Dynamic stability of a propagating crack

In this work we investigate the stability of a straight two-dimensional dynamically propagating crack to small perturbation of its path. Willis and Movchan (J. Mech. Phys. Solids 43 (1995) 319; J. Mech. Phys. Solids 45 (1997) 591) constructed formulae for the perturbations of the stress intensity factors induced by a small three-dimensional dynamic perturbation of a nominally plane crack. Their solution is exploited here to derive equations for the in-plane and out-of-plane perturbations of the crack path making use of the Griffith fracture criterion and the principle of “local symmetry” (i.e the crack propagates so that local K_II=0). We consider a crack propagating in a body loaded by a pair of point body forces and subjected to a remote uniaxial stress, aligned with the direction of the unperturbed crack. We assume that the loading follows the crack as the crack advances and is such that the unperturbed crack is subjected to Mode I loading. We perform an analysis of the stability of the dynamic crack in a similar way as in earlier work (Obrezanova et al., J. Mech. Phys. Solids 50 (2002) 57) on the quasistatically advancing crack. We present numerical results illustrating the influence of the crack velocity on the crack stability. Numerical computations of the possible crack paths have been performed which show that at velocities of crack propagation exceeding about one-third of the speed of Rayleigh waves the crack may admit one or more oscillatory modes of instability.     

Correlations predicting frictional pressure drop and liquid holdup during horizontal gas-liquid pipe flow with a small liquid holdup

Experimental data and correlations available in the literature for the liquid holdup ε_L and the pressure gradient ΔP_TP/L for gas-liquid pipe flow, generally, do not cover the domain 0 < ε_L < 0.06. Reliable pressure-drop correlations for this holdup range are important for calculating flow rates of natural gas, containing traces of condensate. In the present paper attention is focused on reliable measurements of ε_L and ΔP_TPIL values and on the development of a phenomenological model for the liquid-holdup range 0 < ε_L < 0.06. This model is called the “apparent rough surface” model and is referred to as the ARS model. The experimental results presented in this paper refer to air-water and air-water + ethyleneglycol systems with varying transport properties in horizontal straight smooth glass tubes under steady-state conditions. The holdup and pressure gradient values predicted with the ARS model agree satisfactorily with both our experimental results and data obtained from the literature referring to small liquid-holdup values 0 < ε_L < 0.06. Further, it has been shown that in the domain 38 < < 72 mPa m the interfacial tension of the gas-liquid system has no significant effect on the liquid holdup. The pressure gradient, however, increases slightly with decreasing surface tension values.     

Experimental Investigation and Finite Element Analysis of Fatigue Crack Growth in Pipes Containing a Circumferential Semi-elliptical Crack Subjected to Bending

In this paper, a circumferential external surface flaw in a metallic round pipe under cyclic bending loading is considered. Because of very rapid changes in the geometrical parameters around the crack front region, the mesh generation of this region must be done with great care. This may lead to an increase in the run time which makes it difficult to reach valid results and conclusions. Because of the advantages of the sub-modeling technique in problems which need very high mesh density, this method is used. Stress intensity factors in mode I condition are determined using three-dimensional finite element modeling with 20 node iso-parametric brick elements in the ANSYS 9.0 standard code and the singular form of these finite elements at the crack front. In order to estimate the analysis error, the structural parameter error in energy norm criterion was used. Because of the advantages of non-dimensional analysis, this method is employed, and the stress intensity factors are normalized. For the analysis of the fatigue crack growth, the Paris law is used. The propagation path of the surface flaw is obtained from the diagram of aspect ratio versus relative crack depth. The fatigue crack growth analysis (the relative crack depth against loading cycles diagram) of different initial crack aspect ratio under cyclic loading is also considered. Fatigue shape development of initially semi-elliptical external surface defects is illustrated. The effect of the Paris exponent (material constant) on fatigue crack propagation is shown as well. Moreover, the fatigue crack growth of several specimens is assessed experimentally using a manually-constructed experimental set up. Finally, the experimental results obtained by cyclic bending loading tests are compared with the numerical results. The experimental results show good conformity with the finite element results.     

