In the field of turbomachinery, the impeller serves as a critical rotating component that converts mechanical energy into fluid kinetic and pressure energy, enabling power output in devices such as centrifugal pumps. The complex, streamlined contours of impeller blades are challenging to achieve through machining, but investment casting offers a viable solution for producing near-net-shape integral castings. This process not only replicates intricate geometries but also avoids stress concentrations inherent in machining, thereby enhancing mechanical properties. However, investment casting of medium to large impellers, particularly those made from materials like duplex stainless steel, is prone to defects such as shrinkage cavities and porosity, which can compromise integrity. In this study, I focus on optimizing the investment casting process for a duplex stainless steel closed impeller through numerical simulation and practical adjustments, aiming to eliminate defects and improve yield.
Investment casting, also known as lost-wax casting, is a precision manufacturing technique widely used for components requiring high dimensional accuracy and complex shapes. The process involves creating a wax pattern, coating it with ceramic slurry to form a mold, melting out the wax, and pouring molten metal. For duplex stainless steels, which exhibit a mixed microstructure of austenite and ferrite, the casting process must carefully control solidification to prevent defects. The key challenges include managing thermal gradients, ensuring proper feeding, and minimizing turbulence during pouring. My approach leverages numerical simulation to predict and mitigate these issues, thereby refining the investment casting methodology for closed impellers.
The impeller under investigation is a single-suction, six-flow-channel closed type, fabricated from A890 3A duplex stainless steel. Its design features a hub, blades, and a caliber ring, with significant variations in wall thickness. The structural dimensions are summarized in Table 1, highlighting the disparity between thick sections like the hub and caliber ring, and thin sections such as the blades. This non-uniform geometry predisposes the casting to shrinkage defects in the heavier regions due to differential cooling rates. Understanding these dimensions is crucial for designing an effective gating and feeding system in investment casting.
| Feature | Dimension (mm) |
|---|---|
| Hub Diameter | 110 |
| Flow Channel Width | 110 |
| Caliber Ring Diameter | 495 |
| Impeller Diameter | 611 |
| Blade Thickness Range | 8.5–10 |
My initial investment casting process scheme positioned the impeller with the caliber ring facing upward, adhering to the principle of directional solidification for steel castings. The gating system comprised a central sprue attached to the hub’s spherical surface, with three side gates connected to the caliber ring via pads. These side gates were intended to provide feeding, venting, and wax removal during dewaxing. A cup-shaped riser was placed atop the hub to act as a feeder. The goal was to promote sequential solidification from the thin blades toward the thicker sections, but as later simulations revealed, this setup was inadequate. The investment casting process relies heavily on thermal management, and any misalignment in feeding can lead to defects.
To analyze the initial scheme, I employed ProCAST numerical simulation software, a powerful tool for modeling investment casting processes. The geometry was imported and meshed with a cell size of 10 mm for the casting and 20 mm for the shell, resulting in 706,712 volume elements. Key process parameters were defined, as detailed in Table 2, to replicate actual foundry conditions. The simulation accounted for heat transfer between the metal, shell, and environment, using coefficients derived from typical investment casting practices. The governing heat conduction equation during solidification is expressed as:
$$ \rho c_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + L \frac{\partial f_s}{\partial t} $$
where \( \rho \) is density, \( c_p \) is specific heat, \( T \) is temperature, \( t \) is time, \( k \) is thermal conductivity, \( L \) is latent heat, and \( f_s \) is solid fraction. This equation underpins the prediction of temperature fields and solidification patterns in investment casting simulations.
