Thermomechanical Fatigue Loading Research Articles

Modeling the influence of initial material inhomogeneities on the fatigue life of notched components

Thermomechanical fatigue (TMF) is a low cycle fatigue process in which material life is correlated to the mechanical strain amplitude. However, it is well known that several other factors influence this life. This paper examines several of these parameters and their influence on life using experiments conducted on a second generation directionally-solidified (DS) Ni-base superalloy. The parameters considered include the influence of the temperature extremums (Tmax of either 750 or 950 °C and Tmin of either 100 or 500 °C), strain ratio (R∊), the strain-temperature phasing (in-phase (IP) and out-of-phase (OP)), the influence of dwells at the high temperature end of the cycle resulting in a creep–fatigue (CF) interaction, and material anisotropy associated with the grain growth direction (longitudinal versus transverse). Results suggest that the phasing has a primary role in controlling the mechanism of degradation. IP TMF is dominated by crack formation in volumes surrounding debonded carbides for both continuously cycling (CC) and CF at 950 and 750 °C, while OP TMF is dominated by surface oxidation and repetitive cracking of the oxide that reforms at the crack tip at 950 °C. Decreasing the Tmax to 750 °C the environmental and creep effects are reduced resulting in virtually pure fatigue exposure under OP conditions. With decreasing Tmin from 500 °C to 100 °C was observed an increase in inelastic strain amplitude and corresponding decrease in life. Variations in R∊ were found to have no significant influence on life or stabilized stress behavior. TMF loading in the transverse orientation resulted in life reductions over the longitudinal orientation due to cracks propagating in a transgranular manner. Lastly, only in material exposed to CF with a Tmax of 950 °C rafting of the γ′ precipitates was observed.

Some aspects of thermomechanical fatigue of AISI 304L stainless steel: Part I. creep- fatigue damage

Thermomechanical fatigue (TMF) tests on the austenitic stainless steel AISI 304L have been conducted under “true≓ plastic-strain control in vacuum. This report considers the damage oc-curring during TMF loading. It is shown how the temperature interval and the phasing (in-phase, out-of-phase) determine the mechanical response and the lifetime of the specimens. If creep-fatigue interaction takes place during in-phase cycling, the damage occurs inside the ma-terial, leading to intergranular cracks which reduce the lifetime considerably. Out-of-phase cy-cling inhibits creep-induced damage, and no lifetime reduction occurs, even if the material is exposed periodically to temperatures in the creep regime. A formula is proposed which allows prediction of the failure mode, depending on whether creep-fatigue damage occurs or not. At a given strain rate, the formula is able to estimate the temperature of transition between pure fatigue and creep-fatigue damage.

A non-linear micromechanics-based analysis of metal-matrix composite laminates

This paper presents a formulation that performs a direct assembly of a set a equations that relates ply stresses to the constituent microstresses and microstrains up through to the laminate level for any ply layup. The basic tenet of the non-linear laminate formulation in this analysis is that assumptions of the classical laminated plate (CLP) theory for the strain variation in the laminate still apply. Complete details of this analysis from the micromechanics level to the laminate level are presented, and appropriate results of the stress/strain response predictions for monotonic and cyclic applied loads are compared to their analytical and experimental counterparts from previous studies. The non-linear effects of viscoplasticity and fiber/matrix debonding are included in the present formulation, and their impact on both crossply and quasi-isotropic laminates of metal-matrix composites undergoing thermomechanical fatigue loads are examined.

The nonisothermal viscoplastic behavior of a titanium-matrix composite

The thermomechanical fatigue (TMF) response of a unidirectional SCS-6/Timetal 21S composite is determined using the unified inelastic-strain model of Bodner and Partom for the matrix response. The Bodner-Partom model captures the strain-rate sensitivity and time-dependent behavior of Timetal 21S for a wide range of temperatures (25–815°C) and strain rates (10 −3-10 −7s −1). For nonisothermal conditions, special terms are added to the inelastic-strain-rate expressions that account for changes in the temperature-dependent material parameters with changing temperatures. This viscoplastic model is implemented into the finite element package ADINA through user-defined subroutines. The unidirectional composite is represented by a concentric cylinder geometry formulated from axisymmetric elements. Numerical simulations predict an increase in residual stresses as the cooling rate from consolidation is increased. The composite response under in- and out-of-phase TMF loading compares well with experimental measurements. Compared to out-of-phase TMF, in-phase TMF shows considerably more ratchetting. For in-phase TMF, both the response and fatigue lives are more sensitive to variations in fiber-volume fraction than the out-of-phase case. The numerical model is consistent with the experimental observations in that the fibers control the in-phase TMF behavior, while the matrix controls the out-of-phase TMF behavior.

