loading page

Loss Characterization and Modeling of Ferroelectric Class II Multi-Layer Ceramic Capacitors in Resonant Converters
  • Yunlei Jiang
Yunlei Jiang
University of Cambridge

Corresponding Author:[email protected]

Author Profile


Class II multi-layer ceramic capacitors (MLCCs) employing ferroelectric materials have been widely utilized as the primary energy storage and transfer element in high-frequency resonant converters due to their ultra-high energy density. In these applications, MLCCs experience large-signal and high-frequency electrical excitations superimposed on biased voltages, burdening converters’ efficiency due to their significant power loss. However, little has been published to reveal the large-signal loss characteristics of Class II MLCCs under such realistic operating conditions. This paper develops a novel characterization method to experimentally extract MLCC power loss using a resonant-Sawyer-Tower (Res-ST) circuit under large-signal and wide-frequency-range excitation with DC bias voltage. A general loss modeling approach based on Steinmetz’s Pre-electricized Graph (SPeG) is also proposed to accurately correlate the large excitation amplitude, wide-range frequency, and DC bias voltage. In this paper, the SpeG model first provides the material-specific volume loss density of four common dielectric materials and then can be easily extrapolated to evaluate the loss of the MLCCs with different rated voltage$\&$capacitance but employing the same dielectric material. In order to extend the material-specific SPeG model to device-level application, this paper also prepares an easy-to-follow tool, dielectric thickness observer (DTO), to estimate the dielectric microstructural geometry by tracking the $C$-$V$ characteristics provided in the product datasheet. The combination of the proposed loss characterization method, SPeG model, and DTO can be used as a very convenient and accurate tool to estimate the power loss of MLCCs in different applications, in particular high-frequency resonant tanks. This article is accompanied by SEM images and MATLAB code demonstrating the effectiveness of the proposed DTO.