Quantitative Comparison Between Precision Closed-Die Forging-Force Data and Computer Simulations

Abstract

This paper presents a study of precision closed-die, isothermal, forgings via both experiments and computer simulation. The closed-die cross-section was an “H” shape and Tin/Lead eutectic solder was used for the billet material. Extensive statistical analysis of the axial force versus displacement history was conducted using replicated forging experiments. The purpose of the experiment was to obtain statistically significant data so that accuracy tests could be conducted on different FEM computer models, e.g., ALPID, EPIC2D, NIKE2D, and DYNA2D. Overall, the forging history exhibited complex behavior consisting of five distinct regions. The experimental results yield a 5.2 percent COV in the required forging force for a specific top-die displacement. A 6.5 percent COV in the “stiffness” of the first region (elastic behavior) of the forging history was also obtained. One set of simulations with one FEM computer model, ALPID Version 2.1 for rigid-thermoviscoplastic behavior, was conducted. The occurrence of all four viscoplasticflow regions was qualitatively predicted by the simulations. Quantitatively, the simulations are within the experimental bounds for the early viscoplastic regions, but out of bounds for the later regions. It appears that, for the eutectic tin/lead billet material, there is no combination of “power-law” material parameters that yield good agreement with the later stages of the forging force history.