In recent years, the demand for large-section ductile iron casting components, such as wind turbine hubs, bases, and nuclear waste containers, has significantly increased due to the growth of clean energy sectors like wind and nuclear power. These applications require exceptional mechanical properties and structural integrity, making the production of heavy-section ductile iron casting a critical area of research and industrial practice. As part of a nuclear energy development project, our team undertook the task of manufacturing a large annular ductile iron casting with a maximum wall thickness of 330 mm and a weight of approximately 33 tons. This component, made of QT400-18 grade ductile iron, features a symmetrical structure with an outer diameter of 3,675 mm and a height of 1,344 mm, including 52 reinforcing ribs. The production of such large-section ductile iron casting poses challenges like prolonged solidification times, severe graphite degeneration, and the risk of defects such as graphite distortion and coarseness, particularly in thermal centers or thick sections. This article details our first-hand experience in utilizing electric arc furnace melting for this large-section ductile iron casting, focusing on process design, numerical simulation, melting control, and treatment techniques to achieve desired microstructures and properties.
The design of the casting process is paramount for ensuring the quality of large-section ductile iron casting. Based on the annular geometry of the component, we adopted a bottom-gating open system with a sectional area ratio set as follows: ∑Ainner : ∑Ahorizontal : ∑Avertical = 2.0 : 1.5 : 1.0. The dimensions of the sprue, runner, and ingate were calculated using the choke section method. Ceramic tubes were employed for the gating system, incorporating filters and slag traps to minimize inclusions. For feeding, necked insulating risers were placed uniformly along the circumference to ensure even solidification and shrinkage compensation. Chills were applied on both the inner and outer circles of the annular ductile iron casting to enhance cooling rates, reduce solidification time, and minimize stress concentration. The chill thickness was determined through empirical guidelines and numerical simulation, ranging from 280 to 320 mm. The distribution of the gating system, risers, and chills is illustrated in the following diagram, which highlights the strategic placement to achieve directional solidification.
To optimize the process, we conducted numerical simulations using Huazhu CAE software to predict solidification behavior and defect formation in the large-section ductile iron casting. The simulation analyzed shrinkage porosity and cavity tendencies under different chill conditions. The solidification time was calculated based on heat transfer principles, with the governing equation for transient heat conduction expressed as:
$$ \rho C_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + Q $$
where $\rho$ is density, $C_p$ is specific heat, $T$ is temperature, $t$ is time, $k$ is thermal conductivity, and $Q$ represents internal heat sources. For sand molds with chills, the boundary conditions were adjusted to account for enhanced cooling. The results, summarized in Table 1, show that chills drastically reduce solidification time compared to sand molds alone. At a chill thickness of 300 mm, the solidification time is minimized to 2.72 hours, representing an 80% reduction. Further increases in chill thickness yield diminishing returns due to thermal equilibrium. The defect analysis, as shown in the simulation outputs, indicates that chills significantly reduce the volume and distribution of shrinkage defects, with 300 mm chills providing the best results. This validated our selection of 300 mm chills for the production of this ductile iron casting.
| Cooling Condition | Solidification Time (hours) | Solidification Time (seconds) |
|---|---|---|
| Sand Mold Only | 13.25 | 47,700 |
| Sand Mold + 250 mm Chills | 2.80 | 10,085 |
| Sand Mold + 300 mm Chills | 2.72 | 9,784 |
| Sand Mold + 320 mm Chills | 2.72 | 9,811 |
| Sand Mold + 350 mm Chills | 2.72 | 9,803 |
The molding process utilized self-hardening sand, with a segmented core assembly approach along the circumferential direction to match the symmetrical geometry of the ductile iron casting. Chills were spaced 20–30 mm apart and coated with a graphite-based alcohol paint approximately 1.5 mm thick. The mold was divided into three parts: bottom, middle, and top. The gating system and attached test blocks (designed per GB/T 1348-2009, with 70 mm thickness) were placed in the bottom box, the ductile iron casting in the middle box, and risers in the top box. After assembly, the mold was dried at 400°C for over 8 hours to achieve a cavity temperature of 80–100°C, ensuring proper venting and reducing moisture-related defects.
Melting and treatment are critical stages in producing high-quality ductile iron casting. For this large-section component, we employed an electric arc furnace to melt nearly 40 tons of metal, accounting for a 10% surplus over the total weight of the casting and risers (36 tons). The charge consisted of high-purity pig iron and scrap steel, with careful control of chemistry and temperature. The target composition for the ductile iron casting is listed in Table 2. To achieve this, we calculated the charge makeup considering carbon loss during melting and used carburizers to adjust carbon levels. The melting process involved batch additions of materials, with each melt cycle followed by slag removal, temperature measurement, and composition analysis. The superheat temperature was maintained at 1,490–1,500°C, and the tapping temperature controlled at 1,440–1,460°C. The importance of precise temperature control in electric arc furnace melting for ductile iron casting cannot be overstated, as it affects graphite nucleation and final properties.
