Ways to improve the analysis of step-stress testing

Abstract

Step-stress testing has a number of worthwhile applications for the analysis of the projected life for components and occasionally systems. These include the common situations of testing when only a small number of systems are available, when extremely long or specialized test equipment are required, when limited environmental chamber capability and/or test fixtures are involved, and lastly when very expensive support equipment is a test limit. Step-stress testing has had limited applicability in the past. This has been due to improperly described degradation or accumulative fatigue, poor control of the test samples and difficulty with the analysis of the failure data. A tight series of step-stress ground rules are proposed in this paper to solve or mitigate these and other common accelerated test problems. This paper also presents methods to improve the analysis of the step-stress test. While these methods are approximate, they lend themselves to analysis on a computer by standard hand techniques. Additionally, the ground rules presented suggest that wider limits may be taken for running a step stress test than has been suggested in the past. These ground rules aid the analysis by helping limit the range of results. The step intervals need not be of the same length nor stress steps uniform in size.