The possibility of measuring the magnitude and number of stress cycles of the machine parts subjected to service loading was examined by the fatigue gauge, made of electroplated copper. It is found that the recrystallized grains formed by cyclic stresses in the foil can be utilized as a measure for stress determination under both constant and variable amplitude stresses. This method is shown enough accurate to evaluate constant stress amplitude. In case of variable stress amplitude, the fatigue damage is correlated with that of the constant stress amplitude in terms of a equivalent stress. This equivalent stress is expressed by (Σ σ^α_iN_i/Σ N_i)^1/α, where σ_i, N_i. and α are the stress amplitude, the number of stress cycles of σ_i and the material constant of the gauge, respectively.

View full abstract

A fatigue life prediction procedure was discussed from the viewpoint of the constraint effect of notches. In this report, fatigue crack initiation lives were used to avoid the fatigue crack propagation process and the prediction procedure was examined experimentally. It was fouud that fatigue crack initiation lives depend on the local strain amplitude at the notch root, and the constraint effect of notches to fatigue crack initiation lives may be considered as the multiaxial effect at the notch root and might be evaluated through the equivalent stress concept as used in the multiaxial fatigue problem. The experimental results showed good agreement with the proposed fatigue life prediction procedure.

View full abstract

Torsional fatigue tests were performed on 60 Sn-40 Pb solder under strain control in order to clarify the effects of cycling frequency on the fatigue life of solder. The type of specimens was a thin-walled cylinder. Four cycling frequencies of 0.03, 0.05, 0.5 and 1.0 Hz were used at 303 K. The results are as follows. (1) When the cycling frequency was reduced, the fatigue life was also reduced in general. The fatigue life was, however, increased inversely in the case of the plastic strain range of over about 1.4%. For each cycling frequency, the plastic strain range and the fatigue life could be correlated by the Coffin-Manson law. (2) The prediction formula of the fatigue life was obtained on the basis of an empirical formula on material constants obtained from the tests and the formula for frequency modified fatigue life. The values calculated jn accordance with the prediction formula nearly coincided with the corresponding test results.

View full abstract

An attempt was made to clarify the effect of nitriding on the fatigue strength (N_f=10⁷) of pure titanium through some critical experiments concerning the fatigue properties of gas-nitrided pure titanium including a sample with a small artificial surface pit. The results are summarized as follows : (1) The reduction of fatigue strength occurs in the nitrided pure titanium with increase in the thickness of the hardened layer. (2) The small pit (d_p-80 μm) has little effect on reducing the fatigue strength of nitrided pure titanium, while such a pit gives a large reduction of the fatigue strength of unnitrided pure titanium. (3) Fractography of the fracture surface of nitrided pure titanium revealed that there exists a microfracture facetlike cleavage at the fatigue crack initiation site beneath the nitrided layer, while a small featureless facet associated with stage-I crack growth can be observed at the fracture edge of an unnitrided pure titanium. (4) The fatigue strength of nitrided pure titanium can be considered as a critical stress at which microcracking of the hardened layer occurs under cyclic loading.

View full abstract

The growth behaviour of small fatigue cracks was studied on both fine- and coarse-grained materials of pure titanium under axial loading (R=-1), and the growth behaviour and its statistical properties on coarse-grained material were also investigated under rotating bending and were compared with the results of fine-grained material in a previous report. As the growth rate data are plotted in terms of an effective stress intensity range, small cracks still grow faster than large cracks in a small crack regime. Small cracks of coarse-grained material show higher growth rates than fine-grained material due to a much smaller effect of microstructures such as the grain boundary and crack deflection. Stage-I facets are observed in all specimens and their depths are less than the maximum grain size estimated by the statistics of extreme values. However, the distribution of Stage-I facet depths corresponds to the maximum distribution of grain size of the materials. The growth rates of small cracks follow a log-normal distribution independent of grain size. The coefficient of variation of the crack growth rate in coarse-grained material is less than that in fine-grained material. The coefficient of variation is significantly large at a/d < 3 ( a : crack depth, d :grain size), showing that the relative size of microstructurally small cracks is not dependent on grain size.

