The following dissertation details the development of a simple defect-tolerant based model for determining the fatigue properties of a 5383-H321 aluminium alloy and metal inert gas and friction stir welded butt joints in the same alloy. The model considers the fatigue life to consist of three regimes, crack initiation and short and long crack growth, in contrast to the typical defect-tolerant approaches which only consider the long crack growth period. Crack initiation was considered to consist of an initial short crack, present prior to fatigue loading, identical in length to the crack initiator. The short crack growth rate was considered to be a function of the crack length and stress amplitude, whilst the long crack growth regime was described through the Paris equation, dcldN = C(AK)m. The model also considers the effect on the crack initiation and growth rates of the microstructure, macrostructure and residual stresses, with the latter being determined both through the conventional hole-drilling approach and the emerging technique of synchrotron strain scanning. The accuracy of the model was verified through integration of the short and long crack growth regimes, with the predicted lifespan being compared to results obtained from S-N testing of identical welded specimens. A good correlation between the experimental and theoretical results was observed for the parent material and friction stir welded joints for lives in the region 105-106 cycles, although the approach, in its current format, appeared less suitable for determining the fatigue properties of the MIG welded joints. 2

The effect of long term pre-exposure on the corrosion fatigue crack growth rates of low alloy steels has been studied in order to provide guidance on the reliability of short term testing. Tests were conducted in aerated 3.5% NaCl at a potential of -1100 mV(SCE) using a sinusoidal loading wave with a stress ratio of 0.25 and a frequency of 0.167 Hz. In order to calculate the pre-exposure periods for the corrosion fatigue tests, the effective diffusivity of hydrogen in low alloy steels was measured using the electrochemical permeation technique. The diffusivity in BS4360 SOD steel in 3.5% NaCl at a potential of -llOOmV(SCE) was 9.7 x 10 " m V compared to 2.0 x 10 m V in 3.5% Ni-Cr-Mo-V steel. Pre-exposure of AISI 4340 for 21 weeks enhanced the corrosion fatigue crack growth I rates by up to a factor of 7 relative to data from conventional 4 week tests. Pre-exposure of 3.5% Ni-Cr-Mo-V steel for 55 weeks increased the crack growth rates by up to a factor of 4 relative to results from a conventional 4 week test. There was some indication of an effect of pre-exposure on the crack growth rates of BS4360 50D at values of AK below 13 MPam^ but the scatter in the data prevented a definitive conclusion. For certain environmental test conditions, conventional corrosion fatigue tests may seriously underestimate crack growth rates in service. The reliability of the data depends critically on the exposure time. In conditions for which hydrogen uptake is greater at the external surface than at the crack tip, the specimen should be pre-exposed prior to testing to allow a steady-state concentration of hydrogen to develop through the specimen.

A dissertation submitted to the Faculty of Engineering, University of the Witwatersrand, Johannesburg, in fulfillment of the requirements for the degree of Master of science in Engineering. Johannesburg, 1992. / This project involves the comparison of different failure criteria with experimental results for fibre composite materials, and investigates the usefulness of acoustic emission in composite testing. Three sets of specimens were tested to failure in tension. The specimens had various ply orientations and were tested using acoustic emission to determine ply failures. Carbon and glass fibre reinforced epoxy pre-impregnated specimens were used. The testing machine was an ESH testing machine, and a physical Acoustics corporation computer and data acquisition unit were used to record data from a piezo-electric sensor. Suitable failure criteria should be chosen on the basis of ply orientation and material type (eg fibre stiffness), a combination of criteria being used if necessary. Acoustic emission was successfully used to detect ply failure in multi-layered laminates. / AC2017

This manuscript predicts the behavior of a doubly reinforced concrete beam with a circular opening at its midspan by closely analyzing traditional beam theory and design. It then confirms these predictions with finite element modeling software while providing design suggestions. The analysis is limited to the tensile and compressive stresses and cracking behavior. The objectives are to determine the stress distribution around a circular opening that agrees with conventional beam theory. The beam behavior is examined from zero load to failure load. ANSYS is utilized in lieu of real world testing, and the appendix includes the finite element results for a beam including design recommendations. The results lay the foundation for a possible new design procedure of concrete beams with single or multiple circular openings. This research offers useful information that was unavailable previously. More research can be conducted to help designers to design lighter, more efficient concrete beams. / by Heather R. Anesta. / Thesis (M.S.C.S.)--Florida Atlantic University, 2010. / Includes bibliography. / Electronic reproduction. Boca Raton, Fla., 2010. Mode of access: World Wide Web.

