In this study, I focus on the development of a robust sand casting process for a critical automotive cushion block. This component serves as a key structural part in vehicle transmission systems, primarily functioning to dampen vibrations and absorb mechanical shock loads. Given its demanding service conditions involving cyclic stresses, the casting must exhibit high integrity, free from defects such as cold shuts, cracks, shrinkage porosity, and macro-shrinkage cavities. The geometry, as shown in the analysis, is a flat-plate type with overall dimensions of 259 mm × 179 mm × 102 mm and a net weight of approximately 7.4 kg. Its complex internal structure features multiple ribs, cavities, and slots, presenting significant challenges for manufacturability through sand castings. The average wall thickness is 8.8 mm, with a minimum of 8 mm, aligning with the typical feasibility limits for ferrous sand castings.
The material specified for this application is ductile iron QT450-10, chosen for its excellent combination of strength, ductility, and castability. Its typical chemical composition range is critical for achieving the desired microstructure and mechanical properties, as summarized in Table 1. The melting process utilized a duplex furnace system (cupola + induction furnace) to ensure precise chemistry control and high-quality molten metal suitable for sand castings.
| Element | Content (wt.%) |
|---|---|
| C | 3.70 – 4.00 |
| Si | 2.15 – 2.93 |
| Mn | 0.46 – 0.66 |
| P | ≤ 0.035 |
| S | ≤ 0.016 |
| Mg | 0.027 – 0.050 |
| RE (Rare Earth) | 0.026 – 0.043 |
1. Casting Process Design
1.1 Parting Plane and Pouring Position Selection
The initial and most critical step in designing the sand casting process is determining the optimal parting plane and pouring position. Four distinct schemes were evaluated. Schemes positioning the two critical functional planes (the bases of the internal cylindrical cavities) at the top were rejected. They would require complex and numerous sand cores, increase the risk of defects on the critical surfaces, and complicate molding. The preferred scheme positions these critical functional planes downward. This offers several advantages for sand castings: it simplifies core placement and support, minimizes the number of cores required, and ensures the critical surfaces are formed against the dense sand mold, enhancing their finish and dimensional accuracy. Furthermore, the top surface, which is non-functional and contains numerous raised pads, is oriented upward. These pads act as natural hot spots and are ideal locations for positioning feeders (risers) to effectively control solidification.
The pouring position was integrated with the parting plane. A bottom-gating system was selected over a top-gating alternative. The governing equations for fluid flow highlight the rationale. The velocity of metal entering the mold cavity is related to the effective head height $h$:
$$ v = C_d \sqrt{2gh} $$
where $v$ is the velocity, $C_d$ is the discharge coefficient, and $g$ is gravity. In a top-gating system, the initial metal stream falls with high velocity, causing severe turbulence, sand erosion (washing), and oxide formation. A bottom-gating system minimizes the drop height, reducing the metal velocity and ensuring a quiescent, progressive fill of the mold cavity, which is essential for high-quality sand castings of complex geometry.
1.2 Gating System Design
A bottom-gated, overlap (step) gating system was engineered. This design further promotes laminar flow. The system consists of a pouring cup, a sprue (vertical channel), a sprue base well/slag trap, a runner (horizontal channel), and ingates (gates connecting the runner to the mold cavity). Ceramic filters with a mesh size of 2.5 mm × 2.5 mm were placed at the sprue base to trap inclusions. The runner cross-section was trapezoidal to aid in slag buoyancy and separation. The ingates were designed as flat trapezoids, distributed to ensure balanced filling. The choke area, which controls the pouring time, was calculated based on the casting weight and desired fill rate. For sand castings, the pour time $t$ can be estimated as:
$$ t = \frac{W}{k \cdot A_c \cdot \sqrt{h}} $$
where $W$ is the casting weight, $k$ is a constant, $A_c$ is the choke area, and $h$ is the effective metallostatic head.
