Development of a monitoring system for non-rechargeable lithium batteries

This thesis is an improvement of the electronic systems that compose WAPPFRUIT, a regional project focusing on smart technologies for water management in fruit growing, aiming to promote rural development in the Piedmont region. The project deployed systems to collect essential soil data (e.g., soil temperature, matric potential, volumetric water content) using LoRa network nodes to characterize and irrigate orchard fields. Efficient and long-lasting power supply is crucial for these systems. They rely on a primary battery in lithium-thionyl chloride (LiSOCl2) chemistry without estimating the SOH (State of Health) battery. This chemistry is challenging in the monitoring of this parameter. When the battery is almost dead, there is a steep voltage drop phenomenon, making end-of-life detection vital in such a way as to prevent an electronic system for real-time monitoring from being switched off suddenly. This work has been addressed two main objectives. Firstly, firmware has been developed for managing battery end-of-life using the BQ35100 IC (Integrated Circuit) from Texas Instruments. Lastly, a PCB (Printed Circuit Board) has been designed to realize a new electronic prototype estimating LiSOCl2 battery end-of-life. The research started with a deep analysis of BQ35100 manuals, revealing its complexity with RAM, flash memory, and internal algorithm for battery end-of-life assessment. Communication is established via the I2C (Inter-Integrated Circuit) protocol. Software development began with implementing the I2C command sequence to read the BQ35100's RAM, considering its timing constraints. The next phase involved the creation of a library for reading and writing data to flash memory, essential for LiSOCl2 gauge operational mode. After the software development step, the challenge shifted to the selection of a suitable testing method to discharge the battery quickly and estimate the battery lifetime. Given the big battery capacity and the low maximum current discharge battery value, a full cycle could take months or years. A stress test solution has been designed to test BQ35100 functionalities and to observe the direct correlation between battery depletion and increasing internal battery resistance. In addition to this work, a PCB has been developed to monitor innovative electronic systems for precision agriculture, guaranteeing the maximum performance from the BQ35100 side. The thesis concluded by verifying effective battery end-of-life management. This value will be extended beyond WAPPFRUIT, offering an accurate way to estimate LiSOCl2 battery lifetime in such a way as to realize long-lasting IoT (Internet of Things) devices for precision agriculture.