Key Engineering Materials Vols. 321-323

Abstract: The electroacoustic measurement model is an explicit model of an entire ultrasonic measurement system, including the pulser/receiver, cabling, and transducers. A summary is given of the measurement procedures needed for characterizing all the system elements contained in this measurement model, including a description of a new pulse-echo method for obtaining the sensitivity and impedance of an ultrasonic transducer. It is demonstrated that these models and measurement procedures can be combined to accurately simulate the measured voltage of a pulseecho immersion system.

Abstract: Any NDE process may be considered to involve three systems, each having a unique set of parameters that define its characteristics viz. (a) The Input to the material, (b) The material itself, and (c) The output response measured by the NDE system. Traditionally, the input and the material parameters are assumed known and numerous Forward Models have been developed that predict or estimate the output response function. Over the years, forward models are very well established and serve the key purpose, for improved interpretation of the, as well as to optimize the input parameters to obtain the desired, output response. The other two scenarios i.e. if the output response function in the form of measured data is available, to obtain one of system parameters, i.e. either the input function or the material properties, while the other one is assumed to be known are classified as Inverse Problems. Due to the availability of computational resources, the inverse problem solutions are becoming increasingly feasible. Typical applications include measurement of material properties such as modulus, viscosity, temperature, hardness and stress profiles, etc. This paper will discuss the different techniques and the kinds of problems that have been successfully addressed in the area of NDE and their implications on the expanding horizons in NDE.

Abstract: A seamless analysis of IT-related products, such as cellular phone, LCD monitor and note PC etc., which are thin and complex geometry is one of great concerns in product design. However, there is a considerable amount of gap between the industrial application and fundamental academic studies due to a time consuming detailed mesh generation. In order to settle the bottleneck, an auto mesh generation program based on a modified grid-based approach is proposed in this paper. At first, base mesh and skin mesh were generated using informations on entities which extracted from an IGES (Initial Graphics Exchange Specification) file. Secondly, a provisional core mesh with a rough boundary geometry was constructed by superimposing the skin mesh into the base mesh generated from CAD model. And then, positions of boundary nodes were redistributed to delineate exact geometry of the provisional mesh. Finally, good qualified meshes were constructed by moving the positions of the nodes and splitting elements along boundary edges. In conclusion, it is anticipated that the developed program can be used as a promising pre-processor for static or dynamic analysis of various IT-related products.

Abstract: It is not easy to solve the troubles around the openings at the detail design stage, because there are a lot of concerning locations. There are not also clear design rules of classification societies. For more detailed strength analysis around holes on the longitudinal girder or transverse web below deck plate, it is also very difficult to decide the boundary condition (B.C) at both ends. Grillage analysis, which considers B.C and more detailed member arrangement, is used in simple analysis. Furthermore, refined mesh method, which use a part of member in the strength analysis, is used to decide S.I.Fs around holes. Component loads are applied to a refined mesh model separately and S.I.Fs according to component loading are calculated. S/F and B/M are decided from the grillage analysis and S.I.F for a given hole location is deduced from database and the results of grillage analysis.

Abstract: NDT(Non-Destructive Testing) is useful for the defect detection of rolling stock because it can be used to detect defects in many invisible parts. In rolling stock one of the components vulnerable to damage is the axle. Fatigue cracks are initiated in press-fitted parts such as the wheel and brake disk seats of the axle. Those parts suffer from fretting fatigue damage. A precise and reliable NDT technique is necessary to detect cracks in the axle. But conventional induced current potential drop methods (ICPD) have some difficulties to detect such cracks in press-fitted parts of axles. In this study, we have introduced an induced current focusing potential drop method (ICFPD), a new concept that can be applied to the press-fitted parts of the axle. And we have measured electromagnetic properties of the wheelsets and performed finite element analysis for wheelsets with defects. Results suggest that the method is very powerful and effective to detect defects.

Abstract: Predicting the behavior of steel during deformation process under service conditions is one of the main challenges in cold drawing. In this paper, finite element method was used to analyze the distribution of residual stress components in a rod with respect to different area reductions in cold drawing process. Cold drawing brought sunken-in deformation at the bottom land of the rod that was in accord with the result obtained from the analysis using ABAQUS. The results obtained from finite element analysis were in close agreement with the determined high-accuracy measurements.

Abstract: To evaluate the 3-D stress field inside a specimen from displacement data on the free surface obtained from the 2-D intelligent hybrid method, we developed the 3-D local hybrid method based on inverse problem analysis. In a previous study, when a uniform load was applied to a structure with a surface crack, it was demonstrated that the stress field was analyzed with high accuracy. In this study, the 3-D local hybrid method was applied to a structure with a surface crack subjected to bending load. However, a suitable solution was not able to be obtained on a bending problem. Therefore, another method was applied. The relative error between the J integral value of the whole model and the local model was compared, and accuracy was investigated. First, the variation of accuracy with width and thickness was examined. If thickness is increased, the relative error decreases as found in the uniform load case. Moreover, as width increases, the relative error decreases. However, even if width and thickness become large, accuracy does not necessarily become better. Therefore, the relative error was compared and a suitable hybrid size was examined.

Abstract: Displacements obtained experimentally normally are including errors that make it impossible to extract stresses strains with high accuracy using the raw displacement data. In the infinitesimal deformation within the elastic region of a steel or an aluminum alloy, the quantity of displacement is less than 1 pixel, and analysis accuracy deteriorates. We have developed a system which employs the 2-D intelligent hybrid method and can analyze the displacement more exactly in the sub-pixel field. By using the Newton-Raphson method in conjunction with 2nd order deformation gradients, it is demonstrated that this method can accurately extract stress intensity factors from a set of measured displacements.

Abstract: This paper presents limit loads for circumferential cracked pipe bends under in-plane bending, based on detailed three-dimensional finite element limit analyses. FE analyses are performed based on elastic-perfectly-plastic materials and the geometrically linear assumption. Both through-wall cracks and part-through surface cracks (having constant depths) are considered, together with different crack locations (extrados and intrados). Based on the FE results, closed-form approximations are proposed for plastic limit loads of pipe bends. It is found that limit loads of pipe bends are smaller than those of straight pipes, but are close for deep and long cracks.

Abstract: This paper presents a micro-mechanical model of ductile fracture for the API X65 steel, using the Gurson-Tvergaard-Needleman (GTN) model. Experimental tests and FE damage simulations using the GTN model are performed for smooth and notched tensile bars with three different notch radii, from which micromechanical parameters in the GTN model are calibrated. The calibrated micro-mechanical model is applied to quantify pre-strain effects on plastic deformation and fracture of the API X65 steel. Good agreements of the FE damage results with experimental data suggest confidence in the use of the proposed micro-mechanical model to simulate ductile failure of pipelines made of API X65 steels.