The casting of complex, safety-critical components like brake calipers presents significant manufacturing challenges, particularly when using materials with specific solidification characteristics such as ductile iron. The inherent volumetric shrinkage during the solidification of ductile iron castings makes them highly susceptible to internal defects like shrinkage porosity and cavities, which can severely compromise mechanical integrity. This study focuses on the process optimization for a QT600-3 caliper body. Leveraging numerical simulation as a core investigative tool, we systematically analyze and refine key casting parameters to minimize defect formation. The methodology involves an initial process design, defect prediction via simulation, followed by a structured orthogonal experimental design to quantify the influence of critical process variables. The ultimate goal is to derive a robust and optimized casting process that ensures high-quality ductile iron castings with minimal internal defects.
1. Component Characteristics and Initial Process Design
The subject of this study is a brake caliper body, a critical structural component subjected to high dynamic loads. Its complex geometry, characterized by uneven wall thicknesses and an internal cavity, inherently creates thermal gradients and isolated hot spots during solidification, making it prone to shrinkage defects. The material is grade QT600-3 ductile iron, with a nominal chemical composition and key thermal properties as summarized below.
| Element | C | Si | Mn | P | S | Mg | Cu |
|---|---|---|---|---|---|---|---|
| Wt. % | 3.0-3.8 | 2.4-2.8 | 0.3-0.5 | <0.1 | 0.03-0.035 | 0.045-0.05 | 0.35-0.40 |
The critical thermal parameters are a liquidus temperature ($T_L$) of 1166°C and a solidus temperature ($T_S$) of 1143°C. The wide solidification range ($T_L – T_S$) is a primary contributor to the mushy zone formation and subsequent feeding difficulties in ductile iron castings.
The initial casting process was designed based on principles for heavy-section ductile iron castings. A horizontally parted sand mold using furan resin-bonded sand was selected. To improve yield, a two-cavity mold layout was adopted. The gating system was designed as semi-pressurized to ensure calm filling. Given the component’s geometry, risers were strategically placed on the thickest sections of the casting to provide adequate feed metal. Furthermore, external chills were designed and placed at specific locations to promote directional solidification towards the risers. This initial setup served as the baseline for subsequent simulation and optimization.
2. Numerical Simulation Methodology and Setup
The finite element-based simulation software ProCAST was employed to model the filling and solidification processes. This approach allows for the virtual analysis of temperature fields, solidification sequences, and defect prediction without the cost and time associated with physical trial runs.
The 3D CAD models of the castings, gating, risering, and mold were imported and discretized into a finite element mesh. A critical aspect of accurate simulation is the mesh density. A finer mesh was applied to the casting and gating system (element size: 5 mm) to capture the detailed thermal history and potential defect formation accurately. The mold and cores were meshed more coarsely (element size: 15 mm) to optimize computational efficiency. The final mesh consisted of approximately 2.49 million volume elements.
The boundary conditions and material properties were rigorously defined. The mold and core material was set as furan sand with appropriate thermal properties. The ductile iron castings material was defined using the QT600-3 database within ProCAST. The interfacial heat transfer coefficient (HTC) between the metal and the mold was set to a constant value of 500 W/(m²·K). The initial simulation parameters for the baseline process were a pouring temperature of 1400°C and a pouring rate of 8.5 kg/s. Gravity was accounted for in the positive Z-direction.
The fundamental governing equations solved during the simulation include the conservation of mass, momentum, and energy. The energy equation, crucial for solidification analysis, is given by:
$$ \rho c_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \rho L \frac{\partial f_s}{\partial t} $$
where $\rho$ is density, $c_p$ is specific heat, $T$ is temperature, $t$ is time, $k$ is thermal conductivity, $L$ is latent heat of fusion, and $f_s$ is the solid fraction. The simulation solves these equations to predict the evolution of the solid fraction field, from which defect criteria like the Niyama criterion can be evaluated to predict shrinkage porosity. The Niyama criterion ($Ny$) is often expressed as:
$$ Ny = \frac{G}{\sqrt{\dot{T}}} $$
where $G$ is the temperature gradient and $\dot{T}$ is the cooling rate at the end of solidification. Regions with a Niyama value below a critical threshold are predicted to be susceptible to microporosity.
