Additive Manufacturing comes with a broad range of challenges. Some of these challenges are related to the thermal effects during the manufacturing process. Thermal challenges include managing cool down rates to guarantee homogenous material properties. Simulation with AdditiveLab can help to optimize intra-layer pauses (dwell-times) to ensure homogeneous cool down rates and the manufacturing of high-quality designs.
The importance of homogeneous cool down rates.
During the powder-bed Additive Manufacturing (AM) process, each individual layer is exposed to heating and cooling cycles over an extended period of time. While a layer is being manufactured, it is exposed to heat from the laser. Post laser exposure, the layer gets to cool down for a bit, until new powder is being deposited for the next layer. When the next layer is being exposed to heat, then the previous layer is subsequently exposed to elevated temperatures from the next layer. This heating and cooling cycle is repeated in a layer-by-layer fashion.
The figure above shows the manufacturing process of the first layer (left) that is exposed to heating by the laser, then gets to cool down, and is partially exposed to heat coming from manufacturing the second layer (right).
The dissipation of heat from the top layer through the manufactured structure becomes more challenging if the prepared build configuration limits the heat flow. For example if the cross-sectional area of the configuration increase with the manufacturing height; a typical, worst case scenario would be a cone that is manufactured upside-down
The figure above shows an illustration of a conical structure being manufactured upside-down, limiting the heat dissipation into the base-plate.
Why is this such a big deal?
For metallic materials, different crystalline structures form depending on the cooling rate. Different crystalline structures result in overall different material properties and for example define if a material is more ductile or more brittle, and allows for little or more elongation. In high-end engineering industries controlled solidification (cool down) is used to create materials that are specifically tailored to certain applications. For example, for certain metallic materials, rapid cooling rates allow an increase of hardness. In addition to that, the better control the manufacturer has over the thermal process and the cool down rates, the better they can manipulate crystalline structures to their liking and ensure homogeneous and failure free material properties in the manufactured design.
This particularly becomes important for dynamically loaded geometries such as engine valves which need to be manufactured flawlessly in order to ensure lifetime durability. Consider the following valve geometry:
The picture above shows an example of a cyclic-symmetric valve geometry with varying cross sections along the height of the valve (right) and the simulated average temperatures throughout the manufacturing process indicating thermal-flow bottlenecks that limit the heat dissipation.
A simulation of the valve with AdditiveLab (see above picture, right) revealed the problem as addressed with the simple cone structure above; thinner cross-sections underneath larger cross-sections limit the heat flow through the valve and cause inhomogeneous average temperatures as well as different cooling rates. The different cooling rates will lead locally to different material properties, essentially to a valve that will have different material properties in the thinner sections than in the thicker sections.
How to ensure homogeneous material properties during the manufacturing process?
The picture above shows an example code of a Python script to optimize the dwell-times during the manufacturing process.
This is where AdditiveLabRESEARCH software comes in handy. With the thermal simulation module and the AdditiveLab Python API, the user can define all sorts of optimization problems, including the optimization of dwell-times to ensure homogeneous cooldown rates in the manufactured design throughout the building process. In other words, one can use AdditiveLab to find what layer longer or shorter pauses are needed to ensure the continuation of the process without allowing for unwanted temperature accumulation.
In order to demonstrate this on the valve case, we have created a Python script that automatically adjusts the intra-layer pauses to ensure homogeneous cooldown rates and to avoid temperature accumulations. The main sections of this script include the preparation and execution of subsequent thermal simulations and the definition of an error function that determines the difference in cooldown rates throughout the entire valve design.
The figure above shows the simulated average temperatures throughout the manufacturing process indicating thermal-flow bottlenecks that limit the heat dissipation with the original process (left) and an improved situation with a more homogeneous distribution of average temperatures in the optimized process (right).
Once the optimization was finished the comparison of the average temperatures calculated over the entire building process revealed a more homogeneous distribution of average temperatures in the optimized process (right) compared to the default process with constant dwell-times between each layer (left).
The figure above shows the dwell-times for the original manufacturing strategy (left) and the dwelltime optimized strategy (left) with longer dwell-times indicated in red.
With this optimization strategy via AdditiveLab, manufacturers can improve the outcome of their manufacturing process and ensure the generation of suitable high-end application parts that need a high level of material quality.
These are a few examples of how we at AdditiveLab use simulation optimizations to improve the manufacturing process. If you are interested in having your design evaluated by us, or having training to help you increase your knowledge in a simulation-based optimization of build configurations, please get in touch with us.
Comments