Investment casting, commonly referred to as lost wax investment casting, is a precision manufacturing process ideal for producing complex components with high dimensional accuracy and superior surface finish. In this study, we employ numerical simulation to optimize the lost wax investment casting process for a rotor disc, a critical component in automotive auxiliary braking systems. The rotor disc experiences significant thermal stresses during operation, and defects such as shrinkage porosity can lead to performance issues like warping or reduced braking efficiency. Therefore, ensuring a defect-free casting through process optimization is paramount. The use of ProCAST finite element software allows us to simulate and analyze the filling and solidification behaviors, predict defect locations, and refine the gating system and process parameters to achieve optimal results.
The rotor disc, made of ZG12CrMoV steel, features a complex geometry with two flanges connected by turbine-like structures and evenly spaced fan blades. Its maximum dimensions are 360 mm in diameter and 48 mm in height, with a minimum wall thickness of 6 mm and a mass of 16.5 kg. This intricate design necessitates the use of lost wax investment casting to achieve the required precision and structural integrity. Traditional sand casting methods are inadequate due to the part’s complexity, making lost wax investment casting the preferred approach. However, initial casting trials revealed significant shrinkage defects, prompting this numerical investigation to identify root causes and implement improvements.
We began by designing an initial gating system based on top-gating principles, which included a pouring cup, a horizontal runner, two inner gates, two feeders, and four venting channels. This configuration aimed to facilitate rapid and stable mold filling while providing adequate feeding during solidification. The choice of top-gating was influenced by its potential for directional solidification, but it also risked turbulence and defect formation in critical areas. To evaluate this, we developed a 3D model using Creo 7.0 and imported it into ProCAST for meshing and simulation. The mesh generation in Visual Mesh involved a element size of 3 mm, resulting in 160,878 surface elements and 1,526,452 volume elements, ensuring sufficient resolution for accurate results. The mold shell was set to a thickness of 8 mm, and material properties were assigned accordingly.
| Gating Method | Description | Advantages | Disadvantages |
|---|---|---|---|
| Top Gating | Inner gates located at the top of the casting | Promotes directional solidification, good feeding | High turbulence, potential for oxide formation |
| Bottom Gating | Inner gates at the bottom | Smooth filling, reduced erosion | Poor feeding, longer filling times |
| Middle Gating | Gates positioned at intermediate heights | Balanced filling characteristics | Complex design, risk of cold shuts |
| Step Gating | Multiple gate levels | Controlled sequential filling | Increased complexity and cost |
For the simulation, we defined key parameters to replicate real-world conditions. Gravity was set along the positive Y-axis, and the alloy material was specified as ZG12CrMoV, with its chemical composition detailed in the table below. The mold material was Refractory Mullite, and interfacial heat transfer was modeled with a coefficient of 750 W/(m²·K). The initial pouring temperature was 1610°C, with a mold preheat temperature of 900°C, and a pouring time of 8 seconds, corresponding to a calculated pouring speed of 309 mm/s. These settings align with standard lost wax investment casting practices and ensure realistic simulation outcomes.
| Element | Cr | Mo | C | Mn | Si | V |
|---|---|---|---|---|---|---|
| Content | 2.25 | 0.30 | 0.12 | 0.50 | 0.20 | 0.20 |
| Parameter | Value |
|---|---|
| Gravity Direction | Y-axis positive |
| Alloy Material | ZG12CrMoV |
| Mold Material | Refractory Mullite |
| Pouring Temperature | 1610°C |
| Mold Preheating Temperature | 900°C |
| Interface Heat Transfer Coefficient | 750 W/(m²·K) |
| Pouring Time | 8 s |
| Pouring Speed | 309 mm/s |
| Mesh Size | 3 mm |
The filling simulation for the initial design showed a relatively stable process, with metal entering the horizontal runner at 0.2 seconds, spreading laterally, and beginning to fill the turbine section by 0.5 seconds. By 4.5 seconds, the casting was nearly full, and complete filling was achieved at 8 seconds. No significant issues like air entrapment or misruns were observed, indicating that the gating system provided adequate flow. However, the solidification analysis revealed critical insights. The solidification process followed a pattern from the bottom upwards and from the exterior inwards, with complete solidification occurring at 1999.1 seconds. Early in the solidification, at 49.1 seconds, isolated liquid zones formed in the fan blades and ribs, indicating potential hot spots due to rapid solidification in thinner sections blocking feeding paths. By 89.1 seconds, similar zones appeared in the top flange, suggesting that these areas would be prone to shrinkage defects. The predicted shrinkage porosity volume within the casting was 14.766 cm³, primarily concentrated in the flanges, blades, and ribs, highlighting the need for design modifications.
