Comparison of Measurement Methods for the Assessment of RMS Spectra in the 9-150 kHz Range

Conducted emissions in the 9–150 kHz range are gaining presence in the low-voltage (LV) distribution grid due to the devices implementing inverters and rectifiers, such as, photovoltaic inverters and electric vehicle charging stations (EVCS). The energy of the conducted emissions is related to the thermal stress that may reduce the lifetime of household appliances. Currently, several measurement methods are available to compute the root-mean-square (RMS) spectra of conducted emissions in the LV grid. Nonetheless, there is no information indicating which method provides the most accurate RMS results in the frequency domain. This study analyses the accuracy of the RMS spectra provided by the standardized IEC 61000-4-7 and CISPR 16 methods, and the research RM-A method. The accuracy analysis has been performed processing two LV grid-like synthetic signals developed by the authors, whose power spectral density (PSD) is known in the entire 9–150 kHz range. These synthetic signals replicate two recordings taken in the LV grid, which contain narrowband emissions, power line communication frames and impulsive emissions. In order to assess the accuracy, the RMS spectra obtained with the IEC 61000-4-7 and RM-A methods, and the RMS-average and average spectra provided by CISPR 16 are computed. Then, the results are compared with respect to the reference PSD of the synthetic signals, via graphical and statistical analyses, to assess the error in the frequency domain outputs. The results indicate that the most accurate methods for measuring the RMS spectra in the LV grid are IEC 61000-4-7 and the RM-A methods.