PSI - Issue 42

Hasan Saeed et al. / Procedia Structural Integrity 42 (2022) 967–976 Hasan et al./ Structural Integrity Procedia 00 (2022) 000 – 000

968 2

2016; Radaj et al., 2006). Hence, the propagation of short cracks has to be investigated in detail to obtain realistic lifetime predictions. Fatigue crack growth rate testing can be executed with different types of specimens. Standardized specimen designs can be found in ISO 12108 (ISO 12108, 2018), BS 7910 (BS 7910-2019, 2019) and ASTM E647 (ASTM E647-15, 2015). Some commonly used techniques for fatigue crack growth monitoring are potential drop (PD), compliance via crack mouth opening displacement (CMOD) (De Waele et al., 2020), strain compliance, and digital image correlation (DIC) technique (Shanmugham and Liaw, 1996). Two kinds of PD techniques are available, i.e., alternating current potential drop (ACPD) and direct current potential drop (DCPD). Although ACPD has been observed to be quite sensitive to crack initiation and short crack propagation (Chaudhuri et al., 2019; Okumura et al., 1981), it has been reported to be sensitive to the applied stress (Gibson, 1987). This could lead to inaccuracies in monitoring of cracks in the long crack regime (Chaudhuri, 2019). The more conventional DCPD technique (Verstraete et al., 2015) is used in this work. DCPD is based on the correlation of crack propagation with change in electric resistance of the material with constant excitation DC current. Strain based compliance techniques are considered reliable and cost effective techniques for fatigue crack monitoring (Mehmanparast et al., 2017; Wen et al., 2019). A back-face strain (BFS) relation for a standard compact tension C(T) specimen was first proposed by Deans and Richards (Deans and Richards, 1979). Their solution provides a correlation between the computed compressive elastic strain per unit load on the back face of the C(T) specimen and the crack length to specimen width ( / ) ratio. Maxwell (Maxwell, 1987) devised an equation linking crack length and BFS, using finite element (FE) simulations for / ratios ranging from 0.2 to 0.8. Shaw and Zhao (Shaw and Zhao, 1994) provided experimental results for 2024-T351 aluminum alloy using the BFS solution proposed by Maxwell. Riddell and Piascik (Riddell and Piascik, 1998) also provided a BFS relation for C(T) specimens calibrated using finite element simulations (Wawrzynek and Ingraffea, 1994, 1987). Their relation was modified by Newman et al (J. Newman et al., 2011) by adopting the CMOD based compliance equation ( / ) used in ASTM E647 (ASTM E647-15, 2015) and replacing the parameter (CMOD) by (absolute back-face strain times specimen width). is the Young ’s modulus, is the applied load, and is specimen width. The validity range for the Newman-Johnston equation is / ranging from 0.2 to 0.8.The context of this paper refers to crack monitoring in specimens subjected to a state of pure bending, i.e., a constant bending moment between two inner loading points. The standard SENB specimen is defined in ISO 12108 (ISO 12108, 2018); a schematic of which is shown in Fig. 1. Saxena and Hudak (Saxena and Hudak, 1978) presented the first BFS compliance relation for compact tension (CT) and wedge opening loaded (WOL) specimens. Huh and Song (Huh and Song, 2000) calibrated that relation for / ratios ranging from 0.15 to 0.6 on the basis of FE simulations. Salem and Ghosn (Salem and Ghosn, 2010) performed tests for validating BFS, measured with different strain gauge sizes, for a range of crack lengths ( 0.3 ≤ / ≤ 0.6 ) in pre-cracked flexure specimens. Garcia (Garcia et al., 2015) used FE modelling to develop a BFS compliance relation for / ratios ranging from 0.2 to 0.95, along with experimental validation on API 5L X70 grade steel.

Fig. 1. Standard SENB geometry according to ISO 12108:2018 (ISO 12108, 2018); = 2 is the width, is the thickness of the specimen, and 0 is the initial crack length.