In the field of engineering machinery, exhaust braking systems play a critical role in auxiliary braking by controlling vehicle emissions, thereby protecting service brakes and extending their lifespan. The exhaust brake valve, characterized by its simple structure and cost-effectiveness, consists of an actuator assembly, connection components, and a butterfly valve assembly. Among these, the butterfly valve is a key part typically manufactured through casting processes, where casting quality directly influences the performance of the exhaust brake valve. Lost foam casting (EPC), also known as expendable pattern casting, offers significant advantages over traditional gravity sand casting, including higher dimensional accuracy, smoother surface finishes, reduced environmental pollution, and greater flexibility in process design. This makes EPC particularly suitable for high-demand components like the brake butterfly valve, which requires mass production. In this study, we employ lost foam casting combined with numerical simulation using ProCAST software to analyze and optimize the casting process for a butterfly valve. By examining the filling and solidification stages, including fluid flow, temperature distribution, solid fraction, and defect formation, we aim to predict and mitigate defects, thereby shortening development cycles and reducing production costs.
The butterfly valve casting, designed with UG software, has overall dimensions of 181.25 mm × 171.30 mm × 184.00 mm and is made of HT250 gray iron, with a total mass of 5.81 kg and an average wall thickness of 8.5 mm. The valve features uniform wall thickness except at the flanges and protruding sections. For the lost foam casting process, the three-dimensional model was imported into MeshCAST for meshing and repair, resulting in approximately 800,000 volume elements and 45,628 surface elements to ensure computational accuracy. The material properties and process parameters were defined based on ProCAST’s internal database and experimental data. The casting material is HT250 gray iron, with chemical composition provided in Table 1. The mold material is permeable sand foam, and the initial temperature of the pouring system is set at 1,450°C, while the pattern, gating system, and sand box are at room temperature (25°C). Thermal physical parameters for the foam pattern and sand mold are summarized in Tables 2 and 3, respectively. The thermal properties of HT250, including enthalpy and conductivity, are calculated and displayed graphically, illustrating key behaviors such as latent heat release during solidification.
| Element | C | Cr | Mn | S | Si | P | Mg |
|---|---|---|---|---|---|---|---|
| Content | 3.240 | 0.430 | 0.810 | 0.009 | 1.660 | 0.014 | 0.005 |
| Parameter | Density (kg/m³) | Thermal Conductivity (W/(m·K)) | Specific Heat (kJ/(kg·K)) | Latent Heat (kJ/kg) | Liquidus Temperature (°C) | Solidus Temperature (°C) |
|---|---|---|---|---|---|---|
| Value | 25 | 0.15 | 3.7 | 100 | 350 | 300 |
| Parameter | Density (kg/m³) | Thermal Conductivity (W/(m·K)) | Specific Heat (kJ/(kg·K)) | Permeability (m²) |
|---|---|---|---|---|
| Value | 1,520 | 0.53 | 1.22 | 1×10⁻⁷ |
The governing equations for the lost foam casting process involve fluid flow, heat transfer, and phase change. The continuity and momentum equations for incompressible flow are given by:
$$\nabla \cdot \mathbf{v} = 0$$
$$\rho \left( \frac{\partial \mathbf{v}}{\partial t} + \mathbf{v} \cdot \nabla \mathbf{v} \right) = -\nabla p + \mu \nabla^2 \mathbf{v} + \rho \mathbf{g}$$
where $\mathbf{v}$ is the velocity vector, $\rho$ is density, $p$ is pressure, $\mu$ is dynamic viscosity, and $\mathbf{g}$ is gravitational acceleration. The energy equation accounting for phase change is:
$$\rho c_p \frac{\partial T}{\partial t} + \rho c_p \mathbf{v} \cdot \nabla T = \nabla \cdot (k \nabla T) + Q$$
where $T$ is temperature, $c_p$ is specific heat, $k$ is thermal conductivity, and $Q$ represents the latent heat source term due to solidification. The solid fraction $f_s$ is modeled using a lever rule or similar method, and defects like shrinkage porosity are predicted based on criteria such as the Niyama criterion, expressed as:
$$N_y = \frac{G}{\sqrt{\dot{T}}}$$
where $G$ is the temperature gradient and $\dot{T}$ is the cooling rate. Values below a threshold indicate potential shrinkage defects.
