In modern industries such as defense, aerospace, and marine engineering, the demand for aluminum alloy castings with complex structures, high dimensional accuracy, and superior internal quality is increasingly critical. Traditional sand casting methods often struggle to meet these stringent requirements due to limitations in mold design, precision, and defect control. To address these challenges, we have developed an integrated approach combining anti-gravity low-pressure casting, 3D printing (3DP) of sand molds, and computational simulation for producing a bridge bracket casting. This article details our first-person perspective on the process design, optimization, and manufacturing, emphasizing the role of sand casting throughout.
The bridge bracket casting, made of ZL114A aluminum alloy according to GB/T1173-2013 with T5 heat treatment, has overall dimensions of 500 mm × 374 mm × 388 mm and a mass of 27 kg. Its wall thickness varies from 9 mm to 34 mm, and it must adhere to a dimensional tolerance grade of DCTG8 per GB/T6414-2017. Key technical requirements include high internal quality in axle hole areas, with no shrinkage porosity, cavities, or sand inclusions allowed, and full X-ray inspection to meet Class II standards of GB/T9438-2013. The casting’s complexity, featuring hollow sections and weight-reduction cavities accessible only through small process holes, posed significant challenges for conventional sand casting. Thus, we opted for a low-pressure casting process to ensure smooth filling and dense microstructure under pressure, complemented by 3DP sand molds for rapid, precise mold fabrication without traditional pattern-making.
Our initial casting process design involved a bottom-gating system with slit feeders to facilitate controlled metal flow. The gating system included four slit gates measuring 35 mm × 20 mm and four slag traps of φ80 mm × 400 mm, along with 12 rectangular ingates at the bottom, each 45 mm in height with an 8° taper. To prevent deformation, we added two process ribs at the top. Additionally, 30 chill blocks, each 20 mm thick, were placed in thick sections to enhance cooling. We used ProCAST simulation software to model the process, setting a pouring temperature of 690°C and a filling time of 45 s. The simulation results indicated stable filling without turbulence, as the molten metal entered through the bottom ingates and progressed steadily. The temperature field during solidification showed that the top sections with chills solidified first, followed by the axle holes, while the bottom ingates solidified last, achieving directional solidification. However, the simulation predicted minor shrinkage defects in the top process ribs and bottom areas, which we deemed acceptable for initial trials.
| Parameter | Value |
|---|---|
| Pouring Temperature | 690°C |
| Filling Time | 45 s |
| Gating System Type | Bottom with Slit Feeders |
| Chill Blocks | 30 (20 mm thick) |
| Predicted Defects | Minor shrinkage in top ribs and bottom |
For the sand casting mold design, we employed a three-box structure comprising upper, middle, and lower boxes, all fabricated using 3DP technology with 100-mesh silica sand and a layer thickness of 0.3 mm. The upper box included pre-formed slots for chill insertion, the middle box housed an integrated core with reinforcement ribs to prevent distortion during drying, and the lower box featured four positioning cores for alignment. A casting shrinkage rate of 1% was applied, with draft angles of 8° and fit clearances of 0.5 mm. The mold’s sand thickness was maintained at 35 mm to ensure strength. After printing, the molds were coated twice with a refractory coating and dried at 120–140°C for 4 hours before assembly. The chill blocks were sandblasted, dried, and fixed using a casting adhesive.
In the initial production trial, we used anti-gravity low-pressure casting with a filling speed of 45 mm/s and a pressure of 50 kPa, followed by a crystallization time of 600 s. The alloy was melted, treated with a ternary modifier (62.5% NaCl, 25% NaF, 12.5% KCl) at 730°C, and refined with argon at 720°C to achieve a density equivalent (DI) of 0.3%, indicating low hydrogen content. After pouring, the castings underwent T5 heat treatment: solution treatment at 535°C for 12 hours, water quenching at 80°C, and aging at 160°C for 6 hours. However, visual inspection revealed localized missing wall sections in the bottom cavities, and dimensional measurements showed that the theoretical size SR217 mm was out of tolerance. X-ray inspection detected shrinkage porosity in the bottom areas. Analysis indicated that the middle box cores, connected only through small process holes, had insufficient strength and fractured during pouring, leading to core floating and misalignment. This sand casting issue resulted in uneven cavity floors and dimensional inaccuracies.
