Precision casting, particularly investment casting, is a specialized forming process widely employed in industries such as aerospace and industrial gas turbines for manufacturing complex and high-accuracy components. This method involves creating a sacrificial pattern, typically from wax, which is coated with ceramic slurry to form a mold. After the pattern is melted out, molten metal is poured into the preheated mold, and upon solidification, the mold is broken away to reveal the final casting. The integration of additive manufacturing with investment casting has revolutionized this process by utilizing 3D-printed patterns instead of traditional wax, significantly reducing lead times and production costs while enhancing design flexibility for intricate parts like impellers.
In this study, we explore the application of fused deposition modeling (FDM) 3D printing to fabricate patterns for investment casting of an impeller, combined with numerical simulation using AnyCasting software to optimize the gating system design. The impeller, with a base diameter of 104 mm, top diameter of 25 mm, and side width of 49 mm, features curved blades with a wall thickness of 2.5 mm. Its complex geometry necessitates careful analysis of the filling and solidification processes to prevent defects such as shrinkage porosity and turbulence-induced imperfections. We designed three gating systems: top-gating, side-gating, and bottom-gating, and simulated their behavior to identify the most effective approach for precision casting.
The numerical simulations were conducted using AnyCasting, with process parameters summarized in Table 1. The mold material was high-temperature resistant gypsum, known for its excellent fluidity and low thermal conductivity, while the casting material was ZL104 aluminum alloy, poured at 750°C into a mold preheated to 650°C. Mesh generation for the simulations ensured accurate resolution of the impeller’s geometry, as illustrated in the grid model for the top-gating system.
| Mold Type | Casting Material | Pouring Temperature (°C) | Mold Preheat Temperature (°C) | Pouring Speed (cm/s) |
|---|---|---|---|---|
| Metal Mold | ZL104 | 750 | 650 | 25 |
For the top-gating system, simulation results indicated turbulent flow during filling, as molten metal entered from the top and accelerated downward due to gravity. This led to instability and potential gas entrapment, resulting in shrinkage defects concentrated at the bottom of the impeller. The combined defect parameter, which integrates temperature gradient and inverse interface velocity, is defined as:
$$ D = \nabla T \cdot \frac{1}{v} $$
where \( D \) represents the defect indicator, \( \nabla T \) is the temperature gradient, and \( v \) is the interface velocity. High values of \( D \) correlated with areas prone to shrinkage porosity, as shown in the simulation outputs.
In the side-gating system, filling was more stable, with solidification progressing upward in a directional manner. However, rapid filling rates caused air entrapment, leading to defects at the junctions between the runners and the impeller blades. The shrinkage porosity fraction highlighted these regions, emphasizing the need for controlled filling speeds in precision casting applications.
The bottom-gating system demonstrated the most favorable outcomes, with sequential solidification minimizing defects. The combined defect map showed no significant issues on the impeller surface, and shrinkage was primarily confined to runner corners. This aligns with the principles of investment casting, where bottom-gating promotes steady metal flow and reduces turbulence. The solidification time \( t_s \) can be estimated using Chvorinov’s rule:
$$ t_s = B \left( \frac{V}{A} \right)^2 $$
where \( B \) is the mold constant, \( V \) is the volume, and \( A \) is the surface area. For the impeller, this ensured uniform cooling and reduced thermal stresses.
Based on the simulation results, the bottom-gating system was selected for experimental validation. A pattern was fabricated using FDM 3D printing with polylactic acid (PLA) material, and the investment casting process was carried out. A gypsum slurry was prepared in a 100:45 ratio with water, stirred thoroughly to eliminate bubbles, and applied to the pattern to form the mold. After setting, the assembly was placed in a steel cup and heated in a resistance furnace to decompose the PLA pattern completely, following a controlled heating cycle: 400°C for 15 min, 600°C for 30 min, and 750°C for 2 hours.
Molten ZL104 aluminum alloy was then poured into the preheated mold, ensuring complete filling. Upon cooling, the gypsum mold was broken away, and the casting was extracted, followed by removal of the gating system and surface finishing. The final impeller casting exhibited no significant defects such as cold shuts or porosity, confirming the effectiveness of the optimized investment casting process combined with additive manufacturing.
This study underscores the synergy between additive manufacturing and investment casting for producing high-quality impellers. The use of numerical simulation in precision casting allows for accurate prediction of defects, enabling proactive optimization. Future work could explore other alloy systems or complex geometries to further advance investment casting techniques. In summary, the integration of 3D printing with traditional casting methods offers a robust solution for rapid prototyping and production, enhancing the efficiency and reliability of precision casting in industrial applications.