Papers by Author: David John Smith

Abstract: This paper presents a novel experiment to quantify both the initial residual stress state in a specimen and its redistribution due to plasticity induced by in-situ loading. The rate of relaxation of the residual stress with respect to permanent deformation is a measure of the elastic follow-up associated with the residual stress field. Residual stress measurements were made using high energy dispersive X-ray diffraction. Digital image correlation, verified by strain gauges, was used to measure full-field deformation on the specimen. The specimen was loaded and unloaded in-situ incrementally to promote plasticity, allowing the relaxation rate of the residual stress to be quantified. An elastic follow-up factor was calculated for the residual stress field, that indicated loading conditions of the residual stress field between fixed-displacement and fixed-load.

Abstract: The role of stress state on the fracture properties of a quasi-brittle material are explored using reactor core Gilsocarbon graphite. The objective of the experiment was to study the initial propagation of cracks. Cruciform specimens have been tested by a biaxial flexural loading method. Pre-slots of 10 mm width and up to a quarter of the depth of the specimen were introduced at the centre of the specimen by electric discharge machining. The slots are located between two through-thickness holes, which are designed to guide crack propagation. A loading jig has been designed and built that allows a range of biaxial loading states to be applied by variation of the length of the loading arms. Clip gauges are used to measure the crack mouth opening displacements. Preliminary tests have studied the fracture of specimens under different loading conditions.

Abstract: A preliminary sensitivity examination of the ductility exhaustion based creep damage prediction model, currently used in the R5 high temperature assessment procedure, showed that material property inputs had significant effects on damage prediction. In the present work, the link between the microstructural factors and the susceptibility to inter-granular high temperature creep failure is considered. The latter was judged to be associated with the low creep ductility. Here, the longitudinal section of a creep specimen and the fracture surface were examined. Auger electron spectroscopy was used to investigate the grain boundary composition in this specimen, which failed after a creep test of 1038h at 550°C under a triaxial stress state. The present results demonstrate that there is a possibility to correlate the susceptibility to high temperature inter-granular fracture from the low temperature fracture investigations. Finally, the susceptibility of the pre-treated 316H stainless steel to inter-granular high temperature failure and the contribution to the creep damage model are briefly discussed.

Abstract: The intention of the present work reported is to provide a detailed characterisation of the factors that contribute to failure of interference fitted joints under combined axial and torsional loading conditions and to provide analytical and finite element based simulations of the slip propagation and interfacial stresses leading to a failure envelope indicating the limits of each loading component in service to avoid premature failure. Small scale tests are conducted to examine failure torque and failure load separately before combined loading regimes are applied. The experimental tests are used to validate each stage of model development and the key parameters such as interference are measured under test conditions for use in the simulation models. The measured probabilistic variation in manufactured specimens will allow parametric studies to be carried out using finite element analysis and ultimately, provide a more comprehensive, validated approach for interference fit design.

Abstract: The main objective of the present study is to validate a simple over-coring deep-hole drilling (oDHD) residual stress measurement technique by utilising finite element simulations of the technique. A number of three dimensional (3D) finite element analyses (FEA) were carried out to explore the influence of material removal and the cutting sequence during the deep-hole drilling (DHD) residual stress measurement process on the initial residual stress field. Two models were considered in the study. First, the residual stress field predicted in a rapid spray water quenched solid cylinder was used as the initial stress field for the DHD FE model. The DHD reconstructed residual stresses were compared with the initial FE predicted stresses. Different cutting sequences and different dimensions were systematically simulated before arriving at an optimum solution for the oDHD technique. The oDHD technique significantly improved the spatial resolution and was applied in a second model consisting of a 40mm thick butt-welded pipe. The DHD reconstructed residual stresses compared very well with the initial FE predicted weld residual stress thereby validating the oDHD technique.

Abstract: Residual stress measurement techniques using mechanical strain relaxation depend on a number of physical quantities and are therefore sensitive to errors associated with the measured data. The resulting stress uncertainties can easily become significant and compromise the usefulness of the results or lead to misinterpretation of the behaviour of the residual stress distributions. It is therefore essential to develop an error analysis procedure for the measurements undertaken. Error analysis procedures for the deep hole drilling (DHD) method are developed to consider triaxial residual stresses. A modified deep hole drilling method, called the incremental deep-hole drilling (iDHD), was applied to measure the near yield residual stress distributions in a cold water quenched aluminium 7010 alloy forged block. The experimental results are used to illustrate the errors.

Abstract: In common with all mechanical strain relief residual stress measurement methods, extra care must be taken when making measurements on components containing highly triaxial residual stress fields which are close to yield. The introduction of a free surface, created as part of the measurement procedure, can lead to plastic redistribution of the residual stress field. Usually, this is not accounted for in the elastic inversion algorithms of the experimental procedure. This paper demon­strates the usefulness and accuracy of deep-hole drilling (DHD) method [1] in a component predicted to contain a triaxial residual stress field. Previous measurements [2] are compared with the results of a DHD simulation on a type 316H stainless steel pipe containing a repair weld offset from an original girth weld. The influence of different material models was also studied.

Abstract: This paper examines the extent to which mechanical shot peening (MSP), ultrasonic impact treatment (UIT) and laser shock peening (LSP) can affect the tensile residual stresses in the fusion zone caused by welding for a 10mm multi-pass 'V' groove weld within a 20 mm thick 304L stainless steel plate. Stresses are measured by deep hole drilling, neutron diffraction and incremental center hole drilling. For the UIT and LSP treated samples, the tensile stresses present in the as-welded plate are reversed to compressive stresses to a depth in excess of 2-4mm. For MSP the affected depth is much less (~0.5mm). The depth of these compressive stresses is similar to those measured in 20 mm thick parent plate test coupons.

Abstract: Using simplified benchmark models, representative of the behavior of real structures, a unified framework for quantification of elastic follow-up (EFU) in structures has been provided. Closed form analytical solutions for evaluation of elastic follow-up are presented for model structures. The impact of elastic follow-up on the relaxation i.e. the redistribution of residual stresses has been explored and hence its significance in the integrity assessment of structures in general and in classification of residual stresses in particular has been highlighted.

Abstract: The formation of fracture process zones in polygranular reactor core moderator graphites subjected to four-point bending has been investigated. The three-dimensional digital image correlation technique has been combined with resistance strain gauge measurements to evaluate, both the localised and the global displacements during testing. The non-linear load-displacement characteristics prior to peak load are correlated with the localised displacements which can extend up to ~3mm (process zone) from the tensile surface of the specimen. At peak load a macro-crack propagates rapidly along an irregular path controlled by the direction of the applied tensile load and the microstructure of the graphite. These cracks arrest prior to complete separation of the specimen. Localised tensile process zones extend for distances of up to ~3mm ahead of the tips of these cracks.