In the realm of engineering vehicles, the transmission system plays a pivotal role, and among its components, axles are fundamental for load-bearing and motion transfer. Axles can be categorized into steering axles, drive axles, steering drive axles, and support axles based on their functions. The axle housing, as a key supporting part of the axle assembly, must withstand substantial torques while enduring forces such as gravity, braking force, and axial loads, operating in complex and harsh environments. Historically, axle housings were often fabricated using welding methods, but cast axle housings have gradually become the preferred manufacturing approach due to advantages like fewer parts, simpler processes, cost-effectiveness, and superior material properties. Cast axle housings are typically made from ductile iron, which undergoes spheroidization and inoculation treatments to precipitate carbon as spherical graphite, offering high tensile strength, good impact toughness, wear resistance, vibration damping, and low notch sensitivity, with comprehensive performance close to that of steel. This study focuses on the casting process for a steering axle housing part used in forklifts, made of QT450-10 ductile iron. We aim to design and optimize the casting process through numerical simulation, ensuring the quality of these critical casting parts. The insights gained can serve as a valuable reference for the production and process optimization of similar casting parts in the automotive industry.
The steering axle housing casting part is structurally designed to house a steering hydraulic cylinder, enabling vehicle steering via steering knuckles mounted at both ends. During service, these casting parts are subjected to significant loads and fatigue stresses, necessitating high mechanical properties such as strength and fatigue resistance. The casting part weighs 70 kg, with overall dimensions of 834 mm × 228 mm × 348 mm. The average wall thickness is 25.26 mm, with a maximum thickness of 63.11 mm and a minimum of 15 mm. The material is QT450-10 ductile iron, whose chemical composition and mechanical property requirements are detailed in Table 1 and Table 2. Although ductile iron has good fluidity, its mushy solidification characteristic makes it prone to shrinkage porosity and shrinkage cavity defects. Therefore, it is essential to select molding materials with high strength and stiffness, and to utilize the graphite expansion during solidification to eliminate defects based on the principle of equilibrium solidification.
| Element | C | Si | Mn | P | S | Cu | Sb | Mg | Re |
|---|---|---|---|---|---|---|---|---|---|
| Range | 3.6–3.9 | 2.0–2.5 | 0.25–0.45 | ≤0.05 | ≤0.02 | 0.1–0.15 | ≤0.01 | 0.035–0.060 | ≤0.02 |
| Property | Tensile Strength (Rm) | Yield Strength (Rp0.2) | Elongation (A) | Hardness (HBW) |
|---|---|---|---|---|
| Requirement | ≥450 MPa | ≥310 MPa | ≥10% | 160–210 |
Sand casting is adopted due to its high production efficiency, low cost, and flexibility. Given the relatively simple structure of the casting part and the need for small-batch production, we use flask molding and core box core-making, with two molds per flask to enhance yield. The molding material selected is furan resin self-hardening sand, which is suitable for single-piece or small-batch production of iron, steel, and non-ferrous alloy castings, offering advantages like fast hardening, high tensile strength, good collapsibility, low gas generation, and high recyclability.
For the pouring position and parting surface, two schemes were considered. Scheme A involves placing the steering knuckle mounting arms vertically, using a horizontal parting plane to split the casting part equally between the cope and drag, with some machined surfaces facing upward, requiring a large integral core. Scheme B positions the steering knuckle mounting arms horizontally, employing a combined horizontal and curved parting surface, with most of the casting part in one flask and machined surfaces vertical, necessitating multiple cores due to challenging draft areas. After evaluation, Scheme B is chosen as it offers better machined surface quality, minimizes mismatch defects, and allows the use of movable blocks to reduce core count. The designed core and movable blocks are shown in Figure 1, with a core for the internal cavity and movable blocks for local protrusions, featuring draft angles for easy removal.
