Noel P. O'Dowd

A unified viscoplastic model for high temperature low cycle fatigue of service-aged P91 steel

The finite element (FE) implementation of a hyperbolic sine unified cyclic viscoplasticity model is presented. The hyperbolic sine flow rule facilitates the identification of strain-rate independent material parameters for high temperature applications. This is important for the thermo-mechanical fatigue of power plants where a significant stress range is experienced during operational cycles and at stress concentration features, such as welds and branched connections. The material model is successfully applied to the characterisation of the high temperature low cycle fatigue behavior of a service-aged P91 material, including isotropic (cyclic) softening and nonlinear kinematic hardening effects, across a range of temperatures and strain-rates.

Microstructural Modeling of P91 Martensitic Steel Under Uniaxial Loading Conditions

The changing face of power generation and the increasingly severe conditions experienced by power plant materials require an improved understanding of the deformation and failure response of power plant materials. Important insights can be obtained through computational studies, where the material microstructure is explicitly modeled. In such models, the physical mechanisms of deformation and damage can be represented at the microscale, providing a more accurate prediction of material performance. In this paper, two approaches are examined to represent the microstructure of a martensitic power plant steel (P91). In one approach, the model is based on a “measured microstructure” with electron backscatter diffraction (EBSD) employed to obtain the orientation of the martensitic grain structure of the steel. The alternative approach is to use a “numerically simulated” model where the microstructure is generated using the Voronoi tessellation method. In both cases, the microstructural model is incorporated within a representative volume element (RVE) in a finite-element analysis. The material constitutive response is represented by a nonlinear, rate dependent, finite strain crystal plasticity model, with the microstructural orientation specified at each finite-element integration point by the microstructural model. The predictions from the two approaches are compared. The stress distributions are observed to be very similar, though some differences are seen in the strain variation within the RVE. [DOI: 10.1115/1.4026028]

Comparison of methods for obtaining crack-tip stress distributions in an elastic-plastic material

Under linear elastic and elastic-plastic conditions the K field and the HRR (Hutchinson-Rice-Rosengren) field respectively are expected to provide an accurate representation of the stress field close to the crack tip in an elastic-plastic material. It has recently been proposed in French and UK defect assessment procedures that the Neuber method, originally developed for sharply curved notches, provides an alternative approach to estimate both notch and crack-tip stress fields, for use in conjunction with the sigma-d (σd) method to predict creep crack initiation times. In this work, the crack-tip stress fields under plane strain conditions, predicted from the Neuber approach, are compared with the HRR and K fields as well as those obtained from full-field finite element calculations. A compact tension and a single edge notched tension specimen have been examined; the material model used is the Ramberg-Osgood, power law plasticity model. As expected, the K field and HRR field have been found to provide a good representation of the near-tip fields at low and high loads respectively. Satisfactory solutions have also been obtained through the use of the reference stress to estimate the amplitude of the crack-tip stress in conjunction with the HRR field. The Neuber approach provides a good estimate of the equivalent (von Mises) stresses over the full range of load levels. However, but the use of the Neuber approach directly to predict the maximum principal stress in the plane of the crack provides a non-conservative prediction. A modified Neuber method, using an appropriate scaling function, has been proposed to determine the maximum principal stress in the plane of the crack from the equivalent (von Mises) stress predicted by the Neuber approach. Using the proposed method, a significantly improved estimate of the crack-tip stresses is obtained.

Methodology for Assessment of Surface Defects in Undermatched Pipeline Girth Welds

The demand for subsea transport of highly corrosive constituents has noticeably increased in recent years. This has driven the requirement for high strength pipelines with enhanced corrosion resistance such as chromium stainless steel or bimetal pipes. The latter are carbon steel pipes with a corrosion resistant alloy lining. Reeling is a cost effective installation method for small to medium size subsea pipelines, up to 457.2 mm (18 in.) in diameter. However, plastic straining associated with reeling has an effect on weld defect acceptance criteria. The maximum acceptable defect sizes are typically developed using engineering critical assessment (ECA), based on the reference stress method, which requires that the weld metal is equal to or stronger than the parent metal in terms of the stress–strain curve. However, evenmatch/overmatch cannot always be achieved in the case of subsea stainless or bimetal pipelines. In this work, a parametric finite-element (FE) study was performed to assess the effect of weld metal undermatch on the crack driving force, expressed in terms of the crack tip opening displacement (CTOD). Subsequently, the fracture assessment methodology for reeled pipes was proposed, where the ECA as per BS7910 is first carried out. These acceptable defect sizes are then reduced, using an analytical formula developed in this work, to account for weld undermatch. [DOI: 10.1115/1.4029190]

High Temperature, Low Cycle Fatigue Characterization of P91 Weld and Heat Affected Zone Material

