In industries such as power generation, petrochemicals, metallurgy, shipbuilding, and equipment manufacturing, large bearing housings are critical components. Examples include turbine rotor supports in nuclear power equipment, blade supports in hydroelectric and wind power units, and support bases for large rolling mills. The ability to produce high-quality large bearing housings directly impacts the quality, safety, and progress of national key engineering projects, holding significant importance for the national economy and people’s livelihoods. This article explores the casting process design and optimization for mass-producing bearing housing steel castings using sand casting methods. We focus on addressing challenges such as shrinkage defects through computational simulations and iterative improvements.
The bearing housing discussed here is a large steel casting with contour dimensions of 2,486 mm × 1,639 mm × 1,340 mm and a wall thickness ranging from 50 mm to 254 mm. As a thick-walled, hollow rectangular-shaped component, it presents difficulties in casting due to significant variations in wall thickness, especially after machining allowances are applied. To tackle this, we employed UG software for CAD modeling, establishing基准 planes and using commands like extrusion, subtraction, and edge blending to create a precise 3D solid model. This model served as the foundation for subsequent sand casting process design.
For the casting process, we selected sand casting due to its suitability for batch production. The material is ZG310-570 steel, and the process involves using a false mold to create upper, middle, and lower molds. We adopted mechanical and manual sand filling with phenolic-modified furan resin self-hardening sand, alcohol-based coatings, and a silver graphite powder parting agent for demolding. The parting surface was designed as a three-box curved surface to accommodate the complex geometry, ensuring dimensional accuracy and production efficiency.
The gating system was designed as an open stepped system, using a bottom-pouring ladle. We calculated the dimensions based on the choke section method, which determines the cross-sectional areas of the sprue, runner, and ingates to ensure proper metal flow and minimize turbulence. The formula for the choke area \( A_c \) is derived from the pouring time \( t \) and the casting weight \( W \):
$$ A_c = \frac{W}{\rho \cdot v \cdot t} $$
where \( \rho \) is the metal density, and \( v \) is the flow velocity. For this sand casting process, we set the pouring temperature between 1,530°C and 1,570°C, with a pouring time of 65–110 seconds. The calculated dimensions for the gating system components are summarized in Table 1.
| Component | Diameter (mm) | Height/Length (mm) |
|---|---|---|
| Sprue (Lower) | 140 | 1,016 |
| Sprue (Upper) | 140 | 527 |
| Runner | 100 | 1,668 / 1,420 |
| Ingate | 140 | 350 / 410 |
Risers were incorporated to compensate for solidification shrinkage, complemented by chills to promote directional solidification. Initially, we used four cylindrical open risers with diameters of 323 mm and 336 mm, and a height of 485 mm. The riser volume \( V_r \) was estimated based on the casting modulus \( M_c \), which is the ratio of volume to surface area:
$$ M_c = \frac{V}{A} $$
For effective feeding in sand casting, the riser modulus \( M_r \) should satisfy \( M_r > M_c \). Additionally, we designed contraction ribs to prevent hot tearing, with thickness \( t \) calculated as:
$$ t = \frac{1}{3} \delta \quad \text{to} \quad \frac{1}{4} \delta $$
where \( \delta \) is the auxiliary wall thickness. For this case, \( t = 65 \, \text{mm} \), length \( l = 8t \) to \( 12t = 520 \, \text{mm} \), and spacing \( d = 15t \) to \( 20t = 975 \, \text{mm} \). Given the spacing constraints, only one symmetrically placed rib was feasible.
Core design involved a cylindrical core for the central cavity and four smaller cores for the mounting seats, each with vent holes to facilitate gas escape during the sand casting process. This ensured the internal geometry was accurately reproduced while maintaining structural integrity.
We performed solidification simulation using Huacast CAE software to predict defects like shrinkage porosity and cavities. The initial results revealed significant shrinkage defects within the casting, indicating inadequate riser performance. This highlighted the need for optimization in the sand casting process. The simulation output showed that the original riser design failed to achieve sequential solidification, leading to isolated hot spots.
To address these issues, we optimized the riser design by switching to two symmetrical腰形 open risers, calculated using the modulus method. The new riser dimensions were determined to enhance feeding efficiency. The modulus for a腰形 riser can be approximated as:
$$ M_r = \frac{V_r}{A_r} $$
where \( V_r \) is the riser volume and \( A_r \) is the surface area. By increasing \( M_r \) relative to \( M_c \), we ensured that risers solidified after the casting, effectively compensating for shrinkage. The optimized sand casting process was re-simulated, showing a marked reduction in defects, with shrinkage concentrated primarily in the risers.
| Parameter | Initial Design | Optimized Design |
|---|---|---|
| Riser Type | Cylindrical Open | 腰形 Open |
| Number of Risers | 4 | 2 |
| Diameter (mm) | 323 / 336 | Custom腰形 |
| Height (mm) | 485 | Adjusted for Modulus |
| Shrinkage Defects | Significant | Minimal |
The optimization process demonstrated that sand casting, when combined with computational tools, can yield high-quality large bearing housings. By iteratively refining the riser design, we achieved a more reliable process that minimizes defects and enhances production efficiency. This approach underscores the versatility of sand casting for complex geometries and batch production, contributing to safer and more economical manufacturing in critical industries.
In conclusion, the sand casting process for large bearing housings was successfully designed and optimized through CAD modeling, systematic parameter calculation, and CAE simulation. The use of sand casting allowed for cost-effective mass production, while optimization efforts focused on riser design effectively mitigated shrinkage issues. This methodology provides a practical framework for similar applications, emphasizing the importance of simulation in advancing sand casting technologies. Future work could explore further refinements, such as alternative gating systems or advanced materials, to push the boundaries of sand casting performance.