Os moinhos de bolas são equipamentos presentes em plantas de mineração, sendo uma máquina importante no circuito de cominuição de minérios. Em função de inúmeros fatores como, por exemplo, projeto, fabricação, sobrecarga no equipamento, falta de manutenção e procedimentos de operação inadequados, são desenvolvidas descontinuidades nos componentes estruturais desse equipamento. Os componentes estruturais dos moinhos, basicamente, corpo, tampas e munhões, além do custo elevado, possuem prazos de fabricação que podem variar de dois a três anos dependendo da demanda do mercado. Portanto é cada vez mais necessário que as descontinuidades detectadas nesses componentes sejam corretamente avaliadas. Neste trabalho analisaram-se pela mecânica da fratura a criticalidade de descontinuidades, tais como trincas constatadas em um munhão de um moinho de bolas e compararam-se os valores teóricos de taxa de propagação (crescimento) dessas descontinuidades com valores reais obtidos por meio de inspeções periódicas realizadas neste componente. A nucleação das trincas foi causada por falta de lubrificação nos mancais do moinho, gerando esforços térmicos circunferenciais no munhão, em que a temperatura estimada do contato munhão e bucha atingiu a faixa de 100 °C a 150 °C. No período analisado, os resultados obtidos por meio da norma BS7910 mostraram-se mais próximos dos valores reais do que a norma ASME Seção XI, Apêndice A. O tamanho da trinca mais crítica foi aceito pelos critérios da norma BS7910 e aprovado apenas na condição de emergência da norma ASME Seção XI, Apêndice A. / The grinding ball mills are equipments present in mining plants, being important in the ore comminution circuits. Depending on numerous factors, such as, for example, design, manufacturing, overloads, poor maintenance and inadequate operating procedures, flaws are developed in the structural components of this equipment. The structural components of a mill, basically, shell, heads and trunnions, besides high costs, have lead times that might vary from two to three years, according to market demand. Therefore, it becomes increasingly necessary that any flaws in those components to be properly evaluated. This paper analyzed the fracture mechanics of the criticality of flaws, such as cracks observed in a ball mill trunnion and compared the theoretical values of growth rate of these defects with actual values obtained through periodic inspections performed in this component. The cracks nucleation was caused by lack of lubrication in the trunnion bearings, generating circumferential thermal stresses, thus the estimated temperature of the trunnion and bushing contact achieved the range 100 °C to 150 °C. During the analyzed period, the results obtained by the standard BS7910 proved to be closest to the actual values than the standard ASME, Section XI, Appendix A. The most critical crack size was accepted by the criteria of the BS7910 and approved only on emergency condition of ASME Section XI, Appendix A.

Este trabalho disserta sobre os métodos usados para determinar a aceitabilidade de defeitos tipo trinca em um bocal de caldeira de geração de vapor. Inicia com uma avaliação sobre as tensões atuantes em bocais e introduz os conceitos de mecânica da fratura para aceitação de defeitos em estruturas. Propõe um método analítico para estimativa de tensões longitudinais em bocais e o compara ao cálculo de tensões por elementos finitos. Explora um caso de um defeito circunferencial e não aflorado em um bocal de uma caldeira real a qual fornece corpos de prova que são submetidos a ensaios de impacto a partir do quais é traçada sua curva de transição ductil-frágil. Analisa os resultados obtidos a partir destes ensaios e os confronta com os valores nominais e de referência. Esta análise embasa discussão sobre probabilidades de falha e fatores de segurança. / This work discusses the methods used to determine the acceptability of crack-like flaws in a boiler nozzle. It begins with an evaluation of the stresses acting on nozzles and introduces the concepts of fracture mechanics for the acceptance of defects in structures. An analytical method for estimating longitudinal stresses in nozzles is proposed and compared with numerical results obtained from the finite element method. It explores a case of a circumferential embedded flaw in a nozzle of a real boiler that provides specimens that are subjected to impact tests from which its ductile-brittle transition curve is drawn. It analyzes the results obtained from these tests and confront them with the nominal and reference values. This analysis bases discussion on probabilities of failure and safety factors.

