Abstract The early detection of micro internal short circuit (ISCr) cells can provide sufficient response time for preventing accidents such as spontaneous thermal runaway in battery packs of electric vehicles, and greatly improve safety. Because the existing models describing ISCr are mainly equivalent circuit models and three-dimensional physics-based models, we build a pseudo-two-dimensional model of micro ISCr cells to make up for the gaps. Using the calculation results of this model, we reveal the phenomenon of electric quantity depletion and the variation of internal electrochemical parameters when a micro ISCr occurs in the cell. We find the effective electrical conductivity of the separator is a crucial parameter describing the ISCr severity and determine reasonable values for this effective conductivity for fault diagnosis and battery design. Moreover, we propose an impedance-identification method that can be used for ISCr diagnostics. Through the simulation and experimental results, we find that the impedance of micro ISCr cells is different from that of normal cells and shows a certain regularity with the increase of ISCr severity.