In modern industrial applications, lost wax investment casting stands as a pivotal near-net-shape manufacturing technique, enabling the production of intricate metal components with complex internal geometries, high melting temperatures, tight dimensional tolerances, minimal machining requirements, and low surface roughness. This process, often referred to as lost wax investment casting, is extensively utilized in sectors such as aerospace, automotive, and marine engineering due to its ability to fabricate small-scale metal castings efficiently. However, the practical implementation of lost wax investment casting involves a multi-step procedure that is inherently complex, time-consuming, and challenging to control, traditionally relying on extensive trial-and-error experiments to achieve optimal results. With the ongoing advancement of industrial technologies, cast components are increasingly trending toward larger dimensions, more complex structures, and thinner walls, making the prediction and mitigation of defects during mold filling and solidification a critical hurdle in lost wax investment casting. The rapid evolution of numerical simulation technologies has provided a powerful tool for investigating these processes, now widely adopted to guide production practices. For instance, researchers have employed software like MAGMA to simulate filling and solidification sequences, predict defect locations, and analyze root causes, leading to significant improvements in casting yield—from 80% to 90% in some cases. Similarly, finite element methods have been applied to forecast shrinkage and macrosegregation in steel ingots, while other studies have optimized gating systems and process parameters for high-temperature alloy components, validating the feasibility of simulation-driven approaches. In this study, we leverage MAGMA numerical simulation software to model and refine the lost wax investment casting process for a bypass valve casting, analyzing fluid flow during mold filling to anticipate defects such as porosity and cold shuts caused by air entrainment and discontinuous flow. Based on these insights, we redesign the pouring cup placement, simulate the optimized filling and solidification stages, and identify potential risks for shrinkage defects. Our results demonstrate that the revised process ensures stable metal flow, promotes directional solidification, eliminates isolated liquid zones, and minimizes shrinkage risks, with practical measures like insulation wraps applied during trial production to address residual concerns. Through X-ray inspection and dissection, the produced castings show no defects, and subsequent chemical composition and mechanical property tests confirm compliance with technical requirements, validating the rationality of the optimized lost wax investment casting methodology.
The bypass valve casting under investigation features a curved pipe structure with overall dimensions of 120 mm × 132 mm × 253 mm, wall thicknesses ranging from 10 mm to 25 mm, and internal cavity diameters between φ51 mm and φ60 mm, weighing approximately 6.7 kg. Manufactured using the lost wax investment casting technique, the component must be free from defects like shrinkage cavities, slag inclusions, and cracks. From a structural perspective, the casting exhibits isolated geometric hot spots and an extended “L”-shaped section with uniform wall thickness, which poses challenges for feeding during solidification. To address these issues in the lost wax investment casting process, we strategically positioned inner gates at hot spots and along the “L”-segment to facilitate adequate feeding and minimize defect formation. The initial gating system design included a horizontal runner measuring 30 mm × 60 mm × 255 mm, two vertical runners (1#: 30 mm × 30 mm × 120 mm; 2#: 40 mm × 40 mm × 120 mm), and multiple inner gates (1#: 15 mm × 25 mm × 20 mm; 2#: φ60 mm × 20 mm semi-cylindrical contour; 3#: two symmetrical gates with 11 mm thickness, 15 mm height, and 55 mm arc length on the ring section; 4#: 25 mm × 50 mm × 40 mm contour on the outer ring wall). A pouring cup with a top diameter of φ90 mm, bottom diameter of φ50 mm, and height of 90 mm was mounted on the horizontal runner, connected via a φ10 mm wax guide rod to aid in dewaxing and prevent shell cracking. Crucially, gate 2# maintained a 30 mm gap from the side flange to avoid shell blockage and localized overheating during the lost wax investment casting process. We developed the three-dimensional model using UG software and performed simulations with MAGMA, configuring parameters such as a total mesh count of 5 million, gravity pouring for lost wax investment casting, material 20Mn5M for the casting, zircon sand for the shell, shell temperature of 950°C, heat transfer coefficient of 500 W/(m²·K), pouring temperature of 1560°C, pouring speed of 3 kg/s, and air cooling. The filling simulation revealed significant issues: at 10% fill, metal entered the cavity primarily through gates 1#, 2#, and 4#, with minimal, intermittent flow through gate 3# and vertical runner 1#, leading to turbulent flow and potential cold shuts at specific locations. Maximum flow velocities reached approximately 1.5 m/s, causing chaotic metal movement due to conflicting streams from different gates. By 30% fill, most of the cavity was filled, but metal backflow into gate 3# indicated poor design, and at 46% fill, the cavity was complete, yet the overall process was deemed unstable for lost wax investment casting, heightening risks of defects.
