In modern foundry engineering, the increasing demand for high-quality castings with short delivery times has driven the adoption of computer-aided numerical simulation. Traditional trial-and-error methods for casting process design are time-consuming and costly, often leading to a high scrap rate. By simulating the solidification process, it is possible to predict sand foundry defects such as shrinkage cavities, porosity, and hot tears before production begins. This article presents a systematic approach to applying solidification simulation technology to plate-like ductile iron castings, integrating three-dimensional modeling, thermal field analysis, and defect criteria to optimize gating and risering systems. The methodology is validated through industrial case studies, demonstrating significant improvements in casting quality and reduction of sand foundry defects.
Background and Significance
Foundry industry forms the backbone of manufacturing, yet many process designs still rely on empirical rules. The complexity of solidification—coupled with the thermal and metallurgical behavior of ductile iron—makes defect prediction challenging. Solidification simulation offers a scientific approach to visualize temperature distribution, identify isolated liquid pools, and predict the formation of shrinkage-related sand foundry defects. The benefits include reduced prototyping cycles, lower material waste, and enhanced product reliability. According to the Engineering Committee of the U.S. National Research Council, simulation can improve product quality by 5–15%, increase material yield by 10–15%, reduce engineering costs by 15–30%, and shorten design cycles by 30–60%.
Fundamentals of Solidification Simulation
The numerical simulation of casting solidification is essentially a heat transfer problem involving latent heat release, phase change, and mass flow in the mushy zone. For ductile iron, the precipitation of graphite during eutectic solidification causes volume expansion, which can compensate for shrinkage if the mold rigidity is sufficient. However, if the expansion is insufficient or the feeding channel is blocked, sand foundry defects like internal shrinkage cavities appear. The simulation process follows a modular workflow:
- Create three-dimensional solid models of the casting, gating system, risers, cores, and chills.
- Export the model data to a simulation software package.
- Define material properties, initial temperatures, heat transfer coefficients, and boundary conditions.
- Run the numerical calculation using finite difference or finite element methods.
- Post-process the results to visualize temperature fields, solidification time, and defect indicators.
Solidification Characteristics of Ductile Iron
Ductile iron exhibits a “pastry-like” solidification mode, with graphite nodules growing within a liquid matrix. The volume changes during solidification are governed by the competition between:
- Liquid contraction before the start of solidification
- Solidification shrinkage (liquid to solid transformation)
- Graphite expansion (precipitation of low-density graphite)
- Mold wall movement
The net volume change determines the need for external feeding. A stiff mold minimizes cavity enlargement, allowing graphite expansion to offset shrinkage. Conversely, a soft mold (e.g., green sand) may expand, increasing feeding demand. Therefore, mold rigidity is a primary factor affecting sand foundry defects in ductile iron.
The governing heat conduction equation during solidification is:
$$ \rho C_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \dot{Q}_L $$
where ρ is the density, Cp is the specific heat, T is temperature, t is time, k is thermal conductivity, and Q̇L is the latent heat source term, expressed as:
$$ \dot{Q}_L = \rho L \frac{\partial f_s}{\partial t} $$
Here, L is the latent heat of fusion and fs is the solid fraction. To handle latent heat, techniques such as the temperature recovery method, equivalent specific heat method, and enthalpy method are applied.
Prediction of Shrinkage Porosity and Cavities
Shrinkage defects in castings are primarily caused by the volume deficit during solidification. When the feeding path is open, concentrated shrinkage cavities form at the highest fluid region or in the riser. When dendrites interlock and block the feeding, dispersed micro-porosity (shrinkage porosity) appears. The simulation predicts these sand foundry defects using several criteria:
Temperature Gradient Method
A low temperature gradient in the liquid region often leads to porosity because the mushy zone is broad and feeding is difficult.
Solid Fraction Gradient
High solid fraction gradients indicate rapid local solidification, which can isolate liquid pockets.
G/F Method
The ratio of temperature gradient G to cooling rate R is used to assess feeding efficiency. When G/R falls below a critical value, porosity is likely.
