||Nowadays the massive forming leads to have smaller production-series and less productions steps, in order to save huge resources, which are spent for following machining processes. This tendency moves the production to Net-shape/Near Net-shape forming, which requires high load conditions. It is therefore necessary to measure the deflections of the press under high loads and to estimate the role of these deflections on the quality of final product. From this critical problem was born the idea of EU-IMPRESS project (Improvement of Precision in Forming by Simultaneous Modeling of Deflections in Workpiece-Die-Press System), whose aim is the development of a Finite Element Code able to predict press deflections.
This master thesis is done as a part of the EU-IMPRESS project, where it represents a part of Work-Package 8. It documents the measuring of press deflections of 1000 tons-hydraulic press placed by Buurskov og Lassen Maskinfabrik in Holstebro, Denmark.
The aim of the thesis is the comparison between experimental measured press-deflections and predicted deflections, coming from the Finite Element Code developed (FORGE3®).
As shape to test for the experiments a hammer-head has been chosen, because it can generate oblique reaction force when forged. In this way the press is stressed also with horizontal load, causing movements of the press-ram. The optimization of this shape was based on criteria, which previewed to maximize the ratio between horizontal and vertical component of the reaction force. A set of simulations have been run in two and three dimension, in order to plot different load-stroke curves for different geometrical proportions of the billet. Real tests have been carried out on the hydraulic press mentioned above and experimental data have been compared with the simulation done for a specific position of the die and the workpiece on the press table, which allowed reaching great displacements values. Only one loading case is analyzed, due to problems at present time for partners in modeling the material characteristic. Results showed a big discordance between measurements and numerically predicted deflections. Thus the first attempt in validating the software did not give a positive evaluation. Other simulations are going to be run at present time and results relative to these new studies will be presented in further works.
Closed to the deadline, a new FORGE3® simulation is still running and part of these new results shown a concordance of the loads with the ones of experimental data. A full description of this new simulation is expected to be presented at the oral defense of this project.