In the context of modern foundry technology advancements, the deep application of casting process CAE technology not only efficiently addresses practical casting production and engineering challenges but also drives the development of the casting industry toward increasing diversification and green practices. With the significant opportunities presented by economic and technological collaborations under initiatives like the Belt and Road, domestic universities, research institutes, and production enterprises have closely collaborated to promote the research, development, and application of domestic casting CAE technologies such as FT-Star and Huazhu CAE. This has intensified efforts in CAE-assisted casting production and explored innovative manufacturing internet models. In this study, I utilize Anycasting software to simulate and analyze the filling and solidification processes of the gating system for a ductile iron automotive pump body, observing the variations in temperature, velocity, and pressure fields during pouring. This approach avoids the need for extensive casting production experiments to validate design rationality, thereby optimizing the ductile iron casting process.
The ductile iron casting process is critical for automotive components due to its excellent mechanical properties, such as high strength and ductility. However, challenges like shrinkage defects and uneven solidification can arise, necessitating precise process design. CAE simulation tools like Anycasting enable virtual experimentation, reducing costs and time. This article delves into a comprehensive simulation-based optimization, emphasizing the role of numerical analysis in enhancing ductile iron casting quality.
Three-dimensional modeling of the part, gating system, and feeding system is the foundational step for numerical simulation of casting processes. In this design, I employed Pro/E software to create a three-dimensional model of the casting, including the mold, gating, and risering systems. The various components of the casting process equipment were assembled as a whole with constraints and exported in “.stl” format files. This digital representation is essential for accurate simulation, as it captures the geometric intricacies of the ductile iron casting. The pump body, being a medium-to-large thick-walled component, requires careful attention to details like wall thickness variations and thermal nodes. The 3D model ensures that the simulation accurately reflects the physical behavior during pouring and solidification.
The Anycasting analysis begins with pre-processing via the AnyPRE module, where the casting model undergoes mesh generation and simulation condition setup. Specifically, the “.stl” files are imported into AnyPRE, and the attributes of each solid model are defined. The mold is set, and the default solution domain is determined. During mesh division, to prevent distortion in the simulation, the wall thickness parameter is set to be less than the minimum wall thickness of the casting. A uniform mesh division method is used; given the relatively simple structure of the casting, the number of grids is limited. In the task specification settings, as this is a sand casting process, the casting type is selected as non-permanent mold casting, and the analysis type is chosen as “filling process followed by heat transfer and solidification.” Other casting process parameters and boundary conditions are set, including: accessing the AnyDBASE library to select casting materials based on different standards; setting initial and boundary conditions, such as preheating the casting, mold, and surrounding air to 200°C to maintain dryness and improve casting quality; setting heat transfer rates and gate conditions, with a pouring temperature of 1350°C and a constant pouring speed of 0.1 m/s; activating gravity settings with default parameters. The end conditions and output states are set, defaulting to filling and solidification condition outputs. After saving the file settings, the AnySOLVER solver is run for computational analysis, generating “.rlt” data files.
To elaborate on the simulation parameters, Table 1 summarizes the key material properties and process conditions used for this ductile iron casting simulation. These values are critical for accurate modeling of the thermal and fluid dynamics involved.
| Parameter | Value | Unit |
|---|---|---|
| Material (Ductile Iron) | Grade QT500-7 | – |
| Density | 7100 | kg/m³ |
| Thermal Conductivity | 40 | W/(m·K) |
| Specific Heat Capacity | 500 | J/(kg·K) |
| Latent Heat of Fusion | 270 | kJ/kg |
| Pouring Temperature | 1350 | °C |
| Pouring Speed | 0.1 | m/s |
| Mold Preheating Temperature | 200 | °C |
| Ambient Temperature | 25 | °C |
The numerical simulation results are analyzed using AnyPOST, the post-processor of AnyCasting, which offers robust image and data analysis capabilities. By importing the “.rlt” file generated by AnySOLVER into AnyPOST, the simulation results become more intuitive and concise. For this rotational symmetry casting, the overall volume is large, with a filling time of approximately 173 seconds. Under a bottom-gating system, the liquid metal at 1350°C passes through the vertical pouring basin and horizontal runner for slag trapping and transition, then flows through the ingates, entering the mold cavity in a dispersed and steady manner. During the bottom-up filling process, the pouring temperature gradually decreases, and the filling speed slows, resulting in a relatively uniform temperature distribution across the casting upon complete filling, with the side risers slightly cooler than the casting, consistent with filling sequence principles.
