Correlation test among EMI measurement sites using a comb generator

Abstract

VCCI had considered that a major reason why we cannot get correlated data on the radiated interference (especially one from the power cord in the frequency range up to 300 MHz) among different measurement sites might be that the characteristics and environment of the power lines to the EUT (equipment under test) are different at each site. Therefore, we performed a correlation test at eight EMI measurement sites using a comb generator as an EUT to whose output terminal a cable was connected, based on an assumption that the correlation among sites would be improved if the common-mode impedance of the power lines is specified at each site. The test results show that the correlation among sites was improved when the common-mode impedance of the power lines was constant. But, the actual laying route and position of the power cord had some influence in the frequency range of 60 to 80 MHz, and the effect of the constant common-mode impedance was less clear than one in the other frequency ranges. Also, the correlation among 14 sites using an actual EUT was improved comparing the standard deviations with the current CISPR22 measurement conditions, when the common-mode impedance of the power lines to an EUT was constant.