4. Modeling approach

hatch spaces has been previously calculated, and keeping a unitary hatch distance will result in a

Figure 9. The geometrical transformation applied to cross-section for two different path rotations.

more reliable intersection procedure.

Figure 10. Results from the path simulator.

The result of the slicing procedure is illustrated in Figure 10.

136 Finite Element Method - Simulation, Numerical Analysis and Solution Techniques

In SLM, the energy needed to melt the powder bed is provided by a laser source. The portion of material under the laser is heated because of the interaction between electromagnetic waves and powder grains. This type of heat transfer occurs in a very short time interval (microseconds) and provokes material modifications due to both phase (liquid and solid) and aggregation (powder and bulk) state changes. When the laser heats the surface, powder grains undergo very rapid heating that melts the material in the localized region surrounding the irradiated spot. After that, the laser is moved forth and the molten pool starts cooling and solidifying. At the end, the material has changed its aggregation state from powder to bulk. Since the path meanders on the surface, the material undergoes multiple reheating processes, sometimes above the melting point.

All previous characteristics would lead to a very complex and cumbersome simulation unless some simplifications are applied to the numerical model. The simulation is formulated taking into consideration all the SLM features, even if they are applied in a simpler way. For example, with the aim of reproducing both phase (solid-liquid-vapor) and aggregation state (bulkpowder) transformations, only material properties are defined, rather than complex thermodynamic models. Applications involving phase change can be approached using ANSYS® through elements with enthalpy property capabilities.

The thermal transient analysis is necessary to take into account the high heating and cooling rate. Moreover, since the laser works in pulsed mode, the analysis is fully solved for each application point. The iterative algorithm forces the analysis to be solved, deleted, and restarted at each step. Consequently, nodal results must be continuously saved and uploaded through a mapping procedure as will be extensively explained later.
