The problem: Within a few months from purchase, the fuel gauge in battery-powered devices, from cell phones to EVs, starts to become inaccurate. This is usually noticeable towards the end of discharge where the battery level may suddenly drop from 30% to 5%. This is not a battery problem but a monitoring problem and while this is an annoyance for cell phones and laptops in most cases, it can be critical for EVs and residential/commercial/utility storage systems where users could remain stranded or without power with little notice.
The mechanism: Although there are accurate ways to measure the State of Health (SOH) of a battery, they are typically costly and impractical as they require some significant downtime (hours if not days). To report the state of charge (SOC) of a battery pack, consumer electronics manufacturers tend to use a simple method which consists of implementing an open circuit voltage (OCV) versus SOC curve to be used as a lookup table in a microcontroller. The SOC is then recalibrated by comparing the voltage of a rested battery to the lookup table. In most cases, this lookup table is never updated because, while it is known that the OCV/SOC relationship is evolving with time, its variations are hard to predict. This makes the SOC estimate progressively more inaccurate as the battery ages and degrades.
The solution: Based on more than a decade of experience in battery degradation monitoring, UH researchers have developed a way to dynamically adjust the OCV/SOC curve without the need for specific sensors or protocols and without downtime. The proposed algorithm requires minimal power consumption and calculation and thus can be easily integrated in a microcontroller and embedded in today’s consumer electronics at no added cost. This technology has the potential to maintain the battery fuel gauge accuracy throughout the battery’s life. Moreover, as an added benefit, the adjustment parameters can be used as information to monitor the evolution of the SOH of the batteries and in most cases to quantify the degradation that the battery experienced. This could allow for the early detection of battery failure and thus enable preemptive maintenance before any safety concern arises.