Austenitic Grain Size Prediction in Hot Forging of a 20MnCr5 Steel by Numerical Simulation Using the JMAK Model for Industrial Applications

Materials Research. 2019; 22(5): e20190230 DOI: http://dx.doi.org/10.1590/1980-5373-MR-2019-0230 Austenitic Grain Size Prediction in Hot Forging of a 20mncr5 Steel by Numerical Simulation Using the JMAK Model for Industrial Applications Thiago Marques Ivaniskia* , Jérémy Eppb, Hans-Werner Zochb, Alexandre da Silva Rochaa a Laboratório de Transformação Mecânica (LdTM), Universidade Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brasil b Leibniz-Institut für Werkstofforientierte Technologien (IWT), Bremen, Germany Received: March 11, 2019; Revised: September 26, 2019; Accepted: October 2, 2019 Yield strength and toughness in steels are directly associated with hot forging processes, especially by controlling austenitic grain size and cooling conditions. The phenomenological JMAK model in macroscale has been applied in different material classes to predict grain size after hot forming. However, on an industrial application, there is still a lack of understanding concerning the synergic effects of strain rate and temperature on recrystallization. This preliminary study aimed at investigating the applicability of coupled semi-empirical JMAK and visco-elastoplastic models in numerical simulation to predict austenitic grain size (PAGS). Hot forging of cylindrical samples of a ferritic-perlitic DIN 20MnCr5 steel was performed followed by water quenching. The main influences, such as temperature, strain and strain rate fields following the recrystallization model were investigated using the subroutine of FORGE NxT 2.1 software. The results were evaluated by comparing experimentally measured and simulated PAGS at process end. The forging process generates different strain and strain rate fields in the workpiece, which in turn lead to a variation in the PAGS and recrystallization fractions. The simulation was able to detect the PAGS variation showing a good agreement between the experimental forging results and the applied model. Keywords: Numerical simulation, JMAK’s model, hot forging, grain size 1. Introduction Microstructure control is a key to the development of high-performance alloy steels, especially for applications requiring toughness, high fatigue strength and hardness in automotive components. Several studies report the phenomena of hardening, dynamic recovery, dynamic recrystallization and austenitic grain growth, and how these phenomena affect the steel mechanical properties. The control of such mechanisms by thermomechanical processing is most commonly implemented in rolling processes1-6. However, they are often more challenging to be implemented for forging processes. As an economically feasible alternative that has excellent potential for forging solutions, Finite Element Method (FEM) using computer simulation seeks to reduce the try-outs in an industrial scenario. Moreover, the classic JMAK model (Johnson-MehlAvrami-Kolmogorov) is mathematically stable and can be applied to predict recrystallization and grain growth phenomena7. The JMAK model is presently available in several numerical simulation software. For more than a decade it has been shown that the semi-empirical analytical model JMAK can describe the global recrystallization kinetics (Eq. 1), where X (t) represents the recrystallized grain fraction as a function of time (t). X (t) = 1 - e -b.t *e-mail: n (1) The exponent (b) represents the Avrami coefficient, therefore very sensitive to variation in temperature. The Avrami exponent (n) is related to the mechanism of phase transformation, for example, if the nucleation rate remains constant or even increases during the transformation progress or if the nucleation rate reaches zero soon after the onset of growth. Many articles report the strong influence of parameters such as temperature, deformation and strain rate on dynamic (XDRX), metadynamic (MDRX) and static (SRX) recrystallization8-10. Innovative studies have considered the quantitative dependence of XDRX activation energy and strain rate exponent on the temperature variation in high carbon steels. They also were successful in minimizing errors between the experimental values and correspondent finite element solutions using optimization tools11-13. However, there are some implications concerning the validation of a robust model for hot forging processes that should be considered, such as thermomechanical history, complex strain fields, fibering zones as well as steady-state conditions in flow curves. These influences are not isolated in forged parts and are driven by industrial demands. Therefore, more efforts must be given to reach more accurate results by FEM simulation models in industrial forging applications. This preliminary study aimed at investigating the applicability of a semi-empirical model of JMAK coupled to the visco-elastoplastic model in numerical simulation. 2 Ivaniski et al. A ferritic-perlitic DIN 20MnCr5 steel microstructural was used to carry out this work. Subsequent steps of the process were performed to represent an industrial process. The main influences, such as temperature, strain and strain rate fields following the recrystallization model were investigated using a subroutine of the FORGE NxT 2.1 software. 2. Materials and Methods 2.1 Acquiring the experimental data Cylindrical samples with a diameter of 25.4 mm and a height of 35 mm were manufactured from a DIN 20MnCr5 steel. The chemical composition of the steel is shown in Table 1. Hot forging (upsetting) experiments were carried out in a hydraulic press with a capacity of 400 kN, as shown schematically in Figure 1. Samples were heated in the furnace to 1200 ºC and then moved to the press where a 60% reduction in height was applied. The temperature evolution in the workpiece was measured by a thermal imager Fluke® Ti400. The temperature frames obtained by the thermal imager were corrected by thermocouple analysis for higher accuracy in the temperature determination. After that, the collected results were used as boundary conditions into the Forge® software. The austenitic grain size was analysed by Optical Microscopy following the ASTM E112 standard. The forged samples were quenched in water directly after forging to stop changes in grain size, in order to preserve the austenitic grain size from deformation end. Samples were etched to reveal the prior austenite grain boundaries (PAGBs) with an etching prepared with 3g of picric acid in a solution of 30% liquid detergent in water. Table 1. Chemical composition of the experimental DIN 20MnCr5. % wt C Si Mn Cr 0.19 0.2 1.25 1.15 Materials Research 2.2 Boundary conditions for Simulation Finite Element Method was carried out using tetrahedral (deformable) mesh with volumetric elements for the billet and triangular (rigid) surface elements for the dies, as illustrated in Figure 2. The average mesh size was enough refinement for high convergence in the calculations. Table 2 shows the boundary conditions used for modelling, including friction (...truncated)