Additive Manufacturing Simulation Types

Three simulation types are available on the AM Calculator – Steady-state, Transient, and Transient with heat source from Steady-state.

Additive Manufacturing Module Theory and About the Heat Source Models.

The keyhole model is available when Gaussian, Core-ring, or Top-hat heat sources are used for either the Steady-state or Transient with heat source from Steady-state types of simulation (see About the Keyhole Model. It is available with or without fluid flow.

Steady-state

For Steady-state simulations, you can also choose to run different Steady-state Calculation Types.

In the Steady-state mode it is assumed that the temperature distribution and the fluid flow around the heat source is in steady state and does not change with time. This is useful to get an estimation of the temperature distribution and size of the melt pool when you assume that the heat source is moving at a constant speed in a given path. In the single-track experiments, temperature distribution around the heat source and fluid flow inside the melt pool reach a steady state very quickly, and you should then perform steady-state simulations to predict melt pool geometry and cooling rates around the melt pool. In a sense, steady-state simulations give you an overall picture of the process but in order to get more precise details and predict temperature distribution in a multi-layer build, as a function of time, you should perform transient simulations. The benefit of the steady-state option is that these simulations are quick, and you get a solution typically within 1-5 minutes, depending on the process parameters and your computational resources. For the steady-state simulations, you can include or exclude fluid flow inside the melt pool due to the Marangoni effect. For the given processing conditions, if convection is the dominant mode of heat transfer, inclusion of fluid flow is crucial to enhance the accuracy of the model by correctly capturing the underlying physical behavior of melting and solidification of material. Furthermore, you can also perform simulations with a powder layer on the top of the substrate having different material properties than the bulk material. The steady-state simulations are performed on a symmetric domain where you specify only the height of the substrate and the thickness of the powder layer, if present. The length and width of the computational domain are determined automatically based on the process parameters. The temperature distribution is computed using the energy equation while the fluid flow is modeled using the Navier-Stokes equation.

Transient

In the Transient mode, you can perform full-scale transient simulations in a 3D rectangular build part and have the possibility to specify a scanning strategy comprising multiple tracks and multiple layers. Here you can enter the height, width, and length of the entire build part or a representative segment of the build part and configure a scanning strategy either for a single track or for multiple tracks (bidirectional or unidirectional). You can also add multiple layers of powder and rotate the scanning pattern between layers.

Similar to the Steady-state mode, here you can also choose to include fluid flow inside the melt pool to correctly capture the underlying physical phenomena of melting and solidification of the material and thereby increasing the accuracy of the model. The inclusion of fluid flow requires coupling the Navier-Stokes equations together with the energy equation which comes at the cost of increased numerical complexity resulting in longer simulation times.

Transient with Heat Source from Steady-state

In order to perform full scale 3D simulations in an efficient manner, with multitracks and multilayers, including fluid flow in the melt pool or with powder layer(s) having different properties than the solid material, you can use the Transient with heat source from Steady-state mode. This mode develops a novel approach where the effect of fluid flow due to Marangoni convection in the melt pool is incorporated without solving for Navier-Stokes equations in the full-scale 3D domain.

The concept involves the assumption that the temperature distribution and fluid flow inside the melt pool instantly reach steady state. First you solve for temperature distribution, using the energy equation and fluid flow due to Marangoni convection, using the Navier-Stokes equations for the given process parameters in the Steady-state mode. Once you have the steady state solution, use this solution inside the melt pool and map it as a “heat source” in the transient simulations. Using this approach, you do not have to solve for the complicated Navier-Stokes equations at each time step in the transient simulation, yet it still captures the effect of fluid flow on the shape of the melt pool. Furthermore, it also reduces the computational time by solving for the most non-linear region in the energy equation i.e., inside the melt pool, only once, and then reusing the solution as a boundary condition in the transient simulations. This approach considerably decreases the numerical complexity of the full-scale transient simulations, which consequently reduces the simulation time without making a significant compromise on the accuracy of the solution.