A unified model for isothermal and non-isothermal phase transformation in hot stamping of 22MnB5 steel

Abstract

The hot stamping of tailored properties is widely established in the automotive industry. However, unlike conventional press hardening, this technique is designed to generate varying mechanical properties by controlling the cooling rate locally. Current material models cannot accurately represent the locally very different thermomechanical paths. This research aims to address this problem by developing a new phase transformation model capable of handling complicated thermal histories and the effects of deformation. A ’unified JMAK‘ model is proposed to capture various thermal-mechanical loading conditions. These conditions include the deformation effect and cooling paths where isothermal conditions range from 450° to 725°C and non-isothermal conditions range from 5° to 40°C/s. Furthermore, for the extension to linear and non-linear cooling paths, correction factors for the phase transformations are introduced into the model. Finally, the model's performance is compared to well-known transformation models, such as the non-isothermal JMAK and the K-V (Kirkaldy and Venugopalan), by comparing the results obtained from isothermal and non-isothermal dilatometer tests and a scaled-down b-pillar experiment. The results from the comparison are classified into 3 cases, isothermal, continuous cooling and non-linear cooling (b-pillar). Case 1 shows that the non-isothermal and K-V models provide low accuracy, while the proposed model yields an error below 2 %. For Case 2, the proposed models and non-iso JMAK are agreeable with the experiment, and the K-V model shows a maximum error of 15 %. Finally, in the b-pillar case study mean absolute errors of hardness predictions are 33.06 HV (mean relative error is 9 %), 43.92 HV (13 %), and 67.70 HV (20 %) for the unified JMAK, non-iso JMAK, and K-V model, respectively.