表面裂纹疲劳扩展的数值模拟(Ⅱ)

根据Paris疲劳裂纹扩展规律,对拉伸和纯弯曲疲劳载荷下表面裂纹扩展进行了数值模拟。数值模型中,用三次样条函数曲线拟合裂纹尖端,在裂纹扩展增量计算中考虑了裂纹闭合影响。裂纹形状演化的模拟结果与Newman和Raju经验公式预测结果进行了比较,表明了所采用的数值模拟方法的实用性。研究发现,裂纹闭合对疲劳裂纹扩展过程中的裂纹形状演化以及裂纹尖端的应力强度因子(SIF)分布都有明显影响。同裂纹形状演化一样,疲劳裂纹扩展过程中裂纹尖端的SIF分布表现出明显的特征。最后,建议了一个简单函数来统一描述表面裂纹尖端的SIF分布。     

The instability of rhombic cell flows

Instability of two-dimensional periodic flows with rhombic cell structure represented by the stream function Ψ=cos kx+cosy is investigated. Stability characteristics are obtained for the Reynolds number R=1, 2, 3 and 4 and the ratio of the diagonals of the cell . Variation of the critical Reynolds number R_c with k is obtained, and the square cell flow (k=1) is found to be most stable (R_c=√2). It is found that R_c → 1 as k → 0, which leads to a finite gap between this limiting R_c and R_c=√2 for K=0 (Ψ=cos y).     

The effect of crack size and specimen size on the relation between the Paris and Wöhler curves

Paris and Wöhler’s fatigue curves are intimately connected by the physics of the process of fatigue crack growth. However, their connections are not obvious due to the appearance of anomalous specimen-size and crack-size effects. In this study, considering the equations for a notched specimen (or for a specimen where failure is the result of the propagation of a main crack) and the assumption of incomplete self-similarity on the specimen size, the relations between the size-scale effects observed in the Paris and Wöhler’s diagrams are explained. In the second part of the work, the behaviour of physically short cracks is addressed and, considering a fractal model for fatigue crack growth, the crack-size effects on the Paris and Wöhler’s curves are discussed.     

对裂纹扩展规律Paris公式物理本质的探讨

首先讨论了著名力学家K.Krausz和A.S.Krausz关于Paris公式物理本质研究的成果，从材料的微观结构和裂纹尖端的应力场出发，应用位错动力学理论，热激活能理论和速率过程理论对疲劳裂纹扩展规律进行了微观到宏观的探讨。最终推导出疲劳裂纹扩展速率的一个解析表示式，该式严格地定了Paris公式的两个试验常数，赋予了Paris公式明确的物理意义，从而真实地揭示了Paris公式的物理本质，为这一经验的普遍规律奠定了理论基础。     

Reconstruction of a quasi-instantaneous image of coherent structures from hotwire signals obtained by a multi-point simultaneous measurement system

The turbulent velocity components (u, v) at 11 points in a reciprocating oscillatory turbulent flow have been measured simultaneously by a set of eleven X-type hotwire probes located in a plane perpendicular to the mean flow. Using a conditional sampling technique and a new method of data analysis for the inverse estimation of flow fields called the “virtual plate/load and MASCON model”, a quasi-instantaneous three-dimensional image of coherent structures of turbulence was first reconstructed directly from the experimental velocity data. The quasi-instantaneous image was expressed in terms of the velocity components u, v, w and the vorticity components ω_x ω_y, ω_z and we found that the large-scale coherent structure was composed of a pair of counter-rotating fluid motions with asymmetry which was quite different from that of the ensemble-averaged one. Flow patterns induced by the large-scale structure have been clarified by perspective representations visualized by computer simulations that produce timelines and streaklines of fluid particle traces. Results showed that the new experimental method was applicable for investigating the three-dimensional feature of coherent structures including asymmetry.     

