The effect of the stress ratio on the near-threshold growth of delamination fatigue cracks was investigated with unidirectional laminates made from ICI APC-2 thermoplastic prepregs (AS4/PEEK). Tests were carried out under mode I opening loading by using double cantilever beam specimens with a special loading device in which the geometrical non-linearity is negligible for the calculation of the energy release rate. The crack growth rate under different stress ratios was a unique function of the equivalent stress intensity range proposed by the authors. The fatigue crack growth rate near the threshold was controlled by the stress range rather than by the maximum stress. The fatigue crack growth resistance of APC-2 laminates was much higher than that of conventional CF/epoxy laminates even near the threshold region. The increase in resistance of fatigue crack growth, however, was smaller than that in fracture toughness. The difference of the growth behavior between CF/PEEK and CF/epoxy laminates was discussed on the basis of the fractographic observation and mechanism consideration.

View full abstract

The effect of a water environment on the growth of delamination fatigue cracks was investigated with unidirectional laminates made from ICI APC-2 thermoplastic prepregs (AS4/PEEK). Tests were carried out under mode I opening loading by using double cantilever beam specimens. The testing environment was water at 23°C and 50°C. Moisture conditioned specimens were also tested in water at 23°C. The water environment did not induce large detrimental effects. The crack growth rate was expressed as two power functions of the equivalent stress intensity range. The exponent of the power function in the low growth rate region was smaller than that in the high growth rate region. Lowering the frequency of stress cycling yielded accelerated crack growth in water, suggesting the time-dependent mechanism of crack growth. The difference of the environmental effects between CF/PEEK and CF/epoxy laminates was discussed on the basis of the fractographic observation and mechanism consideration.

View full abstract

An experimental investigation was conducted to estimate the residual compressive strength of the laminates of carbon fiber reinforced epoxy (CFRP), aramid fiber reinforced epoxy (AFRP) and their hybrid FRP. These were damaged by the penetrative impact under the axial static load, P. After the impact, the specimens were inspected visually and ultrasonically. Both of the velocity of the impact loading and the axial static load give the remarkable effects on the damage (delamination) of the laminate. The residual strength of CFRP is higher than it of the others. However, in the case of CFRP, the reduction ratio of the strength by the impact damaga is highest

View full abstract

It has already been reported by the authors that two "Brinell" indentations facing each other near the tip of a crack are especially effective in retardation of crack propagation. However, it is not easy to form these indentations precisely. Accordingly, this method does not seem to be convenient for application. In this paper, the effect of an indentation pressed at the tip of a crack on the fatigue strength of plate stecimens with a pre-crack is investigated. The results show that the fatigue limit and fatigue lives strongly increase upon application of an indentation. The effect of the indentation can be estimated approximately from the value of residual stress produced by an indentation.

View full abstract

Rotating bending fatigue tests were carried out using electropolished specimens with a small transverse hole in a low-carbon steel and an aluminum alloy. The change in surface state near the hole edge was successively observed by a metallurgical microscope using the replica method, and the difference in crack initiation process between a low-carbon steel and an aluminum alloy was investigated. The main results obtained are as follows. 1) The difference in crack initiation process between them is essentially recognized in the case in which a stress gradient exists, similarly as in the case of plain specimens. That is, in the case of a low-carbon steel, fatigue damage is accumulated at a finite region the size of a grain and the region becoms finally a fatigue crack almost at the same time. On the other hand, in the case of an aluminum alloy, the fatigue crack initiates in a fairly small region and gradually grows with the number of cycles. 2) Notch sensitivity of an aluminum alloy as for crack initiation is higher than that of a low-carbon steel. This is due to the difference in the sizes of the region concerned with crack initiation.

View full abstract

Effective stress intensity factor range ΔK_eff and the J integral range ΔJ which are determined by the crack tip opening or closing point are considered as the likely parameters controling crack propagation behavior. Therefore, a development of the simple measurement method of crack tip opening and closing is considered to be important for the evaluation of crack propagation. In this study, for an elastic-plastic condition in a center-cracked plate, it is proposed that the crack tip opening and closing point can be simply determined with the σ-(ε_y-α)/(ε_x+α) diagram and σ-ε_y diagram, where σ is the cyclic stress, ε_y and ε_x are the strains measured in the vicinity of the crack center by means of a strain gauge and a is a constant value. Then this measurement method is compared with results obtained by finite element method calculations.

