In the realm of high-performance aluminum alloys for aerospace and advanced structural applications, ZL205 (Al-Cu based) alloy stands out due to its exceptional strength-to-weight ratio. However, its wide solidification range and inherent poor castability, particularly in precision investment casting of thin-walled components, present significant challenges. Defects such as shrinkage porosity, hot tears, and misruns are prevalent, directly impacting yield and component reliability. Numerical simulation has become an indispensable tool for designing and optimizing casting processes, yet its predictive accuracy is fundamentally constrained by the fidelity of input parameters, most critically, the interfacial heat transfer coefficient (IHTC) between the molten metal and the ceramic shell.
Traditional simulations often rely on assumed or literature-based IHTC values, which may not capture the dynamic and spatially variable nature of heat exchange in a specific precision investment casting setup. This work, therefore, adopts a methodology that integrates experimental temperature measurement with finite element (FE) analysis to inversely solve for the IHTC, thereby grounding the simulation in empirical data from the actual process. Using this calibrated model, we systematically investigate the filling behavior, solidification progression, and defect formation tendencies of ZL205 alloy under varying shell preheat temperatures and cooling conditions. The ultimate goal is to elucidate the solidification laws governing this alloy and define its limits of formability within the precision investment casting process.
The core challenge in simulating precision investment casting accurately lies in defining the thermal boundary condition at the metal-mold interface. The IHTC, denoted as \( h \), is defined by Newton’s law of cooling at the interface:
$$ q = h (T_{\text{metal}} – T_{\text{mold}}) $$
where \( q \) is the heat flux, and \( T_{\text{metal}} \) and \( T_{\text{mold}} \) are the temperatures of the contacting surfaces. This coefficient is not a material constant but a complex function of interface pressure, air gap formation, surface roughness, and thermophysical properties of the contacting materials. An inverse solution approach circumvents the difficulty of direct measurement. The principle involves using recorded temperature histories at specific locations within the casting and/or mold as input to an optimization algorithm that adjusts the IHTC value in an FE model until the simulated temperatures match the experimental ones.
The experimental setup for IHTC determination was integrated into a standard precision investment casting process. A cluster of test geometries, including cylindrical and plate-like shapes of varying dimensions, was designed within a bottom-gated浇注系统. K-type thermocouples were strategically embedded at three distinct locations: one within a ϕ15.3 mm cylindrical specimen (P1), and two others within the upper (P2) and lower (P3) sections of the main feeding channel. Corresponding thermocouples were also fixed on the external shell surface adjacent to these internal points. A莫来石-based shell with a nominal thickness of 10 mm was used. The alloy was melted, degassed, and grain-refined before pouring at a superheat of 725°C into the un-preheated shell (room temperature). Temperature data was recorded at a high frequency throughout the pouring and solidification cycle.
| Measurement Point | Location Description | Embedment Depth | Primary Purpose |
|---|---|---|---|
| P1 | Within ϕ15.3 mm cylindrical specimen | 8 mm from surface | IHTC for rod-like features |
| P2 | Within upper feeder channel | 20 mm from surface | IHTC for bulk feeder sections |
| P3 | Within lower feeder channel | 20 mm from surface | IHTC for gate/bottom sections |
The recorded cooling curves from these points served as the target for the inverse algorithm in ProCAST. The FE model of the test casting was constructed, and the thermophysical properties of ZL205 alloy were assigned. The initial IHTC values were estimated, and the solver iteratively adjusted them until the sum of squared errors between the calculated and measured temperatures at each point was minimized. The optimal IHTC values obtained were:
| Interface Region | Inversely Solved IHTC (W/m²·K) |
|---|---|
| Cylindrical Specimen (P1) | 542 |
| Upper Feeder Channel (P2) | 911 |
| Lower Feeder Channel (P3) | 875 |
The significant variation in IHTC values underscores the importance of this calibration step. The lower value in the cylindrical specimen likely corresponds to earlier air gap formation due to the geometry’s faster contraction, while the higher values in the more massive feeder channels indicate prolonged intimate contact. These spatially resolved coefficients were then applied as boundary conditions (type COINC in ProCAST) in all subsequent simulations for their respective geometric regions, dramatically enhancing the physical realism of the thermal model for precision investment casting.
With a calibrated thermal boundary condition, comprehensive process simulations were conducted. The baseline process conditions for the multi-geometry test cluster were set: shell preheat temperatures of 200°C and 500°C, combined with natural cooling (air cooling) and forced air cooling (wind cooling). The filling and solidification analysis focused on understanding the limits of the precision investment casting process for ZL205.
The flow pattern for the bottom-gated system under 500°C preheat is characteristic. The metal rises from the sprue base, fills the lower horizontal runner by approximately 3.5 s (38% filled), and begins ascending into the various test cavities. Larger cross-section cavities fill rapidly, while the melt proceeds to the upper runner to fill thinner and smaller features from the top, completing the fill at about 8.2 s. A critical observation is the splashing and vortex formation at the ends of the lower runner, which is a potential source of oxide entrainment and gas pickup, a common concern in precision investment casting of reactive aluminum alloys.