3. Baseline Process Simulation and Defect Analysis
The simulation of the initial process design confirmed a smooth, progressive filling pattern without turbulent splashing or air entrapment. The temperature field during filling showed a logical thermal gradient. Analysis of the solidification sequence, visualized through the solid fraction, revealed that the designed risers solidified last, confirming their intended function as feeders. However, pronounced hot spots were identified at the junctions of ribs and side walls on the caliper body, indicating these areas were potential defect sites.
The porosity prediction module of ProCAST was then applied. The results quantified the total predicted porosity percentage for the casting and localized the areas at highest risk. The initial process yielded a predicted porosity percentage of 6.908%. As anticipated from the solid fraction analysis, the majority of this porosity was concentrated in the thermal centers at the rib-wall junctions. While the risers effectively prevented macro-shrinkage in the main thick sections, these isolated hot spots in thinner yet complex geometries were insufficiently fed. This baseline result clearly identified the need for process optimization to improve the soundness of these ductile iron castings.
4. Orthogonal Experimental Design for Process Optimization
To systematically improve the casting quality, an orthogonal experimental design was employed. This method is highly efficient for studying the effects of multiple factors with a minimized number of experimental runs. Three key process parameters were selected as factors: Pouring Temperature (A), Pouring Speed (B), and Number of Ingates (C). Each factor was assigned three levels, as detailed in the table below.
| Level | A: Pouring Temp. (°C) | B: Pouring Speed (kg/s) | C: No. of Ingates |
|---|---|---|---|
| 1 | 1380 | 7.5 | 2 |
| 2 | 1400 | 8.5 | 3 |
| 3 | 1420 | 9.5 | 4 |
A standard L9(3^4) orthogonal array was chosen, requiring only 9 simulation runs to evaluate the full combination of three factors at three levels. The response variable, or the metric for evaluation, was the Predicted Porosity Percentage of the casting obtained from each simulation run. The design matrix and the simulation results are presented in the following table.
| Run No. | A (°C) | B (kg/s) | C | Porosity (%) | |
|---|---|---|---|---|---|
| L1 | 1380 | 7.5 | 3 | 8.588 | |
| L2 | 1380 | 8.5 | 4 | 6.564 | |
| L3 | 1380 | 9.5 | 2 | 8.335 | |
| L4 | 1400 | 7.5 | 2 | 3.151 | |
| L5 | 1400 | 8.5 | 3 | 6.908 | |
| L6 | 1400 | 9.5 | 4 | 2.945 | |
| L7 | 1420 | 7.5 | 4 | 3.913 | /tr> |
| L8 | 1420 | 8.5 | 2 | 3.812 | |
| L9 | 1420 | 9.5 | 3 | 3.110 |
5. Analysis of Orthogonal Simulation Results
The results from the orthogonal simulation array were analyzed using range analysis (also known as range method or R-method) to determine the primary and secondary order of the influencing factors and to identify the optimal level combination.
First, the average porosity value (K) for each factor at each level was calculated. For Factor A (Pouring Temperature):
$$ K_{A1} = (8.588 + 6.564 + 8.335) / 3 = 7.829 $$
$$ K_{A2} = (3.151 + 6.908 + 2.945) / 3 = 4.335 $$
$$ K_{A3} = (3.913 + 3.812 + 3.110) / 3 = 3.612 $$
Similarly, the averages for Factors B and C were computed. The range (R) for each factor, which is the difference between the maximum and minimum K values for that factor, was then determined. A larger range indicates a greater influence of that factor on the porosity result.
