Four Points Bending Tests on Large Scale Welded Pipes Containing a Through-Wall Defect: Fatigue and Fracture Analysis

Within the framework of European project STYLE (Structural integrity for lifetime management), fracture tests on two large scale pipes containing a through wall crack have been performed. Two Mock-ups have been tested: MU1 is a narrow gap Inconel Dissimilar Metals, provided and designed by AREVA France, and MU2 is a an austenitic steel butt-weld with a thermally aged weld repair austenitic weld, provided by EDF British Energy. The four-points bending tests were carried out by the French Alternative Energies and Atomic Energy Commission (CEA), in order to study the mechanical properties and integrity of component such as welding pipes. A through wall crack was machined in the both pipes. After a fatigue pre-cracking step carried out at RT, the monotonic fracture test was performed (at 300°C on MU1). Optical camera and Electrical Potential Drop Method have allowed following the crack growth during fatigue and final fracture stages. The observations made post-mortem showed ductile tearing of a few millimeters in those pipes. The first part of this paper is devoted to the four-points bending tests. The second part of this paper deals with first numerical analysis related to the Mock-up-1. Previous results concerning the mechanical characterizations of the constitutive materials are discussed. Fracture mechanics small scale specimens are interpreted using FE Analysis to obtain the fracture parameters used in global approaches. First computation is shown on the Mock-up-1 in order to predict the behavior of the large scale test mechanical and fracture behavior.

Improved Compliance Solutions for C(T), SE(B) and Clamped SE(T) Specimens Including Side-Grooves, Varying Thicknesses and 3-D Effects

Fracture toughness and Fatigue Crack Growth (FCG) experimental data represent the basis for accurate designs and integrity assessments of components containing crack-like defects. Considering ductile and high toughness structural materials, crack growing curves (e.g. J-R curves) and FCG data (in terms of da/dN vs. ΔK or ΔJ) assumed paramount relevance since characterize, respectively, ductile fracture and cyclic crack growth conditions. In common, these two types of mechanical properties severely depend on real-time and precise crack size estimations during laboratory testing. Optical, electric potential drop or (most commonly) elastic unloading compliance (C) techniques can be employed. In the latter method, crack size estimation derives from C using a dimensionless parameter (μ) which incorporates specimen’s thickness (B), elasticity (E) and compliance itself. Plane stress and plane strain solutions for μ are available in several standards regarding C(T), SE(B) and M(T) specimens, among others. Current challenges include: i) real specimens are in neither plane stress nor plane strain - modulus vary between E (plane stress) and E/(1-ν2) (plane strain), revealing effects of thickness and 3-D configurations; ii) furthermore, side-grooves affect specimen’s stiffness, leading to an “effective thickness”. Previous results from current authors revealed deviations larger than 10% in crack size estimations following existing practices, especially for shallow cracks and side-grooved samples. In addition, compliance solutions for the emerging clamped SE(T) specimens are not yet standardized. As a step in this direction, this work investigates 3-D, thickness and side-groove effects on compliance solutions applicable to C(T), SE(B) and clamped SE(T) specimens. Refined 3-D elastic FE-models provide Load-CMOD evolutions. The analysis matrix includes crack depths between a/W=0.1 and a/W=0.7 and varying thicknesses (W/B = 4, W/B = 2 and W/B = 1). Side-grooves of 5%, 10% and 20% are also considered. The results include compliance solutions incorporating all aforementioned effects to provide accurate crack size estimation during laboratory fracture and FCG testing. All proposals revealed reduced deviations if compared to existing solutions.

The Influence of Multiple Inclusions on the Cauchy Stress of a Spherical Particle-Reinforced Composite Under Uniaxial Loading

This paper investigates the influence of multiple inclusions on the Cauchy stress of a spherical particle-reinforced metal matrix composite (MMC) under uniaxial tensile loading condition. The approach of three-dimensional cubic multi-particle unit cell is used to investigate the 15 non-overlapping identical spherical particles which are randomly distributed in the unit cell. The coordinates of the center of each particle are calculated by using the Random Sequential Adsorption algorithm (RSA) to ensure its periodicity. The models with reinforcement volume fractions of 10%, 15%, 20% and 25% are evaluated by using the finite element method. The behaviour of Cauchy stress for each model is analyzed at a far-field strain of 5%. For each reinforcement volume fraction, four models with different particle spatial distributions are evaluated and averaged to achieve a more accurate result. At the same time, single-particle unit cell and analytical model were developed. The stress-strain curves of multi-particle unit cells are compared with single-particle unit cells and the tangent homogenization model coupled with the Mori-Tanaka method. Only little scatters were found between unit cells with the same particle volume fractions. Multi-particle unit cells predict higher response than single particle unit cells. As the volume fraction of reinforcements increases, the Cauchy stress of MMCs increases.

Enhanced Charge Carrier Concentration of SiC/CNT With N- and P-Type Doping Agents

Single-walled Carbon nanotubes (SWCNTs) have been shown to have excellent conductive properties. SWCNTs were dispersed in a SiC nanoparticle matrix to form a homogeneous mixture that is both mechanically durable and conductive. The SWCNT amount has been varied. SiC/SWCNT mixtures were then doped with various N- and P-type agents, and the resulting samples were analyzed by scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), and X-ray diffraction (XRD). Raman spectra of the samples were also measured for evidence of structural changes. Seebeck coefficients were measured for the doped samples demonstrating the change in thermoelectric properties. Shifts in the G peak (1580.6 cm-1) of the Raman spectra of the samples provides evidence of an increase in charge carrier concentration in the doped samples, correlating well with the Seebeck coefficient results.

The Role of Multiscale Strain Localizations in Fatigue of Magnesium Alloys

The reliable characterization of fatigue behavior and progressive damage of advanced alloys relies on the monitoring and quantification of parameters such as strain localizations as a result of both crystallographic deformation mechanisms and bulk response. To this aim, this article attempts to directly correlate microstructural strain at specific fatigue life to global strain as well as surface roughness in Magnesium alloys. Strain at the grain scale is calculated using Digital Image Correlation (DIC), while surface topography gradients are computed using roughness data at different stages of the fatigue life. The results are further correlated to Electron Back Scatter Diffraction (EBSD) measurements which reveal the profuse and spatially inhomogeneous nature of the crystallographic deformation mechanisms related to yielding and fatigue crack initiation. Emphasis is given on using multimodal NDE data to formulate first a description of the current state of the material subjected to fatigue loading and on identifying conditions that can probabilistically drive the affected by both local and global response, governing degradation process.

Numerical Modeling of Braided Composites Using Energy Method

Based on the braiding process and force analysis of yarn, a mesoscopic numerical modeling approach was established, which divided the modeling process as follows: establishing the control points according to the braiding process, establishing the fixed points during jamming, adjusting the control points after jamming, changing the position of fiber bundle due to the fiber bundle intertwined each other and establishing the fiber bundle trajectory according to the minimum strain energy. In the process of adjusting the intertwined fiber bundle trajectories, the fiber bundle trajectory was scattered. Using extrapolation adjustment method, discrete points of fiber bundle trajectory intertwined were adjusted in turn from the control points to the fixed points. Adjusted discrete points were equivalent at the corresponding location points of the corresponding trajectory, and at the same time, there was non-interference between the fiber bundle trajectories. Using this method, fiber bundle trajectory and cross section of the models of 2-D woven and 3-D four-directional braided composite materials were established, compared with the experiment result, which were consistent with the electronic microscope scan images and calculated woven structure size was in agreement with the measured data. The maximum relative calculation error of braiding bitch of 3-D four-directional braided structure was about 5%, especially braiding angle was 21° or so, the relative calculation error was below 2%. The maximum relative calculation error of surface braiding angle of 3-D four-directional braided structure was about 4%, especially braiding angle was 21° or so, the relative calculation error was below 2.4%. This modeling approach was fundamental for further analysis of the micromechanical strength and life of braided composites, which was applied to aero-engine hot section.

Explicit and Implicit Lifetime Assessment Methods of 9Cr-1Mo Steel Under Combined Creep and Fatigue Loads Using a Strip Yield Model

Growing demand for clean, affordable energy has driven the power industry towards generation plants with higher thermal efficiency and higher operating temperatures. ASTM Grade 91 is a high chromium (9Cr-1Mo) creep resistant steel commonly used in high temperature pressure vessel and piping applications. These service conditions often involve a combination of stationary and cyclic loads at elevated temperatures. Lifecycle assessments of components under such conditions require modeling of both creep and fatigue behaviors. This paper develops two approaches to modeling mixed creep and fatigue crack growth for lifetime assessment of high service temperature components. Both approaches model fatigue crack growth using the Paris law integrated over the number of lifetime cyclic reversals to obtain crack extension. A strip yield model is used to characterize the crack tip stress-strain fields. The first approach employed an explicit method to approximate creep crack growth using C* as a crack tip parameter characterizing creep crack extension. The Norton power law was explicitly solved to model the primary and secondary stages of creep. The second approach used an implicit method to solve a set of constitutive equations based on properties of the material microstructure to model all creep stages. Constitutive equations were fit to experimental data collected at stresses 10–60% of yield and temperatures 550–650°C. These methods were compared to published experimental data under purely stationary loads, purely cyclic loads and mixed loading. Both models showed good agreement with experimental data in the stress and temperature conditions considered.

Energy Absorption of Thin Walled Members Under Axial Compressive Loading

This study was conducted to investigate the effects of cross-sectional geometry on thin wall axial crushing members for the purpose of improved energy absorption. A total of five geometrically equivalent shapes (same wall thickness area, material, and length) were analyzed namely, triangle, rectangle, square, pentagon, and circle. The deformation modes and energy absorption of the members were studied under compressive loads and compared using ABAQUS/Explicit module, finite element analysis software. The simulations revealed that for the five geometrically equivalent cross sections under equal loading conditions, the pentagon shaped member absorbed the highest amount of energy. As compared to baseline rectangle member, the pentagon member absorbed approximately 25–28% more energy.

Viscoelastic Contact of a Half-Plane Sliding Over a Slightly Wavy Rigid Surface

In this paper, the sliding contact of a rigid sinusoid over a viscoelastic halfplane is studied by means of an analytical procedure that reduced the original viscoelastic system to an elastic equivalent one, which has been already solved in [1]. In such a way, the solution of the original viscoelastic contact problem requires just to numerically solve a set of two integral equations. Results show the viscoelasticity influence on the solution by means of a detailed analysis of contact area, pressure and displacement distribution. A particular attention is paid to the transition from full contact to partial contact conditions.