Crack modelling: A novel technique for the prediction of fatigue failure in the presence of stress concentrations

Finite element (FE) analysis and other computational methods have developed rapidly in recent years, allowing accurate predictions of elastic stresses in components of complex geometry. However, the prediction of fatigue failure in these components is still a non-trivial problem; one reason for this is the difficulty of assessing stress concentrations and regions of high stress-gradient. This paper describes a new technique, called “crack modelling”, which addresses the problem through a modification of linear-elastic fracture mechanics (LEFM). LEFM is designed to deal with cracks in nominally elastic stress fields, using elastic analysis to derive a characteristic stress intensity, K or, for cyclic loading, a range ΔK. This methodology is modified in two ways. Firstly it is shown that LEFM can be extended to predict the fatigue behaviour of bodies containing notches of standard geometry, instead of cracks. Secondly, FE analysis is used in conjunction with a modelling exercise in order to extend the method to include bodies of arbitrary shape subjected to any set of loads. The method was first tested using standard notch geometries (blunt and sharp notches in beams), where accurate predictions of fatigue limit could be achieved. It was then applied to an industrial problem, giving a prediction of high-cycle fatigue behaviour for an automotive crankshaft. The method requires only simple mechanical-property data (the material fatigue limit and stress-intensity threshold) and uses only linear-elastic FE modelling. It allows fracture mechanics theory to be used without the need to specifically model the presence of a crack and uses far-field elastic stresses to infer behaviour in the region of a stress concentration.     

Short crack threshold estimates to predict notch sensitivity factors in fatigue

The notch sensitivity factor q can be associated with the presence of non-propagating fatigue cracks at the notch root. Such cracks are present when the nominal stress range Δσ_n is between Δσ₀/K_t and Δσ₀/K_f, where Δσ₀ is the fatigue limit, K_t is the geometric and K_f is the fatigue stress concentration factors of the notch. Therefore, in principle it is possible to obtain expressions for q if the propagation behavior of small cracks emanating from notches is known. Several expressions have been proposed to model the dependency between the threshold value ΔK_th of the stress intensity range and the crack size a for very small cracks. Most of these expressions are based on length parameters, estimated from ΔK_th and Δσ₀, resulting in a modified stress intensity range able to reproduce most of the behavior shown in the Kitagawa–Takahashi plot. Peterson or Topper-like expressions are then calibrated to q based on these crack propagation estimates. However, such q calibration is found to be extremely sensitive to the choice of ΔK_th(a) estimate. In this work, a generalization version of El Haddad–Topper–Smith’s equation is used to evaluate the behavior of cracks emanating from circular holes and semi-elliptical notches. For several combinations of notch dimensions, the smallest stress range necessary to both initiate and propagate a crack is calculated, resulting in expressions for K_f and therefore for q. It is found that the q estimates obtained from this generalization, besides providing a sound physical basis for the notch sensitivity concept, better correlate with experimental data from the literature.     

Fracture mechanics modelling of constant and variable amplitude fatigue behaviour of field corroded 7075‐T6511 aluminium

For ageing airframe structures, a critical challenge for next generation linear elastic fracture mechanics (LEFM) modelling is to predict the effect of corrosion damage on the remaining fatigue life and structural integrity of components. This effort aims to extend a previously developed LEFM modelling approach to field corroded specimens and variable amplitude loading. Iterations of LEFM modelling were performed with different initial flaw sizes and crack growth rate laws and compared to detailed experimental measurements of crack formation and small crack growth. Conservative LEFM‐based lifetime predictions of corroded components were achieved using a corrosion modified‐equivalent initial flaw size along with crack growth rates from a constant K_max‐decreasing ΔK protocol. The source of the error in each of the LEFM iterations is critiqued to identify the bounds for engineering application.     

