Refine
Document Type
- Article (reviewed) (6) (remove)
Language
- English (6)
Has Fulltext
- no (6) (remove)
Is part of the Bibliography
- yes (6)
Keywords
- Plastizität (6) (remove)
Institute
- Fakultät Maschinenbau und Verfahrenstechnik (M+V) (6) (remove)
Open Access
- Closed Access (4)
- Closed (1)
- Open Access (1)
In this paper, the J-integral is derived for temperature-dependent elastic–plastic materials described by incremental plasticity. It is implemented using the equivalent domain integral method for assessment of three-dimensional cracks based on results of finite-element calculations. The J-integral considers contributions from inhomogeneous temperature fields and temperature-dependent elastic and plastic material properties as well as from gradients in the plastic strains and the hardening variables. Different energy densities are considered, the Helmholtz free energy and the stress-working density, providing a physical meaning of the J-integral as a fracture criteria for crack growth. Results obtained for a plate with two different crack configurations each loaded by a cool-down thermal shock show domain-independence of the incremental J-integral for different energy densities even for high temperature gradients and significant temperature-dependence of the yield stress and the hardening exponent in the presence of large scale yielding. Hence, the derived J-integral is an appropriate parameter for the assessment of cracks in thermomechanically loaded components.
High temperature components in internal combustion engines and exhaust systems must withstand severe mechanical and thermal cyclic loads throughout their lifetime. The combination of thermal transients and mechanical load cycling results in a complex evolution of damage, leading to thermomechanical fatigue (TMF) of the material. Analytical tools are increasingly employed by designers and engineers for component durability assessment well before any hardware testing. The DTMF model for TMF life prediction, which assumes that micro-crack growth is the dominant damage mechanism, is capable of providing reliable predictions for a wide range of high-temperature components and materials in internal combustion engines. Thus far, the DTMF model has employed a local approach where surface stresses, strains, and temperatures are used to compute damage for estimating the number of cycles for a small initial defect or micro-crack to reach a critical length. In the presence of significant gradients of stresses, strains, and temperatures, the use of surface field values could lead to very conservative estimates of TMF life when compared with reported lives from hardware testing. As an approximation of gradient effects, a non-local approach of the DTMF model is applied. This approach considers through-thickness fields where the micro-crack growth law is integrated through the thickness considering these variable fields. With the help of software tools, this method is automated and applied to components with complex geometries and fields. It is shown, for the TMF life prediction of a turbocharger housing, that the gradient correction using the non-local approach leads to more realistic life predictions and can distinguish between surface cracks that may arrest or propagate through the thickness and lead to component failure.
In this paper, the temperature dependent cyclic mechanical properties of the martensitic hot work tool steel 1.2367 after tempering are investigated. To this end, hardness measurements as well as monotonic and cyclic tests at temperatures in the range from room temperature to 650 °C are performed on material tempered for different tempering times and temperatures. To describe the observed time and temperature dependent softening during tempering a kinetic model for the evolution of the mean size of secondary carbides based on Ostwald ripening is developed. Furthermore, mechanism-based as well as phenomenological relations for the cyclic mechanical properties of the Ramberg-Osgood model depending on carbide size and temperature are introduced. A good overall agreement of the measured and the calculated stress-strain hysteresis loops for different temperatures and heat treatments is obtained using the determined material properties of the kinetic and mechanical model.
In this paper, an unconditionally stable algorithm for the numerical integration and finite-element implementation of a class of pressure dependent plasticity models with nonlinear isotropic and kinematic hardening is presented. Existing algorithms are improved in the sense that the number of equations to be solved iteratively is significantly reduced. This is achieved by exploitation of the structure of Armstrong-Frederik-type kinematic hardening laws. The consistent material tangent is derived analytically and compared to the numerically computed tangent in order to validate the implementation. The performance of the new algorithm is compared to an existing one that does not consider the possibility of reducing the number of unknowns to be iterated. The algorithm is used to implement a time and temperature dependent cast iron plasticity model, which is based on the pressure dependent Gurson model, in the finite-element program ABAQUS. The implementation is applied to compute stresses and strains in a large-scale finite-element model of a three cylinder engine block. This computation proofs the applicability of the algorithm in industrial practice that is of interest in applied sciences.
In this paper the yield surface of a recently presented microstructure-based volume element of the gray cast iron material GJL-250 is assessed after different plastic loading histories. The evolution of the yield surface is investigated for different volumetric, deviatoric and uniaxial loadings. The micromechanical material properties of the metallic matrix and the graphite inclusions are validated by means experimental stress-strain hysteresis loops. The metallic matrix is modeled as elastic-plastic with a non-linear kinematic hardening law. The graphite inclusions are described by means of a volumetric strain state dependent Young’s modulus. The results show that the shape of the yield surface does not change significantly in comparison to the initial yield surface after pure deviatoric loadings. After volumetric loadings, the dependence of the material on the Lode angle is significantly reduced. Uniaxial tensile preloadings result in a deformed yield surface, whereby the magnitude of the deformation depends on the applied load. Uniaxial preloadings to compression do not change the shape of the initial yield surface.
Cast aluminum alloys are frequently used as materials for cylinder head applications in internal combustion gasoline engines. These components must withstand severe cyclic mechanical and thermal loads throughout their lifetime. Reliable computational methods allow for accurate estimation of stresses, strains, and temperature fields and lead to more realistic Thermomechanical Fatigue (TMF) lifetime predictions. With accurate numerical methods, the components could be optimized via computer simulations and the number of required bench tests could be reduced significantly. These types of alloys are normally optimized for peak hardness from a quenched state that maximizes the strength of the material. However due to high temperature exposure, in service or under test conditions, the material would experience an over-ageing effect that leads to a significant reduction in the strength of the material. To numerically account for ageing effects, the Shercliff & Ashby ageing model is combined with a Chaboche-type viscoplasticity model available in the finite-element program ABAQUS by defining field variables. The constitutive model with ageing effects is correlated with uniaxial cyclic isothermal tests in the T6 state, the overaged state, as well as thermomechanical tests. On the other hand, the mechanism-based TMF damage model (DTMF) is calibrated for both T6 and over-aged state. Both the constitutive and the damage model are applied to a cylinder head component simulating several cycles on an engine dynamometer test. The effects of including ageing for both models are shown.