The influence of defect shape on the fatigue limit of tempered martensitic steel was investigated using specimens containing either a drilled hole or an electrical discharge-machined (EDM) notch. Fatigue tests confirmed an unconventional defect-shape dependence. At higher stress amplitudes, the EDM-notched specimens exhibited lower fatigue strength, whereas at lower stress amplitudes, the difference between the two defect types diminished, and the drilled-hole specimens ultimately showed a lower fatigue limit. Detailed observations near the defects showed discontinuous crack propagation involving temporary crack arrest and coalescence of sub-cracks, which is characteristic of a damage accumulation mode. Fractographic and crystallographic analyses indicated that the damage accumulation may be distributed more widely around the drilled hole than the EDM notch. These findings suggest that the unconventional defect-shape dependence may emerge at the fatigue limit due to a transition in crack propagation mode from conventional continuous propagation to a discontinuous mode governed by damage accumulation.

The Al Alarea parameter model is widely used for predicting the fatigue limit of materials containing small defects based on the assumption that a small defect can be regarded as a crack. Although the model was successfully applied to various materials, its applicability to high-strength steel requires further validation. In this study, the fatigue limit was evaluated using specimens of vacuum-quenched and tempered martensitic steel in which a drill hole, an electric discharge machined (EDM) notch, and a pre-crack were introduced. The fatigue limit of the specimen with the pre-crack was consistent with the prediction of the Al Alarea parameter model, whereas the specimens with the drilled hole and EDM notch exhibited higher fatigue limits, indicating that these defects could not be regarded as cracks in the fatigue limit evaluation. Fracture surface observations confirmed that the fatigue limits were determined by the crack non-propagation limit rather than the crack initiation limit. Furthermore, finite element analysis indicated that differences in defect-induced stress fields influenced the fatigue crack propagation, leading to deviations in the fatigue limits. These findings contribute to an accurate estimation of the fatigue limit of high-strength steels.

This study conducts compressive fatigue tests with an extended notch on a strain-localized material for quantitative evaluation of damage during compression fatigue and the corresponding effect of loading history on subsequent tensile fatigue limits. Hence, fatigue crack "growth" and "propagation" of two types are found in damage accumulation (DA) mode. The former features several simultaneous multi-crack initiations and independent extensions. Contrarily, the latter features coalescence between the main and secondary cracks. Moreover, the near-crack-tip mechanics causing crack extension and non-propagation in the respective fatigue crack extension types are discussed. Furthermore, a method for subsequent tensile fatigue limit prediction considering the compression fatigue effect is proposed by studying the non-propagating crack length, Vickers hardness, and residual stress in the DA mode during compression fatigue, corresponding to Murakami-Endo's equation parameters for a mechanically small crack. Thus, this study is anticipated to hold great significance for understanding fatigue damage caused by different load blocks and improving Miner's rule.

The fatigue properties of martensitic steels are highly microstructure-dependent, preventing the use of fatigue design criteria for ordinary steels, thus hindering their widespread application. This study focused on fatigue crack extension mode-type and, proposed a material index representing the steels and a new evaluation of fatigue strength. For this purpose, rotating bending fatigue tests were conducted on three types of 18% Ni bcc martensitic steels with different carbon contents. The fatigue crack extension behavior was analyzed by replica observation on the specimen surface, and the relationship among the local plastic strain, microstructure, and crack initiation/extension was analyzed by electron backscatter diffraction (EBSD) on fatigue-interrupted specimens. Moreover, fractographic observation analyzed the fatigue crack extension mode-type transition. As a result, the damage-accumulation mode of fatigue crack growth type (DA-FCG) and the DA mode of fatigue crack propagation type (DA-FCP) appeared as new fatigue extension mode types, with DA-FCP dominating the fatigue limit. In this type, the block size in the microstructure is proposed as the material index of martensitic steel, and the fatigue limit equation is presented in conjunction with the material index.

The fatigue characteristics of a strain-localized material specimen with an inclined notch, subjected to cyclic tension-compression under plane strain conditions, have not been fully elucidated. An unloading elastic compliance test and electron backscattering diffraction analyses were conducted near a fatigue crack tip on inclined notched specimens of JIS-SM490YB hot-rolled steel. Crack extension caused a characteristic change in the local mean strain from zero to a positive value, corresponding to the transition from local plasticity induced by a notch to that induced by a fatigue crack. The sudden increase in local elastic compliance and negative crack opening load were the mechanical indicators of the damage accumulation mode of the fatigue crack extension mode-type.

