An Eulerian thermomechanical elastic–viscoplastic model with isotropic and directional hardening applied to computational welding mechanics

An Eulerian thermomechanical elastic–viscoplastic model with isotropic and directional hardening is used to analyse the residual mechanical state resulting from the arc welding of a multi-pass weld. Details of the weld test plate, weld filler material, and numerical implementation of the model are provided, including integration algorithms and consistent tangent modulus. For the computational welding mechanics analyses, the austenitic ASME stainless steel grade 316L was considered so that no phase transformations of solid states needed to be considered. The maximum residual stresses were found to be about 500–600 MPa, which is of the order of the yield stress of the base material. Variations in the heat input and the resulting weld cooling time had a significant influence both on the residual stress state and on the resulting geometry of the weld. The predicted stress levels were compared to the experimental results. Overall, the proposed Eulerian framework seems to be a promising tool for analysing melting/solidification processes and residual mechanical states.

History Reduction by Lumping for Time-Efficient Simulation of Additive Manufacturing

Additive manufacturing is the process by which material is added layer by layer. In most cases, many layers are added, and the passes are lengthy relative to their thicknesses and widths. This makes finite element simulations of the process computationally demanding owing to the short time steps and large number of elements. The classical lumping approach in computational welding mechanics, popular in the 80s, is therefore, of renewed interest and is evaluated in this work. The method of lumping means that welds are merged. This allows fewer time steps and a coarser mesh. It was found that the computation time can be reduced considerably, with retained accuracy for the resulting temperatures and deformations. The residual stresses become, to a certain degree, smaller. The simulations were validated against a directed energy deposition (DED) experiment with alloy 625.

Two Different Approaches to Simulate J-Groove Joints

The large thickness of most of the heavy components in Pressurized Water Reactors often lead to use multi-pass welding processes. Distortion tolerance and maximal residual stress requirements if any are difficult to fulfill by simply adjusting welding processes with a trial-and-error procedure. This is the main reason why AREVA has developed a robust methodology to perform welding numerical simulations leading to have a better understanding of the phenomena involved during the process. The present work details a 3D method successfully used to simulate a peripheral adapter J-Groove attachment weld in a vessel head [1] and compares the results to those obtained with a 3D simplified method. The first method is the state of the art method use for solving Computational Welding Mechanics problems. It is a transient “step by step” 3D simulation using an equivalent moving heat source as input. The second method is a simplified one. First, an equivalent thermal cycle is obtained from a 3D stationary thermal simulation. This thermal cycle is representative of the welding parameters. This thermal cycle is then prescribed to all nodes in a given sector of the weld. This simplified method is called “prescribed thermal cycle” method. A comparison between displacements and stresses obtained by both methods is completed in order to validate the hypothesis of the simplified approach. The results show a good agreement between the transient “step by step” method and the simplified one. Furthermore, the simplified method speeds up the calculations by a factor of 10. These performances offer the possibility to simulate a multi-pass welding of Pressurized Water Reactor components in a limited calculation time, providing an efficient decision making tool for engineering purpose. This work is the result of a fruitful collaboration between AREVA and ESI-FRANCE. All the computations are performed with SYSWELD™ software [3].