Microchip Technology has integrated Asahi Kasei Microdevices' CZ39 and CZ3K series coreless current sensors into a new reference design. This system utilizes machine learning to identify arc faults, aiming to improve detection accuracy compared to traditional threshold-based methods.
The design centers on Microchip's dsPIC33A digital signal controller. This hardware performs signal processing and inference algorithms locally, eliminating the need for external processing resources. The sensors feature a 100 ns response time, which helps maintain signal clarity for the machine learning model.
The collaboration involved the engineering team at AKM Semiconductor, the US subsidiary of Asahi Kasei Microdevices. The companies worked together on the current-sensing configuration during the development and validation phases of the project.
The reference design is currently available through Microchip's program. Potential applications for this technology include EV chargers, solar PV systems, energy storage, smart ignition systems, e-Fuse designs, and safety switches for residential or industrial use.
Looking forward, the companies are exploring the adoption of this design for data center applications. As AI workloads increase, the demand for higher power density and high-voltage power architectures continues to grow.