For three-dimensional simulations, we use the Niyama criterion:
$$ N = \frac{G}{\sqrt{\dot{T}}} $$
where G is the temperature gradient and Ṫ is the cooling rate. A small N value indicates a high risk of micro-porosity. The critical value depends on the alloy system and mesh size.
Residual Melt Modulus
The residual melt modulus Mr is defined as:
$$ M_r = \frac{V_{melt}}{A_{melt}} $$
where Vmelt is the volume of isolated liquid and Amelt is its surface area. A large Mr suggests a large isolated pool, which increases the risk of shrinkage cavities. This parameter is directly output by most simulation codes and is useful for comparing alternative designs.
Dynamic Isolation of Melt Pools
In complex castings, multiple hot spots may become isolated as solidification progresses. To accurately model feeding, it is necessary to dynamically partition the remaining liquid into isolated pools. The algorithm works as follows:
- Identify all cells with solid fraction below the critical feeding solid fraction (typically 0.7–0.8).
- Assign a temporary temperature to one liquid cell and propagate a search to its six neighbors (in a Cartesian mesh) to find all connected liquid cells.
- Mark this cluster as one isolated melt pool.
- Repeat for unmarked liquid cells to identify all pools.
- Compute shrinkage volume separately for each pool and remove liquid from that pool only, not from others.
This approach prevents erroneous feeding across solidified regions and dramatically improves the accuracy of shrinkage cavity prediction in multi-feeder castings. It also helps in identifying the exact locations where sand foundry defects are likely to form.
Case Study 1: Engine Bearing Cap – Process Optimization
Original Process Description
A ductile iron bearing cap (grade QT400-18) was produced on a high-pressure molding line using green sand. The casting weight was approximately 6.8 kg, with four cavities per mold. The pouring temperature was 1380–1420 °C. After trial production, X-ray inspection revealed shrinkage porosity near the bolt holes, as shown by the dark regions in the simulated solidification time plot. The original riser design could not maintain a feeding channel until the end of solidification.
The three-dimensional model of the original casting and gating system was prepared using CAD software. The solidification simulation was performed with a finite difference solver. The results indicated that at a certain time (e.g., 120 seconds after pouring), the liquid metal in the bolt hole region became isolated from the riser due to premature solidification of the riser neck. This isolation caused a negative pressure, leading to the formation of shrinkage cavities.
Improvement Scheme 1 – Enlarged Riser
In the first attempt, the riser height was increased from 30 mm to 40 mm while keeping other parameters unchanged. The simulation was repeated using the same pouring temperature (1380 °C). The residual melt analysis showed:
| Parameter | Original Design | Scheme 1 (Enlarged Riser) |
|---|---|---|
| Residual melt volume (mm³) | 8,450 | 8,720 |
| Residual melt surface area (mm²) | 3,600 | 3,650 |
| Residual melt modulus (mm) | 2.35 | 2.39 |
| Probability of shrinkage defects | High | High |
Enlarging the riser did NOT eliminate the isolated pool at the bolt hole area. The feeding distance remained too long, and the riser neck still solidified early. Therefore, Scheme 1 was rejected.
Improvement Scheme 2 – Relocated Gate and Vent
The second scheme involved repositioning the ingate to the top of the casting, directly under the riser, reducing the feeding distance to the hot spot. Additionally, a small vent was placed near the bolt hole to accelerate cooling and promote directional solidification. The simulation results after this modification are summarized below:
| Parameter | Original Design | Scheme 2 (Relocated Gate) |
|---|---|---|
| Residual melt volume (mm³) | 8,450 | 1,200 (mostly in riser) |
| Residual melt surface area (mm²) | 3,600 | 950 |
| Residual melt modulus (mm) | 2.35 | 1.26 |
| Probability of shrinkage defects | High | Low |
As shown in the solidification sequence, the casting solidified before the riser, and the feeding channel remained open throughout. No isolated liquid pool remained inside the casting. Production trials confirmed that the castings produced with Scheme 2 were free from shrinkage porosity, as evidenced by sectioning and dye penetrant inspection. This case demonstrates that simulation-guided modification of the gating system can effectively eliminate sand foundry defects without increasing material consumption.