To address these issues, we revised the gating system by eliminating the central hole in the turbine region (to be machined later) and implementing a combined feeder and venting design. This improved system featured eight feeders on both the top and bottom flanges, connected via subsidies to enhance feeding, and venting channels linked to the pouring cup to facilitate gas escape. The feeders were strategically placed to promote directional solidification and provide sufficient molten metal for补偿. This approach leverages the principles of lost wax investment casting to achieve better thermal management and reduce defect formation. The mesh for the revised model consisted of 2,103,149 elements, and simulation parameters remained unchanged for consistency.
The filling simulation of the improved design demonstrated enhanced stability, with metal first filling the turbine area, then spreading to the ribs and bottom flange by 0.8 seconds, and completing the top flange by 2.7 seconds. Full filling was achieved within 8 seconds, similar to the initial design, but with a more controlled flow pattern. Solidification analysis showed that the venting channels solidified early, around 23.2 seconds, while the flanges solidified inward. By 43.2 seconds, some liquid discontinuity was noted in the blades and ribs, but overall, the solidification sequence was more favorable, following a bottom-up and outside-in pattern. Complete solidification occurred at 1843 seconds, and the shrinkage porosity volume was reduced to 2.295 cm³, a significant improvement. However, residual defects in the blades and ribs indicated the need for further parameter optimization.
We conducted an orthogonal experiment to optimize key process parameters: pouring temperature, pouring speed, and mold preheating temperature. A three-factor, three-level L9(3^3) design was used, with levels specified in the table below. The response variable was the shrinkage porosity rate within the casting, and simulations were run for each combination to identify the optimal settings.
| Level | Pouring Temperature (°C) | Pouring Speed (mm/s) | Mold Preheating Temperature (°C) |
|---|---|---|---|
| 1 | 1610 | 309 | 900 |
| 2 | 1630 | 354 | 930 |
| 3 | 1650 | 413 | 960 |
| Experiment No. | Pouring Temperature | Pouring Speed | Mold Preheating Temperature | Shrinkage Porosity Rate (%) |
|---|---|---|---|---|
| 1 | 1 | 1 | 1 | 5.941 |
| 2 | 1 | 2 | 2 | 4.658 |
| 3 | 1 | 3 | 3 | 3.812 |
| 4 | 2 | 1 | 2 | 3.201 |
| 5 | 2 | 2 | 3 | 3.485 |
| 6 | 2 | 3 | 1 | 4.347 |
| 7 | 3 | 1 | 3 | 0.000 |
| 8 | 3 | 2 | 1 | 3.537 |
| 9 | 3 | 3 | 2 | 3.179 |
Analysis of the orthogonal experiment involved calculating the sum of shrinkage rates for each factor level (K values) and the range (R) to determine the influence of each parameter. The results indicated that pouring temperature had the greatest impact (R = 7.694), followed by mold preheating temperature (R = 6.529) and pouring speed (R = 2.538). The optimal combination was identified as pouring temperature of 1650°C, pouring speed of 309 mm/s, and mold preheating temperature of 960°C, which corresponded to experiment number 7 with zero shrinkage porosity. This combination ensures minimal defects by enhancing fluidity and promoting uniform solidification in the lost wax investment casting process.
To validate the optimization, we simulated the casting process with the optimal parameters. The results confirmed the complete elimination of shrinkage porosity within the casting, with any residual defects confined to the feeders, which would be removed during post-processing. This outcome underscores the effectiveness of combining gating system improvements with parameter optimization in lost wax investment casting. The solidification behavior can be described using energy equations, such as the heat transfer equation during phase change: $$ \frac{\partial (\rho c_p T)}{\partial t} = \nabla \cdot (k \nabla T) + L \frac{\partial f_s}{\partial t} $$ where $\rho$ is density, $c_p$ is specific heat, $k$ is thermal conductivity, $T$ is temperature, $L$ is latent heat, and $f_s$ is solid fraction. This model helps in predicting temperature gradients and solidification patterns, crucial for avoiding defects in complex castings like the rotor disc.
In conclusion, this study demonstrates the successful application of numerical simulation in optimizing the lost wax investment casting process for a rotor disc. The initial design revealed significant shrinkage defects, which were mitigated through gating system modifications and parameter optimization via orthogonal experiments. The final process parameters—pouring temperature of 1650°C, pouring speed of 309 mm/s, and mold preheating temperature of 960°C—resulted in a defect-free casting, highlighting the importance of integrated design and simulation in enhancing the quality and reliability of lost wax investment casting components. This approach can be extended to other complex castings, providing a framework for reducing defects and improving manufacturing efficiency in precision casting applications.