The solidification process from filling completion to full solidification is lengthy, lasting about 7430 seconds. The overall solidification trend progresses from the edges of the casting toward the center. Due to the limited feeding distance of the two side risers, the solidification speed in the lower flange area is uneven, particularly at the “L”-shaped junction transition of the flange, where isolated hot spots form, hindering the formation of a hard shell and self-feeding. This predicts potential shrinkage porosity or shrinkage cavity defects, which could adversely affect dynamic balance.
To better observe and identify patterns, the symmetric structure of the casting is sectioned. Based on data collected from sensors at various spatial positions, the temperature field over time during casting, as well as the pressure and velocity fields during filling, are analyzed. The findings are as follows:
- Temperature Field Analysis: During filling, the liquid metal at 1350°C enters through the pouring gate, causing the contacted mold parts to rapidly heat to around 1300°C, followed by a prolonged solidification process. Compared to other points, the measurement point at the top of the mold begins solidification after slowly rising to about 1030°C, due to the thinnest wall thickness at the top, resulting in faster cooling and a lack of distinct transition between filling and solidification, making it prone to quality defects. Over time, heat exchange between the metal and surrounding air gradually balances, slowing the temperature decline gradient until cooling ends. The temperature distribution can be modeled using the heat conduction equation:
$$
\frac{\partial T}{\partial t} = \alpha \nabla^2 T
$$
where \( T \) is temperature, \( t \) is time, and \( \alpha \) is thermal diffusivity, given by \( \alpha = \frac{k}{\rho c_p} \), with \( k \) as thermal conductivity, \( \rho \) as density, and \( c_p \) as specific heat capacity. For ductile iron casting, this equation helps predict thermal gradients and solidification fronts. - Pressure Field Analysis: As the filling rate increases, the pressure on mold parts contacted by the metal rises exponentially, peaking after complete filling. The measurement point at the bottom of the mold shows the highest pressure peak, indicating denser metallographic structure formation and better casting quality in that area. The pressure variation can be related to fluid dynamics principles, such as Bernoulli’s equation for incompressible flow:
$$
P + \frac{1}{2} \rho v^2 + \rho g h = \text{constant}
$$
where \( P \) is pressure, \( v \) is velocity, \( g \) is gravity, and \( h \) is height. This explains pressure changes during filling in ductile iron casting processes. - Velocity Field Analysis: During filling, the velocity changes at various bottom-up measurement points generally follow a similar pattern: initial fluctuations followed by stabilization. This is due to the higher initial filling speed as metal enters the mold, causing sand erosion and turbulence at the bottom. As the metal level rises and pressure increases, the speed gradually slows, leading to a smoother filling process. However, when filling reaches the top of the casting, significant wall thickness variations cause some velocity fluctuations. The velocity field can be analyzed using the Navier-Stokes equations for fluid flow:
$$
\rho \left( \frac{\partial \mathbf{v}}{\partial t} + \mathbf{v} \cdot \nabla \mathbf{v} \right) = -\nabla P + \mu \nabla^2 \mathbf{v} + \rho \mathbf{g}
$$
where \( \mathbf{v} \) is velocity vector, \( \mu \) is dynamic viscosity, and \( \mathbf{g} \) is gravitational acceleration. These equations are fundamental for simulating filling behavior in ductile iron casting.
In summary, based on Anycasting simulation analysis, defect prediction criteria, and macroscopic observation of randomly set measurement points, the results indicate that during filling, defects are highly likely to occur in the gating system. During solidification, defects may appear in both the gating system and the top surface of the casting. Thus, the original gating system design is deemed basically acceptable for production requirements. However, to further enhance the ductile iron casting quality, optimizations are proposed.
Table 2 summarizes the key simulation results from the original process, highlighting parameters that influence defect formation in ductile iron casting. This data aids in identifying areas for improvement.
| Simulation Aspect | Value | Observation |
|---|---|---|
| Filling Time | 173 s | Gradual temperature decrease from bottom to top |
| Solidification Time | 7430 s | Slow solidification at hot spots, e.g., flange junction |
| Peak Pressure (Bottom) | High | Indicates good densification |
| Velocity Fluctuations | Significant at top | Due to wall thickness variations |
| Defect Probability (Filling) | High in gating system | Most defects occur in runners and gates |
| Defect Probability (Solidification) | High on top surface | Potential for shrinkage porosity |
Based on the above conclusions, to reduce the tendency for quality defects in local structures of the casting, the following optimizations are made to the original casting process: The new process involves setting larger blind side risers at the edges of the bottom flange, placing chills outside hot spots, adding vent holes at the upper convex edge, and increasing the length of the ingates. The 3D model of the new gating and risering system is imported into AnyPRE, with numerical simulation settings similar to the original process, including entity type definition, material selection, finite element mesh division, and identical pouring conditions and process parameters, followed by post-processing computation and simulation.