| Issue | Location | Cause |
|---|---|---|
| Missing Wall Thickness | Bottom Cavities | Core Fracture and Floating |
| Dimensional Out-of-Tolerance | SR217 mm | Core Misalignment |
| Shrinkage Porosity | Bottom Areas | Local Overheating |
To address these sand casting challenges, we optimized the process in three key areas. First, we modified the process holes in the weight-reduction cavities: the original four φ10 mm holes were changed to two φ20 mm and two φ14 mm holes, resulting in a total of six φ20 mm and two φ14 mm holes to enhance core strength. Second, we added 10 additional chill blocks, each 15 mm thick, to the bottom areas to accelerate cooling and reduce shrinkage. The solidification time in sand casting can be estimated using Chvorinov’s rule: $$ t = C \left( \frac{V}{A} \right)^2 $$ where \( t \) is the solidification time, \( V \) is the volume, \( A \) is the surface area, and \( C \) is a constant dependent on the mold material and casting conditions. By increasing the cooling rate with chills, we aimed to minimize \( t \) in critical zones. Re-simulation with ProCAST confirmed that these changes eliminated the predicted shrinkage defects. Third, we redesigned the sand mold structure, adjusting the parting surfaces to distribute forces more evenly. Specifically, we set a 1 mm clearance at contact surface A between the middle and lower boxes, while maintaining 0.5 mm clearances at other surfaces, ensuring that filling forces were dispersed to prevent core fracture.
After implementing these optimizations, we reprinted the 3DP sand molds and conducted another low-pressure casting trial. The results were significantly improved: the castings exhibited complete walls with no visual defects, and dimensional measurements met DCTG7 standards, exceeding the required DCTG8. X-ray inspection showed no shrinkage porosity in the bottom areas, confirming the effectiveness of the added chills. Chemical composition and mechanical properties of attached test specimens complied with GB/T1173-2013, as summarized in the tables below. This successful sand casting approach was then scaled for batch production, yielding consistent, high-quality castings.
| Element | Standard Range (%) | Actual Value (%) |
|---|---|---|
| Si | 6.5–7.5 | 6.87 |
| Mg | 0.45–0.75 | 0.53 |
| Ti | 0.10–0.20 | 0.15 |
| Fe | ≤0.2 | 0.16 |
| Al | Balance | Balance |
| Property | Specimen 1 | Specimen 2 | Specimen 3 | Standard Requirement | Result |
|---|---|---|---|---|---|
| Tensile Strength (MPa) | 316 | 303 | 312 | ≥290 | Qualified |
| Elongation (%) | 3.0 | 3.0 | 3.5 | ≥2 | Qualified |
The integration of simulation, 3DP sand molds, and low-pressure casting proved highly effective for this complex sand casting application. By iteratively refining the process based on initial trials, we achieved castings with internal quality meeting Class II standards and dimensional accuracy up to DCTG7. The use of 3DP sand molds enabled rapid prototyping and production without traditional patterns, reducing lead times and costs while ensuring consistency. In sand casting, factors such as gating design, cooling rate, and mold integrity are critical; our experience underscores the importance of a holistic approach that combines theoretical modeling with practical adjustments. For instance, the pressure during low-pressure filling can be described by: $$ P = \rho g h $$ where \( P \) is the pressure, \( \rho \) is the molten metal density, \( g \) is gravity, and \( h \) is the height difference. By controlling these parameters, we maintained stable filling and minimized defects.
In conclusion, our work demonstrates that advanced sand casting techniques, leveraging 3DP and simulation, can overcome the limitations of conventional methods for producing high-integrity aluminum alloy castings. The bridge bracket case study highlights how iterative optimization—addressing core strength, cooling, and mold design—can lead to successful outcomes in sand casting. This methodology is particularly valuable for low-volume, high-complexity components, offering a scalable solution for modern industrial demands. Future efforts could focus on further automating the optimization process and expanding the application of 3DP sand molds to other alloy systems and geometries.