Damage accumulation comparison for various vibration test profile generation methods applied to a complex payload

Abstract

There are potential sources of uncertainty when using default or generic profiles for vibration testing, particularly for large and complex payloads ( Warren and Joiner, 2019 ). Replacing these severities with custom profiles generated using measured data offers one alternative; however, care must be taken not to select an inappropriate or inaccurate method. This research details the methodology and results of a trial which involved the design and implementation of a series of vibration tests, each to simulate a mixed-terrain environment applied to a complex payload. A systematic approach was used to develop and execute a test plan via the design of experiments method. An estimate of the damage potential for each was formed using Miner’s rule (linear damage accumulation) applied to direct strain gauge data. This approach was compared with damage measured during a time waveform replication style test to be used as a baseline. The results of this test can be used to guide test specifiers and programme-level acceptance test guidelines and may be useful in implementing or guiding standards in this space. In addition, the damage measured in a series of increasing amplitude vibration tests (with an identical power spectral density) was compared with predicted values to assess the gain linearity assumptions often used in accelerated vibration testing. The research is unique as the authors could not source a thorough trial comparing the damage produced using a range of different profile development methods with a complex, real-world payload. Test houses and test specifiers seeking to improve the accuracy of their tests should consider the key findings and guidance in this article.