Search

The combination of additive manufacturing principles and electron beam (EB) technology allows complex metal parts, featuring excellent quality material, to be produced, whenever traditional methods are expensive or difficult to apply. Today, the optimization of process parameters, for a given metal powder, is generally attained through an empirical trial and error approach. Process simulation can be used as a tool for decision-making and process optimization, since a virtual analysis can help to facilitate the possibility of exploring “what if” scenarios. In this work, a new type of modelling has been introduced for energy source and powder material properties and it has been included in a thermal numerical model in order to improve the effectiveness and reliability of Electon Beam Melting (EBM) FE simulation. Several specific subroutines have been developed to automatically calculate the powder properties as temperature functions, and to consider the position of the beam during scanning as well as the material state changes from powder to liquid in the melting phase and from liquid to solid during cooling. A comparison of the numerical results and experimental data taken from literature has shown a good forecasting capability. The average deviations of the simulation from an experimental scan line width have been found to be below about 15%.

Electron Beam Melting (EBM) has the potentiality of being an effective system in terms of time and energy consumption. Among the different additive manufacturing processes that are available, the EBM process has shown the lowest Specific Energy Consumption (SEC) and the highest average Deposition Rate (DRa). However, there appears to be a lack of literature on the correlation between SEC and DRa. Moreover, all the literature studies have only an analysis of energy efficiency during the melting of the bulk material phase and have adopted a fixed job design. Therefore, the aim of this study has been to fill this gap. The EBM process is decomposed into small substeps and a bottom-up approach is adopted to provide models that can be used to evaluate the energy of each process subphase and to perform an energy characterisation at the unit process level. A black-box approach is applied to provide a new model for the energy efficiency of the EBM process. Different jobs have been designed to analyse the effect of a part and of manufacturing designs. Bulk material, support and lattice structures have been included. The design has therefore been aimed at investigating the effect of the building height, melted area and process themes on energy efficiency. The jobs have been produced using Arcam A2X and Standard Arcam Ti6Al4V powders. According to this research, the architecture of the machine and its control of the process have the main impact on the relationship between SEC and DRa. The design features of the part and of the job influence the position of the job on this curve and thus the relative energy and time efficiency. Additionally, the empirical approach applied to the machine subunits has highlighted that only a small part of the total energy demand is needed to power the electron beam during the melting phase, while the remaining part guarantees the good machine working conditions.

In this work, an improved but still rather simple computational analysis is presented for a more detailed prediction of Electron Beam Melting (EBM) process outcomes. A fully coupled thermomechanical analysis is developed in which nonlinearities due to the variation of material properties when the material melts are included. A new analytical approach is developed to emulate the volume variation of the powder bed during heating and melting. Particularly, the expansion of the powder particles and the porosity reduction within the powder bed are considered simultaneously. The thermal expansion and the shrinkage of solid material during heating and cooling and the stress formation within the solid material are also modelled. The model can predict the geometrical transformation of the powder into solid material in an efficient way. A comparison between experimental and simulated cross-sectional areas of melted single lines is presented. Both continues line melting and fractional line melting, multi beam melting, are considered. The model shows a good ability to provide consistent and accurate forecasts. The maximum deviations between experimental and numerical results are approximately 15% for the height and 5% for the width of the melted lines, respectively. A comparison with a pure thermal model is also included, and benefits and differences between the two models are discussed.