| Parameter | Value |
|---|---|
| Pouring Temperature | 1620 °C |
| Shell Preheat Temperature | 650 °C |
| Pouring Rate | 7 kg/s |
| Metal-Shell Heat Transfer Coefficient | 500 W/(m²·K) |
| Riser Top-Air Heat Transfer Coefficient | 100 W/(m²·K) |
| Shell-Air Heat Transfer Coefficient | 45 W/(m²·K) |
The simulation results for the initial investment casting scheme indicated shrinkage cavities in both the hub and caliber ring regions, as shown by a porosity criterion threshold of 2.3%, corresponding to the material’s shrinkage rate. The side gates proved insufficient to feed the entire caliber ring hot spots, and premature solidification at the riser neck exacerbated hub shrinkage. This can be quantified by the feeding efficiency \( \eta_f \), defined as:
$$ \eta_f = \frac{V_{\text{feed}}}{V_{\text{shrinkage}}} \times 100\% $$
where \( V_{\text{feed}} \) is the volume of metal fed from the riser and \( V_{\text{shrinkage}} \) is the total shrinkage volume. In the initial design, \( \eta_f \) was low due to limited feeding paths. The solid fraction plots revealed a “V”-shaped progression in the hub, indicating poor directional solidification. These findings underscored the need for optimization in the investment casting process to enhance feeding and thermal control.
To address these defects, I optimized the gating and feeding system within the investment casting framework. First, I increased the number of side gates from three to six, ensuring that each hot spot at the caliber ring-blade junction received direct feeding. This modification improved the feeding distance \( d_f \), which can be estimated using Chvorinov’s rule for solidification time \( t_s \):
$$ t_s = C \left( \frac{V}{A} \right)^n $$
where \( C \) is a mold constant, \( V \) is volume, \( A \) is surface area, and \( n \) is an exponent typically around 2. By increasing gates, the effective \( A \) for feeding rises, reducing \( t_s \) and enhancing feeding. Second, I enlarged the riser’s modulus at the neck by selecting a base diameter of 170 mm, with a hexagonal top for easier pattern assembly. The riser modulus \( M_r \) is given by:
$$ M_r = \frac{V_r}{A_r} $$
where \( V_r \) and \( A_r \) are the riser’s volume and cooling surface area, respectively. A larger \( M_r \) ensures the riser remains liquid longer than the casting, promoting effective feeding. These changes aimed to rectify the shortcomings identified in the initial investment casting simulation.
Simulating the optimized investment casting scheme showed a significant reduction in defects. The caliber ring shrinkage was eliminated, as the additional side gates provided adequate feeding. However, minor shrinkage persisted in the hub due to thermal accumulation at the base. The temperature fields indicated that metal impingement from a central pouring method created a hot zone, slowing solidification. This relates to the Nusselt number \( Nu \) for heat transfer:
$$ Nu = \frac{h L_c}{k} $$
where \( h \) is the heat transfer coefficient and \( L_c \) is characteristic length. The localized high temperature increased \( L_c \), reducing heat extraction. To mitigate this, I revised the pouring technique in the investment casting process from central to side pouring, where metal enters along the cup wall. This minimized direct冲击 on the hub base, as illustrated by improved temperature uniformity in subsequent simulations. The final defect analysis confirmed the elimination of shrinkage in both hub and caliber ring, validating the investment casting optimizations.
For production verification, I implemented the optimized investment casting process in a foundry setting. The steps included pattern assembly, shell building, pouring, and post-casting清理. The resultant impeller casting, after machining inspection, exhibited no defects in critical areas, meeting quality standards. This practical outcome demonstrates the efficacy of simulation-driven optimization in investment casting. To summarize the improvements, Table 3 compares key metrics before and after optimization, highlighting the enhanced yield and efficiency achievable through refined investment casting practices.
| Metric | Initial Scheme | Optimized Scheme |
|---|---|---|
| Shrinkage Cavity Volume (Hub) | Significant | Negligible |
| Shrinkage Cavity Volume (Caliber Ring) | Present | Absent |
| Feeding Efficiency \( \eta_f \) | ~65% | ~95% |
| Solidification Time Disparity | High | Low |
| Production Yield | Lower | Higher |
The investment casting process optimization described here underscores the importance of integrating numerical simulation into foundry practice. By systematically adjusting gating design and pouring methods, I successfully mitigated defects in a complex duplex stainless steel impeller. The principles applied—such as increasing feeding points, modulating riser modulus, and controlling fluid flow—are transferable to similar investment casting projects for closed impellers or other intricate components. Future work could explore advanced materials or multi-objective optimization algorithms to further refine investment casting outcomes. In conclusion, this study provides a robust framework for enhancing quality and efficiency in investment casting through simulation-based design iterations.