To rectify the filling instability, we optimized the lost wax investment casting process by repositioning the pouring cup directly above inner gate 2# on the horizontal runner. This modification ensured that metal entered the cavity primarily through gate 2#, reducing flow conflicts and promoting laminar flow. Subsequent filling simulations confirmed the improvement: at 15% fill, metal flowed smoothly from the pouring cup through gate 2# into the cavity, with a maximum velocity of about 1.8 m/s and no entry from other gates. By 30% fill, the metal level rose steadily, and at 60% fill, a sectional view showed velocity gradients from approximately 1.8 m/s in the upper ring to 0.6 m/s at the bottom, indicating controlled flow. At 85% fill, the entire casting and gating system were fully filled without turbulence, demonstrating the efficacy of the optimized lost wax investment casting setup. For solidification analysis, we monitored the liquid fraction to identify isolated liquid zones that could lead to shrinkage porosity. At 64% solidification, no isolated liquid areas were observed near gates 3# and 4#, confirming proper gate placement in the lost wax investment casting design. By 77% solidification, the casting had solidified except below gate 2#, with liquid continuously retreating into the gates, ensuring directional solidification. At 98% solidification, liquid was confined to gate 2#, and the absence of isolated zones validated the sequential solidification pattern. Shrinkage prediction analysis indicated that major defects were localized in the pouring cup and runners, while minor shrinkage risks occurred in region “G” of the casting, adjacent to the inner gates. To mitigate this in the lost wax investment casting process, we applied insulation wraps to the runners and gates during trial production to enhance feeding and eliminate residual shrinkage.
In the production validation phase, we employed a full silica sol shell-making process for the lost wax investment casting, using GRJ-30 silica sol compliant with HB5346-1986 standards. The face coat consisted of silica sol-zircon flour (320 mesh, slurry ratio 1:3.55), with intermediate and backup coats of silica sol-shangdian flour (200 mesh). Additives like defoamer and wetting agent were incorporated into the face coat, and slurry viscosity was measured with a Zahn No. 4 cup. The shell-building involved 7.5 layers: two face coats, one intermediate coat, three backup coats, and a seal coat, resulting in a shell thickness of 7–8 mm. Drying times were set at 6–8 hours for face coats, ≥10 hours for intermediate, ≥12 hours for backup, and ≥24 hours for the seal coat, under controlled conditions of 23–25°C temperature and 50–70% humidity. Critical areas such as corners and deep holes were manually brushed to prevent casting fins, and the shell was moistened with 30% SiO₂ silica sol before intermediate coating to avoid face coat detachment. After dewaxing, the shell was dried for 24 hours and fired at 950–1000°C for over 30 minutes to remove residues, achieving a white or rosy color indicative of proper preparation. Prior to pouring, the shell interior was rinsed repeatedly with water, dried at 400°C for ≥1 hour, and insulation wraps (20 mm thick) were applied to runners and gates. The shell was preheated to 1050°C for ≥1 hour, and melting was conducted in a 200 kg medium-frequency furnace using certified raw materials, with adjustments for carbon, silicon, and manganese using low-carbon ferromanganese and ferrosilicon. Pouring temperature was maintained at 1560±10°C, and pre-pouring infrared measurements confirmed an average shell temperature of 946°C, aligning with simulation inputs. Post-casting, the components underwent normalizing heat treatment at 890±10°C with an 80°C/h heating rate, 2.5-hour hold, and air cooling, followed by mechanical testing. Chemical composition analysis via SPECTROM12 spark spectrometry met all requirements, as summarized in Table 1, and mechanical properties, including tensile and impact tests, exceeded specifications, with room-temperature tensile strength of 559 MPa, yield strength of 385 MPa, elongation of 31.5%, reduction in area of 54%, high-temperature tensile strength of 524 MPa at 300°C, and impact energies of 58.0 J, 67.1 J, and 56.4 J at 0°C, all satisfying technical standards for the lost wax investment casting process.
| Element | Required Content | Measured Content |
|---|---|---|
| C | ≤0.25 | 0.17 |
| Si | ≤0.60 | 0.56 |
| Mn | ≤1.20 | 1.03 |
| P | ≤0.020 | 0.016 |
| S | ≤0.020 | 0.004 |
Visual inspection of the castings after gate removal, cleaning, and shot blasting revealed no cracks, shrinkage, sand inclusions, or other defects. Surface roughness, evaluated against GB6060.5-88 standards, measured ≤6.3 μm, well below the required 12.5 μm for the lost wax investment casting. Although internal non-destructive testing was not mandated, we performed X-ray inspection on all four trial castings, which showed no shrinkage, porosity, slag, or cracks, and dissection of one sample confirmed the absence of visible defects in critical areas and thickness transitions. These results affirm the robustness of the optimized lost wax investment casting process for mass production. The integration of numerical simulation in lost wax investment casting not only predicts potential defects but also guides process refinements, as demonstrated by the stable filling and directional solidification achieved here. In summary, the lost wax investment casting approach, enhanced with MAGMA simulations, proves highly effective for producing high-integrity bypass valve castings.