The fatigue properties of martensitic steels tend to differ significantly from those of low- and medium-strength steels. Even if the hardness is the same, the fatigue strength varies substantially depending on the microstructure. Thus, for using martensitic steel rationally and developing materials with high fatigue strength, it is desirable to establish a fatigue strength evaluation method that can consider the microstructural effects on fatigue properties. For this purpose, 18% Ni bcc martensitic steel was used as a model metal to conduct rotating-bending fatigue tests in this study. The fatigue crack behavior on the smooth specimen surface was examined, and the fracture surface morphology on both sides of the broken specimen and the microplastic strain distribution near the fatigue crack tip were analyzed. The results clarified the mechanism underlying fatigue crack extension (FCE). The analytical results revealed three types of FCE mechanisms dependent on the stress amplitude, stress state (i.e., plane stress or plane strain), and martensitic microstructure. Considering these mechanisms, a unique stress-life curve was predicted, and a particular fatigue crack shape and fractography were confirmed. The material indices were speculated to be the representative properties of martensitic steels.

This study aims to evaluate shear-mode crack propagation rates under rolling contact fatigue (RCF) in railway wheel steels. Presently, no standardized evaluation method has been established for shear-mode crack propagation under RCF. In this study, twin-disc-type RCF tests were conducted on railway wheel steels with artificial defects, alongside finite element analyses (FEA) simulating these tests. The RCF test results indicated that the cracks initiated from the leading side of the defects propagated inward, while those initiated from the trailing side propagated toward the surface. The stress intensity factor (SIF) calculations revealed that cracks initiated from the leading side of the defects propagated through a mixed modes I and II, whereas cracks initiated from the trailing side propagated exclusively in Mode II. Additionally, crack propagation rates on the trailing side of the defects were higher than those on the leading side, attributed to a larger Mode II equivalent SIF range on the trailing side compared to the leading side. These results suggest that the FEA results corresponded to those of the RCF tests. Therefore, the relationship between the RCF crack propagation rate and shear mode SIF range could be evaluated by combining twin-disc-type RCF tests using test specimens with artificial defects and FEA.

The alternation of dry and wet contact causes a complex interaction of various damage mechanisms in railway wheels. To study the damage evolution in such condition, rolling-sliding contact tests in alternated dry-wet contact were carried out by a bi-disc machine on specimens extracted from railway wheels. The effect of different durations of the dry and wet sessions was investigated. Two kinds of tests were done: in one case coupling two cylindrical specimens, in the other one coupling a cylindrical specimen with a crowned one. The surface appearance during the tests was acquired by a vision system; the acquired images were elaborated to extract quantitative damage indexes. At the end of the tests, the specimens were cut and analysed, to evaluate the subsurface damage and the strain hardening.In cylindrical-cylindrical coupling, the damage increased in alternated dry-wet sessions as far as the cycle number increased, whereas it was stabilized in long dry sessions. The prevailing phenomenon, occurring during the wet sessions, was fatigue, promoted by the pressurization of the water entrapped in the cracks previously formed in dry contact.In crowned-cylindrical couplings, the damage tended to decrease as far as the cycles increased, in all the test conditions. This was correlated to the widening of the contact patch caused by wear, which led to an overall stress relaxation, this way mitigating the damage.The results highlighted the double role of wear, which alters the optimum wheel profile, but at the meantime relaxes the overall stress state and mitigates rolling contact fatigue.

Rolling contact fatigue tests were performed on specimens containing spheroidal inclusions to clarify the effect of the orientation of inclusions, and the results were compared with those of specimens containing stringer‐shaped inclusions. For all types of inclusion, cracks first formed on the rolling surface and then propagated in the depth direction, and an internal crack parallel to the rolling surface formed from the deepest point of the vertical crack, which led to flaking. The flaking life of the specimens with spheroidal inclusions was longer than that of the specimens with stringer‐shaped inclusions, where the initiation and propagation lives of cracks were affected by the inclusion shape and size. The surface crack initiation life was shorter for specimens with longer surface inclusions, whereas the crack propagation life decreased with the depth of surface inclusion. The greater the scatter of the size of inclusions, the greater the scatter of flaking life.

To evaluate stresses including stress concentration at weld toes for fatigue evaluation of complex shape welded structures such as railway bogie frames, previous studies proposed the stress assessment procedure through measuring strains of actual structures by attaching strain gauges with its edges to be at weld toes. This paper evaluates design fatigue strength in terms of the measured stress from published fatigue data and re-analysis of those fatigue tests. The obtained fatigue strength fell into danger side by 12-29% if compared to the fatigue strength estimated based on the current version of JIS. The result also showed that fatigue strength at 2×106 cycles with 0.13% failure probability had a similar value with that at 1×107 cycles with 2.3% failure probability. The fatigue strength in terms of the aforementioned strain measuring system was comparable with those in terms of structural hot-spot stress approach.

The endurance limit of polycrystalline pure iron is high. To understand its mechanism, tensile strength (σB), cyclic yield stress (σyc), and fatigue limit (σw) were evaluated for both single-crystalline and polycrystalline pure iron. The endurance ratio (σw/σB) of rotary bending test and axial loading using polycrystalline specimens ranged from 0.58–0.68. Moreover, the σw/σB of axial loading using a single-crystalline specimen indicated a high value of 1.06. Even without considering the strain rate dependence, the σw/σB of pure iron was higher than that of general-purpose steels such as low-carbon steel, general structural steel, stainless steel, and martensitic steel. Contrary to these results, σw/σyc of single-crystalline and polycrystalline pure iron showed the same value (0.881) as that of general-purpose steels. Thus, σyc would be more correlated with σw than σB. Furthermore, attempts were made to quantify the effect of strain rate, residual stresses, and work-hardened layers on fatigue limit. Consequently, the change in the endurance ratio (0.58–0.68) was quantitatively explained by the strain rate, residual stress, and work-hardened layer.

