In the field of industrial automation, components like lifting arms play a critical role in mimicking human arm motions for tasks such as grasping, lifting, and transporting objects. The reliability and precision of these parts are paramount, and their manufacturing often relies on advanced casting techniques. Among these, the investment casting process stands out for its ability to produce complex, near-net-shape parts with excellent surface finish and dimensional accuracy. However, the intricate geometry of components like lifting arms can lead to defects such as shrinkage porosity and cavities during solidification. This study focuses on the design and optimization of the investment casting process for a lifting arm, utilizing numerical simulation and experimental methods to enhance quality and efficiency. We will delve into the structural analysis, initial process design, simulation-based evaluation, and systematic optimization of both the gating system and key process parameters. The goal is to minimize defects and establish a robust framework for similar castings.
The lifting arm under consideration is made of ZG0Cr18Ni9Ti, a stainless acid-resistant cast steel widely used in demanding applications due to its corrosion resistance and mechanical strength. Its chemical composition is detailed in Table 1, which ensures the material meets the necessary performance criteria for industrial use. The casting has a relatively uniform wall thickness of approximately 6 mm, but its geometry is complex, featuring a stepped shaft section, a through-hole connection part, and a U-shaped claw working area. The overall dimensions are about 70 mm × 45 mm × 57 mm. Such complexity necessitates meticulous process design to avoid defects like pores, shrinkage, and cold shuts, which could compromise structural integrity. The three-dimensional model of the lifting arm, created using CAD software, highlights these features and serves as the basis for subsequent simulation and optimization.
| Element | Content |
|---|---|
| C | ≤0.8 |
| Mn | 0.8–2.0 |
| Si | ≤1.5 |
| P | ≤0.045 |
| S | ≤0.030 |
| Cr | 17.0–20.0 |
| Ni | 8.0–11.0 |
| Ti | 0.3–0.7 |
The investment casting process begins with the design of the gating system, which is crucial for controlling metal flow, feeding, and solidification. For the lifting arm, a horizontal sprue-vertical runner-in-gate system was initially adopted to ensure relatively stable filling and directional solidification. The pattern assembly was arranged in a cluster of six parts per mold to improve productivity. The three-dimensional model of the gating system was developed based on the casting geometry, and finite element meshing was performed with a mesh size of 4 mm, resulting in 34,928 nodes and 249,485 elements. This model forms the foundation for numerical simulation using ProCAST software, a powerful tool for analyzing fluid flow, heat transfer, and defect formation in casting processes.
In any investment casting process, accurate parameter setting is essential for reliable simulation results. For this study, the key parameters included heat transfer coefficients, boundary conditions, and material properties. The shell mold, composed of quartz sand refractory material and silica sol, had a thickness of 6 mm. The initial shell preheating temperature was set at 850°C, but this was varied later during optimization. The pouring temperature for ZG0Cr18Ni9Ti ranges from 1,500°C to 1,550°C, with a liquidus temperature of 1,344.7°C and a solidus temperature of 1,235.7°C. Gravity acceleration was set to 9.8 m/s² in the negative Y-direction to align with the pouring orientation. The heat transfer coefficient between the casting and the environment was set to 1,000 W/(m²·K), and that between the casting and the shell was 500 W/(m²·K). These parameters ensure that the simulation closely replicates actual foundry conditions.
To determine the pouring speed for the initial investment casting process, the Kalakin formula was employed, which relates pouring speed to casting height, wall thickness, and pouring temperature. The formula is expressed as:
$$v = k \sqrt{h \cdot \delta \cdot (T – T_s)}$$
where \( v \) is the pouring speed (cm/s), \( h \) is the casting height (cm), \( \delta \) is the wall thickness (cm), \( T \) is the pouring temperature (°C), \( T_s \) is the solidus temperature (°C), and \( k \) is an empirical constant. For this casting, with \( h = 7.0 \, \text{cm} \), \( \delta = 0.6 \, \text{cm} \), \( T = 1,530°C \), and \( T_s = 1,235.7°C \), the calculated pouring speed is approximately 344.6 mm/s. Considering practical factors, an initial pouring speed of 350 mm/s was used for simulation. This speed aims to balance filling time and turbulence minimization, which are critical in the investment casting process.
The simulation of the initial investment casting process revealed the filling and solidification behavior. The filling process was generally stable, with molten metal completing the sprue and entering the first set of castings within 1.0 s, and the entire mold filled by about 3.9 s. Temperature distribution during filling showed no excessive冲击, indicating that the gating design avoided severe turbulence. However, the solidification analysis indicated that the casting began to solidify at the outer surfaces within 5.6 s, and by 374 s, the main body was nearly fully solid. The total solidification and cooling time to 1,235°C was around 778 s. The directional solidification sequence, from the casting extremities toward the feeders, was generally favorable, but defect prediction highlighted issues.
Shrinkage porosity and cavities are common concerns in the investment casting process, especially for complex geometries. The initial simulation results, as shown in the defect distribution map, indicated significant shrinkage in the vertical runner and at the junction between the shaft and the connection part of the lifting arm. The porosity volume fraction exceeded 1.0 in these regions, suggesting inadequate feeding during solidification. This aligns with typical defects observed in actual production, validating the simulation accuracy. The root cause was identified as premature solidification at certain junctions, blocking feeding channels and leading to shrinkage. This necessitated a redesign of the gating system to improve feeding efficiency.
