As the general features of fretting fatigue, initiation of fretting fatigue crack is in the very early stage of the fretting fatigue life and there are small non-propagating cracks in the test specimen that doesn't fracture at the fretting fatigue limit. In accordance with these experimental facts, fretting fatigue problem can be considered as a propagation problem of small crack. Thus, a pre-cracked specimen was used in the fretting fatigue test in this study. The objective was to consider the determinant factors of fretting fatigue strength. In the fretting fatigue test, the fretting fatigue limit of the pre-cracked specimen was once reduced and after increased with increase of the contact pressure. The reason was understood by the stress intensity factor of the pre-crack obtained by a finite element analysis. In this study, the fretting fatigue limit can be predicted by the comparison of ΔK of the pre-crack and the propagation threshold of the pre-crack ΔK_th. The effect of the relative location of the pre-crack to the contact edge on the fretting fatigue strength was also discussed by both fretting fatigue test and FEM analysis.

View full abstract

Although a lot of interface crack problems were previously treated, few solutions are available under arbitrary material combinations. This paper deals with an edge interface crack in a bonded finite plate and a semi-infinite plate. Then, the effects of material combination on the stress intensity factors are discussed. To obtain the interface stress intensity factor very accurately, a useful method is presented on the basis of the stress values at the crack tip calculated by the finite element method. Then, the stress intensity factors are indicated in charts under arbitrary material combinations for the bonded structure subjected to bending loading conditions and tension. For the edge interface crack, it is found that the dimensionless stress intensity factors are not always finite depending on Dundurs' parameters α and β.

View full abstract

Ultrasonic fatigue tests were carried out for high carbon chromium bearing steel, SUJ2, under two-step variable amplitude stressing at a frequency of 20 kHz in order to investigate the growth of an interior crack. Since it is important to clarify the damage mechanisms under variable amplitude stressing for practical applications, fatigue tests were conducted under repeated two-step variable amplitude stressing. An inclusion was observed at the initiation site of fracture. Optically dark area (ODA) and fish-eye were observed around the inclusion. Sizes of ODA and fish-eye depended on the stress intensity factor range. Interior crack growth rate was slower than the surface one due to vacuum environment. Cumulative damage values calculated by Miner's rule were relevant to the size of inclusions under repeated two-step variable amplitude stressing. Characteristics of interior fatigue crack growth were determined by the fracture surface observations for beach marks and the ODA sizes. Based on the results, the formation mechanism of ODA was discussed in conjunction with the reports by the other researchers.

View full abstract

Health monitoring is important technique in aerospace and other mechanical engineering structures to prevent structures from breaking down by hidden damages. Damages in a structure change its dynamic behavior such as natural frequencies, damping and mode shapes. Use of this phenomenon is effective for global health monitoring of structures. Flexural vibration analysis of damaged beams is made in Fourier transform to discuss a method of damage detection of beam structures with multiple damages. To analyze an inhomogeneous beam due to existing damages, a damaged beam is modeled as a piecewise continuous beam. Solution of dynamic response of the beam is obtained by taking account of continuity of piecewise connected boundaries. The mode shape of the beam is estimated using value of the peak at first response frequency in Fourier transform under a set of positions of impulse load. It is found that positions of damages can be predicted using wavelet transform of difference between mode shapes for damaged beam and undamaged beam. Validity of the method on the damage detection is also illustrated with experimental result.

View full abstract

In this study, Epoxy resin that is a typical thermosetting resin was used as a matrix resin, and 3-phase composites (Epoxy reinforced by carbon nanofiber (CNF) and PAN-based carbon fiber (CF)) were manufactured. By evaluating mechanical properties, the thermal characteristic and the electrical characteristic of each composite, the effect of filler addition was investigated in detail. As a result of 3-phase composite was compared with 2-phase composite (Epoxy reinforced by CNF or CF), thermal conductivity was deteriorated. However tensile strength, Young's modulus and the electrical conductivity were greatly improved. Moreover, to improve CNF/Epoxy matrix adhesion, the oxidation treatment by heating in the air atmosphere was done to CNF. By the oxidation treatment for CNF, the improvement of the Young's modulus and tensile strength was admitted in only 2-phase composite. On the other hand, thermal conductivity and the volume resistivity were inferior compared with untreated CNF.

View full abstract

This paper deals with the problem of dynamic stability of composite laminated cylindrical shells subjected to periodic hydrostatic pressure. First, the axially symmetric motion of the shell is imposed by the periodic hydrostatic pressure. Subsequently, certain perturbations are superimposed on this motion, and their behavior sequential with time is investigated. The symmetric state of motion of the shell is considered to be stable if the perturbations remain bounded. The solutions for the prebuckling motion and the perturbated motion are obtained by the use of Galerkin's method. Calculations are carried out for composite laminated cylindrical shells and the instability regions are determined by utilizing Mathieu's equation. The inevitability of dynamically unstable behavior is proved analytically and the effects of various factors, such as stacking sequences, fundamental natural frequency, driving amplitude of the vibration and dynamic unstable mode, are analytically clarified.

