Experimental investigation and finite-element modeling of the short-time induction quench-and-temper process of AISI 4140

https://doi.org/10.1016/j.jmatprotec.2019.116485Get rights and content

Abstract

Induction hardening is a widely used surface treatment technique that has extensively been investigated. In contrast, induction tempering with short heating times 1 s has not been investigated thoroughly, nor by experiment neither by simulations. Also, the influence of rapid tempering on the residual stresses after induction surface hardening has only been investigated superficially.

Induction quench and temper experiments were performed, both with heating times 1 s. A significant change of the residual stress state was observed, leading to a shift from compressive to tensile residual stresses in the surface layer. A multiphysical FE-model for the whole quench-and-temper process has been developed and validated. A good agreement with the experiments for the relevant mechanical properties hardness and residual stress could be achieved. The recently reported transformation induced plasticity during tempering has been found to play a key role in the development of residual stresses during tempering. The simulations further indicate that conventional heat treatment leads to more favorable residual stress states after tempering to a prescribed surface hardness. By accounting for tempering processes during austenitization, the hardening simulation could be generalized to different initial states and allows for the prediction of hardness minima in the transition zone.

Introduction

Induction hardening is a frequently used method to enhance the mechanical properties of parts such as gears and crankshafts. One goal of induction hardening is the introduction of compressive residual stresses that are beneficial to the fatigue limit of the material as stated by Withers (2007) in his comprehensive paper on the effect of residual stresses on fatigue.

Tempering is a common heat treatment after surface hardening. Tempering increases ductility and toughness of the very brittle martensitic surface, as it is e.g. stated in the book by Krauss (1989). The underlying metallurgical processes are mainly carbon redistribution and carbide precipitation as well as dislocation annihilation. An overview over the typical temperature ranges in which these processes occur is given by Speich and Leslie (1972). Usually, tempering is done as a conventional heating process in an oven with long tempering times and moderate heating rates. Advantages of rapid tempering using induction are short process times and a good reproducibility due to the induction heat treatment. Even more, studies of the mechanical properties of tempered samples report beneficial effects of rapid tempering. Furuhara et al. (2004) performed tempering experiments with heating rates up to 1000 °C/s and stated an increase of the strength/ductility ratio with increasing heating rate for a given hardness after tempering. Recent investigations by Judge et al. (2018) revealed that the toughness of rapidly tempered steel samples is significantly higher as compared to conventionally tempered samples of the same hardness.

Due to volume changes during carbide precipitation and a decrease of the yield strength, tempering affects the residual stresses that are present after any hardening process. In spite of its importance for practical applications, the change of residual stresses in a surface hardened part during tempering for both rapid and conventional heating methods has not been investigated in much detail yet. Since the measurement of residual stresses is tedious, numerical simulations of the quench-and-temper (QT) process can help to gain understanding of the evolution of residual stresses.

Early simulation models of the conventional QT process were presented by Inoue et al. (1981) who included the volume changes due to phase transformations which were modeled by phenomenological Johnson–Mehl–Avrami–Kolmogrov-type (JMAK) equations to compute the residual stress evolution during tempering. Aubry (1998) developed a comparable tempering simulation model, also based on phase transformations modeled via JMAK equations, which included the effect of auto-tempering for the first time. They found this effect to be present in steels with high martensite start temperatures but it has shown little effect on the resulting residual stresses after tempering. Later, Wang (2006) studied the conventional tempering of carburized steels. The phase transformations during tempering were modeled using physically based nuclation and growth models. All these works on the QT process allowed the calculation of residual stresses after tempering, but only for conventional and thus relatively slow oven heat treatments. Zabett and Mohamadi Azghandi (2012) modeled the induction tempering of carbon steel including an electromagnetic heating model. This work was clearly focused on the heating simulation via induction, the only mechanical quantity that was calculated was the hardness after tempering. The recent study by Tong et al. (2018) reported results on the simulation of induction hardened and subsequently oven tempered specimen which allowed the calculation of phase transformations and residual stresses. The tempering times applied in this work were 2 h, thus it cannot be considered as short-time heat treatment.

As can be seen from the literature survey, no results on experiments and especially simulations of a short-time QT process are reported to this day. The present paper aims at filling this gap. It presents experimental results such as hardness and residual stresses of rapidly hardened and rapidly tempered specimen. A simulation of the complete QT process is presented and validated using the aforementioned experimental results. Special emphasis is put on the stability of residual stresses during the tempering process. For the first time the influence of the transformation induced plasticity during tempering (T-TRIP) due to the precipitation of carbides is addressed.

Section snippets

Samples

The mechanical properties of the material in the tempered state were determined using tensile tests with the sample geometry as shown in Fig. 1.

For the induction heating experiments, ring-shaped samples as shown in Fig. 2 were used.

All samples were manufactured from 90 mm bars of AISI 4140 (German grade 42CrMo4) steel with the chemical composition as given in Table 1. The chemical composition was measured using spark emission spectroscopy.

