As an engineer specializing in aluminum alloy metal mold casting, I have been deeply involved in the production of critical components for aerospace applications. Aerospace castings, such as gear pump housings, are essential parts of aero-engine fuel systems, requiring high precision, complex internal structures, and stringent performance standards. These aerospace castings often feature intricate geometries with varying wall thicknesses, leading to significant challenges in achieving defect-free production. In this study, I focus on a specific aerospace casting—an aluminum alloy gear pump shell—that exhibited severe underfilling defects at the oil filter tank area, resulting in low qualification rates and posing risks to production schedules and quality. Through detailed simulation analysis using AnyCasting software, I aimed to optimize the casting process to mitigate these defects and improve the reliability of aerospace castings.
The gear pump shell is a typical example of aerospace castings with demanding requirements. Its structure includes multiple internal oil passages, thick and thin walls交错, and drastic cross-sectional changes, as illustrated in the following figure. For instance, the “8-shaped cavity” has a wall thickness of 40 mm, while the oil filter tank area is only 4.5 mm thick. This complexity makes the mold cavity irregular, and the large-size thin-walled sections are prone to cold shuts and underfilling defects. The casting material is ZL101A aluminum alloy, and the process employed is metal mold tilt casting, which reduces turbulence and gas entrapment during pouring. However, despite this, the qualification rate remained as low as 22.2%, with underfilling defects accounting for 45.3% of rejections. This highlighted the urgent need for process optimization in aerospace castings production.
To address this issue, I utilized AnyCasting simulation software to analyze the filling and solidification processes. The simulation allowed me to visualize flow fields, temperature distributions, and defect formation in aerospace castings. I began by setting up the numerical model. The 3D geometry of the casting, mold, and cores was meshed with approximately 9 million uniform grids. Key material properties and process parameters were defined, as summarized in Table 1. The aluminum alloy ZL101A and mold material H13 have specific thermal physical properties that influence heat transfer during casting. The pouring temperature was initially set at 720°C, and the mold preheat temperature was 280°C. A ZnO coating with a thickness of 200 μm was applied to the mold cavity to affect heat flow.
| Parameter | Value | Description |
|---|---|---|
| Aluminum Alloy | ZL101A | Material for the aerospace casting |
| Mold Material | H13 | Tool steel with high thermal conductivity |
| Pouring Temperature | 720°C | Initial temperature of molten aluminum |
| Mold Preheat Temperature | 280°C | Initial temperature of the metal mold |
| Cavity Coating | ZnO, 200 μm | Insulating layer to reduce heat loss |
| Core Material | Furan Sand | Used for forming internal cavities |
The tilt casting process involves rotating the mold to control the filling sequence. I divided the filling into seven stages, each with specific tilt angles and speeds, as shown in Table 2. This parameterization is crucial for ensuring smooth filling in aerospace castings, as excessive velocity can cause turbulence and defects. According to J. Campbell’s research, if the alloy flow velocity exceeds 50 cm/s, surface turbulence occurs, leading to gas entrapment and slag inclusions. The simulation confirmed that the tilt parameters kept the flow velocity below this threshold, minimizing such defects.
| Stage | Tilt Angle (°) | Tilt Speed (1/400 L/min) |
|---|---|---|
| 1 | 85 | 50 |
| 2 | 65 | 65 |
| 3 | 45 | 80 |
| 4 | 30 | 90 |
| 5 | 15 | 75 |
| 6 | 1 | 45 |
| 7 | -2 | 10 |
Through simulation, I analyzed the flow and temperature fields during filling. The flow velocity remained below 50 cm/s throughout, indicating a平稳充型过程. However, the temperature field revealed critical insights. At 70% filling, the oil filter tank area showed temperatures around 590°C, significantly lower than other regions at 640°C or above. Since ZL101A has a phase transformation temperature range of 555°C to 615°C, the molten aluminum at the defect site began solidifying prematurely, reducing fluidity and leading to underfilling. This is a common issue in aerospace castings with thin-walled sections.
The solidification simulation further confirmed this. Within 5% of solidification, the oil filter tank area cooled rapidly, becoming the first region to solidify completely by 30% solidification. The temperature differential was stark, as described by the heat conduction equation. For aerospace castings, heat transfer plays a pivotal role in defect formation. The Fourier’s law of heat conduction can be expressed as:
$$ q = -k \nabla T $$
where \( q \) is the heat flux, \( k \) is the thermal conductivity, and \( \nabla T \) is the temperature gradient. For H13 mold steel, \( k \) is approximately 40 W/m·°C, which is relatively high, promoting rapid heat loss from thin-walled areas. Additionally, the solidification time can be estimated using Chvorinov’s rule:
$$ t = C \left( \frac{V}{A} \right)^2 $$
where \( t \) is the solidification time, \( C \) is a constant dependent on material and mold properties, \( V \) is the volume, and \( A \) is the surface area. For the oil filter tank, the high surface-area-to-volume ratio accelerates cooling, exacerbating underfilling risks in aerospace castings.