Fatigue assessment using an integrated threshold curve method - applications

This paper deals with the analysis and prediction of a high-cycle fatigue behaviour in notched and damaged specimens, as well as butt-welded joints by using a threshold curve for fatigue crack propagation that includes the short crack regime (a function of crack length, a). The approach regards the effective driving force applied to the crack as the difference between the total applied driving force defined by the applied stress distribution corresponding to a given geometrical and loading configuration, ΔK(a), and the threshold for crack propagation, ΔK_th(a). Chapetti’s model is used to estimate the threshold for crack propagation by using the plain fatigue limit, Δσ_eR, the threshold for long cracks, ΔK_thR, and the microstructural characteristic dimension (e.g. grain size). Applications, predictions and results, in good agreement with experimental results from the literature, demonstrate the ability of the method to carry out quantitative analyses of the high cycle fatigue propagation behavior (near threshold) of short cracks in different geometrical, mechanical and microstructural configurations.     

Predicting fracture and fatigue crack growth properties using tensile properties

The safe-life assessment of components requires information such as the plane stress (K_c), plane strain (K_Ic), part-through fracture toughness (K_Ie), and the fatigue crack growth rate properties. A proposed parametric/theoretical approach, based on an extended Griffith theory is used to derive fracture toughness properties and generate fatigue crack growth rate data for a range of alloys. The simplicity of the concept is based on the use of basic, and in most cases available, uniaxial stress-strain material properties data to derive material fracture toughness values. However since the methodology is in part based on an empirical relationship a wide ranging validation with actual data is required. This paper uses steel, aluminum and titanium based alloys from a pedigree database to quantify material properties sensitivity to the predictions for K_Ic and K_c and the subsequent estimation of ΔK_th threshold and the Paris constants, C and n values. A sensitivity analysis using experimental scatter bounds show the range of da/dN predictions can be achieved. It is found K_Ic/ΔK_th ratios designated as α has a range of 5-25 irrespective of tensile ductility, ε_f, and is insensitive to it. The value of ΔK_th for all the alloys considered was found to be proportional to the final elongation, ε_f, and an empirical relationship describing ΔK_th as a function of ε_f was established. Furthermore it is suggested that, with the knowledge of appropriate tensile properties and the estimated range of K_Ic/ΔK_th ratios for the different alloys applying this method could be an appropriate tool that can be used to conservatively predict fracture and fatigue in similar alloy categories. Thus helping to reduce costs and optimize the number of experimental tests needed for alloy characterizations.     

Experimental analysis of the practical LBZ effects on the brittle fracture performance of cryogenic steel HAZs with respect to crack arrest toughness near fusion line

Focusing on crack arrest behavior, this study investigates the practical influence of local brittle zones (LBZs) on the brittle fracture resistance of heat-affected zones (HAZs) in advanced 9% Ni cryogenic steel welds, and discusses whether the LBZs of this steel in practice have potentially deleterious effects as previously thought, or not. By analyzing the variations in brittle crack arrest toughness (K_a) and brittle crack initiation toughness (K_c) within actual HAZ, it is found that LBZs of this steel may not be harmful in consideration of crack arrest toughness near fusion line.     

Evaluation of the overload effect on fatigue crack growth with the help of synchrotron XRD strain mapping

The overload retardation effect on fatigue crack growth rate (FCGR) in titanium alloy Ti-6Al-4V is studied. Synchrotron X-ray diffraction strain mapping of near-crack tip regions of pre-cracked fatigued samples is used to determine the effective stress intensity factors experienced by the crack tip. The effective stress intensity factor values are computed by finding the best match between the experimental strain maps and linear elastic fracture mechanics (LEFM) predictions. The dependence of the effective stress intensity factor, K, on the applied load is plotted, and an interpretation of the overload retardation effect is proposed. The present approach permits to reconcile the traditional LEFM fatigue crack propagation prediction and the experimental measurement of strain fields.     

