Search

The optimization of cast aluminum alloy components is proposed using a Through Process Modeling (TPM) methodology applied to the fatigue design. A full manufacturing process simulation of a cast aluminum alloy A356 component is described. The modeling strategy presented in this paper is able to simulate the defect size, microstructure size and thermal history during the casting process. The microstructure and defect simulated are the Secondary Dendrite Arm Spacing (SDAS) and maximum pore size, respectively. Based on the Defect Stress Gradient approach (DSG) published in a previous work, the fatigue limit is predicted as a function of the SDAS and defect size. Tension fatigue tests have been performed on bespoke prototype cast component samples with various SDAS and defect size. The comparison between experimental results and the TPM simulation shows good agreement in terms of simulated heat transfer rates, defect size, SDAS and fatigue limit except for defect sizes greater than 500 µm. [ABSTRACT FROM AUTHOR]

The design of metallic cast parts requires a compromise between the fatigue resistance of the component and the allowable defect size due to the process. Treatment of defect sensitivity coupled with intrinsic length scales of grains or other microstructure attributes is ultimately necessary to form a predictive basis for defect size effects in forming and growing small defect cracks. This work presents experimental results on high cycle fatigue behavior of specimens containing a surface hemispherical defect under uniaxial tension loading for a wide range of notch size to grain size ratios, including cases where the notch size is on the order of, or even smaller, than the grain size. The influence of grain size on the fatigue strength is clearly demonstrated and the corresponding effects are evaluated. This paper shows that for the same specimen geometry, loading conditions and defect morphology, the fatigue limit is directly dependent on the relationship between the defect size and the grain size. Dimensionless Kitagawa diagram shows that the defect size which impacts the fatigue limit is greater than 0.7 times the grain size in the Armco iron and greater than three times the grain size in other steels. [ABSTRACT FROM AUTHOR]

This paper aims at comparing two different modeling to assess the fatigue life of steel under complex multiaxial fatigue spectrum. The first model is based on a multiaxial fatigue criterion able to describe out of phase loading plus non-linear damage rule. The second one is based on an incremental mesoscale plasticity/damage modeling applied directly on the spectrum studied. Experimental results on 1045 steel are obtained under tension/torsion loading for out of phase variable loading spectrum representative of automotive chassis loading type. It is shown that the criterion with a non-linear damage rule is able to describe the experimental result for full spectrum (with overload) but the identification of the non-linearity is a function of spectrum type. The incremental approach gives better results for simplified spectrum (without overload) without any identification of the non-linearity of the damage. [ABSTRACT FROM AUTHOR]

The Direct Current Potential Drop method is one of the possibilities to measure the crack growth during fatigue tests without optical access to the sample. The accuracy of this technique applied to short cracks mainly depends on the calibration curve. In the present work experimental and numerical approaches are proposed to calibrate the potential drop measurement. An optimization of the calibration procedure is supported by finite element calculations. The crack front shape and the location of the potential probes are found to be of great influence. An extensive study is conducted to identify all the mechanisms controlling the potential drop measurement. Plastic deformation remains the last parameter which is not directly considered in this study; therefore a calibration strategy is suggested to take into account this effect and achieve a calibration curve with high accuracy. This strategy shows a very good agreement between the obtained calibration curve and the experimental calibration. [ABSTRACT FROM AUTHOR]

AS7G06-T6 cast aluminum alloy is tested under tension, torsion and tension–torsion fatigue loading for two load ratios. Basquin’s law and step loading method are used to obtain the fatigue limit under multiaxial loading. Crossland criterion and principal stress criterion considering Goodman idea are compared to evaluate the multiaxial behavior. The influence of complex defects on fatigue limit is analyzed under multiaxial loadings. Several artificial defects are machined on fatigue specimen with different distance between edges. A new definition of the equivalent defect size considering the distance between defect edges is proposed. For both tension and tension–torsion fatigue, the competition between single natural defect and complex artificial defects is observed and analyzed. [ABSTRACT FROM AUTHOR]

Highlights: [•] Critical defect size is linked to the grain size rather than to the SDAS on cast aluminum. [•] Defect Stress Gradient (DSG) criterion shows good predictions. [•] DSG allows producing defect size maps on structural component. [ABSTRACT FROM AUTHOR]