After machining, all samples were austenitized at 880 °C

Measurement and modeling of mechanical properties

For the simulation of the tempering process, the mechanical properties are required including elastic modulus E, yield strength σy and the hardening behavior. Generally, the mechanical properties during tempering depend on the current temperature T as well as on the tempering state. Therefore, a model is sought that accounts for the temperature as well as the tempering state. To produce the necessary data to develop such a model, two series of tensile tests were performed. In the first series,

FE-model

The simulation model for the induction QT process is multiphysical including electromagnetic and thermal–metallurgical–mechanical submodels. The model is based on the commercial FE solvers ABAQUS/Standard and ABAQUS/Electromagnetic. The coupling between the electromagnetic and the thermal-metallurgical-mechanical simulation is done via Python scripts.

The main features of the models will be presented in the following.

Simulation and experiments of the induction QT process

In this section, the results from the induction QT experiments are discussed and compared to the results from the simulation model.

Hardening

Fig. 6 depicts the comparison of the simulated and measured temperature evolutions during hardening. The 3 phases of the process heating (t<0.5 s), moving shower (0.5s<t<0.9 s) and quenching (t>0.9 s) are clearly distinguishable.

Surface as well as subsurface temperatures are in a good overall agreement during heating and quenching. The immediate rise of the measured

Influence of T-TRIP

To assess the influence of the tempering-TRIP effect caused by cabide precipitation during tempering, the simulation of tempering experiment 2 as discussed in Section 5.1 was repeated without accounting for T-TRIP. The resulting tangential residual stresses are plotted in Fig. 16. For direct comparison, the results from the simulation with T-TRIP (cf. Fig. 13a) are plotted as well. A striking difference can be seen from the plot. The tensile residual stresses at the surface without T-TRIP are

Conclusion

Short-time induction tempering experiments of induction hardened parts were performed that indicated that a severe change in the residual stresses after tempering occurs. With increasing maximum tempering temperature, the residual stresses are shifted into the tensile regime.

A comprehensive multiphysical FE-model for the simulation of short-time induction QT processes was developed and validated. The model consists of a coupled electromagnetic-thermal-mechanical-metallurgical simulation and was

Acknowledgement

This research was supported by the German Research Foundation (DFG) program with the grant number SCHU 1010/49-1.

References (43)

  • D. Tong et al.

    Numerical simulation on induction heat treatment process of a shaft part: involving induction hardening and tempering

    J. Mater. Process. Technol.

    (2018)
  • A. Zabett et al.

    Simulation of induction tempering process of carbon steel using finite element method

    Mater. Des.

    (2012)
  • P. Zwigl et al.

    A non-linear model for internal stress superplasticity

    Acta Mater.

    (1997)
  • U. Ahrens

    Beanspruchungsabh&ldquo;angiges Umwandlungsverhalten und Umwandlungsplastizit&rdquo;at niedrig legierter St&ldquo;ahle mit unterschiedlich hohen Kohlenstoffgehalten

    (2003)
  • C. Aubry

    Simulation numérique par éléments finis en 3D du comportement thermomécanique au cours du traitement thermique d’aciers: application à la trempe de pièces forges ou couléss

    (1998)
  • B. Buchmayr et al.

    Modeling of the temperature field, transformation behavior, hardness and mechanical response of low alloy steels during cooling from the austenite region

    J. Heat Treat.

    (1990)
  • L. Cheng et al.

    The tempering of iron-carbon martensite; dilatometric and calorimetric analysis

    Metall. Trans. A

    (1988)
  • B. Eigenmann et al.

    R&ldquo;ontgenographische Untersuchung von Spannungszust&rdquo;anden in Werkstoffen. Teil III. Fortsetzung von Matwiss. und Werkstofftechn. Heft 3/1995, S. 148-160 und Heft 4/1995, S. 199-216

    Materialwiss. Werkstofftech.

    (1996)
  • T. Furuhara et al.

    Control of cementite precipitation in lath martensite by rapid heating and tempering

    ISIJ Int.

    (2004)
  • C. Gomes et al.

    Predicting the mechanical properties of a quenched and tempered steel thanks to a “tempering parameter”

    Rev. Metall.

    (2010)
  • G.W. Greenwood et al.

    The deformation of metals under small stresses during phase transformations

    Proc. R. Soc. A: Math. Phys. Eng. Sci.

    (1965)
  • Cited by (24)

    • A model for converting thermal analysis to volume fraction of high carbon bearing steels during low-temperature tempering

      2023, Journal of Materials Science and Technology
      Citation Excerpt :

      Their multi-component precipitation modelling accurately predicted the volume fraction and stereological characteristics of precipitates quantitatively and shed light on the evolution of transition carbides into cementite. Recently, Tong et al. [13] and Kaiser et al. [14] studied the precipitation behavior of tempered carbides in 4140 steel during rapid heating considering the nucleation, growth and coarsening of carbides. However, no attempt was made to address the volume fraction of retained austenite as the precipitation modelling was inappropriate for retained austenite decomposition.

    View all citing articles on Scopus
    View full text