View full abstract

Several attempts have been made to assess the damage induced in materials by crystal orientation distributions identified using electron backscatter diffraction (EBSD). However, the error in crystal orientation measurements makes quantitative evaluation of the local misorientation difficult. Furthermore, the local misorientation depends on distance between the measurement points (step size). For a quantitative assessment of the local misorientation, the error in the crystal orientation measurements must be reduced or the degree of error must be shown quantitatively. In this study, first, the influence of the quality of measurements (accuracy of measurements) and step size on the local misorientation was investigated using stainless steel specimens damaged by tensile deformation or fatigue. By performing the crystal orientation measurements with different conditions, it was shown that the quality of measurement could be represented by the error index, which was previously proposed by the author. Secondly, a filtering process was applied in order to improve the accuracy of crystal orientation measurements and its effect was investigated using the error index. It was revealed that the local misorientations obtained under different measurement conditions could be compared quantitatively only when the error index and the step size were almost or exactly the same. It was also shown that the filtering process could successfully reduce the measurement error and step size dependency of the local misorientations. By applying the filtering process, almost identical local misorientations distributions were obtained under different measurement conditions. Finally, for standardization of the measurement procedure of the local misorientation, present knowledge and points to note about measurements were summarized.

View full abstract

Finite element analysis has been performed based on a mechanical model proposed here to interpret the experimental data of stress-strain curves and thermal shrinkage behavior of poly(ethylene terephthalate) (PET) films observed in the temperature region near the glass transition point. The proposed mechanical model consists of a parallel array of hyper-elastic element and plastic deformation element. The former element was introduced to explain the thermal shrinkage above the glass transition temperature and the latter element was required to explain the permanent strain detected for the oriented PET sample. The numerical calculation was performed to give a good reproduction of the experimental data on the basis of explicit finite element method.

View full abstract

Usage of the conventional bioabsorbable plastic bone fixation devices have been limited, because the mechanical characteristics of them are lower. In this study, poly-L-lactide ( PLA ) which is one of the bioabsorbable plastic was molded into the screw shape and the mechanical characteristics of the PLA screw was investigated. In order to strengthen by orientation of molecular chain, a PLA screw was prepared by extruding a rectangular PLA block followed by forging. The molding conditions were varied to evaluate the formability of PLA. The mechanical properties, such as pulled out strength, shearing strength and the twist strength were measured. The cross-sectional surface of the thread was observed and the crystallinity was measured. Decrease in void in the thread for the screw extruded using lubricants were observed. This results suggested void formation in the thread was caused at higher extrusion ratio. Decrease in shear strength and increase in twist strength were observed for the screw extruded using lubricants, which is due to lower extrusion ratio. Same tendency is observed for the screws molded at higher temperature. These results suggested the lubrication and higher molding temperature have similar effects. Decrease in twist strength at molding temperature of 120°C was obtained, which was caused by embrittlement due to crystallization.

View full abstract

A Three dimensional transparent physical model of a human head with real shape was constructed to visualize and measure relative motion between skull and brain in order to clarify the mechanism of acute subdural hematoma. The model consists of a transparent skull, brain made of silicone gel, cerebrospinal fluid to reconstruct relative motion between brain and skull, and meninges which constraints the motion. The shape was based on real-shaped three dimensional CAD data constructed from CT/MRI images of a specific individual. The results of experiments in the cases of impactor collision to the occipital skull showed that shape of impact force pulse applied to the head was not affected on peak bridging vein stretch ratios because significant relative motion between skull and brain occurred by existence of cerebro spinal fluid.

View full abstract

The nondestructive inspection of a high temperature structure is required to guarantee its safety. However, there are not any useful sensors for high temperature structures. Some of them cannot work at temperatures over 50°C. Another concern is that they are too expensive to use. The sensing system, which can transmit and receive an ultrasonic wave that travels a long distance using a long waveguide, has been studied. We could confirm that an optimal guided ultrasonic wave can travel more than 50cm using a thin 2mm diameter bar as the waveguide. However, we had a difficult problem of receiving a reflected ultrasonic wave from the bottom surface of a test specimen. We tried to improve the trial inspection system using an ultrasonic horn. Finally, an experiment in which the temperature of a test block was heated to about 270°C has been done and the reflected ultrasonic wave from the bottom surface was successfully detected.

View full abstract

The mechanism for a bolt and nut self loosening under repeated bolt axial tensile load has yet to be clarified, despite much investigation into this phenomena. In this paper, the self loosening mechanism is derived from basic strength of materials equations, which results in the following conclusions. (1) When load is applied, a slip occurs on the screw thread surface, and the bolt shank twists clockwise, that is, it descends on the lead angle of the screw thread. At the end of this process, counterclockwise restitution torque T_S1－ is generated by the twisted angle of the bolt shank. However, there is no rotation of the nut. (2) When the load is released, if T_S1－ exceeds the friction torque T_W0 on the nut bearing surface generated by the decreased bolt axial tension, a slip occurs on the nut bearing surface, and the bolt and the mating nut rotate counterclockwise as one unit. To satisfy the relationship T_S1－> T_W0, the maximum bolt axial tension F₁ must be larger than cF₀. Here F₀ is the minimum bolt axial tension and the coefficient c depends on the bolt shape and the friction coefficient. The rotational behavior of the bolt and the nut derived from this analysis concurs with experimental and FEM calculation results of other researchers. For steel joints, it is believed that rotational loosening rarely occurs when there is no separation between joined parts.

View full abstract