View full abstract

Recently, generators and moters undergo higher rotating velocity and frequent stop and start. As the result, the roter receives an alternating high centrifugal force. In this paper, the fatigue crack propagation behavior of S14 silicon steel sheet was investigated, because it is usually used as the iron core of roter. The following results are obtained: Under the mode I condition, no effect of anisotropy appears to the fatigue crack propagation behavior. On the other hand, the effect of anisotropy appears under the mode I and II complex conditions, and it becomes more remarkable as the increase of the mode II component. The reason seems to be based on the discrepancy of the real crack propagation direction and the weakest direction to the fatigue crack propagation resistance which is caused by roll.

View full abstract

Fatigue crack propagation behaviors and crack length distribution on the surfaces of smooth specimens under completely reversed bending were investigated for a ferritized spheroidal graphite cast iron and silicon steel plate. The variations of fatigue crack propagation and the crack density on the ferritized spheroidal graphite cast iron were much higher than those on the silicon steel plate because of the presence of many graphite nodules. The cumulative probability of crack length distribution in both ferritized spheroidal graphite cast iron and silicon steel plate could be presented by Weibull distributions with a function of crack length. Moreover, the cumulative probability of crack length distribution showed a knee on the ferritized spheroidal graphite cast iron in the early stage of fatigue life. It became clear that the crack length on this knee and a minimum coalescence crack length were almost equal numerically.

View full abstract

The dependence of WC grain size and Co content of WC-Co cemented carbides on compressive fatigue strength and monotonic compressive strength was investigated using cylindrical specimens. WC grain sizes from 0.6 to 6μm and Co content from 5 to 20wt% were tested. The S-N diagrams of the range from 10^-1 to 10⁷ cycles were obtained, and they had two straight lines: i.e., a gentle slope in the low-cycle area and a steep slope in the high-cycle area. The cemented carbides with a WC grain size of 0.6μm and a Co content of 5wt% had the steeper slope. The fatigue strength at 10⁶ cycles increased with decreasing WC grain size and Co content until 10wt%. The strength for a Co content of 5wt% markedly decreased as compared to that of 10wt%. These fatigue properties were discussed from the stand point of the fracture morphology, ductility and crack extension resistance of the cemented carbides.

View full abstract

In order to clarify the effect of small prestrain on the fatigue strength of carbon steel, fatigue behavior of plain specimens of eutectoid steel subjected to tensile prestrains is investigated, and compared with that of annealed or strain-aged specimens. It is found that small prestrain of a few percent does not decrease either the fatigue limit or the crack initiation and propagation lives. Fatigue slip initiation, however, is accelerated by the small prestrain. It is also found that no cyclic strain aging occurs. From these results, it is concluded that dislocations are unpinned by the small prestrain and this causes fatigue slip to occur easily, but the fatigue behavior after the slip initiation is not affected because spheroidal cementite acts as obstacles to further slip deformation and the stage I crack propagation.

View full abstract

Metallurgical discontinuities are dominant in the fatigue strength reductions at the welded joints of vessels whose surfaces could be finished. In the welded joints of SUS 304 with TYPE 308 weld metal fatigue strength reductions are caused by strain concentrations as the result of the softening of the weld metal. A combination model of two elastic fully plastic materials is applicable to the structures under thermal stresses where displacements are self-controlled. Metallurgical discontinuities are represented by the difference of the yield strength. The longitudinal welded joint of a large FBR main vessel in the vicinity of the sodium surface was analysed using this model under various conditions related to the design. Strain concentrations at the welded joint could be evaluated using the elastic follow-up model. The maximum value of the elastic follow-up parameter was 3.0 if the yield stress ratio of the weld metal to the base metal was not less than 0.8.

