Through-process characterisation for welding, microstructure evolution and constitutive response of 9Cr steel material for flexible power plant operation
Mac Ardghail, Padraig
Mac Ardghail, Padraig
Loading...
Repository DOI
Publication Date
2020-05-06
Type
Thesis
Downloads
Citation
Abstract
This thesis presents a combined experimental and computational characterisation programme of work for development of a through-process modelling methodology for the welding, post-weld heat treatment and in-service performance of 9Cr steel power plant components. The through-process modelling methodology links welding processes, microstructure evolution and mechanical performance as a step towards a weld-design tool for industry. This modelling methodology utilises finite element welding simulation to represent test specimens and power plant components. Weld metal regions are defined and weld beads are applied in sequence to represent multi-pass welding processes. Empirical, temperature-dependent microstructure-evolution models are developed to account for the evolution of Vickers hardness, prior austenite grain size, lath-width and the diameter and area fraction of carbide precipitates. The predicted microstructures influence subsequent constitutive behaviour via microstructure-dependent constitutive parameters. A visco-plastic constitutive model is designed to predict the mechanical performance of welded 9Cr test specimens under tensile and cyclic loading conditions and power-plant components under flexible operation conditions. Distributions in predicted microstructure (e.g. the separate HAZ regions) lead to distributions of predicted material behaviour without the need to deliberately partition the FE geometry and define specific material properties. The constitutive model is calibrated against tensile and cyclic test data for 9Cr steel at elevated temperature. vi An experimental test programme is conducted consisting of bead-on-plate welding trials and Gleeble physical simulation for manufacture of heat-affected zone material, as well as subsequent thermomechanical fatigue testing of parent material and simulated heat-affected zone. The bead-on-plate trials provide key thermal history data which acts as input for the Gleeble welding simulations, as well as residual stress, hardness and grain size data for weld-affected regions. Application of the through-process model to the weld-on-bead tests provides general agreement with residual stress, hardness and microstructure (prior austenite grain size) distribution. TMF testing of the Gleeble-simulated heat-affected zone and parent material has demonstrated that the simulated as-welded HAZ has a higher cyclic strength and longer life than parent P91 (particularly for the more detrimental out-of-phase condition). Simulated post-weld heat-treatment leads to significant reduction in as-welded cyclic strength and life relative to the as-welded. The through-process model shows general agreement with these relative cyclic strength trends, viz. significant increase due to welding and decrease due to post-weld heat treatment.
Funder
Publisher
NUI Galway
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland