Key Engineering Materials Vols. 525-526

Abstract: Fatigue damage is a localized phenomenon controlled by the near-tip crack behavior. This paper presents an application of a dislocation distribution technique to the simulation of crack tip behavior under fatigue loading. A centre-cracked tension specimen under uni-axial fatigue loading is used in the study. Crack opening and plastic deformation around the crack tip are simulated by distributions of dislocation dipoles in crack plane and four inclined planes ahead of the crack tip. Climb dislocation dipole is used to model the opening and closing of the crack while glide dislocation dipole is used to simulate the backward and forward slip in the inclined planes during loading and unloading of the fatigue cycle. Stress field around the crack tip is obtained by the superposition of the contributions of the applied external load and the distributed dislocation dipoles. Correct boundary conditions of the model are achieved by employing a quadratic programming technique to minimize a properly constructed non-negative object function. It is found that the simulated crack closure variations under the constant amplitude fatigue load agree well with the result of a previously developed modified strip yield model with an appropriate constraint factor.

Abstract: A key factor affecting the use of carbon fibre reinforced composite laminates is the low velocity impact damage which may be introduced accidentally during manufacture, operation or maintenance of the component. Among the several barely visible impact damages, interlaminar delamination is the dominant failure mode and may reduce the post-impact compressive strength of the component significantly. This paper focuses on the study of the delamination threshold load (DTL) above which significant increase of delamination and thus large reduction of the residual compressive strength of the component may occur. Instrumented drop weight tests were carried out under various impact energy levels to determine the delamination threshold load. Efforts are directed to the study of the laminate thickness effect on the reliability of the detection of the DTL. The validity of the concept of DTL has been investigated and possible implications on the measurement of the DTL has been discussed. It is demonstrated that DTL exists but its detection requires proper testing conditions.

Abstract: The observed damage level in field, due to the 1985 earthquakes, is compared in a 10-story RC building in Mexico City with the analytical predicted behavior without and with the available over-resistance effects. The results were compared to those obtained from the conventional seismic analysis and to the observed damage behavior after the earthquake. Facing the observed behavior in many buildings, the necessity of studying in detail the available over-resistance effects merged, in order to widely explain the seismic-resistance behavior participation of such structures. The over-resistance effects are included: slab steel, deformation hardening period of reinforcement steel, average real stresses of concrete and steel, compression slab participation and concrete confinement. Lineal and non-lineal step-by-step dynamic analysis are made. The soil-structure interaction and the P-Δ effects are included in the analysis. The analytical periods are compared to those experimentally obtained. A very good congruency between the analytically predicted behavior and the observed damage level after the earthquake is obtained. The direction and the stories with maximum damages match with the direction and stories with maximum deformations obtained from the analysis. It is noticed that the structure has a superior lateral resistance capacity regarding to the given in the conventional design.

Abstract: A description of stress and displacement fields by means of the Williams power series using also higher-order terms is the focus of this paper. Coefficients of this series are determined via the over-deterministic method from the results of conventional finite element (FE) analysis. A study is conducted into the selection of the FE node set whose results are processed in this regression technique. Coefficients up to the twelfth term were determined with high precision. The effect of the position of the FE node set on the accuracy of the values of the higher-order term coefficients is reported.

Abstract: Laminated composite structures are increasingly finding more applications in various fields thanks to their lower weight if compared with other materials of the same strength. Nevertheless, composites thin plates show a critical behavior in terms of damage propagation mechanisms when subjected to (low velocity) impact. Indeed they tend to produce delaminations which can be hardly detected by optical inspections and can affect the global load carrying capability, leading to a premature structural collapse. The aim of this paper is to assess the capabilities of the Davies-Zhang approach (introduced in 1994 and aimed to the estimation of both the delamination initiation impact load and the size of the impact induced delaminations) by using a multiscale FE model based on the mesh superposition technique. Indeed the impact area has been modeled layer-wise with an element per layer while the rest of the structure has been modeled at laminate level by layered elements by means of a homogenization approach for the determination of the equivalent laminate material properties. The impact induced delamination area has been determined by adopting stress-based criteria. The results (in terms of delamination initiation impact force and delamination size) have been compared to the ones obtained by adopting the Davies-Zhang approach.

Abstract: Barcelona, as well as a large number of cities in the Mediterranean basin, has a housing stock composed of a large number of unreinforced brick masonry buildings. Motivated by different factors, the enlargement of the city (Eixample in Catalan) was held from the second half of the 19th century and the beginning of the 20th, a period in which a large number of buildings of this type were built, many of which are still used as dwellings. Although the buildings were built individually, some of them are linked to adjacent buildings by the side walls. This feature leads to the analysis of the buildings as isolated structures and also as an aggregate. Barcelona is located in a seismic region of low to moderate hazard, with macroseismic intensity between the grades VI and VII of the European macroseismic scale EMS'98. Based on the deterministic and probabilistic response spectra for the different types of soils present in Barcelona obtained in the work of Irizarry (2004), the seismic risk of four individual buildings and an aggregate is evaluated. The buildings are modeled and analyzed using the TREMURI program and MATLAB routines under the guidance of RISK-UE project.

Abstract: Fiber-reinforced composite materials (FRC) are used in a wide range of applications, since FRC exhibits higher strength-to-density ratio in comparison to traditional materials due to long fibers embedded in a matrix material. Failures occurred in FRC components are complicated because of the interaction of the constituents. The aim of this study is to investigate damage behavior in a unidirectional glass fiber-reinforced epoxy on both macro-and micro-levels by using finite element method. The Hashins criterion was applied to define the onset of macroscopic damage. The progression of the macroscopic damage was described using the Matzenmiller-Lubliner-Taylor model that was based on fracture energy dissipation of material. To examine the microscopic failure FE representative volume elements consisting of the glass fibers surrounded by epoxy matrix with defined volume fraction was considered. Elastic-brittle isotropic behaviour and the Coulomb-Mohr criterion were applied for both fiber and epoxy. The results of the macroscopic and microscopic analyses were correlated. As a result, damage initiation and damage development for the investigated FRC could be predicted.

Abstract: The assessment of conditions of crack initiation in a tip of a bi-material notch composed of two orthotropic materials is dealt. The assessment of the bi-material orthotropic notch stability criteria based on standard linear elastic fracture mechanics can lead to incorrect results due to a change of fracture mechanics properties. The change of the fracture mechanics properties are taken into account in the discussed stability criterion. It is shown that the criterion of this kind can qualitatively and quantitatively influence the results, and it contributes to more reliable assessment of components with geometrical and/or material discontinuity.

Abstract: The differences in fracture behavior between the compact tension C(T) and the middle tensile M(T) specimens make structure integrity assessment uncertain. Two different types of specimens C(T) and M(T) specimens made from stainless steel have been used for fracture toughness testing at the room temperature by the principles of the ASTM 1820-05 standard procedure. Stable crack initiation and crack propagation occurred for the C(T) specimens at lower values of crack driving force than for the M(T) specimens. Crack tip opening displacement-CTOD has been directly measured on the surface of specimens by using a stereo-optical grading method. The critical crack tip opening displacement at crack initiation CTODi has been measured as a plastic Stretch Zone Width (SZW) during a post test fractographic inspection. Comparison between the CTOD-R curves of both types of specimens shows some difference between the C(T) and the M(T) specimens, but a more significant difference appeared in the crack driving force, as consequence of different constraint (triaxiality) of the C(T) versus the M(T) specimens. Therefore, the result obtained by test on laboratory C(T) specimens cannot be directly used as fracture toughness material properties in a structure integrity assessment, except as a conservative lower bound estimate.

Abstract: This paper analyzes the propagation of fatigue cracks in pearlitic steel in two forms, hot rolled bar and cold drawn wire. The experimental procedure consisted of fatigue tests on bars under tensile loading, using steps with decreasing amplitude of stress and constant stress range during each step. The curves plotting cyclic crack growth rate versus stress intensity factor range show a main steady-state regime preceded by transient paths. The steady-state regime is associated with the curves of the Paris regime. The cold drawing process improves the fatigue behaviour of steel by retarding the cyclic crack growth rate, and the propagation rate is not dependent on the R-ratio. The transient branches allow one to calculate the plastic zone size, considering that they are a consequence of the overload retardation effect at each step change, and a unique expression is fitted as a function of K_maxΔK product and of the conventional mechanical properties.