In modern automotive manufacturing, the production of critical components like exhaust manifolds relies heavily on sand casting processes. These sand casting parts are integral to engine systems, demanding exceptional dimensional accuracy and structural integrity. However, the intricate geometries involved—characterized by thin walls and complex curved surfaces—pose significant challenges, often leading to defects such as shrinkage porosity, cold shuts, slag inclusions, and gas entrapment. These issues not only compromise the quality of sand casting parts but also increase production costs and延长 lead times. Traditional trial-and-error methods for process improvement are time-consuming and inefficient, especially since internal casting phenomena are difficult to observe directly. To address this, numerical simulation has emerged as a powerful tool, enabling detailed analysis of mold filling and solidification behaviors before physical prototyping. This study focuses on leveraging simulation software to optimize the sand casting process for a complex exhaust manifold, a典型 sand casting part, aiming to minimize defects and enhance production efficiency.
The foundation of casting simulation lies in solving the fundamental governing equations of fluid flow and heat transfer. For sand casting parts, the process involves the flow of molten metal with a free surface into a geometric cavity, accompanied by heat transfer to the surrounding mold. Assuming the metal is an incompressible viscous fluid and neglecting turbulence effects initially, the process adheres to the conservation laws of mass, momentum, and energy. The continuity equation ensures mass conservation:
$$ \frac{\partial \rho}{\partial t} + \frac{\partial (\rho u)}{\partial x} + \frac{\partial (\rho v)}{\partial y} + \frac{\partial (\rho w)}{\partial z} = 0 $$
where \( \rho \) is density, \( t \) is time, and \( u \), \( v \), and \( w \) are velocity components in the \( x \), \( y \), and \( z \) directions, respectively. The momentum conservation is described by the Navier-Stokes equations:
$$ \frac{\partial (\rho u)}{\partial t} + \text{div}(\rho u \mathbf{u}) = \text{div}(\mu \nabla u) – \frac{\partial p}{\partial x} + S_u $$
$$ \frac{\partial (\rho v)}{\partial t} + \text{div}(\rho v \mathbf{u}) = \text{div}(\mu \nabla v) – \frac{\partial p}{\partial y} + S_v $$
$$ \frac{\partial (\rho w)}{\partial t} + \text{div}(\rho w \mathbf{u}) = \text{div}(\mu \nabla w) – \frac{\partial p}{\partial z} + S_w $$
Here, \( \mu \) is the dynamic viscosity, \( p \) is pressure, and \( S_u, S_v, S_w \) are source terms. The energy equation governs heat transfer during solidification of sand casting parts:
$$ \frac{\partial (\rho T)}{\partial t} + \frac{\partial (\rho u T)}{\partial x} + \frac{\partial (\rho v T)}{\partial y} + \frac{\partial (\rho w T)}{\partial z} = \frac{\partial}{\partial x}\left( \frac{k}{c_p} \frac{\partial T}{\partial x} \right) + \frac{\partial}{\partial y}\left( \frac{k}{c_p} \frac{\partial T}{\partial y} \right) + \frac{\partial}{\partial z}\left( \frac{k}{c_p} \frac{\partial T}{\partial z} \right) + S_T $$
where \( T \) is temperature, \( k \) is thermal conductivity, \( c_p \) is specific heat capacity, and \( S_T \) is a heat source term. To account for turbulence in real-world casting flows, the standard \( k-\epsilon \) model is often employed. This model introduces turbulent kinetic energy \( k \) and its dissipation rate \( \epsilon \), with the turbulent viscosity \( \mu_t \) given by:
$$ \mu_t = \rho C_\mu \frac{k^2}{\epsilon} $$
The transport equations for \( k \) and \( \epsilon \) are:
$$ \frac{\partial (\rho k)}{\partial t} + \frac{\partial (\rho k u_i)}{\partial x_i} = \frac{\partial}{\partial x_j}\left[ \left( \mu + \frac{\mu_t}{\sigma_k} \right) \frac{\partial k}{\partial x_j} \right] + G_k + G_b – \rho \epsilon – Y_M + S_k $$
$$ \frac{\partial (\rho \epsilon)}{\partial t} + \frac{\partial (\rho \epsilon u_i)}{\partial x_i} = \frac{\partial}{\partial x_j}\left[ \left( \mu + \frac{\mu_t}{\sigma_\epsilon} \right) \frac{\partial \epsilon}{\partial x_j} \right] + C_{1\epsilon} \frac{\epsilon}{k} (G_k + C_{3\epsilon} G_b) – C_{2\epsilon} \rho \frac{\epsilon^2}{k} + S_\epsilon $$
In these equations, \( G_k \) and \( G_b \) represent generation terms due to mean velocity gradients and buoyancy, respectively, \( Y_M \) accounts for compressibility effects, and \( C_{\mu}, C_{1\epsilon}, C_{2\epsilon}, C_{3\epsilon}, \sigma_k, \sigma_\epsilon \) are empirical constants. Solving these equations numerically allows for the prediction of flow patterns and temperature distributions during the casting of sand casting parts, which is crucial for identifying potential defects.