Results are presented of an investigation into the combined effects of hot water and static uniaxial tensile stress on chopped strand mat reinforced polyester resins. Transport of water has been studied by measuring permeation rates, and the damage has been categorised by optical microscopy. Changes in flexural properties are reported. The resistance to crack propagation of polyester resins in the presence of water has also been studied. Isophthalic-acid based polyester resin/chopped strand mat laminates were prepared by the “contact moulding process”. Rates of permeation of water at a series of temperatures (35-80C) were measured by a gravimetric method for up to 2000 hours. The flow was Fickian in most cases, but there were circumstances where the transport became non-Fickian. The non-Fickian behaviour was observed at higher temperatures on prolonged exposures. A specially designed apparatus was used to stress laminates uniaxially in tension and to measure water permeation rates through the stressed samples. The effect of introducing external stress ([not greater than] 25% of the ultimate tensile strength) was to increase the rate of permeation. The progressive changes in the structures, as observed by microscopy took the form of (i) flat plate like circular disc cracks in the resins and resin rich areas of laminates, (ii) surface cracks in the resin casts and gel coats of laminates and (iv) partial failure of glass monofilaments by buckling. In most cases the damage was confined to regions very near the exposed surface. However, straining to 0.4% elongation led to the onset of debonding in transverse fibre bundles throughout the cross-section. Water damage was accelerated and modified by the presence of external stress. The laminates retained their flexural strength and modulus well. Laminates stressed to 0.3% strain at 80C in water for nearly three months retained 80% of their original strength. No catastrophic failure was observed. Single edge notched polyester resin specimens were exposed to water at three different temperatures for up to three months. The resistance to crack propagation as measured by the critical stress intensity factor was lowered at all exposure temperatures. An examination of the fractured surfaces revealed that no stable crack growth had occurred.

Fatigue tests have been completed using a weldable aluminium alloy NS 8 over a range of alternating and mean stress levels. Specimens have been cut from thick plate material containing either a typical production joint or one of a range of notches. Results have enabled a comparison to be made between the fatigue strengths of welded joints with those for specimen containing stress concentrations. To describe fatigue curves analytically a general empirical fatigue equation has been developed and found to fit results with good accuracy. This equation exhibits the characteristic shape of the S-N diagram and also distributes mean stress curves according to a Goodman (linear) law relationship with the alternating stress at all fatigue lives. An analytical study of random signals has resulted in the development of a general probability density function which is suitable for describing complex random load spectra on a fatigue diagram. This function, together with the empirical fatigue equation and expression obtained from notch specimen results, have been used in a general design computer programme that enables the fatigue life of a component section, of known stress concentration factor, to be predicted for a given complex random load environment. The design programme requires a limited knowledge of the material constant amplitude fatigue behaviour which is obtainable from a small number of fatigue tests. This technique has been confirmed by random load fatigue tests performed on similar specimens.

The problem of crack (tear) propagation in structural nets and fabrics is investigated. A theoretical solution which predicts the crack propagation behaviour under the effect of biaxial stress fields is presented. This was achieved by abstracting the most important interrelations from the Phenomena understudy and expressing them as a partial differential equation, which was solved to give particular solutions corresponding to a given boundary and initial conditions. The proposed expression was modified to account for the behaviour of coated fabrics under a uniaxial stress field. The scope of the investigation included other topics relevant to the crack propagation phenomena such as, the effect of joint slipping on the redistribution of stresses in structural nets. Experimental and analytical work was conducted to evaluate the reliability of the proposed expression and to explore its limitations. The results obtained were found to be in close agreement with the predicted crack propagation behaviour. The excellent correlation between the proposed expression and the different theoretical and experimental work previously published which apparently seemed to be contradictory, established the validity and the reliability of the proposed solution.

In the past, the use of a fracture mechanics approach to describe crack growth in elastomers has been shown to work well for specimens of simple test geometry, simply loaded. This has been the case because elastic strain energy density (e.s.e.d.) functions could reliably be used to calculate both the magnitude of elastic stored energy available to drive a crack and the magnitude of the rate of release of such energy as the crack grows. The aim of this thesis was to investigate the applicability of such a methodology to situations of more complex loading. To this end two novel test-piece geometries were developed. The first consisted of a pure shear geometry with the sample having been pre-strained in the longitudinal direction to varying extents, hence introducing a type of bi-axial deformation. The second consisted of a pure shear geometry test-piece inclined at 30° to the horizontal and loaded in the vertical direction, hence inducing simultaneously pure shear and simple shear loading. Both types of test-piece were used to study the validity of the particular e.s.e.d. functions, the energetics and mechanics of crack growth and crack growth geometries on a macro and micro scale. The constants in particular e.s.e.d. functions were determined by uniaxially deforming in pure shear each of the carbon black reinforced materials used in this study. The resulting functions became progressively less good at predicting the elastic strain energy in the novel geometry test-pieces as the deformation modes became more complex. Anisotropy induced by deforming specimens in one direction was not easily removed even by an imposed large deformation in another direction. Nevertheless, the functions were successfully used to predict crack growth directions in the 30° inclined test-piece. However in the pre-strain pure shear test-pieces the functions significantly underestimated the elastic strain energy. Hence the real energies had to be determined from the forces and extensions measured during cyclic crack growth tests. In these tests crack growth rates for a given tearing energy (elastic energy release rate) increased as the magnitude of the pre-strain increased. This significant weakening was associated with the development of a strain induced molecular and carbon black anisotropy.