The evolution of the temperature field and solid fraction is profoundly affected by shell preheat. At a low preheat of 200°C, the thermal gradient is steep, and solidification initiates rapidly. In contrast, with a 500°C preheat, the cooling rate is significantly dampened. At 12.3 s post-pour, the overall solid fraction is only 0.2%. The 3 mm thin-wall sections, despite their high surface-area-to-volume ratio, remain mostly above the liquidus temperature at this stage. By 20.1 s, these thin sections show a marked temperature drop and begin solidifying, with solid fraction isoclines showing a “peak” shape at their ends, indicating uneven cooling and potential for localized stress concentration. The complete solidification of the entire system is prolonged, occurring around 1954 s. The governing heat transfer during this phase can be described by the transient heat conduction equation with a moving boundary (Stefan problem):
$$ \rho c_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \rho L \frac{\partial f_s}{\partial t} $$
where \( \rho \) is density, \( c_p \) is specific heat, \( k \) is thermal conductivity, \( L \) is latent heat, and \( f_s \) is solid fraction. The source term \( \rho L \frac{\partial f_s}{\partial t} \) is crucial for modeling the precision investment casting solidification accurately.
Defect prediction, primarily for shrinkage porosity, was performed using a feeding-based criterion (e.g., Niyama criterion). The simulations clearly show that process parameters are decisive. With a 200°C shell preheat, extensive shrinkage is predicted in the smaller-diameter rods and thin sections due to premature isolation from the feeder and insufficient feeding. The junctions between the lower runner and the heavier sections also show significant porosity due to hot spots. Increasing the shell preheat to 500°C dramatically improves the feeding efficiency. The thermal gradients are less severe, promoting more directional solidification from the extremities toward the feeders. However, shrinkage remains predicted at the sharp corners of the lower runner ends—a result of the initial turbulent fill and the inherent difficulty of feeding such geometric features. Forced air cooling (wind cooling) was simulated but generally exacerbated shrinkage in thin sections by creating isolated, rapidly solidified zones that could not be fed, demonstrating that simply increasing cooling rate is not a viable strategy for defect mitigation in precision investment casting of ZL205.
The critical role of simulation in modern precision investment casting cannot be overstated, as it allows for the virtual testing of gating designs and process parameters before any metal is poured, saving considerable time and cost.
To validate the simulation findings, actual castings were produced under the key conditions: shell preheated to 200°C and 500°C, followed by natural cooling. The cast clusters were inspected visually for fill integrity and then by X-ray radiography for internal quality. The results corroborated the simulation predictions with remarkable consistency.
| Feature Type | Dimension (mm) | Shell Preheat 200°C | Shell Preheat 500°C | Simulation Prediction (500°C) |
|---|---|---|---|---|
| Rod | ϕ3 | No fill | No fill / Poor fill | Severe shrinkage/misrun |
| ϕ10 | Partial fill | Full fill, visible shrinkage | Full fill, porosity predicted | |
| Plate | 3 (thick) x 40 (wide) | Partial fill | Full fill, some defects | Full fill, defect prone |
| Plate | 6 x 40 | Full fill, sound | Full fill, sound | Full fill, minimal defects |
The experimental data firmly establishes that a high shell preheat temperature is essential for the successful precision investment casting of ZL205 alloy, especially for sections below 6 mm. At 500°C preheat, the alloy demonstrated the ability to fully fill a 3 mm thick plate section, albeit with a higher propensity for internal microporosity. This defines the practical limit of formability under gravity-poured precision investment casting with natural cooling. Thicker sections (≥6 mm) showed excellent integrity under both preheat conditions, confirming their lower sensitivity to the process window. The defect locations, particularly in the lower runner junctions and the upper sections of thin features, matched the simulation forecasts, validating the accuracy of the model calibrated with the inverse-derived IHTC.
This integrated study demonstrates a robust methodology for enhancing the reliability of numerical simulation in precision investment casting. The inverse determination of the interfacial heat transfer coefficient provides a critical, physics-based boundary condition that moves simulations beyond educated guesses. Applying this calibrated model to ZL205 alloy reveals that the primary lever for improving castability and reducing defects is the control of thermal dynamics through shell preheat temperature. A high preheat (~500°C) mitigates premature freezing, promotes better feeding, and extends the alloy’s formability limit down to approximately 3 mm wall thickness. However, attention must be paid to gating design to minimize turbulent entry and to manage solidification sequencing in complex junctions. The strong correlation between simulated and experimental outcomes confirms that this approach—combining targeted experimentation, inverse thermal analysis, and comprehensive process simulation—constitutes a powerful framework for the development and optimization of robust precision investment casting processes for challenging high-strength aluminum alloys.