Solitary waves in initially stressed thin elastic tubes

In the present work, we study the propagation of non-linear waves in an initially stressed thin elastic tube filled with an inviscid fluid. Considering the physiological conditions of the arteries, in the analysis, the tube is assumed to be subjected to a uniform inner pressure P₀ and an axial stretch ratio λ_z. It is assumed that due to blood flow, a finite dynamical displacement field is superimposed on this static field and, then, the non-linear governing equations of the elastic tube are obtained. Using the reductive perturbation technique, the propagation of weakly non-linear waves in the longwave approximation is investigated. It is shown that the governing equations reduce to the Korteweg-deVries equation which admits a solitary wave solution. It is observed that the present model equations give two solitary wave solutions. The results are also discussed for some elastic materials existing in the literature.     

Mixed mode fatigue crack growth and the strain energy density factor

The strain energy density factor S was first proposed by Sih for the prediction of the critical of the load and failure direction under monotonic, mixed mode loading condition. It seems a natural extension to apply the same concept to fatigue crack propagation. However, a close examination of the existing theory indicates that the Strain Energy Density Factor cannot logically account for the phenomena of the R-ratio effect and crack arrest. Thus, modification is necessary before the concept can be applied successfully for the prediction of mixed mode fatigue crack propagation.Based on the concept of hysteresis energy dissipation, an effective strain energy density factor range, ΔS_p,eff, is proposed for the correlation of fatigue crack growth data. ΔS_p,eff is consistent with the concept of crack closure. Experimental investigation indicates that it could predict the crack growth rates and trajectories.     

A study of the causes of cracks in bed plates of a ship's diesel engines

In cargo vessels, large machines such as engines in which part of the assembly consists of cast-iron structures are often exposed to the risk of crack formation. The present study is concerned with two diesel engines of identical type. A number of fractures were observed in the bed plates. Analysis of the type of loading and crack propagation shows that the fractures could have been predicted. The study shows a picture of the fracture causes which indicates fatigue due to pulsating loads. The fractures and cracks emanated from sharp notches. The material of the bed plates was cast iron of low strength. Fatigue with stable crack propagation has taken place.     

Fatigue crack growth in high and low strength steel under torsional loading

During loading of a crack in mode III the crack surfaces in contact slide against each other giving rise to friction, abrasion and mutual support, thereby reducing the effective stress at the crack tip (“sliding mode crack closure”). This phenomenon was investigated in a high strength steel (AISI 4340) and in a low strength steel (AISI C1018) in circumferentially notched specimens under pure cyclic torsion and combined loading (cyclic torsion plus static axial load). The influence of sliding mode crack closure on fatigue crack propagation is shown and “true” crack growth values (without the sliding mode crack closure influence) are determined on the basis of an extrapolation procedure. Explanations are given for causes of the various fracture modes observed, such as “factory roof” fracture, macroscopically flat mode III fracture and “lamella” fracture. Finally the scientific and technical importance of sliding mode crack closure is demonstrated.     

Brush-like modes of fatigue fracture in unidirectional fibre composites

Fatigue fracture of unidirectional fibre composites under tension along the fibres is discussed with account of the interaction between various mechanisms of damage such as single and multiple fibre ruptures, matrix cracking, and matrix-fibre debonding. The case of brittle fibres and a comparatively weak and ductile matrix is considered that exposes non-conventional modes of fracture, named “brush-like” cracks. Growth of such cracks under cyclic quasistatic loading is studied, and the effect of various factors on the crack growth rate is investigated.     

Limitations of growth regimes for cracks initiated in smooth fatigue specimens

The stages of the growth of small cracks initiating at natural flaws in smooth specimens subjected to fatigue loading are characterized and the dominant propagation mechanisms and corresponding fracture paths are described. Characteristic crack lengths are introduced to assess the transition between the regimes of microstructurally small cracks, physically small cracks and long cracks. A log Δσ-log a-diagram is used to derive estimates of these crack lengths. It is shown that simple formulate can be found which relate these characteristic crack lengths to mechanical and material parameters that can be measured using standard fracture mechanics specimens and fatigue tests.     

