We report a systematic investigation on the gravity filling process for a thin‑walled and geometrically intricate conical housing manufactured by the high precision investment casting route. The component is made of ZL205A aluminum alloy, a high‑strength casting alloy widely used in aerospace applications due to its excellent mechanical properties after T5 heat treatment. The casting has a total height of 220 mm, a bottom ring outer diameter of 140 mm, a top ring outer diameter of 46 mm, and an average wall thickness of only 4.5 mm. The intermediate rib region is only 3.5 mm thick, and the transition between the rib and the bottom ring creates a local thickness difference of 6.5 mm, which forms a hot spot prone to shrinkage porosity. Moreover, the alloy composition with 4.6–5.3 wt% Cu leads to a wide solidification range (544–633 °C), poor fluidity, and a strong tendency for segregation and hot tearing. To overcome these challenges, we adopted a bottom‑gating gravity filling system assisted by a horseshoe‑shaped runner and side slot gates, and we optimized the process using finite element method (FEM) simulations. The entire study demonstrates how high precision investment casting integrated with numerical simulation can successfully produce defect‑free thin‑walled complex housings with superior mechanical performance.
Material and Casting Process Design
The nominal composition of ZL205A alloy is listed in Table 1. Its solidification interval of 89 °C promotes mushy‑zone solidification, which can cause severe micro‑porosity if the feeding path is insufficient. The liquidus and solidus temperatures are 633 °C and 544 °C, respectively. The alloy also contains Cd, Ti, Zr, V, and B for grain refinement and strengthening.
| Cu | Mn | Cd | Ti | Zr | V | B | Zn | Al |
|---|---|---|---|---|---|---|---|---|
| 4.6–5.3 | 0.4 | 0.24 | 0.20 | 0.12 | 0.18 | 0.4 | 0.1 | Balance |
We designed a bottom‑gating gravity filling system to address the key forming difficulties. The system consists of a tapered sprue (bottom diameter 12 mm, top diameter 30 mm), a horseshoe‑shaped runner to distribute the melt uniformly, two side slot gates (170 mm × 5 mm) attached to the intermediate rib, and a top riser connected to the top ring. The ingate dimensions are 18 mm × 4 mm × 30 mm. This configuration was chosen after a preliminary top‑gating trial produced severe gas porosity and shrinkage in the bottom and top rings. The new design ensures that the mold is filled from the bottom upward, promoting directional solidification from bottom to top and from inside to outside. The top riser and side slot gates provide feeding during the final stage of solidification, which is critical for the thin rib region. The entire gating system was developed within the framework of high precision investment casting to achieve a sound casting with minimal defects.
FEM Simulation Setup
We employed ProCAST software for coupled fluid flow and solidification analysis. The casting was meshed with 4 mm tetrahedral elements (548 296 elements), while the ceramic shell mold was meshed with 8 mm elements (489 652 elements). The heat transfer coefficient at the casting‑mold interface was set to 500 W/(m²·K). The pouring temperature was 720 °C, and the filling velocity was controlled at 1 cm/s under gravity. The initial mold temperature was 25 °C. The thermophysical properties of ZL205A were taken from the ProCAST database. The numerical model solves the continuity, momentum, and energy equations:
$$ \frac{\partial \rho}{\partial t} + \nabla \cdot (\rho \mathbf{u}) = 0 $$
$$ \rho \frac{D \mathbf{u}}{Dt} = -\nabla p + \mu \nabla^2 \mathbf{u} + \rho \mathbf{g} $$
$$ \rho C_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \rho L \frac{\partial f_s}{\partial t} $$
where \( \rho \) is density, \( \mathbf{u} \) velocity, \( p \) pressure, \( \mu \) viscosity, \( \mathbf{g} \) gravity, \( C_p \) specific heat, \( k \) thermal conductivity, \( L \) latent heat, and \( f_s \) solid fraction. The simulation accounted for both fluid flow and heat transfer, enabling us to predict filling sequence, temperature evolution, and solid phase distribution. This level of modeling is a key enabler for high precision investment casting of complex geometries.
Filling Flow Field Results
The simulated mold filling with the bottom‑gating system is summarized in Table 2. The total filling time was 21 s. At 3 s, the sprue was fully filled. By 5 s, the horseshoe runner was filled and the melt began to enter the cavity from the bottom. At 9 s, the side slot gates started to deliver metal to the rib region. After 14 s, 50% of the cavity was filled, and the top riser began to assist filling from the top. At 18 s, about 85% of the cavity was filled, and the remaining volume was completed at 21 s. The flow front advanced smoothly without turbulence or jetting, which is essential for minimizing gas entrapment and oxide film formation. This stable filling behavior is a hallmark of a well‑designed high precision investment casting process.
| Time (s) | Filling status |
|---|---|
| 3 | Sprue fully filled |
| 5 | Horseshoe runner filled; melt enters cavity from bottom |
| 9 | Side slot gates start delivering metal to rib |
| 14 | 50% of cavity filled; top riser begins to assist |
| 18 | 85% of cavity filled |
| 21 | 100% filled |
Solidification Temperature Field
Temperature field evolution during solidification is shown in Table 3. After complete filling (21 s), the melt remained above the liquidus temperature, ensuring good fluidity. At 44 s, directional solidification began: the inner regions cooled first while the side slot gates stayed hot for feeding. By 88 s, the bottom ring solidified, creating a downward temperature gradient. At 214 s, most of the casting had solidified except the top riser, which continued to feed the bottom areas. Complete solidification of the casting occurred at 374 s. The whole system (including gating) reached full solidification at 408 s. This sequence confirms that we achieved the desired directional solidification from bottom to top and from inside to outside, which is critical for eliminating shrinkage porosity. The successful implementation of such a thermal strategy is a direct result of our high precision investment casting optimization.