View full abstract

A detection of fatigue crack initiation and growth behavior of a blind-holed specimen was carried out by means of the strain interference method. The strains on the center axis of the hole are used for the determination of the strain function 'h'. The crack initiation can be detected using the waveform of the strain function 'h'. The waveform of 'h' of the precracking can be distinguished from that of the postcracking. The penetration of crack depth into the specimen thickness can also be inspected using the waveform of the strain function 'h'. Such variations of the waveform are based on the crack tip opening and closing behavior. Furthermore, the approximate estimation of crack length was carried out using the value of the strain function 'hh' measured at the maximum stress point.

View full abstract

In fracture problems, stress intensity factors obtained theoretically or by experimental analyses have been effectively applied as a quantity of the magnitude of the stress field occurring in the highly stressed region at the vicinity of the crack tip. In actual construction materials, the combined action of stress intensity factors under complex stress conditions often exists in most cases. In this paper, an experimental technique is presented for determining the mixed-mode stress intensity factors K_II and K_III separately for a three-dimensional cracked-body problem. The stress intensity factor K_II is determined by the photoelastic method with an isochromatic fringe loop, and the stress intensity factor K_II, which is often not easy to evaluate with experimental and calculation method, is determined by using the method of reflected caustics in combination with the stress-freezing method. It was found that by using the present experimental technique, the mixed-mode stress intensity factors K_II and K_III and for a three-dimensional crack can be easily separated.

View full abstract

A model for lubricant penetration into cracks under rolling contact was presented. The finite element method was applied to solve the Reynolds equation incorporated with the elastic deformation of the crack. When the lubricant pressure increases at the crack mouth, the pressure propagates toward the crack tip and its distribution becomes constant. The stress intensity factor increases as time passes, and approaches the constant value. The rise time of the stress intensity factor, that is, the time required to penetrate the lubricant, is calculated.

View full abstract

An investigation of the fracture mechanisms of (0/±45/90)_2S aramid fiber/epoxy composies, i.e., Kevlar 49 and HM50 composites, has been carried out, and the delamination characteristics have been quantitatively measured by means of an acoustic microscope. Free-edge delamination occurs initially at -45°/90° and 90°/0° interfaces in static tensile tests, whereas it occurs initially at ±45° interfaces in fatigue tests. Under static tensile loading, resin cracking and interface debonding in the weakest 90° ply are followed by the lowering of the load-carrying capacity of 90° and 45° plies. The final fracture is caused through an increase in 0° ply stress. No difference in the fracture process betwen Kevlar 49 and HM50 composites is observed. Compliance changes in fatigue tests are divided into two stages : stage I where the compliance is almost constant, and stage II where the compliance grows with progression of stress cycles. In stage I, no 0° ply damage occurs, whilst 0° ply damage as well as 0°/45° and 90°/0° delaminations takes place in stage II.

View full abstract

At present, a standard test method for the fracture toughness of high polymer materials has not been fully developed in comparison with that for metallic materials. In this paper, fracture toughness tests using polycarbonate test specimens were carried out. The following conclusions were obtained : (1) In the fatigue precrack introduction, frequency (f) and maximum stress intensity factor (K_fmax) had no influence on the fracture toughness values. (2) The fracture toughness values obtained were almost independent of the crosshead velocity (V), and were completely independent of the load condition P_m/P_Q≤1.1.

View full abstract

The effect of humidity on the fracture toughness of the optical silica fiber was investigated using real fiber specimens 125μm in diameter. Each fiber specimen with a precrack was tested under constant tensile loads at several temperatures and humidities. The stress intensity factor at the onset of final unstable fracture after stable crack growth, i. e., fracture toughness K_IC, was obtained based on the observed crack front on the fractured surface. It was found from this study that the fracture toughness K_IC of the optical glass fiber varied as a function of water-vapor pressure P_H2O in the environment for P_H2O of less than about 3kPa and showed a lower value at the higher lever of P_H2O. For example, K_IC &sime; 0.9MPa√(m) at P_H2O &sime; 0.5kPa and K_IC &sime; 0.7 MPa√(m) at P_H2O &sime; 3kPa.

