This chapter explains why it is sometimes necessary to separate inelastic from elastic strains and how to do it using one of two methods. It first discusses the direct calculation of strain-range components from experimental data associated with large strains. It then explains how the method can be extended to the treatment of very low inelastic strains by adjusting tensile and compressive hold periods and continuous cycling frequencies. The chapter then begins the presentation of the second approach, called the total strain-range method, so named because it combines elastic and inelastic strain into a total strain range. The discussion covers important features, procedures, and correlations as well as the use of models and the steps involved in predicting thermomechanical fatigue (TMF) life. It also includes information on isothermal fatigue, bithermal creep-fatigue testing, and the predictability of the method for TMF cycling.

Strain-range partitioning is a method for assessing the effects of creep fatigue based on inelastic strain paths or strain reversals. The first part of the chapter defines four distinct strain paths that can be used to model any cyclic loading pattern and describes the microstructural damages associated with each of the four basic loading cycles. The discussion then turns to fatigue life prediction for different types of materials and more realistic loading conditions, particularly those in which hysteresis loops have more than one strain-range component. To that end, the chapter considers two cases. In one, the relationship between strain range and cyclic life is established from test data. In the other, a rule is required to determine the damage of each concurrent strain and the total damage of the cycle is used to predict creep-fatigue life. The chapter presents several such damage rules and discusses their applicability in different situations.

This chapter addresses the question of how to deal with multiaxial stresses and strains when using the strain-range partitioning method to analyze the effects of creep fatigue. It is divided into three sections: a general discussion on the rationale used in formulating rules for treating multiaxiality, a concise listing of the rules, and an example problem in which axial creep-fatigue data is used to predict the torsional creep-fatigue life of type 304 and 316 stainless steel. The chapter also includes a brief introduction in which the authors outline the challenges presented by multiaxial loading and set practical limits on the problem they intend to treat.

This chapter compares and contrasts empirical approaches for partitioning hysteresis loops and predicting creep-fatigue life. The first part of the chapter presents experimental partitioning methods, explaining how they can be used to partition any loading cycle into its basic strain-range components. The methods covered include rapid cycling between peak stress extremes, half-cycle rapid loading and unloading, and variations of the incremental step-stress approach. The methods are then compared based on their ability to predict creep-fatigue life. The chapter goes on from there to describe how fatigue life can be estimated from ductility measurements when cyclic data are unavailable or are likely to change. It also explains how cyclic life is influenced by the time-dependent nature of creep-plasticity and the physical and metallurgical effects of environmental exposure.

This chapter explains how the authors assessed the potential risks of creep-fatigue in several aerospace applications using the tools and techniques presented in earlier chapters. It begins by identifying the fatigue regimes encountered in the main engines of the Space Shuttle. It then describes the types of damage observed in engine components and the methods used to mitigate problems. It also discusses the results of analyses that led to changes in design or approach and examines fatigue-related issues in turbine engines used in commercial aircraft.