The infrared thermoelastic method is considered a promising technology for the non-contact measurement of the stress distribution and fatigue limit of structures. In this method, the stress distribution can be detected based on the temperature variation caused by loading the structure, and the fatigue strength can be estimated from the dissipated energy. Additionally, the minute temperature variations of the structures under cyclic load conditions are evaluated. However, there is a concern that appropriate measurement results cannot be obtained using low-frequency cyclic loads owing to the large influence of heat diffusion. Therefore, the effect of the load frequency was investigated using a thermoelastic finite element (FE) analysis on a gusset welded joint specimen commonly used in welded structures. Furthermore, to simulate the thermoelastic effect, a stress field-temperature field thermoelastic FE analysis technique that calculates the heat transfer of the generated heat and heat absorption generated by decreasing and increasing tensile stress was developed. The calculated stress distribution showed consistency with the thermoelastic stress distribution measured using the infrared ray method, implying that the proposed technique is effective in reproducing the thermoelastic effect. A small gradient of temperature distribution was observed in the case employing a load frequency of 1 Hz; therefore, it is important to consider the influence of the load frequency when applying the thermoelastic method to the stress concentration field.

To predict the fatigue limit with a mechanically long crack under mixed modes based on the fracture mechanics, fatigue tests were conducted with an inclined notch. As a result, damage accumulation (DA) mode of fatigue crack propagation was found to occur from the crack tip. The DA mode is caused by the formation and coalescence of micro-voids. Therefore, the method of modifying the initial crack length by considering the DA mode crack propagation length has shown potential application in fatigue limit prediction. Moreover, a classification method of DA and normal modes is proposed.

Rolling contact fatigue cracks propagate in mode II or mixed mode of I and II owing to the cyclic shear stress under multiaxial compressive stress states at the subsurface. No standardized evaluation method for shear mode crack propagation has been established. A method to evaluate the rates and behaviors of rolling contact fatigue crack propagation was studied using twin disc type rolling contact fatigue tests on railway wheel steels with artificial defects and finite element analyses (FEA) simulating the tests. The rolling contact fatigue test results indicated that the cracks initiated from the leading side of defects propagated in the depth direction, whereas those initiated from the trailing side propagated in the surface direction. Calculation results of the stress intensity factor (SIF) indicated that the cracks initiated from the leading side of defects propagated in the mixed mode, whereas those initiated from the trailing side propagated in only mode II. In addition, the crack propagation rates in the trailing side of defects accelerated with increasing number of cycles. This was because of the increase in the shear mode equivalent SIF during the crack propagation. Thus, the results of FEA corresponded to those of rolling contact fatigue tests. Therefore, the relationship between the rolling contact fatigue crack propagation rate and the shear mode SIF range could be evaluated by combining twin disc type rolling contact fatigue tests using test specimen with artificial defects and FEA.

We investigated the effect of load frequency using thermoelastic finite element (FE) analysis on a gusset welded joint specimen commonly used in welded structures. To simulate the thermoelastic effect, we developed a stress field-temperature field thermoelastic FE analysis technique that calculates the heat transfer of heat generation and heat absorption according to compression and tensile stress. The calculated stress distribution showed a good agreement with thermoelastic stress distribution measured using an infrared ray method. The result implies that the developed technique is effective in reproducing the thermoelastic effect. In the case of a load frequency of 1 Hz, a small gradient of temperature distribution appeared. Therefore, when applying the thermoelastic method to the stress concentration field, it is important to consider the influence of the load frequency.

The present study aimed to clarify the relationship between the fatigue life under in-service conditions and the failure probability of welded joints in a railway bogie frame. Considering the effects of the stress ratio and weld toe radius, the authors determined the P–S–N curves by applying the fatigue data of cruciform as-welded joints to the Smith Watson Topper method. An estimation method for the probability of fatigue failure was proposed based on the partial safety factor method (JIS B 9955, 2017, JSME S 018, 2018). Subsequently, the following two concepts were introduced to the fatigue failure issue: one employed the accumulated fatigue damage and its criterion as an evaluation parameter, and the other employed the interval estimation of the scatter of the aforementioned damages based on the t-distribution. Therefore, the proposed method enabled the consideration of the deviation of the in-service cyclic stress. The method was applied to a vehicle using the modified Miner’s rule and the aforementioned P–S–N curves. Consequently, the failure probability increased with an increase in the mileage of the vehicle, and the relationship between the failure probability and the mileage varied because of the deviation of the in-service stress.