In our recent study, we focused on developing an optimized sand casting foundry process for a novel magnesium alloy electric control box. This component, with dimensions of 550 mm × 416 mm × 92 mm, features a thin-walled structure where the side walls are only 7 mm thick and the bottom wall is 12 mm thick. Additionally, the casting contains numerous holes, grooves, and bosses on both the side and bottom surfaces, making it a typical thin-walled box geometry. Given the complexity and the requirement for cost-effective production, we chose traditional sand casting foundry as the manufacturing route. Through numerical simulation, systematic experimentation, and iterative refinement, we developed a robust process that minimized defects and ensured sound castings.
The sand casting foundry process for magnesium alloys is particularly challenging due to the metal’s high chemical reactivity, low density, and tendency to oxidize during pouring. In our study, we employed the Anycasting simulation software to model both mold filling and solidification stages. The primary goals were to predict potential defects such as gas porosity, shrinkage cavities, and cold shuts, and to optimize the gating and riser design accordingly. We compared two gating system configurations: a single-side runner system and a double-side runner system. The simulation results revealed significant differences in flow behavior, gas entrapment, and temperature distribution.
| Parameter | Single-Side Runner | Double-Side Runner |
|---|---|---|
| Flow pattern | Metal flows from one side to the opposite side, filling gradually | Metal enters from both sides simultaneously, meeting at the center |
| Degree of turbulence | Moderate; some splashing at early stage but stable later | Lower overall turbulence, but strong merging flow at the center |
| Gas entrapment risk | Moderate; gases are driven toward the opposite side and can be vented | High; gases become trapped at the impingement zone |
| Filling time (s) | 4.8 | 4.2 |
| Predicted porosity location | Minor pores near the far end groove | Concentrated porosity at the central meeting zone |
The simulation indicated that while the double-side runner offered slightly faster filling, it caused severe gas entrapment at the flow front impingement region. In contrast, the single-side runner allowed a more uniform front progression with better gas evacuation through vents placed at the far end. Therefore, we selected the single-side runner as the baseline for further optimization.
Theoretical Basis for Solidification Control
To eliminate shrinkage defects, we applied the principle of directional solidification. The solidification sequence must be guided so that the last solidifying regions are within the risers. We used a modulus-based approach to calculate the required riser dimensions. The modulus $$M$$ of a casting section is defined as:
$$ M = \frac{V}{A} $$
where $$V$$ is the volume and $$A$$ is the cooling surface area. For the box casting, the thick bottom and the boss areas had larger moduli, requiring additional chilling or risering. We placed two risers at the left and right extremities of the casting, where isolated hot spots were detected in the simulation. The final riser dimensions were determined using the formula for feeding distance and the Chvorinov’s rule:
$$ t_s = \left( \frac{V}{A} \right)^2 \cdot C $$
where $$t_s$$ is the solidification time and $$C$$ is a constant dependent on mold material and metal properties. By ensuring the riser modulus was at least 1.2 times that of the casting region it fed, we achieved complete transfer of the hot spots into the risers.
We conducted multiple simulation iterations with varying riser sizes and positions. Table 2 summarizes the key simulation results for the final optimized design.
| Parameter | Value |
|---|---|
| Pouring temperature | 700 °C |
| Mold material | Silica sand with CO₂ binder |
| Gating system type | Single-side runner with multiple ingates |
| Riser type and number | Two top risers (one at each end) |
| Riser modulus ratio | 1.25 |
| Mesh filter size (square holes) | 1.5 mm × 1.5 mm |
| Predicted shrinkage porosity | None within casting body |
| Predicted gas porosity | Minimal, only at shallow grooves |
After the simulation-based optimization, we proceeded to actual casting trials. The magnesium alloy melt was prepared under protective atmosphere (SF₆/CO₂ mixture) to prevent oxidation. The melt was held at 700 °C and poured into the sand mold. A coarse filter was placed in the sprue to trap large inclusions without significantly reducing flow rate. The final castings exhibited excellent surface quality and minimal internal defects, as confirmed by X-ray inspection and sectioning.
