The relentless pursuit of performance and efficiency in sectors such as aerospace, defense, and advanced optics continuously drives the demand for components that are lighter, stronger, and more complex. Titanium alloys, celebrated for their exceptional strength-to-weight ratio, superior corrosion resistance, and biocompatibility, are at the forefront of this evolution. A significant trend within titanium component manufacturing is the shift towards thin-walled, intricate geometries, which are essential for achieving weight reduction without compromising structural integrity. Producing these delicate yet high-performance shell castings presents formidable challenges, primarily concerning dimensional accuracy, surface finish, and the integrity of thin sections. This work details a comprehensive study and successful implementation of a novel composite mold casting technology specifically engineered for the production of high-quality, thin-walled titanium alloy shell castings.
The primary objective in casting thin-walled titanium structures is to achieve complete mold filling while simultaneously controlling solidification to prevent defects like cold shuts, misruns, porosity, and surface irregularities. Traditional monolithic mold approaches often fall short. For instance, while graphite molds offer excellent thermal conductivity promoting directional solidification, they can be costly for complex cores and may lead to excessive chilling on thin sections, causing cracks. Conversely, ceramic cores used in investment casting provide superb surface finish and detail but are characterized by long lead times, high cost, and difficult removal. This study addresses these limitations by proposing and validating a hybrid approach: a graphite mold exterior combined with a precision-engineered expendable sand core. This composite mold strategy synergizes the benefits of both materials. The graphite mold provides the necessary structural rigidity and thermal management, while the specialized sand core, coated with a refractory layer, allows for the creation of complex internal passages with good collapsibility, acceptable surface finish, and significantly reduced cost and lead time compared to ceramic alternatives. The focus of this investigation is a representative thin-walled cylindrical shell casting with internal ribs, serving as a benchmark for the process.
Structural Analysis and Technical Specifications of the Shell Casting
The subject component is a cylindrical shell casting made from ZTA15 (Ti-6Al-2V-1.5Mo-2Zr) titanium alloy. Its key structural characteristics and the associated technical demands define the complexity of the casting task.
The geometry is fundamentally a thin-walled cylinder. The primary challenge lies in its wall thickness distribution. The nominal wall thickness is minimal, and the structure features an array of internal reinforcing ribs. These ribs are crucial for maintaining stiffness but are themselves extremely thin, acting as potential barriers to metal flow and increasing the risk of mistun. The internal cavity of the shell casting is non-axisymmetric, further complicating core design and metal flow dynamics. The specification requires that all internal surfaces, which form the intricate cavity, be “as-cast” finish, meaning no post-cast machining is permitted. This demands that the casting process itself delivers exceptional dimensional accuracy and surface quality. The dimensional tolerances must conform to CT6 level as per GB/T 6414-1999, and surface roughness must be controlled within a low micrometer range. The chemical composition and mechanical property requirements for the ZTA15 alloy are stringent, as detailed in the tables below.
| Element | Al | V | Mo | Zr | Ti | Fe ≤ | Si ≤ | C ≤ | N ≤ | H ≤ | O ≤ | Other (each) ≤ | Other (total) ≤ |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Content | 5.5-6.8 | 0.8-2.5 | 0.5-2.0 | 1.5-2.5 | Bal. | 0.30 | 0.15 | 0.13 | 0.05 | 0.01 | 0.16 | 0.10 | 0.30 |
| Condition | Tensile Strength (Rm), MPa ≥ | Yield Strength (Rp0.2), MPa ≥ | Elongation (A), % ≥ | Reduction of Area (Z), % ≥ |
|---|---|---|---|---|
| HIPed | 885 | 785 | 5 | 12 |
Casting Process Design and Rationale for the Composite Mold
The selection of the overall casting process is critical. Vacuum arc skull melting combined with centrifugal casting was chosen. This combination is ideal for reactive metals like titanium, as it operates under high vacuum to prevent contamination, and centrifugal force aids in filling thin sections and reducing gas porosity. The composite mold concept is the cornerstone of this research. The external mold is machined from high-purity, fine-grained graphite. Graphite provides excellent thermal conductivity, which can be leveraged to promote directional solidification. Its machinability allows for high precision in the external dimensions of the shell castings. However, machining complex internal cavities directly into graphite is impractical and wasteful.
This is where the sacrificial sand core is introduced. The core is fabricated using a high-alumina sand blended with a silica sol binder system. This formulation provides adequate green strength for handling and precision, while maintaining excellent collapsibility after casting to allow for easy core removal from the intricate internal passages of the shell casting. To prevent metal-core reaction and improve surface finish, the cured sand core is coated with a slurry of yttria (Y2O3), a highly refractory and thermodynamically stable oxide relative to molten titanium. This coating acts as a barrier, insulating the sand from the melt and resulting in a much smoother as-cast surface on the internal features of the shell casting.
