Experimental and numerical evaluation of heat transfer in the press hardening process

In recent decades, the use of high- and ultra high strength steels in modern car bodies has increased drastically. This is due to both severe legislation of passenger passive safety and the recent year's effort to reduce fuel consumption to decrease environmental harmful emissions. Many new materials and technologies have been developed to meet these new demands, where press hardening, or also called hot stamping, has been one of the most successful technologies to produce complex components with superior mechanical properties. In the product development process, thermo-mechanical coupled forming simulations are used to predict the final components properties. In order to obtain accurate results, correct models of the physics involved in the simultaneous forming and quenching is needed. The objective of current work is to investigate the heat transfer between the hot blank and the cold tools in the press hardening process. The transfer of heat is the key process that affects the products formability, final geometry, residual stresses and the development of mechanical properties. The two most important contact criteria are investigated in this thesis, heat transfer at a thin contact gap and heat transfer at mechanical contact under an applied contact pressure. An experimental setup to investigate these two contact criteria is developed. It consists of an upper and lower cylindrically shaped tool were the hot blank is quenched between cold tool surfaces. The results from experiments consist of measured temperature histories in the tools. A finite element model of the experiments in combination with an inverse simulation algorithm is used to predict the heat transfer at each contact condition. The inverse technique called improved advance retreat and golden section method is used to solve the ill conditioned inverse heat conduction problem that arise when solving for the heat transfer at the contact interface. The result from inverse simulations in combination with regression analysis is used to develop a general model of the heat transfer coefficient. The outcome is two regression models, one for each contact criterion, were the parameters affecting the heat transfer coefficient are identified. It is found, that a heat transfer coefficient depending on contact pressure or contact gap as well as contacting surface temperatures provide a good match between experimental and simulated temperature response in the tools. The regression model captures the main characteristics of the heat transfer coefficient and has been implemented as a subroutine in the finite element code LS-DYNA.Godkänd; 2009; 20091108 (persal); LICENTIATSEMINARIUM Ämnesområde: Hållfasthetslära/Solid Mechanics Examinator: Professor Mats Oldenburg, Luleå tekniska universitet Tid: Fredag den 11 december 2009 kl 10.00 Plats: E 243, Luleå tekniska universitet