Return to search

3-D simulation of the fullering process in hot forging

Elongated parts make up a considerable percentage of forged components used in various industries. In many cases, their production by means of a multi-impression forging operation involves fullering and rolling processes. These are two types of open-die forging process, the main function of which is to properly distribute the metal in the longitudinal direction of the component. Several sets of empirical rules for fuller and roller die design have been proposed previously and have also been incorporated into CAD/CAM systems. Since in these processes the metal can flow freely in certain directions, there is no guarantee that the desired preform shape will be produced and some means of predicting this shape would be beneficial. During the initial stages of this project, various methods of metalworking analysis were reviewed. The model material technique and the finite-element (FE) method were selected for the analysis of the deformation during the fullering process. A gravity drop model hammer was designed and constructed for the physical simulations of the fullering process using plasticine as the model material. Ring tests were performed to find a suitable lubricant and compression tests were used to obtain the flow stress of the plasticine as a function of strain and strain rate. For the physical simulation of different types of the fullering process, a typical elongated forging component was used. These tests were performed using die shapes described as flat-flat and crowned-flat. The main features of deformation studied by the model tests were the distribution of deformation, variations of elongation and maximum sideways spread and the elongation achieved during each blow in each of the processes. The results were discussed, compared to each other and employed for the validation of the FE results. The fullering processes that were modelled physically were also simulated numerically using an elastic-plastic thermo-mechanical code (EPFEP3). Different FE models were developed to investigate the effects of mesh density on material flow. Separate sets of material properties, namely those of hot steel and plasticine, were employed in the simulations. For each set, separate FE analyses of ring compression tests were conducted to ensure that the appropriate friction condition was provided for the simulations of the fullering processes. The material flow during the fullering process, and the effects of various parameters influencing this, were investigated. To validate the results obtained from the FE simulations, they were correlated with the available experimental data as well as with the results obtained from physical modelling of the process. In most cases there was very good agreement. Also, to evaluate the empirical design rules for the forging component under consideration, the total elongation and the required minimum fuller width gained from different physical and FE simulations of various fullering processes were compared to the mass distribution requirement and to the suggestions made by some investigators. There was good agreement between various estimated fuller widths. However, it was found that to improve the amounts of total elongation, the geometries of the designed fuller dies should be modified. To avoid a trial and error method of die modification which has economical disadvantages, it was decided to employ the FE results gained from the first simulation of the process together with a numerical predictive approach. The effects of two important parameters which influence the total elongation, namely fuller gap and fuller length, were studied in the fullering process of a square bar with flat-flat dies. A method of fuller gap modification was introduced and extended to other types of fullering processes. Also, the effect of fuller length on longitudinal and transverse flow of metal was interpreted based on previous experimental observations. This investigation has also shown the feasibility of developing the current fuller CAD/ CAM system into an expert system.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:633094
Date January 1997
CreatorsFereshtehi-Saniee, Faramarz
PublisherUniversity of Birmingham
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation

Page generated in 0.002 seconds