The total linear contraction for the shell casting must be accurately predicted to scale the pattern dimensions. Based on empirical data and the geometry, a contraction allowance ranging from 0.5% to 1.5% was applied, varying across different sections of the part. Furthermore, machining allowances were added to critical external surfaces. The thermal interaction between the graphite mold and the sand core is a key consideration. The heat flux, \( q \), across the interface can be described by Fourier’s law, considering the different thermal properties:
$$ q = -k_{\text{eff}} \frac{dT}{dx} $$
where \( k_{\text{eff}} \) is an effective thermal conductivity influenced by both the graphite mold and the sand core with its Y2O3 coating. The coating’s lower conductivity relative to graphite helps moderate the cooling rate on the internal surfaces of the shell casting, reducing thermal stress and the risk of hot tearing in thin sections.
Gating System Design and Numerical Simulation with ProCAST
A well-designed gating system is paramount for the success of thin-wall casting. The objectives are to: 1) achieve rapid, uniform, and turbulence-free filling to avoid cold shuts, 2) establish a favorable temperature gradient for directional solidification towards the feeders, and 3) minimize volume shrinkage porosity. For this centrifugal casting process, a bottom-gating system was designed. The system features a central downsprue that branches into two horizontal runners positioned at the bottom of the mold cavity. From these runners, multiple in-gates rise vertically to feed the shell casting at its lower circumference. This design promotes upward filling against centrifugal force, helping to vent gases ahead of the metal front and reducing the velocity of metal entry into the thin sections.
To optimize this design before costly physical trials, numerical simulation using ProCAST software was employed. A three-dimensional model of the shell casting, gating system, graphite mold, and sand core was created. The governing equations for fluid flow, heat transfer, and solidification were solved. The Navier-Stokes equations for incompressible flow with a free surface (Volume of Fluid method) were used for filling:
$$ \frac{\partial \vec{v}}{\partial t} + (\vec{v} \cdot \nabla) \vec{v} = -\frac{1}{\rho} \nabla p + \nu \nabla^2 \vec{v} + \vec{g} + \vec{F}_{\text{cent}} $$
where \( \vec{v} \) is velocity, \( p \) is pressure, \( \rho \) is density, \( \nu \) is kinematic viscosity, \( \vec{g} \) is gravity, and \( \vec{F}_{\text{cent}} \) is the centrifugal force term. The energy equation incorporating latent heat release was solved for solidification:
$$ \rho C_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) – \rho L \frac{\partial f_s}{\partial t} $$
where \( C_p \) is specific heat, \( k \) is thermal conductivity, \( L \) is latent heat, and \( f_s \) is solid fraction.
Initial simulations of the solidification sequence for the shell casting alone identified potential hot spots at junctions where internal ribs met thicker mounting bosses. These areas were prone to macro-porosity, as predicted by the Niyama criterion, a widely used index for predicting shrinkage porosity:
$$ N_y = \frac{G}{\sqrt{\dot{T}}} $$
where \( G \) is the temperature gradient and \( \dot{T} \) is the cooling rate. Regions with a low Niyama value are susceptible to shrinkage defects. The designed gating system was then simulated. The filling sequence showed a controlled, bottom-up progression with minimal free surface turbulence. The solidification simulation, incorporating the composite mold materials, demonstrated that the thermal mass of the graphite mold successfully pulled heat away from the external walls, while the insulating sand core slowed cooling on the interior. This created a desired solidification front moving from the thin walls towards the heavier sections fed by the in-gates, effectively eliminating the isolated hot spots. The final predicted shrinkage was limited to the feeder heads, confirming the gating design’s efficacy for the shell casting.
| Component | Material | Thermal Conductivity (W/m·K) | Specific Heat (J/kg·K) | Initial Temperature (°C) |
|---|---|---|---|---|
| Shell Casting | ZTA15 (Ti-6Al-2V-1.5Mo-2Zr) | ~7 (liquid), ~17 (solid) | ~550 | 1750 (Pour) |
| Mold Body | High-Purity Graphite | ~100 | ~710 | 300 |
| Core (Coated) | Alumina Sand / Y2O3 | ~1.5 | ~1050 | 300 |
Mold and Core Fabrication Process
The physical realization of the composite mold requires precise fabrication steps for both the graphite mold and the sand core.
1. Graphite Mold Manufacturing: The outer mold was CNC-machined from a block of ISO-63 or equivalent fine-grained graphite. Machining provided the high dimensional accuracy required for the external profile of the shell castings. After machining, a critical step is vacuum degassing. Graphite is porous and can absorb moisture and gases from the atmosphere, which would be released during casting, causing gas porosity in the titanium. The mold was heated in a vacuum furnace according to a specific thermal profile, typically ramping to 800-1000°C under high vacuum, to desorb these contaminants.
