PSI - Issue 60

S. Mahesh et al. / Procedia Structural Integrity 60 (2024) 382–389 Mahesh et al. / Structural Integrity Procedia 00 (2019) 000 – 000

383

2

1. Introduction Structural failure due to fatigue loads is a long-standing problem for over decades and seems to get more complicated due to large-scale application of newer materials in various engineering sectors. Even though advancement in design tools have led to a better understanding of the material behaviour under uncertain loading conditions, design using damage tolerance (DT) approach needs newer tools for robust and quicker predictions. Structures generally fail due to poor workmanship, use of inappropriate or substandard materials, ignorance of standard protocols, human errors, and possibly unaccounted loading parameters (Anderson (2017)). It was observed that structures built using well researched and established materials did not always perform as expected and often led to premature catastrophic failure. Gradually, discoveries in the field of fracture mechanics indicated that the defects in the material, in the form of pre-existing flaw could initiate cracks and eventually fracture. These flaws induce stress concentration within the material due to fatigue loads eventually leading to crack initiation at this point. Further growth of these cracks leads to premature failure even though the fatigue loads acting on the structure is well below the maximum loading limit of the material (Boyer (1985)). Thus, safe life of any engineering structure is determined by the process of damage tolerance approach, wherein the structure is periodically monitored for defects and crack growth. The structure is usually either replaced or repaired before the crack reaches critical phase. Fatigue life of the structure is determined either by generating the S-N curve or by evaluating the fatigue crack growth rate (FCGR) curve (Ritchie (1979)). S-N curve gives the total fatigue life of the material whereas the FCGR curve is used to determine the crack growth rate due to cyclic loads acting on the structure/component. Thus, the FCGR data gives the crack growth life of the structure. A typical FCGR curve is a representation of crack growth rate per cycle (da/dN) with respect to applied stress intensity factor range, ΔK on a log -log scale as shown in Fig. 1. The sigmoidal shape of the FCGR curve may be divided into 3 regimes viz., - • Regime 1 – Near-threshold region • Regime 2 - Linear crack growth / Paris region • Regime 3 - Critical / unstable / high  K region Practically it is quite difficult to detect and measure crack growth in the threshold regime. As crack grows beyond the threshold region and further into the Paris region it becomes easier to measure and monitor the crack. In the Paris region, the crack continuously grows although the component appears intact. Once the crack growth reaches the critical region it becomes unstable leading to catastrophic failure (Broek (2012)).

The FCGR test is usually performed using a compact tension (CT) specimen as per ASTM E647 standards (ASTM E647 (2008)). The entire testing process is split into two separate stages. The first stage is the generation of decreasing ΔK curve and the second stage is the generation of increasing delta K curve. In both the stages the constant amplitude fatigue loads are applied on the specimen at a particular stress-ratio (R). During the decreasing Fig. 1. Typical FCGR curve of a metallic alloy