Modeling fatigue crack growth for a through thickness crack: An out-of-plane constraint-based approach considering thickness effect
Graphical abstract
Introduction
The prediction of crack growth is critical in the damage tolerance design of engineering components. Varying thickness design in engineering structures is applied more and more widely, such as turbine blades and discs with air-cooling techniques. For these complex geometrical configurations, the thickness effect on crack growth requires to be evaluated to seek for a more accurate and reliable prediction. Several studies have examined this effect with contradictory results. Several scholars [1], [2], [3] proposed that thickness increases the crack growth rate, while Kim et al. [4] stated that thickness has a negligible effect on it. Roughly associating specimen thickness (e.g., the crack thickness for a through crack) with crack growth rate can lead to ambiguous conclusions. Therefore, a more comprehensive explanation of the mechanism behind thickness is needed to develop an accurate and reliable crack growth model. Yu et al. [5] proposed that the microcosmic mechanism about the thickness effect on nickel-base superalloy's creep and crack behavior was associated with the depth of γ’ depleted region. Many studies [6], [7], [8], [9] have similarly suggested that the stress state associated with the thickness has a significant effect on crack propagation. However, the traditional crack propagation theory framework is limited when considering the high stress triaxiality close to the crack, as a quantitative description of the relationship between the stress state and thickness is lacking.
Conventional quantitative models mainly employ the crack closure mechanism to explain the effect of thickness on fatigue crack growth by introducing the parameter of thickness B into the crack closure correction factor U. The non-dimensional load ratio parameter U is defined in Eq. (1), often used to represent the effects of the R ratio (R = Fmin/Fmax) on crack closure:where ΔKeff is the effective stress intensity factor (SIF) range and ΔK (= Kmax - Kmin) is the traditional stress intensity factor range. Novelman et al. [10] employed the parameter α, a mathematic parameter based on fitting the experimental data of crack growth with different thicknesses, in his crack closure formula to describe the thickness effect. As this model can conveniently be applied in fracture mechanics analyses, further studies [11], [12], [13], [14] used it and built on it. Roychowdhury and Dodds [15] and Kotousov et al. [6] proposed another type of model that introduces the governing parameter into the crack closure correction factor formula U = f(η, R). However, all these crack growth models rely on the concept of crack closure. Although thickness affects crack closure, these two effects are not synonymous. When the closure effect is weak, such as under a high stress ratio R, these crack growth models cannot invalid to account for the thickness effect.
Another common method to explain the thickness effect involves to examining the size of the plastic zone around the crack tip [6,[16], [17], [18]]. Variation of the crack growth rate is attributed to change in the plastic zone size, induced by thickness variation. Plane stress or plane strain is assumed to calculate the plastic zone size [17]. However, the stress state of an actual crack in engineering structures cannot be simply determined. Moreover, it is difficult to detect the plastic zone of an actual crack, and there are no practical formulas to apply the size of the plastic zone in crack growth models. Therefore, this method is generally limited to qualitative or theoretical analyses.
Current theories rely on the assumption of plane stress or plane strain condition. However, actual cracks in engineering structures, even simple straight-through ones, are always under a triaxial stress state, as illustrated by the phenomenon of a straight crack front tending to become a bowed curve during crack propagation [19]. The presence of cracks introduces complex triaxial stress states in the region surrounding the crack tips, and the stress state varies from the surface to the internal region in the transverse direction [20]. The region close to the surface is less constrained, so the plane stress state dominates, while the internal region is highly constrained by both sides, so the plane strain state dominates. The stress state transition is highly affected by crack thickness [20,21]. Therefore, the stress state should be part of the crack growth model to reflect the mechanical principle of the thickness effect.
In the last decades, the crack constraint parameters were employed to characterize the effect of thickness on the fracture behavior. The constraint parameters have been introduced into the fracture toughness expressions for different thicknesses [22,23]. The influence of the constraint on the creep crack growth were evaluated [3,24,25]. Guo et al. [24] used the constraint factor Tz to define a new creep SIF. Xu et al. [25] proposed an in-plane constraint Q* to quantify the crack tip constraint level under creep conditions, and stated that the thickness has a significant effect on the crack constraint.