View full abstract

In order to improve the mechanical reliability of devices in which thin films are used, it is important to investigate the fundamental aspects of the mechanical properties of thin films and also to establish methods for their evaluation. This paper is concerned with indentation by a spherical indenter for quantitative evaluation of interfacial fracture resistance of the film-substrate and fracture strength of a thin film itself. An elastic stress analysis was carried out for indentation by an elastic sphere of a film-substrate system by means of FEM, and whether or not the contact pressure profile can be regarded as elliptical was examined. The influence of elastic properties of the indenter, the film and the substrate on the pressure profile, the maximum pressure and the contact radius was also investigated.

View full abstract

In order to investigate a method for evaluating the fracture strength of a film itself and the interface of a film-substrate system by means of indentation of a sphere, a finite-element analysis was carried out and the stress distribution was obtained. The analysis was made by assuming an elliptically distributed contact pressure. This assumption is almost valid within the range of parameters used in the present analysis. The ratio of the film thickness to the radius of the contact circle and the elastic properties of the film were varied. Dependences of the stresses relevant to microfracture of the film and the interface on the relative thickness and the elastic properties of the film were revealed. The results are useful in predicting and understanding the microfracture behavior of a film-substrate system.

View full abstract

Adhesive/rivet combined bonding has attracted special interest recently as a joining technique of high-strength steel because of its high joint efficiency. In this study, the effects of steel strength on the tensile and fatigue strength of adhesive/rivet combined and adhesive joints were investigated. In addition, the stress distributions of these joints were analyzed by finite-element methods, taking into consideration the plastic deformation of adherend steels. With the increase of steel strength, the tensile strength of combined and adhesive joints increased and tensile strength was improved by the combination with the rivet. However, irrespective of the steel strength, the fatigue strength of combined and adhesive joints was constant and the fatigue strength of the combined joint was similar to that of the adhesive joint. These findings could be explained from the difference of stress distribution between static and fatigue load conditions.

View full abstract

This paper deals with the strength of joints that combine adhesion and tap bolts. When an external load was applied to a combination joint formed after joining two hollow cylinders by adhesion, they were clamped by a tap bolt with an initial clamping force, and the stress distributions in the adhesives and the variation of the bolt axial force were analyzed using an axisymmetric theory of elasticity. In addition, a method to estimate the strength of the combination joint was proposed. Experiments were performed. The analytical results are consistent with the experimental results concerning the variation of the bolt axial force and the joint strength. It is seen that the strength of the combination joint increases as the length of the tap bolt increases. Furthermore, the strength of the T-type butt adhesive joint combined with tap bolts was examined from both analyses and experiments. The characteristics of the combination joint were clarified.

View full abstract

Thermal stress distribution and the strength of the joint due to thermal stress were examined when butt adhesive joints having circular hole defects in an adhesive were in the steady-state temperature distribution. In the analyses, the thermal stress distrbution in the adhesive was given using the two-dimensional theory of elasticity in the case that adherends of similar size and material were kept at constant temperature and side surfaces of the joint were surrounded wich fluid of different temperature. The effects of the location and the size of hole defects on the thermal stress distrbution were clarified by numerical calculations. Then, comparing the principal stress at the periphery of a hole with that at the interface between an adherend and an adhesive, the strength of the joint due to thermal stress was discussed. For verification, photoelastic experiments were performed, and the analytical result was found to be in fairly good agreement with the experimental one.

View full abstract

The effects of the composition gradient in the radial direction on thermal stress have been analyzed for thick-walled tubes of functionally gradient material (FGM) under uniform thermal loading. The combinations of aluminum and silicon carbide were assumed to be the same as the plaster/corundum model. The distributions of the stress and strain components depended mainly on the composition gradient. Its qualitative estimation seems to make it possible to consider the variations in the linear expansion coefficient and elastic modulus. The circumferential stress at the inner surface varied from tension to compression according to the increase in composition gradient accompanied by a decrease in the linear expansion coefficient and an increase in the elastic modulus. For generalized plane strain conditions (axial strain, ε_z=constant ≠0), circumferential stress and radial stress are independent of axial strain. The optimum composition gradient generated by compressive circumferential stress at the inner surface is estimated.

