High cycle thermo-mechanical fatigue: Damage operator approach

The stress-life approach is standardized and widely accepted for determining fatigue damage under stress-controlled high cycle fatigue (HCF) loading. It was first extended to non-isothermal cases by introducing an equivalent temperature approach (ETA). The paper presents its extension, that is, the damage operator approach (DOA), enabling online continuous damage calculation for isothermal and non-isothermal loading with mean stress correction. The cycle closure point, cycle equivalent temperature, threshold temperature and separate rainflow counting, obligatory for the ETA, are not necessary for the DOA any more. Both approaches are equivalent for the second and subsequent runs of block loading if temperature is constant. However, for non-isothermal cases the DOA is within the worst- and the best-case scenarios of the ETA. The approaches are compared with the simple stress history and several thermo-mechanical fatigue (TMF) cycle types.     

Experimental and theoretical studies on thermo‐mechanical fatigue test for aluminium cast alloy

Evaluation of the thermo‐mechanical behaviour and prediction of the service life of cast aluminium alloys are important for the design of automobile engine cylinder heads. In this study, cast Al alloy specimens are extracted from cylinder heads and subjected to in‐phase thermo‐mechanical cyclic loading. The hysteresis curves related to stress and strain were recorded under the individual thermo‐mechanical loading conditions. The number cycles to failure corresponding to multiple mechanical strain and temperature ranges were obtained. It is found that the cyclic stress amplitude decreases and the cyclic softening rate increases with increasing maximum temperature rise. A modified fatigue‐creep model based on energy conservation has been developed for prediction of the fatigue life of cylinder heads. The proposed method shows good agreement with the well‐established Ostergren model and low standard deviations. In summary, the proposed method described in this study provides an option for prediction of the thermo‐mechanical behaviour of metals.     

Experimental and computational study on thermo‐mechanical fatigue life of aluminium alloy piston

To assess the life of a new diesel aluminium alloy piston under thermal shock loads, thermo‐mechanical fatigue (TMF) testing was conducted to characterise the TMF properties of the piston alloy, and an empirical model based on the constraint ratio concept was proposed to predict the TMF life of the piston. Considering that the empirical model required expensive experimental support, a platform with high‐frequency induction heating was established to simulate the force on the piston under thermal shock loads to calculate the piston life using the thermal shock test. Additionally, a finite element method was developed to compute the distributions of temperature, strain, and stress during this process. The characteristics of crack initiation and propagation in TMF test rods and piston mock‐ups were also investigated. The results showed that the TMF test rod suffered brittle fracture with brittle quasi‐cleavage features. The microcracks mainly occurred in primary Si particles due to stress concentration around the primary Si particles induced by the difference between the thermal expansion coefficients of Si and Al. From a macro perspective, the piston initially cracked at the rim above the pinhole, where the stress is larger than that along other directions. From a micro perspective, the protrusions of various sizes on the piston rim were induced by the compression stresses at high temperature. The piston cracks usually initiate around primary Si particles, propagate along the edge of primary Si in a straight line, bifurcate and then stop at a certain depth. If the piston was only heated, cracks or plastic deformations were not produced. The piston life can be assessed using the proposed empirical model based on the constraint ratio concept or thermal shock testing based on the developed platform. The difference between the predicted and experimental life was not greater than 7%.     

Improved fatigue resistance of pseudo‐elastic NiTi alloys by thermo‐mechanical treatment

Shape memory alloys are susceptible to two types of fatigue in addition to classical fatigue: 1. Pseudo‐elastic fatigue leads to an increase in the slope of the pseudo‐elastic plateau and final loss of pseudo‐elasticity 2. A change in transformation temperature. Usually the martensite temperature is lowered with the number of cycles until final loss of transformability. This paper describes measures to improve stability against both types of fatigue. Such methods are simple ageing in order to achieve precipitation in austenite, and thermo‐mechanical treatments such as ausforming that introduce lattice defects into austenite, which transforms subsequently into martensite. Another method consists in the introduction of defects into martensite by marforming plus subsequent ageing. This ageing treatment has two purposes. It increases the classical strength and restores the β‐phase from residual martensite and consequently it recreates transformability. It is shown that the last mentioned method leads to the greatest effect in respect to stabilisation against both types of fatigue. An additional effect of these treatments is a transition of localised to more homogeneous strain. Its relevance for fatigue resistance is still under investigation.     