The gating system is designed as a semi-closed type with a cross-sectional area ratio of $$A_{\text{sprue}} : A_{\text{runner}} : A_{\text{ingate}} = 1.2 : 1.5 : 1$$. The runner has the largest area to reduce flow velocity and enhance平稳性, minimizing mold erosion. A middle pouring system is used for simplicity, placed on the parting surface. Based on the “large orifice outflow” theory, the cross-sectional areas are calculated using the following formula:
$$A_{\text{ingate}} = \frac{G_L}{\rho_L \mu \tau \sqrt{2g h_p}}$$
where $$A_{\text{ingate}}$$ is the total cross-sectional area of the ingates (cm²), $$G_L$$ is the total mass of metal in the mold including the gating system (kg), $$\rho_L$$ is the metal density (7100 kg/m³ for ductile iron), $$\mu$$ is the flow loss coefficient (0.6), $$\tau$$ is the pouring time (s), $$g$$ is the gravitational acceleration (980 cm/s²), and $$h_p$$ is the pressure head at the ingate (cm). The casting part mass is 70 kg, and with a yield of 65–70% for small-batch production, the total pouring mass $$G_L$$ is determined as 187 kg. The pouring time $$\tau$$ is calculated as:
$$\tau = S \sqrt{G_L}$$
where $$S$$ is a coefficient depending on wall thickness, taken as 2.2. Thus, $$\tau = 2.2 \sqrt{187} \approx 30$$ s, but considering the characteristics of ductile iron casting parts, the pouring time is reduced by one-third to 20 s. The pressure head $$h_p$$ is derived from:
$$h_p = \frac{k_2^2}{1 + k_1^2 + k_2^2} H_P$$
with $$k_1 = A_{\text{sprue}} / A_{\text{runner}} = 0.8$$, $$k_2 = A_{\text{sprue}} / A_{\text{ingate}} = 1.2$$, and $$H_P$$ as the average static pressure head (cm), calculated by:
$$H_P = H_0 – \frac{P^2}{2C}$$
where $$H_0 = 30$$ cm is the pressure head above the choke, $$P = 21$$ cm is the mold cavity height above the choke, and $$C = 34.5$$ cm is the total cavity height. Substituting values, $$H_P = 23.61$$ cm and $$h_p = 11.04$$ cm. Then, $$A_{\text{ingate}} = 15$$ cm², $$A_{\text{sprue}} = 18$$ cm², and $$A_{\text{runner}} = 22.5$$ cm². The gating system is verified by checking the metal rise velocity $$v_L$$ and minimum residual head height $$h_M$$:
$$v_L = \frac{h_C}{\tau}$$
where $$h_C = 34.5$$ cm is the casting height, giving $$v_L = 1.725$$ cm/s, within the reference range of 1.0–2.0 cm/s. For $$h_M$$:
$$h_M = L \tan \alpha$$
with $$L = 884$$ mm as the metal flow distance and $$\alpha = 8^\circ$$ as the pressure angle, yielding $$h_M = 124$$ mm. The actual designed minimum residual head height is 160 mm, satisfying the requirement. The gating system includes a tapered sprue (3° taper, small end diameter 50 mm), a hemispherical sprue well, a tapered runner with sections of 22 cm², 15.5 cm², and 8.1 cm², and six ingates each of 2.5 cm² area located at the bottom of the runner to improve slag trapping. The initial casting process layout is shown in Figure 2, emphasizing the arrangement for these casting parts.
Numerical simulation is conducted using ProCAST software to analyze the filling and solidification processes. The model is meshed with 159,960 surface elements and 5,073,703 volume elements. The molding material is set as resin-bonded sand with an initial temperature of 25°C, and the alloy composition follows Table 1 with a pouring temperature of 1350°C. The heat transfer coefficient between the casting part and mold is 500 W/(m²·K), and the mold is set to air cooling. The simulated filling process, completed in 19 s, shows平稳 metal flow without evident air entrainment or turbulence, aligning with the semi-closed gating design. However, solidification simulation reveals shrinkage defects in the steering knuckle mounting arms on both sides, with a total shrinkage volume of 6.22 cm³, and a depression defect at the top conical boss due to liquid shrinkage. The solidification time distribution and fraction solid slices indicate that isolated liquid zones form in the arms around 748 s to 828 s after pouring, leading to shrinkage porosity as graphite expansion is insufficient to compensate for contraction in these thin-walled areas. The top conical boss, as the highest point, suffers from liquid contraction without adequate feeding, resulting in a concave shape. This analysis underscores the need for optimization in producing such casting parts.
To address these defects, the process is optimized by adding risers. For the top depression, two open-top cylindrical necked risers are placed on both sides of the conical boss to provide feeding and venting. For the shrinkage in the arms, four side risers are designed as hot risers to feed the isolated liquid zones. The riser dimensions are calculated using the modulus method, considering varying wall thicknesses. The optimized layout includes risers with dimensions as shown in Figure 3, ensuring adequate feed metal for these casting parts. A subsequent simulation confirms that defects are eliminated from the casting part, with shrinkage now confined to the risers and gating system. The revised process is validated through actual production, where melting is performed in a medium-frequency induction furnace, with treatment using silicon-barium inoculant and rare-earth silicon-iron-magnesium spheroidizer. Pouring is done at 1350 ± 10°C, followed by shakeout after 4 hours and cleaning. The produced casting parts are tested for chemical composition, surface quality, dimensions, and mechanical properties, all meeting specifications. A sectioning test on a sample casting part, as shown in Figure 4, reveals no shrinkage or porosity, verifying the feasibility of the optimized process for high-quality casting parts.
In summary, this study successfully designs and optimizes the casting process for a steering axle housing part made of QT450-10 ductile iron. The initial design involves sand casting with a semi-closed gating system, but numerical simulation identifies defects in critical areas. By incorporating risers, the optimized process eliminates shrinkage and depression defects, as validated by simulation and practical production. This work highlights the importance of numerical simulation in refining casting processes for complex casting parts, ensuring their reliability in demanding applications. The methodology can be extended to other casting parts in the automotive sector, contributing to enhanced manufacturing efficiency and product quality. Future work may explore advanced materials or further process refinements for even better performance of these essential casting parts.
| Parameter | Value |
|---|---|
| Pouring Temperature | 1350°C |
| Mold Material | Furan Resin Sand |
| Initial Mold Temperature | 25°C |
| Heat Transfer Coefficient | 500 W/(m²·K) |
| Simulated Pouring Time | 19 s |
| Total Solidification Time | ~1300 s |
The optimization of casting processes is crucial for producing defect-free casting parts, especially in automotive applications where performance and durability are paramount. Through this case study, we demonstrate how integrated design and simulation can lead to robust solutions for challenging casting parts. The repeated emphasis on casting parts throughout this article underscores their significance in manufacturing, and the techniques discussed here can be adapted for various other casting parts to achieve similar success.