To further elucidate the fluid dynamics in lost wax investment casting, we can model the metal flow during filling using the Navier-Stokes equations, which describe the motion of viscous fluids. The general form for incompressible flow is given by:
$$ \rho \left( \frac{\partial \mathbf{v}}{\partial t} + \mathbf{v} \cdot \nabla \mathbf{v} \right) = -\nabla p + \mu \nabla^2 \mathbf{v} + \mathbf{f} $$
where \( \rho \) is the density of the metal, \( \mathbf{v} \) is the velocity vector, \( t \) is time, \( p \) is pressure, \( \mu \) is the dynamic viscosity, and \( \mathbf{f} \) represents body forces such as gravity. In the context of lost wax investment casting, this equation helps simulate the flow behavior through complex gating systems, with boundary conditions set by the shell geometry and pouring parameters. For solidification analysis, the heat transfer equation governs the temperature distribution:
$$ \frac{\partial T}{\partial t} = \alpha \nabla^2 T $$
where \( T \) is temperature, \( t \) is time, and \( \alpha \) is the thermal diffusivity, defined as \( \alpha = \frac{k}{\rho c_p} \), with \( k \) being the thermal conductivity, \( \rho \) the density, and \( c_p \) the specific heat capacity. This equation, coupled with phase change models, predicts the formation of isolated liquid zones and shrinkage defects in lost wax investment casting. The optimization of process parameters, such as pouring temperature and shell preheat, can be summarized using empirical relationships; for instance, the solidification time \( t_s \) for a casting can be approximated by Chvorinov’s rule:
$$ t_s = B \left( \frac{V}{A} \right)^n $$
where \( V \) is volume, \( A \) is surface area, \( B \) is a mold constant, and \( n \) is an exponent typically around 2. In our lost wax investment casting study, adjusting the gating design based on these principles minimized \( t_s \) in critical regions, reducing defect risks. The mechanical properties of the final casting, validated through tests, relate to the microstructure influenced by the lost wax investment casting process; for example, yield strength \( \sigma_y \) can be expressed via the Hall-Petch equation for grain size strengthening:
$$ \sigma_y = \sigma_0 + \frac{k_y}{\sqrt{d}} $$
where \( \sigma_0 \) is the friction stress, \( k_y \) is the strengthening coefficient, and \( d \) is the average grain diameter. The fine microstructure achieved in the lost wax investment casting contributed to the high mechanical performance observed. Overall, the iterative simulation and validation cycle in lost wax investment casting underscores the importance of integrating numerical tools with practical expertise to achieve optimal outcomes in industrial applications.
| Test Type | Temperature | Property | Required Value | Measured Value |
|---|---|---|---|---|
| Tensile | Room Temperature | Rp0.2 (MPa) | ≥275 | 385 |
| Rm (MPa) | ≥485 | 559 | ||
| A (%) | ≥20 | 31.5 | ||
| Z (%) | ≥35 | 54 | ||
| Tensile | 300°C | Rp0.2 (MPa) | ≥210 | 238 |
| Rm (MPa) | ≥435 | 524 | ||
| Impact | 0°C | KV2 (J) | ≥40 | 58.0, 67.1, 56.4 |
In conclusion, the application of MAGMA numerical simulation in the lost wax investment casting process for the bypass valve casting enabled precise prediction and mitigation of defects through systematic analysis of filling and solidification. By optimizing the pouring cup location, we achieved stable metal flow and sequential solidification, eliminating isolated liquid zones and minimizing shrinkage risks in the lost wax investment casting. Practical measures, such as insulation wraps, further enhanced the process during trial production. The resulting castings, free from defects and compliant with all technical specifications, demonstrate the effectiveness of simulation-driven optimization in lost wax investment casting. This approach not only reduces reliance on traditional trial-and-error methods but also paves the way for efficient, high-quality production in complex casting applications, reaffirming the value of lost wax investment casting as a advanced manufacturing solution.