Highlights: [•] The fatigue limits under multiaxial fatigue loadings are identified. [•] Crossland criterion gives reasonable equivalent stress under multiaxial loadings. [•] A new definition of the equivalent defect size for complex defect is proposed. [•] The competition between single and complex defects in fatigue failure is explained. [Copyright &y& Elsevier]

Abstract: The multiaxial fatigue behaviour of PBT–PET GF30 (polybutylene terephthalate–polyethylene terephthalate with 30% mass short glass fibres) and PA66 GF35 (Polyamide 66 with 35% mass short glass fibres) is studied under constant amplitude force loading. Standard flat specimens are used to establish the influence of microstructure on the tension fatigue behaviour. Tubular specimens are used to study the influence of loading path on the tension–torsion fatigue behaviour. Finally an experimental device is developed to test dedicated flat samples in order to study the influence of microstructure on the pure shear fatigue behaviour. All fatigue tests are performed at two load ratios: −1 and 0, 1. Experimental results are analysed using a Through Process Modelling (TPM) based on three major steps. The first one is based on the simulation of the whole process in order to get the fibre orientation tensor at each point of the part. The second one uses the orientation tensor to perform a two-step homogenization procedure in order to estimate local effective properties and compute the stress–strain response as a function of local anisotropy. Mechanical fields thus obtained are used as input data for the application of a fatigue criterion in the third step. Results show that energetic fatigue criterion is an excellent compromise to get good fatigue life assessment for only one experimental input fatigue data. [Copyright &y& Elsevier]

Abstract: Experimental Kitagawa analysis has been performed on A356-T6 containing natural and artificial defects. Results are obtained with a load ratio of R =−1 for three different loadings: tension, torsion and combined tension–torsion. The critical defect size determined is 400±100μm in A356-T6 under multiaxial loading. Below this value, the microstructure governs the endurance limit mainly through Secondary Dendrite Arm Spacing (SDAS). Four theoretical approaches are used to simulate the endurance limit characterized by a Kitagawa relationship are compared: Murakami relationships [Murakami Y. Metal fatigue: effects of small defects and nonmetallic inclusions, Elsevier; 2002], defect-crack equivalency via Linear Elastic Fracture Mechanics (LEFM), the Critical Distance Method (CDM) proposed by Susmel and Taylor [Susmel L, Taylor D. Eng Fract Mech 2008;75:15] and the gradient approach proposed by Nadot and Billaudeau [Nadot Y, Billaudeau T. Eng Fract Mech 2006;73:1]. It is shown that the CDM and gradient methods are accurate; however fatigue data for three loading conditions is necessary to allow accurate identification of an endurance limit. [Copyright &y& Elsevier]

Abstract: Multiaxial fatigue behaviour of polypropylene pipes is investigated under tension and torsion loading with or without mean stress. Fatigue limit is experimentally determined and compared to self heating curve method. A multiaxial fatigue criterion is proposed and shows that the fatigue behaviour of this semi-crystalline polymer seems to be governed by the von Mises maximum stress. [Copyright &y& Elsevier]

Abstract: The aim of this work is to study the influence of both defect and tension and torsion loading on stress number of cycles (S–N) 25 curves for C35 steel. A spherical artificial defect has been machined at the surface of gauge length of fatigue samples. The crack initiation mechanisms have been identified based on several observations on Scanning Electron Microscope (SEM) at different stage of fatigue life. The initiation crack length definition is proposed for defect free and defective material. For defect free material, torsion loading allows relatively earlier initiation compared to tension loading. In the case of defective material, it is observed that, for the both types of loading, initiation fatigue life is not negligible by comparison to total fatigue life. It is also concluded that defects are much more deleterious to fatigue life in the range of high cycle fatigue regime. However, for the limited fatigue lives (between 104 and 105 cycles), the defect free and defective material S–N curves are relatively comparable. [Copyright &y& Elsevier]

