Influence of post-weld heat treatment (PWHT) on the tensile behaviour of P91 weldments

Thesis / Dissertation

2023

Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
License
Series
Abstract
A key factor for successful installation of Grade 91 steel for high pressure-high temperature applications in power plant is optimised welding and post weld heat treatment (PWHT) which restores homogeneous mechanical properties after welding. Although optimum PWHT is prescribed in standards, deviations in the field practices can potentially lead to accelerated component degradation and failure. Consequently, a full understanding of cross-weld behaviour as a function of deviations in PWHT is required. This study investigates the influence of PWHT on the occurrence of strain localisation in cross-weld tensile specimens tested at temperatures up to typical power plant service temperature. Localised strain maps were obtained during high temperature tensile deformation. These assist in understanding the influence of PWHT on the overall performance of the cross-weld, and the identification of the location most susceptible to failure in the welded specimens. Specimens used in this work were extracted from as-welded new Grade 91 pipe material. Wire cut EDM was used to extract rectangular blocks with the weld in the centre which were machined to required geometry. Specimens for as-welded (AW) tests were not subjected to PWHT. The recommended PWHT for P91 steel is tempering at 760°C for 2 hours. Post weld heat treatments that were performed on specimens included soaking at 720°C, 760°C and 800°C for 2 hours respectively, as well as so-called excursion heat treatments that included extended treatment at 760°C for up to 6 hours. In addition, excursion heat treatments also included over-heating up to 840°C followed by soaking for 2 hours at 760°C. A Gleeble 3800D thermomechanical simulator was used for high temperature tensile testing at a strain rate of 103 s-1 . High temperature tensile tests were performed at 300°C and 535°C. Three-dimensional (3D) digital image correlation (DIC) was used for non-contact strain measurement (i.e., application of virtual strain gauges) to accurately map the occurrence of strain localisation. DANTEC Istra 3D DIC software was used for capturing and post processing the DIC data for strain analysis across the weldment. Specimen subjected to PWHT exhibited reduced resistance to deformation at room temperature having a lower yield strength (YS) and ultimate tensile strength (UTS) compared to the AW condition, indicating that PWHT lowers the materials resistance to the onset of plastic deformation. Despite an overall reduction of UTS at elevated temperatures, this observation was extended to tensile strength at elevated test temperatures. Over tempering resulted in diminished tensile properties as observed in specimens tempered at 800°C for 2 hours. A comparison of specimens subjected to excursions and extended heat treatments shows that excursions resulted in poor tensile properties, exhibiting very low yield strength and ultimate tensile strength compared to the recommended PWHT. Specimens subjected to excursions and over tempering in PWHT exhibited diminished tensile properties compared to the recommended PWHT. The change of tensile behaviour across weldments as a function of PWHT contributes to understanding the metallurgical risk associated with deviations in PWHT field practices. Localised strain maps showed that the weld metal has high resistance to deformation at room temperature which decreases as test temperature increases. At room temperature the peak localised strain occurred in the base metal (BM) for all PWHT conditions. The peak localised strain migrated further into the BM with increased PWHT temperature, while it occurred in a hot zone at elevated test temperatures. Based on experimental results obtained at room temperature, strain localisation was established as a function of heat treatment condition. The non-uniform temperature profile remains a challenge to reliably conclude the extent to which strain localisation is influenced by heat treatment and material property.
Description
Keywords

Reference:

Collections