Search

The UK High Temperature Power Plant Forum have completed a long term cross-weld creep test programme on a Grade 92 steel pipe. This paper describes an analysis of the creep test data. The weld strength factor (WSF), i.e. the ratio of cross-weld strength to base material strength for a specified rupture time and temperature, is shown to be a unique function of a dimensionless parameter S, the creep test stress divided by ultimate tensile strength at the test temperature. The key practical finding is that, since a single universal cross-weld creep life parametric is applicable to all temperatures, long-term behaviour at service temperatures should reliably be predictable from short-term tests at accelerated temperatures. [ABSTRACT FROM AUTHOR]

To unravel their long-term creep properties at simulated reservoir conditions, we conducted constant stress deformation experiments at elevated confining pressures, pc = 50–115 MPa, and temperatures, T = 75–150 °C, on Posidonia (GER) and Bowland (UK) shale, which exhibit varying petrophysical and mechanical properties. Depending on applied pc–T conditions and sample composition, recorded creep curves exhibit either only a primary (decelerating) or additionally a secondary (quasi-steady state) and a tertiary (accelerating) creep phase during deformation. At high temperature and axial differential stress and low confining pressure, creep strain is enhanced and a transition from primary towards secondary and tertiary creep behavior is observable. Creep strain of Posidonia shale, which is rich in weak constituents (clay, mica, and organic content), is enhanced when compared to creep strain recorded during deformation of either carbonate- or quartz-rich Bowland shale. Electron microscopy observations revealed that creep strain is mainly accommodated by the deformation of weak minerals and local pore space reduction. In addition, microcrack growth occurred during secondary creep. An empirical correlation between creep strain and time based on a power law was used to describe the decelerating creep phase, also accounting for the influence of confining pressure, temperature, and axial differential stress. The results suggest that the primary creep strain can be correlated with mechanical properties determined from short-term constant strain rate experiments such as static Young's modulus and triaxial compressive strength. [ABSTRACT FROM AUTHOR]

This paper presents a numerical study of the performance of two instrumented surface footings at the Bothkennar Clay research site in the UK. Footing A was loaded to failure over 4 days, reaching a net bearing capacity of = 138 kPa; footing B was loaded to 89 kPa, at an identical loading rate, and left to consolidate under maintained load for about 11 years. The preloaded footing was then loaded to failure over 3 days, reaching = 204 kPa. The increase in bearing capacity was significantly larger than that expected due to consolidation effects alone, and it is anticipated that the occurrence of creep and other ageing processes may have played a major role in the observed response. The complete loading history of the two footings is simulated by means of coupled axi-symmetric finite-element analyses in which the foundation soil is described using an elastic-viscoplastic model that mimics isotach viscosity. The ground profile and the model parameters are derived based on the extensive laboratory and field test data available in the literature. The numerical analyses are able to describe accurately the footings behaviour during first loading, the development of delayed settlement under maintained load and the increase in bearing capacity due to preloading. The paper emphasises various issues regarding the application of elastic-viscoplastic models to model boundary value problems in conditions close to failure. [ABSTRACT FROM AUTHOR]

• The developed CPFE-interface element modelling framework of creeping grains and sliding grain boundaries allows calibration and validation of the model at various scales. • Triple junction constraint formulation ensures compatibility is satisfied at these locations. • For Type 316 steel where a population of grain-boundary particles exists (φ 0 > 1), only a moderate increase <50 MPa in local normal stresses at triple junctions in the presence of sliding was predicted. • Boundaries bounded by plastically-harder grains, i.e. those with relative crystallographic orientations away from the [001]-direction, are expected to carry higher stress. [Display omitted] Previous studies demonstrate that grain-boundary sliding could accelerate creep rate and give rise to large internal stresses that can lead to damage development, e.g. formation of wedge cracks. The present study provides more insight into the effects of grain-boundary sliding (GBS) on the deformation behaviour of realistic polycrystalline aggregates during creep, through the development of a computational framework which combines: i) the use of interface elements for sliding at grain boundaries, and ii) special triple point (in 2D) or triple line (in 3D) elements to prevent artificial dilation at these locations in the microstructure with iii) a physically-based crystal plasticity constitutive model for time-dependent inelastic deformation of the individual grains. Experimental data at various scales is used to calibrate the framework and compare with model predictions. We pay particular consideration to effects of grain boundary sliding during creep of Type 316 stainless steel, which is used extensively in structural components of the UK fleet of Advanced Gas Cooled Nuclear Reactors (AGRs). It is found that the anisotropic deformation of the grains and the mismatch in crystallographic orientation between neighbouring grains play a significant role in determining the extent of sliding on a given boundary. Their effect on the development of stress within the grains, particularly at triple grain junctions, and the increase in axial stress along transverse boundaries are quantified. The article demonstrates that the magnitude of the stress along the facets is highly-dependent on the crystallographic orientations of the neighbouring grains and the relative amount of sliding. Boundaries, transverse to the applied load tend to carry higher normal stresses of the order of 100-180 MPa, in cases where the neighbouring grains consist of plastically-harder crystallographic orientations. An approach to estimate the amount of sliding as a function of GB particle morphology and diffusive accommodation is developed and employed. The possible implications of these findings on the initiation of creep damage are commented on. [ABSTRACT FROM AUTHOR]

The R5 procedures have been developed within the UK power generation industry to assess the integrity of nuclear and conventional plant operating at high temperatures. Within R5, there are specific procedures for assessing creep-fatigue crack initiation in initially defect-free components (Volume 2/3) and for assessing components containing defects (Volume 4/5). The current R5 Volume 2/3 procedure for assessing creep-fatigue initiation in weldments involves analysis assuming homogeneous parent material properties with a single Fatigue Strength Reduction Factor (FSRF) used to account for both the reduction in fatigue endurance and enhancements in strain due to material mismatch and local geometry effects. A new development of this approach involves splitting the FSRF into a Weldment Endurance Reduction (WER), which accounts for reduced fatigue endurance due to weld imperfections, and a Weldment Strain Enhancement Factor (WSEF), which accounts for material mismatch and local geometry effects. The new approach results in less conservative predictions of creep damage than are obtained using the existing approach as it is only the WSEF that affects the stress at the start of the creep dwell rather than the larger FSRF in the existing approach. However, the new approach has little effect on the calculated fatigue damage as the combined effects of the WER and the WSEF are essentially equivalent to the previous FSRF. In addition to methods for predicting creep and creep-fatigue crack growth for load-controlled situations, R5 Volume 4/5 gives advice on the prediction of crack growth for situations involving combined primary and secondary loading. This advice has recently been extended to cover cases involving significant amounts of crack growth, C(t) estimation methods for elastic-plastic-creep behaviour and explicit advice on the treatment of combined loading for situations involving significant welding residual stresses. This extended advice can be used to predict the growth of defects in non-stress relieved austenitic weldments operating at elevated temperatures. This paper briefly outlines the current R5 Volume 2/3 and Volume 4/5 procedures and then focuses on these two significant recent developments in R5, both of which are highly relevant to real instances of creep and creep-fatigue cracking that have been observed in high temperature austenitic plant components. [ABSTRACT FROM AUTHOR]