Previous work presented residual stress measurements in an electron beam weld in a thick section ferritic forging [1]; this weld was also modelled using finite element analysis. Due to the tool used to model the heat source, the mesh density in the region of the weld was limited. This work improves on the previous work by using a DFLUX subroutine to provide a mesh-independent heat source input, allowing a better mesh in the region of the weld. The modelling was carried out in Abaqus[2] using the VFT[3] user material model to allow phase transformation effects to be included. This however does not include creep properties and so the as-welded stresses were seeded on to a model that used Abaqus built-in material properties in order to model the heat treatment. The results of this analysis have been compared with analyses run using just the VFT material model (with no creep) and using just the Abaqus properties (with no phase transformation) in order to investigate the sensitivity of the stresses predicted to the material model used. The results of all three analyses have also been compared to the results of the original analysis and with the deep hole drilling residual stress measurements.

Considering the significant role that residual stresses play in determining the lifetime-service of materials, it is mandatory to have a good understanding of and a means of predicting those that develop during welding processes. For this purpose, a User MATerial subroutine (UMAT) is developed to study the effects of various parameters that influence solid state phase transformations and residual stress evolution during welding of SA508 ferritic steel. The temperature dependent elastic and kinematic hardening parameters for each of the individual phases that can potentially develop during cooling from elevated temperatures are measured and are used for calculating stress development during low (75 mm/min) and high (300 mm/min) speed gas-tungsten arc welding (GTAW) on SA508 grade 3. These two speeds are selected to cover a wide range of cooling rates in the heat affected zone so that different phase proportions would be present. The results of the numerical simulations for residual stresses are compared against those measured by neutron diffraction. It is shown here that a low speed weld results in bainite formation whereas a high speed weld results in bainitic as well as subsequent martensitic phase transformations where each welding rate results in different residual stress development.

This paper describes the results of weld model analysis and deep hole-drilling measurements undertaken to evaluate residual stress distributions in austenitic and ferritic steel thick section electron beam welds. The work was undertaken in support of a Rolls-Royce and TWI development programme in NOMENCLATURE DHD Deep Hole Drilling EB Electron Beam FE Finite Element FZ Fusion Zone HAZ Heat Affected Zone ID Inner Diameter OD Outer Diameter PS Proof Strength PWHT Post-Weld Heat Treatment

Small bore austenitic stainless steel pipework is used in a number of nuclear plant systems. Many of these locations are subjected to large thermal shocks and therefore have high fatigue usage factors. Their justification therefore often includes a fatigue crack growth and fracture assessment, for which a key input is the residual stress associated with the welding process, in UK assessments these are typically taken from the R6 compendium. A common process used for these welds is manual tungsten inert gas welding, due to access difficulties each pass is usually completed in two halves. The stop-start locations for each weld run are sometimes stacked, especially in horizontal pipe runs where each weld operation starts at the bottom of the pipe and progresses upwards. The stack up of stop-start locations is likely to lead to considerable circumferential variation in weld residual stress, potentially resulting in stresses that locally exceed the R6 profiles. This paper presents results from a series of FE models for a single small bore pipe weld. The simulated weld is a 3-pass manual TIG weld with an EB insert in a 2 inch (50 mm) nominal diameter pipe. Both 2D and 3D models were run. The results of the modelling are then compared with measurements of weld mock-ups of the same weld (both with and without the stop-start stack-up). The results show that, local to the assumed stop location the predicted stresses do exceed even the R6 level 1 profile (a membrane stress equal to the 1% proof stress of the material). However, the locally enhanced stresses drop off quickly away from the peak location, so for defects of a size that may be a concern for a defect tolerance assessment, the R6 Level 1 and 2 profiles remains appropriate or bounding.

Where a crack has grown such that it crosses a material interface, such as in a clad pressure vessel, there is an additional assessment requirement, above the surface and depth points to evaluate the material interface region. This is typically in terms of fatigue crack growth behaviour and crack extension by ductile tearing. Generally, elastic stress intensity factor (KI) solutions, such as those in R6, only consider the surface and deepest point locations, and therefore do not provide a methodology to account for any interactions at material interfaces. Therefore, for a material interface, an assessment may look to describe the region by the maximum of the surface or deepest point values, or it may be argued to be represented by the value from either the surface or deepest point, depending on which is geometrically closer. However, this has the potential to lead to overly conservative interpretations in terms of KI and the elastic-plastic equivalent, KJ, at the material interfaces, depending on the fracture toughness and growth laws used. The work presented here investigates the influence of the clad to plate interface on stress intensity factor under elastic (KI) and elastic-plastic (KJ) conditions for a number of defect geometries, loading conditions and material properties. The results presented in this paper seek to provide guidance on the suitability of KI and KJ for the surface and deepest points, extracted from models containing only plate material properties, to describe the behavior of a crack which crosses a material interface region in a clad pressure vessel wall.

This paper describes some of the outcomes of the development of finite element modelling guidelines for the stress analysis of bolted joints in pressure vessels and piping. The modelling methods originally developed at Rolls-Royce typically used 2D axisymmetric models as this was deemed adequate at the time. However, computing software and hardware improvements have subsequently been made which enable more realistic 3D bolted joint models to be solved where a greater level of geometric detail is required. For example the bolts, nuts and perforated flanges can now be represented more realistically reducing the degree of geometric abstraction that is required. Also, modern finite element codes such as ABAQUS and ANSYS now offer gasket elements which enable the initial compression, in-service performance and unloading of the joint to be modelled more realistically. Additionally, contact techniques can also be used to simulate the axial and radial distribution of thread load in the joint which will affect the stress distribution remote from the threaded region. Consequently, the modelling guidelines have been updated and provide guidance for stress engineers to decide which degree of model complexity is warranted.

No abstract

The inner surfaces of ferritic steel pressure vessels are often coated with a weld-deposited layer of austenitic cladding for the purpose of corrosion resistance. UK practice of assessing the defect tolerance by postulating defects in components leads to the requirement for a fracture assessment that includes cladding. Section XI requires that the cladding be considered in the fracture analysis, including cladding-induced stresses in the KI evaluation; these stresses largely arise due to the different coefficients of thermal expansion of the cladding and base material, and can also be known as cladding residual stresses. It is thought that in some assessments of clad pressure vessels, the cladding residual stress is either not included, or assumed to be of a significantly reduced magnitude — due to it decreasing at typical operating temperatures and/or being relaxed by prior overloads such as a hydrostatic test. This paper describes finite element analysis to determine the effect of cladding residual stress, comparing J values for loading with and without cladding stresses. Comparisons are also made with J-estimations using the R6 fracture assessment procedure. It is argued that cladding-induced stresses would not contribute to ductile failure as they would be reduced by prior loading, with remaining residual stresses being overcome under large plasticity before failure, due to their secondary nature. Further, it may be misleading to include them in an R6 assessment when using simplified approaches, as their contribution can be exaggerated through pessimistic calculations of the V factor.