Case Study 2: Large Bearing Cap – Process Design
Initial Design and Simulation
A large bearing cap weighing approximately 45 kg was designed for a heavy-duty engine. The material was QT400-18, and the solidification characteristics required a riser at the thickest section (max wall thickness 65 mm). The initial gating design used a side riser with a neck diameter of 35 mm. Simulation predicted a solidification time distribution where the riser neck solidified prematurely. The simulation output showed a “no liquid region” near the top of the casting, indicating a shrinkage cavity would form beneath the riser.
To quantify the risk, the following values were extracted:
| Time (s) | Liquid Volume in Casting (cm³) | Isolated Pool Volume (cm³) | Defect Indicator |
|---|---|---|---|
| 50 | 850 | 0 | None |
| 150 | 420 | 85 | Potential |
| 250 | 120 | 65 | High risk |
| 350 | 0 | 0 | Cavity formed |
The defect formed because the riser volume was insufficient to feed the local solidification shrinkage after the feeding path was blocked. Since increasing the riser size was impractical (due to sand box dimensions and low yield), the process was improved by adding external chills.
Modified Design with Chills
Chills were placed at the bottom and side surfaces of the casting near the thick section. The chill dimensions were designed based on the modulus ratio:
$$ \frac{M_{chill}}{M_{casting}} \approx 0.8 $$
where M is the modulus (volume/surface area). The modified simulation showed:
| Parameter | Initial Design | Modified with Chills |
|---|---|---|
| Casting solidification time (s) | 620 | 560 |
| Riser neck solidification time (s) | 480 | 520 |
| Temperature gradient near hot spot (K/mm) | 0.02 | 0.08 |
| Shrinkage cavity predicted | Yes | No |
The chills increased the local cooling rate, creating a steeper temperature gradient and shifting the final solidification zone into the riser. Figure 5 shows the solidification sequence after modification; the casting surface is fully solidified while the riser still contains liquid. Production validation confirmed that the modified process yielded sound castings with no internal cavities.
Role of Simulation in Reducing Sand Foundry Defects
The two case studies illustrate different strategies for minimizing sand foundry defects:
- Case 1: Relocating the gate and increasing the feeding efficiency of an existing riser.
- Case 2: Using chills to control the thermal gradient and reduce the feeding demand.
In both cases, simulation provided quantitative data that allowed engineers to make informed decisions without costly physical trials. The key simulation outputs used as defect criteria were:
- Solidification time distribution
- Residual melt volume and surface area
- Temperature gradient at critical sections
- Niyama criterion values
- Isolated pool dynamics
The following table summarizes the defect criteria applied in each case:
| Defect Criterion | Mathematical Expression | Application |
|---|---|---|
| Solidification time | $t_s = \frac{M^2}{K^2}$ | Identify hot spots |
| Niyama value | $N = G / \sqrt{\dot{T}}$ | Predict micro-porosity |
| Residual melt modulus | $M_r = V_{melt}/A_{melt}$ | Evaluate isolated pools |
| Temperature gradient | $G = |\nabla T|$ | Assess feeding path |
Using these criteria, foundry engineers can identify the exact locations where sand foundry defects are likely to form and then modify the process accordingly. This proactive approach replaces the “cast and inspect” mentality with a “simulate and optimize” methodology.
Conclusion
This article has demonstrated the application of solidification simulation to the production of plate-like ductile iron castings. By simulating the temperature field and dynamically partitioning isolated liquid pools, it is possible to predict shrinkage cavities and porosity with good accuracy. The case studies on bearing caps confirmed that simulation-guided modifications can eliminate sand foundry defects while reducing material waste and development time.
The main conclusions are:
- Solidification simulation is a powerful tool for optimizing casting processes and reducing sand foundry defects.
- Residual melt parameters (volume, surface area, modulus) are reliable indicators of feeding problems.
- The dynamic isolation of melt pools is essential for accurate shrinkage prediction in multi-feeder castings.
- Both gating design changes and chill placement can be effectively evaluated through simulation before physical trials.
Future work should focus on integrating microstructure simulation with solidification models to predict mechanical properties, and on expanding the application to more complex geometries and alloy systems. The continuing advancement of simulation technology will further reduce the occurrence of sand foundry defects, improve foundry efficiency, and enable the production of high-integrity castings at lower cost.