| Element | Target Range |
|---|---|
| C | 3.60–3.80 |
| Si | 1.90–2.30 |
| Mn | < 0.10 |
| S | < 0.02 |
| P | < 0.03 |
| Mg | 0.04–0.06 |
For treatment, a 50-ton bottom-pour dam-type ladle was used. In one side of the ladle dam, 1.4% nodulizer (e.g., magnesium-ferrosilicon) was placed, covered with 0.4% high-calcium barium inoculant and iron chips. The other side contained 0.15% high-calcium barium inoculant. During tapping, two-thirds of the molten iron was rapidly poured into the ladle to initiate nodularization and primary inoculation. After reaction completion, the remaining iron was added, followed by slag removal. Secondary inoculation was performed by adding 0.15% high-calcium barium inoculant to the ladle surface. After slag skimming, the temperature was measured at approximately 1,350°C, and pouring commenced with stream inoculation using sulfur-oxygen inoculants. Post-pouring, insulating materials like cenospheres were sprinkled on risers to slow cooling. The ductile iron casting was shaken out below 300°C to prevent cracking.
The challenges in electric arc furnace melting for such a massive ductile iron casting included ensuring accurate metal weight, consistent composition, and uniform temperature. To address weight accuracy, the furnace lining was repaired to allow complete tapping. Temperature uniformity was achieved by stirring before measurements and regulating furnace power. Carbon adjustment was critical; initial carbon after melting was 3.5%, and carburizers raised it to 3.7% to meet targets. The ladle design included a channel in the dam to ensure complete drainage, facilitating smooth pouring of the ductile iron casting. The final chemical composition of the treated iron is shown in Table 3, which aligns well with the specifications for this ductile iron casting.
| Element | Measured Value |
|---|---|
| C | 3.71 |
| Si | 2.20 |
| Mn | 0.09 |
| S | 0.019 |
| P | 0.022 |
| Mg | 0.05 |
Mechanical properties and microstructure were evaluated from attached test blocks. The results, summarized in Table 4, indicate that the ductile iron casting meets the required standards. The microstructure, as observed in metallographic samples, showed a graphite nodularity of 90%, with graphite size of 5–6 grade and a fully ferritic matrix. The tensile strength and elongation exceed the target values for QT400-18, even though the wall thickness of 330 mm surpasses the 200 mm limit in GB/T 1348-2009. This demonstrates the effectiveness of our process for large-section ductile iron casting. Non-destructive testing via ultrasonic inspection revealed no internal defects, confirming the integrity of the ductile iron casting. The successful production of this component underscores the viability of electric arc furnace melting for heavy-section ductile iron casting applications.
| Property | Target Value | Measured Value (Average of 3 Samples) |
|---|---|---|
| Tensile Strength (MPa) | 350 | 366 |
| Yield Strength (MPa) | 200 | 215 |
| Elongation (%) | 15 | 24.5 |
| Elastic Modulus (GPa) | 150 | 157 |
From a theoretical perspective, the solidification of ductile iron casting involves complex phase transformations. The growth of graphite nodules can be described by diffusion-controlled kinetics, often modeled using the Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation for phase transformation kinetics:
$$ f = 1 – \exp(-kt^n) $$
where $f$ is the transformed fraction, $k$ is a rate constant, $t$ is time, and $n$ is the Avrami exponent. For graphite nucleation in ductile iron casting, factors like inoculation efficiency and cooling rate play crucial roles. The cooling rate $R$ affects graphite nodule count $N$, which can be estimated empirically for large-section ductile iron casting as:
$$ N = A \cdot R^b $$
with $A$ and $b$ as material constants. In our process, the use of chills increased $R$, thereby refining graphite structure and improving mechanical properties. Additionally, the Mg treatment for nodularization follows reaction kinetics that can be expressed as:
$$ [\text{Mg}]_{\text{final}} = [\text{Mg}]_{\text{added}} – \alpha \cdot [\text{S}] – \beta \cdot t $$
where $\alpha$ and $\beta$ are coefficients accounting for sulfur neutralization and fading, respectively. Our strict control of S content (<0.02%) and rapid treatment minimized Mg loss, ensuring high nodularity in the ductile iron casting.
Further considerations in producing large-section ductile iron casting include thermal stress management. The temperature gradient $\nabla T$ during cooling induces stresses $\sigma$ that can be approximated by:
$$ \sigma = E \cdot \alpha \cdot \Delta T $$
where $E$ is Young’s modulus, $\alpha$ is the thermal expansion coefficient, and $\Delta T$ is the temperature difference. By using chills and controlled cooling, we reduced $\Delta T$, thereby minimizing residual stresses in the ductile iron casting. This is vital for components like nuclear waste containers, where dimensional stability is critical.
In summary, the production of large-section ductile iron casting via electric arc furnace melting requires a holistic approach integrating process design, simulation, and precise metallurgical control. Our practice demonstrates that with optimized gating and chilling, rigorous melting practices, and advanced treatment techniques, high-quality ductile iron casting can be achieved even for sections exceeding 300 mm. The key takeaways include the importance of numerical simulation for defect prediction, the role of chills in shortening solidification times, and the necessity of multi-stage inoculation for consistent graphite formation. Future work may focus on further refining cooling strategies and exploring alloy modifications for enhanced performance in ductile iron casting applications. This experience contributes to the broader knowledge base for manufacturing heavy-section ductile iron casting, supporting advancements in energy and industrial sectors.