Analysis of Automobile Crash Test Data and Recommendations for Acquiring and Filtering Accelerometer Data

An attempt is made to define the meaningful frequency content of occupant compartment deceleration data in order to establish effective filtering guidelines which will enhance the important features of the deceleration pulse. Acceleration and displacement spectral distributions of crash test and structural resonance data are compared to assess the presence and effects of resonances in the deceleration time history. A typical accelerometer package-floor pan configuration is modeled to characterize resonant modes associated with current accelerometer package size and mounting. Guidelines are suggested for partitioning data frequency content based on these analyses and also on the comparative effects of high and low frequency decelerations on occupant loading. An alternative method to analog filtering of crash data, which employs a least squared error polynomial curve fitting routing, has been developed and is described