Abstract: Artificial spherical defects (400μm) are introduced at the surface of C35 steel fatigue samples. Fatigue tests are conducted under tension and torsion with different mean stress levels in order to determine the fatigue limit. Experimental results show that the effect of mean stress under both torsion and tension is qualitatively the same than for the defect free material. Under torsion, the fatigue limit decreases when the applied mean stress is higher than the yield stress, below this value, it remains constant. Experimental results are analysed using two multiaxial criteria (Dang Van and Crossland) and Murakami’s models: comparison with experimental results shows that the analysis should take into account for the local stress distribution around the defect. Elastic–plastic FE simulations are conducted: isotropic and kinematic hardenings are compared to determine stress distribution around the defect. A nonlocal criterion based on stress gradient is applied and results using the two different hardening laws are compared. It is shown that the isotropic hardening gives good results when the fatigue test is conducted for a load ratio R =−1; kinematic hardening gives good results when R ≠−1. [Copyright &y& Elsevier]

Samples of a Zr-based bulk metallic glass with a small (1000appm) or very small (less than 300appm) oxygen content are pre-cracked by fatigue and tested for toughness evaluation. It is shown that oxygen trapped in oxide dendrites eases the initiation of a straight crack and embrittles the glass even at such low concentrations. In contrast, when oxygen is dissolved in the glass, fatigue crack initiation becomes difficult and it is not possible to get a crack passing straight through the glass. [Copyright &y& Elsevier]

Abstract: The purpose of this study was to test a vinylester adhesive T-joint without stratification for a structural part of a ship. Tests were conducted under fatigue and monotonic loadings in order to study the durability of such bonded joints. A careful observation of cracking mechanisms during fatigue life showed that fatigue mechanisms were strongly dependant on the geometry of the sample tested. Moreover this cracks can initiate in the bulk. Results also showed that the fatigue lifetime of such adhesive T-joints is separated into two stages, the initiation stage, which is about a third of the fatigue lifetime, and the propagation stage for the rest up to failure. [Copyright &y& Elsevier]

Abstract: The objective of this paper is to quantify the influence of defect on the fatigue limit. Elastic–plastic simulations are conducted to determine the stress distribution around defects for different geometries and different loading. It is shown that a relevant mechanical parameter governing the fatigue limit for defect material could be the gradient of the hydrostatic stress. A multiaxial fatigue criterion is identified with three parameters and validated for different metallic materials under multiaxial conditions. Results are good and show that the gradient of the hydrostatic stress is a good parameter to characterise the influence of a defect on the fatigue behaviour. [Copyright &y& Elsevier]

To study the influence of defect geometry on the fatigue behaviour, experiments are conducted under tension and torsion on a C36 steel containing artificial defects of different sizes and shapes. Results show that the shape of the defect influences the fatigue limit at a given size. A study of fatigue mechanisms shows that cracks propagate in the plane perpendicular to the maximum principal stress direction because of the stress distribution around the defect. Nevertheless, it is shown that fatigue crack initiation in the maximum shear plane seems to be the governing mechanism at the tip of the defect for low stress concentration defect. Stage I of the fatigue process is not negligible for defective material. A multiaxial endurance criterion is used to analyse the experimental results: it appears necessary to calculate the local stress distribution around the defect to characterise the defect morphology in a multiaxial criterion. [Copyright &y& Elsevier]

Recent studies in the Giga Cycle Regime show that the internal crack initiation is largely promoted, compared to the High Cycle Regime. This change of mechanism is still under wide discussion. The purpose of this study is to highlight the influence of the environment on both internal and surface fatigue failure. In that way, experimental results from the literature concerning Giga Cycle Fatigue are presented and crack initiation mechanisms are discussed. High Cycle Fatigue results on nodular cast iron containing casting defects are presented. The position of the defect at the origin of the failure is analysed, considering the environmental effect. Giga Cycle results from the literature are then analysed using simple fracture mechanics assessment. It is shown that the environment has to be considered when computing fatigue life assessment for materials containing surface or internal defects. Our results also indicate that, even if there is a change of mechanism between HCF and GCF, the initiation stage is still predominant in the Giga Cycle Regime. [Copyright &y& Elsevier]

An experimental device have been developed to study fatigue phenomena for nodular cast iron automotive suspension arms. On the base of a detailed fracture analysis, it is shown that the major parameter influencing fatigue failure of casting components are casting defects: the High Cycle Fatigue behaviour is controlled mainly by surface defects such as dross defects and oxides while the Low Cycle Fatigue is governed by multiple cracks initiated independently from casting defects. A methodology is proposed to define the maximum defect size allowable in a casting component. It correlates the empirical method proposed by Murakami to determine the evolution of the fatigue limit with defect size and a multiaxial endurance criterion based on the Dang Van model. The junction between the two approaches gives a concurrent tool for the fatigue design of casting components. Validation of the proposed approach gives encouraging results for surface defects and constant amplitude proportional loading. [Copyright &y& Elsevier]

