The pursuit of high-integrity, defect-free sand castings, particularly for critical structural components, remains a central challenge in foundry engineering. Complex geometries with varying wall thicknesses often lead to internal defects like shrinkage porosity and cavities, compromising the mechanical performance and service life of the final part. This article details a comprehensive case study on the process optimization for a demanding gray iron (HT300) upper rotary disc casting, leveraging numerical simulation as the core tool for virtual prototyping and iterative design refinement. The component under investigation is a large, plate-like structure with numerous bosses, ribs, and through-holes, widely used in heavy machinery where it must withstand significant static and dynamic loads. The primary objective was to develop a robust sand casting process that effectively eliminates shrinkage-related defects, ensuring the required strength, stability, and wear resistance.
The initial step involved a thorough analysis of the 3D CAD model. The upper rotary disc, with an outline dimension of approximately 1281 mm x 1270 mm x 101 mm and a mass of about 502 kg, presented a typical scenario of uneven section modulus. The main challenges identified were:
- A large, thin bottom plate (minimum thickness 15 mm).
- Several isolated heavy sections or “hot spots,” particularly around the central hub and various boss features, with a maximum thickness of 113 mm.
- The requirement for sound metal in load-bearing areas and sealing surfaces.
For such sand castings, uncontrolled solidification would inevitably lead to macro-shrinkage in the heavy sections. The design philosophy adopted was to promote directional solidification towards strategically placed feeders (risers).
Initial Process Design and Numerical Simulation Setup
An initial casting process was designed for these sand castings. Considering the large, flat bottom surface, a horizontal pouring position was selected to avoid the complexities of “flat molding, vertical pouring.” A bottom-gating system was chosen over a top-gating system after preliminary flow analysis, as it offered a more tranquil fill with less turbulence and oxidation, which is crucial for quality sand castings. The initial gating system comprised one sprue, two runners, and seven ingates, designed with a choke area ratio. The mold and cores were assumed to be made of furan resin sand. For simulation, the ProCAST software’s iron (gravity) module was employed. The key material properties and boundary conditions are summarized below.
| Parameter | Value / Specification |
|---|---|
| Casting Material | Gray Iron HT300 |
| Liquidus Temperature | ~1200 °C |
| Pouring Temperature | 1370 °C |
| Mold Material | Furan Resin Sand |
| Heat Transfer Coefficient (Metal-Sand) | 500 W/(m²·K) |
| Heat Transfer Coefficient (Metal-Chill) | 2000 W/(m²·K) |
| Simulation Focus | Filling, Solidification, Defect Prediction (Shrinkage) |
The governing equations for the simulation involve solving the Navier-Stokes equations for fluid flow during filling and the energy equation for heat transfer during solidification. The key energy conservation equation is:
$$ \rho c_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \rho L \frac{\partial f_s}{\partial t} $$
Where \( \rho \) is density, \( c_p \) is specific heat, \( T \) is temperature, \( t \) is time, \( k \) is thermal conductivity, \( L \) is latent heat, and \( f_s \) is the solid fraction. The term \( \rho L \frac{\partial f_s}{\partial t} \) represents the release of latent heat during the phase change, critical for accurately modeling the solidification pattern of sand castings.
Analysis of Initial Simulation Results and Defect Prediction
The filling simulation confirmed a smooth, wave-front advancement of the metal with no severe impingement or air entrapment, validating the bottom-gating design for these sand castings. The total fill time matched the designed value closely. However, the solidification simulation and shrinkage prediction analysis revealed significant issues. The Niyama criterion, a common indicator for predicting microporosity in castings, was calculated post-solidification. The criterion is often expressed in its simplified form for evaluation:
$$ N_y = \frac{G}{\sqrt{\dot{T}}} $$
Where \( G \) is the temperature gradient (°C/m) and \( \dot{T} \) is the cooling rate (°C/s). Regions with a Niyama value below a critical threshold (specific to the alloy) are prone to shrinkage porosity.
The results clearly showed that the heavy sections, isolated from easy feed paths, were the last to solidify. These areas developed isolated liquid pools, leading to volumetric shrinkage. The defect prediction module highlighted major shrinkage cavities and dispersed porosity in the central hub and several thick bosses. This confirmed that the natural solidification pattern of these sand castings, without adequate feeding, was unacceptable. The table below summarizes the critical findings from the initial simulation.
| Aspect | Result | Assessment |
|---|---|---|
| Filling Behavior | Stable, complete fill. No cold shuts or misruns. | Acceptable |
| Solidification Sequence | Non-directional. Thin walls solidified first, trapping heavy sections. | Unacceptable |
| Major Defect Locations | Central hub region, multiple thick bosses. | Unacceptable |
| Primary Cause | Lack of effective feeding to compensate for solidification shrinkage in hot spots. | Needs Redesign |
Process Optimization Strategy for Sand Castings
Based on the simulation findings, the optimization strategy focused on two main pillars: 1) Implementing an effective feeding system to promote directional solidification, and 2) Modifying the gating system to better align with the new thermal gradients. The goal was to redesign the process so that the thermal center of the sand castings would shift from the problematic hot spots into the feeders.
