The continuous evolution of the rail transit industry imposes increasingly stringent demands on vehicle components, particularly in terms of performance-to-weight ratio. Lightweight design has become a paramount objective, as it directly contributes to reducing vehicle mass, lowering energy consumption for traction and braking, and conserving raw materials. Traditionally, structural casting parts for rail transit, such as housings and shells, have been manufactured from conventional ductile iron grades like QT400 to QT600. While offering good castability and reasonable mechanical properties, these materials present limitations for advanced lightweight applications where higher strength and toughness are required without a significant increase in section thickness.
This research focuses on overcoming these limitations by developing a high-performance rail transit shell using Austempered Ductile Iron (ADI), specifically targeting the grade EN-GJS-800-10 (equivalent to QT800-10). ADI is a unique class of materials produced by subjecting ductile iron to an austempering heat treatment. This process yields a microstructure of acicular ferrite and high-carbon, stabilized austenite, known as ausferrite. This microstructure provides an exceptional combination of high tensile strength, good ductility, high fatigue strength, and wear resistance, often surpassing the properties of conventional quenched and tempered steels at a lower density. The successful implementation of ADI for complex, safety-critical casting parts like transit shells requires a holistic approach, integrating optimal alloy design, robust casting process engineering, and precise heat treatment control.
The development target was a structural shell casting part with a mass of approximately 36.7 kg. A primary technical challenge stemmed from its geometrically complex design featuring significant variations in wall thickness, ranging from a minimum of 9 mm to a maximum of 58 mm. This non-uniformity poses substantial difficulties for both the foundry process and the subsequent heat treatment. During casting, thick sections are prone to shrinkage defects and slower cooling rates, which can lead to undesirable microstructural constituents like pearlite or coarse graphite. During austempering, varying section sizes risk non-uniform transformation kinetics, potentially resulting in inconsistent mechanical properties throughout the casting part. Therefore, the initial phase of this project involved a detailed structural analysis to identify critical regions requiring special attention in process design.
Material Design and Alloying Strategy
The base material for ADI is ductile iron. The chemical composition must be carefully tailored to ensure: 1) excellent graphite nodulization, 2) sufficient hardenability to avoid pearlite formation during the quenching stage of austempering, and 3) stabilization of the high-carbon austenite in the final ausferritic matrix. Silicon plays a crucial role in inhibiting carbide formation and raising the austenite transformation temperature. However, excessive silicon can embrittle the matrix. Carbon content influences both the graphite phase and the carbon available for enriching the austenite during the austempering hold. Manganese, while improving hardenability, tends to segregate at cell boundaries and can promote the formation of brittle carbides. Copper is an effective alloying element that enhances hardenability with less proneness to segregation compared to manganese.
For the rail transit shell casting parts, the target chemical composition was designed as follows:
| Element | Target Range (wt.%) | Primary Function |
|---|---|---|
| C | 3.60 – 3.80 | Graphite formation, Austenite Carbon Enrichment |
| Si | 2.35 – 2.55 | Ferritizer, Carbide Inhibitor |
| Mn | ≤ 0.50 | Hardenability (limited due to segregation risk) |
| Cu | 0.40 – 0.70 | Hardenability, Matrix Strengthening |
| Mg | 0.035 – 0.055 | Graphite Nodulization |
| P | ≤ 0.05 | Minimized (embrittling element) |
| S | ≤ 0.02 | Minimized (interferes with nodulization) |
The mechanical property targets for the final ADI casting parts were set according to the EN-GJS-800-10 specification:
| Property | Minimum Requirement |
|---|---|
| Tensile Strength (Rm) | 800 MPa |
| Yield Strength (Rp0.2) | 500 MPa |
| Elongation (A) | 10 % |
| Brinell Hardness (HBW) | 260 – 320 |
Casting Process Development and Numerical Simulation
The production of high-integrity ADI casting parts begins with a sound and defect-free casting. The geometry of the shell mandated a carefully engineered foundry process. A preliminary process was designed for a horizontal molding line with a two-cavity mold. The system featured a vertical sprue, horizontal runners with filters to reduce turbulence and slag inclusion, and three ingates to ensure a controlled fill. To address solidification shrinkage, a feeder (riser) was placed on the top of the casting, sized according to the modulus principle where the feeder modulus should be approximately 1.2 times the casting modulus: $M_{riser} \approx 1.2 \times M_{casting}$. The casting modulus $M_{casting}$ is calculated as the volume-to-surface area ratio of the relevant section.
