Therefore, in the present investigation, we aimed to accurately estimate the heat transfer coefficient at the die– workpiece interface as a first step toward improving die design. We used an inverse approach that estimates heat transfer coefficients based on temperature measurements at select locations.

The second step involved evaluating the fatigue life of a hot press forming die based on the obtained heat transfer coefficient. The detailed temperature distribution of the die was evaluated using heat transfer coupled forming analysis. The stress distribution of the die was computed using FEM analysis by assuming the die to be an elastic body. Based on the information on temperature and stress, the fatigue life of the die was assessed in terms of dynamic failure under repeated loads. Furthermore, the fatigue life of the die for hot press forming conditions was compared with that for cold forming conditions. The work flow of the present investigation is summarized in Figure 1.

2. EVALUATION OF INTERFACE HEAT TRANSFER COEFFICIENT

2.1. Temperature Measurement of Die during Hot Press Forming Process

A hat-type part was designed based on the cross-sectional shape of the side sill of an automobile, as shown in Figure 2. The side sill is a safety part of an automotive body that protects passengers during side collisions. High-strength steels have been used recently for side sill parts to improve their safety performance. As an alternative, hot press forming is an excellent technique to improve the strength of parts through a die quenching effect on the part material.

The die for hot press forming process was designed based on the geometry of the hat-type part. A number of water cooling channels were installed in the forming die to suppress the rise in the temperature of the die and produce the die quenching effect on the heated workpiece during

Figure 3. (a) Hot press forming die with cooling channels and thermocouple locations for temperature measurement,

(b) measured temperature history at 11 thermocouple locations.

the hot press forming process. The size and locations of the cooling channels were designed to provide sufficient cooling effect on the die. The initial design of the hot press forming die with cooling channels is shown in Figure 3 (a).

The overall hot press forming process is composed of the following steps: First, the sheet metal workpiece was heated to the austenitizing temperature of 950oC at a heating rate of ~4.32oC/s and was then held for 1 min to remove thermal gradients. Subsequently, the workpiece was transferred from the furnace to the press, as quickly as possible, in order to take advantage of the enhanced formability at high temperatures. During transport, the workpiece was cooled down to ~650oC. During forming, the workpiece was deformed at a forming speed of 54 mm/s and was then held for 5 s. Finally, the formed part was taken out of the die and air cooled at room temperature. The step-by-step process images are shown in Figure 4.

In order to evaluate the interface heat transfer coefficient by inverse analysis, the temperature histories for some select locations of the die were measured. To measure the temperature of the die, K-type thermocouples were

LIFE ESTIMATION OF HOT PRESS FORMING DIE BY USING INTERFACE HEAT TRANSFER COEFFICIENT 287

attached to the tab holes, which are marked as the red dots with the numbers 1~11 in Figure 3 (a). The tab holes were machined at the central locations between the channels on the die surface. The measured temperature history is shown in Figure 3 (b).

2.2. Evaluation of Interface Heat Transfer Coefficient by Inverse Approach

The goal of the inverse heat transfer analysis by the FEM is to find the boundary condition, i.e., the heat transfer coefficient, that can provide temperature predictions best fitted to temperature measurements with respect to the select sensor locations (Kim and Oh, 2001). Because an inverse analysis is similar to solving an optimization problem, we can obtain the solution of this inverse analysis by applying an appropriate optimization method using temperature sensitivity to boundary conditions or by conducting sufficient numbers of iterative FEM simulations. In the present investigation, we performed iterative FEM simulations for simplicity.