Neutron diffraction measurements of weld residual stresses in three-pass slot weld (Alloy 600/82) and assessment of the measurement uncertainty

Vasileios Akrivos,Robert C Wimpory,Ondrej Muransky,Mike C Smith,John Bouchard,Beverly Stewart,Michael Hofmann

doi:10.1107/s1600576720009140

Abstract

This paper describes in detail two neutron diffraction residual stress measurements, performed on the ENGIN-X neutron scattering instrument at the ISIS facility in the UK and on the SALSA instrument at the Institut Laue-Langevin in Grenoble, France. The measurements were conducted as part of the NeT Task Group 6 (TG6) international measurement round robin on an Alloy 600/82 multi-pass weldment - a slot in an Alloy 600 plate filled with three Alloy 82 weld beads, simulating a repair weld. This alloy/weld combination is considered challenging to measure, due to the large grain size and texture in the weld, and large gradients in the stress-free lattice parameter between the parent and weld metal. The basic principles of the neutron diffraction technique are introduced and issues affecting the reliability of residual stress characterization are highlighted. Two different analysis strategies are used for estimation of residual stresses from the raw data. Chemical composition studies are used to measure the mixing of parent and weld metal and highlight the steep lattice parameter gradients that arise as a consequence. The inferred residual stresses are then compared with three sets of measurements performed on the same plate by other NeT partners on E3 at the HZB in Berlin, STRESS-SPEC at the FRM II in Munich and KOWARI in Sydney. A robust Bayesian estimation average is calculated from the combined five-instrument data set, allowing reliable best estimates of the residual stress distribution in the vicinity of the weldment. The systematic uncertainties associated with the residual stress measurements are determined separately in the weld and parent materials, and compared with those in the NeT TG4 benchmark. This is a three-pass slot-welded plate fabricated from American Iron and Steel Institute AISI 316L(N) austenitic stainless steel, and is normally considered less challenging to measure using diffraction techniques than all nickel welds. The uncertainties in the stress measurements by neutron diffraction for these two weldments seem to be comparable.

Highlights

Residual stresses within a component can adversely affect its structural integrity and reduce its lifetime, as it becomes more susceptible to degradation mechanisms such as stress corrosion cracking, fatigue and creep (Bouchard, 2001)
It assesses the residual stress (RS) profile through the thickness of the weldment starting from the top surface and measuring the RSs in all the weld beads, sampling more weld metal which is always more challenging to measure due to the large grain size and texture
A number of neutron diffraction RS measurements have been performed on the Neutron Techniques Standardization for Structural Integrity (NeT) Task Group 6 (TG6) specimen using neutrons from reactor and spallation sources

Summary

Introduction

Residual stresses within a component can adversely affect its structural integrity and reduce its lifetime, as it becomes more susceptible to degradation mechanisms such as stress corrosion cracking, fatigue and creep (Bouchard, 2001). Substantial efforts have been made to understand RS development in nickel alloy dissimilar metal welds, via both measurement and simulation (White et al, 2007; Lippold et al, 2009; Norring & Engstrom, 2008; Bruemmer et al, 1999; Marlette et al, 2010; Brust et al, 2010; Fredette et al, 2011; Shim et al, 2010; Rathbun et al, 2011; Smith, Muransky, Bendeich & Edwards, 2010; Smith, Muransky, Goodfellow et al, 2010) These have highlighted the need both to make RS measurements using multiple techniques with different characteristic errors and to interpret simulation predictions in the light of reliable measured estimates of RS. This approach is mandated in the R6 structural integrity assessment procedure in the UK (EDF, 2015; Bate & Smith, 2016)

Results

Discussion

Conclusion