1.3 Determination of Process Allowances
Key process allowances were determined according to industry standards for medium-volume sand castings. The casting tolerance grade was set to CT11. Machining allowances (MA) were applied to all functional surfaces. The allowance value depends on the casting dimension $D$ and the tolerance grade:
$$ MA = f(CT, D, \text{Surface Position}) $$
For this ductile iron casting, the allowances were specified as follows: 2.0 mm for the bottom (drag side) surfaces, 1.5 mm for the top (cope side) surfaces, and 0.5 mm for vertical side walls. A linear shrinkage (patternmaker’s) allowance of 0.8% was applied to all relevant dimensions to compensate for the contraction of ductile iron during solidification and cooling. Draft angles of 0°35′ were incorporated on all vertical faces to facilitate pattern and core removal from the molds and core boxes.
2. Core Design and Assembly
The internal complexity of the cushion block necessitates the use of multiple sand cores. A monolithic core would be extremely difficult to produce and handle. Therefore, a split-core design strategy was adopted, breaking down the internal cavity into five separate core pieces. This is a common and effective practice in complex sand castings to improve manufacturability and accuracy.
The cores were numbered 1# through 5#. Cores 1#, 2#, and 3# are positioned in the same horizontal plane, while cores 4# and 5# are vertically oriented and interlock with the former set. Precision positioning is achieved through integrated locators: mating grooves and corresponding protrusions on the contact faces of cores 1#, 2#, and 3#, and stepped surfaces for cores 4# and 5#. This interlocking design is crucial for maintaining strict alignment within the sand mold assembly, preventing core shift, and ensuring the dimensional fidelity of the final sand castings.
Core ventilation is paramount. As the hot metal surrounds the cores, the binder in the sand thermally decomposes, generating large volumes of gas. If not vented, this gas can penetrate the molten metal, causing blows or porosity. Ventilation channels were incorporated into the design of all five cores. Cores 4# and 5# are vented through their upper surfaces into the cope. Cores 1#, 2#, and 3# are vented both upwards and laterally through the sides of the mold. This multi-path venting strategy ensures efficient gas evacuation during the pour, a critical factor for sound sand castings.
| Core # | Primary Function | Orientation | Locating Feature | Venting Path |
|---|---|---|---|---|
| 1# | Forms central cavity & ribs | Horizontal | Groove/Protrusion with 2# & 3# | Upward & Lateral |
| 2# | Forms central cavity & ribs | Horizontal | Groove/Protrusion with 1# & 3# | Upward & Lateral |
| 3# | Forms central cavity & ribs | Horizontal | Groove/Protrusion with 1# & 2# | Upward & Lateral |
| 4# | Forms side cavity | Vertical | Stepped surface | Upward |
| 5# | Forms side cavity | Vertical | Stepped surface | Upward |
3. Numerical Simulation and Defect Prediction
To validate and optimize the initial sand casting process design, I employed computational modeling using AnyCasting software. A three-dimensional model of the casting, complete with the gating system, was created, meshed, and assigned material properties. The mold material was set as silica sand, and the casting material was QT450-10. The key simulation parameters were: a pouring temperature of 1,350 °C, an interfacial heat transfer coefficient of 0.1 W/(m²·°C) between the casting and the sand mold, and a solid fraction cutoff for feeding.
3.1 Filling and Initial Solidification Analysis
The simulation of the filling phase confirmed the efficacy of the bottom-gating design. The metal front advanced smoothly without significant turbulence or air entrainment. The total filling time was approximately 4.0 seconds. The subsequent solidification analysis, however, revealed areas of concern. The temperature gradient and solidification sequence indicated that while the thin walls and outer sections solidified rapidly, several isolated hot spots remained at the junctions of thicker sections, particularly in the lower region around the cylindrical bosses. These are potential sites for shrinkage defects.
A more precise defect prediction was obtained by analyzing the Niyama criterion or the residual melt modulus, which is a function of temperature gradient $G$ and cooling rate $\dot{T}$:
$$ NY = \frac{G}{\sqrt{\dot{T}}} $$
Areas with a low Niyama value are prone to micro-porosity. The initial simulation predicted a 0.18% volume fraction of the casting as being at high risk for shrinkage porosity, distributed across several locations, confirming the need for process modification to direct solidification more effectively.