Fracture of quasi-isotropic composite sheets with sharp notches

Continuous fibre composites are materials that exhibit rather linear elastic deformation behaviour: suggesting brittleness and notch sensitivity. However, notched composites may sustain significant mechanical load. The notch resistance of composites is investigated on quasi-isotropic composite sheets with sharp crack like notches. This allows the use of analytic solutions of the stress field around a crack in a similar way as is used for linear elastic fracture mechanics (LEFM) in homogeneous isotropic solids. Similar to the small scale yielding boundary condition in fracture mechanics, applied on homogeneous isotropic solids, a small-scale non-linear damage condition should be fulfilled for valid LEFM application on quasi-isotropic composites. Indeed, it appeared to be possible to define critical stress intensity factors (K_1c) for the quasi-isotropic composite. Moreover, K_1c values can quantitatively be related to laminate parameters and to the related damage and deformation processes occurring in a small near crack tip zone with intense non-linearity and strain gradients in the thickness direction. Before the final explosive fracture occurred, stable crack growth was observed. This could be described with R-curves, as done for homogeneous metal sheet specimens. Indeed, also in this case, the R-curves were identical, independent of the length of the initial crack-like notch. The R-curves can be estimated adopting a crack-bridging model. Crack growth occurs at the notch tip in the 0° plies. The other plies bridge the fractured 0° plies. The fracture mechanisms, determining the K_1c-values and the shape of the R-curve, are quite different for composites and metals. Yet, the method of fracture mechanics, well established for metals, can obviously also be applied to quasi-isotropic composites.     

Fatigue behaviour of friction stir welded AA2024‐T3 alloy: longitudinal and transverse crack growth

The fatigue crack growth properties of friction stir welded joints of 2024‐T3 aluminium alloy have been studied under constant load amplitude (increasing‐ΔK), with special emphasis on the residual stress (inverse weight function) effects on longitudinal and transverse crack growth rate predictions (Glinka's method). In general, welded joints were more resistant to longitudinally growing fatigue cracks than the parent material at threshold ΔK values, when beneficial thermal residual stresses decelerated crack growth rate, while the opposite behaviour was observed next to K_C instability, basically due to monotonic fracture modes intercepting fatigue crack growth in weld microstructures. As a result, fatigue crack growth rate (FCGR) predictions were conservative at lower propagation rates and non‐conservative for faster cracks. Regarding transverse cracks, intense compressive residual stresses rendered welded plates more fatigue resistant than neat parent plate. However, once the crack tip entered the more brittle weld region substantial acceleration of FCGR occurred due to operative monotonic tensile modes of fracture, leading to non‐conservative crack growth rate predictions next to K_C instability. At threshold ΔK values non‐conservative predictions values resulted from residual stress relaxation. Improvements on predicted FCGR values were strongly dependent on how the progressive plastic relaxation of the residual stress field was considered.     

Fatigue growth of short cracks in Ti-17: Experiments and simulations

The fatigue behaviour of through thickness short cracks was investigated in Ti-17. Experiments were performed on a symmetric four-point bend set-up. An initial through thickness crack was produced by cyclic compressive load on a sharp notch. The notch and part of the crack were removed leaving an approximately 50 μm short crack. The short crack was subjected to fatigue loading in tension. The experiments were conducted in load control with constant force amplitude and mean values. Fatigue growth of the short cracks was monitored with direct current potential drop measurements. Fatigue growth continued at constant R-ratio into the long crack regime. It was found that linear elastic fracture mechanics (LEFM) was applicable if closure-free long crack growth data from constant K_Imax test were used. Then, the standard Paris’ relation provided an upper bound for the growth rates of both short and long crack.The short crack experiments were numerically reproduced in two ways by finite element computations. The first analysis type comprised all three phases of the experimental procedure: precracking, notch removal and fatigue growth. The second analysis type only reproduced the growth of short cracks during fatigue loading in tension. In both cases the material model was elastic-plastic with combined isotropic and kinematic hardening. The agreement between crack tip opening displacement range, cyclic J-integral and cyclic plastic zone at the crack tip with ΔK_I verified that LEFM could be extended to the present short cracks in Ti-17. Also, the crack size limits described in the literature for LEFM with regards to plastic zone size hold for the present short cracks and cyclic softening material.     