View full abstract

Residual strengths after repeated thermal shocks were studied using Al₂O₃ ceramics specimens. Thermal shocks were repeated by the water quench method under two temperature difference conditions less than the critical temperature difference ΔT_C at which the residual strength degradation was recognized after a single thermal shock. Residual strengths were significantly reduced by repetition of thermal shock and show wide scattering. After a relatively large thermal shock was repeated, the residual strentgth of the specimen showed a minimum value which was lower than that by single thermal shock. The results were discussed by means of the stable and the unstable crack growth behavior based on the thermal shock fracture theory proposed by Hasselman.

View full abstract

The θ projection method proposed by Evans et al. can describe a creep curve even in the tertiary stage. It can calculate the creep rate at all creep strains, and also provides knowledge of long-term creep behaviour. In this paper, it was examined whether the θ projection method can be applied to the results obtained from constant-load creep tests on SUS 304 stainless steel. As each θ value varied with the cutoff strain in a creep curve, the scaled standard error Ses was defined to determine the values systematically. Consequently, creep curves could be drawn to the extent of the tertiary stage. The four parameters in the θ projection method and the rupture strain were represented as a function of testing temperature and initial stress. Creep properties, e.g. minimum creep rate/creep rupture time for longer lives, were easily extrapolated. Furthermore, estimation curves for the creep rupture time showed good agreement with the experimental data up to 10 times the longest rupture data.

View full abstract

The three-dimensional axisymmetric elastic problems of two materially dissimillar wedges of arbitrary angles that are bonded together along a common edge are treated. The dependence of the order of the singularity in the stress field at the apex on the wedge angles and material constants is considered. The order of the stress singularities is determined by use of local coordinates at the top of the wedge. The expression to determine the order of stress singularities in these problems is obtained.

View full abstract

Creep damage is evaluated considering stress relaxations during steady operations in the design standard developed for the construction of a prototype FBR "Mouju" in Japan, but the creep damage evaluated at structural discontinuity portions is considerably conservative because of limited knowledge about the local stress-strain behaviors in the gross structural discontinuities during stress relaxations. Inelastic analyses of a round bar with a U groove and a stepped one were performed assuming power law materials whose exponent n was assumed to be 5. As a result, the local stress-strain behavior without the gross structural discontinuity is found to be estimated by using an elastic follow-up parameter q_F-K^(n-1)/(n+1), where K is an elastic stress concentration factor and n is an exponent of the power law. The overlap effect of the gross discontinuity is found to be expressed as q_eff= q_n×K^(n-1)/(n+1), where q_n is an elastic follow-up parameter for the gross structural discontinuity.

View full abstract

Microscopic stress, strain and the out-of-surface deformation of the plain woven-fabric-reinforced plastic (PW-RP) lamina were analyzed using our recent 3-D FEM program developed for analysis of PW-RP lamina subjected to a uniform strain (deformation) in an arbitrary direction. The results of analysis showed that the location of the highest value in the maximum principal stress moved slightly depending on the deformation direction, and a minimum surface depression appeared when the lamina was deformed at a direction of 45 degrees. These phenomena were commonly obtained for lamina regardless of the composition of warp and weft in PW-RP.

View full abstract

In finite element analysis using the zooming method, the error of the bending moment at the selected point is expected to be smaller than the sum of two errors. One is the error caused by the mesh diameters. The other is the error of displacement or gradient on a zooming boundary. Using the self-adaptive mesh refinement technique, the former can be made smaller than the allowable error. The latter can be estimated from the results of preceding two analyses. A computer program using this error estimation method was developed and applied to many plate-bending problems with various shapes. The usefullness of this accuracy estimation method was illustrated by these application results.

