PSI - Issue 17

Konstantinos Kouzoumis et al. / Procedia Structural Integrity 17 (2019) 347–354 Konstantinos Kouzoumis / Structural Integrity Procedia 00 (2019) 000 – 000

348

2

1. Introduction

Engineering components such as pressure vessels are usually subject to pressure, residual stresses generated from welds during manufacturing and thermal stresses. Such complex loading often creates a multiaxial state of loading. Surprisingly, the effects of load multi-axiality is generally ignored in standard assessments of integrity. Substantive research has been conducted to address the effect of load biaxiality on integrity (Bass et al., 1994; Lidbury et al., 2006; Mostafavi et al., 2009). On materials that show a moderate level of plasticity prior to fracture load biaxiality increases the crack tip multi-axiality and thus increases plastic constraint, or in other words, it suppresses the flow of plasticity. It has been shown (Bass et al., 1994) that there is a decrease in the critical value of the energy release rate ( J c ) at fracture with increasing plastic constraint. This is a well-established effect of biaxiality on fracture on the lower transition region. However, many structures, such as offshore pipelines carrying ambient temperature fluids under a combination of internal pressure and axial loading, experience biaxial loads in the plastic collapse dominated-upper shelf. Experimental evidence showing biaxiality effects on plastic collapse (Østby and Hellesvik, 2008) strongly suggest a correlation between variation in limit load/strain capacity/plasticity flow and load biaxiality. Thus, many researchers have studied the load bearing capability of cracked components under biaxial (or combinations of) loading and its effect on assessing them (Meek and Ainsworth, 2014; Miura and Takahashi, 2010). The current advice on biaxiality in the Fitness for Service (FFS) procedures, like BS 7910 and R6 (BSI, 2013; EDF Energy Nuclear Generation Ltd., 2000), is limited and emphasizes on the constraint effects on fracture toughness. In pursuit of better understanding and more accurately assessing the effects that biaxial loading has on the integrity of a structural component, several of the biaxial and uniaxial tests conducted previously in TWI (Phaal et al., 1995) are being reanalyzed. These tests prove to be suitable for reanalysis given that they are high in number, include both surface breaking and through thickness flaws and have been tested throughout a spectrum of temperatures, biaxiality ratios and thicknesses. This work analyses four, tested at temperatures corresponding to lower shelf fracture toughness. The assessments include FALs created with Option 1, which requires knowledge of the yield stress and ultimate tensile strength as well as with Option 3 which is a tailor made FAL corresponding to the material and geometry studied and is generated via FEA. In addition to the Option 3 FALs, a constraint modified Option 3 FAL is created for all specimens that experience constraint relaxation. Initial Option 1 assessments were conducted with the limit load and stress intensity factor solutions of uniaxially loaded plates (assessments “A”) as well as with using the limit load solution which takes biaxiality into account (assessments “B”). The Option 3 and constraint modified Optio n 3 analyses have assessment points that originate from the FEA calculated values of limit load and stress intensity factor. Overall 20 large scale tests including uniaxial and biaxial loading had been previously conducted on A533B pressure vessel steel plates by TWI. From these tests, four will be analysed here. The tests’ numbering from the original reports (Andrews and Garwood, 1994; Challenger and Andrews, 1996; Garwood and Andrews, 1995) and the one used in a previous analysis (Hadley, 2018) are shown in Table 1. All the plates assessed here include through thickness cracks. The geometry of the biaxially loaded plates was that of a cruciform specimen, while the geometry of the uniaxially loaded was similar with the two loading legs of the specimen removed; a schematic of the specimens is shown in Figure 1.The thickness and crack size values of each specimen along with test temperatures, biaxiality ratios and failure loads are presented in Table 1. The biaxiality ratio k is defined according to equation (1). 2. A533B Plate Tests

(1)

2 1 k P P =

The four results comprise three discrete tests, since #41a did not reach failure during initial loading [#41a(1)], where a biaxiality ratio k=2 was applied, and was reloaded to failure with a ratio k=0.5 [#41a(2)]. The failure load reported in Table 1 for k=2 corresponds to the maximum load prior to unloading.