Abstract
To address the critical challenges in characterizing the dynamic responses and identifying the instability precursors of sandstone under repetitive impacts, this study establishes a comprehensive experimental framework. Multi-gradient cyclic impact testing (50-160 cm drop heights) was implemented to simultaneously monitor the dynamic mechanical responses and acoustic emission (AE) signals. Through an integrated analysis of the impact mechanical parameters, power-law statistics of the AE absolute energy, damage stage classification, and b-value/S-value correlation, we reveal the intrinsic relationship between the impact rate, damage progression, and instability precursors. The experimental results demonstrate the following: (1) exponential relationships exist between the impact height and both the peak impact force and time-to-peak in sandstone; (2) under moderate strain rates, the absolute energy probability density of acoustic emissions follows a power-law distribution, with the power-law index correlating with the impact height under equivalent impact cycles; (3) damage evolution manifests through three characteristic stages, showing a negative linear correlation between the bmin values and impact rate; and (4) damage instability exhibits coupled precursors featuring a synchronous b-value surge and S-value collapse. The developed b-S dual-parameter criterion enables precise identification of damage acceleration thresholds. These findings provide new laboratory evidence for dynamic disaster monitoring in rock engineering applications.