Battery management systems cannot detect capacity fade -- the most common degradation mode -- because voltage and resistance stay normal

Battery management systems (BMS) in EVs, grid storage, and consumer electronics rely on voltage, temperature, and internal resistance measurements to estimate battery State of Health (SoH), but these signals do not reliably correlate with capacity fade, the most predictable and most common form of battery degradation. A battery can lose 50% of its usable capacity while its voltage profile and internal resistance remain within normal ranges, causing the BMS to report a healthy battery that unexpectedly dies mid-use. Why it matters: Because the BMS gives a clean bill of health to a degraded battery, users and fleet operators cannot plan replacements or adjust usage patterns before capacity drops below functional thresholds. So EV drivers experience sudden range drops that do not match their dashboard estimates, eroding trust in electric vehicles. So grid-storage operators cannot accurately bid into energy markets because they do not know their true available capacity, leading to penalties for failing to deliver contracted energy. So second-life battery buyers cannot reliably assess whether a retired EV pack is worth repurposing, killing the economics of the $5+ billion second-life battery market projected by 2035. So the entire battery industry suffers from a measurement gap where the single most important health metric -- how much energy the battery can actually store -- is the one the BMS is worst at estimating. The structural root cause is that accurate capacity measurement requires a full charge-discharge cycle under controlled conditions (taking hours), which is impractical during normal operation. BMS designers instead use proxy signals (voltage, resistance, temperature) that are computationally cheap and real-time but fundamentally do not capture capacity fade, and traditional algorithms like Coulomb counting and Kalman filters compound errors over time due to the non-linear, temperature-dependent behavior of lithium-ion chemistry.