A Development of the Rosenthal Equation for Predicting Thermal Profiles During Additive Manufacturing
William H. Keeley, Richard Turner, Bashir Mitchell, Nils Warnken
Abstract
Thermal modelling of additive manufacturing is a key method for furthering the quality of the components produced, as it allows for analysis that is not possible via experimental methods due to the difficulties involved with in situ monitoring. The thermal gradients present during the additive manufacturing process have a large impact on the formation of defects, such as porosity, residual stress, and cracking. The thermal gradients also have a large impact on material properties by controlling the microstructure formed. Thermal modelling methods are often based on numerical solutions of the heat conduction equation. Whilst numerical methods can be more accurate, they are often very slow because of the fine mesh requirements to capture high thermal gradients and iterative solvers to approximate the real-world solution to the required thermal field equations. An analytical model was developed to provide a fast solution to the problem. The analytical model used in this research was based on the Rosenthal equation and was analysed under a range of process parameters. A temperature-dependent Rosenthal model was also created with the aim of improving the results. The analytical model was then compared with a finite element numerical model to act as verification for the results. The analytical model accurately predicted the meltpool width over a range of process conditions. The analytical model underestimated the meltpool length compared to the numerical model, especially at high velocities. When using the standard Rosenthal model, the use of room-temperature or high-temperature thermal conductivities underestimated or overestimated the cooling rates from the meltpool, respectively. A temperature-dependent Rosenthal model was shown to produce more accurate cooling rates compared to the original Rosenthal equation.