Low‐Frequency Spectral Representation of Space Shuttle Flight Data

Abstract

Eighteen accelerograms associated with the lift‐offs of the space shuttle are processed digitally. The recording instruments are calibrated for the 0–50 Hz band. To account for the nonstationary nature of the space shuttle lift‐off, each of the accelerograms is split into three time slices. The first represents the space shuttle main‐engine‐thrust buildup, the second represents the period just following the solid‐rocket boosters' ignition, and the third represents the subsequent period involving intense acoustic activity due to exhaust‐plume engagement on the launch pad. The spectral treatment of the data involves the use of autoregressive (AR) and autoregressive‐moving‐average (ARMA) filters. The filters provide analytical representations of the spectra for all three time slices in terms of a set of coefficients. These coefficients can also be used to synthesize artificial acceleration time histories, which are necessary in base drive qualification studies of critical payloads exhibiting linear or nonlinear behavior. In this context average shock spectra of actual flight data and of pertinent synthetic accelerograms are produced to expedite the dynamic analysis of future payloads.