PSI - Issue 42

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

971

5

Fitting a polynomial using the least squares technique to the numerically obtained compliance function values, the corresponding calibration function is given in equation ( 2 ). A fifth-degree polynomial was used to represent the function as the coefficient of determination (R 2 ) was found to be satisfactory and relative improvement was not found with further increment of the polynomial order. This calibration function describes the compliance data with an error less than 1% ( 2 =0.9998). ⁄ = 0 + 1 ( ′) + 2 ( ′) 2 + 3 ( ′) 3 + 4 ( ′) 4 + 5 ( ′) 5 ( 2 )

Table 1. Extended back-face strain compliance function. Degree, i Coefficient, N i 0 0.049 1 1.883 2 -6.192 3 17.464 4 -24.504 5 12.874

3. Experimental study

3.1. Specimen preparation and strain gage instrumentation In order to experimentally validate the FE based BFS compliance relation, SENB-4P bending fatigue tests were performed. The specimens were extracted from a 50 mm thick steel coupon having a K-weld configuration made by Flux Cored Arc Welding (FCAW 138); a cross-section of the weld is shown in Fig. 5.

Fig. 5. Microstructure of the weldment. The specimens were subsequently machined to dimensions =40 mm and =20 mm. The exact outline of the weld was then identified using an etching process (as shown in Fig. 6 - left), to allow placing a notch within the heat affected zone (HAZ). Notches were machined to a depth of 300 μm and a width of 200 μm (see Fig. 1Fig. 6 – right). The milled notches were next electropolished as per the recommendation given in ASTM E-647 (ASTM E647-15, 2015) for minimizing residual stresses that could have been introduced due to machining. A solution of 65% methanol, 32% butoxyethanol and 3% perchloric acid was used as electrolyte.