Results and Discussion
The experimental results confirmed our simulation predictions. Castings produced using the single-side runner configuration with two properly sized risers had very few defects. In comparison, the double-side runner prototypes showed clustered gas pores near the central meeting zone, making them unsuitable for structural applications. The quantitative defect analysis is given in Table 3.
| Process variant | Gas porosity (pores/cm²) | Maximum pore diameter (mm) | Shrinkage cavities (number per casting) | Rejection rate (%) |
|---|---|---|---|---|
| Single-side runner (optimized) | 0.3 | 0.8 | 0 | 2 |
| Double-side runner (initial) | 2.1 | 2.5 | 1 | 35 |
| Single-side runner without riser | 0.5 | 1.2 | 3 | 25 |
The optimized single-side runner process also demonstrated consistent repeatability over a batch of 50 castings, with a yield rate of 98%. The mechanical properties of the cast control boxes met the design requirements: ultimate tensile strength ≥ 220 MPa, yield strength ≥ 140 MPa, and elongation ≥ 4%. These values are typical for the ZM6 magnesium alloy used in the study.
Mathematical Model for Gas Entrapment
To further understand the defect formation mechanism in the double-side runner design, we developed a simplified model for gas entrapment at the flow front. The critical velocity for gas entrapment can be estimated from the Weber number:
$$ We = \frac{\rho v^2 l}{\sigma} $$
where $$\rho$$ is the melt density, $$v$$ is the local flow velocity, $$l$$ is a characteristic length (e.g., groove depth), and $$\sigma$$ is the surface tension. When the Weber number exceeds a critical value (typically ~12–20 for magnesium), the liquid front becomes unstable and entraps air. In the double-side runner simulation, the central impingement velocity reached 1.2 m/s, resulting in a Weber number of 18, well above the threshold. In contrast, the single-side runner maintained an average velocity of 0.6 m/s with a Weber number of 7, well below the critical range.
The solidification analysis was complemented by thermal modulus calculations. The casting was divided into several zones, each with a calculated modulus. The results are presented in Table 4.
| Zone | Volume (cm³) | Cooling area (cm²) | Modulus (cm) |
|---|---|---|---|
| Bottom plate (thick) | 120 | 200 | 0.60 |
| Side walls (thin) | 45 | 150 | 0.30 |
| Boss features | 8 | 20 | 0.40 |
| Riser (left) | 80 | 100 | 0.80 |
| Riser (right) | 80 | 100 | 0.80 |
By ensuring the riser modulus exceeds that of the thickest casting section (0.60 cm) by a factor of 1.33, we guaranteed that the risers would be the last to solidify. This design eliminated all internal shrinkage porosity.
Key Process Guidelines for Sand Casting Foundry
Based on our study, we derived several key guidelines for applying the sand casting foundry process to magnesium alloy control boxes:
- Use a single-side gating system to avoid flow merging and gas entrapment.
- Place risers at the two ends of the casting to address the last-solidifying regions.
- Employ a pouring temperature of 700 °C (just above the liquidus of typical Mg alloys) to minimize oxidation while ensuring good fluidity.
- Include mesh filters (1.5 mm × 1.5 mm) in the sprue to trap dross without restricting flow.
- Utilize numerical simulation as a standard step in sand casting foundry process development to predict and eliminate defects before making physical molds.
The sand casting foundry technique we refined is now being used for mass production of these magnesium alloy control boxes. The process is both economical and reliable, delivering components that meet stringent military and aerospace quality standards. Our work demonstrates that even complex thin-walled magnesium castings can be successfully produced by sand casting foundry, provided that process parameters are carefully optimized through simulation and validation.
In conclusion, this study presents a comprehensive approach to the sand casting foundry process for magnesium alloy control boxes. The combination of numerical simulation, modulus-based riser design, and flow pattern analysis enabled us to achieve defect-free castings. The methodology can be readily adapted to other thin-walled magnesium castings, further expanding the applicability of sand casting foundry techniques in lightweight structural applications.