The present paper focuses on developing a methodology to propose a fatigue crack growth model that considers the thickness effect for through cracks. A characteristic parameter to quantify the stress state variation depending on thickness is defined based on the crack constraints, then, it was introduced into crack growth models to account for the thickness effect. The paper is organized as follows. Section 2 gave a brief background of two out-of-plane constraints used in this paper. Section 3 described the outline and working flow of the approach to illustrate the procedure of model development. In Section 4, three finite element (FE) models of compact tension (CT) specimens are set up according to an experiment. In Section 5, parameter identification of the formulas used in the approach are carried out. Section 6 is the discussion of the approach. In summary, the out-of-plane constraint factors are used to characterize the crack stress state, and a novel crack propagation model is set up by introducing the crack stress state level.
Section snippets
Theories of out-of-plane constraint approach
The framework of a single driving force parameter such as the SIF or J-integral sometimes has limited ability to fully characterize the near-tip crack fields [26]. Thus, the constraint parameters were employed as the second argument to improve the characterization [27]. There are two types of constraints used to describe the stress field near the crack tip, i.e., in-plane and out-of-plane constraints. The out-of-plane constraints are related to the thickness and strongly influenced by the crack
Outline of the approach to establish the model
To model the crack growth of the alloy material with different thicknesses using the approach in this paper, experimental data of the crack growth rate from three thicknesses of specimens and out-of-plane constraint distributions of these cracks are required. The outline of the approach will be described in the following section.
In context of the actual 3D crack, there is a “nonuniform distribution” of the stress state through the thickness: the plane stress state close to the surface and plane
Fatigue crack growth experiment
The crack growth data employed here come from the experiment in [34]. The experiment utilizing CT specimens made of aluminum alloy 6082 with three thicknesses, the stress ratio R is 0.1, and the maximum loadings are 2, 6 and 12.5 kN for thickness B = 3, 10 and 25 mm respectively. The details of the crack growth experiment can be found in reference [34]. The crack growth rate curves of the experiment are shown in Fig. 6, and the specific parameters of the crack growth rate formula Eq. (15),
Crack constraint distributions through the thickness
In this section, the constraint distributions are calculated, and appropriate empirical formulas are determined to express the distributions through the thickness. Then, a parameter is proposed to characterize the crack stress state conditions. With this parameter, the relationship between the crack thickness and stress state was expected to be quantitatively assessed. Additionally, for the convenience of unified analysis, the distances along the thickness of all specimens are normalized to 0
Discussion about characteristic parameter P
A character parameter P was proposed in this paper to describe the phenomena of a steady phase in the out-of-plane constraint distribution curves in the middle position (Figs. 12 and 16), so the physical meaning of P is the stress state level assessment for a crack front.
The values of P were determined from the slope curve (Figs. 18 and 22). A criterion line paralleling to the X-axis, indicating a critical slope value, needs to be selected beforehand. The criterion line for Tz was selected from
Conclusion
The thickness effect on crack growth is related to the physical nature and mechanism behind the phenomenon. In the present paper, an approach was proposed to model fatigue crack growth for through cracks considering the thickness effect. Following the hypothesis that cracks thickness changes the stress state of the crack front, and thus the crack growth rate, the relationship between the thickness and the crack stress state was investigated.
Crack growth simulations of three thicknesses of CT
CRediT authorship contribution statement
He Liu: Conceptualization, Methodology, Software, Data curation, Writing - original draft, Visualization. Xiaoguang Yang: Conceptualization, Writing - review & editing, Supervision, Funding acquisition, Project administration, Funding acquisition. Shaolin Li: Writing - review & editing, Supervision, Funding acquisition. Duoqi Shi: Resources, Project administration, Funding acquisition. Hongyu Qi: Resources, Project administration, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This research is sponsored by the Aeronautics Power Foundation(No. 6141B09050332), National Science and Technology Major Project (2017-IV-0012-0049) and the Fundamental Research Funds for the Central Universities (YWF-19-BJ-J-338).