Effects of geometry of notched specimens on the local cleavage fracture stress σf of C-Mn steel

An elastic-plastic three-dimensional finite element method analysis is used to determine the stress and strain distributions ahead of notches of four-point bending (4PB) specimens with various sizes (W, B and a) and widths (B). By measuring the location of the cleavage initiation sites for a C-Mn steel, the local cleavage fracture stress σ_f is accurately determined. With increasing specimen sizes and widths the fracture load P_f increases considerably, but σ_f remains nearly constant. The reason that the σ_f of the specimen with minimum size is slightly larger than that of the other specimens is analyzed by an active zone model of cleavage fracture for notched specimens. The critical event for cleavage fracture is the propagation of a ferrite grain-sized crack into the neighboring matrix, and is independent of specimen sizes and widths. σ_f is mainly determined by the length of the critical microcrack, and the specimen sizes and widths have little effect on it.     

Weakly nonlinear fracture mechanics: experiments and theory

Material failure occurs at the small scales in the immediate vicinity of the tip of a crack. Due to its generally microscopic size and the typically high crack propagation velocity, direct observation of the dynamic behavior in this highly deformed region has been prohibitively difficult. Here we present direct measurements of the deformation surrounding the tip of dynamic mode I cracks propagating in brittle elastomers at velocities ranging from 0.2 to 0.8C _s . Both the detailed fracture dynamics and fractography of these materials are identical to that of standard brittle amorphous materials such as soda-lime glass. These measurements demonstrate how Linear Elastic Fracture Mechanics (LEFM) breaks down near the tip of a crack. This breakdown is quantitatively described by extending LEFM to the weakly nonlinear regime, by considering nonlinear elastic constitutive laws up to second order in the displacement-gradients. The theory predicts that, at scales within a dynamic lengthscale ? _nl from the tip of a single crack, significant logr displacements and 1/r displacement-gradient contributions arise, and provides excellent quantitative agreement with the measured near-tip deformation. As ? _nl is consistent with lengthscales that appear in crack tip instabilities, this “weakly nonlinear fracture mechanics” framework may serve as a springboard for the development of a comprehensive theory of fracture dynamics.     

Fatigue crack propagation work coefficient—a material constant giving degree of resistance to fatigue crack growth

Study on fatigue crack growth in steels was carried out from energetic point of view, i.e. taking account of plastic work around the fatigue crack. Based on the examination of the relation between fatigue crack growth rate (da/dN) and the plastic work around the fatigue crack tip (W_0.02 in SUS304, Fe-3Si and HT 60 steels, a material constant-fatigue crack propagation work coefficient-Q_0.02 is proposed. It is the ratio of W_0.02 to da/dN and means the degree of the resistance to fatigue crack growth. Numerical expression of Q_0.02 by mechanical properties was derived, which is given by

$\text{Q}{}_{0.02}\text{=}\text{9.3x10}{}^1\text{(σ}{}_{\mathrm{y}}\text{+σ}{}_{0.2}\text{)}\text{σ}{}_{\mathrm{y}}{}^{1.3}$

Comparison of Q_0.02 of various steels showed that Q_0.02 of high strength steels is very small compared with that of low strength steels. Graphical representation of the relation between Q_0.02 and da/dN at various values of ΔK/σ_y for steels revealed that da/dN at given value of ΔK/σ_y increase with decreasing Q_0.02. It is shown that fatigue crack growth behaviour of a steel (da/dN-ΔK relation) can be obtained from the Q_0.02-da/dN diagram by knowing the mechanical properties. Discussion on design stress level of the steels is also given.     

Discrete fractal fracture mechanics

A modification of the classical theory of brittle fracture of solids is offered by relating discrete nature of crack propagation to the fractal geometry of the crack. The new model incorporates all previously considered theories of fracture processes, in particular the Griffith [Griffith AA. The phenomenon of rupture and flow in solids. Philos Trans Roy Soc Lond 1921;A221:163-398] theory, its contemporary extension known as LEFM and the most recently developed Quantized Fracture Mechanics (QFM) by Pugno and Ruoff [Pugno N, Ruoff RS. Quantized fracture mechanics. Philos Mag 2004;84(27):2829-45]. Using an equivalent smooth blunt crack for a given fractal crack, we find that assuming that radius of curvature of the blunt crack is a material property, the crack roughens while propagating. In other words, fractal dimension at the crack tip is a monotonically increasing function of the nominal crack length, i.e., the presence of the Mirror-Mist-Hackle phenomenon is analytically demonstrated.     