| Time (s) | Event |
|---|---|
| 21 | Filling completed; melt still above liquidus |
| 44 | Inner regions start to solidify; side gates remain hot |
| 88 | Bottom ring solidifies |
| 214 | Most casting solidified; top riser still feeding |
| 374 | Casting fully solidified |
| 408 | Entire system solidified |
The solidification front velocity can be approximated by the one‑dimensional heat conduction model:
$$ v_s = \frac{k \left( \frac{\partial T}{\partial x} \right)_{solid}}{\rho L} $$
In our simulation, the temperature gradient at the solid‑liquid interface was maintained at 2–4 K/mm, which is sufficient to drive the solidification front upward. The directional solidification condition was satisfied when the local temperature gradient \( G \) and solidification velocity \( R \) obeyed the criterion for columnar growth:
$$ \frac{G}{R} \geq \frac{\Delta T_0}{D_L} $$
where \( \Delta T_0 \) is the solidification range and \( D_L \) is the solute diffusion coefficient in the liquid. Our simulation confirmed that this inequality held throughout most of the casting, ensuring a sound structure. This level of process control is a hallmark of high precision investment casting.
Solid Phase Distribution
The evolution of solid fraction in the rib region is summarized in Table 4. At 21 s, the rib was completely liquid. Nucleation started at 46 s in the central part of the rib cross‑section. By 94 s, clear solid nuclei appeared. At 148 s, the solid fraction reached about 30% in the core. After 226 s, the rib was mostly solid (90% solid fraction). Complete solidification of the rib and side slot gates occurred at 424 s. The solidification sequence was from the center of the rib outward, which is consistent with the temperature field results. This gradual solidification feeding path avoided hot spots and prevented hot tearing in the thin rib, especially at the 2 mm radius bend. The avoidance of such defects is a direct benefit of applying simulation‑based optimization in high precision investment casting.
| Time (s) | Solid fraction in rib (%) | Key event |
|---|---|---|
| 21 | 0 | Filling completed |
| 46 | ~5 | Nucleation starts at center |
| 94 | ~20 | Clear solid nuclei |
| 148 | ~30 | Solid fraction increases |
| 226 | ~90 | Rib nearly solid |
| 424 | 100 | Fully solidified |
Microstructure and Mechanical Properties
After casting, we performed T5 heat treatment (solution at 538 °C for 14 h, water quench at 40 °C, then aging at 155 °C for 9 h in air). The as‑cast grain size averaged 112 µm, while the T5‑treated grain size was 94 µm. No porosity, gas holes, or inclusions were observed in the optical micrographs, confirming the soundness achieved by the optimized high precision investment casting process. We then cut test specimens from the rib region at 1/3 height and tested their mechanical properties. The results are given in Table 5. The average ultimate tensile strength (UTS) was 453 MPa, yield strength (YS) 400 MPa, elongation 8.9%, and Brinell hardness 114 HBS. All values exceeded the design requirements (UTS ≥ 420 MPa, YS ≥ 380 MPa, elongation ≥ 4.5%). The high ductility and strength are attributed to the fine grain structure and the absence of casting defects, which are direct outcomes of the controlled directional solidification and proper feeding design. This demonstrates that a well‑executed high precision investment casting route can meet the most demanding aerospace specifications.
| Property | Measured value | Design requirement |
|---|---|---|
| Ultimate tensile strength (MPa) | 453 | ≥420 |
| Yield strength (MPa) | 400 | ≥380 |
| Elongation (%) | 8.9 | ≥4.5 |
| Brinell hardness (HBS) | 114 | — |
The tensile properties can be modeled using the Hall‑Petch relationship for the yield strength:
$$ \sigma_y = \sigma_0 + \frac{k_y}{\sqrt{d}} $$
where \( d \) is the average grain size (94 µm), \( \sigma_0 \) is the friction stress, and \( k_y \) is the Hall‑Petch coefficient. For ZL205A‑T5, typical values are \( \sigma_0 \approx 250 \) MPa and \( k_y \approx 0.2 \) MPa·m1/2, giving a predicted YS of 402 MPa, which matches our experimental result well. This further validates the quality of the high precision investment casting.
Conclusion
We have successfully developed a gravity filling process for a complex thin‑walled conical housing using ZL205A alloy via high precision investment casting. The key achievements are:
- A bottom‑gating system with a horseshoe runner, side slot gates, and a top riser was designed to ensure stable filling and directional solidification.
- FEM simulation predicted a filling time of 21 s, complete solidification time of 374 s for the casting, and 424 s for the entire system. The solidification proceeded from bottom to top and from inside to outside, satisfying the directional solidification criterion.
- The as‑cast grain size was 112 µm, and after T5 treatment it reduced to 94 µm. No defects such as porosity, gas holes, or hot tears were observed.
- Mechanical properties of specimens cut from the rib region far exceeded the design targets: UTS 453 MPa, YS 400 MPa, elongation 8.9%, and hardness 114 HBS.
This work confirms that a combination of alloy‑specific process design, numerical simulation, and precise control of the casting parameters is the cornerstone of high precision investment casting for demanding thin‑walled aerospace components. The methodology can be extended to other complex castings requiring high integrity and repeatability.