Dislocation distributions in monocrystalline nickel‐based superalloys subjected to thermo‐mechanical fatigue with and without hold times

The characteristics of dislocation configurations under thermo‐mechanical fatigue cycling were investigated in [001] oriented nickel‐based single‐crystal superalloys. Thermo‐mechanical fatigue tests were performed on TMS‐75 (without hold time) and TMS‐82 (with hold time) superalloys. The specimens were subsequently studied by transmission electron microscopy under two‐beam conditions. In TMS‐75 superalloy, cross‐slipping is the main characteristic for the low number of dislocations. In TMS‐82 superalloy, more complex process of dislocation configurations has been demonstrated in detail, involving five stages: after the first stress relaxation, after the first tensile plastic deformation, after the second stress relaxation, after 30 cycles and after rupture. In addition, for TMS‐82 superalloy, there is a reversible movement behaviour of stacking faults that occur in compression and disappear in tension. After rupture, the number of dislocation is related to the hold time. Longer hold time could generate a higher degree of stress relaxation and produce more dislocations with climbing characteristic.     

A new energy‐based isothermal and thermo‐mechanical fatigue lifetime prediction model for aluminium–silicon–magnesium alloy

In this paper, a new fatigue lifetime prediction model is presented for the aluminium–silicon–magnesium alloy, A356.0. This model is based on the plastic strain energy density per cycle including two correction factors in order to consider the effect of the mean stress and the maximum temperature. The thermal term considers creep and oxidation damages in A356.0 alloy. To calibrate the model, isothermal fatigue and out‐of‐phase thermo‐mechanical fatigue (TMF) tests were conducted on the A356.0 alloy. Results showed an improvement in predicting fatigue lifetimes by the present model in comparison with classical theories and also the plastic strain energy density (without any correction factors). Therefore, this model is applicable for TMF, low cycle fatigue (LCF) and both TMF/LCF lifetimes of the A356.0 alloy. Furthermore, this model can be easily used for the estimation of thermo‐mechanical conditions in components such as cylinder heads.     

A weight function-critical plane approach for low-cycle fatigue under variable amplitude multiaxial loading

Low‐cycle fatigue data of type 304 stainless steel obtained under axial‐torsional loading of variable amplitudes are analyzed using four multiaxial fatigue parameters: SWT, KBM, FS and LKN. Rainflow cycle counting and Morrow's plastic work interaction rule are used to calculate fatigue damage. The performance of a fatigue model is dependent on the fatigue parameter, the critical plane and the damage accumulation rule employed in the model. The conservatism and non‐conservatism of predicted lives are examined for some combinations of these variables. A new critical plane called the weight function‐critical plane is introduced for variable amplitude loading. This approach is found to improve the KBM‐based life predictions.     

Multiaxial thermo‐mechanical fatigue on material systems for gas turbines

Material systems made from nickel based superalloys with protective coatings have been tested in thermo‐mechanical fatigue with superposed thermal gradients, which generated multiaxial stress states. The testing conditions were selected for simulating the fatigue loading in the wall of an internally cooled gas turbine blade of an aircraft engine. After thermo‐mechanical testing the damage behaviour of the materials has been investigated by means of microscopic methods. The laboratory experiments have been accompanied by numerical simulations. Based on the results of the simulations and observed damage features the test parameters in subsequent laboratory tests have been controlled to facilitate the validation of models describing the initiation and propagation of damages. This contribution gives an overview over results on the influence of multiaxial stress states on (i) oriented deformation and coagulation of γ’‐precipitates (‘rafting’) in the substrate, (ii) on morphological instabilities of the surface of metallic oxidation protection coatings (‘rumpling’), and (iii) on crack initiation and growth in material systems with additional ceramic thermal barrier coating.     

Modelling of reinforced polymer composites subject to thermo‐mechanical loading

A coupled finite element model is developed to analyse the thermo‐mechanical behaviour of a widely used polymer composite panel subject to high temperatures at service conditions. Thermo‐chemical and thermo‐mechanical models of previous researchers have been extended to study the thermo‐chemical decomposition, internal heat and mass transfer, deformation and the stress state of the material. The phenomena of heat and mass transfer and thermo‐mechanical deformation are simulated using three sets of governing equations, i.e. energy, gas mass diffusion and deformation equations. These equations are then assembled into a coupled matrix equation using the Bubnov–Galerkin finite element formulation and then solved simultaneously at each time interval. An experimentally tested 1.09 cm thick glass‐fibre woven‐roving/polyester resin composite panel is analysed using the numerical model. Results are presented in the form of temperature, pore pressure, deformation, strain and stress profiles and discussed. The maximum normal stress failure criterion is used in order to establish the load‐bearing capability of the composite panel. Significant pore gas pressure build‐ups (to 0.8 MPa and higher) have been perceived at high thermo‐chemical decomposition rates where the material experiences a complex expansion/contraction phenomenon. It is found that the composite panel experiences structural instability at elevated temperatures up to 300°C but retains its integrity even under moderate external loading. Copyright © 2005 John Wiley & Sons, Ltd.     