| Factor | Level 1 Avg. (K1) | Level 2 Avg. (K2) | Level 3 Avg. (K3) | Range (R) |
|---|---|---|---|---|
| A: Temp. | 7.829 | 4.335 | 3.612 | 4.217 |
| B: Speed | 5.217 | 5.761 | 4.797 | 0.964 |
| C: Ingates | 5.099 | 6.202 | 4.474 | 1.728 |
The analysis of the ranges clearly shows that Pouring Temperature (A) has the most significant effect on the porosity of these ductile iron castings, with a range of 4.217. The Number of Ingates (C) has the second-largest influence (R=1.728), followed by Pouring Speed (B), which shows a relatively minor effect (R=0.964). The optimal level for each factor, based on minimizing porosity, is the level with the smallest K value: A3 (1420°C), B3 (9.5 kg/s), and C3 (4 ingates). Therefore, the theoretically optimal parameter combination from the orthogonal analysis is A3B3C3.
However, practical foundry knowledge for ductile iron castings must be integrated. A pouring temperature of 1420°C, while optimal in the simulation, approaches the upper limit and may promote issues like dross formation, excessive mold reaction, and degraded graphite morphology in practice. Examining the individual run results, combination L6 (A2B3C3: 1400°C, 9.5 kg/s, 4 ingates) yielded the absolute lowest porosity value of 2.945%. This represents a 57.4% reduction compared to the baseline porosity of 6.908%. Therefore, a modified optimal combination of A2B3C3 (1400°C, 9.5 kg/s, 4 ingates) was selected as the final optimized process. This balances superior simulated performance with practical manufacturability for high-integrity ductile iron castings.
6. Validation of the Optimized Casting Process
The selected optimized parameters were used in a final, comprehensive simulation to validate the overall casting process performance. The results confirmed excellent filling behavior with the four-ingate system, distributing metal more evenly. The solidification sequence showed an even more pronounced directional solidification towards the risers.
Most importantly, the porosity prediction for the optimized process confirmed a drastic improvement. The predicted shrinkage porosity was concentrated almost exclusively within the riser heads, which is their intended function, and was virtually eliminated from the critical rib-wall junctions on the caliper body. The total predicted porosity percentage was confirmed at 2.945%, validating the effectiveness of the optimization. The improvement can be attributed to the combined effect of the parameters: a sufficiently high pouring temperature (1400°C) maintains fluidity for feeding, a faster pouring speed (9.5 kg/s) reduces heat loss in the gating system, and an increased number of ingates (4) promotes a more uniform temperature distribution during filling, reducing the severity of isolated hot spots in these complex ductile iron castings.
The relationship between pouring temperature and solidification feeding can be conceptually framed. The pressure head available for feeding a mushy zone is related to the metallostatic pressure and the fraction of solid. A modified feeding equation highlights the importance of temperature-dependent fluidity:
$$ P_{feed} = \rho g h – \frac{\mu}{K(f_s)} v_{feed} $$
Where $P_{feed}$ is the feeding pressure, $\rho$ is density, $g$ is gravity, $h$ is metal height, $\mu$ is viscosity, $K$ is permeability (a strong function of solid fraction $f_s$), and $v_{feed}$ is the feeding flow velocity. Higher superheat (from a higher pouring temperature) delays the formation of a coherent solid network (reduces $f_s$ for a given time/temperature), increasing permeability $K$ and reducing the flow resistance during the critical feeding stage.
7. Conclusion
This study demonstrates a systematic methodology for optimizing the sand casting process of complex ductile iron castings. By integrating numerical simulation with orthogonal experimental design, the effects of key process parameters—pouring temperature, pouring speed, and number of ingates—were quantitatively evaluated with high efficiency. For the QT600-3 caliper body, the analysis revealed that pouring temperature is the most influential factor on internal shrinkage porosity. An optimized parameter set of 1400°C pouring temperature, 9.5 kg/s pouring speed, and a four-ingate system was identified and validated. This optimized process reduced the simulated porosity by over 57% compared to the initial baseline design. The approach provides a robust, data-driven framework for improving the soundness and reliability of ductile iron castings, reducing the need for costly physical trials and minimizing the risk of defect-related failures in critical applications. Future work could involve refining riser and chill design based on the optimized thermal conditions and conducting physical verification casts to correlate simulation predictions with actual microstructure and mechanical properties.