Introduction to filters for EMC test chambers


The problem with connecting cables

        Emission from - and pickup onto - the associated cables is a significant factor in the radiated EMC of electronic products.  Accordingly, the standards require that the cables should be positioned to achieve a worst-case layout.  However, this is an impossible requirement when there is more than one cable since the radiation (or pickup) onto one cable may cancel that of the other at some frequencies and add to it at others.  Put another way, two cables will act as a two-element antenna,  but the frequency and phase response of each element is likely to be quite different so the worst-case layout for one frequency and direction will not be the worst-case for another.

Furthermore, real-life installations of electronic equipment have “untidy” cable connections to the outside world.  These are characterised by random - but not usually very tight - coupling to other cables and to conductive and dielectric structural elements.  However, these cables never pass through a definite boundary surface as must be the case for product EMC testing.

        Standard test methods are either silent or unsatisfactory about  how cables should  leave the test chamber.  For example, CISPR22 makes no statement about data or control cables.  It does suggest that the mains cable exit via a junction box or artificial mains network bonded to the floor. Such an arrangement provides a near-zero common-mode impedance at this point - hereinafter called the “exit impedance”.  This is much lower than is likely to be encountered in practical situations, and has different effects according to the cable length expressed in terms of wavelength.  At frequencies below that at which the combination of EUT and cable is quarter-wave resonant, the low exit impedance increases the interference current on the cable, and so results in systematic over-estimation of emission or underestimation of immunity.  There will be several higher frequencies at which the cable will exhibit resonance which will be accentuated by reflection at the mismatch due to zero exit impedance.  Such resonance will be critically dependent on cable length and layout.  It may mask or accentuate critical emission or immunity problems; it certainly contributes significantly to the measurement uncertainty.



The switched cable decoupler

        One solution is to leave the cables in fixed positions,  and radically disturb the radiation pattern of each cable in turn  by adding series impedance in common-mode.  The effect of this impedance may be to reduce radiation from the cable concerned and so avoid cancellation effects.  Alternatively, it may change the feed impedance of the cable “antenna” so that its resonant frequency is drastically altered producing the effect of a substantial change of cable layout.

        RML’s series 16Dxxswitched cable decoupler head comprises a ferrite sleeve whose impedance may be switched in and out of circuit via a fibre-optic link that does not disturb the test environment.   A switch on the associated controller allows the worst case to be selected quickly and without having to enter the test chamber.  Alternatively the "worst case" of cable emission or pickup may be determined as testing progresses by pulsing the SCD on and off, maintaining each state for a sufficiently long period to allow the quasi-peak detector of the measuring receiver to settle to its final value.

        The paper “Switched Cable Decoupler Measurements”, presented at the EMC2000 conference at York on 10th & 11th July by authors from the NPL and from Richard Marshall Limited, gives results obtained in both these modes of operation.



Exit filters

        Richard Marshall Limited has developed a range of “Exit Filters” that present the standard 150 ohm common-mode impedance to cables leaving the test chamber.  This impedance damps the resonance effects discussed above so as to maximise test realism and repeatability. These filters are all electrically non-intrusive to the cable and so avoid any risk of disturbing EUT function.  The “Duct” versions clamp around the cables, allowing any kind of cable to leave the test chamber with any desired length remaining inside.  This avoids the need to “bundle” cables in the test volume and further enhances test consistency.

        Critical to this development has been the provision of substantial shunt capacitance from the cable  to ground,  and carefully-optimised magnetically-coupled series elements. The arrangements used are the subject of patent applications.

Technical background is available in the paper  “Chamber Exit Filters for EMC testing” by R C Marshall,  presented at the EMC2000 Conference  held at York on 10th/11th July 2000.