Abstract

Nowadays, a main task in the production sector is the optimization of components and production processes in order to reduce costs and improve the quality of products. Nu- merical simulations have become a fundamental tool, as they allow the optimization even before the actual production starts. For the simulation of the mechanical behavior of component parts, an appropriate constitutive material model is necessary. Since ma- terial behavior is usually very complex, models are developed for a specific application area. The zinc die casting alloy Zamak 5 is widely used in the automotive industry because of its excellent castability and its good mechanical properties. However, it exhibits a complex thermo-mechanical behavior, which additionally changes gradually over the course of time. This effect is known as aging and is caused by microstructural changes such as diffusion, phase transformation and precipitation in the material. These pro- cesses are temperature activated. This thesis proposes a material model of thermo-viscoplasticitywith the goal of repro- ducing the thermo-mechanical behavior of the alloy Zamak 5 considering the influence of natural aging in finite element computations. This is performed on the basis of an extensive experimental campaign, in which the thermo-mechanical behavior is charac- terized with tension, compression and torsion tests at different loading conditions and for different aging times. The material shows a strong rate and temperature dependence and a moderate dependence on natural aging. Additionally, several thermo-physical properties which are necessary in the heat equation are also measured. Moreover, the microstructure of the alloy is investigated with a scanning electron microscope and X- ray diffractometry for different aging times in order to describe the microstructural pro- cesses which take place during aging. At first, the material model is developed for the small deformation case since, in this case, the torsion tests can be considered as one-dimensional and purely deviatoric, which is advantageous during the identification process. Moreover, several temperature and aging dependent material functions are developed during the identification process. The model is extended later on to the finite strain case in order to be able to compute processes in which the strains are larger than 5%. The models are based on the additive decomposition of the strain tensor (for the small strain model) or multiplicative decom- position of the deformation gradient (for the finite strain model). In both cases, there is a component representing each of the different effects contained in the model: aging, temperature and rate-dependence. Furthermore, the total stress is decomposed into an equilibrium and an overstress part in order to identify them in a partitioned manner. The model is implemented in the finite element code Abaqus. The behavior of the model is demonstrated using several simulation examples. Finally, simulations of a real component part (steering-wheel lock of a car), and a cylinder with a hole are compared with complex experiments in which inhomogeneous stress and strain states are reached for different loading conditions. With this information, the developed material model is validated.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call