This study investigates pulse-based diagnostic methods for identifying faulty cells within lithium-ion battery packs in electric and hybrid vehicles. Diagnostic measurements were conducted using a fully electric Volkswagen e-Golf under both dynamic driving (WLTP) and laboratory-based Hybrid Pulse Power Characterization (HPPC) test conditions. Voltage deviations among individual battery cells were analyzed to evaluate system reliability under constant load, peak load, and regenerative braking scenarios. Results revealed that peak load conditions provided the most informative insights for fault detection due to fewer but more significant voltage deviations. In contrast, constant load and regenerative braking conditions frequently exhibited minor deviations. These findings suggest that transient, high-current events are particularly valuable for early identification of cell degradation and faults. Future studies should further investigate the long-term relationships between detected deviations and overall battery health to enhance predictive diagnostics and optimize battery management strategies.

Optimizing the operating temperature of lithium-ion batteries is critical for safe, reliable, and efficient cell operation. Manufacturers’ recommendations vary in this area, which is primarily determined by the cells’ chemical composition and internal structural characteristics. Most manufacturers define the maximum charging voltage level as the same or close to the same value, but there are significant differences in the lower threshold voltage. Lithium-ion cells exhibit increased internal resistance at lower state-of-charge levels, resulting in elevated heat generation during operation, with intensity proportional to the depth of discharge. However, using a too low voltage threshold causes a significant loss of usable capacity, which reduces the cell’s energy utilization. The present research aims to define and analyze the optimal value of the lower voltage threshold more precisely, considering both thermal development and usable capacity aspects. A further objective is to determine an optimal energy safety margin level that provides a suitable compromise for longer-term storage. Different 18650 and 21700 standard lithium-ion cell types were tested using various load profiles. The results show that the two cell formats have different electro-thermal behaviors. The 21700 cells show a clear increase in thermal efficiency at around 3.1 V. In contrast, the 18650 cells have a heating pattern that depends heavily on the load. This requires selecting a cutoff that adapts to the discharge rate to prevent excessive thermal stress. These findings indicate that a fixed lower threshold voltage for all cells is not ideal. Instead, we need cutoff strategies that are specific to each cell and can change dynamically. The TER-based evaluation introduced in this work provides a practical framework for defining these adaptive limits. It may improve control in battery management systems in real-world applications.