View full abstract

This paper describes an application of statically indeterminate elastic-plastic fracture mechanics to a single-edge cracked member subjected to combined tension and bending which is typical of structures in fracture mechanics analysis. The author extend statically indeterminate fracture mechanics to elastic-plastic fracture mechanics methodology, because its application has been limited to elastic fracture mechanics problems. The proposed method enables us to evaluate the J-integral value of the cracked member under displacement controlled condition.

View full abstract

The J-integral-based technique for measuring strain-softening curves is applied to determine the fracture toughness of granite. The measuring method has been recently proposed by Li for concrete. This method provides information on the fracture process zone formation and can be used for fracture toughness determination in strain-softening materials. CT specimens of three different sizes, 1.5inch, 2.5inch and 6inch CT were tested to examine the specimen size effect of measured strain-softening curves. Strain-softening curves measured on 2.5inch CT specimens are shown to be in good agreement with that determined from 6 inch CT specimens, indicating the possibility of determining the valid fracture toughness value by means of the J-integral-based technique. The test results for 1.5inch CT specimens, however, yield a lower fracture toughness value than that obtained on the other specimen sizes, suggesting the need of detailed investigations on specimen size requirements for the reliable use of subsized specimens. Finally, the strain-softening curves obtained by the J-based technique are compared with existing data determined by uniaxial tensile tests on several types of granite.

View full abstract

Since the tensile fracture is governed by the load rate, the unstable crack fracture must be observed under the condition in which the load rate is smaller than crack speed. In this paper, the time-dependent stress caused by the compression wave emanated from the rupture section of a notched round bar is observed by the strain gauge method, while the problem of the same condition is theoretically analysed by postulating the crack growth models. The crack speed is determined from a comparison of the experimental result with theoretical result. Furthermore, the present paper experimentally distinguishes the effect of various factors, load rate and stress intensity factor, strain hardening, and area contraction of the material on the crack speed in tensile fracture, and suggests the experimental condition in observing the unstable crack fracture by the tensile test.

View full abstract

The mechanical property of cellulose acetate in the range of room temperature is affected remarkably by the strain rate and temperature. In this paper, the effect of strain rate and temperature on the breaking strength and the fracture surface of notched cellulose acetate plates are examined. As a result, the breaking strength of notched cellulose acetate plates increased with increased strain rate and decreased with increased temperature. The breaking strength is approximated by the power law incorporating the strain rate and the temperature. The equation proposed gives a good agreement with the experimental results, so the breaking strength of notched cellulose acetate plates could be predicted. The observation of the fracture surface shows the fact that the fracture surface is influenced by the strain rate but not by the temperature.

View full abstract

The relationship between ductility at creep rupture and the change in the creep fracture mechanism has been discussed. The Ni-NiO eutectic alloys have considerably greater elongation at failure than pure Ni. A creep rupture strain has been divided into three groups of strains; i. e., strain for primary creep regions, for steady-state creep regions and for tertiary creep regions. Both stress and temperature dependence of the amount of these three strains have been investigated. The strain corresponding to the tertiary creep region increases with increasing stress. Consequently, the rupture strain for both materials increases. The rupture strain for both materials linearly increases with the strain for the steady-state creep region. It is suggested that greater elongation in Ni-NiO alloys may be attributed to the high ductilitiy in the nickel matrix which is produced by the addition of nickel oxide. The change in the fracture mechanism from intergranular creep fracture to transgranular creep fracture results in the increase of strain corresponding to the tertiary creep regime.

View full abstract

The theoretical prediction of the FLD (Forming Limit (Strain) Diagram) or the forming limit strains of sheet metals by the author's method is reviewed giving some new and useful informations on it, in which the J2G (J₂-Gotoh's corner theory; or MG-) constitutive equation previously proposed by the author is used together with the localized necking condition as the failure condition. Theoretical FLD nets of sheet metals subjected to various proportional loadings are illustrated with the n-value or ρ-value as the parameter for the convenience of determining the most appropriate value of ρ which is the material constant involved in the J2G. Further discussions on ρ- and n-values and on the material properties of the pre-strained sheet are given. The secondary FLD and the sheet thickness effect on forming limit strains are also discussed.

