cases, theoretical analysis and/or simulation is robust enough to act as justification. In reality, a particular combination of emission and immunity requirements is justified by an acceptably low level of complaints of interference (even though only a small percentage of interference incidents result in a complaint to the spectrum management authorities). But that does not address the 'balance' between emission and immunity; for example, 'Why shouldn't 6 dB more emission be allowed, since we can show that 6 dB more immunity is easy to achieve?'
A problem with relying on complaints is that the damage has already been done by the time the complaints come to notice; it then takes several years to produce and apply whatever change to the standards is necessary to address the issue. On the other hand, making requirements more stringent in anticipation of future problems is not only economically problematical but intensely controversial. The CISPR investigations concentrate on data collection; both retrieval of information from past decades and appraisal of relevant current research and other reliable publications.