State‐of‐the‐art review on extended stress intensity factor concepts

The stress intensity factor concept for describing the stress field at pointed crack or slit tips is well known from fracture mechanics. It has been substantially extended since Williams' basic contribution (1952) on stress fields at angular corners. One extension refers to pointed V‐notches with stress intensities depending on the notch opening angle. The loading‐mode‐related simple notch stress intensity factors K₁, K₂ and K₃ are introduced. Another extension refers to rounded notches with crack shape or V‐notch shape in two variants: parabolic, elliptic or hyperbolic notches (‘blunt notches’) on the one hand and root hole notches (‘keyholes’ when considering crack shapes) on the other hand. Here, the loading‐mode‐related generalised notch stress intensity factors K_1ρ, K_2ρ and K_3ρ are defined. The concepts of elastic stress intensity factor, notch stress intensity factor and generalised notch stress intensity factor are extended into the range of elastic–plastic (work‐hardening) or perfectly plastic notch tip or notch root behaviour. Here, the plastic notch stress intensity factors K_1p, K_2p and K_3p are of relevance. The elastic notch stress intensity factors are used to describe the fatigue strength of fillet‐welded attachment joints. The fracture toughness of brittle materials may also be evaluated on this basis. The plastic notch stress intensity factors characterise the stress and strain field at pointed V‐notch tips. A new version of the Neuber rule accounting for the influence of the notch opening angle is presented.     

A COMPARISON OF THE STRAIN INTENSITY AND CYCLIC J APPROACHES TO CRACK GROWTH

Two parameters describing the growth of fatigue cracks are compared. They are the cyclic J integral ΔJ and the strain intensity expressed as an equivalent stress intensity ΔK_eq-. By referring to cyclic stress-strain data obtained from hysteresis loops in high strength ferritic steels at room temperature and austenitic and ferritic steels at elevated temperature it is shown that: (i) for short cracks the parameters are simply related and (ii) both parameters adequately link fatigue crack growth rates observed in the separate high strain fatigue (HSF) and linear elastic fracture mechanics (LEFM) regimes. Correction factors for thumbnail cracks and the conditions under which the relations need further modification are discussed.     

Fatigue analysis of non-load-carrying fillet welded cruciform joints

The goal of this investigation was to study the effect of local geometrical variations of the weld on the fatigue strength. Therefore the fatigue behaviour of non-load-carrying cruciform fillet welded joint under tensile loading has been studied parametrically. Several two-dimensional (2D) finite element models of the joint were analysed using plane strain linear elastic fracture mechanics (LEFM) calculations in order to get the magnification function M_k. A maximum tangential stress criterion was used to predict the crack growth direction under mixed mode K_I-K_II conditions. The derived M_k solution was then applied both for continuous weld toe cracks and also for semi-elliptical toe cracks at the deepest point of the crack front. An experimental crack aspect ratio development curve was used for propagating semi-elliptical cracks. The as-welded condition was assumed with the result that no crack initiation period was considered and stress ranges were fully effective. The Paris crack growth law was used to predict the growth rate. The effects of weld toe radius, flank angle and weld size on the fatigue strength were systematically studied. Finally, predicted fatigue strength values corresponding to different assumed crack sizes were compared with the available test results.     

Studies on crack growth rate under high temperature creep,fatigue and creep-fatigue interaction—II: On the role of fracture mechanics parameter aeffσg

