How can the cooling channels of additive manufacturing and casting tools be optimized for free surfaces in die casting?
In casting tools and additive manufacturing, an algorithm was developed using an accompanying method to optimize the position and cross-section of the 3D printing tools used in the aluminum die-casting process, and the internal cooling channels of the steel inserts.
The algorithm enables the development of optimized complex industrial molds at relatively low computational cost, and the transient model has been experimentally verified to provide a suitable interfacial heat transfer coefficient.
Based on the casting cycle and the thermal behavior of the mold surface, a steady-state thermal model was developed to evaluate the spatial distribution of temperature and serve as an initial solution for subsequent optimization phases.
The adjoint model is then applied to optimize the cooling channel, emphasizing the minimization of the standard deviation of the mold surface temperature. The original transient model is applied to the optimized mold configuration by calibrating using experimental data obtained from a dedicated aluminum furnace.
The optimized cooling channel geometry uses an uneven cross-section over the entire pipe surface area, improving pressure drop and cooling uniformity at the mold/casting interface by 24.2% and 31.6%, respectively.
This model has been used to optimize the cooling channels of a range of industrial high pressure aluminum die casting (HPADC) inserts. This significantly increases the service life of the mold, increasing to nearly 130,000 injections compared to the 40,000 injections previously designed.
Injection molding and die casting are mature areas of manufacturing, and design opportunities brought about by development are gaining renewed attention thanks to additive manufacturing (AM) methods.
These novel additive manufacturing techniques are not limited by the shape limitations imposed by traditional computerized numerical control (CNC) machining methods, enabling the creation of complex internal voids, such as conformal cooling channel layouts.
This feature supports the design of conformal channels in mold inserts to maximize thermal performance, such as gravity and high-pressure die casting (HPDC), which has some similar characteristics to plastic injection molding, including material injection, cooling, and part ejection.
However, metal tool casting requires more extreme operating conditions than polymer injection molding, requiring further understanding of physics and cooling processes.
Nevertheless, some design and optimization concepts for plastic injection molding may have potential value in die casting applications.
Major challenges in the die casting process include uneven heat distribution, high-temperature gradients, and excessive variation in the heat flux at the boundary at the mold-to-casting interface.
These factors can lead to adverse effects such as product shrinkage, warpage, and thermal residual stress, so developing the optimal cooling channel layout for rapid prototyping mold inserts is of great interest in the industry.
The optimization strategy for additive manufacturing casting tools includes both external and internal methods, and the development of external optimization includes advanced rapid tooling (RT) technologies such as direct metal laser sintering (DMLS) (also known as SLS), stereolithography (SLA), and powder bed melting (PBF).
Other innovations include replacing molds with highly thermally conductive materials or alloys, or adding intermediate layers between cavities, all of which demonstrate improvements in thermal efficiency and mechanical properties of finished injection molds.
The addition of lubrication and spray processes also contributes to more uniform cooling of the release process, however the control of lubricant spray is a challenging task when targeting specific mold surfaces and is often considered a separate operation applied between injection cycles.
Although it contributes 20-50% to the cooling in the old HPDC system, it is not considered here as part of the casting modeling process.
While external optimization is still limited by manufacturing processes and material types, optimization of internal cooling channel design remains the most widely used and effective method in injection molding development.
Here, parametric studies are extensive, especially with the development of additive manufacturing technology, according to the specifications of the casting material and the complexity of the part.
This allows different design parameters such as the number of cooling channels, pipe diameter, pitch, distance between parallel pipes, and depth between channels/casting surfaces.
An example describes a casting mold design that minimizes cooling time by considering the number of cooling channels, location, and flow rate. Including additional structures, such as internal bezels or lattices, or secondary cooling layers, may also be beneficial.
Traditional circular cross-section channel geometry has been used in most cooling design cases to avoid overheating around corners or sharp edges.
However, more advanced studies have shown that performance can be improved by using alternative cross-sectional geometries, and an aluminum-filled epoxy injection mold study achieved an 18% reduction in cooling times using profiled cross-sections.
Another design using square and rectangular groove geometries reduced cooling time and warpage by 65% and 54%, respectively, although various non-circular geometries have been proposed.
However, cross-sectional profiles are generally kept constant, and few studies have reported the use of non-uniform cross-sectional profiles in conformal cooling channels in injection molds.
The idea of using non-uniform cross-sectional profiles can be linked to the study of the Response Surface Method (RSM), a powerful numerical method for calculating predefined mesh sensitivities based on selected objective functions.
By the mid-80s of the 20th century, more complex Euler and Navier-Stokes equations were used for aerodynamic flow simulations.
The adjoint approach provides independence between computational cost and the number of design variables, ultimately making it a useful tool for solving problems in large design spaces.