To optimize the investment casting process, two modified gating systems, denoted as Scheme A and Scheme B, were proposed. Scheme A retained a single vertical runner but added排气口 at locations prone to shrinkage and connected adjacent castings at the U-claw areas to reduce thermal isolation. Scheme B further enhanced Scheme A by introducing additional in-gates at the top of the castings and enlarging the runner diameters to improve metal flow and feeding. Both schemes aimed to promote better temperature distribution and sequential solidification. The finite element models for these schemes were simulated under the same parameters as the initial process. The results showed that Scheme B significantly reduced shrinkage porosity compared to Scheme A and the initial design. The total porosity volume fraction dropped to 2.92 in Scheme B, versus over 10 in Scheme A, demonstrating the effectiveness of added in-gates and enlarged runners in the investment casting process.
Beyond gating design, process parameters play a vital role in the quality of the investment casting process. Key parameters include pouring temperature, pouring speed, and shell preheating temperature. To systematically optimize these, an orthogonal experimental design was employed. This method allows for evaluating multiple factors with a reduced number of trials. Three levels were selected for each factor based on practical ranges: pouring temperature (1,500°C, 1,530°C, 1,550°C), pouring speed (350 mm/s, 450 mm/s, 550 mm/s), and shell preheating temperature (1,000°C, 1,050°C, 1,100°C). The orthogonal array L9(3^4) was used, resulting in nine experimental combinations, as detailed in Table 2.
| Level | A: Pouring Temperature (°C) | B: Pouring Speed (mm/s) | C: Shell Preheating Temperature (°C) |
|---|---|---|---|
| 1 | 1,500 | 350 | 1,000 |
| 2 | 1,530 | 450 | 1,050 |
| 3 | 1,550 | 550 | 1,100 |
For each combination, simulations were conducted using the optimized gating system (Scheme B), and the results were analyzed based on total porosity volume fraction (shrinkage rate) and filling time. The experimental outcomes are summarized in Table 3. The shrinkage rate serves as the primary quality indicator, with lower values denoting better casting integrity. Filling time is also recorded as it relates to production efficiency.
| Experiment No. | A: Pouring Temperature | B: Pouring Speed | C: Shell Preheating Temperature | Filling Time (s) | Shrinkage Rate (%) |
|---|---|---|---|---|---|
| L1 | 1,500 | 350 | 1,000 | 3.8078 | 1.2453 |
| L2 | 1,500 | 450 | 1,050 | 3.8431 | 1.1471 |
| L3 | 1,500 | 550 | 1,100 | 4.0174 | 1.3162 |
| L4 | 1,530 | 350 | 1,050 | 3.9403 | 1.2634 |
| L5 | 1,530 | 450 | 1,100 | 3.7968 | 1.2852 |
| L6 | 1,530 | 550 | 1,000 | 4.0297 | 1.2307 |
| L7 | 1,550 | 350 | 1,100 | 3.8659 | 1.3176 |
| L8 | 1,550 | 450 | 1,000 | 3.9273 | 1.2759 |
| L9 | 1,550 | 550 | 1,050 | 3.9167 | 1.2831 |
Analysis of the orthogonal experiment involved range analysis to determine the optimal level for each factor. The mean shrinkage rate for each level was calculated. For factor A (pouring temperature), the means were: level 1 (1,500°C) = (1.2453 + 1.1471 + 1.3162)/3 = 1.2362%, level 2 (1,530°C) = 1.2598%, level 3 (1,550°C) = 1.2922%. Thus, lower pouring temperature reduces shrinkage, as higher temperatures increase liquid contraction. For factor B (pouring speed), the means: level 1 (350 mm/s) = 1.2754%, level 2 (450 mm/s) = 1.2361%, level 3 (550 mm/s) = 1.2767%. The intermediate speed of 450 mm/s yields the lowest shrinkage, balancing flow dynamics and feeding. For factor C (shell preheating temperature), the means: level 1 (1,000°C) = 1.2506%, level 2 (1,050°C) = 1.2312%, level 3 (1,100°C) = 1.3063%. A preheating temperature of 1,050°C is optimal, as lower temperatures may cause premature solidification, while higher ones reduce thermal gradients. Consequently, the optimal parameter combination is A1B2C2: pouring temperature 1,500°C, pouring speed 450 mm/s, and shell preheating temperature 1,050°C.
To validate the optimized investment casting process, a final simulation was run with the optimal gating system (Scheme B) and the optimal parameters. The results showed a significant reduction in defects, with shrinkage porosity virtually eliminated from the casting body and confined to the main runner, where it does not affect part functionality. The shrinkage rate dropped to 1.1471%, and the filling time was 3.8431 s, indicating improved quality and efficiency. This confirms that the synergistic optimization of gating design and process parameters is crucial for enhancing the investment casting process. The methodology demonstrates how numerical simulation can guide practical improvements, reducing trial-and-error in foundry operations.
In conclusion, this study successfully addresses the challenges in the investment casting process for a complex lifting arm component. Through numerical simulation using ProCAST, the initial process was analyzed, revealing shrinkage defects due to inadequate feeding. The gating system was optimized by adding in-gates and modifying runners, which improved thermal management and feeding. Furthermore, an orthogonal experiment identified the optimal process parameters: pouring temperature of 1,500°C, pouring speed of 450 mm/s, and shell preheating temperature of 1,050°C. These optimizations collectively minimized shrinkage porosity and enhanced production efficiency. The findings provide a valuable reference for similar castings in the investment casting process, underscoring the importance of integrated design and parameter optimization in achieving high-quality cast components. Future work could explore additional factors such as cooling rate control or alternative alloy compositions to further refine the process.