Finite element studies have been carried out simulating 3-dimensional deposition of weld metal on a flat plate. The purpose of this work is to further understand and develop the modelling requirements in order to accurately predict post weld residual stresses when compared to residual stress measurements. A single bead on plate specimen has been chosen as it has similar characteristics to those occurring in a repair weld which is of engineering interest. This analysis makes use of detailed fabrication records such as thermocouples, recorded heat inputs and etched macrographs which detail the fusion boundary profile. Analysis of the macrographs indicate a reduced weld penetration towards the end of the weld bead and especially along the centre-line of the bead where a “double-lobed” appearance was noted when viewed on a plane perpendicular to the bead. At the start position, a deeper fusion boundary was observed relative to the majority of the weld bead. The aim of this piece of work was to match the observed and predicted fusion boundaries and to analyse the influence on the predicted residual stresses when compared to a constant fusion boundary, along the length of the bead, as modelled in earlier analysis of the bead-on-plate. Two mechanical simulations were conducted based upon the matched fusion boundary thermal solution. These mechanical simulations utilised two hardening models, both of which are based on a non-linear kinematic hardening model but derived from different test data. A baseline kinematic hardening model has been used that is derived from monotonic and single pass weld bead specimens. A mixed hardening model has also been used which has been derived from both monotonic and cyclic test data. Furthermore this analysis takes advantage of some new features to weld modelling. In particular these include the Dynamic Fusion Boundary (DFB) approach where material properties are assigned to elements according to their temperature history.
A research project has recently been launched in the UK investigating residual stress (RS) in nuclear power plant [1]. At the outset there is a need to review techniques available for modifying/relieving residual stress levels in weldments, since it is well known that large tensile RS levels generated in welds can be detrimental in terms of fatigue, fracture resistance and environmentally assisted cracking (EAC). Therefore current RS mitigation methods have been reviewed. Mitigation methods can be categorised into three main groups as follows: a) Surface treatment to induce compressive skin stress; b) Stress relief through thickness; c) Weld design optimisation to produce low/favourable RS levels and minimize distortion. A brief description is provided of how each method works, together with the capability and potential benefit in terms of RS reduction, as well as references for further information. Metallurgical effects of treatment are also an important consideration. The practicality of application to nuclear plant is considered, both in manufacture and in-service, together with any limitations and risks. Several techniques are identified that are likely to be beneficial and warrant funding for further development. RS mitigation should be targeted at key/critical weld locations in the plant, where loadings and degradation mechanisms (such as corrosion, fatigue, EAC or fracture) are most significant. Treatment would be carried out in order to improve plant integrity and reliability (eg safety margins). There are potentially substantial cost savings since through-life inspection/maintenance work could be reduced and expensive repairs and shutdowns avoided. Note that it is important to understand whether the benefits in terms of RS improvement are likely to be long term. In certain systems large thermal transients are applied that might generate additional surface plastic strains, thereby modifying RS magnitudes and distributions.
The inherent complexity of modelling welding processes and the lack of computational power available to analysts has resulted in simplified methods being commonly utilised when predicting residual stresses. Despite considerable advances in computational power, it is still often not possible to run detailed 3D analyses of complex welded geometries within practical timescales. Against this background, a programme of work has been undertaken to develop a weld modelling procedure which can be followed by analysts. This procedure will account for how various modelling simplifications affect the predicted values of residual stress. One common geometry, which it is often necessary to analyse using modelling simplifications is that of a thin-walled pipe butt weld. Typically this geometry is simulated using a 2D axisymmetric analysis. Despite the popularity of this modelling simplification the effects of its use are not fully understood. In order to feed into this procedure, work has therefore been conducted to better understand the effects modelling simplifications will have on the residual stress levels that are predicted when simulating multi-pass pipe butt welds. The geometry considered in this study is the thin walled austenitic pipe butt weld specimen originally studied in VORSAC 5th Framework European Union project. This paper presents the results of a number of finite element analyses conducted of this geometry. These analyses have been conducted using a combination of the finite element codes SYSWELD and ABAQUS. The aim of this study was to understand the effect that the use of 2D axisymmetric analyses, and other modelling simplifications, namely block dumping and bead lumping will have on the predicted values of residual stress.
A programme of work was undertaken to gain an understanding of the residual stress levels in the tube penetration J-groove attachment welds in a hemispherical head of a large stainless steel clad ferritic pressure vessel. In this first part, of a two part paper, the finite element analyses that were carried out to model the centre nozzle penetration are described. Two axisymmetric residual stress finite element models were developed. One used an accurate representation of the weld bead deposition sequence and the other a bead lumping approach to model bead deposition. The results from the finite element analyses were compared with both surface and through thickness stress measurements. These measurements were taken on a mock-up weld that was representative of the actual component. The surface measurements were taken by using an incremental centre hole drilling technique (ICHD). The through thickness values were obtained from deep hole drilling (DHD) measurements. The DHD measurements were taken before and after the cladding of the mock-up. The analytical results from the two axisymmetric models showed the simpler blocked dump model approach to be reasonable in capturing the general level of stress. The finite element analysis results showed good agreement with the measurements in the radial direction, but predicted greater than the measured values in the hoop direction.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.