Metrological Reliability and Long-Term Stability of Digital MEMS Accelerometers: Investigating Calibration Factor Drift, Output Data Rate (ODR) and Signal-to-Noise Ratio (SNR) Consistency Under Accelerated Thermal-Hygrometric Aging

This study investigates changes in three metrological attributes of eight nominally identical digital MEMS accelerometers from one production lot. After approximately six years of laboratory storage, six devices underwent three unpowered thermal-hygrometric stress protocols, while two devices provided a natural-aging baseline. Calibration factors were compared for one axis at 10 Hz, together with effective output data rate (ODR) and static signal-to-noise ratio (SNR). Arrhenius, Norris–Landzberg and Peck relationships were examined as conditional exposure-equivalence models rather than validated lifetime predictors. Two thermally cycled devices exhibited calibration-factor increases of 7.9% and 37.6%; the other four stressed devices remained compatible within the reported expanded uncertainties. ODR changes remained below 0.5% for all eight devices. The six stressed devices showed SNR decreases of 0.34–1.43 dB, averaging 0.89 dB, whereas the two baseline decreases were 0.34 and 0.20 dB. Stable ODR therefore did not guarantee stable calibration factor. The contribution is a paired assessment of these selected output attributes, not a complete characterization of the accelerometer or identification of the degrading subsystem. The limited sample size precludes population-level reliability estimates.