Enhancing automotive comfort and fuel efficiency relies heavily on advanced transmissions. The 9AT gearbox, featuring a nested planetary gearset design within compact dimensions, offers smoother shifting and 10%-16% better fuel economy compared to 6AT transmissions. However, producing its complex main housing via high-pressure die casting presents significant challenges in minimizing gas entrapment defects. This work details the systematic optimization of injection speed switching points using simulation and experimental validation to achieve defect-free castings.
Material, Geometry, and Initial Die Casting Setup
The ADC12 aluminum alloy (Table 1) was selected for its high strength-to-weight ratio, essential for supporting transmission gears and bearings under dynamic loads. The main housing (Fig. 1) measures approximately 470 mm × 400 mm × 400 mm, weighs 12.6 kg, and features uneven wall thickness (4-30 mm), deep cavities, ribs, and oil galleries. This complexity creates inherent die casting challenges like turbulence, gas entrapment, shrinkage, and stress concentrations.
| Si | Cu | Fe | Zn | Mn | Mg | Ni | Ti | Al |
|---|---|---|---|---|---|---|---|---|
| 9.6-12.0 | 1.5-3.5 | ≤1.3 | ≤1.0 | ≤0.5 | ≤0.3 | ≤0.55 | ≤0.3 | Bal. |
The gating system (Fig. 2) utilized a multi-branch design with the main ingate positioned near the deep cavity region and a secondary branch targeting the intricate valve plate face. Initial die casting parameters were established based on industry standards and empirical knowledge:
- Melt Temperature: 680 °C
- Die Temperature: 200 °C
- Shot Sleeve Diameter: 150 mm
- Low Speed ($v_{low}$): 0.2 m/s
- High Speed ($v_{high}$): 3.5 m/s
- Injection Pressure: 80 MPa
- Clamping Force: 30,500 kN (Bühler Carat 305 die casting machine)
The critical parameter under investigation was the high-low speed switching position ($S_{switch}$), defined as the plunger displacement where velocity transitions from $v_{low}$ to $v_{high}$. Three distinct $S_{switch}$ values were simulated and tested:
- Scheme 1: $S_{switch}$ = 480 mm (Metal reaches ingates)
- Scheme 2: $S_{switch}$ = 520 mm (Metal fronts meet centrally)
- Scheme 3: $S_{switch}$ = 560 mm (Metal fronts meet near branch junction)
Flow Simulation: Impact of Switching Position
Flow-3D simulations provided detailed insights into melt flow behavior and gas entrapment probability for each die casting scheme. The velocity field ($\vec{v}$) and volume of fluid (VOF) were solved using the Navier-Stokes equations and continuity equation:
$$ \rho \left( \frac{\partial \vec{v}}{\partial t} + \vec{v} \cdot \nabla \vec{v} \right) = -\nabla p + \mu \nabla^2 \vec{v} + \rho \vec{g} $$
$$ \frac{\partial \phi}{\partial t} + \nabla \cdot (\phi \vec{v}) = 0 $$
where $\rho$ is density, $\mu$ is dynamic viscosity, $p$ is pressure, $\vec{g}$ is gravity, and $\phi$ is the fluid fraction.
Scheme 1 ($S_{switch}$ = 480 mm): Premature high-speed initiation caused severe jetting (Fig. 3b) as molten metal entered the cavity. Turbulence dominated the flow front, leading to significant splashing and air entrainment, particularly in the shallow left cavity (marked area). Gas entrapment probability exceeded 30%, reaching 50% in critical zones.
Scheme 2 ($S_{switch}$ = 520 mm): Delaying the switch improved flow stability. Metal fronts coalesced more smoothly before high-speed filling commenced. While jetting reduced substantially, residual turbulence in the shallow left cavity still trapped gas (probability ~50% locally), and end-of-fill regions showed moderate entrapment (Table 2).
Scheme 3 ($S_{switch}$ = 560 mm): Optimal metal front coalescence occurred before the high-speed phase. Flow transitioned smoothly into a near-laminar pattern (Fig. 5), filling deep and shallow regions synchronously. Minimal jetting and uniform advancement drastically reduced air entrainment. Gas entrapment probability fell below 13% in the shallow cavity and under 30% at the end-of-fill.
| Switching Position (mm) | Shallow Cavity (%) | End-of-Fill Regions (%) | Overall Flow Stability |
|---|---|---|---|
| 480 | >30 (up to 50) | >30 (up to 50) | Poor (Severe Jetting) |
| 520 | ~30 | ~50 (localized) | Moderate (Residual Turbulence) |
| 560 | <13 | <30 | Excellent (Near-Laminar) |
Experimental Validation and Results
Scheme 3 parameters ($v_{low}$ = 0.2 m/s, $v_{high}$ = 3.5 m/s, $S_{switch}$ = 560 mm) were implemented in production trials using the Carat 305 die casting machine. The resulting ADC12 transmission housings (Fig. 7) exhibited excellent surface finish, dimensional accuracy, and no visible surface defects like cold shuts or oxides.
Microstructural analysis (Fig. 8) revealed a fine-grained, dense microstructure consisting of primary α-Al phase (dendritic and globular morphologies) and α-Al + Si eutectic with needle/platelet Si particles. Minor blocky intermetallic phases were also present. Faster solidification at the end-of-fill (farthest from ingates) promoted a finer, more globular α-Al structure compared to regions near the ingate.
Tensile testing (Table 3) confirmed the mechanical integrity met specifications (Tensile Strength > 190 MPa, Elongation > 1%). The near-ingate region, benefiting from slower cooling and potential minor porosity reduction due to optimized die casting flow, showed superior properties:
- Near Ingate: UTS = 272.0 MPa, Elongation = 3.4%
- End-of-Fill: UTS = 230.6 MPa, Elongation = 2.7%
| Sampling Location | Tensile Strength (MPa) | Elongation (%) |
|---|---|---|
| Near Ingate | 272.0 ± 5.2 | 3.4 ± 0.3 |
| End-of-Fill | 230.6 ± 4.8 | 2.7 ± 0.2 |
Conclusions
This study demonstrates the critical influence of the high-low speed switching point in high-pressure die casting of complex, thin-walled components like 9AT transmission housings. Flow simulation proved essential for visualizing flow instabilities and gas entrapment risks associated with premature switching. Key findings are:
- Switching too early ($S_{switch}$ = 480 mm) induces severe jetting, turbulence, and high gas entrapment (>30-50%), detrimentally impacting casting quality in the die casting process.
- A moderate delay ($S_{switch}$ = 520 mm) improves flow stability but may leave localized turbulence, particularly in thin sections or end-of-fill areas during die casting.
- Optimal switching ($S_{switch}$ = 560 mm, melt fronts fully coalesced) promotes near-laminar flow, minimizing jetting and reducing gas entrapment probability to <13% in critical zones. This die casting parameter set yielded sound castings with excellent surface finish.
- Microstructures were uniformly fine-grained and dense. Mechanical properties significantly exceeded requirements, with near-ingate strength (272 MPa) and elongation (3.4%) reflecting the quality achievable with optimized die casting parameters.
The relationship between switching position ($S_{switch}$), flow front position ($L_{front}$), and gas entrapment probability ($P_{gas}$) can be summarized for this geometry as:
$$ P_{gas} \propto \frac{1}{L_{front}(S_{switch}) $$
This work provides a validated methodology for optimizing injection parameters in high-pressure die casting, particularly for intricate automotive components where minimizing internal defects is paramount for performance and durability.