View full abstract

The paper describes the experimental results and the theoretical analysis concerning the fractures of plaster hollow cylinders with a transverse hole under compression. This analysis has been carried out with the surface layer theory applied to the stresses around the circular hole in a strip under compression. The analytical results coincide well with the experimental data and the conclusions are as follows: (1)There are two types of fractures, categorized depending on the diameter of the hole; one is the fracture caused by compressive stress and the other is that caused by the tensile stress. (2)The fracture stresses of the hollow cylinders are influenced not only by the diameter of the hole, but also by the ratio of inner to outer diameter of the cylinder. (3) The thickness of the surface layer of the plaster in 0.25mm

View full abstract

A new method is proposed for the estimation of the so-called breakdown pressure in the hydraulic fracturing tectonic stress measurements. The stress field around the well bore was analyzed taking into account the influence of the fluid penetration into the porous rock due to the pressurization of the well bore. By the use of the stress field, the breakdown pressure, i. e., the well bore pressure required to induce cracks on the well bore was estimated based on a newly constructed fracture criterion. The criterion determines the breakdown pressure as a pressure value at which the maximum tensile effective stress at a characteristic depth beyond the well bore surface, reaches the tensile strength of the rock. To verify the new method, laboratory hydraulic fracturing experiments were conducted on cubical rock specimens under uniaxial compression. Results show that the breakdown pressures predicted by the classical method are erroneously lower than the experimental data when the uniaxial compressive stress is large. On the other hand, the breakdown pressures predicted by the present method perfectly agree with the experimental data independent of the magnitude of the uniaxial compressive stress.

View full abstract

The copper electroplating method of stress analysis utilizes the phenomenon that slip bands appear when the maximum shearing stress in the copper plating attains the proper value to the plating which is a function of the number of stress cycles. Accordingly, in the case of combined stress, it is generally impossible to obtain the second principal stress by the copper electroplating method. In this paper, a method that makes it possible to measure the second principal stress at the bottom of a circular groove subjected to bending is proposed, and measuring accuracy of this method is briefly discussed. Based on this method, stress concentration factors of the second principal stress are obtained. The results are in good agreement with previous results, and it is concluded that the method presented here can be used satisfactorily for the measurement of the second principal stress at the bottom of a groove.

View full abstract

A frictionless, axisyrnmetric contact problem of an elastic half-space pressed onto a rigid foundation with a shallow concavity is solved. The problem in transformed into the solution of infinite systems of simultaneous equations using Papkovich-Neuber stress functios in oblate spheroidal coordinates. Numerical results are given for the distributions of the normal displacement in a spherical-ended concavity and the contact stress. The variation of the contact area in the concavity with the applied stress to the elastic half-space in also studied.

View full abstract

In the second-order theory which is one of the analytical methods of three-dimensional elastic problems in finite deformations, the induced second-order scalar term in the stress-strain relations for incompressible materials has two kinds of expressions. These expressions of scalar term in conventional theory include certain unclear contents mathematically and physically. In this paper, we propose a clear expression for the second-order scalar term by using a strain invariant. Further, we show that the second-order term of hydrostatic pressure can be easily determined by means of strain invariant. Finally, the analytical method of boundary-value problems for incompressible materials is examined by two simple examples.

View full abstract

The effect of geometrical uncertainty and thickness variability on the stress distribution of a thin-walled pressure vessel is analysed by the stochastic finite-element method based on the second-order perturbation technique. Numerical analyses are carried out in regard to an axisymmetric model and 3-D shell model with the circular dent due to the welding joint under internal pressure. Moreover, as an attempt to estimate the reliability of a thin-walled pressure vessel, the stochastic safety factor is evaluated with respect to the calculated stress by using the stochastic finite-element method, and it is compared with the conventional safety factor divided minimum strength of material by calculated maximum stress.

View full abstract

The large deflection of the annular plates with simply supported boundary conditions is analyzed by the finite element method using the nonlinear programming theory. Both fixed-hinged and movable-hinged supports are taken into account, while the line loads or the uniform distributed loads are applied. The distributions of the out-of-plane and in-plane displacements and stress distributions in the radial cross section are presented for some load ratios in six cases. It is concluded that the out-of-plane displacements with movable-hinged supports were smaller than those with the fixed-hinged support, and that the shrinkages between the inner and outer edges in the radial direction were almost the same under the two simply supported boundary conditions.

