A comparative analysis of DPI and BCI methods for EM immunity of automotive ICs

Abstract Direct power injection (DPI) and bulk current injection (BCI) tests are widely used for integrated circuit (IC)-level and system-level immunity evaluation. While both aim to assess susceptibility to conducted radio frequency (RF) disturbances, their injection paths, test configurations, and resulting immunity metrics differ fundamentally. Even throughout different BCI setups, the results change. This makes BCI immunity highly non-predictable. This paper presents a comparative study of DPI and BCI testing using an automotive high-side switch as the device under test (DUT). Immunity results, injection path attenuation, and disturbance voltages at the IC supply pin are measured and analyzed for both methods. The results show that overall immunity metrics from DPI and BCI have limited direct comparability. In particular, the common current measurement used in the BCI setup fails to provide a reliable indicator of IC failure, as system current resonances do not directly correlate with differential-mode (DM) disturbance voltages at the IC pin. In contrast, disturbance voltage measurements at the IC pin offer a more meaningful metric for predicting IC failure and enable a more direct comparison between BCI and DPI immunity characterization.