The high temperature low cycle fatigue behavior of P91 weld metal (WM) and weld joints (cross-weld) is presented. Strain-controlled tests have been carried out at 400 °C and 500 °C. The cyclic behavior of the weld material (WM) and cross-weld (CW) specimens are compared with previously published base material (BM) tests. The weld material is shown to give a significantly harder and stiffer stress–strain response than both the base material and the cross-weld material. The cross-weld tests exhibited a cyclic stress–strain response, which was similar to that of the base material. All specimen types exhibited cyclic softening but the degree of softening exhibited by the cross-weld specimens was lower than that of the base material and all-weld tests. Finite element models of the base metal, weld metal and cross-weld test specimens are developed and employed for identification of the cyclic viscoplasticity material parameters. Heat affected zone (HAZ) cracking was observed for the cross-weld tests.

Effect of residual stress on high temperature deformation in a weld stainless steel

This paper considers the measurement of residual stresses induced by mechanical loading in a weld Type 347 stainless steel. The work is based in part on an ongoing Round Robin collaborative effort by the Versailles Agreement on Materials and Standards, Technical Working Area 31, (VAMAS TWA 31) working on ‘Crack Growth of Components Containing Residual Stresses’. The specific objective of the work at Imperial College London and HMI, Berlin is to examine how residual stresses and prior straining and subsequent relaxation at high temperature contribute to creep crack initiation and growth for steels relevant to power plant applications. Tensile residual stresses have been introduced in the weld by pre-compression and neutron diffraction measurements have been carried out before and after stress relaxation at 650 oC. Significant relaxation of the residual stresses has been observed, in agreement with earlier work on a stainless steel. Preliminary results suggest that the strains local to the crack drop by over 60% after 1000 h relaxation at 650 oC for the weld steel. The results have been compared with finite element studies of elastic-plastic pre-compression and stress relaxation due to creep.

Cyclic Viscoplasticity Testing and Modeling of a Service-Aged P91 Steel

A service-aged P91 steel was used to perform an experimental program of cyclic mechanical testing in the temperature range of 400 °C–600 °C, under isothermal conditions, using both saw-tooth and dwell (inclusion of a constant strain dwell period at the maximum (tensile) strain within the cycle) waveforms. The results of this testing were used to identify the material constants for a modified Chaboche, unified viscoplasticity model, which can deal with rate-dependant cyclic effects, such as combined isotropic and kinematic hardening, and time-dependent effects, such as creep, associated with viscoplasticity. The model has been modified in order that the two-stage (nonlinear primary and linear secondary) softening which occurs within the cyclic response of the service-aged P91 material is accounted for and accurately predicted. The characterization of the cyclic viscoplasticity behavior of the service-aged P91 material at 500 °C is presented and compared to experimental stress–strain loops, cyclic softening and creep relaxation, obtained from the cyclic isothermal tests.

A comparison of measurement and modelling of plastically induced residual stresses in a 316H and a weld 347 stainless steel

Neutron diffraction measurements on two types of stainless steel have been carried out on Compact Tension (CT) specimens containing plastically induced residual stresses at the blunt notch root. The materials were a type 316H stainless steel parent material and a type 347 stainless steel weld material. The former exhibited a high creep ductility of ∼25% and the latter exhibited brittle behaviour under operating conditions with less than 10% creep ductility. The work is based in part on an ongoing collaborative effort by the Versailles Agreement on Materials and Standards, Technical Working Area, VAMAS TWA 31 Committee working on ‘Crack Growth of Components Containing Residual Stresses’. The objective of this paper is to examine how residual stresses and/or prior straining and subsequent relaxation at high temperature (550 °C for 316H and 650 °C for 347 weld) contribute to creep crack initiation and growth in the two steels. Elastic/plastic/creep finite-element results and neutron diffraction measurements are presented for the CT specimens before and after elevated temperature exposure. The results suggest that the mechanical induced normalised stresses and strains profiles ahead of the crack tip are insensitive to material, however the relaxation response of the materials appear to be dependent on the creep behaviour and ductility. Localised cracking in the plastically deformed material has been observed in both materials due to the redistribution of the residual stress field and associated creep deformation at elevated temperature.Copyright

Predicting the Effect of Compressive and Tensile Residual Stresses in Fracture Mechanics Specimens

This paper considers the prediction of the effects of tensile and compressive residual stress in fracture mechanics specimens by the application of a mechanical pre-load. This is considered in the context of a ‘C’ shape specimen which is mechanically pre-tensioned or pre-compressed to produce, respectively, a compressive or tensile residual stress in the region where the crack is introduced. Finite-element analysis is performed to simulate the pre-loading and the subsequent fracture loading of the cracked specimens. The finite-element predictions are compared with experimental data including residual stress measurements using neutron diffraction. A discussion is presented on modelling and material issues pertaining to the use of mechanical pre-loading as a means for introducing residual stress.