A cumulative damage model for fatigue life estimation of high‐strength steels in high‐cycle and very‐high‐cycle fatigue regimes

A cumulative fatigue damage model is presented to estimate fatigue life for high‐strength steels in high‐cycle and very‐high‐cycle fatigue regimes with fish‐eye mode failure, and a simple formula is obtained. The model takes into account the inclusion size, fine granular area (FGA) size, and tensile strength of materials. Then, the ‘equivalent crack growth rate’ of FGA is proposed. The model is used to estimate the fatigue life and equivalent crack growth rate for a bearing steel (GCr15) of present investigation and four high‐strength steels in the literature. The equivalent crack growth rate of FGA is calculated to be of the order of magnitude of 10^?14–10^?11 m/cycle. The estimated results accord well with the present experimental results and prior predictions and experimental results in the literature. Moreover, the effect of inclusion size on fatigue life is discussed. It is indicated that the inclusion size has an important influence on the fatigue life, and the effect is related to the relative size of inclusion for FGA. For the inclusion size close to the FGA size, the former has a substantial effect on the fatigue life. While for the relatively large value of FGA size to inclusion size, it has little effect on the fatigue life.     

A rapid path‐length searching procedure for multi‐axial fatigue cycle counting

In this paper, a series of advanced searching algorithms have been examined and implemented for accelerating multi‐axial fatigue cycle counting efforts when dealing with large time histories. In a computerized calculation of the path‐length dependent cycle counting method, most of the central processor unit's (CPU) time is spent on searching for the maximum range or distance in a stress or strain space. A brute‐force search is the simplest to implement, and will always find a solution if it exists. However, its cost, in many practical problems, tends to grow exponentially as the size of the loading spectrum increases with a search time measured in the order of O(n²), where n is the number of spectrum data points. In contrast, a form of Andrew's monotone chain algorithm, as demonstrated in this paper, can remarkably reduce the solution time to the order of O(n log n). The effectiveness of the new path‐length searching procedure is demonstrated by a series of worked examples with a varying degree of non‐proportionality in multi‐axial loading history.     

A thermo‐damage strength model for ceramic material including an embedded crack with an arbitrarily shaped front

A fracture strength model applied at room temperature for the embedded crack was obtained based on the stress intensity factor model for embedded crack with an arbitrarily shaped front and Griffith fracture criterion. With further research on temperature effect on strength of ultra high temperature ceramics (UHTCs), a thermo‐damage strength model including the effects of temperature and crack shape was applied to each temperature phase. The fracture strengths were studied and analyzed. The conclusions were compared with the results obtained using the finite element software ABAQUS. This study will provide a theoretical basis and guidance on the design and preparation of the UHTCs.     

ANN approach to sorption hysteresis within a coupled hygro‐thermo‐mechanical FE analysis

Non‐linear deformable porous media with sorption (capillary condensation) hysteresis are considered. An artificial neural network with two hidden layers is trained to interpolate the sorption hysteresis using a set of experimental data. The performance of the ANN, which is applied as a procedure in the FE code, is investigated, both from numerical, as well as from physical viewpoint. The ANN‐FE code has been developed and tested for 1‐D and 2‐D problems concerning cyclic wetting–drying of concrete elements. In general, the application of the ANN procedure inside the classical FE program does not have any negative effect on the numerical performance of the code. The results obtained indicate that the sorption isotherm hysteresis is of importance during analysis of hygrothermal and mechanical behaviour of capillary‐porous materials. The most distinct differences are observed for the saturation and displacement solutions. The ANN‐FE approach seems to be an efficient way to take into account the influence of hysteresis during analysis of hygro‐thermal behaviour of capillary‐porous materials. Copyright © 2001 John Wiley & Sons, Ltd.     

