Measurement uncertainty and cable balance - with implications for the CDNE-M and CMAD

The emulation of the real-world of mains power cables in EMC test methods can be improved. We show that for devices with significant differential-mode emission in the frequency range 150 kHz to 30 MHz, the impedance balance of the mains cable should be specified, even when the actual measurement is of data cable emission. For the same reasons radiated emission testing between 30 MHz and 200 MHz may be seriously in error. Neither the proposed CDNE-M (Coupler Decoupler Network for Emission - Mains) or the CMAD (Common-Mode Attenuation Device) captures the real-world interference due to DM because they do not model the DM to CM conversion and reflection at points remote from the EUT. It is recommended that the development of Standards should assume that one-tenth of mains-cable DM is converted to CM. An appropriate small modification to the CDNE-M design is presented.