4. Process Optimization: Risers and Chills
Based on the simulation results, the process was optimized to achieve directional solidification, where the casting sections solidify progressively toward locations where extra metal (feed metal) is available. This is a fundamental principle for producing sound sand castings.
4.1 Riser (Feeder) Design
To feed the thermal centers at the top of the casting, nine small cylindrical risers were placed on the non-functional raised pads. The riser design must satisfy two main requirements: it must remain molten longer than the casting section it feeds (via its modulus, which is Volume/Surface Area ratio), and it must contain sufficient feed metal volume. The modulus of the riser $M_r$ must be greater than the modulus of the casting hot spot $M_c$:
$$ M_r > M_c \quad \text{where} \quad M = \frac{V}{A} $$
For this application, risers with a height of 14.4 mm were designed to meet the modulus requirement. Additionally, an atmospheric vent was added at the highest point of the mold to allow air to escape during filling and to prevent back pressure that could impede the flow of metal into the risers.
4.2 Chill Design
To accelerate the solidification of the isolated hot spots in the lower part of the casting, external chills were employed. Chills are metal inserts placed in the sand mold that rapidly extract heat, creating a steep temperature gradient. Eight steel chills were strategically positioned in a circular arrangement around the problematic thick sections at the base of the cylindrical features. The chill action forces these regions to solidify earlier, effectively turning them into extensions of the solidified skin and eliminating them as isolated hot spots. This promotes a more favorable solidification sequence toward the risers located at the top.
4.3 Analysis of the Optimized Process
The simulation was re-run with the optimized system including risers and chills. The results demonstrated a dramatic improvement. The solidification pattern became distinctly directional, progressing from the chilled regions at the bottom and the thin walls toward the risers at the top. The risers functioned as the last points to solidify, as intended. Analysis of the defect prediction parameter showed that the volume of potential shrinkage defects was reduced by 50%, down to 0.09% of the casting volume. Furthermore, the remaining at-risk areas were now concentrated in the risers themselves, which are removed during subsequent machining. This optimization directly translates to a higher yield and more reliable sand castings. The calculated casting yield improved to over 90%, which is excellent for a complex ductile iron sand casting.
| Metric | Initial Design | Optimized Design | Improvement |
|---|---|---|---|
| Predicted Shrinkage Volume (%) | 0.18% | 0.09% | -50% |
| Solidification Pattern | Isolated Hot Spots | Directional to Risers | Corrected |
| Riser Function | Not Present | Active Feeding, Last to Freeze | Established |
| Estimated Casting Yield | Lower | > 90% | Significantly Higher |
5. Conclusions
This comprehensive study successfully demonstrated the complete design and optimization cycle for producing a complex automotive cushion block via sand castings. The key conclusions are:
- Process Design: A bottom-gated, overlap gating system combined with a carefully selected parting plane proved optimal for this geometry, ensuring a tranquil fill and protecting critical surfaces. The use of a split-core assembly with integrated locators was essential for achieving the required internal complexity and dimensional accuracy in the sand castings.
- Simulation-Driven Optimization: Numerical simulation was an indispensable tool for identifying inherent solidification defects in the initial design. It provided a clear rationale for implementing a combined feeding strategy using risers and chills.
- Optimization Efficacy: The strategic placement of chills at lower hot spots and risers at upper thermal centers successfully transformed the solidification pattern into a controlled, directional sequence. This engineering intervention halved the predicted shrinkage volume and confined the remaining defect-prone areas to the sacrificial riser bodies.
- Production Outcome: The final optimized sand casting process is capable of producing high-integrity ductile iron castings with excellent dimensional precision, minimal internal defects, and a high yield, fully meeting the stringent performance requirements for automotive applications. This work underscores the synergy between foundational sand casting principles and modern simulation technology in advancing manufacturing reliability.