View full abstract

The large deflection analysis of the annular plates with clamped inner edges is investigated using the nonlinear programming technique. The distributions of out-of-plane and in-plane displacements and stresses are shown in the three cases. Then, the coefficients of the design formulae for the large deflection problem are proposed for the ten cases given in the previous and the present reports. The three ratios are called a nonlinear stress reduction coefficient, a nonlinear deflection reduction coefficient and an in-plane displacement coefficient, respectively. By multiplying these coefficients by the linear solutions or the thickness of the plates, we can calculate the maximum stress and displacements. Finally, it is concluded that the reduction of the maximum stress is smaller than that of the maximum deflection when the out of plane displacements are of the same order as the plate thickness.

View full abstract

The large deflection of rectangular plates is investigated by the finite element method using a nonlinear programming method, considering the coupling effect between the bending and the in-plane deformation. The basic equations and the energy gradient used in the calculating process of the Davidon-Fletcher-Powell method are given for a rectangular element. Numerical examples of the rigidly clamped square plates with three aspect ratios subjected to the distributed loads are presented. The distributions of the out-of-plane and in-plane displacements and stresses are obtained for the square plate. It is concluded that the maximum deflections obtained by the present method are slightly larger than those of Way's solutions, and that the maximum stresses obtained by this method show lower values compared with those calculated by the small elastic deformation theory.

View full abstract

An inverse analysis method to obtain the contact stress distributions between two elastic bodies was proposed in the previous paper. In this paper, the boundary element method is applied for the inverse method and by using the method, a practical contact problem, that is, the elastic contact problem of rigid punch and square bar made of silicone rubber, is analyzed. The square bar is put on a sensor for measuring forces. The sensor is made of piezoelectric ceramics and can be assumed to be a rigid body in comparison with the silicone rubber. The forces measured at several points of contact surface between the square bar and the sensor are used as the input data for the inverse analysis method. The estimated contact stress distributions between the rigid punches and the square bar are compared with the corresponding theoretical ones.

View full abstract

In this paper a sensing system measuring semidynamic contact pressure distributions is proposed. The system is the development of the sensing system proposed previously for static contact pressure distributions. The sensing material is made of piezoelectric ceramics and a special electric circuit is designed to measure electric charge in the material. The semidynamic contact pressures are determined using the repeated static measurement at high speed. The sampling time is 25 ms and the measurement is repeated in 255 cycles. Therefore, we can obtained the dynamic behavior of contact pressure distribution for about 6 seconds by using this system. The semidynamic contact pressure distributions for several problems are determined. The obtained dynamical total force variations are compared with that of the corresponding force.

View full abstract

Light intensity equations for holographic photoelasticity, conventional photoelasticity and other holographic or photoelastic method which correlate experimental fringe patterns with distributions of principal stresses, strains and displacements, are described in unified form. A new method for producing theoretical fringe patterns using a personal computer interfaced to a dot printer is also presented. Comparison of theoretical fringe patterns generated by this method with experimental ones facilitates examination of analytical solutions of elastic problems, and it shows that in holographic photoelasticity, when sensitivity g for (σ₁+σ₂) is much greater than sensitivity f for (σ₁-σ₂), the light intensity equation should be simply given by I=(1/2)+(1/2) cos2β to avoid complicated treatment.

View full abstract

Coupling effects between the temperature field and the structural changes are considered in the analysis of the laser transformation hardening process. The volume fractions of the structural components, the latent heat during the phase transformation and structure dependence of the thermophysical properties are introduced during the iterational calculations of the steady-state temperature distribution and the structural changes. The carbon content of a steel, mean value of the grain size of the prior austenite, and the size and the interlamellar spacing of the pearlite colony are input for the initial parameters. Finite-element calculations are carried out to obtain the temperature distribution and the structural changes in the process, and the microhardness distribution in the hardened layer. An example of the numerical computations is shown for a 0.45%C carbon steel. The calculated results are compared with those of the experiments on the laser transformation hardening and found to be in good agreement with them.

View full abstract