Using Monte Carlo Simulation for Prediction of Tool Life

Abstract

Hot metal forming (rolling, forging, extrusion, wire drawing) constitutes a very large proportion of manufacturing activity. Of all the equipment and tooling involved in a hot forming process, the most critical component is usually considered to be the die due to its superior precision and reliability requirement and the associated high cost. Dies and ancillary tooling are exposed to high pressures, elevated temperatures, and both mechanical and thermal fatigue. Cost and engineering difficulty are obviously high because of factors such as special material and processing, very fine tolerances, and high demands on repeated thermo-mechanical performance. How often a die has to be scrapped and replaced with a new one directly contributes to the commercial viability of producing a certain profile, not only because of the large cost of die replacement, but also because of losses due to interrupted production and reduced product quality in the case of a defective die. Due to this critical importance and high cost of metal forming tools, one of the major goals is a longer tool life. Continued research in tool and process design is therefore targeted at minimization of tool failure. However, tool failure is a complex phenomenon, governed by an interaction of various mechanisms, and not easy to control or restrain. A more manageable approach is to make die failure predictable. Estimation and prediction of tool life thus become crucially important. Extrusion is one of the most popular metal forming processes. Because of its wide-ranging and abundant application in the automobile, aircraft, and construction industries, aluminum has been called the metal of the millennium. As extrusion is the primary manufacturing process for aluminum alloys, the popularity and importance of extrusion has increased even more. The three leading failure mechanisms for extrusion dies are fracture, wear, and plastic deformation (Bauser et al., 2006). Fatigue fracture and gradual wear are the more dominant of these failure modes.