In the initial lost foam casting simulation, the filling process was analyzed to assess temperature distribution and potential defects. The results showed that the sprue was fully filled at 5.6 seconds with a fill percentage of 20%, and the total mold filling time was 43.1 seconds. The metal front advanced steadily without significant turbulence, minimizing risks of gas entrapment and slag inclusion. The temperature field during filling indicated a layered progression, with the metal maintaining adequate superheat. The voids analysis, which tracks foam degradation, revealed complete pattern decomposition without residual foam, ensuring smooth gas evolution and reducing the likelihood of surface defects. This is critical in lost foam casting, as incomplete decomposition can lead to voids or folds.
The solidification process was examined to identify isolated liquid regions prone to shrinkage defects. The sequence showed that the upper protrusions solidified first, followed by the thin-walled sections, and finally the distal flanges. By 70% solidification, isolated liquid pools formed in the flange areas, as illustrated by the solid fraction distribution. These regions, devoid of feeding paths, are hotspots for shrinkage porosity and cavities. The bottom-gating system contributed to thermal segregation, with early solidification of the gating and riser systems blocking compensation channels and increasing internal pressure. The defect prediction module in ProCAST highlighted significant shrinkage volumes of approximately 6.45 cm³, concentrated in the flanges, aligning with the solidification analysis. This underscores the importance of optimizing the gating and riser design in lost foam casting to ensure directional solidification and effective feeding.
To address these issues, the lost foam casting process was optimized by shortening the ingate length to reduce gas back-pressure and improve metal flow. Additionally, side risers were incorporated near the distant flanges to enhance feeding, and chills were placed beneath the protrusions to accelerate cooling and reduce hot spots. A rectangular riser was also added near the top protrusion to facilitate overall compensation. The modified gating system, with revised dimensions, aimed to promote sequential solidification toward the risers. The thermal properties and boundary conditions remained unchanged, but the geometry adjustments were simulated to evaluate their impact.
The optimized filling process demonstrated a reduced total filling time of 48.974 seconds, with metal reaching the pattern earlier at 10.51 seconds, compared to 14.13 seconds initially. This decrease alleviated gas-related resistance, as evidenced by the steady, layered filling and complete foam decomposition. The temperature field during filling showed improved thermal uniformity, reducing the risk of cold shuts and misruns. The solidification analysis confirmed that the side risers remained liquid longest, acting as effective feeding sources, and the chills helped dissipate heat efficiently. The solid fraction plots indicated a more directional solidification pattern, minimizing isolated liquid zones. Defect prediction post-optimization showed a substantial reduction in shrinkage volume to 3.62 cm³, representing a decrease of over 42%. The defects were redistributed to less critical areas, validating the effectiveness of the risers and chills in the lost foam casting process.
The success of this optimization highlights the value of numerical simulation in lost foam casting for complex components like the brake butterfly valve. By integrating ProCAST analysis, we achieved a robust process design that minimizes defects and enhances production efficiency. Future work could explore advanced materials or real-time control systems to further refine the EPC process. In conclusion, lost foam casting, combined with predictive modeling, offers a reliable pathway for high-quality casting manufacturing, with broad applications in automotive and machinery sectors.
Further mathematical modeling can enhance the understanding of lost foam casting dynamics. For instance, the foam degradation kinetics can be described by an Arrhenius-type equation:
$$\frac{dm}{dt} = -A e^{-E_a / (RT)}$$
where $m$ is the foam mass, $A$ is the pre-exponential factor, $E_a$ is the activation energy, $R$ is the gas constant, and $T$ is temperature. This relates to the gas evolution during EPC, affecting the metal-foam interface. Additionally, the feeding efficiency during solidification can be quantified using the feeding capacity equation:
$$V_f = \int_{t_1}^{t_2} A_r v_r \, dt$$
where $V_f$ is the fed volume, $A_r$ is the riser cross-sectional area, and $v_r$ is the feeding velocity. Optimizing these parameters in lost foam casting ensures adequate compensation for shrinkage. Overall, the iterative simulation approach enables precise control over the EPC process, leading to superior castings with minimal trial-and-error.