2. Expendable Sand Core Fabrication: The process for the core is multi-stage:
- Mixing & Molding: High-alumina sand was mixed with silica sol binder and a hardening agent in a controlled ratio. This mixture was packed into a precision corebox (with a designed draft angle) and compacted.
- Curing & Drying: The core was allowed to chemically cure at room temperature, developing its green strength. It was then extracted from the corebox and air-dried to remove residual moisture.
- Coating Application: The dried core was dipped or sprayed with the Y2O3 slurry. The coating thickness is crucial; too thin fails to protect, too thick can crack or impede collapsibility. A thickness of 0.2-0.5 mm was targeted.
- Firing: The coated core was fired in an air furnace to ~1000°C. This process sinters the coating, creating a coherent refractory layer, and burns out any organic residues from the binder, enhancing the core’s collapsibility after casting.
The fired core was inspected for cracks, dimensional accuracy, and coating uniformity before assembly into the graphite mold. Assembly used locator pins and square registers to ensure precise alignment between the core and the mold cavity, which is essential for achieving uniform wall thickness in the final shell casting.
Melting, Casting, and Post-Processing Parameters
The integrated process parameters are summarized below. The vacuum arc skull melting ensures a clean, superheated titanium pool. The centrifugal parameters (rotational speed) are calculated based on the mold radius and the required mold-filling pressure to overcome surface tension in thin sections. The relationship can be simplified as the pressure head, \( P \), generated:
$$ P = \frac{1}{2} \rho \omega^2 (r_2^2 – r_1^2) $$
where \( \rho \) is melt density, \( \omega \) is angular velocity, and \( r_1 \) and \( r_2 \) define the radial position of the melt in the mold. This pressure must be sufficient to force metal into the thinnest sections of the shell casting.
| Process Stage | Parameter | Value / Range |
|---|---|---|
| Mold Preparation | Preheat Temperature | 250 – 300 °C |
| Graphite Degas Temperature | ~900 °C (under vacuum) | |
| Core Firing Temperature | ~1000 °C (in air) | |
| Melting & Casting | Melting Method | Vacuum Arc Skull Melting |
| Pouring Temperature | ~1750 °C | |
| Casting Method | Centrifugal Casting | |
| Centrifugal Speed | 150 rpm | |
| Chamber Vacuum | ≤ 6 Pa | |
| Post-Casting | Cooling Time in Vacuum | 180 – 240 min |
| Heat Treatment | Hot Isostatic Pressing (HIP) | |
| Surface Finishing | Sand Blasting (Alumina grit) |
Results, Verification, and Discussion
Following the developed process, the thin-walled titanium alloy shell castings were successfully produced. Visual inspection confirmed complete filling of all thin sections and internal ribs, with no visible cold shuts or surface tears. The as-cast surface finish on the internal cavity, defined by the sand core’s Y2O3 coating, was notably smooth. Dimensional verification was performed using 3D optical scanning. The point cloud data from the actual shell casting was compared to the nominal CAD model. The results demonstrated that the dimensional profile of the critical non-machined internal surfaces was controlled within ±0.2 mm, successfully meeting the CT6 tolerance specification. Surface roughness measurements via profilometry or comparison samples confirmed a surface roughness (Ra) in the range of 3.2 to 6.3 µm for the as-cast internal surfaces, which is acceptable for many non-wearing functional applications and validates the effectiveness of the refractory coating.
The composite mold approach proved its advantages conclusively. The graphite mold provided the necessary thermal control and dimensional stability for the outer surface. The expendable sand core, with its engineered coating, was successfully removed after casting, leaving a complex internal geometry with good surface quality. This combination avoided the high cost and long lead time of monolithic ceramic cores while solving the chilling and core removal problems associated with using graphite for complex internal features. The process demonstrated excellent reproducibility for the shell castings.
Conclusion
This study successfully developed and demonstrated a robust composite mold casting process for manufacturing thin-walled, complex titanium alloy shell castings. By integrating a precision-machined graphite outer mold with an expendable, refractory-coated sand core, the process effectively balanced the conflicting requirements of thermal management, dimensional accuracy, surface finish, and economic feasibility. Numerical simulation with ProCAST was an indispensable tool for optimizing the gating and feeding system, predicting and eliminating shrinkage defects. The final shell castings met all stringent technical specifications, including complete filling of thin sections, dimensional accuracy within ±0.2 mm on as-cast surfaces, and a surface roughness between 3.2 and 6.3 µm. This composite mold strategy presents a highly viable and efficient manufacturing route for advanced thin-walled titanium components, offering a compelling alternative to traditional investment casting with ceramic cores for a wide range of applications in high-performance industries. The principles established here are directly applicable to the production of other complex, thin-walled shell castings where precision, quality, and cost are critical factors.