Preparation for simulation is a critical step in analyzing sand casting parts. The exhaust manifold, as a complex sand casting part, was designed using CAD software to create a precise 3D model. This component features four branching pipes connected to a central manifold, all with thin walls and intricate curvatures. The geometry was then integrated with a gating system designed for sand casting. To optimize production, a “one mold, two cavities” approach was adopted, meaning two identical parts are cast simultaneously. The gating system included a sprue, runner, and ingates with flat trapezoidal cross-sections to facilitate slag trapping and removal. Given the thick sections at the pipe ends, five risers were strategically placed to provide feeding during solidification and prevent shrinkage defects in these sand casting parts. The assembly of the casting, gating system, and risers was imported into simulation software for meshing. Due to the complex surfaces, local mesh refinement was applied, resulting in approximately 12 million cells to ensure accuracy. Key material properties and process parameters were defined, as summarized in the tables below.
| Element | Content (wt.%) |
|---|---|
| Carbon (C) | 3.47 |
| Silicon (Si) | 3.61 |
| Molybdenum (Mo) | 0.62 |
| Manganese (Mn) | 0.18 |
| Phosphorus (P) | 0.035 |
| Sulfur (S) | 0.017 |
| Magnesium (Mg) | 0.039 |
| Temperature (°C) | Density (g/cm³) | Specific Heat (cal/g·°C) | Thermal Conductivity (cal/s·cm·°C) | Viscosity (g/cm·s) |
|---|---|---|---|---|
| 100 | 7.1 | 0.13 | 0.0883 | 0.0207 |
| 500 | 7.0 | 0.14 | 0.0812 | – |
| 1140 | 6.9 | 0.19 | 0.0764 | – |
| 1600 | 6.6 | 0.23 | 0.0620 | – |
Additional parameters included a latent heat of 62.097 cal/g for the alloy. The mold material was ordinary sand, and boundary conditions were set accordingly. The pouring temperature was 1450°C, with a pouring time of 12 seconds and a gravity acceleration of 980 cm/s². Heat transfer coefficients were defined as 0.1 cal/(cm²·s·°C) between metal and mold, and 0.001 cal/(cm²·s·°C) for metal-air and air-mold interfaces. The simulation solver utilized successive over-relaxation (SOR) methods to compute flow and temperature fields over time steps, enabling a detailed analysis of the sand casting process for these complex parts.
The simulation results provided deep insights into the mold filling and solidification behaviors of the sand casting parts. During filling, velocity field analysis revealed that metal entered through the sprue and runner, then rapidly filled the ingates due to their扁平 design. The flow initially concentrated in areas远离 the sprue, with velocities decreasing as metal ascended under gravity. As filling progressed, metal spread into the branching pipes, with slower velocities at curved junctions promoting gas evacuation. By 75% fill, most thin-walled sections were filled, but thicker regions at pipe ends required additional time. This analysis helped predict potential cold shuts or misruns in sand casting parts, which are common defects in complex geometries. The filling pattern can be described by the dimensionless Reynolds number \( Re = \frac{\rho u L}{\mu} \), where \( L \) is a characteristic length. For sand casting parts, maintaining \( Re \) below critical thresholds helps avoid turbulent inclusions. In this case, velocities in thin sections were kept low to ensure smooth filling.