谱载荷下疲劳裂纹扩展随机规律的实验研究

简要介绍了谱载荷下疲劳裂纹扩展试验的理论基础、方法和过程。讨论了各种数据处理方法对参数估计的影响,研究了疲劳裂纹扩展的随机规律。由试验得到的疲劳裂纹扩展试验数据估计了裂纹扩展方程的参数,计算出结构可靠度曲线,通过对试验结果的分析难证了以下结论：以时间为参量的裂纹扩展随机过程模型和以裂纹长度为参量的模型在一定条件下是统一的;数据处理方法的选择与可靠性分析的结果有密切的联系;裂纹扩展寿命受裂纹扩展队机     

Effect of crack position on stress intensity factor in particle-reinforced metal-matrix composites

In this study, a numerical model was developed to study the effects of mechanical properties of the particle and matrix materials, the crack position (in particle/in matrix) and loading conditions (mode 1 and mixed-mode) in particle-reinforced metal-matrix composites. The finite element technique was used to calculate the stress intensity factors for crack at and near-interface. The Displacement Correlation Method was used to calculate the stress intensity factors K₁ and K₂. In the present model, the particle and matrix materials were modeled in linear elastic conditions. The interface crack was considered between the particle and matrix, without the presence of the interface. For near-interface crack problem, two different crack positions (in particle/in matrix) were selected. The obtained results show the key role on the stress intensity factors played by the relative elastic properties of the particle and matrix. The results also show that loading condition has an important effect on the K₂ stress intensity factor and the crack deflection angle.     

The effect of contact stress intensity factors on fatigue crack propagation

Using the technique of Dimensional Analysis the phenomenon of crack closure is modelled using the concept of a contact stress intensity factor Kc. For constant amplitude loading, a simple expression, Kc_max = g(R) ΔK, is obtained without making idealized assumptions concerning crack tip behaviour. Further, by assuming that crack closure arises from the interaction of residual plasticity in the wake of the crack and crack tip compressive stresses, the function g(R) is shown to be constant for non-workhardening materials. This implies that any dependency of Kc_max on R must be attributed to the workhardening characteristic of the material. With Kc known, an “effective” stress intensity factor Ke may be calculated and incorporated into a crack growth law of the form da/dn = f(ΔKe). From analysis, it can be deduced that for a workhardening material, Kc_max will decrease as R increases and the effective stress intensity factor will increase. This means that the fatigue crack propagation rate will increase with R, in accordance with experimental observations.     

降载勾线法在高强钢疲劳断裂试验中的应用

制作高强钢特征试件,在压、弯应力共同作用下进行疲劳断裂试验;结合疲劳辉纹产生机理,采用降载勾线法在试件断面制造出“海滩状花纹”;利用CAD软件描绘断面形貌并测量裂纹长度,用Newman-Raju理论进一步研究表面裂纹扩展的规律,并与试验值比较.结果表明降载勾线法可以解决高强度钢表面裂纹不易测量及断面难以观测到疲劳辉纹的问题;采用Newman-Raju公式模拟压弯组合应力下表面裂纹扩展形貌是合适的,但由此计算的疲劳寿命偏于保守.另外,当压弯应力比值较小时,计算扩展形貌时压应力可以忽略,而计算疲劳寿命时压应力不可忽略.     

Exact solutions to the Korteweg-de Vries-Burgers equation

We introduce a transormation which reduces the Korteweg-de Vries-Burgers (KdVB) equation, u_t + 2auu_x + 5bu_xx + cu_xxx = 0, to a quadratic form involving a new dependent variable and its partial derivatives. Exact solutions of the KdVB equation can be obtained by solving this equation. The exact form of the travelling wave solution to the KdVB equation is obtained in this paper, and its nature depends on the direction of propagation of the wave.