The high cycle fatigue behaviour of a nodular cast iron has been investigated under tension loading. Casting defects are at the origin of crack initiation for all samples tested. Critical defects (shrinkage) are located either at the surface or within the bulk. The role of the size and position of natural defect on the fatigue limit has been established. Results show that, for a given size, internal defects are less damaging than surface located ones. SEM observations revealed the existence of non-propagating surface cracks below the fatigue limit. By means of a marking technique, the 3D shape of these cracks initiated around the defects was fully described. In an attempt to simulate the evolution of the fatigue limit with defect size, some existing approaches are compared to experimental data; it is shown that 2D analysis of defects (crack or notch) is conservative. [Copyright &y& Elsevier]

Although Ti-6Al-4V is the most commonly used titanium alloy in the aerospace industry, the mechanisms governing crack initiation in the different fatigue regimes are not well understood yet. This situation partly pertains to a competition between multiple crack initiation mechanisms. In particular, applied loading conditions were identified as a key parameter governing the transition between mechanisms. The fatigue behavior of Ti-6Al-4V with a bi-modal microstructure was investigated in the present study using different waveforms, load ratios and frequencies to clarify this feature. A detailed characterization of the main crack initiation sites was carried out to identify microstructural configurations governing crack nucleation in low-cycle dwell-fatigue, low-cycle fatigue and high-cycle fatigue regimes. While lifetimes and fracture surfaces showed a good consistency with data from prior studies, a single microstructural configuration was found involved with the formation of crack initiation facets. All investigated cracks leading to specimen failure were nucleated along (0001) twist grain boundaries exhibiting similar features. Based on this finding, a criterion is proposed to identify candidates for crack initiation. This also demonstrates no major sensitivity of the critical microstructural configurations to environment, free surface, loading conditions and, as shown in previous studies, microstructure and composition. However, conventional fatigue failure results from one, or a few, surface or subsurface crack initiation sites while dwell-fatigue failure involves the formation of multiple internal cracks. This is accompanied by a significant dwell-fatigue life debit at peak stress magnitudes close to the yield strength. Features of microstructural configurations prone to crack nucleation are finally compared with data reported in existing literature to propose a mechanism of crack formation in Ti alloys. [Display omitted] [ABSTRACT FROM AUTHOR]

Abstract: The aim of this work is to model short crack growth under fretting fatigue loading conditions by considering a criterion based on linear elastic fracture mechanics quantities, which also accounts for the first non-singular terms of the asymptotic expansion, namely the T-stresses. The Modes I and II Stress Intensity Factors and the T-stresses were computed by the finite element method under plane strain hypothesis. To assess the model fretting fatigue tests were carried out using two cylindrical fretting pads, which were loaded against a flat dogbone tensile test piece, both made of a Ti–6Al–4V titanium alloy. The model was capable to correctly estimate short crack arrest and to find the threshold fretting conditions separating failure from infinite life (here defined by tests which reached one million cycles). An optimization technique was implemented to the numerical model so that it could also estimate crack path. [Copyright &y& Elsevier]

Highlights: [•] Nearby defect stress field computation by Equivalent Inclusion Method (EIM). [•] EIM vs Finite Element (FEM) comparison for internal defect: good agreement. [•] Internal vs surface defect by FEM using Defect Stress Gradient (DSG)/fatigue approach. [•] Legitimacy of the EIM employment within upgraded DSG. [•] Evaluation of the upgraded DSG/EIM using experimental fatigue database (C35 steel). [Copyright &y& Elsevier]

Abstract: A multiaxial criterion for high cycle fatigue based on invariants of macroscopic stress tensor is proposed. The presence of J 2, mean allows capturing accurately effects of phase shift and frequency on fatigue limit of material under multiaxial loading. Prediction capacity of the proposed criterion is tested on 119 iso-frequency axial–torsion experiments and some other more complex loadings: biaxial loading and asynchronous loading. Results show that the criterion is in good accordance with the experimental data. Under proportional loading, the assessment of the criterion can be carried out from an analytical solution. Under others loading cases, the numerical implementation of the criterion is very simple and can be easily integrated in a damage model or in a finite element code. [Copyright &y& Elsevier]