Based on this analysis, I identified three main factors contributing to the defect in aerospace castings: (1) the large-size thin-walled geometry of the oil filter tank, (2) the inability to place a riser above this non-machined surface for heat supplementation, and (3) the high thermal conductivity of the H13 steel core and mold. To address these, I proposed targeted process improvements aimed at enhancing the temperature of the molten aluminum and reducing heat loss during filling.
The improvements included increasing the pouring temperature, elevating the mold preheat temperature, and applying an insulating coating to the steel core. I conducted experiments to quantify the effect of the coating. As shown in Table 3, the cooling rate of the mold without coating was 30°C/min, whereas with the coating, it reduced to 10°C/min, improving insulation efficiency by three times. This is vital for maintaining adequate fluidity in aerospace castings.
| Condition | Cooling Rate (°C/min) | Improvement Factor |
|---|---|---|
| Without Coating | 30 | 1 |
| With Coating | 10 | 3 |
The specific measures implemented were: (1) raising the pouring temperature from 730°C to 740°C (the upper limit per工艺规程), (2) preheating the steel core and mold cavity locally to 380°C–400°C, compared to the previous overall preheat of 300°C–350°C, and (3) ensuring timely application and reapplication of the insulating coating. Previously, the coating was applied once per batch and patched only when脱落; after improvement, the coating was stripped and reapplied every five castings, with a thickness of 0.2 mm–0.3 mm, and any脱落 was immediately repaired.
To validate these improvements, I reran the simulation with updated parameters. The results showed a marked increase in temperature at the oil filter tank area. After 67% filling, the temperature was around 700°C, much higher than before. During solidification, while the sequence remained similar, the time to reach 5% solidification increased from 42.5 s to 46.6 s, indicating slower cooling and better fluidity. This aligns with the Navier-Stokes equations for fluid flow, which govern the motion of molten aluminum in aerospace castings:
$$ \rho \left( \frac{\partial \mathbf{v}}{\partial t} + \mathbf{v} \cdot \nabla \mathbf{v} \right) = -\nabla p + \mu \nabla^2 \mathbf{v} + \mathbf{f} $$
where \( \rho \) is density, \( \mathbf{v} \) is velocity, \( p \) is pressure, \( \mu \) is dynamic viscosity, and \( \mathbf{f} \) represents body forces. Higher temperatures reduce viscosity, improving flow characteristics and reducing defects in aerospace castings.
The production data after implementing these changes demonstrated significant improvement. The underfilling defect rate dropped from 45.3% to 8.4%, and the overall qualification rate for aerospace castings increased from 22.2% to 68.7%. This is summarized in Table 4, highlighting the effectiveness of simulation-driven optimization.
| Metric | Before Optimization | After Optimization |
|---|---|---|
| Underfilling Defect Rate | 45.3% | 8.4% |
| Overall Qualification Rate | 22.2% | 68.7% |
In conclusion, this study underscores the value of simulation technology in optimizing processes for aerospace castings. By using AnyCasting to analyze flow and temperature fields, I identified the root causes of underfilling defects in a complex gear pump shell and implemented targeted improvements. The integration of higher pouring temperatures, enhanced mold preheating, and effective insulating coatings proved successful in reducing defects and boosting qualification rates. These findings offer valuable insights for the production of similar aerospace castings, emphasizing the importance of thermal management and process control. As aerospace castings continue to evolve in complexity, simulation-based approaches will be indispensable for ensuring quality and efficiency in manufacturing.
Looking ahead, further research could explore advanced materials for coatings or dynamic control of tilt parameters using real-time simulation feedback. The principles discussed here—such as heat transfer modeling and fluid dynamics analysis—are broadly applicable to other types of aerospace castings, including turbine blades and structural components. By continually refining these methods, we can achieve higher performance and reliability in aerospace castings, supporting the advancement of aviation technology. This work also highlights the need for interdisciplinary collaboration in aerospace castings production, combining metallurgy, mechanical engineering, and computational simulation to overcome manufacturing challenges.