| Parameter | Value | Role in Sand Casting Parts Production |
|---|---|---|
| Pouring Temperature | 1450 °C | Ensures proper fluidity for complex sand casting parts |
| Pouring Time | 12 s | Controls filling speed to reduce turbulence in sand casting parts |
| Mesh Cells | ~12 million | Enables accurate resolution of thin walls in sand casting parts |
| Heat Transfer Coefficient (Metal-Mold) | 0.1 cal/(cm²·s·°C) | Governs cooling rate of sand casting parts |
| Gravity Acceleration | 980 cm/s² | Drives flow in sand casting process |
Solidification analysis through temperature fields was crucial for defect prediction in sand casting parts. The simulation showed that thin sections solidified first, followed by thicker areas, aligning with directional solidification principles. At 15% solid fraction, thin walls in the manifold pipes began to solidify, while risers and gating remained液态. By 35%, most thin walls were solid, but thick pipe ends formed isolated molten pools due to premature solidification of surrounding regions, cutting off feed from risers. This isolation led to shrinkage porosity, a common issue in sand casting parts with varying wall thickness. The temperature gradient \( \nabla T \) in these regions can be expressed as \( \nabla T = \frac{\partial T}{\partial x} \mathbf{i} + \frac{\partial T}{\partial y} \mathbf{j} + \frac{\partial T}{\partial z} \mathbf{k} \), where high gradients exacerbate defect formation. The simulation predicted shrinkage locations precisely, which were later validated experimentally. To quantify solidification progress, the fraction solid \( f_s \) is often used, derived from energy equations: \( f_s = \frac{H – H_l}{H_s – H_l} \), where \( H \) is enthalpy, and subscripts \( l \) and \( s \) denote liquid and solid states. For sand casting parts, controlling \( f_s \) evolution is key to minimizing defects.
Based on these findings, process optimizations were proposed to enhance the quality of sand casting parts. The primary issue was shrinkage in thick pipe ends, caused by unfavorable temperature gradients. To address this, design modifications were implemented, such as adding chills or cooling channels near problematic areas to promote uniform cooling. Additionally, adjustments to riser size and placement were simulated to improve feeding. The gating system was also optimized by modifying ingate dimensions to achieve more balanced filling. These changes aimed to reduce isolated molten pools and ensure better feeding paths for sand casting parts. The optimized process was re-simulated, showing significant reduction in predicted defects. Experimental trials confirmed these results, producing sound castings with minimal shrinkage. This demonstrates the power of numerical simulation in refining sand casting processes for complex components. The overall effectiveness can be assessed using a quality index \( Q \) for sand casting parts, defined as \( Q = 1 – \frac{V_d}{V_t} \), where \( V_d \) is defect volume and \( V_t \) is total casting volume. Through optimization, \( Q \) approached 1, indicating high-quality sand casting parts.
| Defect Type | Simulation Prediction Method | Optimization Strategy for Sand Casting Parts |
|---|---|---|
| Shrinkage Porosity | Temperature field analysis, solidification time | Add chills, optimize riser design, modify wall thickness |
| Cold Shut | Flow velocity analysis, temperature at flow fronts | Increase pouring temperature, adjust gating geometry |
| Slag Inclusion | Flow turbulence tracking, velocity vectors | Use trapezoidal gating, include filters, control pouring speed |
| Gas Entrapment | Air pocket prediction, pressure distribution | Improve venting, reduce flow turbulence in sand casting parts |
In conclusion, numerical simulation proves indispensable for optimizing the sand casting of intricate automotive components like exhaust manifolds. By solving governing equations for fluid flow and heat transfer, simulation software enables detailed analysis of filling and solidification, accurately predicting defects in sand casting parts. This study demonstrated that through模拟-driven design changes—such as gating modifications, riser optimization, and chill application—the formation of shrinkage and other defects can be minimized. The close agreement between simulation predictions and experimental outcomes underscores the reliability of this approach. For industries producing sand casting parts, adopting simulation technologies not only enhances product quality but also reduces development costs and time, fostering more efficient manufacturing processes. Future work could explore advanced turbulence models or multi-scale simulations to further refine predictions for sand casting parts with even greater complexity.
To generalize the findings, the simulation methodology can be extended to other sand casting parts. Key equations remain applicable, but material properties and boundary conditions must be adjusted. For instance, the energy equation can be adapted for different alloys used in sand casting parts by updating \( k \), \( c_p \), and latent heat values. Similarly, the \( k-\epsilon \) model constants may require calibration for specific flow conditions in sand casting processes. By systematically applying these principles, manufacturers can achieve consistent quality across a range of sand casting parts. Ultimately, the integration of simulation into sand casting practice represents a paradigm shift, moving from经验-based tuning to科学-driven optimization, ensuring that sand casting parts meet ever-increasing performance demands in automotive and other industries.