Fractional step methods for thermo‐mechanical damage analyses at transient elevated temperatures

In the interest of computational efficiency this paper describes the implementation of a coupled thermo‐damage constitutive model into a coupled time‐stepping analysis using fractional step methods. To begin it is demonstrated that a thermo‐damage model can be presented in a thermodynamic framework with the evolution equations satisfying the first and second laws of thermodynamics. The equations of evolution are partitioned in two ways, thus defining two fractional step methods: an isothermal method and an isentropic method. When implemented into a time‐stepping algorithm the isentropic method maintains a precise energy balance for the entire analysis where as the isothermal method can only provide an energy balance at the end of each thermal time step. In addition, a stability analysis shows that the isentropic analysis is unconditionally stable while a isothermal analysis is at best conditionally stable. Simulations of thermal fracture in a restrained specimen under heat show stable growth of damage to failure. Copyright © 2000 John Wiley & Sons, Ltd.     

Online multiaxial fatigue damage evaluation method by real‐time cycle counting and energy‐based critical plane criterion

To realize online multiaxial fatigue damage assessment for the mechanical components in service, an online multiaxial cycle counting method is proposed coupled with the segment processing technique and Wang‐Brow's relative equivalent strain concept. Meanwhile, considering all the stress and strain components, which contribute to the fatigue damage on the critical plane, a multiaxial fatigue damage model without any weight coefficients is also proposed in an equivalent form of shear strain energy. Then, an online fatigue damage evaluation method for multiaxial random loading is developed by combining with the proposed damage model and online cycle counting method. The experimental results showed that the proposed online cycle counting method can be successfully applied to the calculation of multiaxial fatigue damage under random loading. Moreover, the proposed online multiaxial fatigue damage evaluation method can provide satisfactory predictions.     

An evaluation of limited‐memory sparse linear solvers for thermo‐mechanical applications

We consider the performance of sparse linear solvers for problems that arise from thermo‐mechanical applications. Such problems have been solved using sparse direct schemes that enable robust solution at the expense of memory requirements that grow non‐linearly with the dimension of the coefficient matrix. In this paper, we consider a class of preconditioned iterative solvers as a limited‐memory alternative to direct solution schemes. However, such preconditioned iterative solvers typically exhibit complex trade‐offs between reliability and performance. We therefore characterize such trade‐offs for systems from thermo‐mechanical problems by considering several preconditioning schemes including multilevel methods and those based on sparse approximate inversion and incomplete matrix factorization. We provide an analysis of computational costs and memory requirements for model thermo‐mechanical problems, indicating that certain incomplete factorization schemes can achieve good performance. We also provide empirical evaluations that corroborate our analysis and indicate the relative effectiveness of different solution schemes. Our results indicate that our drop‐threshold incomplete Cholesky preconditioning is more robust, efficient and flexible than other popular preconditioning schemes. In addition, we propose preconditioner reuse to amortize preconditioner construction cost over a sequence of linear systems that arise from non‐linear solutions in a plastic regime. Copyright © 2007 John Wiley & Sons, Ltd.     

Gradient‐enhanced damage modelling of high‐cycle fatigue

Continuum damage mechanics can be used to model the initiation and growth of fatigue cracks. However, finite element analyses using standard fatigue damage formulations exhibit an extreme sensitivity to the spatial discretisation of the problem. The mesh sensitivity is caused by the fact that the underlying continuum model predicts instantaneous, perfectly brittle crack growth as soon as a crack has been initiated. The growth of damage localizes in a vanishing volume during this instantaneous growth. This localization is not so much due to loss of ellipticity of the problem, but is caused by the fact that the damage rate is singular at the crack tip. The damage rate singularity can be removed by the introduction of higher‐order deformation gradients in the constitutive modelling. As a result, crack growth at a finite rate and with a positive amount of energy dissipation is predicted. Finite element analyses converge to this solution and are thus no longer pathologically dependent on the spatial discretization. Copyright © 2000 John Wiley & Sons, Ltd.     

A fatigue damage indicator parameter for P91 chromium‐molybdenum alloy steel and fatigue pre‐damaged P54T carbon steel

Two grades of structural steel were subjected to fully reversible, constant stress amplitude cyclic loading. The local strain response of the material was measured and recorded during the test, with the applied testing technique enabling the monitoring of hysteresis loop variation for the narrowest cross‐section of the hourglass specimen. Changes in hysteresis loop width, representing the local inelastic response of the material, were recorded in order to monitor the density of structural imperfections. Material ratcheting behaviour was observed as changes in the mean strain for selected load cycles. Ratcheting was attributed to local deformation of the material in the vicinity of imperfections such as voids or inclusions, as well as deformation induced by the propagation of microcracks. Definitions of a damage indicator parameter and damage parameter were proposed. The fatigue behaviour of the two investigated grades of steel was finally illustrated in the form of damage curves for different stress amplitudes and for undamaged and fatigue pre‐damaged material.