To validate and optimize this initial design, numerical simulation using MAGMAsoft was employed. The 3D model of the casting part, including the gating system and feeder, was meshed, resulting in over 450,000 elements. Key process parameters for the simulation were defined as shown in the table below:
| Simulation Parameter | Value |
|---|---|
| Casting Material | ADI (QT800-10) |
| Molding Sand Material | Silica Sand with Resin Coating |
| Pouring Temperature | 1400 °C |
| Pouring Time | 10 s |
| Sand Initial Temperature | 25 °C |
The initial simulation predicted a high risk of shrinkage porosity in the thickest sections of the shell. The thermal analysis showed isolated liquid pools, or “hot spots,” forming within the casting body, indicating that the feeder’s feeding range was insufficient to compensate for shrinkage in these remote, heavy sections. This is a critical defect that would severely compromise the performance of the final casting parts.
To rectify this, the process was optimized by strategically incorporating chills. Chills are metallic inserts placed in the mold that rapidly extract heat from the solidifying metal. By accelerating solidification at specific locations, chills can effectively control the solidification sequence, extending the effective feeding distance of the feeder and promoting directional solidification towards it. The design of a chill considers its chilling power, which is related to its volume and contact area. A simplified approach to sizing involves ensuring the chill has sufficient heat capacity. The required thickness $\delta$ of a simple plate chill can be estimated from its mass $G$, density $\rho$, and contact area $A$:
$$
\delta = \frac{G}{\rho \times A}
$$
In practice, the size and placement are optimized via simulation. For this shell, two different sizes of steel chills were placed in the mold cavity adjacent to the critical thick sections.
The modified simulation, incorporating chills, showed a dramatic improvement. The thermal centers and the last points to solidify were successfully shifted entirely into the feeder. The simulation predicted a sound, shrinkage-free casting part, confirming the effectiveness of the chill-augmented process. This optimized design was approved for physical trial production.
Austempering Heat Treatment Process
The transformation of a standard ductile iron casting part into a high-performance ADI component is achieved through the austempering heat treatment cycle. This is a two-stage isothermal process that must be precisely controlled to obtain the desired ausferritic microstructure.
- Austenitization: The castings are heated to a temperature within the fully austenitic region, typically between 860°C and 920°C. For the developed alloy, a temperature of 880°C was selected. The holding time at this temperature, approximately 2-3 hours, is critical to ensure complete dissolution of carbides and homogenization of carbon in the austenite matrix. The required time $t_{aust}$ can be estimated based on the section thickness $d$ of the casting part, often following a relationship like $t_{aust} \propto d^n$, where n is an empirical constant.
- Quenching and Isothermal Transformation: After austenitization, the castings are rapidly quenched into a molten salt bath maintained at the austempering temperature. The transfer time must be swift (typically less than a few seconds) to avoid the nose of the pearlite transformation curve on the Time-Temperature-Transformation (TTT) diagram. The selected austempering temperature range for achieving EN-GJS-800-10 properties was 350°C to 400°C. At this temperature, the supercooled austenite undergoes the “ausferrite” transformation. The isothermal hold time (1-2 hours) allows for the complete reaction to form the desired mixture of acicular ferrite and carbon-saturated austenite. The kinetics of this transformation can be modeled by the Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation: $$ f = 1 – \exp(-k t^n) $$ where $f$ is the transformed fraction, $t$ is time, and $k$ and $n$ are temperature-dependent constants.
- Cooling: After the isothermal hold, the casting parts are removed from the salt bath and air-cooled to room temperature. No further tempering is required.
Results, Analysis, and Discussion on Alloying Effects
To systematically investigate the influence of alloy composition and inoculation on the final properties of the ADI shell casting parts, three distinct experimental melts were conducted, with variations primarily in Mn, Cu, and the addition of a bismuth-containing inoculant. Test coupons were attached to the actual castings to represent their metallurgical condition. The chemical compositions, mechanical properties, and microstructural characteristics of these trials are summarized below.
