1. Introduction
In modern industrial applications, the demand for high-performance turbomachinery components, such as impeller castings, continues to grow. Among these, the double-suction impeller stands out due to its ability to handle high flow rates and pressures in systems like pumps and compressors. However, traditional sand casting methods often struggle with the complexity of such components, particularly when dealing with materials like aluminum bronze (C95820), which presents challenges like shrinkage porosity, oxide inclusions, and gas defects.
This study focuses on optimizing the impeller casting process by integrating 3D-printed sand molds—a revolutionary approach that overcomes the limitations of conventional sand casting. By leveraging digital design tools like ProCAST and additive manufacturing, we aimed to achieve defect-free, high-precision double-suction impellers while reducing production time and costs.
2. Challenges in Traditional Sand Casting of Aluminum Bronze Impellers
Aluminum bronze alloys, such as C95820, are favored for their excellent mechanical properties and corrosion resistance. However, their narrow solidification range and high susceptibility to gas absorption make them prone to defects during impeller casting. Key challenges include:
- Complex Geometry: The double-suction impeller features asymmetric double-layer blades, thin connecting plates (as narrow as 15 mm), and uneven wall thickness (minimum 5 mm). These characteristics complicate mold preparation and core extraction in traditional sand casting.
- Defect Formation: Oxide inclusions, shrinkage porosity, and gas entrapment frequently occur due to turbulent metal flow and inadequate venting.
- Production Inefficiency: Conventional methods require extensive manual labor for mold assembly and core placement, leading to longer lead times.
3. Methodology: Integrating 3D-Printed Sand Molds
3.1 Material Composition and Properties
The chemical composition of C95820 aluminum bronze is critical for understanding its casting behavior. Table 1 summarizes the alloy’s specifications.
Table 1: Chemical Composition of C95820 Aluminum Bronze
| Element | Cu | Al | Ni | Fe | Mn |
|---|---|---|---|---|---|
| Wt.% | ≥77.5 | 9.0–10.0 | 4.5–5.8 | 4.0–5.0 | ≤1.5 |
3.2 Gating System Design
To ensure smooth metal filling and defect minimization, we designed a bottom-gating system using ProCAST simulations. Key features include:
- Four bottom ingates connected to four blind risers.
- Symmetric placement to balance thermal gradients.
- Steady upward filling to prevent turbulence.
The design achieved a filling time of 120 seconds at a pouring temperature of 1150°C. The simulation results (Figure 1) confirmed stable metal flow without splashing or air entrapment.
Table 2: Critical Parameters for Gating System Simulation
| Parameter | Value |
|---|---|
| Pouring Temperature | 1150°C |
| Mold Initial Temperature | 80°C |
| Heat Transfer Coefficient | 500 W/m²·K |
| Filling Time | 120 s |
3.3 Riser Design and Solidification Analysis
Risers were strategically placed to address shrinkage defects identified through ProCAST simulations. The final design included:
- Six elliptical risers for upper cover plate feeding.
- Cylindrical risers at blade roots and tails.
- Four blind risers at the bottom for directional solidification.
The solidification sequence (Figure 2) demonstrated that the risers solidified last, ensuring effective feeding. Defects were concentrated in risers and gating channels, validating the design.
Mathematical Model for Solidification
The heat transfer during solidification can be approximated using Fourier’s law:∂T∂t=α(∂2T∂x2+∂2T∂y2+∂2T∂z2)∂t∂T=α(∂x2∂2T+∂y2∂2T+∂z2∂2T)
where αα is the thermal diffusivity of the resin sand mold.
4. 3D-Printed Sand Mold Fabrication
4.1 Core Design and Conformal Venting
Traditional venting channels are inadequate for complex geometries. Using 3D printing, we integrated conformal vents along blade cores to ensure efficient gas escape. Key steps included:
- Unibody Core Design: Eliminated assembly errors in critical blade sections.
- Conformal Vents: Followed the curvature of blades to optimize gas flow.
Table 3: 3D Printing Parameters for Sand Molds
| Parameter | Value |
|---|---|
| Laser Power | 900 W |
| Spot Size | 1.1 mm |
| Layer Thickness | 0.3 mm |
| Scanning Speed | 2900 mm/s |
| Material | Resin-Coated Sand |
4.2 Post-Processing and Assembly
Printed molds underwent:
- Surface Hardening: Flame treatment to strengthen the mold surface.
- Sand Backfilling: Loose sections were stabilized with silica sand.
- Thermal Curing: Heated to 180°C for binder activation.
5. Casting Trials and Results
5.1 Pouring and Defect Analysis
The assembled molds were poured at 1150°C, with vents ignited to expel gases. Post-casting inspection revealed:
- Defect Distribution: 95% of shrinkage porosity localized in risers.
- Dimensional Accuracy: Blade thickness met ±0.5 mm tolerance.
Table 4: Quality Metrics for Final Impeller Casting
| Metric | Value |
|---|---|
| Shrinkage Porosity | ≤2% (in risers) |
| Surface Roughness | Ra 12.5 µm |
| Dimensional Deviation | ±0.5 mm |
5.2 Economic and Technical Benefits
- Cycle Time Reduction: 3D printing cut mold production time by 60% compared to traditional sand casting.
- Material Savings: 30% less sand waste due to precise additive manufacturing.
6. Conclusion
This study demonstrates that 3D-printed sand molds are a game-changer for impeller casting, particularly for complex geometries like double-suction impellers. Key achievements include:
- Defect Control: ProCAST-optimized gating and riser systems minimized shrinkage and gas defects.
- Precision Manufacturing: Unibody cores and conformal vents ensured dimensional accuracy.
- Cost Efficiency: Additive manufacturing reduced lead times and material waste in sand casting.
Future work will explore hybrid processes combining 3D printing with advanced alloys to further enhance impeller casting performance.