For high temperature creep, fatigue and creep-fatigue interaction, several authors have recently attempted to express crack growth rate in terms of stress intensity factor $\text{K}{}_{\mathrm{I}}\text{= α}\text{aσ}{}_{\mathrm{g}}$, where a is the equivalent crack length as the sum of the initial notch length a₀ and the actual crack length $\text{a}{}^1$, that is, $\text{a = a}{}_0\text{+ a}{}^1$. On the other hand, it has been shown by Yokobori and Konosu that under the large scale yielding condition, the local stress distribution near the notch tip is given by the fracture mechanics parameter of $\text{aσ}{}_{\mathrm{g}}\text{?(σ}{}_{\mathrm{g}}$), where a is the cycloidal notch length, σ_g is the gross section stress and $\text{?(σ}{}_{\mathrm{g}}$) is a function of σ_g. Furthermore, when the crack growth from the initial notch is concerned, it is more reasonable to use the effective crack length a_eff taking into account of the effect of the initial notch instead of the equivalent crack length a. Thus we believe mathematical formula for the crack growth rate under high temperature creep, fatigue and creep-fatigue interaction conditions may be expressed at least in principle as function of $\text{a}{}_{\mathrm{eff}}\text{σ}{}_{\mathrm{g}}\text{, σ}{}_{\mathrm{g}}$ and temperature.In the present paper, the geometrical change of notch shape from the instant of load application was continuously observed during the tests without interruption under high temperature creep, fatigue and creep-fatigue interaction conditions. Also, the effective crack length a_eff was calculated by the finite element method for the accurate estimation of local stress distribution near the tip of the crack initiated from the initial notch root. Furthermore, experimental data on crack growth rates previously obtained are analysed in terms of the parameter of $\text{a}{}_{\mathrm{eff}}\text{σ}{}_{\mathrm{g}}$ with gross section stresses and temperatures as parameters, respectively.     

Limitations of elastic definitions in Al-Si-Mg cast alloys with enhanced plasticity: linear elastic fracture mechanics versus elastic-plastic fracture mechanics

Linear elastic fracture mechanics describes the fracture behavior of materials and components that respond elastically under loading. This approach is valuable and accurate for the continuum analysis of crack growth in brittle and high strength materials; however it introduces increasing inaccuracies for low-strength/high-ductility alloys (particularly low-carbon steels and light metal alloys). In the case of ductile alloys, different degrees of plastic deformation precede and accompany crack initiation and propagation, and a non-linear ductile fracture mechanics approach better characterizes the fatigue and fracture behavior under elastic-plastic conditions.To delineate plasticity effects in upper Region II and Region III of crack growth an analysis comparing linear elastic stress intensity factor ranges (ΔK_el) with crack tip plasticity adjusted linear elastic stress intensity factor ranges (ΔK_pl) is presented. To compute plasticity corrected stress intensity factor ranges (ΔK_pl), a new relationship for plastic zone size determination was developed taking into account effects of plane-strain and plane-stress conditions (“combo plastic zone”). In addition, for the upper part of the fatigue crack growth curve, elastic-plastic (cyclic J based) stress intensity factor ranges (ΔK_J) were computed from load-displacement records and compared to plasticity corrected stress intensity factor ranges (ΔK_pl). A new cyclic J analysis was designed to compute elastic-plastic stress intensity factor ranges (ΔK_J) by determining cumulative plastic damage from load-displacement records captured in load-control (K-control) fatigue crack growth tests. The cyclic J analysis provides the true fatigue crack growth behavior of the material. A methodology to evaluate the lower and upper bound fracture toughness of the material (J_IC and J_max) directly from fatigue crack growth test data (ΔK_FT(JIC) and ΔK_FT(Jmax)) was developed and validated using static fracture toughness test results. The value of ΔK_FT(JIC) (and implicitly J_IC) is determined by comparing the plasticity corrected elastic fatigue crack growth curve with the elastic-plastic fatigue crack growth curve. A most relevant finding is that plasticity adjusted linear elastic stress intensity factor ranges (ΔK_pl) are in remarkably good agreement with cyclic J analysis results (ΔK_J), and provide accurate plasticity corrections up to a ΔK corresponding to J_IC (i.e. ΔK_FT(JIC)). Towards the end of the fatigue crack growth test (above ΔK_FT(JIC)) when plasticity is accompanied by significant tearing, the cyclic J analysis provides a more accurate way to capture the true behavior of the material and determine ΔK_FT(Jmax). A procedure to decouple and partition plasticity and tearing effects on crack growth rates is given.Three cast Al-Si-Mg alloys with different levels of ductility, provided by different Si contents and heat treatments (T61 and T4) are evaluated, and the effects of crack tip plasticity on fatigue crack growth are assessed. Fatigue crack growth tests were conducted at a constant stress ratio, R = 0.1, using compact tension specimens.