| Trial ID | Key Composition (wt.%) & Change | Rm (MPa) | Rp0.2 (MPa) | A (%) | HBW | Nodularity (%) | Graphite Count (mm⁻²) |
|---|---|---|---|---|---|---|---|
| 1 | Base: C~3.75, Si~2.35, Mn=0.49, Cu=0.66, No Bi | 770-801 | 487-502 | 6.0-6.5 | 318-321 | 80 | 166 |
| 2 | Reduced: Mn=0.47, Cu=0.53, Added 0.10% Bi inoculant | 897-918 | 572-581 | 10.5-13.5 | 309 | 90 | 307 |
| 3 | Optimized: C=3.65, Si=2.52, Mn=0.38, Cu=0.44, Added 0.15% Bi inoculant | 971-984 | 602-610 | 16.5-20.0 | 293-296 | 95 | 345 |
The results clearly demonstrate a profound evolution in properties. Trial 1, with higher Mn and Cu but no specific inoculation enhancement, yielded the lowest performance. The microstructure showed a nodularity of only 80%, a low graphite nodule count (166 mm⁻²), and the presence of some compacted/vermicular graphite. This poor graphite morphology directly led to inferior ductility (6% elongation), despite the tensile strength barely meeting the 800 MPa minimum. The higher hardness suggests a greater amount of martensite or other transformation products due to incomplete austenitization or transformation, likely linked to the segregated alloying elements.
Trial 2, with a reduction in Mn and Cu and the introduction of 0.10% bismuth-based inoculant, marked a significant improvement. Bismuth acts as a potent graphitizing agent and nucleation site for graphite during solidification. This resulted in a finer and more uniform distribution of graphite nodules (307 mm⁻²) and improved nodularity (90%). The refined and more spherical graphite allowed for a more efficient and uniform austenite-to-ausferrite transformation during heat treatment. Consequently, both strength and, more notably, ductility increased substantially, fully meeting the EN-GJS-800-10 specification.
Trial 3 represents the optimized composition. By further fine-tuning the carbon and silicon to 3.65% and 2.52% respectively, slightly reducing Mn and Cu, and increasing the bismuth-containing inoculant to 0.15%, the best overall balance was achieved. The Carbon Equivalent (CE), calculated as $CE = \%C + \frac{\%Si}{3}$, was approximately 4.46, which is in an ideal range for castability and graphite formation. This recipe produced the highest quality graphite structure: 95% nodularity and 345 nodules per mm². This superior starting microstructure translated into exceptional final properties after austempering: a tensile strength approaching 1000 MPa, a yield strength over 600 MPa, and remarkable elongation values between 16.5% and 20.0%, far exceeding the 10% requirement. The hardness was in the lower part of the specified range, confirming a fully ausferritic matrix without brittle phases.
The metallographic examination confirmed that the matrix in the successful trials (2 and 3) consisted entirely of the characteristic ausferrite. The high-carbon austenite (retained austenite) in this structure provides the excellent ductility and toughness through its transformation-induced plasticity (TRIP) effect under stress, while the acicular ferrite provides high strength. The relationship between graphite characteristics (nodularity, count) and mechanical properties in ADI casting parts is crucial; superior graphite quality ensures uniform stress distribution and facilitates the desired phase transformation during heat treatment.
Conclusion
This comprehensive development project successfully culminated in the production of a high-performance rail transit shell casting part in Austempered Ductile Iron grade EN-GJS-800-10. The key to success was the integrated optimization of the entire manufacturing chain:
- Casting Process Integrity: The use of MAGMA numerical simulation was instrumental in identifying and solving solidification-related defects. The optimized process, incorporating strategically placed chills, ensured the production of sound, dense, and shrinkage-free castings, which is the fundamental prerequisite for high-performance casting parts.
- Alloy Design and Inoculation: The systematic investigation revealed that an alloy with approximately 3.65% C, 2.52% Si, 0.38% Mn, and 0.44% Cu, combined with 0.15% of a bismuth-containing inoculant, produced the optimal as-cast graphite morphology. This resulted in high nodularity (95%) and a high nodule count (345 mm⁻²), creating the ideal precursor microstructure for heat treatment.
- Controlled Austempering: The subsequent two-stage isothermal heat treatment (austenitization followed by quenching and holding in a salt bath at 350-400°C) consistently transformed the well-structured ductile iron into ADI with a fully ausferritic matrix.
The final ADI casting parts exhibited a superior combination of tensile strength (~980 MPa), yield strength (~605 MPa), and elongation (~18%), significantly surpassing the standard requirements. The hardness was consistently within the 260-320 HBW range. This demonstrates that ADI is a highly viable and advantageous material for demanding, weight-sensitive applications like rail transit components. The developed process provides a robust framework for manufacturing other complex, high-integrity ADI casting parts, balancing lightweight design with exceptional mechanical performance and reliability.