Riser Design and Placement: A combination of different riser types was designed. For the massive central hub, large top risers (necked-down for easy removal) were placed directly above it. For smaller, isolated heavy sections like bosses, smaller blind risers were positioned. The riser dimensions were calculated using modulus-based methods. The modulus (Volume/Surface Area ratio, \( M \)) of a riser must be greater than the modulus of the casting section it is intended to feed. For a cylindrical riser, the relationship can be simplified for design purposes:
$$ M_{riser} = \frac{D}{4} \quad \text{(for a side-insulated riser)} $$
Where \( D \) is the riser diameter. The riser was designed such that \( M_{riser} > 1.2 \times M_{casting\_hotspot} \).
| Riser Location (Feeds) | Type | Key Dimensions (Diameter x Height) mm | Calculated Modulus \(M\) (cm) |
|---|---|---|---|
| Above Central Hub | Top, Necked | 180 x 300 | ~4.5 |
| Above Major Bosses (4x) | Blind Riser | 80 x 180 | ~2.0 |
Gating System Redesign: The initial bottom-gating system, while good for fill stability, was not optimal for the new directional solidification scheme requiring heat to be directed upward toward the top risers. Therefore, the gating was changed to a top-pouring system. This allows the hotter metal to reside at the top of the mold cavity (where the risers are), establishing a strong thermal gradient from the bottom (which cools first) to the top risers (which cool last). The new system used a larger single sprue, a runner, and four ingates to maintain a controlled fill time. The pouring temperature was maintained at 1370°C.
Validation Through Simulation of the Optimized Sand Castings Process
The entire process—including the new risers, gating, and the modified virtual mold—was remeshed and simulated. The filling analysis for the optimized sand castings process showed a slightly more turbulent but still acceptable fill pattern, with no evidence of slag or sand entrainment into critical areas. The crucial solidification analysis told a completely different story from the initial run.
The temperature gradient and solid fraction plots demonstrated a clear directional solidification sequence:
- The thin bottom plate and peripheral walls solidified first.
- Solidification then progressed upwards through the ribs and lighter sections.
- The heavy sections (hub, bosses) remained liquid, now being fed by the liquid metal in the risers above them.
- Finally, the risers themselves solidified, containing the last liquid pool and the resultant shrinkage cavity.
The “last-to-freeze” region was successfully moved from the casting body into the riser necks and riser bodies.
The definitive proof was in the shrinkage prediction results. The Niyama criterion plot and the direct shrinkage cavity prediction showed a dramatic improvement. The major shrinkage defects previously present in the hub and bosses were completely eliminated from the final sand castings. Minor, isolated regions of potential microporosity were reduced to negligible levels acceptable for the component’s duty. The final defect was confined almost exclusively to the risers, which are removed during post-casting cleanup. This validated the optimization strategy for producing high-quality sand castings.
| Evaluation Criteria | Initial Process | Optimized Process |
|---|---|---|
| Directional Solidification | Absent. Multiple isolated hot spots. | Excellent. Clear progression from bottom to top risers. |
| Major Shrinkage in Casting Body | Present in >5 critical locations (Hub, Bosses). | Effectively eliminated (0 major cavities). |
| Predicted Microporosity Level | High in thick sections. | Very low, within acceptable limits. |
| Riser Efficiency | N/A (No risers used). | High. Shrinkage successfully transferred to risers. |
| Overall Casting Soundness | Unacceptable for structural application. | Fully acceptable, meeting technical requirements. |
Conclusion and Implication for Sand Castings Production
This case study underscores the indispensable role of numerical simulation in the modern development of reliable processes for complex sand castings. For the HT300 upper rotary disc:
- Initial Design is Insufficient: Even a logically designed gating system failed to prevent shrinkage defects in heavy sections, highlighting that intuition alone is inadequate for such sand castings.
- Simulation-Driven Optimization is Effective: By identifying problem areas virtually, a targeted optimization strategy combining strategic riser placement and gating redesign was formulated and validated without costly physical trials.
- Key to Success: The fundamental principle was to enforce directional solidification. Changing from a bottom-gate to a top-gate system was crucial in establishing the correct thermal gradient to work in concert with the top risers.
The final optimized process ensured that the critical structural areas of the sand castings were free from shrinkage defects, guaranteeing the mechanical properties required for its demanding application. This methodology provides a proven framework for tackling similar challenges in the production of high-integrity, heavy-sectioned sand castings, leading to improved quality, reduced scrap rates, and shorter development cycles.