Thermomechanical Fatigue Behavior of a Silicon Carbide Fiber‐Reinforced Calcium Aluminosilicate Composite

Isothermal fatigue and in‐phase thermomechanical fatigue (TMF) tests were performed on a unidirectional, continuous‐fiber, Nicalon®‐reinforced calcium aluminosilicate glass‐ceramic composite ([O]16, SiC/CAS‐II). Monotonic tensile tests were performed at 1100°C (2012°F) and 100 MPa/s (14.5 ksi/s) to determine the material's ultimate strength (σult) and proportional limit (σpl). Isothermal fatigue tests at 1100°C employed two loading profiles, a triangular waveform with ramp times of 60 s and a similar profile with a superimposed 60‐s hold time at σmax. All fatigue tests used a σmax of 100 MPa (40% of σpl), R= 0.1. TMF loading profiles were identical to the isothermal loading profiles, but the temperature was cycled between 500° and 1100°C (932° and 2012°F). All fatigued specimens reached run‐out (1000 cycles) and were tested in tension at 1100°C immediately following the fatigue tests. Residual modulus, residual strength, cyclic stress‐strain modulus, and strain accumulation were all examined as possible damage indicators. Strain accumulation allowed for the greatest distinction to be made among the types of tests performed. Fiber and matrix stress analyses and creep data for this material suggest that matrix creep is the primary source of damage for the fatigue loading histories investigated.

Analysis of a [0°/90°] metal matrix composite under thermomechanical fatigue loading

The response of a [0°/90°] composite thermomechanical fatigue (TMF) loading is computed using a completely strain-coupled analysis via the finite element method. Calculations are made for in-phase and out-of-phase TMF conditions assuming either a perfect or an imperfect fiber-matrix interface. In addition, a simplified composites analysis (S-CAN) procedure is used to determine the response assuming no lateral strain coupling between the [0°] and [90°] plies. Comparison of the results from the two analyses indicates that the uncoupled analysis provides accurate strains and fiber stresses in the longitudinal direction, but individual ply (average) stresses are not as well predicted. The complete three-dimensional finite element computations indicate that results obtained using an imperfect interface provide better correlation with experimental data than those obtained assuming perfect bonding. The micromechanical stresses developed at the fiber-matrix interface are used to identify the critical locations where matrix failure is expected to occur. These locations are validated to a reasonable degree with experimental observations.

Thermomechanical fatigue of a quasi-isotropic metal matrix composite: Majumdar, B.S. and Newaz, G.M. Proc. Conf. Composite Materials: Fatigue and Fracture, Lake Buena Vista, Florida, USA, 6–7 Nov. 1989 vol 3, 732–752

In this paper, effects of strain rate and mean strain on the cyclic behavior and the lifetime of aluminum–silicon alloys are investigated under thermo-mechanical and isothermal fatigue loadings. To achieve these goals, low cycle fatigue tests are accomplished at evaluated temperatures under various strain rates (by changing the loading frequency) and different strain ratios (minimum to maximum strain). Thermo-mechanical fatigue experiments are performed in an out-of-phase condition where the temperature varies between 50 and 250 °C. Various heating/cooling rates are taken into account to assess the strain rate effect and different starting temperatures are considered to study the mean strain effect.

Prediction of thermomechanical fatigue lives in metal matrix composites

To identify the role of silicon carbide participate reinforcement on high-temperature thermomechanical fatigue behavior of Al 2xxx-T4, experiments have been conducted under thermomechanical out-of-phase and in-phase loading conditions. A general constitutive representation, based on Eshelby’s inclusion theory, is used for the determination of volumetric average stresses and strains under cyclic loading of the metal matrix composite. This constitutive representation is used with a life prediction model, based on the matrix stress-strain behavior, which predicts contributions of fatigue, creep, and environmental damages to failure under both isothermal and thermomechanical fatigue loading. In isothermal fatigue experiments at 200 °C and 300 °C, pure fatigue damage and creep damage are the dominant damage mechanisms in the short-life regime. In the long-life regime, however, the stress levels are too low to induce considerable creep damage; so, oxidation damage becomes dominant. When fatigue damage is dominant, the model predicts a decrease in life, based on strain range, with increasing volume fraction of reinforcement. Based on stress range, improved fatigue lives are predicted with increasing volume fraction of reinforcement. The reinforced alloy exhibits longer lives when compressive hydrostatic stresses in the matrix at the high-temperature end of the cycle reduce the creep damage.

Thermo-mechanical fatigue behaviour and life prediction of C-1023 nickel based superalloy

Nickel based superalloys are used for manufacturing turbine blades and vanes components due to their ability to withstand high stress levels at high temperatures. The complex thermo-mechanical fatigue loadings that those components suffer (as a result of start ups and shutdowns) make life assessment a complex task. Working towards this end, the thermo-mechanical fatigue behaviour and life prediction of C-1023 nickel based superalloy was studied in this work. Several tests were conducted at different phase-shifts between mechanical and thermal cycles. It was found out that in-phase tests (where peak temperature coincides with maximum mechanical strain) were prone to intergranular fracture due to creep contribution. On the other hand, out of phase tests showed clear transgranular paths ascribed to fatigue damage. Finally, some simple damage parameters were found to be suitable tools for life prediction.