View full abstract

In order to investigate the output power characteristics of the tilt plate heat engine using the shape memory alloy, a small heat engine model was constructed. The influence of the tilt plate angle, the heating phase angle, the heating angle, the hot water temperature, the wind velocity and the deflection of the shape memory alloy helical spring on the output power of the engine was discussed experimentally. The main results obtained are summarized as follows. (1) The maximum output power is obtained at the heating phase angle of about half of the heating angle. (2) If the hot water temperature is not high, the maximum output power is obtained at the heating angle of about 60°. (3) The maximum output power is obtained at the hot water temperature which is higher than the A_f point of the spring by about 20 K. (4) While the output power increases proportionally to the tilt plate angle, the maximum output power takes an almost constant value at the tilt plate angle of above 60°. (5) While the output power increases with increase in the wind velocity, the increasing rate of the maximum output power decreases at wind velocity above 2 m/s. (6) The maximum output power is obtained at the unit elongation of the spring of about 250%.

View full abstract

In the present paper, a tensile elastic-plastic-viscoplastic constitutive equation is proposed where the concepts of under-stress and over-stress is introduced. Strain-rate-dependence of stress, high strain-rate-dependence of stress, strain-dependence of wave propagation speed and strain-rate-dependence of wave propagation speed are shown. 1-dimensional numerical analysis of a bar subjected to longitudinal impulsive loading is carried out using the constitutive equation proposed. As a result, the existence of the strain plateau in the strain-distance plane, the existence of constant strain and the attenuation in the strain-time plane, strain-dependence of wave propagation speed and strainrate-dependence of wave propagation speed are shown.

View full abstract

The present paper gives a theoretical basis for the stability of solutions in the previous hyperelastic modeling of Green as well as hyperelastic one using rotationless strain proposed by the present author. The rate potential and elasticity tensor relating stress rate to strain rate are derived from the elastic potential. The principle of virtual work and its rate form are obtained by considering internal work due to angular momentum balance, which was neglected in the previous study. Stabilities of displacement and its velocity are examined, and a criterion is proposed. Finally, restrictions on elastic potential deduced from the positive definite rate potential are discussed.

View full abstract

The resonance method is studied for measuring residual stresses in thin plates of aluminum 2024-T361. The resonance frequency is a function of the plate thickness and the elastic wave velocity, both of which vary with the stress because of the Poisson effect and the acoustoelastic effect. PZT elements of the thickness-shear mode are used to obtain the resonance frequencies for the polarizations in the principal stress directions. Their difference, being equivalent to the acoustic birefringence, is found to be linearly dependent on the uniaxial stress up to 300 MPa. The frequency shift induced by mounting the PZT element can be minimized by choosing the mode whose resonance frequency is closest to that of the transducer alone. The two-dimensional stress field in welded plates is then measured by scanning and rotating the transducer. The obtained principal stress difference coincides with the measurements by the conventional sing-around method and the destructive strain gauge method. The accuracy is about ±10 MPa.

View full abstract

In this study, the experiments relating to an efflux by gravity of powder from a cylindrical vessel are carried out by using different particle diameters of silica sand and carborundum (i.e., non-spherical powders). As the results, it is clarified that two experimental equations proposed by Beverloo et al. or Bulsara et al. are established in the cases where particle diameter of powder and orifice diameter are changed systematically.

View full abstract

The push-pull fatigue test and crack propagation tests were performed on aluminum-metal-matrix/continuous Si-Ti-C-O fiber composites. The fatigue behaviour of a smooth cylindrical specimen reinforced with longitudinally directed fibers and fatigue crack growth behaviour of a C.T. specimen reinforced with transversely directed fibers were then examined. The fatigue strength of the composites at the life of N_f=10⁷ cycles was about twelve and one-half times that of the matrix metal. This is slightly higher than that of the static tensile strength ratio of either materials. The crack growth rate, da/dN, plotted against stress intensity factor range, ΔK, forms a linear line on a double logarithmic scale, which depends on stress ratio. The stress-ratio-dependent threshold stress intensity factor range, ΔK_th, was also recognized. The stress-ratio-dependence of the crack growth rate becomes slight in the low stress intensity regime, and remains slight in the high stress intensity regime when da/dN is plotted against the effective stress intensity factor range, ΔK_eff. The energy release rate, ΔK²/E, may be an effective fracture mechanics parameter controlling the crack growth rates of both the composites and matrix metal.

View full abstract