Abstract: This paper attempts to improve the understanding of the multiaxial high cycle fatigue response of micro sized stress concentrations or notches of different geometries. The investigation is composed of an experimental part and a numerical part. In the former, three types of micro-notches or “artificial defects” are compared: spherical, elliptical and circumferential. All types have the same basic dimensions, the difference being the 3D geometry. The notches were machined on the surface of smooth cylindrical specimens made of mild steel. The fatigue limits under reversed tension (push–pull) and reversed torsional loading conditions for different micro-notch sizes have been experimentally determined. In the numerical part, finite elements simulations using a cyclic elasto-plastic material behaviour law show that the mechanical state ahead of the different stress concentrations change drastically with the loading mode and the geometry of the artificial defect. From a fatigue point of view, it is shown that a stress gradient correction is required for all the loading, size and geometry configurations. Once the gradient correction is made and a proper multiaxial criterion is used, it appears that the size effect due to increasing the loaded surface area at the notch tip for the different geometries is negligible compared to the gradient effect. [Copyright &y& Elsevier]

Highlights • Fatigue lifetimes of samples subjected to complex multiaxial spectrum are measured. • Different simulation tools to predict fatigue lifetime are evaluated. • Incremental damage model can forecast lifetime with and without overload. Abstract This study deals with fatigue of a 1045 carbon steel subjected to a cyclic tension-torsion spectrum combining in-phase and out-of-phase loadings and including overload, representative of an automotive chassis loading type. The experimental lifetimes are compared to the results of two different approaches. The first one combines a multiaxial fatigue criterion able to describe out of phase loading and a non-linear damage rule. The second one is based on an incremental mesoscale plasticity/damage model. It is shown that the criterion involving a non-linear damage rule is able to describe the experimental result for full spectrum (with overload) but the identification of the non-linearity is a function of spectrum type. The incremental approach gives better results for both spectra (with and without overloads) and does not need specific parameter identification. [ABSTRACT FROM AUTHOR]

Very high cycle fatigue fracture is often associated with internal crack propagation and one major problem to study the initiation and the propagation of this internal crack is to detect its initiation and quantify its propagation rate. The objective of the present work is to develop an experimental methodology to follow the initiation and propagation of the internal crack. The experimental technique presented here is based on temperature field measurement on the specimen surface during an ultrasonic fatigue test using infrared thermography. Indeed, plasticity in the reverse cyclic plastic zone near the crack front generates heat sources due to the dissipation of plastic energy rate and thus an increase of the temperature which propagates through the specimen thanks to heat conduction. By considering the heat source located in the reverse cyclic plastic zone and the geometry of the crack obtained by tomographic observations, finite element simulation of the heat transfer problem enables us to establish a relationship between the internal crack growth and the temperature field evolution on the specimen surface. The results obtained for a cast aluminum alloy are presented and analyzed. • Initiation and propagation of internal fatigue crack is studied in this paper. • The plasticity in the reverse cyclic plastic zone generates a heat source. • A thermal modeling using the 3D geometry of the crack is developed. [ABSTRACT FROM AUTHOR]

The influence of internal casting defects on the fatigue properties of A357-T6 alloy has been studied. Specimens with artificial casting defects ( ∅ eq ≈ 2 mm) placed at the surface or in the bulk have been produced. A reference S–N curve has been obtained for the material containing only natural defects (microshrinkages). Cycling has been performed in air or under vacuum. In spite of their relatively large size, the internal artificial defects do not induce a reduction of the fatigue life compared to the reference material. When the artificial defects are placed at the surface, a decrease is observed when the specimens are cycled in air but it disappears again when cycling is performed in vacuum conditions. The detrimental effect of the casting defects is discussed in relation to their ability to initiate a crack growing in vacuum or not. It can be concluded that the increased fatigue lives observed for internal defects are the consequence of slower crack initiation and propagation mechanism occurring in vacuum. [ABSTRACT FROM AUTHOR]