References (44)
- et al.
Effect of specimen thickness on fatigue crack growth rate
Nucl Eng Des
(2000) - et al.
Characterizing crack growth behavior and damage evolution in P92 steel under creep-fatigue conditions
Int J Mech Sci
(2017) - et al.
A crack closure model of fatigue crack growth in plates of finite thickness under small-scale yielding conditions
Mech Mater
(2009) Three-dimensional analyses of plastic constraint for through-thickness cracked bodies
Eng Fract Mech
(1999)- et al.
A new constraint based fracture criterion for surface cracks
Eng Fract Mech
(2007) - et al.
3D constraint effect on 3D fatigue crack propagation
Int J Fatigue
(2005) - et al.
A numerical investigation of 3-D small-scale yielding fatigue crack growth
Eng Fract Mech
(2003) - et al.
Stress intensity factor analysis of through thickness effects
Int J Fatigue
(2013) - et al.
Three-dimensional analysis of an edge crack in a plate of finite thickness with the first-order plate theory
Theor Appl Fract Mech
(2018) - et al.
Three-dimensional crack front fields in a thin ductile plate
J Mech Phys Solids
(1990)
Effects of crack front curvature on J–R curve testing using clamped SE (T) specimens of homogeneous materials
Int J Press Vessels Pip
Experimental investigation of out-of-plane constraint effect on fracture toughness of the SE (T) specimens
Int J Mech Sci
Equivalent thickness conception for corner cracks
Int J Solids Struct
Quantifying the creep crack-tip constraint effects using a load-independent constraint parameter Q
Int J Mech Sci
Predictions of fatigue crack growth in aluminium alloy 2024–T351 using constraint factors
Int J Fatigue
Influence of the -stress on the 3-D stress state around corner cracks in an elastic plate
Eng Fract Mech
Elastoplastic three dimensional crack border field—I. Singular structure of the field
Eng Fract Mech
The in-plane and out-of-plane stress constraint factors and K-T-Tz description of stress field near the border of a semi-elliptical surface crack
Fatigue Fract Eng Mater Struct
Determination of elastic T -stress along three-dimensional crack fronts using an interaction integral
Int J Solid Struct
Experimental -stress formulation of test specimen thickness effect on fracture toughness in the transition temperature region
Eng Fract Mech
A numerical approach to simulate 3D crack propagation in turbine blades
Int J Mech Sci
Notched M(T) specimen for plane strain studies
Int J Fatigue
Cited by (11)
An adaptive multi-patch isogeometric phase-field model for fatigue fracture
2024, International Journal of Mechanical SciencesEffect of thermal degradation on mechanical properties of cast Al-Si alloys for cylinder heads
2023, Engineering Fracture MechanicsA new effective stress intensity factor approach to determine thickness-independent fatigue crack growth rate curves
2022, Theoretical and Applied Fracture MechanicsCitation Excerpt :This supports the hypothesis that cracks do not evolve from a specific advanced crack size, as demonstrated by Escalero et al. in a numerical study [17]. Moreover, this findings also aligns with a recent study by Liu et al. [4], in which the researchers found that the approximate crack size to achieve the stable stage is 40% of the specimen thickness. Regarding thickness effect, TD increased with increasing thickness, and the differences between the maximum and minimum specimen thickness were equal to 69% for BM 1 and 76% for BM 2.
Modeling of the fatigue crack growth of nickel-based superalloy using a constraint-based approach considering thickness
2022, Engineering Fracture MechanicsCitation Excerpt :Therefore, we proposed an approach [28] for modeling FCG, which attributes the thickness effect to variations in the stress state induced by the specimen thickness, and the out-of-plane constraint was employed to quantify the stress state level for different thicknesses. The constraint-based approach has been successfully employed to model the thickness-dependent FCG of aluminum alloy AA6082 (BS HE30) [30]. The applicability of the proposed approach to the nickel-based superalloy, GH4169, was verified in the present paper.
Interpretable Machine Learning Method for Modelling Fatigue Short Crack Growth Behaviour
2024, Metals and Materials International