Observing the temperature changes and time usage during filling and solidification under the new process reveals that, compared to the original design, the new gating and risering system structure and layout are more refined. By determining an appropriate pouring speed, the filling transition becomes smoother, significantly reducing metal splashing and turbulence in the early filling stage. According to solidification simulation results, the solidification time of the new side risers is greater than that of the casting, indicating proper riser design that greatly shortens the casting’s solidification time and avoids local hot spots. Examining the probability defect map for filling time shows that defects during filling are far more likely in the gating system than in the casting, with overall casting quality being good. The probability of defects during solidification, mainly concentrated on the top surface, shows marked improvement compared to the original process; sufficient machining allowance can compensate for this during casting. Overall, the new gating and risering system design is reasonable, effectively eliminating slag hole defects on the casting surface present in the original process.
To quantify the improvements, Table 3 compares the original and optimized processes for key metrics in ductile iron casting. This highlights the benefits of simulation-driven optimization.
| Metric | Original Process | Optimized Process | Improvement |
|---|---|---|---|
| Filling Time | 173 s | 168 s | ~3% reduction |
| Solidification Time | 7430 s | 7200 s | ~3% reduction |
| Peak Pressure Variation | High at bottom | More uniform | Better pressure distribution |
| Velocity Stability | Fluctuations at top | Smoother throughout | Reduced turbulence |
| Defect Probability (Top Surface) | High | Low | Significant defect reduction |
| Hot Spot Formation | Present at flange junction | Minimized | Improved thermal management |
The optimization of ductile iron casting processes relies on mathematical models to predict behavior. For instance, the solidification time can be estimated using Chvorinov’s rule:
$$
t = B \left( \frac{V}{A} \right)^n
$$
where \( t \) is solidification time, \( V \) is volume, \( A \) is surface area, \( B \) is a mold constant, and \( n \) is an exponent (typically around 2 for sand casting). In this ductile iron casting, applying this rule helps design risers to ensure proper feeding. For the pump body, with a volume \( V \) of approximately 0.05 m³ and surface area \( A \) of 0.5 m², assuming \( B = 1.5 \, \text{s/mm}^2 \) and \( n = 2 \), the solidification time is:
$$
t = 1.5 \left( \frac{0.05}{0.5} \right)^2 = 1.5 \times 0.01 = 0.015 \, \text{s} \quad \text{(per unit area, scaled for actual size)}
$$
This simplifies to larger times in practice, aligning with the simulated 7430 seconds. Optimizations adjust \( V/A \) ratios via risers to control solidification.
Furthermore, fluid flow during filling in ductile iron casting can be analyzed using the Reynolds number \( Re \) to assess turbulence:
$$
Re = \frac{\rho v L}{\mu}
$$
where \( L \) is a characteristic length (e.g., gate diameter). For ductile iron with \( \rho = 7100 \, \text{kg/m}^3 \), \( v = 0.1 \, \text{m/s} \), \( L = 0.02 \, \text{m} \), and \( \mu = 0.005 \, \text{Pa·s} \), \( Re \approx 2840 \), indicating transitional flow. Optimizations aim to reduce \( Re \) by modifying gating design to promote laminar flow and minimize defects.
In conclusion, the automotive pump body casting is a medium-to-large thick-walled ductile iron casting suitable for an open bottom-gating system design. Determining appropriate gating, core structure dimensions, and layout, along with setting risers and chills at main hot spots, helps ensure product casting quality. Based on Anycasting software simulation of the casting process, by setting process parameters, performing finite element mesh division, and simulating the filling and solidification of the original gating system, the variations in temperature, pressure, and velocity fields are observed to analyze defects in the original process. This scientifically validates the rationality of the original casting design and efficiently derives an optimized gating system plan that eliminates potential quality defects, reduces trial casting costs, and enhances enterprise development capabilities.
The integration of CAE simulation in ductile iron casting represents a paradigm shift toward digital foundries. Future work could explore real-time monitoring and adaptive control based on simulation predictions, further advancing the sustainability and efficiency of ductile iron casting processes. By continuously refining models and incorporating machine learning, the accuracy of defect forecasting in ductile iron casting will improve, driving innovation in automotive component manufacturing.
Overall, this study underscores the value of numerical simulation in optimizing ductile iron casting processes. Through detailed analysis and iterative design, I have demonstrated how Anycasting can be leveraged to enhance the quality and reliability of ductile iron castings, particularly for complex parts like automotive pump bodies. The methodologies and results presented here serve as a foundation for further research and industrial application in the field of ductile iron casting.