Hello Pieter and group. You are right, the measurement has to be done over a full maximum length of 5 meters, dependent of the frequency to be monitored. The Luthi clamp has to be moved along the cable with small increments (f.a. 5 cm ::2") to find resonance. This almost always takes place on a distance not far away from 1/4 lambda. In addition to the problems with current transformers in general, this resonance point moves toward the EUT, and at higher frequencies, this distance becomes problematic physically.
The usage of current clamps is very not-recommended at higher frequncies then 300 Mhz and at even lower frequncies when diameter of the cable (current transformer)goes above 30 mm. (1.2 ") I addition to all all of this, i need to warn you about the high probability of AC and DC currents that flow through any current clamp and saturating the ferrite. One has to be sure that a cable does not feed any common mode currrent of the load, i.e. no measurement can be made on single wires carrying AC or DC at values above a few amps. (and even at less then 1 amp some clapms are saturating or at least show deviations in measurement value ( remember the BH curve and is hysteresis about magnetic field in ferrites) Good ferrites saturate at low H-values. The elimination of reflections that i mentioned (see below) should be interpreted as that the remaining part of the cable extending behind the clamp is of little influence to the measurements, regardless of it's length and signals induced on it , in normal circumstances. Hope this adds to the discussion. Regards, Gert Gremmen Ing. == Ce-test, Qualified testing == Consultants in EMC, Electrical safety and Telecommunication Compliance tests for European standards and ce-marking Member of NEC/IEC voting committee for EMC. Our Web presence: http://www.cetest.nl List of current harmonized standards http://www.cetest.nl/emc-harm.htm 15 great tips for the EMC-designer http://www.cetest.nl/features01.htm -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Pieter Robben Sent: maandag 15 februari 1999 1:01 To: [email protected] Subject: RE: Pre-Compliance testing for radiated emissions using a current probe Hello Gert, Although the ferrite stabilizes the impedance of the cable, which enables somebody to perform reproducible measurements, I do not agree that this would eliminate the effects of cable length and reflections on it. One would still have to make sure that the maximum level of emission from the cable is measured and therefore the cable to be measured must be extended to at least 5 metres (0.5 times the wavelength @ 30 MHz) in order to be able to perform correct measurements of emission levels in the lower part of the frequency range 30-300 MHz (as stated in the CISPR14/EN55014). As I recall the Luthi clamp is specified for use in the frequency range 30-1000 MHz. Although I must admit that performing this type of measurement above 300 MHz is difficult and almost impossible to reproduce, it is also so that this is mostly being caused by the variance in capacitive coupling of the cable with the test setup and its surroundings. Best regards, Pieter Robben At 99/02/14 10:23 AM, you wrote: >The Power Clamp method measures the current flowing on a wire into a fixed >common mode impedance formed by the added ferrites following the current >clamp. >They sum up to form a +/- 150-200 Ohms impedance, independent of the cable >behind it. This eliminates the effects of cable length , reflections on it >and makes a good average reproducible measurement set up. Now this measures >ONLY the EMI effects from the attad cables. > >Of course the EMI from the enclosure does not pass the clamp and is not >taken into account here. Therefore any test engineer should only apply this >test when it is sensible to conclude that the EUT does not radiate except by >it's cables. This is the case when: > >- the wavelength of the measured impedance is much greater than the >enclosures maximum size >- the max size of the enclosure is so small that it is much smaller then the >highest frequencies wave length (most often 300 Mhz :: 1 meter) >- much smaller is at least 1/20th >- The clamp method is useless above 300 Mhz as the included current >transformer CANNOT be constructed in reasonable sizes for frequencies above >that. >- the larger the clamp; the lower the usable frequency. > >If one uses a current clamp only, the remaining cable does influence the >test values. One should vary their lengths or add additional ferrite cores >behind the clamp for common mode impedance stabilization. > >The limit values (40/44 dBpW) most often used with a ferrite clamp >correspond to the 30/37 dBuV/m CISP22 etc. limits, when this power is >thought to be radiated by a resonant dipole antenna. This can be proved very >easily. > >So if used for pré compliance measurements, limit them to 300 Mhz, use the >smallest clamp (or the official luthi clamp) available, add ferrites if >necessary, and make sure the enclosure is much smaller then wavelength. >However, if your clamp measurements show that no power is emitted at all, >you might conclude that no power is radiated either from the enclosure too. > > > >Regards, > >Gert Gremmen Ing. > >== Ce-test, Qualified testing == >Consultants in EMC, Electrical safety and Telecommunication >Compliance tests for European standards and ce-marking >Member of NEC/IEC voting committee for EMC. >Our Web presence: http://www.cetest.nl >List of current harmonized standards http://www.cetest.nl/emc-harm.htm >15 great tips for the EMC-designer http://www.cetest.nl/features01.htm
