PSI - Issue 68

Robert Sundström et al. / Procedia Structural Integrity 68 (2025) 1081–1090 Robert Sundström / Structural Integrity Procedia 00 (2025) 000–000

1086

6

Drilled and reamed surfaces of the hollow specimen have been compared for a X65 pipeline steel (Campari et al. 2024). Drilled surfaces had deep scratches but they did not appear to cause preferential cracking as the pipeline steel had a sufficient ductility to accommodate the deformation. They observed a larger scatter in ductility for drilled compared to reamed specimens in hydrogen, concluding that the lower ductility in the drilled specimens tested in hydrogen was caused by a higher number of cracks, both in the necking region and farther away from the fracture. In reamed specimens, cracks only occurred in the necking region. Roughness measurements have been done on the inner surface of deep drilled hollow specimens in pipeline steels, austenitic stainless steels and ferritic steels (Michler et al. 2022, Michler et al. 2023, Michler et al. 2024). It is generally observed that the surfaces are smooth but with occasional deep scratches. Peak-to-valley roughness values (Rz) ranged from 0.2 to 8 for pipeline steels (Michler et al. 2022). The deeper scratches did not appear to have cracks growing from them because of the high ductility of the pipeline steel. Austenitic stainless steels had values of Rz = 2.03 µm and Ra = 0.35 µm, but with some deeper scratches down to 8 µm (Michler et al. 2023). A pipeline steel and a low alloy steel (Michler et al. 2024), both with UTS < 1200 MPa, showed Ra = 0.14 ± 0.02 and 0.11 ± 0.11 respectively. The low alloy steel had some deep scratches up to 10 µm, but cracks were not observed to have initiated there more than in other places. An ultra-high strength martensitic steels with an UTS of 2000 MPa had Ra = 0.33 ± 0.29 µm with some deep scratches. Drilling had resulted in a 50 µm machined layer which contributed to crack initiation. The higher notch sensitivity of this high-strength steel resulted in failure at a lower strain for the hollow specimen compared to the solid specimen, resulting from a notch effect caused by surface roughness. These roughness measurements (Michler et al. 2022, Michler et al. 2023, Michler et al. 2024) show that alloys with very high UTS are more sensitive to roughness effects at the inner surface than those with lower UTS. This can give dramatically different SSR testing results when compared to a conventional specimen. Drilling can give deep scratches, but the ductility of the alloy will determine whether this causes preferential cracking during SSR testing. Drilling promotes cracking outside the necking region (Campari et al. 2024). The influence of residual stresses on SSR response or fatigue lives Some previous work comparing solid to hollow specimens tested in air or inert environments has been done, focusing on different wall thicknesses and its effect on fatigue. For LCF testing of 316 stainless steel at 866 K with thicknesses ranging from 0.6 to 1.3 mm, LCF lives of tubular specimens were much lower than for solid bars (Van Den Avyle 1983). The inner surface was honed, and the specimen internally pressurized with argon. Fatigue lives decreased as wall thickness decreased. Crack growth rates were higher for thinner wall thicknesses; the number of cycles spent in propagation decreases substantially with decreasing wall thickness. Scatter in number of cracks to reach crack initiation was attributed to surface finish. For 316H stainless steel with thickness 2.77 mm and alloy 800 with thickness 3.20 mm, smaller fatigue life reductions were seen. The results suggest a thickness effects in hollow specimens in low cycle fatigue at elevated temperatures under an internal pressure of inert atmosphere for stainless steel. Hollow specimens of 429LM stainless steel with a wall thickness of 2 mm and a honed inner surface had larger internal plastic strain than solid specimens (Bae and Lee 2011). Larger differences in fatigue lives at high strain amplitudes were seen when tested in air, which was attributed to differences in crack propagation behavior. A literature review on the effects of specimen geometry on fatigue lives in light water reactor (LWR) environments has been performed (Twite et al. 2016), comparing the solid specimen to a hollow one for various stainless steels. As in the context of testing in gaseous hydrogen, both a solid specimen tested in an autoclave and a hollow specimen containing the LWR environment have been used to generate mechanical testing data in this research field. LWR is liquid and contains dissolved hydrogen. Hollow specimens used for LWR testing often have internal diameters of 4 to 6 mm and wall thickness of 1 to 3 mm. Testing in LWR environments is performed at elevated pressures and temperatures. Autoclave testing in LWR environments is fraught with much the same testing difficulties as testing in hydrogen, for instance strain control, alignment, seal friction and load monitoring and long times to change specimens. They (Twite et al. 2016) found mixed results from this review, with some reporting differences in fatigue lives and some not. Investigators that did report differences in fatigue lives used small wall thickness, 0.71 to 2 mm, and has not been studied in the hollow specimen. 2.4. Comparisons of solid and hollow specimens